ECS2604 Textbook

Frans Barker’s
THE SOUTH AFRICAN LABOUR MARKET

Sixth edition
Derek Yu
With Pietman Roos
Van Schaik
PUBLISHERS
Published by Van Schaik Publishers
A division of Media Books
Francis Baard Street, Hatfield, Pretoria
All rights reserved
Copyright © Van Schaik Publishers
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by
any means electronic, mechanical, photocopying, recording or otherwise without written permission from
the publisher, except in accordance with the provisions of the Copyright Act of .
Please contact DALRO for information regarding copyright clearance for this publication. Any unauthorised copying could lead to
civil liability and/ or criminal sanctions.
Tel: DALRO (from within South Africa) or + ( )

Fax: + ( )
Postal address: PO Box , Braamfontein, , South Africa
http://www.dalro.co.za
First edition
Second edition
Third edition
Fourth edition
Fifth edition
Fifth revised edition
Sixth edition
ISBN
eISBN:
Commissioning editor Marike Visagie

Production manager Werner von Gruenewaldt
Editorial coordinator Daleen Venter
Copy editor Paula Marais
Proofreaders Beverlie and Linton Davies
Cover design by Werner von Gruenewaldt
Cover image by Shutterstock
Typeset in Sabon LT Std on by Pace-Setting & Graphics, Pretoria
Printed and bound by ABC Press
eBook conversion by InfoGrid Pacific
Please note that reference to one gender includes reference to the other.
Website addresses and links were correct at time of publication.
Every effort has been made to obtain copyright permission for material used in this book. Please contact the
publisher with any queries in this regard.
This book has been reviewed by independent peer reviewers.

Preface
Since the s there have been some very exciting and intellectually stimulating developments in the
field of labour. Sociopolitical occurrences in recent years have, if anything, brought to the fore the
close relationship between developments in the labour market and progress on the socio-econo-
political terrain. The ideological divides in South Africa are especially apparent in the labour market,
and these compound the basic conflict between the objectives of protecting basic worker rights on
the one hand, and increasing economic growth on the other. There is obviously less conflict between
these objectives over the long term than in the short term.
During the s and the s the writings of academics and practitioners, and also the legislature,
often concentrated on the first aspect, i.e. protecting basic worker rights. However, increasingly
people involved in the labour arena have come to realise that the interaction between economic
issues and the labour market is of critical importance to ensure the success of a “new” South Africa.
High unemployment, skilled labour shortages, a high wage-low productivity economy, the effects of
globalisation, and programmes to address labour market inequalities are all examples of issues that
have the potential to build upon, or to destroy, the new deal agreed on by politicians.
This book aims to contribute to the current and ongoing critical debate on these issues. Established
subjects such as labour law, union structures and collective bargaining form a lesser but still important
part of the subject matter. They are discussed in the context of their impact on the functioning of the
labour market.
The purpose of the book is to provide the general reader with an overview of the South African labour
market, its characteristics and problems, and above all, its functioning. With the possible exception of
a few sections that deal with more complex labour economics theories (referred to as “Deeper
insight”), the contents should be quite easily understood by any person interested in this topic.
In order to make the information accessible, it has not been possible to deal with the subject matter
in great depth. There are inevitable omissions, but the reference list provides further reading matter
on the various issues. To this end, care has been taken to concentrate on South African source
material, though not exclusively.
In some instances the arguments for and against a particular approach are presented briefly. A
specific standpoint has purposely not always been assumed to avoid a trend prevalent in many
quarters these days – taking a superficial view on a very involved topic, without giving enough
opportunity for thorough analysis, debate and even negotiation.
Non-economists. Non-economists may be disheartened to see some complicated graphs in parts of

this book. Even for economics students not yet completely comfortable with graphs, the task of
understanding these may seem daunting. But fear not! Non-economics readers can skip the more
complicated labour economics sections (or “Deeper insight”) without it seriously affecting their
understanding of the book.
Nevertheless, understanding the theoretical working of the labour market will help the reader gain a
better understanding of how the labour market as a whole operates in practice. Therefore,
inexperienced economics students should make the effort to work through the graphs. Students will
encounter similar analyses in many other economics courses and may as well learn of these with
regard to the labour market where basic concepts such as the trade-off between leisure time and
work may very well assume personal application!
Questions and study suggestions for students. At the end of each chapter the book contains a
number of questions and study suggestions for students. These cover the most important parts of the
chapter. Only by understanding the dynamics of the labour market can further developments be
anticipated and their implications grasped.
Quotations. The sometimes inspiring, sometimes light-hearted and sometimes tongue-in-the-cheek

quotations in the text have been drawn from various sources, but mainly from Auden and
Kronenberger ( ), Braude ( ), James ( ), Peter ( ), Prochnow and Prochnow ( ), as well
as Rawson and Miner ( ).
Preface to the sixth edition
The labour market and labour policies are in constant flux, with the result that fairly substantial
changes have been made to ensure that this revised edition is positioned squarely in the middle of
labour debates currently raging in South Africa.
Two chapters are devoted to labour supply and labour demand theories. The mathematical concepts
behind these theories are explained in thorough detail in this edition.
A chapter is especially dedicated to collective bargaining and the impact of unions on the labour
market. It also covers strike action in South Africa and looks at the controversy surrounding
bargaining councils.
Labour market flexibility has been an important and controversial focal point and policy development
discussion within civil society. This forms part of the broader debate on labour market regulation and
ways to increase employment and productivity.
Formal employment is said to have increased substantially since the advent of democracy. The book
considers the evidence and identifies some urgent discussion points about the sensitivity of
employment to economic growth.
Unemployment remains persistently high and shows an upward trend in South Africa. A chapter
specifically focuses on the characteristics of the unemployed, main causes of unemployment and
labour market policy to boost employment.
The book has three appendices that deal extensively with the impact of globalisation on the labour
market, how other countries have managed the challenges of globalisation, consensus-seeking
institutions such as NEDLAC, as well as the South African labour market trends in – .
Acknowledgements
Professor Luther Backer, formerly from the Department of Human Resource Management at the
University of Johannesburg, provided the inspiration for the first edition of the book. I am particularly
grateful to him.
In previous editions I expressed my thanks to a number of people for their many useful comments,
their guidance and suggestions for the first edition, without which the fifth edition would not have
been possible. I want to express my appreciation to the various organisations that have assisted me
with data and information, specifically the Bureau of Market Research at Unisa (Prof. Carel van Aardt),
the National Productivity Institute (Dr Jan de Jager) and the Department of Labour. A special word of
gratitude to the personnel of Van Schaik Publishers, who provided a very professional and
enthusiastic service.
To my soulmate, Hanneli, a very warm word of thanks not only for her support, but also for her active
interest and suggestions. In the past she suggested that a full chapter be devoted to globalisation,
and this time round it was her suggestion to add the “Deeper insight” sections.
Finally, my gratitude to Our Father, who assists with most things.
As always, comments and suggestions will be valued and carefully considered.
Frans Barker
Pretoria
October
I would like to thank Van Schaik Publishers for giving me the opportunity to be involved in the sixth
edition of this book. It is a wonderful learning experience for me to improve my Labour Economics
knowledge. I am also grateful for the support of the colleagues from the publishers and the
contributions by Pietman Roos.
The great help received from Veliswa Dywili and Mmabatho MacKay of Productivity SA in providing
me with the latest statistics on productivity (for Chapter ) is much appreciated.
Derek Yu
Cape Town
November
In memory of Frans Barker –

The National Planning Commission is liberal around their view that employment and education drive
high levels of poverty and inequality, which are in turn our biggest challenges in South Africa and
ultimately inhibit the economic growth that is so needed. They are also implicitly realistic about the
challenges created because of a highly segmented labour market.
We will not be able to alter South Africa’s growth trajectory if we don’t appreciate and aim to
understand the complexities involved in our country, and the labour market is just one example. This
book seeks to address some of these complexities and be but one voice in an ongoing debate around
these issues. We need more voices and more calls to action. The life lived by Frans Barker has inspired
those around him to raise their voices with gravitas, and we trust that the legacy of his work will do
that as well.
Paul Barker
About the authors
The late Dr Frans Barker was Chief Operating Officer of the Chamber of Mines. Dr Barker was initially
appointed at the Chamber of Mines as Deputy Labour Relations Adviser in . Prior to that, he was
president of the Industrial Relations Association of South Africa and chairperson of the National
Manpower Commission, which preceded the National Economic Development and Labour Council.
During his career, he was a deputy president of Business Unity South Africa, vice-president of the
Economic Society of South Africa, a commissioner for the Commission for Employment Equity and
was also involved in NEDLAC in various roles.
An economist by profession, Dr Barker lectured at a number of universities and was the author of
several publications related to labour issues.
Susan Shabangu, Minister of Mineral Resources at the time of his death in August , described Dr
Barker as
a dedicated colleague, eager to get to a solution yet very calm during hard times. His
contribution to the mining industry will forever continue to guide us as the sector again faces
challenges. His expertise contributed immensely in moving us forward and for that we will sorely
miss him.
Derek Yu is an Associate Professor at the Department of Economics at the University of the Western
Cape. He has a decade of teaching experience in undergraduate and postgraduate Labour Economics,
and has published comprehensively in this area. He is also the author of the first edition of Basic
mathematics for economics students: theory and applications.
Pietman Roos has a decade’s experience in different civil society organisations including national
government, news media and organised business. He has worked on economic policy formulation,
commentary, negotiation and advocacy, and has lectured undergraduate economics and
jurisprudence.
List of abbreviations and acronyms
ABET Adult basic education and training

AIDS Acquired immunodeficiency syndrome
ALMP Active labour market policy
ASGI-SA Accelerated and Shared Growth Initiative for South Africa
BBBEE Broad-Based Black Economic Empowerment
BCEA Basic Conditions of Employment Act
BUSA Business Unity South Africa
CCMA Commission for Conciliation, Mediation and Arbitration
CDE Centre for Development and Enterprise
CEE Commission for Employment Equity
COSATU Congress of South African Trade Unions
CS Community Survey
EAP Economically active population
ECC Employment Conditions Commission
EPWP Expanded Public Works Programme
ERP Effective rate of protection
ETIB Employment Tax Incentives Bill
ETQA Education and Training Quality Assurer
FDI Foreign direct investment
FEDUSA Federation of Unions of South Africa
GDP Gross domestic product
GEAR Growth, Employment and Redistribution Programme
GHS General Household Survey
HIV Human immunodeficiency virus
HSRC Human Sciences Research Council
ILO International Labour Organization
IMF International Monetary Fund
JIPSA Joint Initiative on Priority Skills Acquisition
LFPR Labour Force Participation Rate
LFS Labour Force Survey
LRA Labour Relations Act
MLC Millennium Labour Council
MLL Minimum living level
MRP Marginal revenue product
NACTU National Council of Trade Unions
NALEDI National Labour and Economic Development Institute
NDP National Development Plan
NEDLAC National Economic Development and Labour Council
NEPAD New Partnership for African Development
NGP New Growth Path
NIDS National Income Dynamics Study
NMC National Manpower Commission
NPI National Productivity Institute
NQF National Qualifications Framework
NSA National Skills Authority
NSB National Standards Body
NSDS National Skills Development Strategy
NSF National Skills Fund
OECD Organization for Economic Cooperation and Development
OHS October Household Survey
QES Quarterly Employment Statistics
QLFS Quarterly Labour Force Survey
PLMP Passive labour market policy
PSCBC The Public Service Co-ordinating Bargaining Council
PWP Public Works Programme
RDP Reconstruction and Development Programme
SALDRU Southern Africa Labour and Development Research Unit
SALC South African Labour Commission
SAQA South African Qualifications Authority
SARB South African Reserve Bank
SEE Survey of Employment and Earnings
SETA Sector Education and Training Authority
SGB Standards Generating Body
SLL Supplemented living level
SMMEs Small, medium and micro enterprises
STD Sexually transmitted disease
TFR Total fertility rate
UIF Unemployment Insurance Fund
VMP Value of marginal product
Useful websites
Business Unity South Africa http://www.busa.org.za

Bureau of Market Research, Unisa http://www.unisa.ac.za/bmr
Chamber of Mines of South Africa http://www.chamberofmines.org.za
Commission for Conciliation, Mediation and Arbitration (CCMA) http://www.ccma.org.za
COSATU (trade union federation) http://www.cosatu.org.za
Department of Basic Education http://www.education.gov.za
Department of Labour http://www.labour.gov.za
Economic Society of South Africa http://www.essa.org.za
FEDUSA (trade union federation) http://www.fedusa.org.za
Human Sciences Research Council http://www.hsrc.ac.za
Industrial Relations Association of South Africa http://www.irasa.org.za
International Labour Organization http://www.ilo.org
Institute for the Study of Labor http://www.iza.org
NEDLAC http://www.nedlac.org.za
New Partnership for Africa’s Development http://www.nepad.org
Policy, Law, Economics and Politics http://www.polity.org.za
Productivity SA http://www.productivitysa.co.za
Seifsa (employer organisation) http://www.seifsa.co.za
South African Government http://www.gov.za
South African Labour Guide http://www.labourguide.co.za
South African Qualifications Authority http://www.saqa.org.za
South African Reserve Bank http://www.resbank.co.za
Southern Africa Labour & Development Research Unit http://www.saldru.uct.ac.za
Statistics South Africa http://www.statssa.gov.za
World Bank http://www.worldbank.org
World Economic Forum http://www.weforum.org
Glossary of important terms
active labour market policies

Policies that aim to improve the operation and results of the labour market so as to maximise quality
employment and minimise unemployment and underemployment, and while so doing improve
efficiency, equity, growth and social justice (section . . ).
affirmative action
Legislation, policies and programmes aimed at redressing social, economic or educational imbalances
arising out of past discriminatory laws or practices (section . ).
atypical or “non-standard” employment

Employment that is not full-time (part-time work) or is of limited duration (temporary). It is often
employment at a workplace other than that of the employer (section . . ).
bargaining council
A collective bargaining institution established in terms of the Labour Relations Act on a voluntary
basis by registered employer and union organisations in a specific industry to negotiate conditions of
employment and resolve disputes (section . ).
capital intensive
Any production process requiring a higher proportion of capital relative to other production factors
(such as labour) per unit of output (section . . ).
capital productivity
The number of units of output that are produced per unit of capital input (section . . ).
corporatism
An institutional framework that incorporates to varying degrees the pinnacles of employer and
employee organisations in the economic and social decision making of society (Appendix B. ).
discrimination
The inferior treatment of some workers with respect to hiring, promotion, wages, etc. owing to factors
not related to ability, seniority, skills level, experience or other labour market factors (section . ).
dual labour market

A labour market divided into two non-competing markets, i.e. a primary and secondary labour market,
with the primary market characterised by the choice jobs, high earnings and security of tenure. The
secondary market is characterised by low-wage, unstable, dead-end jobs; workers in this market have
little hope of acquiring a job in the primary market (section . . ).
economically active population
Total number of working-age population aged – years who furnish their labour for the production
of economic goods and services, whether employed or not; this includes workers formally or
informally employed, the self-employed, employers, and the unemployed wishing to work (section
. ). See also labour force.
efficiency wage
A wage increase that results in a productivity increase at least equivalent to the increase in wages,
which makes it advantageous for the employer to pay higher rather than lower wages (section . . ).
elasticity
A measure of the degree to which one variable changes in response to changes in another variable.
The wage elasticity of the demand for labour, for instance, is the degree of responsiveness of the
quantity of labour demanded to changes in the wage rate (section . . ).
employment
All persons years and older who during a specified brief period (for example seven days) have
worked for one or more hours for a wage or salary or for profit or family gain, in cash or in kind. The
self-employed are included, as are persons who have been temporarily absent from work but still
have a formal job attachment (Chapter ).
employment coefficient
The extent to which employment will change as a result of changes in economic growth (section
. . ).
human capital
The skills, knowledge and other acquired characteristics (usually through education and training) of
workers that make them more productive (section . ).
index number; index

An index number is a single number which gives the value of that variable as a percentage of the
value of that variable at some base period. For example, if is the base period (i.e. = ) and
the index number of is , it means that the value of the variable concerned is % of the value
in , i.e. an increase of % over the period. An index is a time series of index numbers.
informal sector
Unorganised, unregulated and mostly legal but unregistered economic activities that are individually
or family owned and use simple, labour-intensive technology (section . ).
internal labour market

An organisational unit where the pricing and allocation of labour are governed by a set of
administrative rules rather than by market forces. Jobs within the internal market are usually filled by
the promotion of workers who are already within the internal labour market and these jobs are thus
shielded from normal competitive forces. Entry is normally from the bottom only (section . . ).
labour absorption capacity
The percentage of new entrants to the labour market who find a job in the formal sector of the
economy (section . . ).
labour force
Synonymous with the economically active population (EAP – section . ), although the ILO states
( : ) that the EAP can be divided into the “usually” active and the “currently” active population,
and that the labour force is in fact equal to the currently active population. The difference is important
when there are a significant number of seasonal workers.
labour force participation rate

The percentage of the working-age population furnishing their labour for the production of economic
goods and services, whether employed or not (section . ).
labour intensity
Any production process requiring a higher proportion of labour relative to other production factors
(such as capital) per unit of output (section . . ).
labour market flexibility

The extent to which an employer can alter various aspects of the enterprise’s work and workforce to
meet the demands of the enterprise, for example the size of the workforce, the content of jobs,
working time, etc. (section . ).
labour productivity
The number of units of output that are produced per unit of labour input (section . . ).
law of diminishing marginal product

As successive units of one input (e.g. labour) are added to a fixed amount of another input (capital), a
level of total production is reached beyond which the marginal product of that variable input declines
(section . . ).
marginal revenue product

The change in total revenue resulting from the employment of an additional worker (section . . ) in
the case of monopoly.
minimum living level

The minimum financial requirements of members of a family or household if they are to maintain their
health, have acceptable standards of hygiene and sufficient clothing for their needs (section . . ).
money wages
See nominal wages.
monopoly
A situation where a single seller controls the entire output of a particular product or service. The
enterprise can thus set its own price and production level, subject to demand conditions (section
. . ).
monopsony
A situation where there is only one buyer in a market, for example when only one enterprise is the
buyer of labour in a particular labour market (section . ).
multifactor productivity
The number of units of output that are produced by the combined input of both labour and capital
(section . . ).
nominal wages
Wages measured in current prices. An increase in nominal wages will not lead to any increase in
purchasing power if prices increase at the same rate as nominal wages.
productivity
The relationship between the quantity of output (e.g. the quantity of goods and services produced)
and the quantity of input (e.g. capital and labour) used to produce that output. It is therefore a
measure of input efficiency (section . ). More recent definitions also take account of quality and
customer satisfaction.
real wages
Wages after having removed the effect of price increases (inflation). An increase in real wages over
time would therefore indicate an actual improvement in purchasing power.
scale effect
The change in employment and output resulting from the effect of the wage change on the
employer’s costs of production (section . . ).
segmented labour market

A labour market divided into a number of segments, each with its own characteristics and mode of
operation. There is often little or limited mobility or competition between the segments (section
. . ).
social wage
The social wage is the total sum of all social benefits received free or partially free, such as state
transfers, subsidies and services (unemployment insurance, state old-age pensions, free public health
services, etc.) and kinship and community transfers (section . . ).
substitution effect
The change in employment when both labour and capital are variable, and a change in the relative
price of labour results in a substitution of one resource (e.g. labour) for the other (e.g. capital) without
a change in production (section . . ).
total fertility rate

The average number of children born alive to a woman in her reproductive years ( – years of age)
(section . ).
underemployment
An unemployed person who either involuntarily works fewer than normal hours or has his or her skills
underutilised (section . . ).
unemployment
All persons in the labour force who are without work (not in employment, as defined above), are
currently available for work, and are seeking work or wanting to work (section . . ).
unemployment rate
The number of unemployed persons as a percentage of the total economically active population
(section . . ).
unit labour cost

The cost of labour to produce one unit of output, which is obtained by dividing the average
remuneration per employee by labour productivity (section . ).
value of marginal product

The change in total revenue resulting from the employment of an additional worker (section . . ).
working-age population
The total number of people aged – years (section . ).
Contents
Preface
Acknowledgements
About the authors
List of abbreviations and acronyms
Useful websites
Glossary of important terms
Chapter Introduction
. Why study the labour market?
. Unique characteristics of the labour market
. Why a theory with unrealistic assumptions?
. . Perfectly competitive labour market
. . Graphic analysis of the labour market
. Important characteristics of the labour market in South Africa
. Structure of the book
Chapter The supply of labour

. Introduction
. The labour force as the supply of labour
. . The South African labour force
. . Labour force by gender
. . Labour force by population group
. . Labour force by age
. . Labour force by education
. Labour force participation rate
. . LFPR by gender
. . LFPR by population group
. . LFPR by age
. . LFPR by education
. Population and population growth
. . Population size
. . Population growth rates
. . Population by age
. . Factors determining population growth rates
. Total fertility rate
. . Factors influencing fertility rates
. The impact of HIV and AIDS
. . Incidence and causes of HIV and AIDS
. . Consequences of AIDS
. Immigration, emigration and migrant labour
. . Immigration and emigration of skilled people
. . Migrant labour within South and southern Africa
. Deeper insight: from the individual decision to work to the market supply of labour
. . Basic concepts and important notations
. . Worker preferences: indifference curves
. . Budget constraint
. . Utility maximisation
. . Income effect with a non-labour income increase
. . Substitution and income effects with a wage increase
. . Reservation wage
. . The individual labour supply curve
. . The elasticity of labour supply
. . The market supply of labour
. . Factors affecting market labour supply
Chapter The demand for labour

. Introduction
. . The measurement of employment
. Employment in South Africa
. . Total employment over time
. . Employment by population group, gender, age and education
. . Employment in the formal and informal sectors
. . Employment by occupation and industry
. . Demand for skills
. . Atypical employment and underemployment
. . Labour absorption capacity
. . Employment elasticity of economic growth
. . Capital and labour intensity
. The informal sector
. . Definition and measurement of the informal sector
. . Importance and size of the informal sector
. Deeper insight: deriving the demand for labour
. . The production function
. . The short-run demand for labour
. . The long-run demand for labour
. . The elasticity of labour demand
. . The elasticity of substitution
. . Marshall’s rules of derived demand
. . The market labour demand curve
. . Factors influencing market labour demand
. . Imperfectly competitive markets: the monopolist
Chapter Wages and the cost of labour
. Introduction
. The determination of wages
. . The determination of wages in a perfectly competitive labour market
. . The determination of wages in a monopsony labour market
. . The determination of wages in a segmented labour market
. . Determination of wages through collective bargaining
. . Government involvement and influence
. . Communes and communism
. . Other methods of wage determination
. Earnings in South Africa
. . Average salaries and wages in South Africa
. . Non-wage labour costs
. . The income of graduates
. . Share of employee remuneration in national income
. Wages and inflation
. Wage differentials
. The trade-off between wages and employment
. . The impact of wage increases on employment
. . The impact of high unemployment on wages
. . Will high wages increase economic growth?
. The impact of labour laws on the cost of labour
. . Basic Conditions of Employment Act
. . Labour Relations Act
. . Other laws
. Deeper insight I: determination of wages in a perfectly competitive labour market
. Deeper insight II: determination of wages in a monopsony market
Chapter Unions, collective bargaining and minimum wages

. Introduction
. Collective bargaining and the labour market
. . The right to bargain collectively
. . Factors that determine the outcome of collective bargaining
. The union movement
. . Reasons for joining a trade union
. . Types of trade union
. . Factors that have an impact on the strength and influence of unions
. . The union movement internationally and in South Africa
. The impact of unions
. . The impact of unions on wages and the cost of labour
. . The impact of unions on flexibility and productivity
. . The impact of unions on equality objectives
. The influence of strike action
. . The Hicks strike models
. . The influence of industrial action in South Africa
. Bargaining councils
. . Background to bargaining councils
. . Importance of bargaining councils
. . Economic analysis of the impact of bargaining councils
. . Extension of agreements
. . Making provision for special circumstances of employers
. . The link between wages and productivity
. Minimum wages
. . Arguments in favour of minimum wages
. . Arguments against minimum wages
. . Minimum wage-setting mechanisms in South Africa
. . Minimum subsistence levels
Chapter Productivity and labour market flexibility

. Introduction
. The importance and measurement of productivity
. . The importance of productivity
. . Definitions and measurement of productivity
. . Shortcomings of productivity data
. . Productivity and employment
. . Some factors influencing productivity
. Productivity in South Africa
. Considering productivity and wages together: unit labour cost
. . The concept of unit labour cost
. . Unit labour cost in various sectors
. . Impact of wage increases on productivity: efficiency wages
. The importance and types of labour market flexibility
. . Flexibility in contracts of employment: job security and hiring decisions
. . Flexibility in contracts of employment: atypical employment
. . Wage flexibility
. . Working time flexibility
. . Work organisation flexibility
. . A skilled workforce
. General considerations regarding flexibility
. . Achieving flexibility through collective bargaining
. . Labour market flexibility compared internationally
. . Balancing the social costs and economic benefits of labour market flexibility
Chapter Unemployment in South Africa

. Introduction
. Definition, types and measurement of unemployment
. . Definition of unemployment
. . Types of unemployment
. . Measurement of unemployment
. Unemployment in South Africa
. . Total unemployment
. . Unemployment by demographic and education characteristics
. . Unemployment rates by demographic and education characteristics
. . Other characteristics of the unemployed
. Reasons for high unemployment
. A strategy to create jobs
. . Economic growth and exports
. . Allowing higher inflation in order to create jobs: the Phillips curve
. . Improving the labour absorption capacity of the economy
. . The cost of labour
. . Labour market flexibility and productivity
. . Active labour market policies
. . A strong skills base
. . Formal and informal small-enterprise sector
. Special employment and income maintenance programmes
. . Public works and special employment programmes
. . Programmes focused on young people
. . Wage subsidies
. . The social plan
. . Unemployment insurance
. . Basic income grant
Chapter Human capital and the demand for skilled workers

. Introduction
. Human capital theory
. Some criticisms of human capital theory
. . Measurement
. . Ability or educational level?
. . The screening hypothesis
. . Family background and financial status
. . Dual labour market theory
. . The radical approach
. Human capital in South Africa
. . Educational level of the labour force
. . Occupational structure of the labour force
. . Skills levels of workers in formal employment
. Education in South Africa
. . Educational shortcomings in South Africa
. . Starting to address the educational shortcomings
. Skills development in South Africa
. . Skills development system and structures
. . Joint Initiative on Priority Skills Acquisition
. . The extent of in-service training activities in South Africa
. Skilled labour shortages
. . Skills shortages in South Africa
. . Projection of the future demand for, and supply of, labour by skills category
Chapter Labour market inequalities and discrimination

. Introduction
. Definitions and types of discrimination
. Theories of discrimination
. . Taste-for-discrimination model
. . Statistical discrimination
. . Market imperfections
. . The crowding model
. . The transaction costs model
. . Legislation and government policy
. Inequalities in the South African labour market
. . Educational level
. . Employment and unemployment inequalities
. . Occupational inequalities
. . Income and wage inequalities
. Addressing inequalities
. . Higher economic growth and employment opportunities
. . Education and training
. . Entrepreneurship
. . Government employment and tender practices
. . Legal remedies
. Equal pay for work of equal value
. Affirmative action
. . The case for affirmative action
. . The case against affirmative action
. . Affirmative action in the Employment Equity Act
. . Quotas versus Employment Equity targets: the BBBEE Codes
Appendices
Appendix A Globalisation and the labour market
Appendix B Social dialogue and codetermination
Appendix C South Africa’s labour market trends, –
Bibliography
Index
Introduction
Outline of this chapter

. Why study the labour market?
. Unique characteristics of the labour market
. Why a theory with unrealistic assumptions?
. Important characteristics of the labour market in South Africa
. Structure of the book
The worker sells his work, but he himself remains his own property.
A M
. WHY STUDY THE LABOUR MARKET?
Any reasonably serious observer of South African society is constantly bombarded by policy questions
that have their origin in the labour market. These policy questions may, for example, be raised in the
newspapers, and discussed in union meetings or at company board level. Or companies may have to
decide about their reaction to some of these issues, and to what extent the company needs to adjust
to face the possible long-term consequences of some labour policy issues. As is apparent from the
examples mentioned below, such questions may assume various guises, but they are nevertheless
often essentially rooted in the labour market.
Most individuals are in some way affected by various economic and social problems being
experienced in the country, such as poverty, inequality, undernutrition, crime as well as physical and
social isolation. Some people say unemployment is the root cause of these problems. But is
unemployment actually that serious, and what can be done about this problem? Should the country
accept fewer immigrants because they appear to be taking the jobs of local workers, provide a
transport subsidy to the unemployed to encourage them to seek work more actively, or launch a
national minimum wage policy to ensure the employed do not end up as working poor despite
having a job?
In pursuit of its policy to correct past imbalances, the government published the draft Broad-Based
Black Economic Empowerment (BBBEE) Codes for comment in March , and in January the
BBBEE Act of was assented to. The primary aim of this act is economic empowerment of all
black people (including women, the youth, people with disabilities and those living in rural areas) by,
among others, achieving an equitable representation in all levels and occupational categories in the
workforce as well as increasing the proportion of black ownership and management control. But does
the country have sufficient skilled black people to appoint? What about the white people currently in
these positions? Has affirmative action worked in other countries?
There have been a number of strikes about wages, and employers complain that increased wages
have a negative impact on employment and their competitiveness. To what extent do unions affect
wage levels? Do sharply increased wages always have a significant impact on employment? What if
the wage increase is complemented by a productivity increase?
Minimum wages have been introduced by means of sectoral determination in numerous relatively
more vulnerable, lowly-paid sectors such as wholesale and retail, contract cleaning as well as for farm
and domestic workers. Is this not interference in the operation of the market? Will it not lead to a
sharp increase in unemployment (for example, unskilled workers whose minimum wages make them
too expensive for prospective employers)?
A number of small businesses have taken the Minister of Labour to court, arguing that the imposition
of bargaining council agreements has an unfair impact on their right to conduct business. What is the
impact of bargaining councils? Do they fairly protect workers against exploitation or do they unfairly
damage small-business interests?
Many of these questions do not relate to facts only (e.g. what is the impact of higher wages on
employment?), but also to normative questions (e.g. what should the government do about the very
low wages in the agricultural sector?). This requires value judgments from those involved, and many
such questions cannot simply be answered by looking at “the facts”. However, to make a good value
judgment, one needs as much information about the facts as possible in order to make an informed
decision. Addressing such normative questions often involves trade-offs, and the facts help to
crystallise the various trade-offs.
Positive economics versus normative economics
In economics, positive questions address the relatively narrow “What is?” while normative questions look at the much
broader “What should be?” questions. Focusing on labour economics, examples of positive questions include “What is
the total number of unemployed in ?”, “What is the youth unemployment rate in ?” and “What is the impact of
the imposition of a minimum hourly wage of R on employment in the hospitality sector?” In contrast, examples of
normative questions are as follows: “Should the government launch a job search subsidy programme to finance the job-
seeking transport cost of the unemployed?”, “Should the eligible age of the youth wage subsidy be extended to
years?” and “Should the small firms in the construction industry with fewer than workers be granted exemption from
the minimum wage determined at collective bargaining?”
A study of the issues in this book will thus greatly assist the reader to acquire an informed opinion
about these issues. It matters not whether the reader is a policy maker, a current or future leader of
business or labour, or simply a citizen of the country. South Africa is a democratic country, with a
custom of wide involvement and interest in policy decisions. Citizens need to be involved in the
formulation of policies and to be responsible voters with knowledge of what the long-term effects of
policy decisions are likely to be.
Too often policy makers are inclined to say that they did not foresee the “unintended consequences”
of some or other policy decision, and that the decision thus needs to be revised (after the damage has
already been done). When it comes to labour matters, there is very little reason why the possible
consequences could not have been foreseen or intended. A proper study of the operation of labour
markets will help to eliminate the need for such revisions of policies only a few years after the policies
have been introduced.
. UNIQUE CHARACTERISTICS OF THE LABOUR MARKET
The labour market is an imaginary marketplace where labour is bought and sold. Although the
labour market is characterised by the same principles of supply and demand as the markets for other
goods, it is quite different from such markets.
The first difference is that the worker is not a product but a person, a human being. The principles of
fairness or equity and humaneness are therefore intrinsic elements of the labour market. Fairness is
not always ensured in the common law of freedom to contract between an employer and a worker,
because of unequal bargaining power. The employer is usually in a much stronger bargaining position
than the individual worker (especially with high unemployment rates such as in South Africa) and for
this reason legislation acknowledges the right of workers to form unions in order to bargain on a
fairer level with employers. This is discussed in section . .
The purchaser of labour (the employer) buys the services of the worker, but the employer does not
buy the worker. However, the services and the worker cannot be physically separated. A contractual
relationship, which differs from what occurs in the product market, therefore comes into being
between the buyer and the seller of labour. Such a relationship is usually an enduring (but not always
an endearing) one, and continues until the relationship is ended by either the buyer or the seller.
Legislation aims to encourage a long-term relationship, for example through measures that prevent
unfair dismissals or discourage hasty retrenchments.
Most products are standardised to a greater or lesser extent. However, the “products” (workers) in the
labour market are not at all standardised and the market is characterised by great diversity or
multiplicity. Different skills make up only one element that causes diversity.
Personality characteristics are also important elements of the product that the employer “buys”. These
cannot always be determined fully before employment starts, and the employer thus does not
necessarily get what he or she intended to buy. Relations with the employer or emotional disposition
may have an impact on the output (the productivity) of the worker.
There are usually also changes over time in respect of any individual worker. His or her skills level may
improve with training and experience as well as further education. Alternatively they may deteriorate
if the worker is demotivated, if his or her skills become obsolete, or when he or she grows older. In
contrast, the products in the goods market depreciate over time in general.
There is also not one single market or clearing house for labour as there is in some product markets
(e.g. the stock exchange). There is a multiplicity of markets in terms of different occupations,
geographical areas and types of economic activity.
A further characteristic is the complexity of the price of labour. This is not usually the case in the
product market. In the labour market the price is determined not only by the employment package
(which includes various fringe benefits), but also by the influence of inflation, personal taxation, the
standard of living of the worker, the provision for health and safety, and other factors.
Labour markets have important functions:
They allocate human resources among alternative users, i.e. among sectors, enterprises, locations
and occupations.
They distribute incomes, either wages or salaries, as incentives and as rewards to workers
(International Labour Organization (ILO, : ).
Labour markets should, by their functioning, contribute to the following objectives:
Efficiency. This means maximum output and maximum income, as well as an improvement in
matching between those seeking work (labour force) and those seeking labour (employers).
Equity. This implies equality of opportunity for all in access to jobs, training, treatment at work and
payment.
Growth. Labour market operations should increase employment in the future and contribute to –
not impede – higher productivity and incomes.
Social justice. This refers to the extent to which society acts to minimise any negative effects the
labour markets may have on workers’ welfare, and to redress any harm that might have been done.
. WHY A THEORY WITH UNREALISTIC ASSUMPTIONS?
Throughout this book, various assumptions will be made about the labour market to illustrate a
particular point. For instance, we may assume that there are no unions and no government
determination of minimum wages to determine the impact of an increase in the number of women
entering the labour market. Although at times these assumptions may appear to be rather unrealistic,
they help us to understand the key issue under discussion better. Although a very complex analysis
may more realistically mirror the labour market, it would then not isolate what really matters (see
Borjas, : ). Furthermore, it is also better to start relatively simply, and then gradually to become
more realistic, to determine the theoretical impact of a specific policy measure or decision.
. . Perfectly competitive labour market

One set of basic assumptions that will often be referred to in this book is that the labour market is a
perfectly competitive market. A perfectly competitive labour market is a theoretical norm against which
the performance of the actual labour market can be measured. It falls within the school of neoclassical
theory. The perfectly competitive labour market is characterised by the following assumptions
(Marshall, King & Briggs, : ; McConnell, Brue & Macpherson, : ):
There is full and perfect knowledge of the market. Workers have full information on jobs available
and wage rates, while employers have full information on wage rates paid by other employers.
Workers and employers are entirely rational. A decision to work or not to work is taken only on the
basis of wage rates, while employers always attempt to maximise their profits.
Neither workers nor enterprises exercise any influence over the market wage. An individual employer
represents a very small proportion of the demand for labour and does not act in concert with other
employers in determining wage levels. Similarly, labour is not organised and each worker makes his
or her own decisions on employment or wages.
There is perfect competition. Many enterprises compete against one another to hire workers and
workers have identical and comparable skills.
Workers are perfectly mobile. They can move easily between positions within or between
organisations or regions.
Taking a closer look at the assumptions of the perfectly competitive labour market
It is expected that not all of these assumptions would hold in the real world. For example:
If a job hunter only reads the Cape Argus but not the Sunday Times to seek work, he or she will not have full and
perfect knowledge of available wage rates.
Some firms in a sector may not fully know the salary levels offered by other firms in the sector (assuming sectoral
determination and bargaining councils do not exist in this sector).
Some workers decide to work at a place due to factors not related to wages, such as loyalty to the firm and good
relationships with colleagues.
Not all employers aim to maximise profits, such as non-profit organisations.
Workers or enterprises may exercise an influence over the market wage. One example is the monopsony market,
which is discussed in section . .
Workers from the same sector could be organised by becoming members of the same trade union (discussed in
Chapter ) to exert strong influence on minimum wage of the sector.
Some employers may attract, treat and reward workers differently or even unfairly, to avoid competing against other
enterprises to hire, retain and retrench workers.
Workers could have different and incomparable skills, for instance, two workers both have a Bachelor’s degree and
work in the same sector, but if one of them pursues postgraduate studies on a part-time basis, they would no longer
have identical skills, at least in terms of educational attainment.
Some workers may not be perfectly mobile, due to reasons like poverty and family responsibility.
Throughout this book, these assumptions would be relaxed, especially when discussing the more advanced labour
market theories.
I don’t meet competition, I crush it.

C R
. . Graphic analysis of the labour market

Throughout the book, graphs are used to illustrate various concepts. Although different graphs are
used, many have their basis in the basic supply-and-demand graph, or may explain how this supply-
and-demand graph is deduced from different decisions in the labour market.
The supply-and-demand graph simply indicates the basic decisions regarding the labour market of
the major actors in the market, i.e. workers and employers. Workers present their labour (the supply of
labour) and employers employ workers (the demand for labour). Although there are many factors
determining the supply and demand of labour, the most important is the wage rate.
A basic graph of the labour market plots the number of workers “supplied” and “demanded” against
wages, as in Figure . .
FIGURE . Equilibrium between supply and demand
You can make even a parrot into a learned political economist – all he must learn are the two words
“supply” and “demand”.
A
The horizontal axis is called the x-axis. This denotes the quantity of labour, i.e. the number of workers.
The vertical axis is called the y-axis and in this case denotes the price (the wage rate). A line or curve,
in this case S, which represents the market supply of labour, indicates the relationship between
quantity and price. As will be explained in this book, but can also be understood intuitively, the higher
the wage rate (for instance in a particular occupation or region) the higher the quantity supplied of
labour (or number of workers supplied). This means that the supply curve is upward sloping to the
right.
The market demand for labour is indicated by the curve D. This curve is downward sloping, because
as the wage rate increases, employers will try to reduce their cost by employing less labour in various
ways, for instance by using machinery instead of workers. In a perfectly competitive labour market,
there would be equilibrium where the quantity supplied of labour is equal to the quantity demanded
for labour, and this would also be the number of workers employed. This is at point e in the graph.
The number of workers employed (E) and the wage rate (W) can be read off the axes.
There can be the following changes in respect of these curves:
(a) Movement along the curves. If there is a change in the wage rate, there will be a movement along
the curves. For example, an increase in wages from W to W along the curve S will generate an
increase in the quantity supplied of workers from E to E . In the case of the demand for labour,
such an increase in wages will decrease the quantity demanded from E to E along curve D. There
will thus be an oversupply of workers who want to work – the quantity supplied is E , whereas the
quantity demanded is only E , that is, there is an excess supply of (E –E ) at W . Workers will then
compete for the available jobs and accept lower wages in order to get work. Wages will then start
declining, but when this happens, employers will employ more workers, until equilibrium is
achieved at point e.
(b) Shift in the curves. The supply curve as a whole can shift to the right or left if influenced by
external (exogenous) factors, for example more women wanting to work or large immigration
gains. An increase in the supply of labour will shift curve S as a whole to the right, which means
that more workers will be available for work at the wage level W. There can be similar shifts in the
demand curve, for example if exports suddenly increase, the demand for labour will increase and
the demand curve will shift to the right. The shift of the curves is discussed in greater detail in
Chapters – .
(c) Change in gradient. The gradient of the curves can also change, depending on how responsive
labour supply or demand is to a wage rate change (this responsiveness is called elasticity). If
demand for labour is very responsive to wage rate changes, a wage increase will result in a sharp
drop in the quantity demanded for workers. In this case, the demand curve will be relatively flat. If
the demand is less responsive to wage rate changes, the quantity demanded for labour will
change less, and the curve will be relatively steep. This is illustrated in Chapter .
. IMPORTANT CHARACTERISTICS OF THE LABOUR MARKET IN SOUTH AFRICA
The following important characteristics have been identified in the South African labour market:
A sharp increase in the supply of labour in – , which can be ascribed to a sharp increase in
the labour force participation rates of women and Africans (Casale & Posel, ; Van der
Westhuizen, Goga & Oosthuizen, ; Yu, ).
Low increases in the demand for unskilled and semi-skilled workers, with restricted growth in sectors
that traditionally employed large numbers of unskilled and semi-skilled workers (such as agriculture
and mining), and across all sectors, an inclination among employers to employ higher-skilled
workers at the cost of semi-skilled or unskilled workers. In addition, the informal sector in South
Africa seems relatively small when compared to the informal sector in other developing countries.
The demand for labour has been most affected by within-sector developments (such as the
increasing preference of employers for higher-skilled employees). Between-sector developments
have also played a significant role, in particular the relatively higher growth of the tertiary sectors
(e.g. services), which has resulted in a higher demand for skilled labour, whereas the primary
sectors, which employ most of the production workers, experienced a relative decline. The losers
have been unskilled labourers, specifically unskilled African workers. The winners, invariably, have
been better-educated, skilled workers who in turn have been disproportionately white and Asian.
The impact of international trade on employment has been relatively benign, with only (semi-
skilled) production workers experiencing actual job losses between and (Bhorat, Van der
Westhuizen & Goga, ).
Unemployment, particularly among the black population and among females. Unemployed persons
also tend to be poorly educated and young, and those living in rural areas are the worst affected.
However, there has also been a disconcerting increase in unemployment among better-educated
individuals, whether with matric or having a tertiary education, and particularly Africans. By far the
majority of the unemployed persons have either never had a job at all, or might have been
unemployed for longer than three years. Such individuals would thus not qualify for unemployment
benefits, and they are likely to have lost any skills that they might have picked up during their
education. Apart from unemployment being a major problem for the country and the individual
concerned, it is also a major contributing factor to two of the country’s other major labour market
problems: inequality and poverty.
High labour costs and low productivity, combined with the general absence of inherent pressures to
increase the level and quality of output. The declining labour intensity of economic growth is
particularly disconcerting, because far fewer job opportunities than in the past are now being
created for every percentage point of economic growth. Where labour productivity has increased, it
has often been at the expense of jobs. Collective bargaining institutions (especially bargaining
councils) and unions have generally resulted in a wage premium for the relevant workers when
compared to workers not falling under a bargaining council or belonging to a union. Government
policy also seems to have contributed to the increased cost of labour and to a reluctance among
employers to employ more labour.
Labour market segmentation is substantial. While the currently employed (or insiders) in the formal
sector are protected by the minimum wage agreement, the views of the unemployed (or outsiders)
are not fully considered during collective bargaining (Nattrass, ). Hence, a lot of unemployed
fail to find employment in the formal sector, and either engage in informal-sector activities or
continue to be searching unemployed. Some of those in the latter group may eventually give up
hope of finding work and become non-searching unemployed (or discouraged work-seekers)
(Fourie, ).
Income inequality is extensive and continues to increase. Whereas some researchers have in the
past found that interracial inequality has diminished (particularly between blacks and whites) over
the past two decades, more recent studies show that it is in fact not the case. Even though high
unemployment among the black population in particular and the high upward mobility of many
Africans into senior positions in the public and private sectors have resulted in a high level of intra-
racial inequality, the high demand for highly skilled labour (consisting of predominantly white
workers) has resulted in interracial inequality remaining the most important contributing factor to
the high level of inequality in South Africa.
. STRUCTURE OF THE BOOK
The structure of the book is presented in Figure . .

FIGURE . Structure of the book
Chapter discusses the supply of labour, i.e. why people enter the labour market, the increased
participation of women, the impact of immigration and emigration and related concepts.
We then turn to a discussion of the demand for labour (Chapter ). The demand for labour is simply
the employment of labour, but over time the characteristics of the demand for labour have changed
somewhat. This is having an impact on, for instance, unemployment.
The interaction of supply and demand is discussed in Chapter , where an indication is given of how
wages are determined, as well as the impact of wages on employment, and the effect of government
policies on the cost of labour.
Unions have a significant impact on wage levels. However, unions can only do so in a collective
bargaining set-up. These concepts are discussed in Chapter . Minimum wages are often determined
by unions, but where there is insufficient collective bargaining, the government has a responsibility to
set minimum wages. The impact of such minimum wages is discussed in this chapter.
Chapter discusses very important related concepts. Wages cannot be discussed without discussing
productivity. Chapter thus discusses productivity and a closely related concept, the flexibility of the
labour market. Part of this chapter is also dedicated to a discussion of the mechanism for considering
wages and productivity together.
Various specific features of the labour market are discussed in the next three chapters, i.e.
unemployment (Chapter ), human capital, the quality of the supply of labour (Chapter ) and
inequalities and discrimination (Chapter ).
The outcome of all of these factors determines to what extent the country is internationally
competitive, which is the subject of Appendix A. Also discussed in this appendix are some case
studies of countries that have successfully addressed the challenges of globalisation.
This is followed by (in Appendix B) a discussion of social dialogue in South Africa, which has
characterised the labour market for the last -odd years. The book ends with Appendix C which
presents the statistics of the country’s labour market trends in – .
KEY CONCEPTS
contractual relationship
demand for labour
elasticity
equilibrium
fairness and equity
perfectly competitive labour market
supply of labour
FOR STUDENTS
. Explain the main differences between the labour market and the goods market.
. Discuss the assumptions of a perfectly competitive labour market, before explaining, with the aid of examples,
why not all of these assumptions may hold in the real world.
SUGGESTED READING*
Department of Labour. Annual. Annual Report. Pretoria: Government Printer.
Department of Labour. . Annual Labour Market Bulletin / . Pretoria: Department of Labour.
Edgren, G. . How Different is the South African Labour Market? International Perspectives and Parallels.
Pretoria: Human Sciences Research Council.
Fields, G.S. . Labor Market Analysis for Developing Countries. Labour Economics, ( ): December.
Fourie, F.C.V.N. . The South African Unemployment Debate: Three Worlds, Three Discourses? SALDRU Working
Paper Series Number . Cape Town: Southern African Labour and Development Research Unit (SALDRU).
National Economic Development and Labour Council (NEDLAC). Annual. Annual Report, submitted to the
NEDLAC summits. Johannesburg: NEDLAC.
* The bibliography contains the full list of references.

The supply of labour

. Introduction
. The labour force as the supply of labour
. Labour force participation rate
. Population and population growth
. Total fertility rate
. The impact of HIV and AIDS
. Immigration, emigration and migrant labour
. Deeper insight: from the individual decision to work to the market supply of labour
The power of population is infinitely greater than the power in the earth to produce subsistence for man. Population, when
unchecked, increases in a geometrical ratio. Subsistence only increases in an arithmetical ratio. A slight acquaintance
with numbers will show the immensity of the first power in comparison with the second.
T R M
. INTRODUCTION
An individual can decide to work or not to work; if he or she decides to work, he or she can decide to
work more hours or enjoy more leisure hours. This is the basis of the supply of labour. Various factors
influence these decisions, for instance: the personal preferences of the individual; the amount of
money he or she needs; the wage rate; non-labour income and so forth. If all the decisions of
individuals are added up, it gives the market supply of labour.
It does not matter whether the persons are working or simply want to work; they are all part of the
supply of labour. Furthermore, the type of work they do is irrelevant, for instance, whether they work
in the informal sector or are an employer of other people. All these people form part of the labour
force, or the supply of labour.
The labour force is thus simply the population of working-age people that is working (the employed)
or that wants to work (the unemployed). It is influenced by factors such as the proportion of the
population that wants to work, the size of the population (as determined by, among other things,
fertility and mortality rates) and net immigration.
Changes in the economic environment – especially wage rates and incomes – can account for a
significant part of the labour supply. This is an aspect that is discussed in this chapter – to indicate
how the labour supply of an individual and of all individuals together (the market supply) is affected
by the decision to work or not to work. Furthermore, numerous factors affect the market labour
supply, ranging from the aforementioned demographic factors (fertility, mortality, net migration) to
changes in the cost of moving, and retirement rules. These topics are discussed only at the end of the
chapter (see section . ) because they are relatively inaccessible to non-economists. Economists
should thus preferably read the last section of the chapter first, as it lays a good foundation for the
rest of the chapter. It is, however, not essential for the reader to understand the theoretical section of
section . in order to comprehend the issues discussed in the rest of the chapter. Nevertheless, this
section does help the reader understand better how the labour market as a whole operates.
The information gained in this chapter will form the basis of analyses in other chapters, for instance,
to find out how the market determines the wage rate, unemployment in the country, and so forth.
. THE LABOUR FORCE AS THE SUPPLY OF LABOUR
The total supply of labour is all those people who are working, whether for themselves or for
someone else, as well as those who want to work and are looking for work, i.e. the unemployed
people.
The total labour force (LF), or the economically active population (EAP) as it is also known, is thus
defined as the total number of working-age population or potential labour force ( – years) who
present their labour for the production of economic goods and services, whether employed or not. This
can be illustrated by the following:
The supply of labour is determined by a number of factors, as graphically illustrated in Figure . .

FIGURE . Determinants of the labour force
Source: Adapted from McConnell et al. ( : )
The individual person’s decision to work or not to work, and how many hours to work. The wage
rate is the most important determinant of this decision. The model of the individual or family labour
supply is discussed in section . .
These various decisions by individuals to work or not to work are measured by the labour force
participation rate (the percentage of the working-age population that is working or wanting to
work), to be discussed later in this section.
The size of the population also has a bearing on the size of the labour force. The population size is,
in turn, determined by factors such as fertility rates, mortality rates and net migration flows, to be
discussed in sections . – . .
Labour supply is not only a matter of quantity but also quality. The skills level of the labour force is
thus also significant. This will be discussed in Chapter .
There are three kinds of lies: lies, damned lies, and statistics.
B D
As discussed above, labour force equals the sum of employed and unemployed. The derivation of
employed will be discussed in section . while the derivation of unemployed will be explained in
detail in section . . The labour force can be derived according to the official, narrow (strict) or broad
(expanded) definition. As explained in section . . , persons who are unemployed and want to work,
but are not actively seeking employment would, in terms of the official definition of unemployment,
not be counted as unemployed. They are the so-called “discouraged work-seekers”. They would,
according to Statistics South Africa, not even form part of the labour force, in spite of the fact that
they might dearly want to work, but are simply not actively looking for work. Nonetheless, these
discouraged work-seekers are included as part of the labour force according to the broad definition.
In other words, the labour force number is bigger according to the broad definition. As the
methodology to derive the discouraged work-seekers has changed drastically since , the broad
labour force aggregates before and since are not comparable, and this will be discussed in
greater detail in section . . .
The discussion below starts with the size of the South African labour force, and considers some of its
most important characteristics. Thereafter the labour force participation rate (LFPR), i.e. the
percentage of the working-age population that participates in the labour market, is discussed. The
focus of the discussion will be on the labour force aggregates according to the official (narrow)
definition, unless stated otherwise.
Finally, the total population is discussed, as well as its various determinants, such as the fertility rate,
mortality due to HIV and AIDS in particular, and immigration and emigration.
. . The South African labour force

Bearing in mind the shortcomings of the data on the labour force, the total labour force of South
Africa, according to the official, narrow definition, nearly doubled from . million in to .
million in , as shown in Figure . . This means that, on average, approximately persons
enter the labour market annually to look for work. The actual number entering the labour market is
greatly influenced by labour market conditions – whether work is available, whether there are
opportunities to continue with studies, emigration, and so forth.
FIGURE . Total labour force, –
Source: Author’s own calculations using the – October Household Survey, – Labour Force Survey
September and – Quarterly Labour Force Survey quarter data
Figure . also shows that the total labour force according to the broad definition increased from
. million to . million during the same period – a rise of . million. The abrupt decline of the
total labour force under the broad definition between and (a decrease of ) was
attributed to the changes in the methodology to define the discouraged work-seekers (to be
discussed in section . . ). The difference between the broad labour force and the narrow labour
force gives the number of discouraged work-seekers, which increased from . million in to .
million in . For the remainder of section . , labour force according to the official, narrow
definition by population group, gender, age and education is discussed.
. . Labour force by gender

With some exceptions, researchers agree that the participation of women in the labour market has
increased quite sharply, as illustrated in Table . . Although not shown in the table, in , women
constituted only % of the labour force. This increased to % in and % in . The
increased labour force participation rate of women, as described in section . . , is primarily
responsible for this phenomenon. The reasons for this development are also discussed in the same
section. They relate to changing norms with regard to the role of women in the labour market,
increased wage levels and lower fertility rates.
TABLE . The labour force of South Africa by gender (’ )
1995 2005 2015 2016

Male 6 689 9 066 11 648 11 972
Female 4 808 7 638 9 586 9 721
Total 11 497 16 711 21 234 21 694
Female labour force 41.8 45.7 45.1 44.8
as proportion of total
Source: Author’s own calculations using the October Household Survey, Labour Force Survey September, and
– Quarterly Labour Force Survey quarter data
Note: Labour force with unspecified gender is not shown separately in this table but is included as part of the total.
However, Yu ( ) makes the point that different surveys cannot simply be compared and
conclusions drawn from such comparisons. In South Africa, the October Household Survey (OHS), the
Labour Force Survey (LFS) and the Quarterly Labour Force Survey (QLFS) differ in various respects and
are not strictly speaking fully comparable. Accordingly, by considering only the trends in the LFS
between and , Yu concludes that there is no strong evidence of a further “feminisation” of
the labour force during this period. In fact, looking at the – period as a whole, the female
labour force as a proportion of total never exceeded % – it peaked at % in .
. . Labour force by population group

The available data on the labour force of South Africa by population group are presented in Table
. . In , more than three-quarters of the labour force were Africans, while there was a sharp drop
in the proportion of whites, from . % in to . % in . In absolute terms, the African labour
force more than doubled between and , but the white labour force stagnated at the . – .
million range.
TABLE . The labour force of South Africa by population group (’ )
Africans Coloureds Indians/ Whites Total Proportion of total

Asians
1995
Total 7 809 1 360 400 1 928 11 497 100.0
Proportion of total 67.9 11.8 3.5 16.8 100.0
2005
Total 12 357 1 710 522 2 087 16 711 100.0
2015
Total 16 431 2 121 576 2 106 21 234 100.0
2016
Asians
Total 16 910 2 128 601 2 054 21 694 100.0
Note: Labour force with unspecified population group is not shown separately in this table but is included as part of the total.
. . Labour force by age

Although South Africa’s total population is very young ( % was under the age of in ), persons
in the -to- -years age category made up only about % of the labour force in , as shown in
Table . . This implies a high dependency ratio (i.e. the ratio of young people depending on other age
groups for a living). In addition, as indicated in Chapter , a high proportion of the youth labour force
is unemployed, which implies a further increase in the dependency on other age groups and a heavy
burden on the taxpayer in terms of educational, health, policing and similar expenditures, and
ultimately, employment opportunities.
TABLE . The labour force of South Africa by age cohort (’ )
15–24 years 25–34 years 35–44 years 45–54 years 55–64 years Total Proportion of total
1995
Total 1 770 4 097 3 224 1 740 667 11 497 100.0
Proportion of total 15.4 35.6 28.0 15.1 5.8 100.0
2005
Total 2 914 5 957 3 970 2 728 1 142 16 711 100.0
2015
Total 2 761 7 122 6 096 3 705 1 550 21 234 100.0
2016
Total 2 761 7 211 6 267 3 889 1 566 21 694 100.0
Compared to about % of the population in South Africa being younger than years, most other
African countries have a much higher proportion of young people (some as high as nearly %). The
proportion of young people in the populations of industrialised countries is much lower, and the main
problem in these countries is a high proportion of elderly people (Population Reference Bureau, ).
HIV and AIDS will also have a significant impact on young people. Many breadwinners are likely to die
of HIV and AIDS, leaving young children to fend for themselves. Such children will most likely be
unable to attend school, making them “unemployable”.
. . Labour force by education
Table . presents the labour force total by highest educational attainment. One encouraging finding
is the decrease in labour force with no education to Grade between and , in both absolute
and relative terms. The labour force with post-Matric qualifications increased by . times (from .
million to . million), while the proportion with at least Matric increased from % to % during the
same period. The results suggest that the labour force has become more educated over the years.
TABLE . The labour force of South Africa by education (’ )
None–Grade 7 Grade 8–11 Grade 12 Grade 12 + certificate/diploma/degree Total Proportion of total

1995
Total 3 618 3 865 2 530 1 393 11 497 100.0
2005
Total 4 041 5 888 4 654 2 014 16 711 100.0
2015
Total 2 948 7 699 6 771 3 608 21 234 100.0
2016
Total 2 800 8 283 6 913 3 533 21 694 100.0
Quarterly Labour Force Survey quarter data
Note: Those with certificates or diplomas but did not have Grade fall under the “Grade – ” category. Also, labour force with
unspecified education are not shown separately in this table but are included as part of the total.
SUMMARY
The labour force (LF), also called the economically active population (EAP), is the total number of people from the
working-age population ( – years) presenting their labour for remuneration on the labour market, whether
employed, selfemployed, employers or unemployed.
The participation of women in the labour market has increased steadily since the economic transition, and they now
form about % of the labour force.
Young people make up a high proportion of South Africa’s total population, but a relatively small proportion of the
labour force. This implies a high dependency ratio (i.e. the ratio of young people depending on other age groups for a
living).
The labour force has become more educated – the proportion with at least Grade increased from % in to
nearly % in .
. LABOUR FORCE PARTICIPATION RATE

Only a proportion of the total working-age population works or wants to work. The rest are
housewives, students, too sick to work, and so forth. The proportion that works or wants to work is
called the labour force participation rate (LFPR). LFPR is thus the percentage of the working-age
population that furnishes its labour for the production of economic goods and services, whether
employed or not. It is thus the percentage of the working-age population that is economically active:
actual economically active population actual labour force
LFPR = × 100 = × 100
total population of working age potential labour force
The Centre for Development and Enterprise (CDE) ( a) states that one of the most important
features of South Africa’s labour market is the very low rates at which adults participate in it. By
international standards, the LFPR in South Africa – at about % – is significantly lower than that of
most other countries, with the only exceptions being countries in which social, cultural and religious
norms keep large numbers of women out of the labour market.
One explanation frequently offered for the relatively low labour force participation likelihood is that
South Africa’s social welfare net discourages people from seeking work. However, there is also
contradictory evidence, for instance that for women the expansion of social grants might have
increased participation rates. This, it is suggested, was because access to cash income in the
household allowed women to leave their children with relatives more easily, freeing them up to look
for work. Furthermore, the LFPR is significantly higher today than it was in the mid- s, when it was
in the mid- % range (Banerjee, Galiani, Levinsohn & Woolard : ). This period has also seen an
increase in access to, and in the value of, social grants which has coincided with rising participation
rates. Social welfare thus does not seem to have kept people out of the labour force.
The CDE states that a more likely explanation for low participation rates is that, after ,
expectations rose and more people set out to look for work. This led to an initial rapid rise in
participation in the labour market. However, the new entrants were unable to find work, and this
discouraged people from actively seeking work. The fact that they didn’t seek work, and are today
overwhelmingly measured as “not economically active”, is a reflection, by this account, not of their
dependence on welfare, but on their (realistic) appreciation of the very low probability that they will
find work if they look for it.
Cyclical nature of the LFPR: This explanation is also confirmed by the decrease in the LFPR
after the economic crisis (see Figure . ), which is found primarily but not solely among the
youth (ILO, : ). Those who lost their jobs as a consequence of the crisis chose to exit the labour
market rather than actively search for employment. Many of those who leave or delay entering the
labour market might use the opportunity to upgrade their human capital, for instance by remaining
longer in schooling.
Narrow LFPR versus broad LFPR: As mentioned before, the labour force number defined according
to the broad definition is greater than the labour force number defined according to the narrow,
official definition. Hence, as expected, the LFPR would be greater according to the broad definition.
For example, Figure . shows that in , the narrow LFPR and broad LFPR were % and %
respectively, while the corresponding figures were % and % in . Note that the abrupt decline
of the LFPR under the broad definition between ( . %) and ( . %) is attributed mainly to
the drastic changes in the methodology to distinguish discouraged work-seekers since .
FIGURE . Labour force participation rate, –
For the remainder of this section, the LFPR according to the narrow, official definition will be the focus
of the discussion.
. . LFPR by gender
The LFPR of women has increased relatively more rapidly over time. This much can be gathered from
Table . . The difference between the male LFPR and female LFPR narrowed from . to .
percentage points between and . The increase in the participation rate among women is
probably the most important change in the labour supply in South Africa that has taken place in
recent years. Nonetheless, the male LFPR remains higher.
TABLE . Labour force participation rates (per cent) by gender

Male Female
1995 58.6 38.4
2000 66.3 52.3
2005 64.0 50.3
2010 63.1 48.1
2015 65.5 52.4
2016 66.0 52.3
Source: Author’s own calculations using the October Household Survey, and Labour Force Survey September,
and , and Quarterly Labour Force Survey quarter data
The following factors are likely to be some of the reasons for the relatively more rapid increase in
female labour force participation likelihood. The analysis at the end of this chapter about the factors
determining the theoretical supply of labour by the household is also relevant in this regard.
It is possible that the increasing coverage of various social grants, especially child grants, could
have contributed to more women participating or wanting to participate in the labour market. This
is because the payment of social grants might make some money available for transport; allow
women to take some risk in small-scale businesses; or allow mothers to pay for childcare while they
work.
The rising relative wages for women as a result of reduced discrimination and a greater demand for
female workers have brought about their increased participation in the labour force. This is related
to the expansion of the services sectors, in which women have traditionally been active.
The rising levels of education among women have increased employment opportunities and
earnings, and would thus theoretically have increased the supply of female workers. In , only
% of the female labour force had completed Grade ; this proportion increased to % in
and a further % in (it exceeded % for the first time in ). However, Casale and Posel
( : ) did not find that increasing education levels could in fact be associated with rising
labour force participation rates, and thus rejected this as a possible reason for higher participation
rates.
Even though they may not be related to education, declining birth rates have enabled more women
(of all educational levels) to participate actively in the labour market for longer periods of time.
There have been changes within the household: firstly, a decreasing proportion of women live with
employed men. This is related to the increase in the number of unemployed men in the economy,
which undermines the resource base of the household, thereby increasing the pressure on women
to generate an income. Secondly, there has also been a decline in the number of women living with
any man, which shows that there has been a decrease in the proportion of women who have
married (Casale & Posel, : – ).
Social conventions about a woman’s role in society (e.g. that a woman’s place is in the home) have
changed. Preferences and attitudes among women themselves have also changed, with more and
more women embarking on professional careers for selffulfilment.
Rising productivity in the household because of advancing technology has made it possible for
women to participate more actively in the labour market.
The implementation of the Employment Equity Act (Act No. of ) and the Broad-Based Black
Economic Empowerment (BBBEE) Act (Act No. of ), which aim at achieving equity in
employment through promoting equal opportunities for the designated groups, including females,
could have a positive impact on the female working-age population’s decision to participate in the
labour market by actively seeking work.
How many women are actually in the labour force?
The data on the female LFPR hide certain important measurement problems. The main difficulty is to distinguish
between people who are in the labour force and those who are not. Usually the distinction is made on the basis of
persons who work for remuneration or who actively seek a paying job – these persons are then considered to be part of
the labour force. However, the categorisation of people who want to work but are not actively looking for work (the
expanded definition of unemployment) will have an important impact on the LFPR. In addition, the categorisation of
homemakers and women in subsistence agriculture is problematic. These women seldom work for remuneration (in the
direct sense of the word), although one cannot deny that their services are economically productive.
When a homemaker, for instance, enters the labour market and employs a domestic worker in her home and in addition
pays a fellow homemaker to look after her children, then all three women who might not have been “economically active”
before, now suddenly become so, and their services are counted as part of the national product of the country. Thus,
there would seem to be an underestimation of production and the EAP with regard to homemakers.
There have also been problems in measuring women in subsistence agriculture and the informal sector. Censuses have
approached women in subsistence agriculture in different ways. The census of , for instance, treated women in
subsistence agriculture mostly as economically active and working. In these women were treated as not being
economically active. The latter approach has been followed in subsequent censuses. Knight ( ) has estimated that
this change of definition led to roughly African workers being reclassified from economically active and working
to not economically active. This resulted in lower measured female employment, hence it has raised the female
unemployment rate (ILO Review, : ; – ; – ). Men in subsistence agriculture are classified as
economically active.
The measurement of the informal sector in surveys such as the OHS and the LFS has also not been consistent, as
indicated in section . . . Since surveys have, for instance, been much more explicit in identifying informal sector
work as employment, and enumerators have asked much more probing questions regarding the informal sector and
subsistence farming. This would also have been a factor in increasing the labour force and thus the LFPR of female
workers in particular.
. . LFPR by population group

Table . shows that while the LFPR increased in all four population groups between and ,
such increase was the greatest for the Africans – a rise of about percentage points, while the
corresponding increase was only between – percentage points for the other three groups. The
relatively greater increase of the African labour force participation likelihood was attributed to various
factors, such as the implementation of the post-apartheid labour legislations, which promote the
previously disadvantaged groups (this is discussed in detail in section . ) and that the African labour
force has become more educated (the proportion of these individuals having completed Grade
increased from % in to % in ).
TABLE . Labour force participation rates (per cent) by population group

Africans Coloureds Indians/Asians Whites
1995 43.5 60.4 56.8 64.0
2000 56.7 65.2 61.9 68.9
2005 54.3 64.3 63.4 68.7
2010 52.7 63.8 62.7 67.2
2015 57.2 63.6 59.0 68.5
2016 57.7 63.1 60.9 67.5
. . LFPR by age
Table . shows that the LFPR increased in all age cohorts between and , but this rate has
always been the highest for the – years cohort ( . % in ), followed by the – years
( . %) and – years ( . %) cohorts. The LFPR was the lowest for the – years cohort, and this
result is not surprising, as the majority of the people in this age cohort are expected to be studying on
a full-time basis and hence inactive in the labour market. The LFPR of the oldest, – years cohort is
always below % – it is possible that some people in this cohort are inactive due to reasons like early
retirement.
TABLE . Labour force participation rates (per cent) by age cohort
15–24 years 25–34 years 35–44 years 45–54 years 55–64 years
1995 21.7 63.9 69.6 62.7 34.4
2000 31.9 76.1 79.2 72.7 48.4
2005 30.7 74.8 75.4 69.1 42.7
2010 26.2 72.0 75.1 67.6 40.8
2015 26.8 75.0 79.8 71.7 44.3
2016 26.8 74.6 80.2 73.0 43.3
. . LFPR by education
An increase in educational attainment is associated with a greater probability of labour force
participation, as shown in Table . , as those with a higher level of education and skills level are the
ones demanded in the labour market, particularly given the structural change of the South African
economy in recent years (this is discussed in section . ). Table . shows that in , the LFPR of
those with post-Matric qualifications ( . %) was nearly twice as high as the LFPR of those without
secondary education ( . %).
TABLE . Labour force participation rates (per cent) by education

None–Grade 7 Grade 8–11 Grade 12 Grade 12 + certificate/diploma/degree
1995 44.8 39.5 61.5 79.6
2000 55.7 50.6 70.9 87.4
2005 48.8 48.7 72.1 85.7
2010 41.8 46.9 69.5 86.8
2015 46.2 48.9 68.7 86.6
2016 45.8 50.5 70.5 86.7
Source: Author’s own calculations using the October Household Survey, and Labour Force Survey
September, and , and Quarterly Labour Force Survey quarter data
SUMMARY
The most important change in the labour supply in recent years has probably been the more rapid increase of LFPR
among women. This can be ascribed to a number of social and economic factors.
Although the LFPR remained the highest for whites, it increased most rapidly for Africans since .
The LFPR was the highest in the – years cohort, followed by those aged – years and – years.
The LFPR was close to % for those with post-Matric qualifications.
. POPULATION AND POPULATION GROWTH
. . Population size
In , South Africa had a total population of about million people (see Table . ). Africans
formed about % of the total population (compared to % of the labour force) and women %
(compared to % of the labour force) in . South Africa’s population is about the same as that of
Myanmar and Tanzania. It is larger than that of Australia (more than double), Canada and many of the
smaller European countries, but less than that of the United Kingdom or France. South Africa’s
population is only % of that of China, the world’s most populous country (World Bank, ).
TABLE . Total population of South Africa, (’ )
Male Female Total % breakdown by population group

Africans 22 119 22 990 45 110 80.7
Coloureds 2 368 2 529 4 897 8.8
Indians/Asians 702 684 1 386 2.5
Whites 2 190 2 325 4 516 8.1
Total 27 380 28 529 55 909 100
% breakdown by gender 49.0 51.0 100
[Due to rounding, figures do not always add up.]

Source: Statistics South Africa, Mid-year population estimates ( a: )
As has already been stated, an important characteristic of the South African population is the high
proportion of young people, and the burden on society and the working-age population that this
causes. This problem is likely to be exacerbated as a result of HIV and AIDS, because the disease will
mainly affect income earners, thereby leaving many households without a breadwinner and, in many
cases, even leaving orphans.
Those who warn of a population explosion picture a world of too many people and not enough food –
sort of like the average cocktail party.
B V
. . Population growth rates

South Africa’s population and therefore its labour force grew quite rapidly in the past – by as much as
. % in the s. However, the growth rate declined rapidly and Statistics South Africa ( a)
recently estimated the rate of natural increase of the population at only . % in . This is related to
factors such as sharply declining fertility rates, the impact of HIV and AIDS and the net loss of skilled
workers owing to emigration.
The population growth rate can be compared to that of some other countries, for example Germany
( . %), the US ( . %), Brazil ( . %), India ( . %), China ( . %) and Mexico ( . %) (World Bank, ). It
is important to note, however, that the data for all countries relate to the natural increase in
population, i.e. disregarding migration.
. . Population by age
The South African population is slowly maturing. The proportion of the population under the age of
years is steadily declining while other age categories, particularly the proportion of elderly people
(over years of age), are showing noticeable growth. The proportion of people (especially Africans)
between the ages of and years of age is expected to grow significantly in the future. There is
therefore likely to be an upsurge in the proportion of people of working age relative to the
proportion of dependents such as children and the elderly, which might provide a demographic
impetus for development. However, the demographic potential might be offset by the poor
educational and skills levels of the population.
There are notable differences between especially the African and the white population by age, which
are likely to have a significant impact on the labour force in the future. The African population shows
the typical pattern found in a developing country, which is that the population is relatively young. In
contrast, the white population is typical of that found in many developed countries, which is an
ageing population.
Figure . (a) and (b) shows the so-called “population pyramid”. These graphs indicate the
percentage of the male and female population in the various age groups.
FIGURE . Population pyramid of the African and white population,
Source: Statistics South Africa, Mid-year population estimates ( : )
Figure . (a) clearly shows the impact of the high fertility rates on the African population until
recently, with the population pyramid having quite a large basis (a high proportion of people in the
younger age groups).
The ageing white population is clearly discernable in Figure . (b) – with higher proportions of the
white population in the older age categories compared to Africans. However, what is also significant
in this graph is the relatively small proportion of the population in the age groups – . This is a
clear indication of the impact of the emigration of young workers, usually with skills, from the country.
. . Factors determining population growth rates

Population growth is a function of three demographic factors: fertility, mortality and migration. This
can be illustrated as follows:
The difference between the fertility and mortality rates is called the natural rate of population increase.
The fertility rate is discussed in more detail in the next section. The mortality rate is greatly affected by
HIV and AIDS; this is dealt with in section . . Immigration, emigration and migrant labour are
discussed in section . .
SUMMARY
The rate of population increase in South Africa is relatively high by international standards, but has dropped sharply
owing to a substantial decline in fertility rates and a high level of emigration among skilled people.
. TOTAL FERTILITY RATE
Of the three demographic factors, the total fertility rate (TFR) of the population is the most important.
The TFR is the average number of children born alive to a woman in her reproductive years ( to ).
The very pessimistic quote at the beginning of this chapter relates to the fertility of a nation. It was
made by Reverend Thomas Malthus in . According to him, as incomes rise above the subsistence
level, men and women become “unrestrained”, and the resultant increase in fertility inevitably leads to
an increase in the population. The available natural resources will not be sufficient to support a larger
population, and the consequent competition for these scarce resources causes incomes to fall again,
which will “restrain” the population. So, according to his theory, income will always be pushed back to
its subsistence level. The theory thus in effect argues that there is a positive correlation between
income and fertility – as income increases, fertility also increases.
Misery, up to the extreme point of famine and pestilence, instead of checking, tends to increase
population.
S L
However, real-world experience does not support this theory; as per capita income and living
standards increase, fertility rates decline rather than increase. This is also borne out by the South
African experience.
To estimate TFRs in developing countries is particularly difficult. This is because there is not complete
reporting of births. As a result, estimates of the level of fertility are derived from census/survey data
by employing indirect demographic estimation techniques. Statistics South Africa ( ) has reported
that although there are differences between the existing studies, there is an agreement among
authors that fertility has been following a downward trend in South Africa since the s. Also, there
is agreement that fertility patterns in the country differ by population group, as well as some other
factors such as province of residence, rural/urban place of residence and level of education.
The TFR in South Africa has declined from an estimated . children per woman in the s, to an
estimated . according to the census. Since then, there has been a further slight decline to
. in (Statistics South Africa, : ). However, the mid-year population estimates of Statistics
South Africa estimate a TFR of only . in , . in and . in .
At the time of writing, the TFR by population group is not yet available, but in , Africans ( . )
and coloureds ( . ) had comparatively higher fertility levels than their white ( . ) and Indian/Asian
( . ) counterparts. In fact, the TFRs for Africans and coloureds are still above the replacement level of
. . This so-called “replacement level” or “breakeven point” is the point where the fertility rate and the
mortality rate are equal so that there is no increase or decrease in the population.
The reduction in the fertility rates since the s, and the consequent reduction in the rate of
population growth that this could bring about, is important for a number of reasons:
A population growth rate that is too high places great demands on social services, such as
education, health, transport and the provision of housing.
Lower population growth will assist in improving living standards. If economic growth increases at a
higher rate than the population increases, the per capita income of the country increases. Although
the “haves” will benefit more than the “have nots”, incomes will rise and, as will be shown
elsewhere, even without state intervention, rising incomes are likely eventually also to benefit the
“have nots”. State intervention in the form of various transfers will speed up this process.
High population growth rates place tremendous pressure on the environment and on natural
resources.
The successful implementation of development programmes depends greatly on the effective
curbing of the population growth rate. A relatively high population growth rate exacerbates the
basic needs backlog South African society faces (ANC, : ).
The decline in the TFR and population growth rates as a whole can be explained by the so-called
“theory of demographic transition” (see accompanying box).
Theory of demographic transition
The theory of demographic transition may explain both the declining TFR in South Africa and the large differences in the
population growth rates of the various races (Marshall et al. : ; Todaro & Smith, : – ). South Africa’s
population is currently progressing through various stages of the demographic cycle. The transition can be explained by
way of a curve as in Figure . .
FIGURE . Theory of demographic transition
The first stage is characterised by high birth rates and high mortality rates and this is typically what happened before
economic modernisation took place. Because both rates are very high, the population may increase slowly or not at all.
In some instances the birth rates could, because of strong cultural traditions and strong family ties, be at relatively
moderate levels.
The second stage is characterised by a steep decline in mortality rates, particularly among the very young, because of
improved health services, improved water and sanitation systems, and better medical care, after modernisation took
place. Birth rates may remain high or even increase because of a breakdown in cultural traditions and family life. The
result is a sharp increase in the natural rate of population growth.
The third stage requires sharply declining birth rates, making it difficult for some less developed countries to move from
the second stage. Stage three is thus characterised by declining fertility and low mortality rates. It has been found that
birth (fertility) rates start declining only under certain circumstances, and these relate especially to improved living
standards. The whites in South Africa made the transition to this stage in the s; the Indians/Asians in the s; the
coloureds in the s and the Africans in the s (Simkins, ).
At some stage birth rates decline to below the replacement level, and the population declines. Some highly industrialised
countries are at this fourth stage of demographic transition, and white South Africans seem to be entering this phase.
. . Factors influencing fertility rates

Goliber ( : – ) describes the determinants of high fertility as: rural residency; poverty; the low
status of women (which leads to childbearing becoming an important way for them to gain status);
women needing children for their labour and for the security they provide in the women’s later life; a
culture which frowns on women remaining childless; low life expectancy; and low educational
standards.
Increased employment opportunities (especially for women), the enhancement of the position and
status of women in the community, improved education and primary health care, the reduction of
poverty and the provision of housing can all play an important role in reducing birth rates. Family
planning and birth control also play a role, but are less important than improved living standards. In
South Africa, the abolition of influx control in seems to have contributed to the reduction in birth
rates, because increased urbanisation also reduces fertility rates. Urbanisation reduces dependency on
children for their labour in agriculture and generally improves living standards. Furthermore, the
decriminalisation of abortion, with the Choice on Termination of Pregnancy Act of being effective
since , may also contribute to the decline in birth rates.
More than any other time in history, mankind faces a crossroads. One path leads to despair and utter
hopelessness. The other, to total extinction. Let us pray we have the wisdom to choose correctly.
W A
SUMMARY
All indications are that fertility rates in South Africa have declined quite substantially in recent years.
Improving the social and economic status of women, increasing living standards and higher urbanisation rates make a
significant contribution to the reduction in fertility rates.
Declining fertility rates reduce high population growth rates. This, in turn, is important as it relieves the pressures on
social services, the environment and employment rates.
. THE IMPACT OF HIV AND AIDS
HIV and AIDS have an impact on a number of elements of the labour market, such as the size of the
population and of the labour force, inequalities and productivity. It has probably become the most
devastating disease humankind has ever experienced, and unfortunately the impact is particularly
severe in South Africa.
The incidence of the human immunodeficiency virus (HIV) is much higher in Africa than elsewhere,
and is most virulent in southern and eastern Africa. In these countries the highest incidence is among
heterosexual persons. The infection rates in South Africa are among the highest in the world. In some
African countries, however, HIV incidence seems to be stabilising. This can be ascribed to the fact that
many people with high-risk behaviour have already been infected, and effective prevention
programmes enable people to reduce their risk of exposure.
. . Incidence and causes of HIV and AIDS

In the case of South Africa, the most important facts regarding the incidence of HIV and AIDS are the
following (Shisana et al. ; Statistics South Africa, a):
There were about . million people with HIV in , which represents a prevalence rate of . %
(the percentage of the total population living with HIV).
Compared to the overall rate of . %, the infection rate among adults (age – years) was higher
at . %, and even higher among adult women ( . %)
People living in informal urban areas were significantly more likely to be HIV positive.
There were no discernable differences in HIV prevalence among the employed and the
unemployed. The same was true with regard to disposable income – there was no discernable
difference between wealthy African households and those that were poor.
The prevalence rates were highest in the mining, transport, construction, government,
accommodation and catering sectors. The lowest incidences were in the financial, insurance,
business and communication services.
The prevalence of HIV increased with increasing levels of schooling, with the highest incidence
occurring among those with Matric. Thereafter the rate dropped quite sharply for persons with
tertiary qualifications.
Demographic projections estimate the life span between infection and death at eight to years (in
the absence of antiretroviral treatment (ART)). During the first part of this period, the HIV-positive
individual may experience relatively few direct symptoms, and employment and productivity may
be only marginally affected. However, with the gradual onset of the acquired immunodeficiency
syndrome (AIDS) (usually during the last two years), the medical symptoms become progressively
worse and eventually lead to death.
Factors that contribute to the high levels of HIV include the high prevalence of untreated sexually
transmitted diseases (STDs), which significantly increases the risk of acquiring HIV; multiple concurrent
sexual partners; relatively high levels of unregulated commercial sex; low rates of condom use (at least
until recently); low rates of male circumcision; rapid rates of urbanisation (resulting in multiple sexual
relationships and STDs); disrupted family and communal life owing to apartheid and migration;
massive refugee movements driven by war, civil conflict and economic distress in neighbouring
countries; good transport systems and high mobility, which allow for the rapid spread of the disease;
low status of women, resulting in situations that place them at risk of HIV infection; and poor health,
hence low resistance levels.
. . Consequences of AIDS
It is important to note that it is not possible to determine the direct impact of HIV and AIDS on
mortality rates in the country. Such statistics are not available, because mortality data are collected
from death notification forms, which give the direct cause of death, such as tuberculosis, influenza,
pneumonia, and so forth. Many of these diseases appear to be HIV and AIDS related. Statistics South
Africa nevertheless estimates the number of AIDS deaths in at . It is encouraging that this
figure is much lower compared to , when the highest number of AIDS-related deaths ( )
was recorded.
The following possible consequences of HIV and AIDS can be identified:
Demographic impact. HIV and AIDS have already resulted in life expectancy dropping from . in
to . in . More recently, however, life expectancy has increased to . (Statistics South
Africa, a).
Dependency ratio. Because HIV and AIDS mostly affects sexually active adults, it results in an
increase in mortality in the -to- -year age group. The likely sharp increase in the number of
orphans is perhaps the most tragic and enduring legacy of AIDS, and it has been estimated that
there are already . million AIDS orphans (Statistics South Africa, ). It will also negatively affect
the dependency ratio in another way, i.e. many elderly persons will lose their adult children and
thus their support structure.
Supply of labour. HIV and AIDS will result in the labour force being smaller than would otherwise
have been the case, seeing that the overwhelming majority of persons affected by HIV and AIDS will
be between the key productive ages of and years. They will affect both employed and
unemployed persons, although the expected impact on unemployment will depend on the mix of
skills demanded by employers.
Employment. Unemployment is unlikely to decline because economic growth rates will not be high
enough to reduce unemployment significantly (Arndt & Lewis, ). Van Aardt ( ) projects that
fewer jobs will be created as a result of investor caution, lower consumer demand due to a smaller
population, higher capital intensity as employers try to avoid the negative consequences of HIV and
AIDS on their businesses, and higher levels of mortality among employers because of AIDS.
Depletion of human capital. The personal and economic returns from years of investment in
schooling and higher education are often cut off before any returns can accrue to the individual,
family or society (Simkins, ). Mortality among breadwinners and other parents weakens
families, and this damages the acquisition of human capital by children. Children, for instance, have
to fend for themselves, and do not receive material or emotional support to attend school.
Poverty and inequality. Many households face dire financial difficulties as their breadwinners fall ill
and die, and household incomes are diverted for the care of HIV-and-AIDS-affected persons and
funerals. This makes the achievement of overall development objectives more difficult, and is likely
to increase the generally high level of inequality already found in the country. As employers
attempt to become more capital intensive by decreasing their reliance on unskilled workers and by
employing more skilled workers, inequalities also increase.
Insurance industry and benefit schemes. Employers face increased contributions to their employees’
life, disability and medical aid schemes. Insurers, and in particular life insurers, are exposed to
increased liabilities on existing policies. Various methods are also being employed to limit future
liabilities.
Direct health costs. The impact of AIDS on medical costs and health services is quite significant.
Hospitals are overburdened with AIDS patients, reducing effective health care. As the provision of
ART is expanded, the effect of HIV and AIDS on hospitals is reduced, but other medical costs,
especially those of pharmaceuticals and associated medical support services to administer the
drugs, are high. Households also redirect their spending to health care spending.
Other consequences for government. Apart from the impact of high health costs on government
expenditure, there is increased pressure from orphan-care expenses and reduced revenues. The
decline in growth and lower profits also affect tax receipts negatively. These, together with higher
social support and health-care expenditure, increase government fiscal difficulties. In addition, the
government loses valuable skilled employees as more civil servants and teachers fall prey to the
epidemic. The epidemic also overburdens social systems and poses challenges for the welfare
system.
Impact on the workplace. It has already been found that HIV and AIDS has a negative impact on the
profitability of companies in the sectors worst affected by the disease. The largest negative effect
has been on workplace morale, labour productivity and absenteeism, followed by the increased
employer contributions to various benefit schemes, such as medical aid contributions, pension, life
and disability insurance, and funeral benefits (Whiteside & Sunter, ; SABCOHA/BER, ).
Companies have also experienced higher labour turnover, have lost experience and skills, and have
incurred higher recruitment and training costs. However, the loss in production will probably be less
than the reduction in the labour force, because companies will strive to maintain production, for
instance by replacing labour losses from the ranks of the unemployed, by importing skills or by
implementing more productive labour-saving technologies. Although AIDS will lead to losses of
labour, a high proportion will be semi-skilled or unskilled workers, who can be replaced. Production
costs will, however, increase because of several of the factors mentioned above. There will also be
an upward pressure on salaries because of skilled workers becoming scarcer and unions exerting
pressure on companies to make up for the increased health-care spending of the workers’
households.
Other macroeconomic effects. The macroeconomic consequences of AIDS are not seen to be
devastating (Bureau for Economic Research (BER), : ). However, estimates of the impact on the
economic growth rates (as measured by increases in GDP) vary greatly. For example, the World Bank
(Arndt & Lewis, ) estimates a significant impact (annual real GDP growth rate was projected to
decline by . % in – ), Burger and De Villiers ( ) estimate a moderate impact (a decline
of . %) and the BER ( ) estimates a smaller impact (a negligible decrease of . %). The
difference might be due to the fact that the BER has assumed that companies will take several steps
to avoid the negative consequences of HIV and AIDS, for example by replacing labour from
unemployed sources, by becoming more capital intensive to maintain production and by
restructuring benefit funds to reduce costs.
Policy issue: Addressing HIV and AIDS (ILO, a)
(a) Prevention
Promotion of healthier and safer sexual behaviour
Programmes to diagnose and treat sexually transmitted diseases (STDs)
Targeted interventions to prevent new infections among teenagers
Programmes to empower women and to eliminate discrimination
Encouragement of the use of condoms
(b) Mitigating the effect in the workplace

Skills investment combined with effective HIV and AIDS prevention
Prevention of secondary infections, such as tuberculosis and pneumonia
Cooperation between employers and unions to develop appropriate strategies
Minimising the risk of infection of co-workers, e.g. provision of protective equipment, training on safe procedures
Anonymous, unlinked surveillance through HIV testing
(c) Care and support

Information on support services and the allocation of scarce resources
Confidential voluntary counselling and testing programmes
More cost-effective modes of therapy for HIV and AIDS sufferers
An efficient grant or other support system, not only for the HIV disabled, but also for the elderly, orphans and others
affected by the death of breadwinners
(d) Elimination of discrimination

Elimination of the stigma and discrimination around HIV and AIDS
Greater opportunity for role models to relate how they deal with HIV and AIDS, but even more importantly, to
advocate safe sexual behaviour
Legislation to eliminate unfair discrimination
Workers and job applicants should not be required to disclose their HIV status
SUMMARY
The HIV epidemic in South Africa is spreading rapidly and it is estimated that nearly % of the total population is HIV
positive. Among adults, the rate could be significantly higher at about %.
AIDS will not “solve” South Africa’s unemployment problem because the medical and economic costs of AIDS will be
so high that they will seriously damage the economy long before the demographic effects of AIDS are felt.
There are various reasons for the high incidence of HIV and AIDS, and therefore also various elements to a strategy to
deal with the epidemic. Probably the most important elements are ensuring safe sexual behaviour and eliminating
discrimination against women and HIV and AIDS sufferers.
. IMMIGRATION, EMIGRATION AND MIGRANT LABOUR
. . Immigration and emigration of skilled people
A two-sentence message received by vice-president Lyndon Johnson came from an American Indian
on a reservation: “Be careful with your immigration laws. We were careless with ours.”
H P
The benefits and costs of migration

Migration inflows have been identified as a factor that increases economic growth to the benefit of
the destination country and all its citizens (United Nations, : xiii). This is more readily accepted in
respect of skilled workers, because when it comes to less skilled workers there are often concerns that
an influx of immigrants will result in job losses and reduced wages. However, empirical studies show
that migrants have only a modest impact on wages and employment (United Nations, : xiii). The
major reason for migrants not having a significant impact on the labour market is that they increase
not only the labour supply but also the demand for goods and services. In addition, some use their
entrepreneurial abilities to set up businesses. These activities increase the demand for labour,
offsetting the initial increase in the labour supply that migrants represent. Migrants also tend to be
net contributors to fiscal revenue: what they pay in taxes is greater than what they cost the state in
welfare payments, education and additional infrastructure.
As pointed out by the CDE ( b: ), the United States provides the best example of the value of
skilled immigrants. In more than a third of engineers and other IT professionals working there
had been born elsewhere. Foreigners were responsible for more than % of biotechnology
inventions; generated a quarter of all global patent applications originating in the United States in
; founded more than a quarter of American companies, including Intel, Sun Microsystems, Yahoo
and Google; and received % of all doctorates awarded in , including % of those earned in
physical sciences and % in engineering. The CDE also refers to equally compelling proof of the
importance of skilled immigrants that can be found in the efforts made by the world’s most
developed countries to attract them. In recent years, the United States, Germany, the United Kingdom,
France, the Netherlands, Canada and Australia have all taken steps to attract highly skilled workers.
The same is true of countries in the Middle East (see also accompanying box).
What the destination countries gain, the countries of origin lose. The most serious loss is the so-called
“brain drain”, i.e. the loss of skilled workers. In effect, the country of origin loses both its earlier
investment in the education and training of the person who emigrates and the future contributions
that person would have made to the economic development of the home country, including future
tax payments by highincome earners – all of which is extremely costly. Recent research has also
identified an indirect effect – such a person’s knowledge not only provides a direct benefit in terms of
available skills, but also has positive effects on the productivity of others (United Nations, : xi).
These people are furthermore often entrepreneurs, i.e. the creators of employment.
Policy issue: a cost-benefit analysis of immigration
Steyn, Meintjies and Haasbroek ( ) undertook a cost-benefit analysis of immigration to South Africa. They concluded
that for the employer it is more cost-effective to make use of immigrants than to train local people when graduates and
technicians are needed. With regard to artisans, operators and other related occupations, the opposite applies and it is
more cost-effective to train locally.
From a purely cost-effective point of view, it is also better for the state to rely on immigration rather than on training,
especially with regard to tertiary qualifications, which are heavily subsidised by the fiscus. However, the criteria for the
state cannot be based purely on cost considerations. Other objectives such as employment creation, upliftment of the
local population, ensuring a stable supply of skilled workers and black advancement should also play at least as
important a role.
In addition to purely cost considerations, the Harvard Group appointed by the South African government to advise it on
economic growth and development (Hausmann, ) pointed out that all empirical studies of labour demand show that
high-skilled and low-skilled workers are strongly complementary, not substitutes. This means that the greater the supply
of skilled workers, the greater the demand for unskilled workers. The shortage of highly skilled workers causes a lower
demand for low-skilled workers: the lack of engineers may cause the loss of hundreds of blue-collar jobs. They state that
since the shared growth strategy involves maximising the job opportunities of the less skilled, it is fundamental that the
high-skill constraint be relaxed, especially in tradables. An obvious policy by which to relax the constraint quickly is
through the liberalisation and encouragement of high-skilled immigration. They point out that greater immigration of
highly skilled workers will help limit salary increases for such workers (and thus the increase in wage inequality).
South African immigration and emigration

The country’s domestic skills production system is grossly inefficient: children fare badly in almost
every international test of literacy and numeracy; the training of technicians and artisans has all but
ground to a halt; and universities produce far too few engineers, managers and other skilled people.
To make matters worse, skilled people have been leaving the country at an alarming rate (CDE, b:
).
Although it is critically important to transform the whole system of human resource development in
South Africa, this will take many years to achieve. The only way in which South Africa can access the
large number of skilled workers that it requires is to import skilled staff from outside the country by
way of changes to its immigration policy.
Since , South Africa no longer keeps data on official immigration and emigration. This means that
one cannot determine whether the country is gaining or losing skills on a net basis – this while the
availability of skills is a critical element for the long-term growth of the country.
In the past, the immigration of skilled people had a significant impact on the supply of labour in
South Africa and especially on the supply of high-level personnel. From until the elections
there was a substantial net gain of migrants as a result of factors such as high economic growth and
the active recruitment of immigrants to South Africa through financial assistance. During the period
to South Africa experienced a total net gain of people or an average of
people per annum.
These gains had a substantial influence on the increase in the supply of skilled labour in South Africa.
Between and , for example, net immigration contributed about a quarter of the increase in
highly skilled and skilled personnel (National Manpower Commission (NMC), : ). However, since
the middle of the s, this no longer seems to be the case and a net loss of migrants seems to have
become the norm.
The CDE ( b: ) refers to studies that estimated that more than South Africans had
emigrated between and , with the numbers growing by about % a year. About of
those emigrants had professional qualifications, amounting to about % of the total stock of
professionals employed in South Africa, and more than eight times the number of professional
immigrants in the same period.
While the census questionnaire did not ask any explicit questions on migration, it captured
information on the birth place of each person as well as the year of moving to South Africa if the
person was not born in South Africa. In the census of , . million ( . % of the total population)
reported they were born outside South Africa; of these . million people, million ( . %) moved to
South Africa during – .
Immigration policy
In spite of the economic benefits of immigration gains, several countries have recently been experiencing
disenchantment with the multicultural approach that migrants inevitably bring about. Feelings of resentment among
pockets of the population, together with increased concerns about national security, the granting of asylum,
undocumented movements and the like have led to a re-examination of immigration policies in several countries (United
Nations, : xv). By , for example, one-third of all countries had policies to reduce immigration compared with
only % in .
This is quite a change from only a few years back when countries such as Britain introduced policies to attract innovative
people with high-tech skills from abroad (Absa, : ) and the US considered tax cuts in part designed to attract more
skilled immigrants into the country.
The resistance to the immigration of skilled workers seems to be much more severe in South Africa. The CDE ( b: )
describes South Africa’s immigration policy as follows:
South Africa’s responses to these pressures and opportunities have been poorly conceived and badly implemented.
We have failed to attract adequate numbers of skilled people in a competitive global labour market, and we have
lost many skilled citizens. We have also failed to manage the influx of irregular migrants looking for jobs, many of
whom lack the formal qualifications that would enable them to obtain residence permits, but who often have some
skills and energy needed to make an economic contribution. At the same time, a flood of asylum-seekers has taken
our asylum and refugee systems to the brink of collapse.
Factors causing immigration and emigration
There is much speculation about the causes of immigration and emigration, but relatively little
empirical research. The factors normally mentioned as causes are crime and violence, affirmative
action and racism. However, in an empirical study, Myburgh ( ) found that the most important
factors causing emigration from South Africa are
real wage differentials between South Africa and the destination country
immigration restrictions in the destination country
political uncertainty in South Africa.
With regard to political uncertainty, Myburgh found that the mid- s seemed to be the only clear
example of emigration induced, in part, by political uncertainty in the last years. Increased
emigration in the s occurred in the context of rising real wage differentials in Australia and the
UK.
Furthermore, Mattes and Richmond (as quoted by Myburgh, ) found that survey evidence does
not support the hypothesis that racism and affirmative action are important drivers of emigration. This
is in contrast to violence, which was found to be an important causal factor (see research quoted by
Myburgh, : ).
If South Africa intends retaining the core of its skills base, it needs to take careful note of the factors
causing an outflow of skills. As predicted by the ILO, the market for highly skilled workers is likely to
become more globally integrated and labour migration is expected to feature more prominently in
the future (as quoted by Strydom, : ).
SUMMARY
The influence of immigration on the supply of labour in South Africa, and especially on the supply of high-level
personnel, was considerable before . However, a net loss of skilled workers seems to have become the norm
since , and this loss is aggravating skills shortages in the country.
The actual number of emigrants seems to be about two or three times higher than that recorded by official statistics.
. . Migrant labour within South and southern Africa

The system of migrant workers from rural areas and from neighbouring countries, who work on a
temporary basis in South African cities or mines, is an important characteristic of the South African
labour market. In this respect a distinction can be drawn between temporary migrant workers and
commuters.
Temporary migrant workers are workers who live in rural areas and neighbouring countries and who
work and live in urban areas on a temporary basis. They return to their area of origin after completion
of their service contract, or on an annual basis for a period of leave. Commuters, on the other hand,
are workers who live in rural areas and commute to work in urban areas, but who return to their
homes every evening or every weekend.
Neighbouring countries
The term “neighbouring countries” refers to Botswana, Lesotho, Malawi, Mozambique and Swaziland.
These countries are referred to as the Southern African Labour Commission (SALC) countries. The total
number of legal migrants from neighbouring countries fell sharply in the s (Whiteside, : ),
with the greatest reduction being in the number of Malawians, followed by Mozambicans. This was
related to developments in the mining industry (see accompanying box).
In recent years the problem of irregular migrants from neighbouring countries has assumed serious
proportions. Large and uncontrolled influxes of irregular migration, together with increasing numbers
of asylum seekers, inevitably impose costs on the receiving country. These include greater
competition for jobs; uncertainties about, and increased pressures on, the planning and provision of
services, especially in poor communities; social tensions that feed xenophobia; the erosion of public
confidence in state agencies that appear unable to cope; and increased opportunities for corruption.
The failure by government to manage the influx of irregular immigrants effectively has also given rise
to a sharp increase in xenophobia, i.e. the fear of foreigners. Foreigners are often accused of taking
away jobs from citizens, engaging in criminal activities and competing for scarce resources and public
services.
Nonetheless, the CDE ( b: ) refers to surveys that demonstrate that irregular workers can and do
make important contributions to the South African economy. For instance, they are much less likely
than the local population to be out of work, and they also seem to compensate for the low level of
entrepreneurial talent among South Africans. Mpedi ( : ) refers to research that found that
migrants actually create jobs at the rate of more than three jobs per migrant. They also often occupy
positions to which unemployed South Africans do not aspire, for example in agriculture or atypical
employment.
Migrant labour on the mines
South African mines have drastically reduced their use of foreign African labour (i.e. from neighbouring countries) since
the s. In , for instance, the foreign African labour component on the mines amounted to % of all workers, and
this had increased to % by . However, in the s the percentage dropped to %, according to the Chamber of
Mines. This percentage dropped further to % in (Mail & Guardian, ).
The first reason for this development was the growing political hostility towards South Africa in the s and s,
which encouraged the mining industry to turn increasingly to the labour force within South Africa’s borders. The second
reason was sharply increasing labour costs on the mines. More and more workers were thus recruited “at the gate” of
the mine, which meant more local labour was hired (unless, of course, migrants posed as locals).
Another important characteristic of the migrant mining labour force is its stabilisation. This is due to the increasing
average duration of employment contracts. For instance, the average age of workers in gold mines increased from
years in to years in (ILO Review, : ).
Penny ( ) estimates that migrants in the gold-mining industry remit more than half of their contract wage to their
home area in cash or in kind, of which the biggest part is remittances sent home informally by the miners. The
remittances constitute more than % of the rural household’s total annual income. Rural economic production forms
only about % of annual income.
Some of these migrant workers even started second families around the mines and hence “their salaries [were] split
between the two families” (Chetty, : ). Furthermore, the migrant workers account for the majority of the lowest
remunerated mineworkers, because they are only hired seasonally for short periods and hence are not unionised.
Internal labour migration
With the lifting of influx control in South Africa, one would expect that the pattern of temporary
migration would be replaced by permanent migration, particularly to urban areas, and that the
migrants’ ties to their households of origin would have weakened considerably. However, Posel and
Casale ( : ) did not find strong empirical evidence to support this. In fact, the number of
households with at least one labour migrant as a household member increased slightly between
and . Posel and Casale found that there were . million households with at least one migrant
worker in . This comprised % of rural African households. A total of % of rural male adults
were migrant workers in that year.
Posel and Casale also found that there was an increase in the number of female migrant workers.
Females comprise about a third of all migrant workers. Some of the possible reasons for the increase
in female migrant workers were that women were less likely to be living with men who had
employment, which implied that there was a greater burden on the women to provide for the
household. There were also fewer women living with men in total, with the result that they had to earn
a living for themselves.
Regarding the remittances of internal migrants, Posel and Casale ( : ) found that the incentives
for remitting income were weaker than before, thus one might have expected remittances to have
been reduced. The weaker incentives were, firstly, that family members could join the migrant worker
in the urban areas, and secondly, that there were fewer income-earning opportunities for migrant
workers and increased pension receipts by rural households. However, the limited available data do
not show any reduction in remittances. Remittance income on average represents about one-third of
total household expenditure.
Statistics South Africa ( a) estimated the provincial migration streams in – , and found that
Gauteng was the province with the highest net migration of (in-migrants and out-migrants
were . and . million respectively), followed by the Western Cape (net migrants equalled
). The results are not surprising, as it is possible that the relatively more educated labour force
participants from other provinces migrate to these two provinces searching for skilled work
opportunities associated with higher remuneration.
Although there may be some arguments in favour of South Africa’s migrant labour system, for
example that it reduces the rapid growth of urban slums and provides a significant income for the
rural areas of origin, there are powerful arguments against the system. Because migrant workers are
often atypical workers (see section . . ), the system discourages training initiatives by the employer,
which limits the earning capacity of migrant workers.
Furthermore, labour migration has negative economic effects on the production potential of rural areas
and thus migrant labour essentially perpetuates its own continuance. There are three main reasons for
these negative effects:
It could be argued that it is usually the better educated, more energetic and younger workers who
migrate to the urban areas. The workers who stay behind are not as productive as those who leave
and they cannot maintain agricultural production.
Agricultural development requires investment, but the remittances from urban areas by migrant
workers and the surplus agricultural production are so low that there is no possibility of investment.
Rural communities have become used to migrant labour as a way of life. Most activities are
therefore directed at preparing sons and daughters for city (migrant) life and not for productive
activity in rural areas.
The result is a breakdown of social values and family life, and social ills and diseases are exacerbated.
SUMMARY
In recent years the problem of illegal migrants from neighbouring countries has assumed serious proportions, and
could damage the economic prospects of both South Africa and the countries of origin.
There does not seem to be a reduction in internal migration, and migration by female workers has in fact increased.
Remittances continue to form an important part – about a third – of the expenditure of rural households.
. DEEPER INSIGHT: FROM THE INDIVIDUAL DECISION TO WORK TO THE MARKET

SUPPLY OF LABOUR
The world is full of willing people; some willing to work, the rest willing to let them.
R L F
In this section, we look at the neoclassical theory of labour supply, which explains how an individual
allocates his or her available time between paid work and leisure to maximise the level of satisfaction
or utility. This decision will be affected by two factors. The first is the individual’s preferences with
regard to work and leisure (one can, for instance, compare the preferences of a workaholic and a
pleasure-seeker), which are embodied in an indifference curve. This decision will be affected by the
income that the individual will earn by working. Secondly, there is the constraint imposed by the total
income available to that individual. This is called the budget constraint. In essence, the neoclassical
framework captures a fundamental trade-off: the greater someone’s enjoyment of leisure, the less he
works.
. . Basic concepts and important notations

In this section, three key concepts are discussed, namely opportunity cost of leisure, substitution
effect and income effect. From introductory microeconomics, we learnt that the opportunity cost is
the value of the best foregone opportunity, and every time a choice is made, opportunity costs are
incurred (Mohr & Seymore, : ). For example, assuming the per-unit selling prices of apples and
hamburgers are R and R respectively, then the opportunity cost of purchasing a hamburger is to
give up the purchase of five apples, ceteris paribus. Using a similar reasoning, the opportunity cost of
an additional hour of leisure is equal to the hourly wage. For example, given an hourly wage of R ,
the opportunity cost of an additional hour of leisure is R , as the individual needs to reduce one
work hour to enjoy this additional leisure hour, that is, a trade-off between leisure and work exists.
The income effect means that if leisure is regarded as a normal good, as income increases while wage
is held constant, leisure hours would increase while work hours would decrease. Finally, the
substitution effect means that if wage increases while utility is held constant, work hours would
increase while leisure hours would decrease. These two effects will be discussed in greater detail in
section . . and section . . .
Table . summarises the symbols of the important variables to be used throughout section . .
These notations will be explained in detail throughout the section. Symbol L is used in some Labour
Economics textbooks to represent the labour quantity or the number of workers, but in this textbook,
we will use the symbols L and E to distinguish clearly leisure hours and number of workers
respectively (instead of using L for both variables to create unnecessary confusion).
TABLE . Symbols of important variables for the labour supply theory
Symbol Variable
U Utility
L Leisure hours (daily)
H Work hours (daily)
T Total hours available for leisure and work (T = L + H) (daily)
W Market wage in rand (hourly)
V Non-labour income in rand (daily)
Y Total income in rand = labour income (WH) + non-labour income (V) (daily)
MUL Marginal utility of leisure (holding income constant) (= ΔU/ΔL)
MUY Marginal utility of income (holding leisure constant) (= ΔU/ΔY)
MRS Marginal rate of substitution of income for leisure

WRES Reservation wage in rand (hourly)
QS Quantity of labour supplied
. . Worker preferences: indifference curves

The notion that an individual enjoys satisfaction from leisure and income is summarised by the utility
function: U = f(Y, L). U represents the individual’s level of satisfaction, happiness or utility, Y stands for
total income and L means hours spent on leisure.
An indifference curve shows the preferences of an individual for various combinations of income and
leisure time that yield the same level of utility. Figure . is an example of such an indifference curve.
The vertical axis indicates the total income per day that the individual might earn, while the horizontal
axis indicates the amount of leisure time. It can be seen from the figure that the maximum leisure
time is hours: although the person has hours available every day, he or she decides to allocate
hours to leisure and work (work hours will be added to the discussion in section . . ), but spend
the remaining hours on other activities such as sleeping, eating and socialising with friends.
FIGURE . The indifference curve
Each point along the curve indicates the same level of satisfaction or utility for that individual. Going
back to Figure . , all points (e.g. A, B, G and H) on the indifference curve yield the same utility level
of U , despite trade-off taking place (less income is given up to enjoy more leisure as we move
rightwards along the curve).
The important properties of indifference curves are as follows:
(a) Negative slope. As leisure time increases, income decreases owing to fewer hours at work. For
example, movement from point D to point E involves a trade-off: the person spends more time
on leisure (from to hours) and earns less (from R to R ). Alternatively, as leisure time
decreases, income increases due to longer hours spent at work. Referring to the same figure,
moving from point B to point A involves a tradeoff, as the person spends less time on leisure
(dropping from to hours) to earn more (total income increases from R to R ).
(b) Convex to the origin. The curve is bowed inward, i.e. convex. It can also be said that the slope of
the curve is not constant, but varies at the different points along the curve. When income is
relatively high but leisure hours are relatively low, leisure is regarded as scarce, so the individual is
willing to give up a large amount of income to enjoy an additional hour of leisure to maintain the
same utility. This is illustrated by the segment AB: the person gives up R to increase his or her
leisure hours from to . As a result, the slope of the indifference curve along this segment is
relatively steeper in absolute terms ( ). Conversely, when income is relatively low but leisure
hours are relatively high, leisure is regarded as abundant, so the individual would give up a small
amount of income for an additional leisure hour to maintain the same utility. This is represented
by the segment DE: the person only earns R but already enjoys hours of leisure at point D,
so he is willing to give up only R to enjoy an extra leisure hour (point E). Hence, the slope of
the indifference curve along this segment is relatively flatter in absolute terms ( ).
Measuring the slope of indifference curves: chord method vs. tangent method
In Figure . , the slope of the indifference curve is derived by inserting a chord between two coordinates, and the slope
Δy
is equal to Δx
(x and y stand for leisure hours and income respectively). This explains why the slopes of the indifference
curve are – and – along AB and DE respectively (or and in absolute terms), as shown in Figure . .
FIGURE . Slope between two x-coordinates
In more advanced books on mathematical economics, the derivative is defined using the concept of a limit and is usually
dy Δy
written in symbols as = lim . Figure . shows a variety of chords, B G, B G and B G, corresponding to
dx Δx
Δx→0
smaller and smaller “widths” Δx. As the left-hand end points (B , B and B ) get closer to G, the “width” Δx tends to zero.
More significantly, the slope of the chord gets closer to that of the tangent at G. As Δx tends to zero, the slope of the
Δy Δy
chord is equal to that of the tangent. This limit is written as lim .
Δx Δx
Δx→0
FIGURE . Slope at a particular coordinate
Although the tangent method measures the slope more precisely, in this book, the relatively straightforward chord method
will be used to discuss the slope of a non-linear curve, such as the indifference curve in this chapter and the isoquant in
section . . .
(c) Higher indifference curves mean higher utility levels. The leisure-income bundles lying on the
indifference curve that yields a utility of U are preferred to the bundles lying on the curve that
yields a lower utility of U , as shown in Figure . : point K must yield more utility (U ) than point J
(U ), as point K is associated with more leisure ( hours) and more income (R ).
(d) Indifference map. Any individual will have a large number of indifference curves, depending on
how much money is available. As the hourly wage rate of the individual increases, the individual
will move to a higher indifference curve by enjoying more income and more leisure
simultaneously (this is discussed in detail in section . . ). This is indicated by the movement
from curve U to U , from U to U , and from U to U in Figure . . The collection of
indifference curves is called the indifference map.
FIGURE . The indifference map
(e) Indifference curves do not intersect. To see why, consider Figure . , where the two indifference
curves are allowed to intersect. Points P and R lie on the same indifference curve (yielding a utility
of U ), so the person would be indifferent between the leisureincome bundles P and R. Points Q
and R also lie on the same indifference curve (yielding a utility of U ), so the person would be
indifferent between bundles Q and R. As the person is indifferent between P and R (utility equals
U ) and between Q and R (utility equals U ), he should also be indifferent between P and Q. This
implies that U equals U . However, this is not possible: Q is more preferable to P as it is
associated with more leisure and more income. Hence, U must be greater than U .
FIGURE . Indifference curves do not intersect
(f) Different work–leisure preferences. The trade-off between work and leisure will not be the same for
different individuals. A leisure lover will more easily give up a larger amount of income to enjoy
more leisure, compared to a workaholic. Figure . (a) shows that the leisure lover’s indifference
curves are steeper, as he does not mind giving up a larger amount of income (R ) to enjoy an
additional leisure hour. In Figure . (b), the workaholic’s curves are relatively flat, indicating that
this individual will only give up a small amount of income (R ) for an additional leisure hour as
work is too important to the person. Hence, the workaholic’s curve is more elastic.
FIGURE . Preferences of the leisure lover versus the
workaholic
Another factor determining each individual’s preference is the type of work undertaken – a person
who enjoys his work, such as a musician, will require only a very small increase in the wage rate to
sacrifice some leisure time. Conversely, a gold miner working underground may require a much
higher wage increase to decrease leisure time because of the unpleasant nature of the work.
“Daddy,” said the bright child accompanying her father on a round of golf, “why mustn’t the ball go
into the little hole?”
A
For the remainder of section . . , the slope of the indifference curve is examined more thoroughly
by introducing the following three concepts: marginal utility of income (MUY), marginal utility of
leisure (MUL) and marginal rate of substitution (MRS).
Marginal utility of income means the change in utility associated with a one-unit (or R ) increase of
income by holding leisure hours constant. Similarly, marginal utility of leisure means the change in
utility associated with a one-unit (or hour) increase of leisure by holding income constant.
In utility terms, the move from R to T in Figure . can be broken down into
ΔU = R → S + S → T = R → T =
This can be rewritten as:
MUY × ΔY + MUL × ΔL =
FIGURE . Derivation of the marginal rate of substitution
(MRS)
The above equation can be explained by means of the following mathematical example. The
movement from R to S is associated with a decrease of income of R . Now we assume that for each
R change of income, on average, marginal utility of income is equal to utils. Therefore, the total
change of utility from R to S is equal to MUY × ΔY = ( ) × (– ) = – utils; this explains why utility
decreases from utils in U (point R) to utils in U (point S).
The movement from S to T is associated with an increase of leisure of hours (from to ). We also
assume that for each hour change of leisure, on average, the marginal utility of leisure is equal to
utils. Hence, the total change of utility from S to T is equal to MUL × ΔL = ( ) × ( ) = utils; this
explains why utility increases from utils in U (point S) to utils in U (point T). Overall, the
movement from R to T results in the person maintaining the same utility level: the person first
experiences a decrease of utility ( to utils) from R to S, before returning to the original utility
level ( to utils) after moving from S to T. In other words, change of utility (ΔU) is zero from R to
T.
Substituting all values into the above equation:
ΔU = MUY × ΔY + MUL × ΔL =
( ) × (– )+( )×( )=– + =
Rearranging the terms of the equation results in the following:
MUY × ΔY = –MUL × ΔL
ΔY M UL
= −
ΔL M UC
In other words, the absolute value of the slope of an indifference curve, or the marginal rate of
substitution (MRS) of income for leisure, is the ratio of the marginal utilities.
It can be easily proven with simple mathematics that the above relationship holds:
ΔY –75
= =– 15
ΔL 5
M UL 30
= =– 15
M UC 2
As explained above, the indifference curve becomes flatter when the person enjoys a lot of leisure but
earns little income. As a result, the slope of the indifference curve in absolute terms declines as we
move rightwards along the curve, as the person is willing to give up less and less income for an
additional hour of leisure. Therefore, it is equivalent to the assumption of diminishing marginal rate of
substitution.
. . Budget constraint
Even though every individual has many indifference curves, he or she is constrained by the amount of
money that is available and the number of hours in a day. This is indicated by a budget constraint line,
which indicates the various combinations of income and leisure hours that an individual can enjoy,
given a specific wage rate.
Assuming the person spends his or her available hours (T) on work and leisure (T = H + L). Total
income is equal to the sum of labour income and non-labour income: labour income is simply equal
to the product of hourly wage (W) and work hours (H). In other words, the person’s labour income
increases the higher the wage or the longer the hours he or she works In contrast, non-labour income
(V) is independent of hourly wage and work hours; examples of non-labour income include lottery
prizes, interests and dividends earned from investment, receipt of remittances. We can now write the
budget constraint equation as follows:
Y = WH + V
Y = W(T – L) + V
Y = WT – WL + V
Y = (WT + V) – WL
The budget line is illustrated graphically in Figure . . First, the budget line is downward sloping but,
unlike the indifference curves, the budget line is linear. That is, the slope of the budget line is the
same across all points in the line. In fact, the slope of the budget line is equal to –W, meaning the
person foregoes W rand in income for each additional hour of leisure the person enjoys. Note that
the hourly wage is constant for the person, regardless of how many hours he or she works.
FIGURE . The budget constraint
Hours of work are introduced in Figure . . While the hours of leisure increase rightwards along the
x-axis, the hours of work rather increase leftwards along the same axis. For instance, if M hours are
spent on leisure, N hours would be spent on work, and M + N equals T.
At point B, the person spends all available T hours on leisure (T = L) and does not work at all, so his
total income only consists of the non-labour income because labour income equals zero. Going back
to the budget line equation, this means:
Y = (WT + V) – WL = WT + V – WL = (WT – WL) + V = +V=V
On the other hand, at point C, the person spends all available T hours on work (T = H) and does not
enjoy leisure at all. His total income is derived as follows:
Y = (WT + V) – WL = WT + V – WL = WT + V – = WT + V
Note that (WT + V) also represents the y-intercept of the budget line.
The budget line shows an upward parallel shift if non-labour income increases. In Figure . , the
wage rate is R per hour, non-labour income is R and the person allocates hours to leisure and
work. At the extremes, the individual can either work for hours at point C and earn R per day
(R of labour income and R of non-labour income), or alternatively enjoy hours of leisure at
point B and earn R (zero labour income and R non-labour income). Hence, the budget line is
derived as BC.
FIGURE . Upward parallel shift of budget line
Now we assume non-labour income increases from R to R . As a result, the budget line shifts
upwards from BC to EF, with the two lines parallel to each other. The distance between EF and BC
represents the change of non-labour income (an increase of R in this example). If the person
decides not to work at all, his or her total income is equal to R (R labour income and R non-
labour income) at point E; if the person decides to spend all hours working, he or she earns R in
total (R labour income and R non-labour income) at point F.
The budget line could also swivel or rotate upwards along the y-axis as the hourly wage increases. In
Figure . , the same original assumptions apply to the budget line BC as discussed above. Non-
labour now remains unchanged at R but hourly wage increases to R . As a result, the budget
swivels upwards from BC to BG. For both budget lines, if the person decides not to enter the labour
market and spends all available hours on leisure, he or she would earn the same total income of
R in both cases at point B. If the person works hours, he or she would earn R (R × hours
=R labour income, plus R nonlabour income) at point G on the new budget line, compared to
R (R × hours = R labour income, plus R non-labour income) at point C on the old
budget line.
FIGURE . Upward rotation of budget line along the y-axis
as wage increases
. . Utility maximisation
By combining the subjective preferences embodied in the indifference curves and the objective
budget line, the individual’s utility-maximising position can be determined. This is illustrated in Figure
. (the same assumptions apply to budget line BC as discussed before). Maximum utility is achieved
at the point where the indifference curve just touches (i.e. is tangent to) the budget line at point O.
This is called the utility-maximising point. At this point, the person spends hours at work and the
remaining hours to enjoy leisure; total income equals R (R labour income and R non-labour
income) and the person attains a utility of U . Also, the slope of the indifference curve equals the
slope of the budget line. In absolute terms, this means M RS ΔY M UL
= = = W.
ΔL M UY
FIGURE . Utility maximisation
All leisure-income bundles on U and U , while feasible, will not lead to utility maximisation. For
example, points M and N, despite lying on the budget line, only result in a lower utility of U
compared with U at point O. Points J and K, which lie to the left of the budget line, are associated
with lower utility, income and leisure hours compared with point O. In other words, points to the left
of the budget line and points that lie on the budget line but fail to meet the tangency criterion are all
feasible but would not maximise utility. In contrast, the individual would prefer to be at point L where
he or she enjoys hours of leisure and earns R income in total, attaining a higher utility of U .
However, given his or her wage and non-labour income, the person could never achieve the bundle at
L. In other words, all points to the right of the budget line are infeasible.
Figure . provides additional information by showing the respective equilibrium of a workaholic

(point R) and a leisure lover (point S). At equilibrium, the workaholic earns R by spending hours
on work while the leisure lover earns less (R ) by working hours.
FIGURE . Utility maximisation: a workaholic versus a
leisure lover
To conclude, the optimal leisure-income bundle for the worker is given by the point where the
indifference curve is tangent to the budget line. This type of solution is known as an interior solution
as the person is not at either corner of the budget line (that is, at point B, working zero hours, or point
C, spending all available hours to work). Furthermore, since work hours are not zero at equilibrium,
this means the person decides to supply his or her labour services to seek work in the labour market.
. . Income effect with a non-labour income increase

In this section, we examine the impact of an increase in non-labour income. Using the same initial
assumptions as discussed in section . . , Figure . shows that initially the person works hours
per day, earning a total income of R and attaining a utility of U at point P. A R increase in non-
labour income leads to a parallel upward shift of the budget line from BC to EF, and consequently the
person will earn a higher total income and attain a higher utility. The impact of the increase in non-
labour income on work hours could be positive or negative, depending on whether leisure is regarded
as a normal good or inferior good.
The income effect represents the impact of the change in non-labour income (holding wage constant)
on work hours. In Figure . (a), it is assumed that leisure is a normal good, meaning demand for
leisure increases as income increases. The new equilibrium point R lies to the right of the original
equilibrium point P, meaning the person works less (reducing from to hours) for more leisure
(rising from to hours). Nonetheless, the person earns a higher income of R (R labour income
and R non-labour income), despite working shorter hours. To conclude, the income effect on work
hours is negative if leisure is a normal good, that is,
ΔH
ΔY
¯
¯
W¯¯ < , as ΔY and ΔH have opposite signs.
In Figure . (b), it is assumed that leisure is an inferior good, meaning demand for leisure decreases
as income increases. The new equilibrium point R lies to the left of the original equilibrium point P.
The person works longer hours (increasing from to ) but gives up some leisure hours (decreasing
from to ), and earns a higher income of R (R labour income and R non-labour income).
Hence, the income effect on work hours is positive if leisure is an inferior good, that is,ΔH
ΔY
¯
W¯¯ >
¯ .
Both ΔY and ΔH have the same sign.
In both cases in Figure . , there is only one effect taking place, namely income effect. Substitution
effect does not take place simply because wage remains unchanged (refer to the discussion in section
. . that substitution effect happens only when there is a wage change). Alternatively, it can be said
that the total effect is equal to the income effect.
FIGURE . The effect of an increase in non-labour income on work

hours
While the derivation of the individual labour supply curve will be the focus of section . . , we
assume at this stage that it is upward sloping. The increase in non-labour income would lead to a
leftward shift of the supply curve if leisure is a normal good, but a rightward shift of the curve if
leisure is an inferior good, as shown in Figure . . The results of the latter figure are derived by
using the information at the two equilibrium points (P and R) in Figure . . At both points, wage
equals R .
FIGURE . Shift of individual labour supply curve
due to increase in non-labour income
. . Substitution and income effects with a wage increase

Now we examine the impact of the increase of hourly wage from R to R (refer to the discussion
relating to Figure . ) by assuming leisure is a normal good. Figure . shows the person initially
works hours per day, earning a total income of R and attaining a utility of U at the equilibrium
point P. The increase in wage causes the budget line to swivel (or rotate) along the y-axis from BC to
BG. Whilst the person would certainly earn a higher total income and attain a higher utility at the new
equilibrium (point R, to be discussed below), the impact of wage increase on work hours could be
positive or negative, depending on whether the substitution effect or income effect is more dominant.
In Figure . (a), at equilibrium, the wage increase results in the person moving from point P to
point R. The person works longer hours (from to ) but enjoys shorter leisure hours (from to ) at
point R. Also, the person’s utility increases from U to U , and total income increases from R to
R , which consists of R labour income (R × hours) and R non-labour income. In contrast,
in Figure . (b), the person also moves from point P to point R as a result of wage increase; at point
R, the person works shorter hours (from to ) but his or her leisure hours increase (from to ),
while a higher utility of U is attained. Nonetheless, his total income rises from R to R , which
comprises R labour income (R × hours) and R non-labour income.
FIGURE . The effect of an increase in wage on work hours

Comparing Figure . (a) and Figure . (b), the results suggest there is an ambiguous relationship
between wage rate and work hours, and this has to do with the relative strength of the income and
substitution effects. While the movement from P to R represents the total effect of wage increase on
work hours, it helps to think of this movement as a two-step procedure to decompose the impact of
wage increase into income and substitution effects.
To distinguish the income effect, suppose a dotted budget line (DD in Figure . ) that is parallel to
the original budget line (BC) but is tangent to the new, highly-lying indifference curve (U ) is drawn.
This dotted line generates a tangency point Q. The movement from P to Q represents the income
effect, and essentially answers the following question: what would the change in work hours be if the
person reaches a new indifference curve with an increase in non-labour income instead of an
increase in hourly wage? In other words, although in actual fact wage increases, the movement from P
to Q assumes that the wage is held constant, but the person attains a higher utility of U by assuming
he or she rather enjoys an increase in non-labour income.
As we assume leisure is a normal good, the income effect is negative, i.e. ΔH

¯
¯
W¯¯ < . In Figure
ΔY
. (a), had non-labour income increased by R (the distance between DD and BC) and the wage
been held constant at R , the person would have attained a higher utility of U at point Q.
Alternatively, this R represents the amount of non-labour income increase the person would need
to be equally well-off as the person is after the wage increase (attaining U ), had there been no
change in wage rate. The income effect would cause work hours to decrease from to but leisure
hours to increase from to . Also, at point Q, total income equals to R , which consists of R
labour income (R × hours) and R non-labour income (the original R plus the additional R
to enable the person to attain U at point Q).
A similar reasoning applies to the isolation of income effect in Figure . (b). Had non-labour
income increased by R (the distance between DD and BC) and the wage been held constant at R ,
the person would have attained a higher utility of U at point Q. The income effect would result in
work hours dropping from to but leisure hours rising from to . Also, at point Q, total income
equals R , which is the sum of the R labour income (R × hours) and R non-labour income
(the original R plus the additional R to enable the person to attain U at point Q).
The movement from Q to R represents the substitution effect, which illustrates what happens to the
individual’s leisure-income bundle as the wage increases, holding utility constant. By moving along
the indifference curve (U in Figure . ), the person’s utility or “real income” is held unchanged. The
substitution effect is always positive, since an increase in wage means an increase in opportunity cost
of leisure. Initially, the opportunity cost of enjoying an additional hour of leisure is R . As wage
increases to R , this opportunity cost becomes more expensive. With the wage increase, the person
would devote more time working to maintain the same utility. In equation terms, the substitution
effect is represented as ΔH
ΔW
U¯
¯¯ > (utility is held constant at U in Figure . , for example). Hence,
the substitution and income effects work in opposite directions, assuming leisure is a normal good.
Going back to Figure . (a), the substitution effect causes the person to increase his or her work
hours from to but leisure hours to decrease from to ; in Figure . (b), this effect results in
the person’s work hours rising from to but leisure hours dropping from to . In both cases,
utility remains unchanged at U at both points Q and R.
The length of the arrow in the figure implies the strength of each effect: the longer the arrow, the
stronger the effect. In Figure . (a), as the (positive) substitution effect dominates the (negative)
income effect, the increase in wage rate increases work hours; in Figure . (b), the opposite
happens as the (negative) income effect dominates the (positive) substitution effect, and hence, the
increase in wage rate decreases work hours.
Finally, the movement from point P to point R in Figure . leads to the movement along the
individual labour supply curve, and this is illustrated in Figure . . Figure . (a) is derived from
Figure . (a): an increase in hourly wage rate from R to R results in an increase in work hours
from to . The labour supply curve is upward sloping due to the positive relationship between
wage rate and work hours. In contrast, Figure . (b) is derived from Figure . (b): a negative or
inverse relationship exists between the two variables because an increase in hourly wage from R to
R leads to a decline in work hours from to .
FIGURE . Movement along the
individual labour supply curve due to
increase in wage
Finally, it is also possible that the substitution effect and income effect are equally strong as a result of
a wage increase; the two effects cancel each other out, and thus the total effect is zero, and there is
no change in work hours and leisure hours. However, this unique case falls beyond the scope of this
chapter and is not discussed further.
. . Reservation wage
The analysis so far assumes that the person works non-zero hours (i.e. he or she is part of the labour
force) both before and after the change in non-labour income or wage. But is it possible that the
person decides not to work and moves out of the labour force?
To answer this question, consider Figure . . Once again we assume that initially the hourly wage
offered by the market is R , and the person decides to work hours at equilibrium point P, attaining
a utility of U and earning R in total. Suppose the person’s reservation wage is R . The reservation
wage represents the wage below which the worker would refuse to accept employment (Ehrenberg &
Smith, : ). A person will supply his or her services in the labour market (i.e. H > , or interior
solution) if market wage is greater than reservation wage, but the person will not work at all (i.e. H =
, or corner solution) if market wage is equal to or smaller than reservation wage.
Looking at Figure . , point P represents an interior solution (H = ) because market wage exceeds
reservation wage (R = W > WRES = R ). If market wage equals reservation wage (W = WRES), or R
in this example, the budget line becomes BK. The person would maximise utility at point B, where the
budget line is tangent to the indifference curve U . This equilibrium stands for the corner solution, as it
occurs at the corner of the budget line where L = T. The person opts to work zero hours and hence
chooses not to participate in the labour force. His or her total income is R , which consists of non-
labour income only. If the person chooses to work, let’s say, hours, as indicated at point O in the
figure, even though he or she earns a higher total income of R in total (R × hours = R labour
income, plus R nonlabour income), the person attains a lower utility of U . The budget line BK is not
even tangent to U . Therefore, the person would be better off at B by not working.
FIGURE . Labour supply equilibrium when market wage

equals reservation wage
The above discussion concluded that work hours equal zero when market wage equals reservation wage. What about a
situation where market wage is below the reservation wage?
Let’s look at Figure . . Assuming reservation wage is still R but now market wage drops from R to R . Hence, the
budget line becomes flatter (BL). At this lower market wage, the person prefers not to work to such an extent that the
budget line is tangent to indifference curve U at point S, and the person would want to enjoy hours but work – hours
(note that the sum of these hours is still equal to , i.e. L + H = + (– ) = = T).
FIGURE . Labour supply equilibrium when market wage exceeds

reservation wage
While this equilibrium point S makes sense from a theoretical point of view (the tangency condition), it is not possible
mathematically for someone to work negative hours. Therefore, the person would reluctantly settle at the corner solution
at point B, working zero hours but enjoying hours of leisure. Utility is at a lower level of U compared to U at point S,
simply because the person is forced to settle at point B by “working” extra hours (from – to ). Note that at point B, the
budget line BL is not tangent to the indifference curve U . In fact, the indifference curve is relatively steeper.
. . The individual labour supply curve

It is now possible to derive the overall labour supply curve for an individual, i.e. to determine what
happens to the labour supply (for instance the number of hours worked) as wages increase or
decrease. (This of course assumes that an individual can choose how many hours he or she wants to
work, but even with fixed work hours, an individual can often choose to take a second job, work
overtime, etc.)
This is illustrated diagrammatically in Figure . . Each higher wage rate (R , R , etc.) results in a
steeper budget line. The tangencies of these budget lines with the indifference curves result in a
series of utility-maximising points (U , U , etc.). When market wage equals reservation wage (R ), the
person is working zero hours. As the wage increases, the individual may prefer to work longer hours
(e.g. moving from point e to e ). However, at some point he chooses more leisure time as wages
increase, and moves from e to e – a wage increase at this point reduces rather than increases hours
of work. Table . summarises what happens at each equilibrium point as wage increases.
FIGURE . Utility-maximising leisure-income bundles at different wage rates
Source Paragraph
TABLE . Work hours, leisure hours, income, utility and total effect at different wage rates
Wage Work hours Leisure hours Labour income Non-labour income Total income (V + Equilibrium Utility Total
(W) (H) (L) (WH) (V) WH) (U) effect
5 0 16 0 50 50 e0 U0 N/A
10 8 8 80 50 130 e1 U1 Positive
20 10 6 200 50 250 e2 U2 Positive
30 12 4 360 50 410 e3 U3 Positive
40 14 2 560 50 610 e4 U4 Positive
50 13 3 650 50 700 e5 U5 Negative

If economists and statisticians had deliberately set out to confuse and perplex our rulers, they could
hardly have been more successful.
J J
Now hours of work can be plotted against wages using the information in Figure . . This is done in
Figure . . If the wage is equal to or below the reservation wage, the individual opts out of the
labour force and decides not to work, that is, the vertical segment e of the individual labour supply
curve. As the wage increases from R to R , the individual chooses to work longer hours. This means
substitution effect dominates and the individual labour supply curve is upward sloping along the
segment e e . As the wage exceeds R , the individual chooses to work fewer hours, meaning income
effect dominates and the individual labour supply curve is downward sloping (or backward bending)
along the segment e e .
FIGURE . Derivation of individual labour supply curve
The same model applies to the household – initially a wife might not work (i.e. outside the labour
market), but might decide to work if the wages she can earn increase. She thus gives up some “leisure
time”. Alternatively, as wages increase, the family might decide to engage in less time-intensive leisure
activities such as playing golf and rather play squash (this would also be the substitution effect, as it
allows more time for work). However, as wages increase and thus also the financial ability to buy and
consume goods (including the ability to go on holidays, watch movies, etc.), more time is needed to
enjoy the goods, and the hours of work of the household might fall – higher wages at this point lead
to reduced work hours (the income effect).
Fewer hours – more jobs?
The passage of the Basic Conditions of Employment Act in resulted in an intensive debate over whether a reduction
in the working week leads to more employment. It is argued in sections . . and . . that this is unlikely to be the
case, and that it might even cause an increase in unemployment because of the resultant higher labour cost.
. . The elasticity of labour supply

The elasticity of labour supply measures the responsiveness of work hours to changes in wage rate, or
the percentage change in work hours associated with a one-percentage change in wage rate. In
equation terms, it is equal to:
%ΔH ΔH /H W
ΔH
= = ⋅
%ΔW ΔW /W ΔW H
The elasticity is positive when the substitution effect dominates but negative when the income effect
dominates. The greater the absolute value of the elasticity, the more responsive work hours are to
changes in wage rate. Also, if the elasticity is greater than (smaller than) one in absolute terms, the
labour supply curve is said to be elastic (inelastic).
Table . presents a hypothetical example on the elasticities of labour supply of four individuals. For
example, person A initially works hours at an hourly wage of R , but increases the work hours to
after enjoying a wage increase to R . Hence, the elasticity of labour supply of this person is equal to
. (positive, inelastic) and is calculated as follows:
TABLE . The elasticity of labour supply – a hypothetical example
Person A Person B Person C Person D

W1 10 10 10 10
W2 15 15 15 15
H1 8 8 8 8
H2 10 14 6 3
Elasticity of labour supply 0.50 1.50 –0.50 –1.25

Dominant effect Substitution Substitution Income Income
Interpretation Positive Positive Negative Negative
Inelastic Elastic Inelastic Elastic
ΔH W (10−8) 10 2 10 20
⋅ = ⋅ = = = = 0.5
ΔW H (15−10) 8 5 8 40
Using a similar approach, the elasticities of the other three persons could be derived, and the results
are presented in the last two rows of the table.
. . The market supply of labour

When it comes to the market supply of labour, all the individual (i.e. backward-bending) supply curves
need to be horizontally added together. However, other factors then also come into play: as the wage
rates increase, individuals who have not worked before might choose to enter the labour market or
people might be attracted away from other labour markets to a labour market that pays higher
wages. All these factors, together with the fact that different persons enter the labour market at
different wage rates, result in the backward-bending part of the labour supply curve disappearing.
Therefore, for the labour market overall, higher wages normally result in a higher number of workers
supplied. In other words, the market labour supply curve is strictly upward sloping. Note that for the
individual labour supply curve, the x-axis variable is work hours (H) but in the market labour supply
curve, the x-axis variable is rather the number of workers or quantity of labour supplied (QS).
For the labour market as a whole, the labour supply curve is therefore positively sloped as indicated
by the line S (for “supply”) in Figure . . Assume a wage rate of W , the quantity of labour supplied
will be Q . If wage increases from W to W , there will be an upward movement along the supply
curve S , and the quantity of labour supplied will increase from Q to Q . Conversely, if wage
decreases from W to W , the quantity of labour supplied will decrease from Q to Q , due to a
downward movement along the curve S .
FIGURE . The market labour supply curve
The magnitude of the changes in supply can vary. This will depend on the responsiveness (elasticity)
of the supply of labour to changes in wages (the gradient of the curve). For the individual, this was
explained above with regard to the workaholic and the leisure lover. This responsiveness is called
elasticity; the more elastic the supply of labour, the more the supply of labour will increase with an
increase in the wage rate. The concept of elasticity is discussed in greater depth in Chapter .
. . Factors affecting market labour supply

In the previous sections, it was discussed that population size, fertility rates, mortality rates and net
migration are crucial demographic variables affecting the market labour supply. However, there are
numerous other factors that also have an impact on labour supply, and they are summarised in Table
. . Graphically, if the market labour supply increases or shifts rightwards from S to S in Figure
. , and if the wage is unchanged at W , the quantity of labour supplied increases from Q to Q .
Conversely, if the market labour supply decreases or shifts leftwards from S to S , at the same
constant wage of W , the quantity of labour supplied decreases from Q to Q .
Looking at some of the factors in Table . in greater detail, for example, it was mentioned earlier
that the abolition of the apartheid-period discriminatory legislations and the implementation of new
legislations (such as the Employment Equity Act) to promote the previously disadvantaged groups
could have a positive impact on market labour supply. For instance, females and Africans could feel
more optimistic about seeking work.
Due to the structural change of the South African economy, those with higher education and skills
level would be more likely to be employed. Hence, if the cost of education and training were lower,
this would motivate more people to pursue further studies and training, and subsequently there
would be an increase in market labour supply.
The market labour supply could also increase if the cost of moving to prospective work places
becomes cheaper (it is possible that some people give up hope of seeking work, knowing they could
not afford the cost to move to locations associated with more work opportunities), the employer
contributes a greater amount towards the employee’s retirement pension funds, and the danger
involved in the work decreases. The spatial legacy of apartheid remains a labour policy issue; many
workers face daily commutes of upwards of hours, which not only has a severe impact on their
quality of life, but also imposes a significant transport cost. Policy innovation along the lines of
affordable inter-modal public transport could play a significant role in increasing the accessibility of
labour supply, among others.
There exists a relationship between labour supply and business cycles: according to the added worker
effect, during an economic recession, if the main breadwinner becomes unemployed, the total
household income decreases. Some household members who used to be inactive might enter the
labour market, hoping to find work and earn income to support the family. For instance, if the
husband is unemployed, his wife and children who used to be inactive may now decide to look for
part-time work. Hence, the added worker effect implies that the market labour supply rises during
recession and falls during expansion.
Conversely, according to the discouraged worker effect, many unemployed persons find it highly
unlikely that they will find employment during an economic recession and eventually give up hope
and leave the labour force. They decide to reenter the labour market only after there are signs of
economic growth. Hence, the discouraged worker effect suggests that the market labour supply
increases during an economic boom but falls during a recession.
TABLE . Factors affecting market labour supply – a summary

Determinant Change Effect on market labour Correct description of the
supply curve effect
Market wage Increase Upward movement along the An increase in quantity of
supply curve labour supplied
Decrease Downward movement along the A decrease in quantity of
supply curve labour supplied
Non-labour income (if leisure is a normal Increase Leftward shift of the supply A decrease in labour supply
good) curve
Decrease Rightward shift of the supply An increase in labour supply
curve
Non-labour income (if leisure is an Increase Rightward shift of the supply An increase in labour supply
inferior good) curve
Decrease Leftward shift of the supply A decrease in labour supply
curve
Fertility rate Increase Rightward shift of the supply An increase in labour supply
curve
curve
Mortality rate Increase Leftward shift of the supply A decrease in labour supply
curve
curve
Net migration (immigration – emigration) Increase Rightward shift of the supply An increase in labour supply
curve
curve
Government’s effort to promote the Increase Rightward shift of the supply An increase in labour supply
disadvantaged groups curve
curve
Cost of education and training Increase Leftward shift of the supply A decrease in labour supply
curve
curve
Cost of moving to prospective work place Increase Leftward shift of the supply A decrease in labour supply
curve
curve
Transport cost to and from work Increase Leftward shift of the supply A decrease in labour supply
curve
curve
Employers’ contributions to retirement Increase Rightward shift of the supply An increase in labour supply
pension fund curve
curve
Danger involved in work Increase Leftward shift of the supply A decrease in labour supply
curve
curve
Business cycles (added worker effect) Economic Leftward shift of the supply A decrease in labour supply
boom curve
Economic Rightward shift of the supply An increase in labour supply
recession curve
Determinant Change Effect on market labour Correct description of the
supply curve effect
Business cycles (discouraged worker Economic Rightward shift of the supply An increase in labour supply
effect) boom curve
Economic Leftward shift of the supply A decrease in labour supply
recession curve
SUMMARY
Utility is maximised when budget line is tangent to the indifference curve. The slopes of the two curves are the same at
equilibrium.
An increase in non-labour income results in a decrease in work hours if leisure is a normal good, but an increase in
work hours if leisure is an inferior good. The income effect could be positive or negative, but there is no substitution
effect.
Assuming leisure is a normal good, an increase in wage rate could lead to either an increase or decrease in work
hours, depending on whether the positive substitution effect or negative income effect is more dominant.
If work hours are positive at equilibrium, this is known as an interior solution. If work hours are zero at equilibrium, this
is known as a corner solution.
The interior solution takes place when market wage exceeds reservation wage, while the corner solution happens
when market wage is equal to or smaller than reservation wage.
The individual labour-supply curve is backward-bending while the market labour-supply curve is upward sloping.
KEY CONCEPTS
added worker effect

backward-bending individual labour-supply curve
budget constraint
corner solution
economically active population (EAP)
dependency ratio
discouraged worker effect
discouraged work-seekers
elasticity of labour supply
fertility rate
HIV and AIDS
hours of leisure
hours of work
immigration and emigration
income effect
indifference curve
interior solution
labour force (LF)
labour force participation rate (LFPR)
labour income
life expectancy
marginal rate of substitution (MRS)
migrant labour
mortality rate
natural rate of population increase
non-labour income
opportunity cost
population pyramid
reservation wage
substitution effect
theory of demographic transition
total fertility rate (TFR)
total effect
utility maximisation
upward-sloping market labour-supply curve
wage
working-age population
FOR STUDENTS
. What has happened to the labour force participation rate of women in South Africa in recent years and what
are some possible reasons for this development?
. Discuss the various stages that a population might be expected to pass through in terms of demographic
developments.
. Discuss the concept of an indifference curve and indicate some of its most important characteristics.
. Given the following leisure-income bundles, use the chord method to calculate the marginal rate of
substitution (MRS), given that the marginal utility of income (MUY) and marginal utility of leisure (MUL) are
two and respectively: bundle A: hours of leisure, income equals R , utility equals utils; bundle B:
hours of leisure, income equals R , utility equals utils; bundle C: hours of leisure, income equals R ,
utility equals utils.
. Michael receives R per day as interest on inheritance. His wage rate is R per hour, and he can work a
maximum of hours per day at his job. Draw his budget constraint.
. Andy insists all humans must spend one-third of their lives sleeping, that is, spending hours per day
sleeping. The remaining hours are allocated to either work or leisure. His hourly wage is R . He does not earn
any non-labour income.
. At equilibrium, he spends hours on leisure. Clearly indicate this equilibrium on the graph. Also, show the
indifference curve, leisure hours, work hours and total income on the same graph.
. Assume Andy won a lottery prize of R income every day. Consequently, the time he spends on leisure
increases to hours. Clearly indicate this equilibrium on the same graph. Also, show the indifference
curve, leisure hours, work hours and total income on the same graph. Also, use arrows to indicate the
substitution effect (if any) and income effect (if any).
. Zoliswa spends hours each day on leisure and work. Her hourly wage is R and she currently earns a daily
non-labour income of R .
. At equilibrium, she works hours. Clearly indicate this equilibrium on the graph. Also, show the
indifference curve, leisure hours, work hours and total income on the same graph.
. Assume Zoliswa enjoys an hourly wage increase of R . Consequently, the time she spends working
increases to hours. Clearly indicate this equilibrium on the same graph. Also, show the indifference
curve, leisure hours, work hours and total income on the same graph. Also, use arrows to indicate the
substitution effect (if any) and income effect (if any).
. Explain, with the aid of a figure, how the labour supply of a person would change if the initially offered wage is
above the reservation wage, but then it drops to exactly the same as the reservation wage, and eventually
becomes lower than the reservation wage.
. Use the information from the following table to calculate the elasticity of labour supply of the four individuals
as hourly wage doubles from R to R . Show all calculations and interpret the answers.
Person A Person B Person C Person D

Original wage 20 20 20 20
New wage 40 40 40 40
Original work hours 6 6 6 6
New work hours 14 10 5 2
. Use a graphic illustration to describe briefly what the influence of each of the following would be on the
market supply of labour:
. An increase in immigrants
. A reduction in wage rates
. More women entering the labour market
. More students studying full-time
SUGGESTED READING*
Arndt, C. & Lewis, J.D. . The macro-economic implications of HIV/AIDS in South Africa: a preliminary
assessment. South African Journal of Economics, ( ): December.
Bhorat, H. . Labour market challenges in the post-apartheid South Africa. South African Journal of Economics,
( ): December.
Casale, D.M. & Posel, D.R. . The continued feminisation of the labour force in South Africa: an analysis of
recent data and trends. South African Journal of Economics, ( ): March.
Festus, L., Kasongo, A., Moses, M. & Yu, D. . The South African labour market, – . Development
Southern Africa, ( ): July.
Oosthuizen, M. . The post-apartheid labour market – . DPRU Working Paper / . Cape Town:
Development Policy Research Unit.
Van Aardt, C. . The impact of HIV/AIDS on the South African labour market from a critical perspective. S A
Journal of Demography, ( ).
World Bank. . South Africa economic update: jobs and South Africa’s changing demographics. Washington DC:
World Bank.
Non-economists are advised to read the comments in the preface before they become too intimidated by
the graphs that economists so love to use. It is not essential for the reader to understand the theoretical
section of the chapter in order to comprehend the issues discussed in the rest of the chapter. Nevertheless,
section . will help the reader understand better how the labour market as a whole operates.
The demand for labour

. Introduction
. Employment in South Africa
. The informal sector
. Deeper insight: deriving the demand for labour
Lord Finchley tried to mend the ’lectric light himself; it struck him dead and serve him right: it is the business of the
wealthy man to give employment to the artisan.
H B
. INTRODUCTION
The demand for labour is a derived demand, which means that it is dependent on the consumer
demand for the good produced by that labour or for the service provided by that labour. The demand
for labour will thus be influenced by the price of labour (which has an impact on the price of the
product or service) and by increases or decreases in the demand for the product or service produced
by the labour. It will also be influenced by other factors such as the market value of the product and
by how much of the product is produced by the labour, i.e. how productive the specific worker is.
This theoretical relationship between the demand for labour and the price of labour – the wage rate –
is discussed in this chapter (section . ). This relationship can change over time because, if the wage
rate increases sharply, an employer can start replacing labour with machinery. The short-run and
long-run demands for labour therefore differ. The more theoretical concepts are discussed at the end
of the chapter (see section . ) because they are relatively difficult for non-economists to grasp. As
stated previously (section . ), economists are advised to start with the last section of the chapter as
it lays a good foundation for the rest of the chapter.
The demand for labour is not only responsive to wage, but also to the price of capital and economic
growth. Hence, throughout the chapter, various elasticities in relation to labour demand are looked at,
namely employment elasticity of economic growth (section . . ), elasticity of labour demand
(section . . ) and elasticity of substitution (section . . ). Also, while section . . of Chapter
discussed factors affecting the market labour supply, section . . of this chapter examines the
factors affecting the market labour demand.
It is not the employer who pays wages; he only handles the money. It is the product that pays wages.
H F
While the majority of workers in South Africa work in the formal sector, there is another source of
work for persons in the labour market – the informal or unrecorded sector, often referred to as a
sponge of the economy. This is because persons who are unable to find a job in the formal sense of
the word, more often than not turn to the informal sector to earn a living. Section . will specifically
focus on the informal sector.
This chapter starts with an overview of employment in South Africa (in effect the employment position
in the formal and informal sectors) and by outlining some of the characteristics of employment in
South Africa. First, however, it is necessary to discuss how employment is measured.
. . The measurement of employment

Employment data are important for a number of reasons. First and foremost, it would not be possible
to determine what factors influence the demand for labour without reliable statistics on employment.
Employment data are also used to calculate productivity, per capita earnings and several other
economic indicators.
However, employment data in South Africa are often poor and selective. In addition, the data are not
comparable over time, owing to various definitional and other changes, such as the exclusion of the
former TBVC territories until . In the following sections, various sources of data will be discussed.
Population census and community survey

Until , a census was conducted by Statistics South Africa every five years or so (in , , ,
and ). There was a -year gap between the census and the census, while the next
census is expected to take place in . To fill the gap between the censuses, Statistics South Africa
conducted the Community Survey (CS) in and . Both the census and CS yield abundant
information on the demographic, education and health characteristics of the population, as well as
nonincome welfare of the households, such as ownership of assets (television, computer, refrigerator
and motor car) and access to facilities and services (water, fuel source for cooking and lighting, refuse
removal, sanitation).
While there is one section in both the census and CS questionnaires devoted exclusively to capturing
labour market status and activities of the population aged years or above, the questions were not
asked as comprehensively as in the household labour surveys (to be discussed below), and therefore,
the employment aggregates derived by the census and CS may not be as precise and reliable. For
instance, in a comparison of the manufacturing census and the population census of the same
year, Wittenberg ( : ) found that the population census had counted fewer
manufacturing workers. Based on a detailed analysis, he concluded that the manufacturing census
figure was probably the more correct one. He also found that too few mineworkers and possibly too
few construction and domestic workers were recorded in the population census. The reasons for
understating some sections of the population are related to difficulties in counting homeless people,
single-person households and other fairly mobile sections of the population. Furthermore, the
population census (even for the most recently conducted one in ) does not measure informal
sector employment properly, because it does not ask detailed enough questions to determine
informal sector activities.
Sample surveys among businesses
Statistics South Africa uses the Quarterly Employment Statistics (QES) to measure employment among
businesses in the formal sector in South Africa. Prior to June , there was the Survey of Employment
and Earnings (SEE) and before that the Survey of Total Employment and Earnings. The main difference
between the QES and the SEE is that the QES covers more employers, and in particular provides better
coverage of small firms (all firms registered for income tax are supposed to be covered). However, the
QES excludes working proprietors and own-account workers and all other employees who are not
remunerated through the firm’s payroll, such as commission-only workers. The QES, therefore, can
also not be compared with the Labour Force Survey (LFS), which is discussed in the next section.
However, the SEE before is also subject to significant limitations, and it underestimated total
employment owing to the exclusion of agricultural and domestic workers; the high nonresponse rates,
for example in the retail trade; and incomplete coverage of, for example, parts of construction, private
transport and communication, social and personal services (ILO Review, : ; Statistics South
Africa Statistical Release, b; Altman, ). There is thus a break in the employment series before
and after , which has resulted in a number of incorrect conclusions being made about a falling
wage share, falling manufacturing employment and a weakening employment–growth relationship
(see Altman, ).
Furthermore, it is difficult to measure atypical forms of employment in these surveys (see section
. . ). Employment by small enterprises and casual, subcontracted and other irregular forms of
employment are often not recorded. This is a serious shortcoming, because more and more
enterprises are subcontracting or outsourcing large sections of their work.
Blessed is he who has found his work. Let him ask no other blessedness.
T C
Household labour surveys

Another source of data on employment is the household labour surveys, which primarily focus on
thoroughly capturing the labour market activities of the population, and are undertaken on a regular
basis by Statistics South Africa. There have been a number of changes to the surveys over time, the
most important of which resulted in different nomenclature being used to indicate the use of different
survey methodologies.
The October Household Survey (OHS)
The principal vehicle for collecting labour market information for the whole country over the period
– was the annual October Household Survey (OHS). However, the OHS also collected
information from respondents about a diverse range of issues relating to births and deaths, health,
crime, education and training initiatives, as well as the services and amenities available to the
dwelling(s) in which households lived and so on.
With regard to employment in particular, Altman ( ) indicates that the OHS figures for the period
– need to be questioned. She maintains that the series for agriculture, mining and
community, social and personal services needs to be approached with much circumspection. In these
years, it was not possible to estimate informal sector employment precisely, as employees were not
asked to declare if they worked in the formal or informal sector. Furthermore, unpaid family workers
were not included as workers in the OHS (which led to an undercount of more than ). The
inclusion of subsistence agriculture would also have resulted in serious variations in employment
estimates (between and million in different years). Since , the latest survey (the QLFS)
has excluded subsistence agriculture and emphasised commercial agriculture more when capturing
employment.
The Labour Force Survey (LFS)
The first LFS was conducted in and since then it has been undertaken on a six-monthly basis in
March and September each year. The LFS is more focused on labour issues than its predecessor (the
OHS) since the bulk of the non-labour questions were channelled to the General Household Survey
(GHS), which was introduced in . The LFS also asked many more probing questions regarding the
informal sector and subsistence farming. This changeover from the OHS to the LFS probably resulted
in the jump in the number of self-employed workers in the informal sector.
The Quarterly Labour Force Survey (QLFS)
The decision to redesign all aspects of the LFS emanated from criticisms by data users and these are
documented in the report written by International Monetary Fund (IMF) consultants in June .
These criticisms related to the scope, coverage, timeliness and frequency of the survey (Statistics
South Africa, a). In addressing these issues, Statistics South Africa decided to embark on a
quarterly cycle for the collection of labour market information from .
Non-market production activities in the QLFS: The international guidelines require that non-market
production activities be included as employment only if they make a substantial contribution to the
total consumption of the household. Because of the difficulty of establishing robust criteria for their
inclusion as employment in the QLFS, individuals who are engaged in non-market production
activities, such as subsistence farming, fetching wood and collecting water, making clay pots, etc., for
the use or consumption of their own household are thus routed to questions that determine whether
they are either unemployed or economically inactive (Statistics South Africa, ). The two
components are then identified as follows:
1. The unemployed (i.e. persons engaged only in non-market production activities, who actively
sought work and were available for work in the reference period).
2. The not economically active (i.e. persons engaged in non-market production activities who did not
engage in any type of job-search activity during the reference period).
National Income Dynamics Study
The National Income Dynamics Study (NIDS) is the first national panel study in South Africa. Since its inception in ,
five waves of the survey were undertaken every two years by the Southern Africa Labour and Development Research
Unit (SALDRU) based at the University of Cape Town. The five broad areas covered by NIDS are as follows: income and
expenditure; the assets owned by the household and the household’s access to services; educational attainment level
and health status; labour market status of the individuals; and membership of community groups. While it falls beyond
the scope of this chapter, one advantage of the NIDS data is that it enables the examination of labour market dynamics
of the working-age population. For instance it could identify people who were employed or unemployed in all waves and
also those whose labour market statuses changed across the waves (for example, someone could be employed in ,
unemployed in , inactive in , before being employed again in and ).
Difficulties in comparing surveys
Yu ( ) refers to a number of studies that discuss the comparability of the OHS and LFS. These
include, among others, the oversampling of informal sector workers in , the overestimation of the
earnings of the self-employed in the OHSs, and the continuous improvement of the questionnaire by
Statistics South Africa. In general, Yu’s ( ) conclusion is that the OHS and LFS are incomparable in
many respects, given changes in the sampling frame, inconsistencies in questionnaire design, coding
errors, changes in methodology to capture employment status and outliers in wage earnings data. In
addition to the possible oversampling of the informal sector, the measurement of the informal sector
also changed over time, which has had an impact on changes in employment and unemployment
rates over time, as outlined in section . . . In particular, this is true if Casale and Posel ( : ) are
correct in estimating that the LFS may be measuring about . million more people as working in
subsistence agriculture and the informal sector than the OHS as a result of the amended
questionnaires. This could artificially lower the unemployment figure in the LFS by about two
percentage points.
With regard to the differences between the LFS and the QLFS, the historical information of the LFS
was revised to make the LFS estimates from to comparable with the QLFS data starting in
. However, there were some qualitative improvements to the QLFS questionnaire, which would
probably not be fully captured by the historical revision.
In addition, Yu ( ) points out that the labour market status is defined very differently in the QLFS
from the LFS, particularly under the broad definition as a result of a different approach to distinguish
discouraged work-seekers in the QLFS.
The difference between the various surveys leads Yu ( ) to a very important conclusion, namely
that if trends are to be correctly interpreted from the available labour data, selected household
surveys cannot simply be compared with each other and conclusions drawn about the “trends” in the
labour market for the entire period between surveys. It is argued that such a methodology is
imperfect and hence the results should be interpreted with caution.
SUMMARY
The level of employment in South Africa is difficult to measure owing to problems in measuring informal activities, small
enterprise employment, irregular forms of employment and employment in some important service sectors.
. EMPLOYMENT IN SOUTH AFRICA
. . Total employment over time

Before the employment aggregates are looked at, it is important to know how someone is classified
as employed. Although the questions used to identify the employed were not asked in a consistent
manner in the labour surveys, there has been an increasing effort to capture low-income employment.
In OHS , the respondents were asked what they did most during the last seven days, whereas in
OHS respondents were asked whether during the past seven days they had actually worked for
pay, profit or family gain. Note that “working full-time” and “working part-time” were allowed as
answers. The and surveys added a third alternative, “casual work”, and in “seasonal
worker” was also added as an option.
From onwards, respondents were asked if they had engaged in any one of numerous specific
(and mostly low-income) activities such as “guarding cars”, “making things for sale”, “doing any work
as a domestic worker” and “working on the household plot” (Yu, : ; Yu, : ). It is therefore
clear that increased effort was made to capture informal and lowincome employment and an
implication of this increased effort may well be apparent fluctuations in the number of employed
between surveys.
The definition of employment also became rather wider since – as long as the respondent
reported to have worked at least one hour in the last seven days, he or she was immediately defined
as employed. Those who were absent from work at the time of the survey but claimed they would
definitely want to work were also distinguished as employed.
South Africa’s employment rate, which represents the percentage of the working-age population that
is working, is very low by international standards. The line chart in Figure . shows that it was as low
as . % in , hovered between % and % in – (peaking at . % in just before
the global economic recession took place), before declining to – %. In countries in Latin America,
Eastern Europe and East Asia at similar levels of development, the proportion is about % or higher,
with China at more than % (CDE, a: ).
While not shown in Figure . , total employment (including the informal sector, but excluding private
households) increased at a modest rate between and , namely, at an average annual rate of
increase of . %. However, there is some doubt about the strict comparability of data in the two years
at the start and end of the period, for instance the data were based on the census, whereas the
data were based on the household survey (the October Household Survey). There are also other
reasons for the data not being strictly comparable, for instance, the shortcomings in measuring the
employment in both the formal and informal sectors as discussed above. Therefore, for the rest of
section . , employment statistics post- will be the focus.
The bar chart in Figure . shows the total employment (including employees, employers and self-
employed, as well as both formal and informal sector workers) in – . During this period,
employment increased at a more respectable rate of . % per annum. In absolute terms, total
employment increased from . million in to . million in , before declining abruptly by
million in (this is mainly attributed to the over-estimation of subsistence agricultural
employment in ). A continuous upward trend took place again in – , before a job loss of
almost million took place in – . Total employment enjoyed another round of continuous
upward trend again since then before reaching . million in .
FIGURE . Total employment and employment rates, –
The growth in employment must also be considered in light of the sharp increase in the labour force.
Compared with the increase of . million in the total number of jobs between and , the
labour force increased by . million. Therefore, just above % of the new entrants to the labour
market actually found work successfully.
Table . shows that more than % of the employed work as employees for pay, while the rest are
either employers or own-account workers (self-employed). For the remainder of section . , all three
groups of workers (employees, employers, self-employed) from both the formal and informal sectors
are included for the analysis, unless stated otherwise.
TABLE . Type of work, selected years
1995 2005 2015 2016

Working for someone else for pay (employees) 8 103 9 816 13 586 13 504
Employer or own-account worker 1 368 2 396 2 233 2 320
9 471 12 232 15 819 15 824
1995 2005 2015 2016
Employees as proportion of total 85.6 80.3 85.9 85.3
Source: Author’s own calculations using the October Household Survey, Labour Force Survey September and –
. . Employment by population group, gender, age and education

There has, overall, been a sharp increase in the employment of African workers since as opposed
to a small decline in the employment of white people (see Figure . ). Between and ,
African employment increased by . % per annum; it was only . % for Coloured employment,
whereas the employment of white people decreased by . % per annum. This becomes even more
significant if one takes into account that the increase in the employment of African people was off a
much higher base than that of whites. This development probably indicates the impact of the
employment equity legislation in the country, the low birth rate among whites and the continued
emigration of white skilled workers. Table . also shows that the African share of total employment
increased from . % in to . % in , whereas the white share dropped from . % to . %.
FIGURE . Index of employment by race – (base

year = )
Source: Author’s own calculations using the – Labour Force

Survey September and – Quarterly Labour Force Survey quarter
data
TABLE . The employed of South Africa by population group (’ )

Asians
1995
Total 6 117 1 143 358 1 853 9 471 100.0
2005
Total 8 454 1 326 440 1 999 12 232 100.0
2015
Total 11 697 1 637 504 1 982 15 819 100.0
2016
Total 11 758 1 641 522 1 904 15 824 100.0
Note: Employed with unspecified population group are not shown separately in this table but are included as part of the total.
The employment of whites also declined in the s. Bhorat ( b) describes the loss of close to
white workers over the period – as an “extremely serious trend” because the country’s
key skills reservoir is located essentially among white workers. Bhorat finds the loss of skills from
South Africa extremely worrying, and describes it as one of the “most serious constraints” this country
is facing.
Table . shows the gender profile of the employed; despite the increasing evidence of feminisation,
the male share of the employed remains more dominant. On the other hand, Table . presents the
employment aggregates by age cohort, and it can be seen that in , more than % of the
employed were aged – years at the time of the survey.
TABLE . The employed of South Africa by gender (’ )
1995 2005 2015 2016

Male 5 766 7 014 8 911 8 956
Female 3 704 5 214 6 908 6 869
Total 9 471 12 232 15 819 15 824
Female as proportion of total 39.1 42.6 43.7 43.4
Note: The employed with unspecified gender are not shown separately in this table but are included as part of the total.
TABLE . The employed of South Africa by age cohort (’ )
15–24 25–34 35–44 45–54 55–64 Total Proportion of

years years years years years total
1995
Total 1 124 3 276 2 858 1 587 625 9 471 100.0
Proportion of 11.9 34.6 30.2 16.8 6.6 100.0
total
2005
Total 1 415 4 150 3 249 2 373 1 046 12 232 100.0
Proportion of 11.6 33.9 26.6 19.4 8.6 100.0
total
2015
Total 1 382 4 966 4 895 3 184 1 392 15 819 100.0
Proportion of 8.7 31.4 30.9 20.1 8.8 100.0
total
2016
Total 1 265 4 893 4 920 3 321 1 425 15 824 100.0
Proportion of 8.0 30.9 31.1 21.0 9.0 100.0
total
Finally, as mentioned in Chapter , those with a higher level of education are the ones demanded
more greatly in the labour market, given the structural change of the South African economy. The
results in Table . confirm this argument, as the employed have become increasingly more
educated: in , only . % of the employed had at least a Matric; this proportion increased to
. % in . Section . . will provide more discussion on the demand for skilled labour.
TABLE . The employed of South Africa by education (’ )
None- Grade 8-11 Grade 12 Grade 12 + Total Proportion of total

Grade7 certificate/
diploma/
degree
1995
Total 2 929 3 040 2 091 1 330 9 471 100.0
2005
Total 3 028 3 911 3 342 1 856 12 232 100.0
2015
Total 2 309 5 194 4 931 3 200 15 819 100.0
2016
Total 2 144 5 436 5 044 3 066 15 824 100.0
None- Grade 8-11 Grade 12 Grade 12 + Total Proportion of total
Grade7 certificate/
diploma/
degree
Note: Those who had certificates or diplomas but also reported not having Grade fall under the “Grade – ” category. Also,
the employed with an unspecified education are not shown separately in this table but are included as part of the total.
. . Employment in the formal and informal sectors

The breakdown of employment in the formal and informal sectors is presented in Table . . It shows
that the formal sector is still the biggest job provider, with about % of all employment being in this
sector. The informal sector (excluding domestic and agricultural workers) was responsible for only
about % of total employment in but this share increased to % in . The informal sector
will be discussed in more detail in section . .
TABLE . Employment in the formal and informal sectors, – (’ )
Sector 1997 2003 2009 2015 2016 % change p.a. 1997–2016 Proportion of 2016 total
1 000 %
Formal sector (excluding agriculture) 6 421 7 347 9 779 10 924 11 023 4 601 2.9 69.7
Informal sector (excluding agriculture) 1 039 1 891 2 106 2 719 2 640 1 601 5.0 16.7
Domestic workers 825 892 1 253 1 280 1 281 455 2.3 8.1
Agriculture 711 1 190 681 897 881 170 1.1 5.5
Total employment 9 070 11 374 13 819 15 819 15 824 6 755 3.0 100.0
Source: Author’s own calculations using the October Household Survey, Labour Force Survey September, ,
and Quarterly Labour Force Survey quarter data
. . Employment by occupation and industry

This section examines employment by broad occupation and industry category respectively, and the
results strengthen the argument that there is an increasing demand for more skilled workers. First,
Table . shows that during both – and – , employment growth was most rapid for
the skilled occupations (approximately % per annum), particularly for the two most skilled
occupation categories (the required education level is a Bachelor’s degree, according to the South
African Standard Classification of Occupations (SASCO)), namely managers and professionals. In
absolute terms, total skilled employment nearly doubled from . million in to . million in .
TABLE . Employment by occupation (’ )
1995 2005 1995–2005 2015 2016 2005–2016

1995 2005 1995–2005
% change 2015 2016 2005–2016
% change
% change % change
Managers 496 851 5.5 1 283 1 351 4.3
Professionals 325 584 6.0 800 834 3.3
Technicians 1 057 1 188 1.2 1 470 1 464 1.9
Skilled 1 878 2 624 3.4 3 554 3 650 3.0
Clerks 1 133 1 185 0.4 1 669 1 650 3.1
Service and sales workers 1 080 1 601 4.0 2 404 2 472 4.0
Skilled agricultural workers 113 291 9.9 99 72 –11.9
Crafts and related trades 1 114 1 736 4.5 2 000 1 946 1.0
Operators and assemblers 1 104 1 123 0.2 1 274 1 311 1.4
Semi-skilled 4 543 5 936 2.7 7 446 7 451 2.1
Elementary occupations 2 337 2 796 1.8 3 790 3 692 2.6
Domestic workers 694 855 2.1 1 029 1 032 1.7
Unskilled 3 031 3 651 1.9 4 819 4 723 2.4
Total 9 470 12 232 2.6 15 819 15 824 2.4
Note: Those employed with unspecified occupation are not shown separately in this table but are included as part of the total.
Table . presents employment by broad industry category. The two industries in the primary sector,
agriculture and mining, with decreasing shares in GDP, have suffered huge employment losses. In
, each industry accounted for . % and . % of total employment, but these respective shares
dropped rapidly to . % and . % in .
TABLE . Employment by industry (’ )
1995 2005 1995–2005 2015 2016 2005–2016

% change % change
Agriculture 1 227 909 –2.9 897 881 –0.3
Mining 440 411 –0.7 446 438 0.6
Primary sector 1 667 1 320 –2.3 1 342 1 318 0.0
Manufacturing 1 432 1 701 1.7 1 773 1 682 –0.1
Utilities 84 97 1.4 127 118 1.8
Construction 442 931 7.7 1 459 1 490 4.4
Secondary sector 1 959 2 730 3.4 3 359 3 291 1.7
Trade 1 660 3 008 6.1 3 198 3 196 0.6
Transport 475 614 2.6 898 915 3.7
Finance 577 1 289 8.4 2 159 2 322 5.5
Community and social services 2 168 2 182 0.1 3 580 3 497 4.4
Private households 796 1 060 2.9 1 280 1 281 1.7
1995 2005 1995–2005 2015 2016 2005–2016
% change % change
Tertiary sector 5 676 8 154 3.7 11 114 11 210 2.9
Total 9 470 12 232 2.6 15 819 15 824 2.4
Note: Those employed with unspecified occupation are not shown separately in this table but are included as part of the total.
In manufacturing, employment increased by an average of . % per annum in – but

decreased by . % in – . The major uptake in employment was therefore in the tertiary sector.
The largest absolute increase in employment took place in the financial and business services and
wholesale and retail trade industries (an increase of . million in each industry, between and
). In relative terms, the financial and business services industry experienced the most rapid growth
– an annualised growth rate of . % in – and . % in – .
. . Demand for skills

The sectors where employment declined sharply are also the sectors that are the most intensive in
unskilled labour. Their relative decline has therefore had a significant impact on the opportunities
available to those with lower educational attainment. Jobs have been lost in the sectors whose skills
are more closely matched with those of the millions of working-age South Africans who are out of a
job. The parts of the economy that have grown the most, such as finance and business services, tend
to be the most skill-intensive.
Bhorat et al. ( : ) show that the highest increase in the relative demand for labour was
consistently for skilled workers. There is, however, one major change. In the period – , there
was a relatively large increase in the relative demand for unskilled labour, while in the period after
the relative demand for unskilled elementary workers experienced a small decline. This result
shows the increased preference for skilled and semi-skilled workers in the – period.
Bhorat et al. ( : ) also show that within-sector forces were more dominant in explaining
changing labour patterns than betweensector forces for both the – and the –
periods. This implies that the sources of within-sector shifts such as technology change and
outsourcing of non-core functions were more important than the effect of demand and, in particular,
international trade flows, in influencing the skills shift in the demand for labour. In addition, when the
changes in relative demand for labour are decomposed to isolate the specific impact of international
trade, it is found that only agricultural workers in – and production workers in –
experienced actual job losses. Indeed, it is not evident whether the impact of trade has been
statistically significant. In terms of the South African labour market, however, given the various
influencing factors, the losers have been unskilled labourers, specifically unskilled African workers. The
winners, invariably, have been better-educated skilled workers who in turn have been
disproportionately white and Indian/Asian.
Bhorat and Oosthuizen ( : ) have found that in agriculture, mining and quarrying and private
households, employment has shifted significantly in favour of semi-skilled occupations at the cost of
unskilled occupations. In manufacturing, utilities, finance and community, social and personal services,
there has been a sharp increase in the employment of skilled workers at the cost of both semi-skilled
and unskilled workers.
Even more important than the absence of specific skills in the labour market is the absence of work
experience among prospective workers. About half of the unemployed persons are new entrants to
the labour market, which means they have never worked before.
Even in Grandpa’s time there was something to make you sleep. They called it work.
A
. . Atypical employment and underemployment

Atypical (or “non-standard”) employment is defined as employment that is not full-time (parttime
work) or is of limited duration (temporary). It could also be employment at a workplace other than
that of the employer, for instance if labour brokers are used (Department of Labour, : par . ).
There has been a sharp increase in the utilisation of such employees. This is discussed in more detail
in section . . .
Underemployment is similar to atypical employment to some extent, and in general refers to the
underutilisation of labour. There are two types of underemployment:
Visible underemployment, in which persons involuntarily work fewer than normal hours and are
seeking, or have the desire for, fulltime or additional work.
Invisible underemployment, which is a misallocation of labour resources, for example the
underutilisation of skills (e.g. of black persons during the apartheid era), or workers having very low
productivity (ILO, : ).
Since the introduction of QLFS in , Statistics South Africa define visible (or time-based)
underemployed as those who worked fewer than hours during the past week, but are willing and
available to work additional hours within the next four weeks. On the other hand, while Statistics
South Africa has yet to release results on invisible underemployment, Beukes, Fransman, Murozvi and
Yu ( : ) attempted to use the over-qualification approach by defining someone as invisibly
underemployed if a worker’s education years in a particular broad occupation category are at least
one standard deviation above the mean of education years of all workers in this occupation category.
Table . shows that the invisibly underemployed are always greater than the visibly underemployed.
After controlling for double-counting (some workers are defined as underemployed under both
definitions), the total number of underemployed was just above . million, accounting for . % of all
employed in .
TABLE . Underemployment (’ )
1995 2005 2015 2016

1995 2005 2015 2016
Visible underemployment: time-based definition 498 413 745 749
Invisible underemployment: over-qualification definition 1 153 1 177 1 216 1 234
Total underemployed 1 588 1 537 1 911 1 939
Total underemployed as a proportion of total employed (%) 16.7 12.5 12.0 12.3
. . Labour absorption capacity

Employment in South Africa has not been increasing sufficiently to accommodate the high number of
new entrants to the labour market, which manifests itself in the growing number of unemployed
persons in South Africa. In fact there has been a decline in the percentage of new entrants to the
labour force able to find work. The percentage of new entrants to the labour force that finds a job in the
formal sector of the economy is called the labour absorption capacity of the economy. In other
words, only formalsector employees are counted as employed when deriving this capacity.
Loots ( ) found that the labour absorption capacity of the formal economy was declining
dangerously and rapidly until , as indicated by some of the data shown in Figure . : whereas the
South African economy provided jobs for nearly three out of every four entrants to the labour market
before the s, this dropped to only three in persons in the s. The situation deteriorated
even further in – as the labour absorption capacity was – %. This means that not only were
hardly any new entrants to the labour market able to find formal employment in the s, but many
of those previously employed lost their jobs as well (Schlemmer & Levitz, : ).
FIGURE . Labour absorption capacity (%), –
Source: Loots ( ) for – , and author’s own calculations using the – October Household Survey, –
Labour Force Survey September and – Quarterly Labour Force Survey quarter data for –
The situation has changed somewhat since (it is not possible to calculate the capacity in –
, as mentioned, employees were not asked to declare their formal/informal sector work status in
OHS – ). In – , the capacity was a mere %. It increased to above % in – ,
meaning nine out of entrants to the labour market were hired in the formal sector. The capacity
dropped rapidly to below % in – .
If jobs in the informal sector are added, then a higher percentage of the new entrants found some
type of work. For example, the labour absorption capacity in – would have increased from
% to % had informal sector employees been included. Even though this makes little impression
on unemployment, it is a vast improvement on the situation that existed in the s, and especially in
the s.
. . Employment elasticity of economic growth

An increase in production normally leads to an increase in the demand for labour. This means that
higher economic growth should also lead to an increase in the demand for labour. However, the
impact of economic growth on employment is not necessarily the same over time – more or fewer
jobs might be created even though economic growth might be unchanged.
These changes can be measured by the employment elasticity of economic growth, or as it is also
called, the employment coefficient or the employment-output ratio. The employment elasticity of
economic growth indicates the sensitivity of employment to changes in economic growth. In other
words, it is the extent to which employment will change as a result of changes in economic growth, or
the degree of responsiveness of employment to changes in economic growth. This is measured as
follows:
% change in employment
Employment elasticity of economic growth =
% change in economic growth
If the coefficient (or elasticity) is positive and greater than one (elastic), it means that a % economic
or real GDP growth rate will result in a more than % increase in employment. Similarly, if the elasticity
is positive but smaller than one (inelastic), it means a % in economic growth rate will result in less
than a % increase in employment. If the elasticity is negative, it implies the rare case of jobless
growth – real GDP growth is associated with job loss.
Table . indicates that the employment elasticity of economic growth was slightly above one in
– , regardless of whether informal sector employment and self-employment are included or
not. Looking at each five-year interval, the elasticity is the highest ( . ) during – , followed
by – (. ) if all workers are counted. However, it is important to interpret the elasticities by
also checking the annualised job creation and real GDP growth rates. For instance, despite the –
elasticity ( . ) being the second highest across the four five-year periods, both employment
( . %) and real GDP ( . %) annualised growth rates were low.
TABLE . Employment elasticity of economic growth
1997–2016 1997–2001 2002–2006 2007–2011 2012–2016

Employment elasticity of economic growth 1997–2001 Including all employed 1.045 1.810 0.620 0.785 1.456
Including formalsector employees only 1.042 0.891 0.807 1.408 1.317
Annualised percentage change (%) Real GDP 2.8 2.5 4.4 2.6 1.6
Total employment 2.9 4.5 2.7 2.1 2.3
Formal-sector employees 2.9 2.2 3.6 3.7 2.1
September, – Quarterly Labour Force Survey quarter , and the South African Reserve Bank data
During the whole -year period as well as all five-year intervals, the elasticity was positive, meaning
jobless growth did not take place. This is also confirmed in a study (Bhorat et al., : ) where the
authors state that the notion of aggregate “jobless growth” in the South African economy is
erroneous. The economy, in the aggregate, has been creating jobs rather than shedding them, albeit
not at a rate sufficient to accommodate the new entrants to the labour market. However, they do
point out that there is some doubt whether the increase in employment could not simply be ascribed
to the growth of the informal sector. Also, there does seem to have been so-called jobless growth in
mining and manufacturing, where employment declined in spite of increases in production.
Furthermore, it seems as if such increases in formal-sector employment that might have occurred,
might have benefitted the higher skills levels, with employment among lower-skilled workers (such as
production workers) declining.
One possible reason for the employment growth not being highly elastic to economic growth is that
changes in labour legislation and other factors have driven employers towards methods of production
that rely more heavily on a reduced number of skilled workers and automation in some industries
such as agriculture and construction (CDE, : ).
Another reason is that relatively highly skilled economic sectors such as financial services have grown
more rapidly than sectors employing a large number of unskilled workers.
. . Capital and labour intensity

As stated at the beginning of this chapter, employers are able to substitute capital for labour or
labour for capital. The extent to which an economy is becoming more or less labour intensive can be
measured in two ways, firstly by considering how many jobs are being created by every million rand
of production. The other method is to consider changes in the capital-to-labour ratio over time.
The capital-to-labour ratio is the ratio of the capital employed at any particular date to the number of
workers at that date, as shown by the following:
fixed capital stock

Capital/labour ratio =
total employment
In , the then National Productivity Institute (now known as Productivity SA) estimated the capital
intensity of the South African private non-agricultural economy over the period – , and it was
found that the capital intensity of the South African economy increased sharply, as shown in Figure
. (owing to changes in Statistics South Africa surveys the data are not always comparable over
time). Almost two million jobs were lost during the period as a result of investment being channelled
increasingly into capital-intensive sectors and technologies.
FIGURE . Capital-to-labour ratio in the private non-
agricultural economy, –
Source: National Productivity Institute ( )
* Series not comparable from rd quarter
In , Productivity SA estimated the average annual growth rate of the capital-labour index by
sector in – , and Figure . shows that all sectors experienced positive growth rates during
the period, ranging from as low as . % in agriculture to as high as . % in mining. For the economy
as a whole (excluding agriculture, government and community sectors), Productivity SA ( : )
estimated that the capital-labour ratio grew by . % in and . % in .
FIGURE . Average annual growth rates of capital-to-labour ratio index
by sector, –
Source: Productivity SA ( )
Bhorat et al. ( : ) confirm the sharp increase in the average capital-to-labour ratio. The primary
sectors (agriculture and mining) experienced the most extensive capital deepening followed by
construction and manufacturing, while the service sectors saw only a modest increase in their capital-
to-labour ratios.
One of the most important factors influencing the capital-to-labour ratio, or the substitution of labour
by capital, is the relative cost of labour and capital. The more expensive labour becomes relative to
the cost of capital (i.e. interest rates), the more labour will be substituted by capital. Behar’s research
( : ) has shown that a reduction in wages relative to the cost of capital would expand
employment as a result of the substitution of labour for capital. Some caution is thus called for by
those advocating lower interest rates as a solution for unemployment, because the lowering of
interest rates would probably lead to a fall in employment in favour of capital acquisition. However,
the corollary is not necessarily valid either, namely in order to raise employment there should be a rise
in interest rates.
The relative cost of capital versus that of labour is not the only factor the influences the capital-to-
labour ratio. Any of the factors that inhibit the demand for labour are to a greater or lesser extent also
likely to result in an increase in the capital-to-labour ratio. However, there are also technological
factors. As machinery and information technology are developed that are able to perform functions
previously reserved for human endeavours, employers are likely to employ more capital-intensive
production methods.
It should also be taken into account that capital-intensive production has another disadvantage
besides reducing the labour-absorption capacity of the economy. This is that capitalintensive
techniques are often less flexible than labour-intensive production. People are more versatile than
machines, and in industries where the life of products is relatively short, labourintensive production is
sometimes more costeffective than a capital-intensive system.
Prostitution is a labour-intensive activity involving little in the way of intermediate consumption.

OECD
One could argue that labour-intensive production techniques should be encouraged. However, there
are two important caveats:
Capital intensity is not necessarily inimical to the economy. Modern production structures are
largely technology driven, which means they are capital-intensive. In order to compete successfully
on overseas markets, South Africa must utilise the most modern production techniques, which
often means high capital intensity.
The sharp increase in capital intensity is not necessarily spread over all industries. Specific projects
were often responsible for the sharp increase in manufacturing capital intensity, for example the
Sasol synthetic fuel development from – and major new heavy investment in the pulp and
paper industry in the period – .
The adoption of new technologies is partly a result of trade liberalisation, more specifically to enable
firms to compete more effectively and to satisfy consumer tastes. Therefore trade liberalisation might
indirectly be responsible for more job losses than could be apparent by simply considering the direct
effect. However, even local consumer tastes have changed, and consumers prefer products produced
by high technology. In addition, the microelectronic revolution has also contributed to this trend
towards more capital-intensive production. Furthermore, the relative prices of capital and labour have
also had an impact on the replacement of labour by capital.
SUMMARY
The number of jobs increased by about . million between and , but this increase is not rapid enough to
keep up with the extent of increase in the labour force ( . million).
More than % of the employed work as employees for pay.
The increase in employment was relatively more rapid for female Africans aged – years with at least a Matric,
working in the skilled occupations in the tertiary sector.
Labour absorption capacity was at its highest level (above %) in – .
Employment elasticity of economic growth is slightly above one in – , which means employment growth is not
too highly responsive to real GDP growth. Nonetheless, there is no indication of jobless growth taking place.
The capital-to-labour ratio in general shows an upward trend in – , meaning the capital intensity of South
African economy increased.
. THE INFORMAL SECTOR
. . Definition and measurement of the informal sector

Some economic activities would clearly be regarded as informal-sector activities. For instance, street
hawkers who are not registered with any regulatory authority and who do not pay any other tax than
value-added tax could be regarded as informal. However, the situation becomes more difficult if that
same street hawker is licensed, and as part of the licensing process is required to register as an
individual taxpayer. Then there might be some people in the informal sector who have full-time,
formal jobs and may undertake their informal activities only on a part-time basis to supplement other
income. Such activities might also be undertaken on a temporary basis to bridge a period of
unemployment. Many children are also involved in the informal sector and some criminal activities
might or might not be seen as informal sector activities.
Consequently, it is not surprising that the way in which the informal sector has been defined and
measured has changed substantially over time, and this would have an impact on changes in
employment over time. Researchers have, for instance, referred to the following definitions of the
informal sector:
According to the dualistic approach, the informal sector is seen as being separate from the formal
sector. It is distinguished on the basis of various characteristics, such as family ownership, skills that
are acquired outside the formal school and training systems, and markets that are unregulated
(either not officially recognised or encouraged, or illegal).
Another method of defining the sector is by the continuum approach, in terms of which the
economy as a whole must be seen as a continuum of activities – from full-time, permanent wage
employment at the one extreme to the informal gathering of wood for own use at the other
extreme, with a number of variants in between. In this regard, the informal sector is increasingly
being referred to as the informal economy so as to get away from the idea that informality is
confined to a specific sector of economic activity but rather cuts across many sectors.
According to another approach, the informal sector can be regarded as that part of the economic
activity of the country that is not recorded in its national accounts, in other words the statistically
unrecorded part.
The definition of the ILO has also changed over the years (see Essop & Yu, : – for a detailed
discussion). When the informal sector was first defined in , the ILO regarded it as the activities of
the hard-working poor, who were not recognised, recorded, protected or regulated by the public
authorities. About years later, the definition for the informal sector was based on the so-called
enterprise approach, in other words the characteristics of the enterprise in which the activities took
place. This meant the informal employment only included those working in the informal sector.
However, the definition suggested at the th International Conference of Labour Statisticians in
was that informal employment should be defined in terms of the employment characteristics of the
persons involved. This means that informal employment should include not only those working in
informal enterprises but people whose employment displays informal characteristics, despite working
outside the informal sector (Yu, : ). According to this definition, the informal economy “is seen as
comprised of informal employment (without secure contracts, worker benefits or social protection)
both inside and outside informal enterprises:
1. Informal employment in informal enterprises (small unregistered or unincorporated enterprises),

including employers, employees, own-account operators and unpaid family workers in informal
enterprises.
2. Informal employment outside informal enterprises (for formal enterprises, for households or with
no fixed employer), including domestic workers, casual or day labourers, temporary or part-time
workers, industrial outworkers (including home-based workers) and unregistered or undeclared
workers” (ILO, : in Essop & Yu, ).
In South Africa, the informal sector in the – October Household Survey (OHS) was determined
with reference only to those working as self-employed in unregistered businesses. However, since
, employees who worked for unregistered businesses were included. According to this
measurement about % of all employees worked for the informal sector.
Further changes took place when the OHS was replaced by the Labour Force Survey (LFS), where the
latter introduced more probing questions to determine the size of the informal sector (see Casale &
Posel, : ). In , some women and men were involved in subsistence farming
according to the OHS. In the LFS pilot survey in February , these numbers suddenly increased to
women and men. Employment and participation rates thus increased quite
substantially, but simply as a result of the more efficient capture of informal-sector and subsistence
farming. However, the general approach used in both the OHS and the LFS could be described as the
enterprise approach (Yu, : ).
In the QLFS, which was introduced in , a fundamentally different approach was added. The one
approach remained the enterprise approach. According to this definition, the informal sector has the
following two components:
1. Employees who are not registered for income tax and who work in establishments that employ
fewer than five people.
2. Employers, own-account workers and persons helping unpaid in their household business who are
not registered for either income tax or value-added tax
The second, broader concept that was also introduced in the QLFS was that of informal
employment. This included both those employed in the informal sector as well as informal
employment in the formal sector. This is in line with the approach of the ILO, as outlined above.
According to the QLFS, formal-sector employees were defined as informally employed if they were
not entitled to medical aid and pension funds, and did not have a written contract with the employer.
This meant that the job-related characteristics instead of the characteristics of the enterprise were
considered in defining whether someone was informally employed: in other words, was the person
employed in an “informal job”?
. . Importance and size of the informal sector

The informal sector plays a very important role in any developing country, especially where there is no
developed system of social security. If unemployed persons cannot rely on unemployment insurance
or some form of state support during spells of unemployment, they must find some other means of
survival. This results in many unemployed persons entering the informal sector. The income in the
informal sector is usually very low and the employment conditions very poor.
The generally unregulated nature of the informal sector presents a double-edged sword to business
owners who operate within that space; there is less administrative burden and costs, which in many
cases form a significant cost of doing business, but the business is less protected and cannot access
key support such as trade credit, insurance and financial services to the same degree as a registered
business.
Although the informal sector does to some extent alleviate the problem of economic inactivity, it does
not generally solve the problem of poverty and low standards of living. Therefore, it should not be
seen as the solution to the unemployment problem, even though it does contribute to an increase in
income. For instance, based on the survey by Statistics South Africa, Van der Berg ( ) has
estimated that the per capita income of Africans is % higher because of informal sector activities.
The incomes of Africans as a percentage of those of whites would therefore increase from . % to
. % as a result of the informal sector.
God gives every bird its food, but He does not throw it in the nest.
A
Because the informal sector provides some kind of income, and because it can serve as a stepping
stone for future “formal-sector” entrepreneurs, the previous government, in the late s and early
s, eased many conditions, which had severely hampered the development of the informal sector.
This was taken even further by the new government after . The result of this policy change can
clearly be seen in the development of, for instance, the minibus taxi industry, hawking activities and
spaza shops. Nevertheless, the authors of the ILO Review ( : ) maintain that the relaxation of
regulations has had little effect on the incomes of those working in the informal sector.
Some key statistics on informal sector employment in – are presented in this section. Note
that the enterprise characteristics approach is adopted for QLFS – . First, Figure . shows
that informal sector employment more than doubled between ( . million) and ( . million),
whereas this employment as a proportion of total employment increased from . % to . % during
the same period. Although not shown in the figure, informal employment (which includes workers in
the informal sector as well as those displaying strong informal characteristics working in the formal
sector) fluctuated between . and . million in – .
FIGURE . Informal sector employment (‘ )
Over the years, approximately half of informal sector workers are self-employed, % work as
employees and the remaining % are employers. In contrast, the corresponding shares are %, %
and % when it comes to the formal sector workers. Table . shows the top detailed occupation
categories of informal sector workers in ; the majority of these workers are involved in wholesale
and retail trade, construction and personal service activities.
TABLE . Top detailed occupation categories of informal sector workers (as a proportion of total
informal sector employment), rd quarter,
Percentage (%)
Street food vendors and related workers 12.16
Street vendors, non-food products 6.17
Brick layers and stonemasons 6.11
Taxi drivers 4.84
Hairdressers, barbers, beauticians and related workers 3.89
Shop salespersons, petrol-pump and filling-station attendants 3.11
Motor vehicle mechanics and fitters 3.04
Percentage (%)
Spaza shop owner 2.88
Cooks 2.86
Building construction labourers 2.41
47.47
Source: Author’s own calculations using Quarterly Labour Force Survey quarter data
Several studies have found that the level of the informal sector remains quite low by the stanless
dards of developing countries (International Monetary Fund, : ). This is confirmed by a doctoral
thesis of Saunders ( ), where he found that the informal sector in South Africa constituted about
% of GDP, compared to to % in other developing countries and to % in transitional
economies. However, Yu ( : – ) refers to studies that find that South Africa is in the mid-range
size of informal markets in developing countries and that it is a regional outlier when compared to
African countries.
However, this excludes the “broad” definition adopted as one of the measures in the QLFS, since
international comparisons are not available.
The researchers who have found that the size of the informal sector in South Africa is in fact small
ascribe this to, among other things, the following factors:
The legacies of the apartheid regime, which made it first illegal and then difficult for blacks to move
to larger urban areas unless they already had a certified job (Rodrik, : ). South Africa might
still be sitting with the legacies of this policy.
Another possibility is the prevalence of high levels of crime, which acts as a particularly onerous tax,
or cost of doing business, on small-scale enterprises.
The low population density in rural areas means that there is not a sufficient demand, which
militates against businesses being established.
Many sectors are dominated by a few very large companies, which actively or otherwise might be
discouraging competition by small business including the informal sector.
The existing system of social grants theoretically sets a higher “reservation wage” level in South
Africa than in comparable countries, acting as a deterrent to low-productivity informal activities
(e.g. street peddling) (Rodrik, : ) (see section . . for a discussion of reservation wages).
The informal sector appears to be at the forefront of job losses in times of crisis, for instance after the
crisis (with job losses at a rate about double that of the rest of the economy). However, this is
not surprising, given that the informal economy provides less security than the formal sector (Bhorat,
). On the other hand, one would have thought that the informal sector might serve as a cushion
for those losing their jobs in the formal sector, and that it would thus grow in times of economic crisis
where many formal-sector jobs are lost.
SUMMARY
Different studies have resulted in widely divergent estimates of the informal sector. Several studies have found that the
informal sector in South Africa is relatively small compared to other developing countries, but this conclusion is by no
means shared by all analysts. The informal sector nevertheless remains a particularly important source of income for
those without a formal job.
. DEEPER INSIGHT: DERIVING THE DEMAND FOR LABOUR
As already stated, the demand for labour is a derived demand, i.e. it is derived from the demand for
the product made or the service provided by the employee. The demand for labour will therefore be
affected indirectly by economic conditions, such as expansionary or restrictive fiscal and monetary
policies, and domestic and foreign competitive conditions that have an impact on the demand for
goods and services. These macroeconomic factors that affect the demand for labour fall outside the
scope of this book.
However, the demand for labour is also directly affected by the cost of labour (including the wage
rate), factors that influence the cost of labour (such as unions and collective bargaining structures)
and by the productivity of labour, as these factors influence the price at which the final product can
be sold, i.e. the employer’s competitive position. These factors are discussed in more detail in
chapters and , and to some extent in Appendix A. To determine what the effect of these factors
might be, however, it is necessary first to determine how the theoretical market demand for labour
can be deduced from the demand for labour by the individual employer. As will become apparent,
this also influences the actions of trade unions when it comes to wage demands.
Table . summarises the symbols of the key variables to be used throughout section . . These
symbols will be explained in detail throughout the section. Note that in some labour economics
textbooks, the symbols y and Y are used to represent the firm’s and the industry’s total output
respectively. However, in order not to confuse the students (as y also stands for the y-axis variable
while Y was already used in Chapter to represent total income), this textbook would rather use q
and Q to distinguish the firm’s and the market’s total output respectively.
TABLE . Symbols of important variables for the labour demand theory
Symbol Variable
TP (= q) Total product of the firm
APE Average product of labour (= TP/E)
MPE Marginal product of labour (holding capital constant) (= Δq/ΔE)
MPK Marginal product of capital (holding labour constant) (= Δq/ΔK)
P Per-unit selling price of the good produced by the firm

MR Marginal revenue
VMPE Value of marginal product of labour (p × MPE)
MRPE Marginal revenue product of labour (MR × MPE)
Q Total product of the market or industry

Symbol Variable
E Employment (work hours)
K Capital (units)
W Market wage in rand (hourly)
R The cost of capital in rand (per unit)
C (= TC) Total long-run production cost (= WE + RK)
MCE Marginal cost of labour
Π Total profit of the firm (= TR – TC = Pq – WE – RK)

MRTS Marginal rate of technical substitution of capital for labour
QD Quantity of labour demanded
In this section, the derivation of short-run labour demand of the firm will be examined first, involving
first-year microeconomics theories. This is followed by a discussion on how the firm’s labour demand
in the long run is derived, with the aid of the isoquants and isocost lines, which are highly similar to
the indifference curves and budget lines we learnt about in Chapter . The derivation of the market
labour demand and factors affecting this demand will also be looked at.
Finally, unless stated otherwise, in this section it is assumed that perfect competition takes place in
both the goods market and the labour market. In other words, the firms in the goods market are price
takers (selling their goods at the market price) while participants in the labour market are wage takers
(all firms would pay all workers the market wage), because the price of output (P) is unaffected by
how much this firm produces and sells, whereas the costs of labour and capital are unaffected by how
many units of labour and capital the firm hires. It will be discussed later how the firm minimises
production cost and maximises profit by hiring the right quantities of the two inputs.
. . The production function

For the sake of simplicity it is assumed that there are two inputs in the production process, namely
number of work hours hired by the firm (E) and the aggregate number of units of capital (K), which
include land and machines, for instance. In equation terms, the production function is:
q = TP = f(E; K)
We start off by examining short-run production and the derivation of short-run labour demand by
assuming capital is the fixed input. This means K is fixed at a particular level. The total production (TP)
of an employer will normally increase if additional labour hours (or more employees) are appointed.
However, if other factors such as the physical space, machinery and so forth are not increased, the
employment of additional labour hours or employees will result in smaller and smaller additions to
total production as the workers start crowding each other. Eventually there will come a point where
the employment of further workers will not lead to increased production – the superfluous workers
will probably simply stand around with nothing to do, because there will not be enough machines or
even space.
This fairly obvious conclusion is the basis of the law of diminishing marginal product.
However, the concept of marginal production needs to be explained first. In economic terms,
marginal product of labour (MPE) refers to the change in production associated with employing an
additional unit of labour, holding capital constant. In equation terms, it is equal to Δq/ΔE. For the sake
of simplicity, it is assumed that the labour provided by all workers is equal, even though in actual fact
labour skills and productivity can greatly differ.
The law of diminishing marginal product is a well-known concept in economics. This law refers to the
amount of additional (i.e. marginal) production that can be produced when successive additional extra
units of one variable input (labour in this case) are added while keeping the other inputs (capital in
this case) constant. For instance, assuming the product being produced is wheat, and that labour
inputs are varied, whereas there is a fixed piece of land and a fixed piece of equipment (e.g. one
tractor and a plough) to produce the wheat. The capital (i.e. the land and the tractor) is kept constant,
and additional workers are added in succession. The additional wheat that is produced after the
addition of every worker is then calculated. This is called the marginal product of labour.
Work expands so as to fill the time available for its completion.

P ’
Table . presents a numerical example to better understand how marginal product of labour is
calculated. Let us assume that capital is fixed at five units. We also assume that the first hour of labour
struggles to perform all the necessary tasks such as driving the tractor, controlling the plough, etc.,
and produces only units (e.g. tons) of wheat (marginal product of units). The addition of another
labour hour makes things much easier and production is now increased to units, which means that
the second worker’s marginal product is units. A third labour hour also helps to increase
production, and this time by an even greater units, to a total of units. Marginal product of
labour starts declining when the fourth labour hour is hired – this additional hour adds a slightly lower
units to total product. If more and more work hours are added, they eventually become
superfluous and total production starts decreasing when the twelfth labour hour is hired (output
drops from to , while marginal product becomes negative for the first time at – as there are
simply too many workers on the piece of land, they start crowding each other, interfere with normal
production processes, etc.
The law of diminishing marginal product thus simply states that as successive units of one input (e.g.
labour) are added to a fixed amount of another input (capital), a level of total production is reached
beyond which the marginal product of that variable input declines.
The last column of Table . shows the average product of labour (or APE), which simply means the
amount of output produced per work hour by holding capital constant, that is, q/E.
TABLE . Calculating the marginal and average product of labour of a hypothetical firm
Capital unit (K) Work hours (E) Total product (TP) Marginal product (MPE) Average product (APE)
5 0 0 – –
5 1 11 11 11.00
Capital unit (K) Work hours (E) Total product (TP) Marginal product (MPE) Average product (APE)
5 2 27 16 13.50
5 3 46 19 15.33
5 4 64 18 16.00
5 5 81 17 16.20
5 6 96 15 16.00
5 7 109 13 15.57
5 8 120 11 15.00
5 9 128 81 4.22
5 10 133 5 13.30
5 11 134 1 12.18
5 12 133 –1 11.08
Total product, marginal product and average product can be illustrated graphically. In Figure . (a),
the increase in production by the addition of each work hour is clear – the first work hour produces
units, the first and second hours jointly produce units, and so on. In Figure . (b), the marginal
product curve shows how the marginal product increases with the addition of the second work hour.
The marginal product of the first hour is units, of the second hour units, of the third units, and
so forth. After the third hour, the marginal product starts to decline – the fourth work hour only adds
more units of production, and total production is units. Total product peaks at units at the
eleventh work hour. If another work hour is added after the eleventh hour, the marginal product of
that work hour will be negative, i.e. that work hour will cause total production to decline to lower than
units.
The average product is also shown in Figure . (b), and it can be seen that the marginal product
curve lies above the average product curve when average product is increasing, but the marginal
product curve lies below the average product curve when average product is decreasing. This means
the two curves intersect where average product reaches its maximum.
FIGURE . Total production and marginal production
The segment ZZ in Figure . (b) is the basis for the short-term demand for labour, but only after the
addition of another factor – the price of the product, which is discussed next.
Finally, the firm’s objective is to maximise its profit, and the profit function (π) is given as:
π = TR – TC = Pq – (WE + RK) = Pq – WE – RK
W and R represent labour wage and rental price of capital respectively. As mentioned before, we
assume perfect competition in both goods and labour markets, and hence P, W and R are constant.
The firm needs to hire the right amount of labour and capital to maximise profit.
. . The short-run demand for labour

Thus far only the physical quantity produced has been considered. Prices are now added – of the
product and of labour (the wage). By multiplying the price of the product by the physical quantity
produced, one can derive the revenue of the producer.
The value of marginal product of labour (VMPE) will then be the change in total revenue resulting from
the employment of an additional labour hour or worker, holding capital constant – in other words, the
marginal product of labour multiplied by the price of the product (Marshall et al., : ). In
equation terms, this means:
VMPE = P × MPE
Because of the operation of the law of diminishing marginal product, the increase in revenue (the
VMP) becomes smaller and smaller as more workers are employed, i.e. every additional labour hour or
worker benefits the employer less and less. Also, as VMPE equals MPE times constant price of the
output, it can be said that the VMPE curve (to be shown later) is simply the “blown-up” version of the
MPE curve.
Under the assumption of a perfectly competitive labour market and fixed capital, if the firm expands
production, the marginal cost of the firm by hiring more labour (MCE) would simply be equal to the
market wage (W).
To maximise profitability, the employer will employ additional labour hours or workers for as long as
their contribution to the total income of the firm is more than what they cost the firm (the firm thus
continues to make a profit). Intuitively, this makes sense: the employer will not employ additional
workers if they cost more than they produce. To put it more technically, the firm will employ an
additional unit of labour for as long as that worker’s VMP is higher than his or her marginal cost (i.e.
the market wage). The employer will thus employ additional labour up to the point where VMPE is
equal to marginal cost, providing the VMPE is declining, that is:
VMPE = W and VMPE is downward-sloping.
However, should marginal cost (the wage) decline, the employer would be willing to employ more
labour, because the marginal cost would then be lower than the VMPE, and the employer would make
additional profits by employing additional workers or labour hours.
Table . uses the information on MPE from Table . to derive VMPE, assuming the perunit price
of the good is R . It can be seen that VMPE peaks at when the third labour hour is hired (MPE also
peaks at the same additional unit of labour hour) from it keeps declining as more than three work
hours are hired, and reaches a negative value when the twelfth labour hour is hired.
TABLE . Calculating the value of marginal product of labour of a hypothetical firm
Labour hours (E) Marginal product (MPE) Per-unit price (P) Value of marginal product (VMPE)
0 – – –
1 11 5 55
2 16 5 80
3 19 5 95
4 18 5 90
5 17 5 85
6 15 5 75
7 13 5 65
8 11 5 55
9 8 5 40
10 5 5 25
11 1 5 5
12 –1 5 –5
The VMPE in Figure . therefore has a downward slope between E = and E = , indicating that as
each next worker is employed, his or her additional contribution to total production will be smaller
than that of the previous worker.
FIGURE . The firm’s hiring decision in the short run
If the market wage rate is R , the employer will hire three labour hours. The employer makes no
more profit if he or she employs more than three labour hours because the VMPE of the third work
hour is exactly equal to the wage rate (VMPE = W = ). In fact, by employing more work hours, the
employer will reduce his profits because the VMP of the next work hour will be lower than the
prevailing wage rate of R . For instance, assume the firm owner stubbornly insists on expanding
production by hiring the fourth labour hour. It would cost R to hire this additional labour hour, but
its VMPE is only R . From a profit-maximising perspective, therefore, it is not worth hiring more than
three work hours at the market wage rate of R .
If the wage rate declines to R , the VMPE of labour hours four to nine is more than R , which
means more profit for the employer. When nine labour hours are employed, the employer cannot
make a higher profit by employing more labour. At that point, the wage rate of R will again be
equal to the VMPE. The VMPE will therefore also be the demand curve for labour.
One thing that has still not been discussed is why only the upward-sloping segment of the VMPE is
not considered when deriving the short-run labour demand of the firm. Consider a market wage of
R . It can be seen in both Table . and Figure . that VMPE equals W at E = (when VMPE is
upward-sloping) and E = (when VMPE is downward-sloping). Despite the fact that VMPE = W = R
when the first labour hour is hired, it would not lead to profit maximisation, because if the firm hires
another unit of labour (E = ), this second labour hour would contribute even more to the firm’s
revenue than the first worker (VMPE = > = W). It means there is room for the firm to expand
production, given the fact that MPE (and VMPE) is still going up as labour quantity increases. In
contrast, when the eighth labour hour is hired, it would lead to profit maximisation, because if the
firm hires the ninth labour hour (E = ), this additional labour hour would contribute less to the firm’s
revenue than the eighth hour (VMPE = < = W).
The firm’s short-run demand for labour can thus be indicated by the downward-sloping VMPE curve. In
our hypothetical example, the labour demand curve is equal to the segment of VMPE between E =
and E = , as shown in Figure . . Although VMPE is downwardsloping at E = , this segment is
excluded since VMPE is negative (– ) and market wage cannot be negative.
FIGURE . The firm’s short-run labour demand curve
In the real world, employers will not base their hiring decisions directly on the marginal product of
labour theory; in fact, very few have ever heard of this theory. However, as already stated, employers
intuitively realise that they will make maximum profits if they employ labour for as long as the wage
rate of the worker is lower than the additional production that can be produced by employing that
worker.
What will happen to the firm’s short-run labour demand when P or MPE increases?
It was discussed previously that quantity of labour demanded in the short run is derived when W = VMPE = P × MPE,
providing VMPE is downward-sloping and VMPE is positive. Now we want to show that the short-run labour demand
curve would shift rightwards if P or MPE increases.
The first four columns of Table . show what originally happened to VMPE, given the MPE schedule and a constant
price of R . The labour demand curve is derived as D in Figure . . Now assume there is a constant increase of of
VMPE at each additional unit of labour as a result of labour productivity increase (see a constant increase of MPE of
units of output at each labour quantity). For instance, at E = , originally VMPE was but now it increases to . The
labour demand curve shifts rightwards from D to D as a result. At the end, the quantity of labour demanded increases
as each wage level (that is, there is an increase in labour demand, as a result of the rightward shift of the labour demand
curve). For example, at W = , originally four labour units were demanded (W = VMPE = when E = ; see point H
in Figure . ). As a result of increase in labour productivity and subsequently the VMPE, at the same wage of R five
labour units are now demanded (W = VMPE = when E = ; see point J in Figure . ).
TABLE . Impact of increase in price and marginal product on short-run labour demand
Before any increase After MP increase After P increase

E MPE P VMPE MPE P VMPE MPE P VMPE
1 100 5 500 120 5 600 100 10 1000

2 80 5 400 100 5 500 80 10 800
3 60 5 300 80 5 400 60 10 600
4 40 5 200 60 5 300 40 10 400
5 20 5 100 40 5 200 20 10 200
FIGURE . Rightward shift of short-run labour demand curve as a result of price or

marginal product of labour increase
Now we instead assume MPE is unchanged, but price doubles from R to R . It can be seen from the last column of
Table . that the VMPE increases accordingly, and the firm’s short-run labour demand curve shifts rightwards from D
to D . At the end, the quantity of labour demanded increases as each wage level. For instance, at W = , originally two
labour units were demanded (W = VMPE = when E = ; see point E in the figure). After the price increase, four
labour units are demanded (W = VMPE = when E = ; see point G in the figure).
. . The long-run demand for labour

In the long run, the firm’s capital stock is no longer fixed. Hence, in the long run, the firm aims to
maximise profit by choosing how many units of labour and capital to hire and invest in respectively.
Before the firm’s long-run profit maximisation is discussed, it is important to examine the isoquant
and isocost curves.
Isoquant curve
An isoquant curve indicates the possible combinations of labour and capital that produce the same level
of output, by assuming the two inputs can be substituted. This is illustrated in Figure . . Points A, B,
D, E and G produce the same levels of production ( units) by using different combinations of
capital and labour. To obtain a higher level of output of units, the firm can increase the
combinations of both capital and labour, and will move from isoquant q to q . Furthermore, it is
actually possible to derive the short-run marginal product of labour in the figure; assuming capital is
fixed at units in the short run, the horizontal dotted line AA’’ shows that total product increases
from to units when labour increases from four (point A) to six (point A’) hours, that is,
marginal product of labour equals (Δq/ΔE = / ). Similarly, as total product rises from to
when after increasing work hours from six to nine, marginal product of labour equals . (Δq/ΔE =
/ ).
FIGURE . Isoquant curves
Isoquants have the following properties, which correspond to those of the indifference curves in
Chapter :
They are downward-sloping, which means that as the input of employment increases, the input of
capital declines in order to produce the same output.
They are convex to the origin, indicating that capital and labour are not perfect substitutes. As the
utilisation of labour increases (decreases) to very high (low) levels, proportionally less (more) capital
will be required to produce the same output.
Higher isoquants are associated with higher levels of output.
Isoquants do not intersect.
For the remainder of section . . , the slope of the isoquant curve is examined in greater detail by
introducing the following three concepts: marginal product of capital (MPK), marginal product of
labour (MPE) and marginal rate of technical substitution (MRTS). As in section . . , the relatively
straightforward chord method (instead of the tangency method – point G in Figure . is an
example) is adopted to derive the slope of the non-linear isoquant curve.
While section . . already defined marginal product of labour, marginal product of capital (MPK)
means the change in production associated with employing one additional unit of capital stock, holding
labour constant.
The move from R to T in Figure . can be broken down into

Δq = R → S + S → T = R → T =
This can be rewritten as:
MPK × ΔK + MPE × ΔE =
FIGURE . Derivation of the marginal rate of technical

substitution
The above equation can be explained with the aid of the following mathematical example. The
movement from R to S is associated with a decrease of capital of five units. Now we assume that, on
average, for each one unit change of capital, marginal product of capital is equal to units of
output. Therefore, the total change of output from R to S is equal to MPK × ΔK = ( ) × (– ) = – .
This explains why output decreases from units at point R to units at point S.
The movement from S to T involves an increase of employment of eight hours (from eight to ). We
assume that, on average, for each one unit change of work hours, marginal product of labour is equal
to . units. Hence, the total change of output from S to T is equal to MPE × ΔE = ( . ) × ( ) = ;
this explains why output increases from at point S to at point T. Overall, this means the
movement from R to T would result in the firm maintaining the same production level: the firm
initially experiences a decline of production ( to units) from R to S, before returning to the
original level ( to units) after moving from S to T. This means the change of the firm’s output
(Δq) is zero from R to T.
Substituting all values into the above equation:
Δq = MPK × ΔK + MPE × ΔE =
( ) × (– ) + ( . ) × ( ) = – + =
Rearranging the terms of the equation results in the following:
MPK × ΔK = –MPE × ΔE
ΔK M PE
= −
ΔE M PK
In other words, the absolute value of the slope of an isoquant curve, or the marginal rate of technical
substitution (MRTS) of capital for labour, is the ratio of the marginal products.
It can be easily proven with simple mathematics that the above relationship holds:
ΔK −5
= = − 0.625
ΔE 8
M PE 12.5
− = − = − 0.625
M PK 20
The isoquant curve becomes flatter as labour quantity increases, meaning the slope of this curve in
absolute terms declines as we move rightwards along the curve, as fewer units of capital are given up
for hiring an additional labour unit to maintain the same output level. Therefore, it is equivalent to the
assumption of diminishing marginal rate of technical substitution. For example, going back to Figure
. , it shows that the slope of the isoquant q equals – between A and B (five units of capital are
given up to hire an additional unit of labour to produce the same output of units), while the slope
equals – between D and E (only one unit of capital stock is reduced to hire an additional unit of
labour).
Isocost curve
The firm’s costs of production consist of the cost of capital (R) and the cost of labour (W). Hence, the
firm’s total production cost, denoted by C, is given by:
C = WE + RK
This can also be shown in Figure . . The line BD shows all the combinations of capital and labour
that are equally costly. If the firm only hires labour, it could employ a maximum of C/W units (that is,
the x-intercept of the isocost curve); if the firm decides to hire only capital, it could hire a maximum of
C/R units of it (this represents the y-intercept of the curve).
FIGURE . An isocost curve
The slope of the isocost curve equals –W/R, by rearranging the total cost equation as follows:
C = W E + RK
RK = −W E + C
RK WE C
= − +
R R R
W C
K = − E +
R R
Higher isocost curves suggest higher costs, and this can be explained with the aid of Table . and
Figure . . Given an initial cost outlay of R and the per-unit costs of labour being R and R
respectively, the original isocost curve of BD is derived. It shows that the maximum units of labour and
capital that could be hired are equal to and respectively. If the firm can now afford to spend a
higher R on the two inputs, the isocost curve shifts outwards to FG; now a maximum of units
of labour and units of capital could be hired. Similarly, if the cost outlay rises further to R , the
isocost curve shifts outwards to MN, and the x-intercept and y-intercept increases to and
respectively. In all three isocost curves (BD, FG and MN), as wage and rent remain unchanged, the
slope of all isocost curves is the same, equalling – . (= –W/R = – / ). This explains why these
curves are parallel to one another, as shown in the figure.
FIGURE . Upward parallel shift of isocost curve
TABLE . Impact of changes in wage, rent and input expenditure on the isocost curve
Isocost curve
BD FG MN FG MN MD
Cost (C) 1 000 1 500 2 000 1 000 1 000 1 000
Wage (W) 50 50 50 33.33 25 25
Rent (R) 100 100 100 66.67 50 100
X-intercept of isocost curve 20 30 40 30 40 40
Y-intercept of isocost curve 10 15 20 15 20 10
Slope of isocost curve –0.5 –0.5 –0.5 –0.5 –0.5 –0.25
Now we rather assume that, instead of an increase in cost outlay, both the prices of capital and labour
decrease by the same extent. If cost remains constant at R , but the wage decreases from R to
R . while the rental price of capital drops from R to R . (that is, a . % price decrease in
both variables), there will be a parallel shift of the isocost curve from BD to FG. Similarly, if labour
wage and rental price of capital drop to R and R respectively (that is, a % price decrease in
both variables), there will be a parallel shift of the isocost curve from BD to MN. Finally, Figure .
shows that if cost outlay and the rental price of capital are unchanged at R and R respectively
but it is only the hourly wage that drops from R to R , the isocost curve would swivel or rotate
rightwards along the x-axis from BD to MD, as the maximum quantity of labour that could be hired
increases from to but the maximum quantity of capital that could be hired remains constant at
units.
FIGURE . Rightward swivelling of isocost curve along the
x-axis as wage decreases
Profit maximisation
As in the analysis of utility maximisation in Chapter , profit will be maximised at the point where a
firm produces a specific level of output at the lowest cost. Figure . illustrates an isoquant q , as
well as an isocost curve C (which is the same as BD, using the information from Table . ). Unless
the price of labour or capital declines, the employer cannot produce units at a lower cost than R
. The employer thus cannot be at any point below the curve C . Therefore, with an isocost curve of
C and an isoquant q , point e will be the best combination of capital and labour (i.e. with the lowest
cost) to produce q ( ) units of production. At this point, the isoquant curve is tangential to the
isocost curve, or the slopes of the two curves are equal. In absolute terms, it means
. This could be rewritten as which means factors the last
ΔK M PE W M PE M PK
MRTS = = = =
ΔE M PK R W R
rand spent on labour yield as much output as the last rand spent on capital.
Points T and S will not represent a profitmaximising position, because they lie on a higher isocost line
(and represents a more costly option – costing the firm R instead of R ), despite the fact that
they also lead to the production of units.
If the firm wants to increase production from to units, the employer could minimise the
production cost at e where isoquant q is tangential to the isocost curve C ; similarly, the cost-
minimising point is located at e where isoquant q is tangential to isocost curve C if the firm aims to
increase production further to units. Ultimately, the locus of all such cost-minimising points (e , e
and e ) gives the firm’s longrun expansion path, as indicated by the dotted line in Figure . .
FIGURE . The firm’s optimal combination of inputs
The labour-capital combinations at each cost-minimising equilibrium point can be summarised as

follows (W = and R = at all points):
Point e : TP = q = ,E= , K = , C = WE + RK = ( )( ) + ( )( ) = R
Point e : TP = q = ,E= ,K= ,C=( )( ) + ( )( ) = R
Point e : TP = q = ,E= ,K= ,C=( )( ) + ( )( ) = R
You must deodorise profits and make people understand that profit is not something offensive, but as
important to a company as breathing.
S P P
Finally, while the derivation of the firm’s longrun labour demand curve will be discussed later, we
assume at this stage that it is downward-sloping. While the hourly wage remains unchanged at R ,
the increase in cost outlay to produce more would lead to a rightward shift of the demand curve from
D to D , as the quantity of labour demanded increases from to (as the firm’s total input cost
increases from R to R to increase production from to units), as shown in Figure
. . The demand curve shifts rightwards again from D to D as the firm hires labour hours to
increase output from to units while total input rises from R to R .
FIGURE . Shift of the firm’s labour demand
curve due to an increase in input spending
Is it more costly to expand in the short run?
Once again, assuming initially W = ,R= , and the firm minimises production at e (spending a total of R on
the two inputs, with E = and K = ) to produce units of the good. In the long run, since it is possible to change both
labour and capital quantities, the firm would be able to produce units of the good at e . The isoquant (q ) and isocost
(C ) curves are tangential to each other, and production cost is minimised to R , as explained earlier.
FIGURE . The firm’s short-run and long-run

expansion paths
Now we assume that in the short run, capital is fixed at five units; under this constraint, the firm could only produce
units of the good at e (see Figure . ). This point is associated with the isocost curve, C , which lies to the right of C .
Also, the isoquant curve (q ) is not tangential to C . This means that compared to the long-run equilibrium point e , a
higher total cost is involved in the short run to produce the same output quantity, due to the fact that the two inputs
cannot be substituted. It can also be seen from the figure that e is associated with a much higher labour quantity
compared with the long-run equilibrium point e . For instance, e is associated with the total cost of R (when W =
,E= ,R= and K = ), while e is associated with a higher total cost of R (when W = ,E= ,R=
and K = ). To conclude, it is relatively more costly to the firm to expand production in the short run.
Now we examine what happens when the wage decreases, holding the rental price of capital constant,
with the aid of the information on the isocost curves BD and MD as shown in Table . . The
downward-sloping short-run labour demand of the firm will be derived in the process.
Over the short term, the demand for labour is affected only by a change in wages. This influences the
employer’s production costs, which in turn influence the price of the product and therefore the
demand for the product. The employer cannot really replace (or substitute) labour with machinery at
short notice to compensate for any change in wages. However, over the long term, the employer
might replace labour with capital (machinery). These two effects are called the scale effect and the
substitution effect.
Scale effect. This is the expansion of output that takes place when an input becomes less expensive. For
example, as wages fall, the marginal cost (wage) falls below the value of marginal product. This
creates a profit incentive for the employer to expand output by employing more labour and more
capital in the long run. In equation terms, this means < , < and < , meaning E, K
ΔE ΔK Δq
ΔW ΔW ΔW
and q all increase as a result of the decline in W.
Substitution effect. This relates to the change in employment when both labour and capital are
variable, and a decrease in the relative price of labour results in a substitution of one resource (e.g.
labour) for the other (e.g. capital) without a change in production. In equation terms, this means q̄
ΔE
ΔW
< and ΔK
ΔW
q̄ > , meaning E increases, K decreases and q remains constant after the decline in W.
Figure . illustrates the impact of the wage decline (from R to R ; the rental price of capital
remains unchanged at R ) on hiring of capital and labour as well as output in the long run. At the
original equilibrium point P, the isoquant q is tangential to the isocost line C . Production cost is
minimised at this point to produce q units of output by hiring the labour-capital combination at this
point. The focus of our discussion is on labour, and it is seen from the figure that at point P, work
hours are hired. The isocost curve becomes flatter as a result of the wage decrease. The firm moves to
point R, producing a higher q output level and hiring more labour hours, and the input cost is
minimised at this point, where q is tangential to C .
FIGURE . The effect of a decrease in wage on employment
in work hours in the long run
While the movement from P to R represents the total effect of wage decrease on the hiring of labour
(and even capital and output), it helps to think of this movement as a two-step procedure to
decompose the impact of a wage decrease into scale and substitution effects.
To distinguish the scale effect, suppose a dotted line (DD) that is parallel to the original isocost line
but is tangential to new highly-lying isoquant curve (q ) is drawn. This dotted line generates a
tangency point Q. The movement from P to Q stands for the scale effect, and essentially answers the
question: what would have been the change in labour and capital if the firm reached a higher
isoquant with an increase in cost outlay instead of a decrease in wage? As long as labour and capital
are “normal” inputs, the scale effect would result in the firm hiring more of both inputs. In other
words, although in actual fact wage decreases, the movement from P to Q assumes that wage is held
constant, but the firm produces a high output level of q by simply assuming the employer can afford
to spend more on both inputs. It can be seen from both Figure . (a) and Figure . (b) that the
scale effect causes labour to increase (from to in the first figure, and from to in the second
figure); output level increases from q to q ; capital level also increases.
On the other hand, the movement from Q to R represents the substitution effect, which illustrates
what happens to the firm’s hiring of labour and its capital as the wage decreases, holding output
constant. By moving along the isoquant q , the firm’s total output remains unchanged at both points
Q and R. As the wage halves from R to R , labour becomes relatively cheaper, so the firm would
hire more labour but invest less capital to maintain the same output level. Focusing on labour, Figure
. (a) and Figure . (b) show that the substitution effect results in the firm hiring more labour
(from to , and from to units, respectively). Finally, the capital level decreases as a result of
the substitution effect.
In both figures, the wage decrease causes both labour and output quantities to increase. However,
there is a big difference between the two figures with regard to the impact of a wage decrease on
hiring of capital. In Figure . (a), since the substitution effect is more dominant (the length of the
arrow represents the strength of each effect), the firm hires less capital at the end. This implies capital
and labour are gross substitutes. In contrast, in Figure . (b), since it is rather the scale effect that
dominates, the firm hires more capital at the end. In other words, both capital and labour increase at
the end (when considering the total effect – movement from P to R) only in the second figure. This
means capital and labour are gross complements.
Table . summarises what happens to labour, capital and output quantities as a result of a wage
decrease in each case.
TABLE . Impact of a wage decrease on quantities of labour and capital demanded
Case [I] – Figure 3.19(a) Case [II] – Figure 3.19(b)

Scale effect
Labour (E) Positive; labour increases Positive; labour increases
Capital (K) Positive; capital increases Positive; capital increases
Output (q) Output increases Output increases
Case [I] – Figure 3.19(a) Case [II] – Figure 3.19(b)
Substitution effect
Capital (K) Negative; capital decreases Negative; capital decreases
Output (q) Output remains constant Output remains constant
Total effect
Capital (K) Negative; capital decreases Positive; capital increases
Output (q) Output increases Output increases
Relationship between labour and capital Gross substitutes Gross complements
Finally, the long-run labour demand curve of the firm could be plotted by using the information at
points P and R in Figure . . In both cases, the demand curve is downward sloping (see Figure
. ), as the wage decrease causes an overall increase in the quantity of labour demanded. However,
such an increase is relatively greater in Figure . (a), when the wage decrease causes the firm to
hire less capital. That is, the firm’s long-run labour demand is flatter (more elastic to wage change) if
the two inputs are gross substitutes.
FIGURE . Movement along the firm’s labour demand curve

due to a decrease in wage
. . The elasticity of labour demand
Over the short term, the demand for labour is affected only by a change in wages. This influences the
employer’s production costs, which in turn influence the price of the product and therefore the
demand for the product. The employer cannot really replace (or substitute) labour with machinery at
short notice to compensate for any increase in wages. However, over the long term, the employer
might replace labour with capital (machinery), with the above-discussed substitution and scale effects
taking place.
This means that over the long term the demand for labour will be more sensitive (or more elastic) to
changes in wages than in the short term. If wages decrease, the employer is able to replace capital with
labour in the long run, and this means that the wage change has had a greater effect on the demand
for labour. The firm’s long-term curve is thus described as being more elastic than the short-term
curve. This is shown in Figure . by a change in the gradient of the demand curve.
FIGURE . The firm’s short-run and longrun
labour demand curves
The line DS represents the short-term (inelastic) demand for labour and DL the long-term (elastic)
demand. If the market wage decreases from W to W (e.g. because of the abolition of minimum
wages), then employment over the short term will move along the line DS, i.e. will rise from E to E . In
the long term, however, employers might be able to switch to labourintensive technology, which will
increase employment even further. Employment will move along the line DL, i.e. the demand for
labour will reduce even further to E .
The elasticity of labour demand measures the responsiveness of the quantity of labour demanded to
changes in the wage rate, or the percentage change in work hours demanded by the firm associated
with a one percentage change in wage rate. In equation terms, it is equal to:
%ΔE ΔE/E ΔE W
= = ⋅
%ΔW ΔW /W ΔW E
The elasticity must be negative in both the short and long run, due to the inverse relation ship
between the wage and quantity of labour demanded, as already proved in sections . . and . . .
The greater the absolute value of the elasticity, the more responsive the quantity of labour demanded
is to changes in the wage rate. If the elasticity is greater than (smaller than) one in absolute terms, the
labour demand is said to be elastic (inelastic). Finally, as the long-run labour demand is more elastic,
the elasticity of labour demand in absolute terms is greater.
Table . presents a hypothetical example of the elasticities of labour demand of two firms. For
example, in the short run, firm A demands five labour hours at an hourly wage of R , but increases
the quantity of work hours demanded to six as a result of a wage decline to R . Hence, the short-
run elasticity of labour demand of this firm is equal to – . (negative, elastic) and is calculated as
follows:
TABLE . The elasticity of labour demand – a hypothetical example
Firm A Firm B
Short run Long run Short run Long run
W1 120 120 200 200
W2 100 100 100 100
E1 5 5 5 5
E2 6 8 7 8
Elasticity of labour demand –1.2 –3.6 –0.8 –2.8

Interpretation Elastic Elastic Inelastic Elastic
W (6−5) 120 120 120

ΔE 1
⋅ = ⋅ = ⋅ = = −1.2
ΔW E (100−120) 5 −20 5 −100
In the long run, the firm is more responsive to this wage decline, and increases the quantity of work
hours demanded to a higher eight hours. The long-run elasticity of labour demand of this firm is
equal to – . (negative, elastic) and is calculated as follows:
W (8−5) 120 3 120 360

ΔE
⋅ = ⋅ = ⋅ = = −3.6
ΔW E (100−120) 5 −20 5 −100
Comparing the two answers, the firm’s labour demand is more elastic in the long run.
Using a similar approach, the elasticities of labour demand of firm B can be calculated, and once
again, demand is more elastic in the long run.
. . The elasticity of substitution

The size of the firm’s substitution effect is dependent on the curvature of the isoquant. While section
. . illustrated the typical convex, downward-sloping isoquant curve, two extreme situations are
discussed in this section.
First, Figure . (a) presents a linear isoquant (we want to remind you again that this is an isoquant
but not an isocost curve) with a slope of – . . This means output remains constant at q whenever two
units of labour are laid off and are replaced by one unit of capital stock. This “exchange rate” between
the two inputs is constant ( K: E) along the whole isoquant. That is, the marginal rate of technical
substitution is constant instead of diminishing.
Figure . (b) shows the L-shaped, right-angled isoquants curves. The isoquant q suggests that
using work hours and capital unit yields q units of output (for example, a bus driver needs to
drive one bus carrying passengers for hours to reach a destination). If work hours are held
constant at , adding more capital has no impact on output; this makes sense as no one is available
to drive the second, third and fourth buses. Similarly, if capital is fixed at one unit, adding more work
hours has no impact on output – if there is only one bus available, the second (which represents the
th to th work hour) and third (which represents the th to th work hour) drivers are simply
redundant, as only hours (one driver) are needed to drive the only bus that is used by the firm.
FIGURE . Isoquants when inputs are perfect substitutes or

perfect complements
When the two inputs can be substituted at a constant rate (a linear isoquant curve), these two inputs
are called perfect substitutes; when the two inputs cannot be substituted at all (an L-shaped isoquant
curve), these inputs are called perfect complements. In between these two extremes there are
numerous substitution possibilities, depending on the curvature of the isoquant curve. The more
curved the isoquant, the smaller the size of the substitution effect.
To measure the size of this effect, the elasticity of substitution is calculated. This elasticity means the
percentage change in the capital-to-labour ratio resulting from a one percentage change in the
relative price of labour, holding production constant. In equation terms, it is presented as:
%Δ(K/E) Δ(K/E)/(K/E) Δ(K/E) W /R

= = ⋅
%Δ(W /R) Δ(W /R)/(W /R) Δ(W /R) K/E
As the relative price of labour decreases (for instance, due to wage decline), the substitution effect
tells us that the capital-to-labour ratio decreases (that is, the firm hires less capital but more labour).
The elasticity of substitution, therefore, is defined so that it is a positive number. If it is smaller
(greater) than one, substitution is inelastic (elastic) to the relative labour price change. If the elasticity
is equal to zero, the two inputs are perfect complements.
Table . presents a mathematical example of the elasticities of labour substitution of two

hypothetical firms. The elasticities are calculated as follows for each firm:
Firm A:
Δ(K/E) W /R (0.5−1) 1 −0.5
⋅ = ⋅ = = 2
Δ(W /R) K/E (0.75−1) 1 −0.25
Δ(K/E) W /R (0.33−1)
Firm B: 1 −0.67
⋅ = ⋅ = = 2.67
Δ(W /R) K/E (0.75−1) 1 −0.25
TABLE . The elasticity of substitution – a hypothetical example
Firm A Firm B
Before wage After wage Before wage After wage decrease
decrease decrease decrease
Q 50 50 50 50
W 200 150 200 150
R 200 200 200 200
E 100 160 100 180
K 100 80 100 60
W/R 1.00 0.75 1.00 0.75
K/E 1.00 0.50 1.00 0.33
Elasticity of 2.00 2.67
substitution
Interpretation Elastic Elastic
Hence, the elasticity of substitution is more elastic for Firm B. A possible reason is that the substitution
effect is stronger, so when the wage decreases, the extent of capital decrease and labour increase is
relatively greater when compared to firm A. Hence, the K/E ratio after the wage decline is much
smaller in firm B, and the change of K/L is greater, thereby causing a more elastic substitution
estimate.
. . Marshall’s rules of derived demand

The higher elasticity of labour demand over the long term has important implications for the labour
market. To determine to what extent wage increases (and other factors increasing the cost of labour)
affect employment, account must be taken not only of short-term influences, but also of the longer-
term effects of changes in the cost of labour and capital. The Marshall’s rules of derived demand
explain that the labour demand is more elastic under the following circumstances:
The greater the elasticity of substitution

As already explained in section . . , the employer might replace capital with relatively cheaper
labour. However, the extent to which this is technically possible will also determine the elasticity of the
demand for labour. The greater the elasticity of substitution, the greater will be the elasticity of the
demand for labour.
The greater the elasticity of demand for the good

The elasticity of labour demand will be higher if the consumers’ demand for the product is more
sensitive (i.e. more elastic) to price changes. A wage decrease can reduce the price of the output, and
if the demand for that output increases significantly because of the reduced price, the demand for
labour will also increase significantly.
The greater the labour’s share of total input costs

If the share of labour cost is high, a decrease in wages will lead to a proportionately bigger decrease
in the price of the product, which will encourage consumers to increase their purchases. This will have
a greater impact on the demand for labour.
The more elastic the supply of other factors of production (e.g. capital)
For instance, assume a wage increase takes place and firms are eager to replace labour with capital. If
the supply curve of capital is elastic (flat), an increase in demand for capital would only result in a
small increase in the rental price of capital. This means the economic incentives for moving along the
isoquant (hiring more capital but less labour) are greatly enhanced. Therefore, it can be said that it is
quite profitable for the firm to replace labour with capital, as there is no substantial increase in the
cost of capital as a result of increase in demand for capital stock.
. . The market labour demand curve

While section . . showed that the market labour supply curve is equal to the horizontal
summation of the individual labour supply curves, one may think that the same approach could be
adopted to derive the market labour demand curve. For example, suppose that there are two firms in
an industry, and the number of workers demanded by each firm increases from to as the market
wage drops from R to R . It would seem that the market or industry labour demand curve can be
derived by simply summing up the quantity of labour demanded across the firms at each wage, and
one may conclude when the wage drops from R to R , the number of workers hired in the industry
rises from to , as shown in Figure . .
FIGURE . The derivation of the market labour

demand curve
However, the market demand for labour cannot simply be an aggregation of the individual demand
curves, because as the output of the market as a whole changes, it will have an impact on the price of
the product. For instance, let us assume that the wage rate in an industry as a whole declines for some
reason. The industry increases its output and thus its quantity demanded for labour (there is a
downward movement along the VMPE or the firm’s labour demand curve). However, the industry will
have to reduce the price of the product to sell the higher output. The lower price reduces the VMPE
and thus shifts the labour demand of each firm to the left again.
Going back to Figure . , when the wage rate is reduced from R to R , each individual firm will
increase the quantity of labour demand from to (that is, a downward movement along the
labour demand curve D ), and the output will also increase. However, because the industry as a whole
needs to sell this higher output, the price of the product is reduced to stimulate additional demand.
Since the price of the product declines, the VMPE curve (the firm’s labour demand curve) shifts left
from D to D in Figure . (a). This means that the quantity of labour demanded by each firm
increases only to but not to . This also means that for the market as a whole, when the wage
drops to R , the quantity of labour demanded by the firms in the market increases to (= + )
instead of (= + ).
The “real” market labour demand curve is the steeper, less elastic ZZ instead of DD, as shown in
Figure . (b). This has important implications for understanding the impact of unions. Normally, if
wages are increased, employment reduces. The flatter (i.e. more elastic) the slope, the more jobs will
be lost as wages increase. However, if the unions control the whole industry, they face the market
demand for labour, which is steeper (ZZ). The steeper curve means that the demand for labour will
not be affected as much by a wage increase as when an individual employer is involved. This is
discussed in greater depth in Chapter .
Wage increases through sectoral bargaining institutions, such as bargaining councils, thus reduce
employment to a lesser extent than if the same wage increases were achieved through bargaining at
individual company level. However, this ignores the broader picture: precisely because employers are
able to increase prices in response to a wage increase across the sector, it has a negative impact on
the non-participating consumer and work-seeker, and eventually probably the economy. This issue is
discussed in greater depth in section . .
. . Factors influencing market labour demand

The market labour demand curve is downward sloping, meaning there is an inverse relationship
between the wage and the quantity of labour demanded, as shown in Figure . . If wage increases
from W to W , the quantity of labour demanded decreases from Q to Q along the market demand
curve D ; conversely, if wage decreases from W to W , the quantity of labour demanded increases
from Q to Q along the same curve. In other words, there is a movement along the market labour
demand curve as a result of the market wage change.
FIGURE . The market labour demand curve
The market labour demand curve could also shift. For instance, if labour becomes more productive,
the labour demand increases and the market labour demand shifts rightwards from D to D . At the
market wage of W , the shift of the curve increases the quantity of labour demanded from Q to Q
units. If labour productivity declines due to factors like continuous absenteeism from work, the market
labour demand shifts leftwards from D to D , and consequently at the market wage of W , the
quantity of labour demanded drops from Q to Q .
There are numerous factors causing the shift of the market labour demand curve, and they are
summarised in Table . . For example, assuming there is no jobless growth, economic growth (an
increase in real GDP as a result of, for instance, an increase in consumption, investment, government
spending and net exports, according to the Keynesian multiplier model we learnt in introductory
macroeconomics) should lead to an increase in labour demand, that is, real GDP growth is associated
with job creation. It was also discussed in section . . that the firm’s labour demand would increase
if price or marginal product of labour increases, and so this would consequently lead to an increase in
the market labour demand. Also, firms in the market are generally discouraged from hiring more
workers as the strike incidence increases (they may prefer to replace labour with capital, if this is
possible). Finally, if capital becomes relatively more expensive, the market demand for labour would
increase if the two inputs are gross substitutes.
TABLE . Factors affecting the market labour demand – a summary
Determinant Change Effect on the market labour Correct description of the

demand curve effect
Market wage (holding cost of capital constant) Increase Upward movement along the A decrease in the quantity of
demand curve labour demanded
Decrease Downward movement along An increase in the quantity of
the demand curve labour demanded
Economic growth (assuming jobless growth Increase Rightward shift of the An increase in labour
does not take place) demand curve demand
Decrease Leftward shift of the demand A decrease in labour demand
curve
Labour productivity Increase Rightward shift of the An increase in labour
demand curve demand
curve
Per-unit selling price of the good Increase Rightward shift of the An increase in labour
demand curve demand
curve
Incidence of strikes Increase Leftward shift of the demand A decrease in labour demand
curve
Decrease Rightward shift of the An increase in labour
demand curve demand
Indirect cost of labour (e.g. due to introduction Increase Leftward shift of the demand A decrease in labour demand
of new labour legislation) curve
demand curve demand
HIV and AIDS infection rate Increase Leftward shift of the demand A decrease in labour demand
curve
demand curve demand
Cost of capital (assuming capital and labour are Increase Rightward shift of the An increase in labour
gross substitutes) demand curve demand
Determinant Change Effect on the market labour Correct description of the
demand curve effect
curve
Cost of capital (assuming capital and labour are Increase Leftward shift of the demand A decrease in labour demand
gross complements) curve
demand curve demand
The demand for labour in South Africa has not been sufficient to provide enough jobs for those
wanting to work and has, except for a few years, not even been sufficient to create enough jobs for
those entering the labour market for the first time. Over time, therefore, both the unemployment level
and rate have generally been increasing, despite the fact that employment in absolute terms has been
increasing. For the remainder of this section, four of the factors that influence the labour demand in
South Africa are discussed in some detail.
The rate and nature of economic growth

Had the country’s economic growth been higher than the annual average of . % experienced since
the transition to a full democracy, the country would probably have been able to create many more
jobs. However, the economic growth rate has not been high enough to reduce unemployment.
Nevertheless, it should be noted that economic performance improved after – the economic
growth rate before the democratic elections was dismal, so the transition to democracy has not been
the cause of low economic growth rates. Furthermore, it seems as if the negative consequences of
adjusting to a globalised world might have dissipated, especially since , and the country seems to
have experienced much higher increases in formal and informal employment.
An important caveat that should be mentioned here is that the job creation ability of the economy
declined substantially during the s and s. This was partly a result of higher capital intensity.
Trade liberalisation
This is discussed in greater detail in Appendix A where it is indicated that trade liberalisation might
have contributed to the loss of unskilled jobs (Nattrass, a). However, it is unlikely to have been
the main cause of job losses overall. In fact, it is not so much increased imports (and thus domestic
producers losing markets) that have caused job losses, but that the export sector became much more
capital intensive and thus shed jobs. This occurred probably to increase productivity and enable
exporters to compete better against foreign competitors. Trade liberalisation also increased the
demand for skilled workers, while simultaneously resulting in fewer jobs for unskilled workers.
The impact of unions and bargaining councils on labour costs

High wage increases have also contributed to the reduced demand for labour. As section . . shows,
most studies indicate that an increase in wages will reduce employment, and this is particularly true in
South Africa. The high real wage increases over the past number of years can probably be ascribed to
strong unions in the country and to the system of bargaining councils, as discussed in sections . .
and . . Appendix A will also show that the countries that have been most successful at creating jobs
recently have done so by introducing wage moderation as part of a package of policy measures. High
wage increases in South Africa have especially discouraged the use of unskilled or lower-skilled
workers, i.e. those with the greatest need for job creation.
Even the ILO has criticised the system of bargaining councils in this country, as shown in section . .
Several studies have found that the South African system of partially centralised wage bargaining
results in the poorest outcomes in terms of job creation and inflation compared to highly centralised
systems, partially centralised but highly coordinated systems, or highly decentralised bargaining
systems.
The relatively high strike incidence of unions in South Africa will also have discouraged the use of
unskilled labour in particular. As indicated in section . , the incidence of strikes in this country has
declined recently, but is still relatively high.
Labour legislation that increases the indirect cost of labour or hinders productivity increases
There are also other reasons for the increase in labour cost, which has reduced the demand for labour.
The introduction of new labour laws has significantly increased the indirect cost of labour. The most
significant is the Basic Conditions of Employment Act, which has increased the hourly cost of labour
by, among other things, reducing hours of work, increasing overtime premiums and various other
provisions. More details on the increased cost of labour are discussed in section . .
In several ways the labour laws have reduced flexibility, thus making it more difficult for employers to
increase productivity. For instance, many rules increase job security with regard to individual
dismissals and retrenchments. The fact that unions may strike about retrenchments limits the ability of
employers to adjust rapidly to changing circumstances.
The impact of labour legislation also differs based on the size of an enterprise. Larger enterprises can
more easily comply with regulatory requirements as their scale allows, e.g. in-house legal advice.
Smaller enterprises are disproportionately affected, as the owners often do not have the resources to
hire or outsource legal advice. This is especially problematic with CCMA referrals, as aggrieved
employees who have been dismissed can institute action without having to prove prima facie fault by
the employer, which differs from the conventional approach to civil litigation, which requires some
proof before a court can hear the case. Employers who face frivolous CCMA referrals still require time
and resources to attend and participate. Again, the impact on smaller enterprises is far more
pronounced, as business owners may have no recourse but to attend in person, often forcing the
business to close on the day of the hearing.
However, certain assumptions about the impact of labour legislation are not always fully borne out in
practice. For instance, Lewis ( : ) reports on a survey done in by the World Bank in
conjunction with the Greater Johannesburg Metropolitan Council among large firms across eight
manufacturing firms. Around % of all the firms initially reported that individual regulations had had
no effect on employment decisions. However, when they looked at the combined impact of all
regulations, % indicated that they had hired fewer workers, had substituted labour with capital,
and/ or had hired more temporary workers. While % is a substantial percentage, the labour laws
seemed to have had no impact on the employment decisions of the rest of the employers, i.e. the
majority.
. . Imperfectly competitive markets: the monopolist

If the firm operates in a perfectly competitive product market, and hence has many competitors and
no control over product price, the marginal revenue per unit of output sold is equal to product price,
that is, MR = P. However, if the firm has some degree of monopoly power in its product market,
additional units of output can be sold only if the firm lowers the price of the product (as the firm faces
the market demand curve for its particular product); in this case, MR < P. In other words, in the case of
a wage reduction, the monopolist will employ more labour to produce more output, but will also
reduce the price of the product to sell the higher production.
Now the firm’s labour demand curve is no longer VMPE = P × MPE, the value of marginal product.
Instead, it is marginal revenue product of labour, MRPE = MR × MPE, where MR is the marginal
revenue from selling an additional unit of product. Perfect competition in the goods market (as
discussed in section . . ) is a special case in which VMPE equals MRPE, because P = MR. Since MR <
P in monopoly, this means the monopoly firm’s MRPE curve lies to the left of the perfectly competitive
firm’s VMPE curve, as shown in Figure . .
FIGURE . Labour demand in a

noncompetitive product market
The figure also indicates that when wage declines from W to W , employment increases from E to E
for the perfectly competitive firm, but employment increases to a lesser extent in the case of a
monopoly firm (a smaller increase from E to E ). Employment by the monopolist is thus less sensitive
to wage reductions (and similarly to wage increases). Finally, the employment before and after a wage
decrease is lower in the case of monopoly, and this result is expected, as the monopolist
underproduces compared to a perfectly competitive firm, as we learnt in introductory
microeconomics.
Why is there only one Monopolies Board?

G
SUMMARY
The market demand for labour is a derived demand; it depends on the demand for the product or service produced by
that labour.
The labour quantity demanded will decline as wages increase, and the decline will be larger over the long term,
because employers have more options to adjust to the higher wages, for instance by becoming more capital intensive.
The long-run demand for labour is thus more elastic (sensitive) than the short-run demand.
The short-run and long-run labour demand curves are both downward sloping, but the long-run curve is flatter.
Assuming a perfectly competitive labour market and goods market, in the short run, the quantity of labour demanded is
derived at the point where the wage equals the value of the marginal product of labour. In contrast, in the long run, the
quantity of labour demanded takes place at the point where the isoquant curve is tangential to the isocost curve.
Production cost is minimised at this tangency point.
The market demand for labour is less elastic (less sensitive) to wage changes, because the resultant changes to the
price of output reduce the impact of wage changes on the demand for labour. In other words, the market labour
demand curve cannot be derived simply by the horizontal summation of the individual firm’s labour demand curves.
KEY CONCEPTS
capital-to-labour ratio
cost minimisation
derived demand
elasticity of labour demand
elasticity of substitution
employment elasticity of economic growth
employment rate
employees
employers
expansion path
formal sector
gross complements
gross substitutes
imperfect competitive market
informal sector
isocost line
isoquant
labour absorption capacity
marginal rate of technical substitution (MRTS)
Marshall’s rules of derived demand
monopoly
perfectly competitive market
perfect complements
perfect substitutes
profit maximisation
scale effect
substitution effect
value of marginal product
FOR STUDENTS
. Discuss the law of diminishing marginal product and illustrate graphically how this law can be used to derive
the short-term demand for labour.
. The table below shows the number of cakes that could be baked daily at a local bakery in the short run (i.e.
capital is the fixed input while labour is a variable input).
Labour hours (E) Number of cakes (TP) MPE VMPE

0 0 – –
1 12
2 20
3 25
4 28
5 30
. Calculate the marginal product of labour by filling in the third column of the table.
. Suppose each cake sells for R . Calculate the value of the marginal product of labour by filling in the
fourth column of the table.
. Draw the value of the marginal product of labour curve, which is also the demand curve for bakers.
Clearly label the x-axis and y-axis.
. If the hourly wage is R , how many labour hours will be demanded, given that the goal is to maximise
profits? How many cakes will be baked and sold each day?
. Discuss the concept of an isoquant curve and indicate some of its most important characteristics.
. Mo-Mo Rappy, a famous R&B singer, failed her undergraduate Labour Economics when she was a student.
Until today, she still gets confused between labour supply elasticity, short-run and long-run labour demand
elasticities, and elasticity of substitution between labour and capital. Explain each elasticity concept, with the
aid of the following tables:
Wage Work hours Work hours demanded in the short run Work hours demanded in the long run
(W) (H) (E) (E)
Period 1 200 8 15 15
Period 2 300 13 12 10
Firm (A) Wage Rent (R) Labour (E) Capital (K)
(W)
q = 50 200 200 50 10
q = 50 200 150 40 13
Firm (B) Wage Rent (R) Labour (E) Capital (K)
(W)
q = 50 200 200 50 10
Wage Work hours Work hours demanded in the short run Work hours demanded in the long run
(W) (H) (E) (E)
q = 50 200 150 30 18
. Jimmy opens a shop to sell sushi. In the long run, both labour and capital are variable inputs. The rent for
using machinery is always R per day, while the hourly wage of labour decreases from R to R .
. Use the information from the graph below to plot the long-run demand curve for labour. Clearly label the
wage and quantity of labour demanded at each equilibrium.
. Explain the substitution and scale effect on labour, capital and production as a result of the wage
decrease, with the aid of the above figure.
. Explain the Marshall’s rules of derived demand.

. Describe what influence each of the following would have on the market demand for labour, and illustrate
graphically:
. Recession in the South African economy.

. The workers take long lunch and bathroom breaks at work.
. Trade unions demand for workers being paid an end-of-the-year trip to African countries annually by the
employers.
. The hourly wage of labour increases from R to R .
. The selling price of the good increases
. What is the difference between the demand for labour by the individual employer and the demand for labour
by the market as a whole? Illustrate graphically.
. Explain the difference between visible and invisible underemployment with the aid of an example in each case.
. Discuss the differences in the demand for labour by an employer in a perfectly competitive market and that
of a monopoly employer. Illustrate graphically.
SUGGESTED READING*
( ): December.
Bhorat, H. b. Employment trends in South Africa. Johannesburg: Friedrich Ebert Stiftung.
Casale, D., Muller, C. & Posel, D. . “Two million net new jobs”: a reconsideration of the rise in employment in
South Africa, – . South African Journal of Economics, ( ): December.
World Bank. . South Africa economic update: Jobs and South Africa’s changing demographics. Washington DC:
World Bank.
Non-economists are advised to read the comments in the preface before they become intimidated by the
analysis in this section, and particularly the graphs that the economists love to use. It is not essential to
comprehend this section fully in order to understand the rest of the chapter, but will help to achieve a
better grasp of how the labour market as a whole operates.
Wages and the cost of labour

. Introduction
. The determination of wages
. Earnings in South Africa
. Wages and inflation
. Wage differentials
. The trade-off between wages and employment
. The impact of labour laws on the cost of labour
. Deeper insight I: determination of wages in a perfectly competitive labour market
. Deeper insight II: determination of wages in a monopsony market
I would like to see a fair division of profits between capital and labor, so that the toiler could save enough to mingle a little
June with the December of his life.
R GI
. INTRODUCTION
One of the most important outcomes of labour markets is the determination of wages. Wages play a
major role in respect of the allocation of labour between occupations, sectors, regions and even
countries. They also play an important role with regard to decisions about education, and higher
education in particular.
Since they play such an important role in determining supply and demand, as discussed in previous
chapters, labour market outcomes, such as the oversupply of labour in the form of unemployment or
an undersupply in the form of labour shortages, are influenced by developments with regard to
wages and wage levels.
Therefore, careful note should be taken of how wages are determined, and the extent to which the
determination of wages in a particular country deviates from the perfectly competitive model, which
theoretically should not allow unemployment to rise to the levels at which it currently is in South
Africa.
This chapter begins to look at some of these factors, and some of the interactions between wages and
other factors, for instance inflation and employment. More of these factors are discussed in greater
detail in Chapter , i.e. the impact of unions and collective bargaining, bargaining councils and
minimum wages.
. THE DETERMINATION OF WAGES
Wages in a market economy are usually determined in different ways, and the strength of the various
factors that influence wage levels are different in each case. In some instances, the impact of market
forces can be quite strong, whereas in others, it may be less so. Some of the most important
mechanisms of wage determination are discussed below.
. . The determination of wages in a perfectly competitive labour market

The first method of wage determination is by agreement between an individual employer and an
individual employee, for example in a contract of employment. This is the classic example of the
operation of market forces and is also one of the assumptions of the model of a perfectly competitive
market. It is the predominant method for wage determination in sectors such as agriculture and
domestic service and in smaller enterprises not covered by bargaining councils or trade unions, as
well as among many casual and unregistered employees. However, even in these markets wages
cannot be set freely because of the obligation to pay a certain minimum wage to the lowest skilled
workers.
In perfectly competitive markets, equilibrium between the demand for labour and the supply of
labour will determine employment and wage levels. If there is excess demand for labour due to an
increase in production, it will drive up the wage rates and thus also the quantity of labour supplied
until equilibrium has been achieved. If there is unemployment (an excess supply of labour), it will drive
down the wage rates, the quantity of labour supplied will decline and the quantity demanded will
increase until eventually equilibrium is achieved.
The theoretical explanation of how the perfectly competitive labour market works is discussed
towards the end of this chapter in section . .
. . The determination of wages in a monopsony labour market

A market is called a monopsony market when the employer is a wage setter and can set wages by
adjusting the number of workers he or she hires. This occurs where there is only one employer or
where there is collusion or cooperation between employers in a certain market to fix wages at a
certain level (McConnell et al., : ). For example, a monopsony can develop if there is a large
company employing the majority of workers in a specific occupation, or where employers form a
united front in respect of wages, for instance in the case of a bargaining council.
As is shown in section . , equilibrium in this market might be reached at a point where both wages
and employment are lower than they would have been in a perfectly competitive market.
. . The determination of wages in a segmented labour market

The traditional analysis of the labour market does not succeed in explaining certain phenomena –
several segments of the labour market are not characterised by the models outlined above. Access to
the labour market for medical doctors, for instance, is very tightly controlled. Furthermore, within
many individual enterprises, factors and rules internal to these organisations rather than general
market forces often determine wages and the allocation of labour. For instance, appointment to a
specific job is more often determined by internal promotion than by open competition among all
possible candidates within and outside an enterprise. The wages of such workers are usually more
favourable than the market would determine. In addition, in direct contradiction to the traditional
analysis, widespread discrimination against some groups persists over time (even in supposed non-
discriminatory countries).
The segmented labour market theory was developed to explain these phenomena. According to this
theory, the labour market is permanently divided into a number of segments, each with its own
characteristics and mode of operation, and with little or no mobility between them. Market forces do
not always play the major allocative or pricing role in these segments, and can be replaced by
institutional forces, for example collective bargaining, which introduces rigidities such as employment
security and seniority rules. The interplay between employer and union also introduces wage
structures that are not always related to economic factors (e.g. market forces or worker productivity)
and these wage structures differ from one labour market segment to the next, depending on relative
union and employer strengths.
Another characteristic of segmented labour markets is that access to the better segments is controlled
and mobility between segments is very restricted. Good examples of such segments are the labour
markets for various professions, such as medical doctors and lawyers, where access is tightly
controlled.
The dual labour market

A special case of the segmented labour market is the dual labour market. This market is characterised
by two distinct non-competing markets, i.e. the primary and secondary labour markets. Each of these
is described here.
The primary segment. This segment is characterised by formal employment, high earnings, good
working conditions and employment stability. It often consists of large enterprises or segments of the
labour market where trade unions play a major role. The primary market can also be divided into a
number of internal labour markets, which could refer to individual enterprises or occupations, for
example multinational corporations or the labour markets for lawyers or medical doctors. These
markets are characterised by the following:
Access is usually severely restricted and is only possible at the bottom level.
Skilled jobs are more often than not filled by promotion from within the enterprise.
Promotion is often determined by bureaucratic and rigid rules of the company or occupation, which
may have been established through collective bargaining.
Since there is substantial job security, the internal labour market tends to lack market mechanisms
for the allocation and pricing of labour. This is because workers in these markets are protected
against the competitive market forces of supply and demand and their behaviour is shaped by
internal rules (McConnell et al., : ). The wage rate is thus not necessarily the rate that would
have been determined by demand and supply conditions, i.e. market forces, if they had operated
normally.
Employers in the primary market spend large amounts of money on training and even further
education of their workers, partly because of pressures to improve skills and productivity, which can
be ascribed to high and rising wages. These high wages also result in more laboursaving and
productive equipment being used. This, in turn, generates improved advancement and earning
opportunities for the remaining workers, creating a self-reinforcing process of progress and
development in the primary market, although at the expense of employment creation (McConnell et
al., : ).
The workers are trained mainly for the “internal” market, and they can often be utilised only by that
specific employer and by close competitors. This means that employers rely mainly on the internal
market, i.e. on workers within the enterprise or workers of close competitors, to fill vacancies.
The secondary segment. Workers “on the outside” often have little or no job security, little prospect of
promotion and generally poor conditions of employment. This is the so-called “secondary segment”,
which consists of jobs that are informal, low-paying and have unstable patterns of employment.
Secondary segment enterprises are typically over-competitive, small in scale and labour-intensive with
a low level of unionisation. There are low levels of skills in this sector, which means employers have
little incentive to encourage stable employment patterns. Employers also have little inclination to train
workers properly – wages are low enough for the employer not to pay much attention to productivity.
Therefore, in the secondary segment, “bad” jobs produce “bad” workers, while in the internal or
primary market, “good” jobs make “good” workers. Examples of secondary labour market jobs are
domestic workers, dishwashers in restaurants, cleaners, janitors and other menial jobs where a low
level of skill is required.
Workers in the secondary segment have little chance of entering the internal labour markets. They are
trapped “on the outside” and education and training by the government will do little or nothing to
increase their earnings.
According to the dual labour market theory, inequality in earnings cannot be drastically reduced by
investment in education and training unless workers in the secondary markets are integrated through
various measures into the primary labour market. This can be done, for example, through policies to
reduce discrimination, better labour market information and the reorienting of employers.
The use of the dual labour market theory to explain differences in earnings not necessarily related to
skills or qualifications is discussed in section . . , and to explain discrimination against certain
groups in section . . .
Hofmeyr ( ) points out that South Africa experienced an extreme form of institutionalised
segmentation of the labour market because of apartheid. He contends that this disappeared with the
labour market reforms of the s, but has been replaced by a form far more typical of some
advanced industrial economies. This later form of segmentation is between the unionised and non-
unionised parts of the formal sector, and between the formal and informal sectors. There are, for
instance, large differences in the earnings of unionised and non-unionised workers, as discussed in
section . . , and he concludes that “the principal source of segmentation is associated with
unionisation” (Hofmeyr, : ).
It has been found that transitions from the informal sector (as a proxy for the secondary labour
market) to the formal sector (the primary labour market) are surprisingly rare, especially in the case of
the youth. It is therefore crucial for a young person to obtain that first job to ensure future
involvement in the labour market. If one is in the one section, the formal section, one has a very good
chance of staying in this section. About % of the entrants into a formal sector job come from
another formal sector job. The other section consists of unemployed persons, discouraged work-
seekers, persons who are not economically active, workers in some types of atypical employment and
the informal sector. Within this chunk of the labour market, there is substantial dynamism, but none
of the labour market statuses in this category are very economically attractive (Banerjee, Galiani,
Levinsohn & Woolard, : ). A more recent study by Essers ( ), using the first two waves (
and / ) of the NIDS data, found that % of entrants into regular wage employment (a proxy for
the primary labour market) in the first wave remained in this sector in the second wave, but for those
initially involved in casual, non-regular employment (secondary labour market), only % eventually
found regular employment.
Policy issue: a dual labour market as a strategy to deal with the labour market crisis
Some years ago, the South Africa Foundation ( ) stated in a policy document that a two-tier labour market should be
encouraged to address South Africa’s unemployment problem. It argued that large numbers of relatively low-wage jobs
should be created by eliminating the extension of bargaining council agreements to non-parties (see section . );
maintaining only those minimum standards that do not hurt the poor and unemployed; avoiding minimum wages that
threaten jobs; and encouraging the creation of new jobs and the employment of the youth. Youth and regional
differentiation should be allowed in all wage contracts. According to the Foundation, the rights and privileges of existing
workers would “not be affected much”, as bargaining councils and strong centralised trade unions protect them.
The ANC’s Discussion Document to revitalise small and medium-sized enterprises (ANC, ) has also raised the
possibility of a dual labour market to accommodate more flexibility. The following are raised as possible elements of such
a policy:
Allowing younger people a more flexible set of arrangements with regard to minimum wages, limits on overtime and
dismissal requirements
Allowing industrial development zones where labour laws would be more flexible
Allowing more flexibility in labour-intensive sectors, such as tourism, textiles and clothing, the household and
agricultural sectors
Having more flexible labour laws for companies that employ fewer than people, as opposed to the current cap of
.
The debate boils down to one question: to what extent should certain labour laws be relaxed (e.g. bargaining council set
wages, and dismissal procedures) in order to grow low-skilled employment. It can be seen as implying that South African
labour legislation impedes job creation.
In the National Government released the National Development Plan (NDP), which is a long-term policy document
that also speaks to the labour market. The NDP, for example, proposes reviewing regulations and standards for small-
and medium-sized enterprises and an approach that simplifies dismissal procedures for poor performance or
misconduct.
SUMMARY
The dual labour market consists of two parts: the primary segment characterised by high earnings, good working
conditions and employment stability, and the secondary segment characterised by low-paying jobs and unstable patterns
of employment. Access to the primary market is restricted and often open only from the bottom.
. . Determination of wages through collective bargaining

Another way in which wages are determined is by collective bargaining, where unions and employers
or groups of employers bargain collectively on wages and other conditions of employment. This can
take place at enterprise or sectoral level, for example within a bargaining council. This is discussed in
detail in Chapter . There are also some sectoral collective bargaining arrangements where parties
bargain at sectoral level but not within a bargaining council, for instance gold mining, coal mining and
automobile manufacturing. There are many factors that determine the outcome of collective
bargaining. One of the most important is the power balance between the parties but, as will be
discussed, this does not mean that market forces do not play an important role.
. . Government involvement and influence

Government involvement in, and influence on, wage determination can take a variety of forms, of
which the following are pertinent in South Africa:
(a) The government sets an example as the single biggest employer in determining the wages of its
employees. In , wages in the public-sector bargaining councils were significantly higher –
nearly double – than in private sector bargaining councils (Bhorat, Van der Westhuizen & Goga,
: ).
(b) Legislative measures that determine minimum wages. In South Africa, the Minister of Labour can
make sectoral determinations for some lowly-paid, vulnerable sectors that are regarded as having
insufficient collective bargaining for the determination of wages. Section . deals with the issue
of minimum wages and is relevant here, as it is one of the methods through which the
government influences wage determination.
(c) Legislative measures that protect and encourage centralised collective bargaining. For instance,
the Labour Relations Act not only provides for the registration of formal collective bargaining
arrangements between representative employer organisations and unions in a particular sector,
but also provides for the Minister of Labour to extend these agreements to all employers in the
sector, subject to certain conditions. This is discussed in more detail in Chapter .
Legislation affecting other employment-related issues can also have an effect on wages. For instance,
legislation reducing work hours can have an impact on wages as unions attempt to prevent a
reduction in the take-home pay of their members or employers try to avoid an increase in the hourly
cost of production. The impact of reduced hours of work on the cost of labour is discussed in section
. . .
. . Communes and communism

Another way in which the allocation and pricing of labour can be organised is the commune, its
ultimate form being communism (Fearn, : ).
The “pure” commune (and therefore pure communist philosophy) is based on two principles (Fearn,
: ):
There is social, collective or public ownership of physical capital (e.g. land, machinery) and property.
This system is properly called socialism. In the former Soviet Union (under communist rule) this
principle was extended to include state ownership of the economy, i.e. authoritarian central
economic planning and decision making.
Consumption (including payment for productive effort, i.e. wages) is separated from productive
effort. The pure commune operates in terms of the principle from each according to his ability – to
each according to his need. Wages would therefore not be determined in the normal way, i.e.
through the joint operation of market forces, productivity and collective bargaining.
Under capitalism man exploits man; under socialism the reverse is true.
P
Regardless of the theory, labour markets in the Soviet Union under communist rule were a curious
blend of free choice and government direction. Workers were generally free to move, but there were
exceptions, for example: the assignment of graduates to a specific location for three years; the
mobilisation campaigns of workers and students for special tasks; state control over education; and
state determination and manipulation of wages. Trade unions were controlled by the Communist
Party and could exercise only such freedom as the party permitted (Marshall et al., : ).
Centralised economic planning also affects the application of capital, and by extension, the range and
scope of industries one is able work in relative to a capitalist free market system.
. . Other methods of wage determination

Various methods of worker participation and worker incentive payment systems can be used to
determine wages, for example: representation at the decision-making levels in the company; merit
awards; productivity or production-linked pay; or profitability-linked remuneration.
SUMMARY
The most important methods of determining wages are contracts of employment, collective bargaining, government
regulation and methods that link wages to productivity or profitability.
. EARNINGS IN SOUTH AFRICA
The structure of actual earnings in South Africa comes from the combined effect of the various ways
in which wages are determined. The main source of data on these earnings is the Statistics South
Africa surveys – SEE and QES for businesses and the OHS, LFS and QLFS for households (see sections
. . and . . ). Another source is the record of graduates’ earnings compiled by the Human Sciences
Research Council (HSRC). Private organisations also compile data on wages or earnings in South
Africa, for example the Labour Research Service (wage settlement data) and PE Corporate Services
(average earnings by occupation). Such data are usually available only to subscribers.
Money doesn’t buy friends, but it allows a better class of enemies.

L M
The original sampling frame of the SEE (which took place from to the first quarter of )
included private and public-sector firms with an annual turnover exceeding R , but excluded
firms in some industries, ranging from agriculture, restaurants and telecommunication services to real
estate and business services, educational services and health services (Altman, : ). In , the
sampling frame was extended to include all the previously omitted formal industries, apart from
agriculture and domestic services.
The QES, which was introduced in the last quarter of and replaced the SEE in , broadened
the sampling frame by including all firms registered for income tax, regardless of annual turnover.
Firms that pay some form of income tax without being registered for VAT would now be included in
the QES sampling frame. The QES sampling frame was revised in due to various factors, such as
the extension of boundary of the formal sector due to legislative changes, greater compliance with
administrative requirements to register as an employer, and better capturing of small businesses in
the registration process (Statistics South Africa, ).
Regarding the three household-level surveys (OHS, LFS and QLFS), information on earnings was
collected from all workers, regardless of whether they worked in the formal sector or informal sector,
and whether they were employees, employers or self-employed.
In section . . , average wages of formalsector employees are examined by using the QES data. In
section . . there is a further discussion on wages in South Africa, with the emphasis on wage and
income inequalities between different population groups.
. . Average salaries and wages in South Africa

Formal sector
Figure . gives an indication of the increases in current average monthly earnings during the period
– and compares these with inflation. It shows that current earnings increased much more
rapidly than inflation. This means that workers enjoyed a relatively sharp increase in average real
earnings (i.e. after allowing for inflation), as is apparent from the dotted line in the same figure. In fact,
the average real earnings increased by almost one-third (R in November versus R in
August , in December prices).
FIGURE . Indices of consumer price index, current average monthly earnings and real average monthly
earnings (including bonuses and overtime payments) in the formal nonagricultural sector, fourth
quarter to third quarter
Source: Author’s own calculations using the – Quarterly Employment Statistics and consumer price index data.
The increase in real earnings over the period – amounts to about . % per annum (the
annualised growth rates of the consumer price index and nominal earnings are . % and . %
respectively). However, as indicated in Figure . , labour productivity also showed an upward trend in
general. The relationship between nominal earnings, real earnings and labour productivity are
discussed in detail in Chapter .
Table . compares the real average monthly earnings levels in various sectors in August and
August . In both years, average earnings in mining, manufacturing, finance and real estate, and
community, social and business services are more or less in line with the average of all non-
agricultural sectors. The average earnings in construction as well as the wholesale and retail trade
sectors are, however, far below the overall average. Electricity is far above the overall average
(especially because of the high percentage of skilled workers in this sector). The earnings in the
transport and communication sector are also significantly above the overall average. Finally, looking
at the annualised growth rate of the average real monthly earnings, this rate is the highest in
construction ( . %) and mining ( . %), but the lowest in the transport and communication sector
( %). For the remaining five sectors, their respective annualised growth rates (ranging between .
and . %) are close to the overall annualised growth rate of . %.
TABLE . Average monthly earnings of non-agricultural formal employees by sectors in December

prices, and
2005 August 2016 August Annualised growth rate (%)

Rand Ratio to total Rand Ratio to total
Mining 13 812 101.0 18 001 122.3 4.5
Manufacturing 12 687 92.8 13 314 90.5 2.5
Electricity, gas and water 27 732 202.9 30 244 205.5 2.8
Construction 8 393 61.4 11 085 75.3 4.6
Trade 9 446 69.1 10 004 68.0 2.6
Transport and communication 20 350 148.9 18 272 124.1 1.0
Finance, real estate 14 499 106.1 15 693 106.6 2.8
Community, social and personal services 16 514 120.8 17 169 116.6 2.4
Total 13 671 100.0 14 719 100.0 2.7
Source: Author’s own calculations using the Statistics South Africa Quarterly Employment Statistics
. . Non-wage labour costs

According to a survey undertaken in the Greater Johannesburg Metropolitan Area in , in South
Africa non-wage costs formed about % of wage costs (Bhorat & Lundall, : ). In this survey,
non-wage costs refer to worker benefits such as overtime pay, housing allowances, transport
allowances and contributions to unemployment insurance, pension funds and medical aid funds. In
contrast, non-wage labour costs in the US and European countries form between and % of total
labour costs (Clarke, : ; Eurostat, ).
It cannot be denied that benefit funds contribute to substantial long-term social gains for workers
and protection against unstable economic circumstances, dangerous working conditions, high
medical costs and other eventualities. In addition, with the exception of unemployment insurance and
insurance for occupational injuries and diseases, there is no overarching social security or social
insurance system in South Africa, although a basic income grant has been proposed by some union
and community interests (see section . . ).
. . The income of graduates

Since , the HSRC has undertaken surveys to monitor the income structure of graduates who work
full-time and earn positive income. These data are gathered by postal survey among graduates, and
the income is presented according to occupation, employer sector, age and work hours. Data on
additional earnings, the value of fringe benefits and the income package are also provided.
The results of the most recent publicly available HSRC remuneration of graduates survey conducted in
found that graduates in the public sector seem to be earning less (about three-quarters) than
their counterparts in the private sector, although the gap appears to be narrowing. The remuneration
of white male graduates in the public sector is about half that of white male self-employed graduates.
Also, self-employed graduates appear to earn more than those in the private sector. However, the
self-employed graduates have an average of years’ work experience compared with years for
graduates in the private sector.
As the QLFS captures information on monthly earnings from the main job (before taxes and other
deductions for employees, and after expenses in the case of employers and self-employed) and
educational attainment, it is an alternative data source to examine the remuneration of graduates. By
defining graduates as those with post-Matric qualifications and after excluding graduates who
reported zero or extremely large earnings (i.e. the outliers), Table . shows that in , graduates
who are employers earn more than those who are employees and self-employed. Interestingly, the
graduates working in the private sector earn higher mean income (R ) but lower median income
(R ) compared to those working in the public sector. Finally, as expected, a higher educational
attainment leads to a higher remuneration, and hence those with postgraduate qualifications earn the
highest (R per month on average).
TABLE . Mean and median monthly income of graduates in December prices,
Mean Median
Employee 15 810 12 807
Employer 22 284 16 186
Self-employed 13 564 6 474
Private sector 16 661 11 526
Public sector 15 553 13 874
Post-Matric certificate or diploma 12 945 9 072
Bachelor’s degree 18 296 16 009
Postgraduate degree or diploma 21 753 18 143
Total 16 298 12 807
Source: Author’s own calculations using the Quarterly Labour Force Survey quarter data.
. . Share of employee remuneration in national income

South Africa’s national income and production accounts measure the total share of employee
remuneration in the national product. These data indicate that, in – , remuneration of
employees comprised between and % of gross value added at factor cost, with a downward
trend in general over the whole period (albeit showing a tendency to increase since – see Figure
. ). This contrasts with the net operating surplus, which in effect is the profits of enterprises, which
generally show an increase as proportion of the gross value added.
FIGURE . Compensation of employees compared to net operating surplus of enterprises, as
proportion of gross value added at factor cost, –
Source: South African Reserve Bank Data series , and
However, these data do not necessarily relate directly to the average wages of individual workers. The
remuneration of employees measures total remuneration of all workers together. If there has been a
general decline in employment or even a relatively small increase (as was the case for many years), the
share of employee remuneration will tend to decline, even though wage levels may have increased
substantially.
SUMMARY
Real average monthly earnings of the non-agricultural formal sector employees from all sectors increased by . % per
annum between and . This annualised growth rate is the highest in two sectors (about . %), namely mining
and construction.
Graduates who are employers earn more than those who are employees or self-employed. Also, as expected,
graduates with postgraduate qualifications enjoy higher earnings.
Remuneration of employees as proportion of gross value added at factor cost shows an overall downward trend in
– , despite showing a tendency to increase since .
. WAGES AND INFLATION
As stated above (Table . ), the real wages of workers in South Africa increased over time, meaning
that money wages (or nominal wages) increased at a more rapid rate than the inflation rate. The
question that arises is whether the high increase in money wages (often caused by the outcome of
collective bargaining – see Chapter ) might not cause inflation to increase, or whether the wage
increases are high simply because inflation is high.
In order to analyse this question, it is important firstly to consider how inflation is measured (see
accompanying box).
Inflation means that your money won’t buy as much today as it did when you didn’t have any.
A
The measurement of inflation
Inflation stands for the continuous and considerable rise in prices in general (Mohr, Yu & Mollentze, : ), while the
inflation rate is the annual rate of increase in the general price level. Price levels are determined by using a price index,
namely the Consumer Price Index (CPI), which is the most common measure of inflation and shows how the average
price level of all those goods and services bought by a typical household (a “basket” of goods and services) changes
over time. It is generally used to determine the national inflation rate of the country.
The basket of goods and services and the weighting afforded to each item in the basket are based on the Income and
Expenditure Survey (IES), which is conducted every five years by Statistics South Africa. The weights of the items in the
basket are based on the typical spending patterns of representative household samples in different areas of the country.
These expenditure patterns are updated roughly every five years. The inflation rate obtained by including the complete
basket is called Headline CPI, or the official inflation rate.
Since , Statistics South Africa has applied the concept of owners’ equivalent rent (or imputed rent) to measure the
cost of housing. This concept reflects the cost of the accommodation services derived by owner occupiers from their own
homes. It excludes, as it should, the investment component of home ownership. Owners’ equivalent rent measures the
opportunity cost to the owners of forgoing a rental income by living in, rather than renting out, the house they own (see
Statistics South Africa, c). This means that there is no longer a need to distinguish between different measures of
inflation in order to exclude volatile items such as interest rates.
The Producer Price Index is another measure of inflation that might be used for specific purposes such as to determine
the inflation rate for producers (as opposed to consumers), but it falls beyond the scope of this book and will not be
discussed further.
Wage levels are usually adjusted upwards during inflationary times to keep up with increases in the
cost of living. Price increases therefore result in wage increases. At the same time, an increase in
wages might also cause prices to rise because wages are an important cost element in determining
prices. Even if unionised sectors do not comprise the majority of workers, wage and therefore cost
increases might be transmitted to other sectors (the spillover effect).
Wages and price levels are, however, influenced by a multitude of other factors. Wages, for instance,
are not only influenced by prices but also by educational levels, discrimination, productivity,
legislation and many other factors. Prices are influenced by, for instance, the exchange rate, monopoly
conditions, prices set by public enterprises, high government expenditure and high growth in money
supply. Exogenous factors, such as crop failures and increases in oil prices, can also increase price
levels. It is therefore not possible to fully isolate the influence of a single variable such as the money
wage level on price levels.
The Reserve Bank and managing second-round inflation
What then is the role of monetary policy in managing inflation? The South African Reserve Bank (SARB) is tasked with
protecting the value of the rand, which by extension is interpreted to mean to keeping inflation in check. The SARB
adopted a CPI annual inflation target range of % to % in and will measure its own actions against this range.
A very important distinction should be maintained between first-round and second-round inflation. First-round inflation
refers to external price shocks, such as a rapid currency exchange depreciation or a spike in the internationally-set price
for oil. There is very little the SARB or any central bank can do to stop that external shock. However, second-round
inflation refers to how the domestic economy responds to the external price shocks, and the SARB can influence this by
way of the interest rate.
Second-round effects, such as a reaction in inflation expectations or wages to higher headline inflation, would cause a
further increase in headline inflation after the initial effect of the shock had dissipated. In a peak in food-price
inflation averaged , %. In four of the five subsequent years, wage inflation exceeded both food and headline inflation.
If the SARB foresees inflation pressure in months’ time (this is how long it takes for an interest rate change to have a
full impact), it will increase the interest rate. By doing so, the cost of capital is increased, or to put it plainly, the domestic
economy has less money. Because of this, consumers need to rein in spending and this influences storekeepers and
sellers to moderate their price increases to keep on selling products. In the labour market, because there is less money,
employers need to moderate the wage increases passed on to workers.
The extent of the influence of money wage levels on price levels is determined, first, by the proportion
that the wage bill accounts for the total costs of an enterprise and, second, by productivity trends. If
wage increases exceed productivity increases, production costs will rise, and this might eventually lead
to increased prices. However, as already stated, there is also an interrelationship between wages and
productivity.
McConnell et al. ( : ) come to the following general conclusions regarding wages and inflation:
Over the long term, expansionary fiscal or monetary policies or supply shocks appear to be a
significant factor in determining the rate of inflation.
Unions do not appear to be the basic cause of inflation.
Unions do appear to perpetuate existing inflation and increase the difficulty of reducing inflation
once it has begun.
As stated earlier, nominal wages in South Africa are increasing more rapidly than inflation, which
means that workers are enjoying increases in real wages (refer to Figure . ).
The inflation phenomenon is a very complex one, and a simplistic approach, for example to blame
unions for inflation, is certainly not helpful. In a country such as South Africa the influence on inflation
of factors such as imported inflation, an earnings spiral caused by skilled labour shortages, the high
degree of economic concentration and restrictive business practices (i.e. reduced competitive forces),
in addition to union influences, cannot be ignored.
SUMMARY
High wage increases do not appear to be the initial cause of inflation, but they do seem to perpetuate existing inflation
and increase the difficulty of reducing inflation once it has begun.
. WAGE DIFFERENTIALS
A competitive labour market is characterised by wage or pay differentials, i.e. differences in wage rates
between industries, different classes of employees in the same industry or between geographic areas.
Wage differentials can also refer to differences in wages paid for similar work, in which case it is likely
to be discriminatory (Barker & Holtzhausen, ). Fallon and Lucas ( ) state that non-productive
factors such as discrimination, union membership and employment status (i.e. being employed or not)
have resulted in larger differentials occurring in South Africa than in other countries.
There are many reasons for such differentials, and a few are highlighted below.
Unfair discrimination
Differences in the average or mean wages of workers of different races or genders do not necessarily
indicate unfair discrimination by employers, because they may be due to differences in occupation,
education, experience and similar factors. However, such factors can point to before-the-market
discrimination, for example where African workers occupy less-skilled positions because of poorer
educational standards.
A better indicator of employer unfair discrimination would be differences in wage levels in the same
occupation, after controlling for differences in education or experience. This is discussed in more
detail in section . . , where it is shown that there is still a large measure of wage discrimination in
South Africa that cannot be ascribed to education, occupation, level of skill or some other non-
discriminatory factor.
The workers spend what they get and the capitalists get what they spend.
M. K
Educational differentials
The level of schooling is an important cause of large wage differentials. Bhorat ( a: ) found that
the most important factor determining higher earnings is education, especially Matric or tertiary
education. Bhorat ( a) shows that the median wage for those with a Matric is close to % of
those with a degree. However, the median wage of people whose highest education is Grade to is
only % of the wage earned by a person with a tertiary education. A more recent study by Statistics
South Africa ( ) found that the median wages of workers with incomplete and complete secondary
education are only % and %, respectively, of the median wage of workers with tertiary education.
Bhorat ( a) concludes that the most important element in an employment equity strategy would
be to equalise access to, and the quality of, educational provision.
However, educational level is not the only education-related factor of importance. Bhorat ( a) and
Pauw, Oosthuizen & Van der Westhuizen ( ) show that there are at least two other important
factors. The first is that there appears to be a disparity (either actual or perceived) in the quality of a
tertiary qualification from historically white and black universities, which is probably mainly as a result
of different resource allocations to the universities. The second important factor is that a tertiary
degree is a heterogeneous product. Some diplomas and degrees are in higher demand than others;
for instance, a person with an engineering degree will probably earn a higher salary than a person
with a humanities degree.
Intersectoral differentials
Table . shows the sharp differences in the earnings of different sectors. This is partly related to
differences in the skills structures of workers and partly to the rate of unionisation in the various
sectors. In addition, Fallon and Lucas ( ) have found that bargaining council activities have also led
to increased wage differentials, i.e. higher wages in sectors with bargaining councils compared to
those without (this issue is discussed in more detail in section . ). The authors of the ILO Review
( : ) state that although there is some similarity to the pattern found in other countries, the
extent of the differential between agriculture and other sectors in South Africa is very high.
The degree of unionisation

It is argued in section . . that unions have caused wages to peak at between and % above
those in non-unionised situations, and that they have caused a higher differential in South Africa than
in some other countries.
Size of the enterprise

According to a survey undertaken in the Greater Johannesburg Metropolitan Area in , on average,
large firms pay more than small firms across all occupational categories. The premium for being in a
large firm seems to be in the order of to % (Bhorat & Lundall, : – ). Similar findings are
observed when examining the earnings of formal-sector employees by firm size using the QLFS
data (the median wage of those working in enterprises with at least employees is about %
higher than that of workers from smaller firms with one to employees).
Smaller enterprises are usually more labour-intensive and tend to compete more on the basis of
labour costs. Therefore wages tend to be lower in smaller enterprises. This would normally not be so if
a bargaining council were involved. This means that small enterprises might lose their competitive
advantage vis-à-vis larger enterprises.
An important reason for the higher pay in larger firms is that these firms are probably unionised,
whereas it is often difficult for unions to organise workers in smaller enterprises.
Another reason for this differential is the efficiency wage hypothesis (see section . . ). Since higher
wages may result in increased effort, reduced shirking and lower monitoring costs, larger firms are
willing to pay higher wages in line with their greater ability to pay such wages. Smaller firms may have
the same priorities, but their opportunities to gain the benefits of efficiency wages are restrained by
their lower ability to pay higher wages. Furthermore, workers in larger firms may indirectly share in
the higher profits of larger companies by being paid higher wages.
Age or experience
Employees with more experience are usually paid higher wages. In some cases this may amount to
older workers being paid higher wages. In some cases, employees receive a salary increment based on
years of service. For example, Statistics South Africa ( b) found that in , the median monthly
earnings of employees aged – years (R ) was much higher than those aged – years (R
) and – years (R ).
Working environment
Circumstances such as night work, unpleasant physical working conditions and hazardous work can
justify higher wage rates.
Occupational differentials
There is a large difference between the wage level of the average production worker and that of the
chief executive officer (CEO) of the enterprise. This is sometimes referred to as the wage gap. The
Employment Equity Act refers to it as income differentials. The Labour Market Commission believes
that income differentials between the highest and the lowest paid workers in South Africa are high,
both by developed and developing country standards (Labour Market Commission, : ).
In a survey of very large companies, the Labour Research Service has estimated that the average
CEO remuneration (including salary, benefits and bonus) is times that of the average low-wage
worker. The average executive director earns times what the average low-wage worker earns.
However, PE Corporate Services have determined that the average CEO earns times more than
junior workers in his or her company, and that there has been a decline in the wage gap in recent
years (http://www.pecs.co.za). Using the QLFS data, Table . indicates that those working as
highly skilled managers and professionals earn much higher compared to others.
TABLE . Average monthly earnings of non-agricultural formal employees by broad occupation

category in December prices,
Mean Ratio to total

Managers 22 782 265.7
Professionals 22 408 261.3
Technicians 10 931 127.5
Clerks 7 882 91.9
Service workers 5 942 69.3
Craft and related trade workers 6 923 80.7
Plant and machine operators 6 320 73.7
Elementary occupations 3 863 45.1
Domestic workers 1 375 16.0
Total 8 575 100.0
Source: Author’s own calculations using the Quarterly Labour Force Survey quarter data.
Whatever the true state of affairs, the fact of the matter is that the wage gap in South Africa is large,
but this is by no means a South African phenomenon – the general secretary of UNI Global Union,
which claims to represent million workers worldwide, has indicated that chief executives in the US
earn times more than the average American, up from or times more in the s (Labour
Research Service, : ). In the UK, the New Economics Foundation has stated that the average CEO
earns times more than a teacher (Labour Research Service, : ).
There are a number of factors that should be taken into account when interpreting this information.
The first is that until recently the typical executive’s package was made up almost entirely of cash and
a few benefits. Today it is likely to be split among basic salary, annual incentive (e.g. performance
bonuses) and long-term incentives (e.g. share options). A substantial portion of a senior executive’s
pay packet is made up of variable components, i.e. components that are, theoretically at least, at risk
and dependent on the performances of the employee and the enterprise. This is very seldom the case
with production workers, as most of their remuneration is guaranteed pay.
Another factor that should be taken into account is the skills profile of workers in different countries.
Production plants in European countries and Japan are usually highly mechanised and very capital
intensive. These plants have a very small proportion of production workers and such workers are
usually highly skilled operatives using very sophisticated machinery. This means that their productivity
is also very high, with commensurate pay packages. South Africa is at the opposite end of the
spectrum, with many workplaces still being very labour intensive, in spite of increases in capital
intensity in recent years. This is particularly the case in the resource-based industries.
Demand and supply factors can also not be disregarded. In South Africa there is an oversupply of
workers at the lower skills levels, whereas people with the necessary skills and abilities to occupy
executive, managerial and other senior positions are in very short supply. Many of these skills are
tradable internationally and therefore very mobile. Leading enterprises can retain their competitive
edge only by attracting and retaining the best leadership and managerial skills, which means high
remuneration packages.
The Employment Equity Act compels an employer to reduce so-called “disproportionate income
differentials”. The Employment Conditions Commission must investigate proportionate norms and
benchmarks, and may make recommendations to the minister in this regard. This provision, if applied
rigorously, could have negative consequences for the economy – the wage gap between the highest
and lowest categories of workers has arisen, among other reasons, because of the double impact of
skills shortages and an oversupply of unskilled labour. To address only the symptoms rather than the
causes of these imbalances in the labour market will have various unintended consequences.
First, if an artificial ceiling is placed on the earnings of highly skilled workers, there is likely to be a
skills flight from the country, especially of those skills that are internationally very mobile. This brain
drain will be aggravated by the fact that share-option schemes and profit sharing for senior
executives might become less popular, because they are likely to increase rather than reduce the
income differential (especially if the enterprise is financially successful). As indicated in section . . ,
such a skills flight is certainly not in South Africa’s best interests.
The wages of the lowest categories of employees cannot simply be increased without taking into
account the wage curve (i.e. the wage levels of the next level of employees, and so forth), as such a
step might not only lead to inconsistencies, but might also be unfair if unskilled workers receive the
same salary as workers with more skills or responsibilities. Furthermore, if the wages of unskilled
workers are increased without any regard to their productivity, it will lead to upward pressure on the
total wage bill, and this will damage competitiveness.
Companies might also reduce the wage gap by getting rid of the lowest-paid employees, for example
by mechanisation or outsourcing. This is likely to increase unemployment and lead to more insecure
forms of employment.
Bhorat ( a: ) emphasises that a simplistic legislative approach to reducing the wage gap would
be “foolhardy”. The wage gap has arisen due to a fairly intricate interplay of factors, and a blunt
instrument to reduce the gap could prove to be “both extremely disruptive to incentives in the labour
market, as well as resulting in highly inefficient outcomes” (e.g. an exodus of skills from the domestic
market, thereby aggravating the skills shortage). Wage differentials can only be reduced by
addressing the root causes (e.g. by improving education and training, as well as productivity) and not
by simply addressing the symptoms.
SUMMARY
Apart from discrimination, there are many reasons for wage differentials between workers, among others, differences
in skills levels, occupations, sectors, experience and level of unionisation.
The wage gap between the highest- and lowest-paid employees in an enterprise has arisen due to an intricate
interplay of factors, and to address only the symptoms of this phenomenon rather than the causes would have
extremely disruptive consequences for the economy.
. THE TRADE-OFF BETWEEN WAGES AND EMPLOYMENT
. . The impact of wage increases on employment

Hypothesis: an increase in wages leads to a decrease in employment
It is generally accepted among mainstream economists that union policies of aggressively pushing up
wage rates will cause unemployment. An artificial increase in wages will reduce employment. Marshall
et al. ( : ) summarise the impact of wage increases as follows:
When the demand for labour remains constant, any wage increase will tend to reduce employment.
When the demand for labour rises, a wage increase will reduce employment below what it would
have been in the absence of a wage change, and the rate of growth of employment will therefore
be reduced.
When the demand for labour falls, any wage increase will magnify the reduction in employment.
Critique: other factors do play a role

What are ignored in this analysis, however, are developments with regard to productivity, which,
through its impact on unit labour cost, will obviously also influence the net impact of a wage increase
on employment.
The outcome of a wage reduction is not necessarily the mirror image of a wage increase, as wage
changes can have an asymmetrical effect. This simply means that a wage increase might adversely
affect job prospects by reducing the number of jobs available, whereas a wage reduction might not
necessarily improve job prospects by increasing the number of jobs (ILO Review, : ). A
reduction in wage rates can lead to smaller declines or larger increases in employment, depending on
the circumstances.
The South African experience: evidence of a link between wages and unemployment
In the case of South Africa, studies by the World Bank and one prepared for the Labour Market
Commission by Bowles and Heintz both found that a % increase in real wages correlates with a %
decline in employment levels. The authors of the ILO Review ( : ) remain sceptical, but they
choose their words very carefully: “the available studies have not demonstrated either that real wages
have been rigid or that they have had a strong [sic] negative effect on employment.”
An econometric analysis of South African data for – by Lawrence and Van der Westhuizen
( : ) shows that “an increase of nominal wages of, say % would lead to a fall in employment
levels of around %”. In addition, the wage increase would also lead to a loss of foreign exchange as it
would result in a % drop in exports.
Two researchers from the World Bank, Fallon and De Silva ( : ), assessed the implications of a
% real-wage increase in owing to union pressure, with no real-wage declines after that. They
found that such a scenario would increase the number of persons without jobs by as much as % in
most years from – . Inflation would be higher by about percentage points, and exports and
GDP real growth lower.
In a World Bank study done for the Labour Market Commission, Fallon and Lucas ( ) found that
“rising real product wages have had a substantial dampening effect on the demand for black
workers”. Econometric analyses have indicated that a % increase in the real product wage would
eventually lead to a . % decrease in African employment. However, employment changes take a
significant time to adjust to the effects of wage changes, and this period can, in fact, be several years.
More specifically, after nearly three years only slightly more than half of the adjustment to a wage
change will have taken place. This is an important fact, and makes research work in this regard more
complicated.
Mazumdar and Van Seventer ( : ) found that during the s the gains in manufacturing
output could be attributed almost equally to an employment increase and a wage increase. This
changed quite substantially in the second half of the s when, in spite of a relatively small increase
in output, the real-wage increase “bumped up to a whopping . %” (per annum) and had to be
balanced by a fall in employment (quite sharply so, at a decline of % per annum).
Fedderke and Mariotti ( : ) have similarly found that real wages are likely to be important in
determining employment trends in South Africa’s labour markets. More specifically, where the growth
rate in labour remuneration has outstripped growth rates in labour productivity, there is a tendency
for labour inputs to decline.
Furthermore, the large employment sectors of the economy that have experienced strong declines in
employment also show strong negative correlations between real labour remuneration and
employment. These sectors are agriculture, household domestic services, gold mining, transport, and
building construction. These were also the very sectors that showed negative growth rates in
employment.
South Africa: a weakening of the link

However, Banerjee et al. ( : ) come to a different conclusion when they say that there is no
evidence that the increase in unemployment in South Africa was driven by wage growth. In fact, the
collapse of employment in the mining and agriculture sectors during the s and the high turnover
rates we observe in the microdata would mute the power of wage push by dampening union
demands.
A more recent descriptive study by Yu ( ) examined the relationship between average real monthly
earnings and formal employment by sector in – , and found that an increase in real earnings is
associated with a downward formal employment trend in the agricultural and mining sectors (these
two sectors also showed the lowest real gross value-added growth rate during the period), but both
variables show an upward trend in the construction sector and the community, social and personal
services sector.
Although most research points to a wage– employment trade-off, it does not follow that lower wages
will necessarily “solve” the unemployment problem. As discussed in Chapter , what happens to
employment depends very much on the elasticity of the demand for labour. If the demand for labour
is very inelastic, that demand is not very “sensitive” to wages, and wages might have to drop
substantially before increasing the demand for labour significantly.
It is also important to note that unemployment is not necessarily “caused” by high wage levels. As
indicated in Chapter , unemployment is the result of a complex interaction of various factors, and it
cannot be addressed by a simplistic approach. In any event, reducing wages is politically not feasible.
Nevertheless, slower increases in wages would have fewer negative consequences for employment
than wage increases significantly above the rate of inflation, which seems to have become the norm in
several sectors.
SUMMARY
Labour cost increases that are not associated with equivalent productivity increases will reduce the demand for labour
to a level that is lower than would otherwise have been the case.
In the case of South Africa, it seems as if a % increase in real wages may lead to a % decline in employment, but
will take some years to take full effect.
. . The impact of high unemployment on wages

As indicated by the theoretical discussions, unemployment will lead to competition for jobs among
unemployed people. This will result in wages declining, which in turn will both increase the quantity of
labour demanded and reduce the quantity of labour supplied. In other words, if high unemployment
levels cause wages to fall, unemployment will reduce and eventually be eliminated. This will happen in
perfectly competitive labour markets, but the reality is that markets are imperfect as a result of a
number of factors, in particular artificial wage determination as a result of the impact of the
government, unions, monopoly employers and bargaining councils.
Fallon and Lucas ( ) found that the labour markets for skilled workers in South Africa clear well, but
the same cannot be said of the markets for unskilled workers. Using the wages of Africans and whites
as representative of unskilled and skilled wages respectively, they found that “unskilled wages have
been much less responsive to unemployment levels than those of the skilled”. The positions of skilled
Africans and unskilled Africans are different in that there is clear evidence that the markets for skilled
Africans do in fact clear. Another finding was that African wages do not adjust to changes in the
inflation rate as quickly as those of whites, suggesting greater rigidities in wage setting.
SUMMARY
Unskilled wages have been much less responsive to unemployment levels and inflation than skilled wages, which
suggests greater rigidities in wage setting among unskilled workers.
. . Will high wages increase economic growth?

There is an argument that if wages are increased on a sufficiently wide scale, consumption and
eventually economic growth will increase because workers will have more spending power. The higher
economic growth will, so the argument goes, lead to an increase in employment.
However, if it were this easy to increase employment, no country in the world would experience
unemployment. For a thorough critique of these populist theories, consult Moll ( ). The following
general remarks, however, suffice to illustrate the fallacy of this argument.
If wages are increased, the higher wage bill has to be financed in some way. If it is financed by passing
the costs on to consumers in the form of higher prices, the consequent higher inflation rate will
reduce real wages and workers will be no better off; in fact, they may be worse off because inflation
has negative redistributive effects and feeds on itself. If employers dip into their profits (if any) to
finance the wage bill, surpluses available for reinvestment will be reduced, and this will affect
economic growth negatively. If the employer attempts to avoid a higher wage bill by reducing
employment or by replacing workers with machinery, employment will be reduced. Total spending
power is therefore not necessarily increased.
Nobody kin talk as interestin’ as th’ feller that’s not hampered by facts or information.
K H
. THE IMPACT OF LABOUR LAWS ON THE COST OF LABOUR
Various laws, and labour laws in particular, can increase the cost of labour and thus have a negative
impact on employment. In many cases, this may be entirely justified on the basis of the improved
protection of those in employment. However, the broad impact of the legislation should be
considered more thoroughly than has been the case up to now. What is important is for the legislator
to compare the benefits (in terms of the greater protection) and the cost (in terms of increased labour
cost and the impact thereof on employment and competitiveness). Such a comparison can help
prevent the unintended consequences of laws.
By way of example, the impact of some provisions of labour laws on the cost of employment is
discussed below. Chapter will consider the impact of labour laws on productivity and flexibility.
Private enterprise is ceasing to be free enterprise.

F D. R
. . Basic Conditions of Employment Act

Hours of work
Some years ago, there was an intensive debate in South Africa about the impact of a reduction in the
hours of work on the cost of labour and on employment (for a full discussion, see Barker, a). This
debate was fuelled by the reduction in hours of work in terms of the Basic Conditions of Employment
Act (Act No. of ) (BCEA), with Chapter Two of this act stipulating that an employer may not
require an employee to work more than hours per week and hours per day (if the employee
works for days or fewer in a week); the act also establishes procedures for the progressive reduction
of the maximum ordinary weekly work hours to per week and per day.
When considering the impact of reduced ordinary hours through legislative decree, the argument is
usually that reduced hours have social benefits, for example, increased time for education and
training, family and social responsibilities, and leisure time (Department of Labour, : ). This, of
course, considers only the social benefits of those in employment, while ignoring both the costs to
the economy and the social costs to those without employment.
A legislated reduction in work hours will, in most instances where hours are reduced, increase the unit
cost of labour and production costs in general, because of the impact on fixed labour costs, variable
labour costs and capital costs. The fixed cost of labour (e.g. leave pay, office space, protective clothing,
recruitment costs, costs associated with education and training, and severance pay), as well as capital
costs, do not change in relation to hours of work. Therefore, if employees work fewer hours, the fixed
costs associated with their employment are spread over fewer hours, i.e. the hourly fixed costs
increase.
With regard to the variable cost of labour (i.e. labour costs associated with the number of hours
worked), the impact of the reduced hours will depend on what happens to the hourly cost of labour. If
workers are paid by the hour, the hourly cost of labour remains constant, regardless of hours worked.
The hourly variable cost of labour for the employer will therefore remain constant if hours are
reduced, and the employees’ total take-home pay will be reduced because of the reduced hours.
If workers are paid on a weekly or monthly basis, and the weekly or monthly rates remain unchanged,
reduced hours will increase the hourly cost of labour. The take-home pay of employees will remain
constant, but because reduced hours increase production costs and therefore inflation, and reduce
employment, the real disposable income of households will generally decline (see the survey of
studies by the Commission of the European Communities, quoted by Barker, : – ).
Besides the significant impact of reduced hours on the cost of production, the BCEA, introduced in
, increased the cost of labour in several ways, such as increases in paid annual leave, paid family
responsibility leave, longer notice periods, several obligations relating to night work, and the broad
definition of remuneration when calculating leave pay, notice pay and severance pay.
The act might also have an indirect impact via the determination of minimum wages by the
Employment Conditions Commission. This matter, however, pertains more to the level of the
minimum wages than to the impact of the BCEA itself. It is discussed in section . .
With regard to the impact of reduced work hours on employment, there are those who believe that
reduced work hours will increase labour absorption, among them COSATU ( : ). It argues that
existing work will be shared among more employees. This is the classic “lump of labour” argument. Its
fundamental flaw is, however, that the number of jobs available is regarded as fixed, and can simply
be divided up in different ways. In particular, it ignores the impact of reduced work hours on
production costs. Reduced hours of work will, in most instances, increase production costs and this is
likely to have a negative effect on the demand for labour – there is thus no fixed number (or “lump”)
of jobs available.
In contrast, there is good evidence that employment will either reduce or at most remain constant if
work hours are reduced (see the survey of studies on this topic by Barker, : – ). Roche,
Fynes and Morrissey ( : ) state: “Although calculating the employment effects of a reduction in
standard working time is a problematic exercise, the international experience appears to have been
disappointing, with some commentators suggesting the possibility of a negative effect.”
More recent studies by Börsch-Supan ( ) as well as Cahuc and Zylberberg ( ) conclude there is
no strong indication that a reduction in work hours would improve employment, despite the fact that
the worker’s utility may increase.
With regard to the effect of a reduction in work hours on other macroeconomic variables, a number
of studies have found that it generally leads to higher inflation (see sources quoted by Barker, :
– ). Because a reduction in work hours reduces competitiveness, it also has a negative effect on
the balance of payments.
SUMMARY
A reduction in work hours will increase the hourly cost of production and unit production costs unless there is a
commensurate increase in productivity. The net effect of a reduction in work hours is likely to be a lower demand for
labour than would otherwise have been the case. Nonetheless, the results of recent empirical studies do not strongly
indicate that a reduction in work hours would have a definite negative impact on employment.
Minimum wages
The BCEA provides for the establishment of the Employment Conditions Commission. This body
makes recommendations to the Minister of Labour in respect of minimum wages and other
conditions of employment.
When making wage recommendations the Employment Conditions Commission has to take into
account the ability of employers to carry on their undertakings successfully, the cost of living, and the
operation of small and new enterprises. The Employment Conditions Commission should also
consider the likely impact of any proposed condition of employment on current employment or the
creation of employment. Having considered a recommendation, the minister may make a sectoral
determination.
Although sectoral determinations have had a significant impact on some sectors, for instance the
agricultural sector and domestic work, the minimum wages set by such determinations are generally
relatively low compared to actual wages in the particular sector. In this sense, they do not appear to
have had a significant impact on the cost of labour. The impact of minimum wages on, among other
things, the employment of farm workers, is dealt with in section . .
. . Labour Relations Act

The Labour Relations Act (LRA) firstly provides for bargaining councils and the extension of bargaining
council agreements to all parties in a particular sector. Bargaining councils have been much criticised,
not only in respect of their impact on the cost of labour, but also for introducing rigidities in the
labour market. These criticisms as well as some counter-arguments are discussed in section . .
As discussed in section . , the LRA also actively promotes unionisation and, according to some
authors, does not properly apply the principles of positive labour relations, i.e. voluntarism for all
parties, equality before the law for all parties, and non-competitive universal human rights. Although
unions are essential for various reasons, they do cause wages to be higher than in non-unionised
situations – international researchers have put this differential at between and %. Union action in
South Africa has resulted in larger wage differentials in this country than in other countries (see
section . . ). Fallon and Lucas ( ) found that employment was reduced by about . % by the
effects of unions raising wages. As indicated in section . . , the level of unionisation in South Africa
is high by international standards and very high by the standards of developing countries (ILO Review,
: ). This is one reason for the significant impact that unions have on the labour market in this
country.
The LRA also increases the cost of labour in various other indirect ways, for instance by making
dismissals and retrenchments more difficult and expensive. This is discussed in more detail in section
. . .
. . Other laws
Another potentially negative effect of labour policies on wage costs is the provisions relating to
income differentials in the Employment Equity Act (EEA). This issue is discussed as part of occupational
differentials in section . above, where some of the unintended consequences are pointed out.
SUMMARY
Provisions relating to bargaining councils in the LRA are likely to increase the cost of labour, as will the provisions that
have encouraged and protected the growth towards strong unionism in the country. However, the potential positive effect
of unions should not be ignored.
. DEEPER INSIGHT I: DETERMINATION OF WAGES IN A PERFECTLY COMPETITIVE

LABOUR MARKET
Wages in a market economy are usually determined in different ways, and the strength of the various
factors that influence wage levels will be different in each case. This section deals with the perfectly
competitive market, and certain imperfect conditions are introduced in the next section.
Wages are determined by the bitter struggle between capitalist and worker.
K M
The functioning of the perfectly competitive market can be studied by considering the interaction of
the supply-and-demand curves that were developed in the previous chapters. In this market, all
workers and all jobs are homogeneous. Furthermore, information on the labour market and the
availability of skills and of vacancies is immediately available without cost, and workers are perfectly
mobile. In such a market, there will be no wage differentials. Clearly, these assumptions are unrealistic,
but the assumption of a perfectly competitive labour market helps to understand the operation of the
labour market.
In this market, and as shown by Figure . , the wage level will be set by the market. If the supply
curve is assumed to be S and demand D, then the market will stabilise where supply and demand are
equal, i.e. the market will stabilise where the two curves intersect at point e. The wage will be W* and
employment E*. This is called equilibrium, or the market-clearing level.
FIGURE . Equilibrium between supply and
demand
A higher wage than W* means an oversupply, excess supply or surplus of labour. Consider a wage
level of W . Reading down from points A and B, quantity supplied has increased to E and quantity
demanded has reduced to E . The oversupply of labour is the distance between points B and A, i.e. E –
E . Some of the people looking for work will be prepared to accept a job at a lower wage, and at a
lower wage employers will also be willing to take on more workers. This reduces both wages and the
oversupply of labour, until equilibrium is eventually reached at point e.
Similarly, a lower wage rate than W*, such as W in the figure, will mean an undersupply, excess
demand or shortage of labour (which is equal to E –E at W , by examining points G and H), and
employers will bid wages up to the equilibrium or market-clearing level.
Shifts in supply-and-demand curves

Sections . . and . . discussed the factors causing the shift of market labour supply and labour
respectively. Let us assume that an exogenous (external) factor, for example higher immigration,
moves the labour supply curve from S to S in Figure . . There will thus be an oversupply of labour
to the extent of E –E at wage level W . Workers will bid down wages and the quantity of labour
supplied will decrease. Because of lower wages, employers will employ more workers, i.e. the quantity
demanded for labour will increase. Eventually a new equilibrium level will be reached where quantity
supplied and quantity demanded are equal, i.e. at point e . Wages will be at level W and
employment at level E , that is, wage decreases and employment increases as a result of the increase
of market labour supply.
FIGURE . Impact of increase of market labour supply on
wage and employment
Similarly, the demand curve can also move to the right or left. For example, the increase in labour
productivity will increase the demand for labour, causing the rightward shift of the labour demand
curve. This will cause an excess demand for labour, which will increase the wage rate (from W to W )
and employment (from E to E ), as shown in Figure . .
FIGURE . Impact of increase of market labour demand on
wages and employment
It is also possible for both market labour supply and labour demand to change simultaneously. For
example, if both labour supply and labour demand increase, both forces have the same positive
impact on employment but their combined impact on the equilibrium wage is uncertain, as an
increase in labour supply reduces wages, ceteris paribus, while an increase in labour demand increases
wages, ceteris paribus. Whether the equilibrium wage would rise, fall or remain unchanged depends
on the relative magnitudes of the changes in labour supply and demand.
Table . summarises the results of the impact of the changes in labour supply and labour demand
on wages and employment.
TABLE . Changes in labour supply and labour demand
Change in labour supply Change in labour demand Change in wages Change in employment
Increase No change Decrease Increase
Decrease No change Increase Decrease
No change Increase Increase Increase
No change Decrease Decrease Decrease
Increase Increase Uncertain Increase
Increase Decrease Decrease Uncertain
Decrease Increase Increase Uncertain
Decrease Decrease Uncertain Decrease
SUMMARY
In a perfectly competitive labour market, markets will move towards equilibrium through the wage mechanism, and
equilibrium is defined as the point at which quantity supplied equals quantity demanded.
. DEEPER INSIGHT II: DETERMINATION OF WAGES IN A MONOPSONY MARKET
A monopsony market occurs where there is only one employer (in certain occupations) or where there
is collusion or cooperation between all or most employers in a certain market to fix wages at a certain
level and not to hire away each other’s employees (Filer et al., : ; McConnell et al., : ).
The monopsonist is therefore a wage setter and can influence wages by adjusting the number of
workers he or she hires (Ehrenberg & Smith, : ). The customary example is that of a huge steel
mill in a very small town, where the employer would have to increase wages every time he wants to
employ another worker in order to attract, for instance, housewives or students to come to work for
him. This situation is not entirely theoretical and may arguably be present in South Africa, for example
with large companies employing the majority of workers in a specific occupation. A bargaining council
is also a form of monopsony because all the employers in bargaining councils form a united front in
respect of wages and can thus be regarded as a single employer.
For the sake of simplicity, only the single-employer situation is considered. Since it is the only
employer, a monopsonist enterprise will face an upward-sloping labour supply curve, similar to the
cumulative supply curve in a perfectly competitive labour market. Whereas an employer in the perfect
labour market can employ as many workers as he or she wants at a constant wage rate (because of
the assumption that no single employer is large enough to influence the market wage), the
monopsonist enterprise drives wages up if it expands employment. The monopsonist thus faces the
market supply of labour and will have to entice every next worker into the labour market by paying a
higher wage than the previous wage, because the reservation wage of that worker is higher than that
of the previous worker (Hamermesh & Rees, : ). This is reflected by the same market curve that
was discussed in Chapter (see S curves in Figure . ).
For the sake of simplicity, we continue to assume that all workers have equal skills. As far as the wage
levels are concerned, the monopsonist can do one of two things. Firstly, the monopsonist can decide
to pay all workers the same wage, which means he would have to increase the wages of all current
workers to the level at which he employed the last worker (in this case the monopsonist is non-
discriminating). Alternatively, the monopsonist can discriminate between employees, and simply pay
them their reservation wages, which means each next worker will get more than the previous worker
(the monopsonist is “perfectly” discriminating). The latter case is the simplest and will be dealt with
first.
The discriminating monopsonist

The marginal cost (MC) of the discriminating monopsonist will be equal to the labour supply curve,
since the employer simply pays the worker his or her reservation wage and can hire different workers
at different wage rates, as shown in Figure . . This can be explained with the aid of the information
in Table . . At a daily wage of R , no one is willing to work and hence the quantity of labour
supplied is zero. At a wage of R , the firm attracts one worker (whose reservation wage is R ), and
hence both the total labour cost and marginal labour cost equal R . As the wage increases to R ,
this attracts the second worker. The total cost of labour is equal to R : the employer pays R to the
first worker (whose reservation wage equals R ) and R to the second worker (whose reservation
wage equals R ). Also, the marginal cost of hiring the second worker is equal to R . Using similar
reasoning, at a wage of R , the third worker is hired, resulting in a total cost and marginal cost of
R and R respectively; when the wage increases to R , the fourth worker is hired, and the total
cost and marginal cost of labour are R and R respectively. Finally, the second-last column of
Table . derives the average cost of labour, which simply equals total cost divided by employment.
TABLE . Labour supply of a hypothetical firm operating as a discriminating monopsonist
Wage Employment Total cost of Marginal cost of Average cost of Value of marginal product of
(W) (E) labour (TCE) labour (MCE) labour (ACE) labour (VMPE)
50 0 0 – – –
60 1 60 60 60 120
70 2 130 70 65 100
80 3 210 80 70 80
90 4 300 90 75 60
FIGURE . The perfectly discriminating

monopsonist
The value of the marginal product of labour (VMPE) is also indicated in the figure and table, and as
outlined in the previous chapter, equilibrium will be where the value of the marginal product obtained
by employing another worker is equal to the marginal cost of employing that worker (the wage of the
worker) – therefore where VMPE = MCE. This is indicated by point e. Only the last worker will earn this
wage, and all other workers will earn a lower wage. In other words, referring to the information in
Table . and Figure . , at equilibrium, E* (three) workers are hired, but only the last (third) worker
would be paid the equilibrium wage of W* (R ), which is also equal to this worker’s reservation wage;
the first two workers would be paid their respective reservation wages of R and R .
The non-discriminating monopsonist

As discussed above, non-discriminating monopsonists are employers who pay all their workers the
same for the same job, even if they were able to recruit some at a lower wage rate. The cost of
employing an additional worker becomes quite expensive for non-discriminating monopsonists. First,
monopsonists have to pay the additional worker a higher wage to attract him or her. Second, as
monopsonist employers are also non-discriminatory employers, they will, in addition, have to increase
the wages of all their other workers to that same level. The enterprise’s total cost of employing one
additional worker (the marginal cost) will therefore be the wage of the additional worker plus the cost
of increasing the wages of existing workers. The marginal cost of employing that additional worker
will therefore be higher than the wage of the particular worker. The wage of this worker is also the
wage of all the workers, which is the average wage. If the marginal cost of the last worker is higher
than his wage, then the marginal cost will also be higher than the average cost of all the workers.
This is reflected in Figure . , where MCE indicates the marginal cost curve and ACE the average cost
curve. The MCE curve lies above the ACE curve to indicate the higher marginal cost involved in
employing additional workers. Also, whilst the MCE of the discriminating monopsonist is equal to the
labour supply curve (see Figure . ), it is rather the ACE that is equal to the labour supply curve in the
case of non-discriminating monopsonists (see Figure . ).
The information in Table . could aid in explaining Figure . . At a wage of R , the firm attracts
one worker (whose reservation wage is R ); as the wage increases to R , this attracts the second
worker whose reservation wage is R . The total cost of labour is equal to R : although the
monopsonist could hire the first worker at a lower wage of R , he pays this worker R because he
pays the same wage to the second worker. The marginal cost of labour equals R (R – R ). As
the wage increases to R , the third worker, whose reservation wage is also R , is hired. The total
labour cost equals R , as the same wage is paid to all three workers (R × ), despite the fact that
the monopsonist was able to hire the first two workers at a lower wage. The marginal cost of labour
equals R (R – R ) when hiring the third worker.
TABLE . Labour supply of a hypothetical firm operating as a non-discriminating monopsonist
50 0 0 – – –
60 1 60 60 60 120
70 2 140 80 70 110
80 3 240 100 80 100
90 4 360 120 90 90
FIGURE . The non-discriminating monopsonist
To maximise profits, the monopsonist enterprise must ensure that its marginal cost is equal to the
value of marginal product (which is equal to R from the information in Table . ). In Figure .
this means that it will employ E* (three) workers. Profits will be maximised because if the monopsonist
employs fewer than E* workers, the value of marginal product produced by the employment of an
additional worker will be greater than the marginal cost and the employer can make more profit by
employing more labour. However, if he employs more than E* workers, the marginal cost of the last
worker will be higher than the value of marginal product. The employer will then reduce the number
of workers to reduce costs until the marginal cost of a worker is equal to the value of marginal
product produced by that worker.
If an employer employs E* (three) workers, he will pay the same wage of W* (R ) to all his workers.
However, if it were a perfectly competitive market, equilibrium would have been at point q, i.e. where
supply and demand intersect, in which case wages would be higher at W** (R ) and not W*, and
employment higher at E** (four workers) and not E* (three workers). It is clear that in a monopsony
market both wages and employment could be lower than they would be in a perfectly competitive
market (Hamermesh & Rees, : ).
SUMMARY
In a monopsony market, i.e. where there is only one employer or where there is collusion between employers to fix wages
at a certain level, equilibrium might be reached at a point where both wages and employment are lower than they would
have been in a perfectly competitive market.
KEY CONCEPTS
consumer price index

discriminating monopsonist
dual labour market
equilibrium or market-clearing level
inflation
internal labour markets
labour market segmentation
monopsony
nominal (money) wage/earnings
non-discriminating monopsonist
oversupply/excess supply/surplus of labour
primary segment
real wage/earnings
secondary segment
undersupply/excess demand/shortage of labour
wage differentials
FOR STUDENTS
. Discuss the theory of labour market segmentation (also known as the dual or internal labour market theory),
and then discuss whether you would describe parts of the South African labour market as a dual labour
market.
. Discuss the reasons for wage differentials in the South African labour market, and possible measures to reduce
these differentials.
. Discuss the three main ways that the South African government gets involves in and influences wage
determination.
. Explain, with the aid of a graph in each case, the Marshall et al. argument on the relationship between
wage increase and employment:
. Wage increases tend to reduce employment when labour demand remains constant.
. Wage increases magnify the reduction in employment when labour demand falls.
. When labour demand rises, a wage increase would reduce employment below what it would have been in
the absence of a wage change, and the rate of growth of employment will therefore be reduced.
. Explain, with the aid of a figure, the impact on wages and employment in a perfectly competitive labour
market in the construction sector in a hypothetical country, assuming the danger involved in the work
increases and labour productivity in the sector declines.
. In the finance sector of a hypothetical country, there is an increase in the number of immigrants with
qualifications in finance seeking work, while this sector experiences a rapid decline in strike incidence. Explain,
with the aid of a figure, their combined impact on wages, and employment in this sector, assuming there is a
perfectly competitive labour market in the sector.
. It is found that the majority of jobs in the “legislators, senior officials and managers” occupation are located in
province X of a hypothetical country. It is also found that the cost of relocating from other provinces to
province X has become more expensive, while the workers in this occupation are well known to be associated
with low absenteeism due to a low HIV and AIDS infection rate. Explain, with the aid of figure, their combined
effect on wages and employment in this occupation category in province X, assuming there is a perfectly
competitive labour market.
. It is found that a lot of medical graduates leave South Africa to work in Europe (i.e. brain drain from South
Africa), while the cost of the capital equipment in the medical sector has become cheaper. Explain, with the
aid of a figure, their combined impact on employment and wage in the medical sector in South Africa,
assuming that there is a perfectly competitive labour market.
. Assume that Magic Mining Corporation is the only mining company in its region in a hypothetical country. In
addition, the firm pays employees their reservation wages, which implies that each next worker will get more
than the previous worker. With the aid of a well-labelled diagram, explain how the number of workers
employed to maximise profits at equilibrium is determined in this firm. Also explain the wage of each worker
employed.
. Use the following graph to answer .– . :
. Explain what a discriminating monopsonist is and, using the graph above, state how many workers it
would employ.
. What wage would the discriminating monopsonist pay its workers? Explain.
. For a non-discriminating monopsonist, the MC curve lies above the AC curve. Explain why for a non-
discriminating monopsonist MC is greater than AC.
. How many workers would the non-discriminating monopsonist employ and at what wage rate?
. How many workers are employed and at what wage rate in the case of a perfectly competitive market?
SUGGESTED READING*
Bhorat, H. a. Wage premia and wage differentials in the South African labour market. Submission to the
Employment Equity Commission. Pretoria: Department of Labour.
Burger, R. & Yu, D. . Wage trends in post-apartheid South Africa: constructing an earnings series from
household survey data. Development Policy Research Unit Working Paper / , University of Cape Town.
Filer, R.K., Hamermesh, D.S. & Rees, A.E. . The economics of work and pay. New York: HarperCollins Publishers.
Presidential Commission to Investigate Labour Market Policy. . Restructuring the South African labour market.
Pretoria: Department of Labour.
Standing, G., Sender, J. & Weeks, J. . Restructuring the labour market: the South African challenge. Geneva:
International Labour Organization (referred to as ILO Review).
Statistics South Africa. . Monthly earnings of South Africans, . Statistical release P . . Pretoria: Statistics
South Africa.
Yu, D. . An overview of real earnings trends of employed in post-apartheid South Africa. In Hofmeyr, J. &
Nyoka, A. (Eds). Transformation audit : confronting exclusion. Cape Town: Institute for Justice and
Reconciliation: – .

Unions, collective bargaining and minimum wages

. Introduction
. Collective bargaining and the labour market
. The union movement
. The impact of unions
. The influence of strike action
. Bargaining councils
. Minimum wages
Trade Unionism is not Socialism: it is the Capitalism of the Proletariat.

G B S
. INTRODUCTION
Chapter discusses the determination of wages in a more or less perfectly competitive labour
market. It shows that in this market wages are determined through the interaction of demand and
supply, and would, theoretically at least, be determined at levels that will eliminate unemployment.
Some imperfections are raised in that chapter, for instance the impact of labour laws on the cost of
labour and monopsony labour markets.
This chapter takes the discussion of labour market imperfections further. One of the most important
factors that has an impact on the free operation of market forces is labour unions. They negotiate
wages and conditions of employment with one or more employers. The outcome of such collective
bargaining usually determines minimum wages in an enterprise or even in a whole sector. However,
where unions are absent or not strong enough to ensure proper collective bargaining, the
government might intervene to set minimum wages for vulnerable workers.
Although the role of unions in preventing the exploitation of workers is generally accepted, there is
discomfort in some quarters about unions becoming too strong. It is feared that strong unions might
cause wide differentials between the wages of unionised and non-unionised sectors, and increase
unemployment. At the same time, there is much debate about whether unions promote or inhibit
productivity and flexibility in the workplace.
These are the issues that are discussed in this chapter. The first section is dedicated to an analysis of
collective bargaining. The union movement and the impact of unions on wages and productivity are
then discussed. This is followed by a discussion of strike action and its particular consequences.
The discussion then shifts to bargaining councils, which are both important forums for centralised
bargaining and a source of much controversy in South Africa. The pros and cons of these councils are
outlined. Among the functions of bargaining councils is the determination of minimum wages, which
are then extended to apply to the workers in a whole sector. Other ways in which minimum wages are
determined in South Africa are mentioned, before the arguments for and against minimum wages are
considered.
. COLLECTIVE BARGAINING AND THE LABOUR MARKET
. . The right to bargain collectively

An individual employee and an individual employer do not have equal bargaining power. This
situation is exacerbated when the worker is unskilled or when unemployment is high. Because of this
unequal bargaining power, it is accepted in most countries and in terms of international labour
standards that mechanisms should be implemented to protect the individual employee against
exploitation. This includes the right to form unions and employee organisations that can bargain with
the employer for acceptable conditions of employment. Principles such as freedom of association, the
right to bargain collectively and to be protected against victimisation because of union activities are
acknowledged worldwide as fundamental human rights.
The right to bargain collectively is also acknowledged in the Constitution of South Africa and in labour
legislation. Collective bargaining therefore plays a major role in the determination of wages in this
country.
As summarised in Figure . , collective bargaining is a process in which representatives of the

employees (e.g. trade unions) and representatives of the employer (e.g. management) interact and
negotiate in an attempt to reach agreements and uphold these agreements regarding matters that
relate to or may have an impact on the employment relationship. Once the terms of a collective
agreement are recorded and signed by the parties, the agreement is applied uniformly across a
specified group of workers. Employers, unions and their members are bound by the terms of this
agreement. Finally, collective bargaining is structured through legislation or by agreement between
unions and employers. In other words, the state does play a role in collective bargaining in some way
(Finnemore, : ; Nel, Kirsten, Swanepoel, Erasmus & Poisat, : ).
FIGURE . Collective bargaining
Bargaining power is not only determined by the formation of unions, but also by the extent to which
bargaining is centralised within different sectors, for instance in a bargaining council. As indicated in
section . , South Africa’s system of bargaining councils probably allows unions and employers
substantive control over the labour and product markets, which might be to the eventual
disadvantage of the consumer, the small enterprise and the unemployed worker.
. . Factors that determine the outcome of collective bargaining

Although the outcome of collective bargaining is more often than not determined by the respective
bargaining powers of the union(s) and the employer(s) involved, market forces and the economic
environment continue to play a role. This also applies to strike action, which is the ultimate sanction of
collective bargaining. It appears, however, that factors such as the inflation rate, profitability,
standards of living, market conditions and, to a lesser extent, productivity have featured much more in
collective bargaining recently.
Some of the most important factors that determine the relative power of the collective bargaining
parties and, therefore, the outcome of collective bargaining are discussed below.
The unemployment position and the effect of the outcome of collective bargaining on employment
do not always seem to be important considerations in collective bargaining in South Africa. This
contrasts with the approach in many other countries, where employment considerations appear to
be much more prominent. Reasons for this may be that the unemployment problem in this country
is already so severe that it seems an accepted fact of life, or that the employment effects often lag
so far behind the wage settlements that the parties to the settlement (in particular the union, but
also the employer) do not recognise or acknowledge the link between wages and employment
(Marshall et al., : ). Furthermore, unions often argue that employees need to receive high
wages when unemployment is high because they have to support the unemployed. The effect of
wage increases on employment is considered in section . . .
Demand conditions, profitability and productivity also have an important influence on collective
bargaining. In buoyant market conditions or when labour productivity is increasing, the employer
will more readily grant wage increases. In recessionary conditions, an employer might be more
willing to face a strike than give a substantial wage increase. Strikes during such periods might
induce the employers to consider retrenchment or reduce the working hours of employees. Strike
activity during recessions pressurises employers less than during boom periods and is therefore less
prevalent than during boom periods or seasonal peaks.
Collective bargaining without the right to strike becomes collective begging.

A
The elasticity of demand for the employer’s product is also an important consideration. If the
demand is inelastic, i.e. the quantity demanded does not vary much with price increases, the
employer will grant wage increases more readily and will pass the cost increases on to the
consumer through higher prices. If the demand is very elastic, any wage increase leading to a price
increase will sharply reduce the quantity demanded for that product and the employer will be more
resistant to wage demands. For instance, consumers tend to be very price sensitive (i.e. the demand
for the product is elastic) if there is intense competition between producers, which means that any
individual producer will attempt to keep wage and cost increases to a minimum. This is discussed in
more detail in sections . . and . . .
In monopolistic conditions employers will also grant wage increases more readily because they can
usually recover higher costs arising from wage increases by increasing the prices of their products
(Ehrenberg & Smith, : ). This is discussed fully in section . . . In a competitive market, an
enterprise might price itself out of the market if the price of the product is increased because of
high wage increases. For the same reason, in a relatively open economy, imported products can
compete with local products and this reduces the local employers’ ability or willingness to grant
wage increases.
The export orientation of both employers and unions has a significant influence on collective
bargaining. If both employers and unions are very conscious of the importance of exports and aim
to keep local products highly competitive in international markets, market forces will play a bigger
role and bring more economic realism to collective bargaining than otherwise.
The employer’s capacity to deal with a strike directly affects collective bargaining. If production can,
for instance, be easily varied or stopped, or sufficient inventories have been built up to provide for
uninterrupted marketing during the strike, the employer will be less willing to grant high wage
increases. Similarly, if the employer does not have a large component of highly specialised workers
but can make use of temporary workers during a strike, the employer might decide to take his or
her chances with a strike (Marshall et al., : ).
The standard of living of the workers and inflation rate are also important considerations. Relatively
low living standards and a high inflation rate might influence the union to fight harder for a high
wage increase and might also make the employer more sympathetic in this regard.
SUMMARY
The outcome of collective bargaining is determined by the bargaining power of the parties involved, market forces and the
economic environment.
. THE UNION MOVEMENT
While there is no single definition of a trade union, typically a trade union is considered as an
association of employees that seeks to represent its members’ interests within the workplace and
regulates the relationship between employees, employers and employers’ organisations (Venter & Levy,
: ).
It obviously hurt him to wear the dinner jacket of respectability instead of the boiler suit of revolt.
S W C
In the early days, unions were formed when industrialisation forced workers into positions of
dependency in which their earnings, working conditions and job security were largely beyond their
control as individuals. As a result, workers increasingly bonded together to prevent exploitation and
promote their interests. In any modern economy, unions are protected through legislation because
they are important mechanisms to protect individual workers and help to reduce the possible
negative impact of market forces on the most vulnerable workers. If unions did not play this role, the
state would have to interfere in the market mechanism to ensure the protection of workers to a much
greater extent than is currently the case.
Unions are critical in addressing some of the shortcomings of a market economy at the level of the
enterprise (the exploitation of individual workers by employers being the most obvious example). In
this regard, legislation should aim to strengthen and protect unions.
Another important role for unions, which is not always recognised, is in the protection of democratic
values (see accompanying box). While legislation should protect unions, it should also aim to achieve
the correct balance between voluntarism and compulsion, and between individual rights (e.g. the
discretion that individual employers and employees have to agree on conditions of employment) and
collective rights.
A recent policy discussion on the back of the platinum industry strike was to introduce
mandatory secret ballot voting when deciding on, among others, whether or not to engage in
industrial action, in order to ensure that no intimidation can take place. The argument is that union
leadership may intimidate members to support a strike, even when this does not reflect the majority
view of the rank-and-file membership. Since the broader union movement is closely aligned to
politics in South Africa, such an abuse of agency would be motivated out of political expediency,
rather than the needs of the membership.
Policy issue: unions as important protectors of democratic values
Kochan (as quoted by Maree, ) has argued that “no democracy can long prosper or indeed even survive in the
absence of an independent, strong, and forward-looking labour movement”. One lesson emerging from the experience of
unions in industrialised countries is that civil and political liberties are essential preconditions for exercising labour rights,
and that only a liberal democracy can provide the institutional environment for fulfilling these rights. In some developing
countries the evolution of democratic institutions such as unions is accepted as part of democracy, but in others union
activities are frowned upon as being “unpatriotic” or economic sabotage. In a number of countries and in South Africa in
particular, trade unions have been instrumental in accelerating the pace of transformation through their sustained
support and solidarity with the struggle for liberal democracy. Only prolonged struggle and profound sacrifice have
brought them closer to the goal of guaranteeing civil and political liberties to a broad spectrum of society. Unions have a
special challenge to continue playing this role in countries throughout the world.
. . Reasons for joining a trade union

There are numerous reasons why workers join a trade union, including the following (Finnemore, :
– ):
Protection from exploitation: it was discussed above that workers bond collectively with other
workers to join a trade union to stand against exploitation.
Economic needs: workers focus on the “bread-and-butter” issues and the contents of their
remuneration packages to ensure they earn enough to cover the increasing living costs.
Job security: many workers join unions to protect their jobs, particularly against dismissals,
retrenchments, unfair discrimination and unilateral changes to working conditions by the employer.
Political reasons: unions played an important role in bringing about the democratisation of South
Africa. For example, during apartheid, unions provided an important forum for black South Africans
to express their frustrations and demands of a disenfranchised workforce. Today, unions still use
their influence to exert political power, particularly given that the LRA made provisions for unions to
embark on sociopolitical protests to protect the socioeconomic interests of their members (Venter
& Levy, : ).
Social needs: some workers join unions for comradeship and community acceptance. Interaction
among union members takes place both within and outside working hours, which improve
friendship and solidarity. Peer pressure may play a role, especially to new employees.
Self-fulfilment: some trade union members undergo training and may develop skills that they would
not have been able to develop in their ordinary daily jobs. For instance, some trade union
representatives or shop stewards moved into management positions within their companies, or
even entered politics and the public service after .
. . Types of trade union

Trade unions can generally be classified into the following three categories (Finnemore, : – ;
Nel et al., : – )
Craft or skills-based trade unions: these are unions that organise workers who possess specific skills
or perform defined types of work, regardless of the area or industry in which the work is performed.
Examples include unions for plumbers, carpenters and electricians.
Industrial or sectoral trade unions: these represent unions that organise workers at an industry level,
irrespective of trade or profession. Examples include unions for industries like the chemical,
construction and mining industries. In the South African legislative context, due to the promotion of
the centralisation of collective bargaining with sectoral-based bargaining councils, sectoral trade
unions have been strengthened in recent years. Note that sectoral trade unions exist in both the
private and public sectors.
General trade unions: these unions organise workers who share common interests and an identity
as workers (regardless of occupation, skill or sector), to stand against the strong power of
employers. While a general union has the potential to develop a large membership, it may not be
sufficiently concentrated in a particular sector, thereby diluting the union’s power base.
. . Factors that have an impact on the strength and influence of unions

The strength and impact of unions are influenced by these factors (Jose, , Webster, ;
European Foundation for the Improvement of Living and Working Conditions, : – ):
Union density
This is the extent to which unions have built a solid base by representing a significant proportion of
the workforce. This in turn is influenced by a wide variety of factors as indicated below.
The union capacity and organisational capability

This is the extent to which unions, irrespective of numerical strength, have the capacity to mobilise
labour successfully, for instance to bring workers out on strike, or organise national protest action
against unpopular social or economic policies. This also relates to whether the unions have the
necessary capacity to engage employers or other social partners at various levels (for instance in the
workplace), in sectoral bargaining structures such as bargaining councils, or at national level on policy
issues. Often one finds a lack of strong and capable union leadership, due to the leaders being
poached by employers or government, and this has a negative effect on the strength and capacity of
the union. Another important factor is the extent to which the union or union federation acts
independently from government or political parties. Even though this might at times be influenced by
legislation, this is not always the case, and the independence of the union movement is then based on
the approach by the union itself.
Labour institutionalisation
This is how much worker and union rights, as well as union participation in decision-making
processes, have been institutionalised through labour legislation, collective agreements and union
participation in the administration of benefits. As far as legislative provisions are concerned, unions
have a priority to ensure job security, unemployment insurance and special benefits on termination of
employment, as well as social security providing for health care and pension schemes.
Globalisation
Globalisation has led to intense and competitive pressure in product markets, accelerated the mobility
of capital and added to the vulnerability of labour. Unions are therefore under pressure to participate
in the development of wage policies that accommodate productivity differentials for greater efficiency
in resource allocation. The increasing mobility of capital means that “the cost of labour is back … in
the competitive sphere” (ILO, ). Modern human resources management policies require a union
to facilitate cooperation-based methods such as teamwork, and modern incentive payment systems
or similar policies are used to counter the influence of unions.
More insecure and atypical employment

Flexible labour market policies have been gaining greater legitimacy and political support than before,
owing to factors such as globalisation, technological change and, in some cases, worker preferences.
Practices such as subcontracting, outsourcing and the hiring of temporary and part-time workers,
long considered as atypical employment, are becoming more common, especially at the lower end of
the labour market. The net outcome is an increased segmentation of labour markets, with one group
of workers having very insecure jobs on the one hand and, on the other hand, (often unionised)
workers having more secure employment in the primary labour market (see section . . ).
Atypical employment, contract work, temporary employment or labour brokering has steadily grown
in South Africa. The South African government has historically been one of the biggest users of
temporary employment; in the / financial year, nine national government departments
admitted to making use of labour brokers, at a cost of R million (Democratic Alliance, a), with
the industry in total generating R billion in taxes in the same period (Democratic Alliance, b).
Some argue that the rise can be ascribed to the opportunity it presents for employers to side-step
rigorous labour legislation. The counter-argument is that atypical employment exposes workers to
abuse and an implied waiver of their rights. In response to these concerns, legislation was passed in
, which provides that any fixed-term contract for work (i.e. the typical legal vehicle used for
atypical employment) over three months could be deemed to be indefinite should the contract not be
linked to the completion of a task or event, among others.
A politically important task for unions is thus to build distributive institutions to defend the interests
of workers at the lower end of the market, particularly the vast reservoir of workers in the informal
sector. Ideally, this could be attained through macrolevel framework agreements encompassing
minimum standards of employment, minimum wages, portable benefits including health care and
safety nets, which workers are entitled to irrespective of the location of employment. The question is
whether unions can empower themselves to ensure a secure income, decent working conditions and
social justice for all, and not focus on their members only. Such empowerment is a precondition for
unions to emerge as credible partners ensuring social cohesion. Webster ( : ) calls this role of
unions “social unionism”. In this regard unions can also play an exceedingly important role in ensuring
democratic principles in countries (see preceding box – Policy issue: unions as important protectors of
democratic values).
One of the ways in which unions have responded to this challenge is to extend their membership
drives to non-traditional constituents. These may be grouped into several distinct but overlapping
categories: (i) new entrants at the higher end of labour markets, including professional and white-
collar workers; (ii) casual workers, who are either part-time or temporary; (iii) home-based workers
and those in the informal sector; and (iv) women workers.
Trade unions are ideally placed to lead an initiative for a social minimum wage, consisting of the right
to income security and other entitlements such as education, health, shelter and a safe environment.
Small business
Technological changes have made it possible to reshape production through new forms of industrial
organisation. This has often resulted in a number of smaller enterprises undertaking production or
other activities that large companies might previously have undertaken themselves.
The larger number of production units makes it more difficult for unions to organise workers and to
bargain collectively. Often there is a more personal relationship between workers and employers, and
unions have greater difficulty in recruiting members. This is one of the reasons for unions striving to
achieve multi-employer agreements on minimum standards, for instance through bargaining councils
in South Africa (see section . ).
Geographic dispersion of production

Technological changes have resulted in production being geographically dispersed even outside the
boundaries of urban labour markets. In addition, production or other activities have been outsourced
to other regions or even other countries. The geographic dispersion within a country makes it more
difficult for unions to gain access to every workplace. However, a bigger threat is the ease with which
employers can move production to other countries. This often gives employers more leverage in
negotiations – if the union does not agree to moderate wage increases, the employer will move
production out of the country.
Skills composition
The skills composition of the workforce is changing and workers are increasingly being differentiated
by their competence. At the higher end of the scale, workers tend to be better educated, career
minded, individualistic and less motivated by class interests and solidarity. These workers often have
incentivised pay arrangements and they generally have less need for trade unions. Owing to
technological developments and consumer preferences, employers are also often more intent on the
development of human resources than before. This encourages the trend towards a more skilled
workforce. This is also reflected in the membership of COSATU, which has seen growing numbers of
members with higher skill and education levels (Naledi, a: ).
On the other hand, there is a discernible concentration of workers (mainly women or migrants) at the
lower end in service industries or occupations. This trend has resulted in greater wage disparities and
inequalities between different classes of workers, and this has increased the tensions within the union
movement between the higher-paid and lower-paid employees. Union solidarity might thereby have
been reduced.
The change in skills composition is clear upon analysing the labour survey data. In , % of union
members were semi-skilled or unskilled, with the percentage in being %. The proportion of
members with at least a Matric increased from % in to % in .
Structural adjustments
The shift in focus from the manufacturing sector, which normally has a relatively high union density,
to the services sector, which is less unionised, has contributed to the decline in union density. The
shift towards the information economy, which has no tradition of supporting collective bargaining,
also accelerates the decline in union membership. Initially, the growth of union membership in the
public sector compensated for this decline, but recently there has been a reshaping of the public
sector in many countries, which has reduced the overall union density (European Foundation for the
Improvement of Living and Working Conditions, : ).
SUMMARY
The changing economic environment, often linked to globalisation, has created a number of threats for unions and the
unions have responded in a variety of ways to such threats.
. . The union movement internationally and in South Africa

The profound changes in the social and economic environment in many countries, as outlined above,
have had a negative effect on the position and influence of trade unions.
To determine union strength and influence, it would be more accurate to use union density than
union membership. Union density is defined as union membership as a percentage of the number of
employees as defined in national labour force surveys.
The European Foundation for the Improvement of Living and Working Conditions ( : ) found a
declining trade union density in most European countries between and . Of the
countries for which the relevant information is available, all but two – Belgium and Greece –
experienced a decline in density. Slovakia and Sweden experienced the sharpest declines, followed by
Austria, Bulgaria, Denmark, Hungary, Latvia, Lithuania and Malta. While trade union density is still
decreasing in the central and eastern European countries, the rate of decline has slowed considerably
in most cases when compared with the previous period – .
In spite of the declines in union density in Sweden and Denmark, the union density in these countries
remains at about %. Other countries with equally high union density figures are Finland, Belgium
and Norway. Countries such as the UK, the Netherlands, Portugal, Spain, Germany and several central
and eastern European countries have a union density of below %.
Unions in South Africa are still relatively strong. Using the OHS, LFS and QLFS data and only
considering formal non-agricultural employees, Figure . shows that the number of trade union
members increased steadily from . million in to . million in – representing an
annualised growth rate of . %. Nonetheless, trade union density peaked at % in before
showing a downward trend in general and dropping to . % in .
FIGURE . Trade union membership and density
Note: The question on trade union membership was not asked in the – surveys.
Although not shown in the figure, the number of unions increased sharply between (there were
about registered trade unions) and (this number peaked at ). This was probably
attributed to changes to the LRA, which made union registration easier. In , there were further
changes to the LRA aimed at eliminating consultancies fronting as trade unions. This made union
registration more difficult and thereafter the number of unions started to decline. In December ,
there were registered trade unions (Department of Labour, ).
Figure . shows the proportion of formal non-agricultural employees who reported to be trade
union members by sector in , and , and it can be seen that the mining sector is
associated with the strongest union density – . % of the workers were trade union members in
. This is followed by the utilities as well as community and social services sectors. The
construction sector had the lowest union density in ( . %).
FIGURE . Trade union density by sector, selected years
Source: Author’s own calculations using the October Household Survey, Labour Force Survey September and
Largest union federations

Three trade union federations comprise the bulk of union membership in South Africa. They play a
significant role in the formulation of policy. There are also a number of smaller labour movements
that play a lesser role, although certainly not irrelevant. The most important union movements in the
country are the following (Barker & Holtzhausen, ):
The Congress of South African Trade Unions (COSATU) claims to have approximately . million
paid-up members and affiliates (see accompanying box) (website: http://www.cosatu.org.za). It
was founded in and is the largest labour federation in South Africa. Even by international
standards it has been among the fastest-growing unions in the world. It is founded on the
principles of non-racialism and non-sexism; one industry, one union; worker control of the union;
paidup membership to ensure self-sufficiency; one country, one federation (i.e. to try to unite all
union federations in the country into one federation), and international worker solidarity. COSATU
adopted the Freedom Charter and formed an alliance with the ANC and the South African
Communist Party to ensure “a democratic, non-racial society” in South Africa. COSATU is committed
to a socialist economic system.
The Federation of Unions of South Africa (FEDUSA) was formed in by the merger of the
Federation of South African Labour Unions (FEDSAL) and the Federation of Organisations
Representing Civil Employees (Force). FEDUSA consists of nearly affiliated trade unions with a
claimed membership base of about (website: http://www.fedusa.org.za). FEDUSA is founded
on the principles of: party political independence (it is therefore not aligned to any political party);
equal opportunities and non-discrimination; a democratic labour environment; professionalism,
disciplined action; and support for balanced economic policies. At the time of writing, FEDUSA and
NACTU (see below) had just agreed to start the process to revive the South African Confederation
of Trade Unions (SACOTU).
The National Council of Trade Unions (NACTU) has about affiliated trade unions. It believes in
non-affiliation to political parties, non-racialism and autonomy of affiliates.
Although not a federation, another important union movement is Solidarity (website:
https://www.solidariteit.co.za/en). It claims a membership of . In Solidarity affiliated
with the Confederation of South African Workers’ Unions (CONSAWU). This membership enables
the trade union to take part in the labour debate both nationally and internationally.
Against this background, the impact of unions on the cost and productivity of labour is discussed in
the next few sections.
Largest trade unions in South Africa
The four largest trade unions in South Africa are all affiliated to COSATU. Their unconfirmed membership is as follows
(http://www.cosatu.org.za):
Trade union Claimed membership

National Union of Mineworkers (NUM) 300 000
National Education, Health and Allied Workers Union (NEHAWU) 235 000
South African Democratic Teachers Union (SADTU) 254 000
National Union of Metalworkers of South Africa (NUMSA) 340 000
SUMMARY
Employees join trade unions due to various economic, political, social and personal reasons.
Trade unions are broadly categorised into craft unions, sectoral unions and general unions.
Unions in South Africa are very powerful and members of registered unions form about % of the formal non-
agricultural employment. This is high by international standards and very high by the standards of developing
countries.
The principal bodies representing organised labour in South Africa are three labour federations: COSATU, FEDUSA
and NACTU.
. THE IMPACT OF UNIONS
. . The impact of unions on wages and the cost of labour

Most researchers accept that unions cause an increased wage differential between unionised and
non-unionised sectors. A theoretical explanation is given before some research results are discussed.
If a union bids up wages in the unionised sector, it may lead to lower employment in that sector. In
turn, this may mean an oversupply of labour in the non-unionised sector, which will lead to lower
wages there. Unions may thus cause an increase in the wage differential between unionised and non-
unionised sectors. This is illustrated by the graph in Figure . .
FIGURE . The influence of unions on wages and employment
Source: Adapted from McConnell et al. ( : ) and Ehrenberg & Smith ( : )
Upon being asked what unions wanted, Samuel Gompers, the founder of the American Federation of
Labor, replied: “More”.
A
We assume that exactly half the labour market is unionised and the other half is not. Furthermore, let
us assume that labour supply is perfectly inelastic in both sectors, and the total labour market is in
equilibrium to start with: in both halves wages are W and the number of workers employed N. In
other words, the overall employment equals N.
Let us now further assume that the union in the unionised half (Figure . (a)) succeeds in increasing
wages to WU. The real union/ non-union wage gap is thus (WU – W). However, the wage increase
results in employment in the unionised half of the market being reduced from N to NU workers. This
will leave (N–NU) workers unemployed. Assuming mobility is costless, all of the unemployed in the
unionised sector now migrate and look for work in the non-unionised half of the market, and hence
the labour supply curve in the non-unionised sector shifts rightwards as shown in Figure . (b). This
will force wages down to WN in the non-unionised part (which is clearly lower than WU) in the
unionised part. Employment in the non-unionised sector increases from N to NN. Since labour supply
is perfectly inelastic, all unemployed from the unionised sector successfully find work in the non-
unionised sector, that is, (N – NU) equals (NN – N). Hence overall employment remains unchanged at
N.
Instead of the union/non-union wage gap of (WU – W), it now seems as if the union has caused a
wage gap of (WU – WN) between the unionised and non-unionised sectors. However, this is the
measured gap, which has been caused by surplus labour in the unionised sector migrating to the
non-unionised sector. The measured differential of (WU – WN) is much larger than the actual
differential of (WU – W). It therefore overstates the impact of the union.
As outlined by McConnell et al. ( : ), this impact of the unionised sector on the non-unionised
sector is called the spillover effect. The spillover effect refers to the decline in non-union wages caused
by displaced union workers supplying their services to the non-unionised labour market – they will
“spill over” to the non-unionised sector. It thus appears as if the union has caused a much bigger
wage differential than is in actual fact the case.
However, there are also other effects that union wages might have on non-union wages. One is the
so-called threat effect. This refers to the increase in the non-union wages that is offered by an
employer in the non-unionised sector in response to the threat of unionisation. The employer thus
hopes to discourage workers from joining the union by paying wage rates that are more comparable
to union wage rates. To return to Figure . (b) – the threat effect might increase the wages in the
non-unionised sector from W to WT. The differential now appears to be only (WU – WT). In this case,
the true effect of the union is understated. Nonetheless, at WT, employment in the non-unionised
sector would decrease from N to NT. Note that WT must be below WU, or the workers in the
unionised sector would ask for their wage to increase (for example, during collective bargaining) until
it is higher than the wage in the non-unionised sector.
A final effect is the superior worker effect – the higher wages paid by the union firms will cause
workers to queue up for the “good” union jobs. This will enable the employers to select the most
productive workers (which equals NU) and, in time, the firms will benefit. Such workplaces might thus
acquire better skilled workforces, and there might be a higher wage paid in union firms because of
the higher skills available rather than the presence of the union.
Impact of the elasticity of demand

The extent to which unions affect employment by increasing the cost of labour depends on the
elasticity of the demand for labour, i.e. the sensitivity of the labour demand to wage changes.
This is illustrated in Figure . . Two labour demand curves are presented – DELASTIC, which is
relatively elastic, and DINELASTIC, which is relatively inelastic. The supply of labour is S, which results in
an equilibrium wage of W and employment of E . If the union increases the wage to W , the impact
on employment differs, depending on the elasticity of the labour demand curve. If demand is elastic
(DELASTIC), employment will decline relatively sharply to E . However, if the demand for labour is
inelastic (DINELASTIC), employment will show a smaller decline to E .
FIGURE . Impact of a union where there are different

demand elasticities
The elasticity of labour demand has a direct bearing on union behaviour, as is apparent from the
following (see also Borjas, : – ):
(a) The elasticity will be higher if the firm is easily able to substitute capital for labour as wages
increase. This is one reason why unions may at times be inclined to resist the introduction of
technological advances. Such new technologies increase the possibilities of labour being
substituted by capital.
(b) The elasticity of labour demand will be higher if the demand for the output is more sensitive (i.e.
more elastic) to price changes. A wage increase can increase the price of the output, and if the
quantity demanded for that output declines significantly because of the increased price, the
quantity demanded for labour will also decline significantly. Unions thus sometimes may resist
imported products that compete with locally produced products, because the availability of the
alternative goods will result in the demand for the products being more elastic to price changes.
The demand for luxury goods is also usually more elastic than the demand for basic necessities
because consumers can more easily postpone their purchases or even desist from purchasing
luxury products subject to high price increases.
(c) The elasticity of labour demand will be higher if labour cost constitutes a large proportion of the
total cost of the product. If the share of labour cost is high, an increase in wages will lead to a
proportionately bigger increase in the price of the product, which will encourage consumers to
cut back on their purchases. This will have a greater (negative) impact on the demand for labour.
(d) The elasticity of the demand for labour will be higher if the supply elasticity of other factors of
production is higher. For instance, if the price of capital only increases slightly as a firm demands
more capital to replace the more expensive labour, then the supply of capital is elastic and it is
more profitable to substitute capital for labour. If a company outsources certain functions in
order to obtain labour at a lower cost, the unions will try to ensure that the contractors are paid
the prevailing rate, for instance by ensuring that they are also covered by a bargaining council
agreement.
Research results on the wage advantage of unions

As can be gathered from the above, it is difficult to isolate the influence of a union from other
influences, in particular the substitution effect, the threat effect or the superior worker effect. Other
factors such as inequalities in education, variable demand conditions, imperfect knowledge,
monopsony conditions and productivity improvements may also play a role.
To try to determine the influence of unions on wages, researchers usually use two methods of
empirical studies.
Different enterprises within an industry. These studies concentrate on a single industry and compare
wages for different enterprises under varying degrees of unionisation in that industry. The
shortcoming of this approach, however, is that unionised and non-unionised enterprises influence
each other, as indicated above.
After considering various empirical studies, Marshall et al. ( : ) conclude that the craft unions
(those for specific occupations) are capable of securing relatively large wage increases because they
have control over the labour supply. In this instance the wage differential because of unions is
about %. In industrial union situations, the wage advantage seems to be much less marked.
Interindustry. This type of study compares different industries with varying degrees of unionisation
to determine the impact unions have on wages. The shortcoming of this approach is that it is very
difficult to find industries with varying degrees of unionisation but where other conditions, such as
the skills level of the workforce, the occupational groups, the labour supply and the market
conditions for the product, are the same. It would not be sensible, for instance, to compare the
agricultural and manufacturing sectors.
This type of study estimates that the average percentage difference between unionised and non-
unionised wages was about % in the period – in the US (Marshall et al., : ). The
differences were most notable in recessionary periods, during which unions caused downward
rigidity in wages, whereas in non-unionised industries wages tended to decline. The union wage
advantage was eroded during times of inflation and buoyant economic conditions because all
employers then competed for workers by offering higher wages.
In general, however, one should take into account that the extent to which unions are able to
influence wages depends on the extent to which economic considerations influence collective
bargaining. This issue is discussed in section . .
International evidence
Hamermesh and Rees ( : ) concluded that unions have caused wages to peak at between
and % higher than in non-unionised situations. Another study by Blanchflower ( ) indicated an
average union wage premium of . % from – in the UK and % in the US.
Hamermesh and Rees ( ) emphasise, however, that results are inconclusive because productivity is
probably higher in unionised enterprises since, over the long term, unionised enterprises can compete
with non-unionised enterprises only if productivity is higher. However, in many cases the increases in
productivity might have been gained through reduced employment. This issue is discussed in section
. . .
McConnell et al. ( : ) refer to studies that found a much lower wage advantage of unions. In
, for instance, the wage advantage was only %. In the s, the advantage was higher, peaking
at % in , with some other researchers finding a wage advantage of between and % in the
s. This large wage advantage can be ascribed to two factors: the high inflation caused by the oil
price shocks resulted in high union wage increases, whereas the high unemployment resulted in low
wage increases in the non-union sectors.
These findings are in the main confirmed in a review of more than a thousand studies of the impact of
unions and collective bargaining by the World Bank ( ). Union members on average get
significantly higher wages than workers not affiliated to a trade union. The wage markup can be as
high as % in the US, with most other industrial countries having a markup of between and %. In
South Africa, the World Bank found a relatively high markup for union members, as is discussed
below.
Demand on strike picket’s placard: “ % of nothing is nothing … We want %”.

A
Borjas ( : ) refers to different wage advantages in different sectors (with sectors such as
construction experiencing high advantages). Furthermore, lower-skilled workers experience a higher
union wage gap than skilled workers.
Wages may have an impact on levels of unionisation
It should be kept in mind that the extent of unionisation does not only influence wage increases – successes with
increasing wages also influence the extent of unionisation. There might be a correlation between highly unionised
enterprises and high wage increases, but this does not necessarily indicate a causal relationship: wages might increase
sharply, but not simply because of a high degree of unionisation. In fact, it may be that more employees join the union
because of the union’s successes, which means that high wage increases have resulted in an enterprise being highly
unionised, rather than the other way around. Van Heerden and Van Tonder ( : ), for instance, have found that
among whites, coloureds and Asians the rate of growth of unions can be explained by the increase in real wages as well
as union density (i.e. the higher the density, the lower the trade union growth).
Impact on inflation
As outlined in Chapter , economists are generally divided on the role of wages in causing,
perpetuating or merely transmitting inflation. This will equally apply to the impact of unions on
inflation – once inflation begins, it becomes nearly impossible to determine whether the higher wages
caused by unions are pushing prices upwards or whether wages are being pulled upwards by price
increases. However, what does seem clear is that unions appear to perpetuate existing inflation and
increase the difficulty of reducing inflation once it has begun.
South African evidence

It seems that unions have caused a higher wage differential in this country than in other countries,
and there are a number of relevant research results in this regard.
Most of the studies arrive at more or less similar results (see for example Hofmeyr, ; Moll, ;
Hofmeyr, ; Fallon & Lucas, ; Armstrong & Steenkamp, ). After controlling for observable
variables such as industry, personal characteristics and region, the research results can be summarised
as follows:
(a) Union pressure resulted in a relatively sharp increase in the wages of lower-skilled Africans
relative to non-union workers.
(b) The reason that wages in sectors such as manufacturing and mining increased more rapidly than
in domestic service and agriculture can to a significant extent be ascribed to the higher union
density in the former sectors.
(c) The union wage premium (i.e. the percentage by which the wages of the average unionised
workers exceeds that of the average non-unionised worker) has been estimated at anything
between and %, for instance between and % in (Hofmeyr, ), between and
% over the period – (Armstrong & Steenkamp, ); in the region of % in
(Moll, ); and between and % in the period – (Fallon & Lucas, ).
(d) Even though the union wage premium varied from year to year, there seems to have been an
almost doubling of the premium between and . In , the premium appeared to be
just below %, whereas in it was between and % (see Armstrong & Steenkamp, ;
Banerjee, ) and seems to have peaked in . However, Bhorat et al. ( : ), in
considering unionised workers not belonging to a bargaining council, found that the union wage
premium might have declined slightly over the same period. In the most recent study of Bhorat et
al. ( ), it was estimated that the union wage premium could be as low as %.
(e) Other factors might or might not have played a role in explaining this differential. Armstrong and
Steenkamp ( : ) found that the wage gap in was not driven by differences in the
characteristics of workers, but that this might have started to change towards . Hofmeyr
( ) argues that non-observable characteristics such as decision making and other factors (e.g.
motivation and discipline) might have resulted in workers finding themselves in the various
unionised or non-unionised segments of the labour market.
(f) In recent years, unions in the public sector have been considerably more effective at increasing
the wages of their members in the public sector than was the case for their members in the
private sector. Bhorat et al. ( : ) found a wage premium for unionised workers in the public
sector not belonging to a bargaining council of . %, compared with a lower wage premium of
. % for those in the private sector.
(g) Unions also tended to compress wages across skills levels by securing higher minimum wages for
the unskilled. This resulted in the union wage premium for unskilled workers being higher than
for semi-skilled and skilled workers (see Table . ).
TABLE . Mean monthly earnings of formal sector employees by skills level and trade union
membership ( December prices), and
Sector [A]: Unionised employees [B]: Non-unionised employees Ratio: [A]/[B]

2011
Unskilled 5 509 3 065 1.798
Semi-skilled 8 216 5 950 1.381
Skilled 16 035 18 263 0.878
2015
Unskilled 5 878 3 113 1.888
Semi-skilled 8 414 5 895 1.427
Skilled 16 672 17 156 0.972
Sources: Author’s own calculations using the and Quarterly Labour Force Survey quarter data.
Note: Employees who reported zero or extremely high earnings (outliers) are excluded.
SUMMARY
International evidence shows that the wages of unionised employees have on average been between and %
higher than those of non-unionised employees.
Unions have generally caused higher wage differentials between unionised and non-unionised workers in South Africa
than is the case in other countries. However, the union wage premium is smaller after taking bargaining council
agreements into consideration.
The wage premium is higher for the unskilled unionised workers.
. . The impact of unions on flexibility and productivity

Even if unions are responsible for an increase in wages, the cost effect can be neutralised or reduced if
unions also bring about an equal or higher increase in productivity, as this will reduce unit labour
costs. The impact of unions on productivity therefore needs to be considered.
On the negative side, there are the following possible influences (see McConnell et al., : ):
Unions may impose work rules that diminish productivity in an enterprise. They may, for instance,
impose restrictions on the kinds of jobs workers may perform or the assignment of workers to jobs.
They may also engage in so-called “featherbedding” practices by forcing an employer to employ
more employees in a particular occupation than required, for instance a certain ratio of apprentices
per artisan.
Unions may refuse to have new shift systems implemented or may hamper the introduction of new
technology.
Unions may insist on promotion in line with certain criteria unrelated to merit and efficiency, such
as seniority, affirmative action or the person having strong union sympathies.
Strikes can have a negative influence on output and eventually on productivity, although as
indicated in section . . , the data on strikes may be misleading as a measure of the cost of a
strike.
The possible positive influences of unions include:
Union wage increases may accelerate the search for cost-reducing and productivity-increasing
technologies, or may prompt employers to ensure that workers are more productive because of the
higher direct and indirect expense entailed in employing unionised workers.
Unions may improve job satisfaction by acting as a collective voice for dissatisfied employees.
Unions also tend to emphasise training by which productivity in general can potentially be
increased.
Enterprises may be forced to adopt better methods of dealing with personnel and production to
meet union wage demands and to maintain profitability.
As indicated in the box in section . . , a wage increase can, under certain circumstances, lead to
an equivalent productivity increase (the efficiency wage hypothesis). If the union is responsible for
the wage increase, it will also indirectly be responsible for the productivity increase.
Unions may also encourage further training initiatives by the employer either in response to union
demands in this regard, or because the employer tries to improve productivity to counter the
impact of high wage increases. Armstrong and Steenkamp ( : ) have for instance found that
unionised workers appear to have invested more in human capital, illustrated by a higher mean
level of education and experience among union members relative to non-union workers. However,
the results could also be interpreted differently, namely, that employers operating in a unionised
environment prefer to appoint better educated workers or workers with more experience because
of the wage premium associated with a unionised environment.
The results of overseas empirical studies with regard to whether unions improve or detract from
productivity are inconclusive. McConnell et al. ( : ) list a number of studies indicating that
productivity was both higher and lower in unionised industries. In one study both positive and
negative effects were found for different industries. However, if there was an impact, it seems that the
positive impact on productivity is the greatest in industries where the wage advantage is the highest.
This is probably due to firms responding to the high cost of unionised employees by becoming more
productive. Ingram and Lindop ( ) also reviewed evidence and found that productivity growth was
slower on average in unionised plants than in non-unionised ones during the late s. In the early
s there was a faster growth in productivity in unionised plants, while in – , very rapid
productivity growth was recorded in unionised and non-unionised manufacturing establishments,
with the non-unionised plants recording faster average growth overall.
In a study that reviewed more than a thousand investigations on the effects of unions and collective
bargaining, the World Bank ( ) found that in firms where industrial relations are of a “high” quality
(in terms of a low number of unsolved grievances, low strike activity, etc.) the presence of unions
tends to increase productivity levels.
One difficulty in measuring the influence of a union is that unionisation is higher in larger plants and
in more mature enterprises. These plants and enterprises are likely to experience slower productivity
growth for reasons unrelated to the presence of trade unions. In South Africa, some wage
negotiations since show that both management and unions accept the importance of
productivity improvements. In some instances, such as the mining industry in , wage increases
have been linked directly to productivity.
What is clear is that the industrial relations climate is critical in determining whether unionisation is
accompanied by higher or lower productivity. If industrial relations are good, with management and
unions working together to produce a “bigger cake”, productivity is likely to be higher under
unionism. In an atmosphere of conflict, productivity is likely to be lower in unionised enterprises.
SUMMARY
Unions may have a positive or negative effect on productivity. Empirical studies conducted overseas in this regard are
inconclusive.
. . The impact of unions on equality objectives

Without unions, employers may tend to pay differential wages to individual workers on the same job,
or to pay much higher wages to workers with better skills than workers who are in oversupply,
irrespective of their marginal productivity. This may also relate to differential wages being paid owing
to the dual labour market, as discussed in earlier chapters.
Unions promote equality by insisting on higher wage increases for lower-skilled workers, thereby
reducing the gap between skilled and unskilled workers.
A review of more than a thousand studies on the impact of unions and collective bargaining by the
World Bank ( ) confirms this contention – union membership reduces wage differences between
skilled and unskilled workers and between men and women.
In a country such as South Africa, bargaining councils also allow unions to seek standard wages for
given occupational classes within the same industry, thereby promoting equality among firms with
different productivity and profitability profiles within the industry.
Hofmeyr ( ) has found that from onwards there was a sharp increase in the wages of those
with lower skills levels. In fact, wage rates rose in inverse proportion to skills level, and this reduced
inequalities. Hofmeyr ( ) concludes on the basis of available evidence that union pressure was the
major factor driving African wage levels after . Hofmeyr’s findings are substantially confirmed by
Moll ( ).
The South African Reserve Bank ( : ) states that the wage compression policies of trade unions
have resulted in lower inequality in the unionised sectors than in the non-unionised sectors.
This is confirmed by the findings of Armstrong and Steenkamp ( : ). They found that the
advantage of belonging to a union was greatest for workers at the lower end of the wage distribution.
Union membership may therefore have decreased the level of inequality within the union sector. They
also found that the inequality-reducing characteristic of unions appeared to be increasing over time.
However, even though unions seem to reduce inequalities in unionised firms or sectors, they appear
to increase inequalities between unionised and non-unionised sectors. Because they raise the average
wage in the unionised sectors relative to the non-unionised sectors, they increase the inequality
between these sectors as discussed above.
SUMMARY
Unions in general reduce the level of wage inequality within a firm and also among firms in a particular sector. However,
they increase inequalities between unionised and non-unionised firms, and between unionised workers and the
unemployed.
. THE INFLUENCE OF STRIKE ACTION
The ultimate sanction of unions to enforce their demands for higher wages, i.e. strikes, has both costs
and benefits. The media especially are fond of referring only to the high cost of strikes to the
economy. One cannot deny that strikes do entail a cost to the economy (and to the employers and
workers involved), but the possible benefits of a dispensation that acknowledges strikes as part of the
basic rights of workers should not be ignored.
Strike activity is an integral part of collective bargaining, and the power relationship between
employer and employee would be severely distorted without the right to strike (see section . for a
discussion of bargaining power). The overall benefit of collective bargaining is that workers have more
influence over their working environment, and this can conceivably lead to improved productivity. As
stated before, collective bargaining also helps to avoid exploitative actions by the employer, thus
reducing the need for government intervention. The system of collective bargaining should be seen as
a process of long-term relationship building between employer and employee (see Marshall et al.,
: Chapter for a full discussion).
Strikes are a healthy sign of a free economy, i.e. of a market-oriented economy. To severely limit the
right to strike would increase the necessity of introducing other measures to protect workers, which
would mean more central control of the economy. In general, the efficient operation of the economy
would then be negatively affected.
It is so that industrial action is, by its very nature, disruptive. – Justice Maya in SATAWU v Moloto
A
Strikes certainly do have major cost implications for enterprises, employees and often the economy
(discussed in section . . ). At the same time, the acknowledgement of the right to strike does have
certain benefits. In addition, the loss of working time because of strikes does not necessarily imply a
proportional loss of production. Whether this is the case or not depends on a number of factors. For
example, the long platinum sector strikes of were costly and the cost extended beyond the cost
of production outage. However, it seems probable that the loss of production through strikes and the
loss of striking workers’ earnings are usually much less than would appear from the loss of working
time (see Marshall et al., : – for a full discussion). Some of the factors that have a bearing
on the extent of losses suffered as a result of strikes are referred to in section . . For instance, losses
will be reduced if employers are able to stockpile production before the strike, if they are monopolist
producers, if overtime worked after the strike can catch up on some of the backlog, or if replacement
workers can continue production during the strike.
A survey of businesses that employ at least employees by the South African Chamber of
Commerce (SACCI) showed that (SACCI, ):
Unionisation within a business follows an “all-or-nothing” trend with businesses likely to be either
completely unionised or have no unionised workers at all.
A correlation between unionisation levels and the frequency of strikes. Businesses with zero
unionisation levels had a . % probability of not experiencing a single strike in the previous three
years and a . % chance of one strike occurring. Firms with near-universal unionisation levels have
a . % chance that there was a strike every year for the previous three years.
As soon as a firm has any level of unionisation, there is a near certainty that strike activity will occur
once every three years, but employers with zero unionisation levels are nonetheless exposed to
strike activity, albeit at lower frequencies.
The harmful effects to a single firm of general strike activity in the economy or sector are worse
than a strike of workers in the same firm, which suggest that firms are better able to plan for their
own strikes than a general strike that affects the entire economy.
. . The Hicks strike models

Figure . explains that strikes are not Pareto optimal. Let us assume that there are R worth of
rents to be shared between the trade union and the firm. The downward-sloping line shown in the
figure illustrates the various ways in which these rents can be shared. Now the firm makes an offer at
point RF, keeping a bigger share of R and giving the union the remaining R . In contrast, the
union prefers point RU by getting R for its members (employees) and giving the firm R .
Obviously, the two parties do not reach an agreement and so a strike takes place.
FIGURE . The Hicks paradox
Source: Adapted from Borjas ( : )
As a result of the strike, workers may lose their income; the firm’s profit may decline and even lose
customers permanently. Due to these costs, the size of the economic rents decreases from R to
R . If the parties eventually agree to share the rents equally, they would settle at point R (each
party gets R ). If each party insists on getting a bigger share, the strike will persist and become more
costly, and the size of the economic rents will shrink further, let’s say from R to R . Assuming the
parties finally agree to share the rents equally, the final settlement takes place at point R , with each
party getting R .
Had the two parties agreed to a prestrike settlement at point R by sharing the rents equally (R
each), both parties would have been better off, compared to points R and R . Therefore, it can be
said that strikes are not Pareto optimal.
Figure . presents another Hicks model, which relates to the determination of strike duration. The
curve C is the employer’s concession curve, representing the firm’s highest wage offer at each point
during the strike. At the start of the dispute (or t = ), the firm is willing to pay a low wage of WF, due
to the fact that the firm may still have plenty of unsold inventories that can be used to satisfy the
customers. However, as the strike takes place and its duration increases, these inventories are
exhausted and the firm starts to worry about issues like customer dissatisfaction, lower labour
productivity and lower profits. Hence, the firm would increase its offered wage as strike duration
increases. In other words, the employer’s concession curve is upward sloping.
FIGURE . Hick’s strike duration model
Source: Adapted from Ehrenberg & Smith ( : )
The curve R is the union’s resistance curve, which stands for the lowest wage that is acceptable to the
union at each point during the strike. Initially at t = , the union only accepts at least WU. However, as
the strike continues, the union may fear that the firm might hire replacement workers or even
retrench workers due to economic losses. Hence, the union would eventually decrease its asking wage
as the strike duration increases. In other words, the union’s resistance curve is downward sloping.
At the end, a settlement is reached at point e where the two curves intersect, indicating that the strike
comes to an end after t* days, with a wage settlement of W*. Finally, the shift of the curve(s) could
take place, but this falls beyond the scope of this textbook and is not discussed.
. . The influence of industrial action in South Africa

It is common knowledge that strike activity in South Africa increased considerably in the s and
especially in the second half of that decade, as shown in Figure . . Compared to previous years, it
appears that industrial action in South Africa increased to a new, higher plateau after the Wiehahn
reforms to labour legislation in the early s, and especially after the formation of COSATU in .
FIGURE . Strikes and work stoppages in South Africa, –
Source: Department of Labour
The four high points of industrial action in this graph are also significant. The first is in , the
mining industry strike. The number of work days lost as a result of the strike is quite apparent. This
shows that the strike had a significant impact on the mining industry. The second high level of strikes
took place in , when a large number of workers were involved. However, not many work days
were lost, which is an indication that the strikes were of relatively short duration. The third and fourth
high points happened in and respectively due to the public sector workers strikes. In
particularly, nearly . million workers took part and a huge . million work days were lost in .
Even though the cost of strike activity to the economy cannot be denied, the right to strike does play
an important role in voice regulation, i.e. employers and employees determining, without government
interference, conditions of employment through collective bargaining. The right to strike and lockout,
provided they are not abused, improve the probabilities that employers and unions will agree on
economically justifiable and socially acceptable conditions of employment.
He uses statistics as a drunken man uses lamp-posts for support rather than illumination.
A L
The cost of strike activity should not be exaggerated. The total average annual loss of work days in
the period – , when a large number of work days were lost, made up only about . % of
total working time. This was equivalent to each worker taking off minutes once every month. Such
time loss can be made up relatively easily by improved management practices to ensure efficient
utilisation of the time at work. The loss of labour days through strikes in the period – was
also only one-tenth of the loss of labour days through industrial accidents.
The industrial action in the platinum sector lasted five months and cost an estimated R billion in lost revenues for
the mining companies involved. The Association of Mineworkers and Construction Union (AMCU) originally demanded
an immediate doubling of basic wages to R per month but the ultimate settlement was a R per month
increase. Taking into account the adjustments to existing benefits, the increase came to . %. Reports of widespread
violence and intimidation against workers continued throughout the period. The severe economic and human cost of the
industrial action has since prompted senior representatives from government and the private sector to suggest
fundamental changes to the South African industrial relations framework. One salient policy suggestion that emerged
was a so-called “forced arbitration” by which an official can force the parties to return to negotiations and suspend the
industrial action. This measure would be triggered by evidence of protracted industrial action with broader economic
impact. The urgency of a policy amendment was emphasised when the steel manufacturing sector entered into industrial
action just weeks after the platinum sector reached agreement. The ultimate impact of the industrial actions in both
sectors was a severe loss of investor confidence as illustrated by the downgrading of South Africa’s internationally traded
government bonds by two credit rating agencies. In October , the International Monetary Fund (IMF) downgraded
South Africa’s GDP forecast from . % to . % growth based largely on the malign impact that industrial action in the
platinum sector had on the broader South African economy.
SUMMARY
Although industrial action does have a cost to the economy, this cost should not be exaggerated, particularly as the
right to strike is an integral part of collective bargaining, with such bargaining reducing the need for government
intervention.
Strike activity in South Africa reached high points in the mid- s, , and .
. BARGAINING COUNCILS
. . Background to bargaining councils

Bargaining councils are permanent collective bargaining institutions established on a voluntary basis
between one or more employers’ organisation(s) and one or more trade union(s) for a specific sector
or area, in order to facilitate collective bargaining to negotiate wages and other conditions of
employment and to resolve disputes (Barker & Holtzhausen, ; Venter et al., ). The union(s)
and employer organisation(s) that form the council have to be sufficiently representative to form a
council. What constitutes sufficiently representative is not specified in the LRA, but in respect of the
unions it is usually the percentage of employees in the sector that are union members, and in respect
of employers the percentage of employees in the sector employed by them.
Bargaining council agreements can also be made applicable (the so-called extension of agreements)
by the Minister of Labour to parties that were not party to the agreement, but that fall within the
council’s jurisdiction. The minister is obliged to do so if the employer parties employ, and the union
parties have as members, the majority of employees within the scope of the council. If this is not the
case, the minister may still extend the agreement if the parties are “sufficiently representative”, and if
he is satisfied that failure to extend the agreement may undermine collective bargaining at sectoral
level. This particular provision in the LRA is highly controversial, as will be pointed out below.
In the High Court ruled in Free Market Foundation versus the Minister of Labour on a
constitutional challenge section of the Labour Relations Act (LRA), which permits the extension of
collective bargaining agreements concluded at sectoral level to persons not directly involved in the
collective negotiations and not party to the agreement concluded in the bargaining forum. The
argument advanced before court was that, sectoral bargaining and the extension of the products of it
to non-participants, far from advancing the protection of vulnerable workers, are an impediment to
the growth of small businesses resulting in less job creation and a higher rate of unemployment.
The application hinged on Section ( ), which provided that the minister must extend the collective
agreement as requested by a bargaining council to any non-parties to the collective agreement that
are within its registered scope and are identified in the request. The argument of the application was
that “must” should be replaced by “may” in order to ensure that the minister exercise his or her
judgment. The action was unsuccessful and strengthened the principle of bargaining council
agreements in South African jurisprudence.
Coverage of bargaining councils

There are about councils in South Africa and about one million workers are covered by bargaining
council agreements. However, this covers only the private sector, and the bargaining councils in the
public sector (central government and local authorities) should also be included. The Public Service
Co-ordinating Bargaining Council (PSCBC), for instance, covers another one million workers. There are
a further workers in the bargaining council covering the local authorities. The number of
employees within the jurisdiction of bargaining councils increased substantially after the inclusion of
the public sector under the scope of the LRA and the consequent registration of the public sector
bargaining council in .
Bhorat et al. ( : ) found that while the institutionalised collective bargaining system covered
substantially more formal sector workers in ( %) compared to ( %), this still meant that
less than a third of the formally employed were covered by bargaining councils. Notwithstanding this,
the overall rise in the number of workers covered by bargaining council agreements between
and was driven almost entirely by the introduction of public sector councils. Thus, bargaining
council coverage in the first decade of democracy was characterised by an erosion of coverage within
the private sector bargaining council system on the one hand and the rapid rise of this system of
bargaining in the public sector. This is confirmed by Godfrey, Maree and Theron ( ).
Although it is true that the coverage of bargaining councils is not extensive, it seems to be especially
high in sectors with a high proportion of mid- to lower-skilled workers, who often have permanent
employment contracts. These are also the sectors where permanent jobs for those with poor
educational levels could most likely have been created. Further research is also required about the
overall economic impact of bargaining councils in the tradable sectors, i.e. those sectors that have to
compete with international competitors, that is, where competing products can easily be imported, or
where South African products are struggling to compete on the international markets, or where
services are provided that are an important input to exporting companies. Apart from the clothing
sector, this could include sectors such as motor components, chemical, metal and engineering,
leather, road freight and Transnet (the latter two would relate to transport costs).
. . Importance of bargaining councils

(a) Probably the most important reason for bargaining councils is that minimum conditions of
employment are set by the parties themselves without government intervention. Bargaining
councils are probably the only reasonably effective mechanism to ensure that atypical employees,
in particular, are also covered by collectively agreed wages and conditions of employment. This is
because it is especially difficult for unions to organise atypical employees (see Chapter ), with
the result that the normal mechanism of collective bargaining between an employer and a group
of employees to set wages and conditions of employment does not apply to atypical employees.
The determination of minimum wages and minimum conditions of employment across an
industry could eliminate exploitation by employers, and is an important advantage of bargaining
councils. This is discussed more fully as part of the discussion on minimum wages in section . .
A bargaining council is an important mechanism to protect workers in a manner that balances the
interests of employers and employees without state interference. However, whether it is the best
or only mechanism, especially when taking into account the imperatives of competing in
international markets and creating employment, needs more intensive debate and research.
South Africa should also be much more cognisant of the reasons for the worldwide trend towards
decentralised (i.e. workplace) wage bargaining, whereas South Africa is moving in the opposite
direction (see for instance, Ouchi & Takashi, ).
(b) Conditions of employment are standardised for the industry and competition on the basis of
labour costs is eliminated. This compels employers to compete on other bases, for example,
higher productivity and better quality. There is thus not the “race to the bottom” in respect of
labour standards, where every employer attempts to reduce costs by reducing wages and other
conditions of employment. Employers thus have to concentrate on other methods to reduce
costs and increase productivity and workers do not have to bear the brunt of these efforts.
(c) Closely related to the previous point is that bargaining councils remove wage conflict from the
shop floor, which means that employer and employee relations at this level are not damaged by
the bitterness that often characterises wage bargaining. Bargaining councils may also promote
labour peace as there is much more pressure on employers and unions to reach agreement
because of the significant impact that a strike in a bargaining council can have on the whole
sector. Bargaining at the central level can also be more professional, and it is easier to make
financial information on general conditions in the industry available without having to release
information on individual enterprises, which may damage their competitiveness.
(d) Bargaining councils play a very important role in the provision of pension and provident benefits as
well as medical benefits.
(e) Bargaining councils assist with the settlement of disputes. They settle on average about a quarter
of the approximately disputes referred to them every year (information supplied by the
Department of Labour). This lightens the load of the Commission for Conciliation, Mediation and
Arbitration (CCMA), thus saving taxpayers’ money.
However, bargaining councils also appear to have significant disadvantages when considered from an
economic point of view. This is discussed in more detail in the three sections that follow.
. . Economic analysis of the impact of bargaining councils

In economic jargon, bargaining councils may be regarded as a form of bilateral monopoly. In the case
of employers, there is a monopsony because of solidarity among them to fix wages at a certain level.
At the same time, however, labour has a monopoly in the sense that the union bargains a certain wage
level for all employees. In this case, wages are determined by the relative strength of each party.
Because neither party has all the power, some compromise in respect of wages must be reached. The
final wage outcome in such a market will therefore be indeterminate – the negotiated wage may be
either above, below or equal to the competitive wage rate (Marshall et al., : ).
South Africa’s cost competitiveness has been jeopardized by insider-dominated wage bargaining.
OECD
Some economists (as quoted by McConnell et al., : ) reject the proposition that countervailing
or balancing power determines the wage outcome in the case of a bilateral monopoly. These
economists maintain that there is tacit collusion between employers and the union to control both the
labour and product markets, i.e. to suppress competition. In such circumstances, product prices and
wages will be higher than determined by the market. Such collusion therefore has as its objective the
maximisation of the income of employers and union members to the detriment of non-participative
employers (especially small employers), consumers and jobseekers. As discussed below, the impact
could be potentially severe in a sector that is subject to intensive foreign competition (the so-called
tradable sectors), or where the wage rates are set by the larger, formal employers, and do not allow
scope for the smaller or less formal employers to grow and expand.
It is thus not surprising that the type of mesocentralised wage bargaining system found in bargaining
councils does not perform well in terms of employment, economic growth and price stability.
Significantly, the authors of the ILO Review ( : ) point out that because of their macroeconomic
properties for influencing employment, unemployment, economic growth and inflation, highly
centralised and highly decentralised bargaining systems perform much better than intermediary
systems such as bargaining councils. This refers to the so-called “inverted U-shaped relationship” (see
Figure . ) between a country’s labour market outcomes (in terms of wages and unemployment) and
its level of collective bargaining, meaning that both highly centralised and decentralised bargaining
give better labour outcomes than systems such as the bargaining councils (Calmfors & Driffel, ).
The OECD similarly states ( : ) that high levels of coordination – either through greater
centralisation or greater decentralisation – are associated with better employment outcomes by
avoiding inflationary wage demands, reducing the real costs of disinflation and increasing the scope
for interest rate reductions.
FIGURE . The Calmfors and Driffel model of the
impact of extent of collective bargaining on wages
and unemployment
Currently, South Africa is characterised by an intermediate level of wage coordination, which is found
elsewhere to be associated with poor employment outcomes. It also seems as if decentralised but
highly coordinated bargaining has a positive effect on economic performance (Auer, : ). The
ILO questions whether the system of sectionalism contained in bargaining councils is appropriate in
South Africa (ILO Review, : ).
However, the World Bank ( ) has found another interesting outcome: where trade unions are
weak, decentralised bargaining may be the most economically efficient form of collective bargaining,
but where trade unions are strong, centralised bargaining is a better level of bargaining measured
against the criterion of economic performance. In this instance, the World Bank also refers to highly
coordinated collective bargaining, which then tends to be associated with lower and less persistent
unemployment, lower earnings inequality, and fewer and shorter strikes.
If the labour costs of all employers increase in tandem, the possibility of all enterprises passing on the
increased cost to the consumer is much greater. This form of indirect price fixing is in the interests of
employer parties to the council. Because wages are standardised throughout the industry, it might
happen that employers, in the interests of labour peace, agree to higher wages than they would have
done in the absence of a bargaining council. As all employers have to pay the same wages, and all can
pass the higher costs on to the consumer, the incentive to resist wage demands is reduced. The
setting of a relatively high minimum wage could price workers out of employment and
unemployment would consequently increase (ILO Review, : , – ).
The International Panel on Growth appointed by the South African government refers to the existence
of wage rigidities that bind more at the bottom of the pay scale (Hausmann, ). The panel states
that increases in the cost of labour can only in some cases be passed on to the consumer. This is
where all competitors must pay the increased wages, which is the case if the activity is non-tradable
(i.e. not traded on international markets). However, in tradable activities, such as manufacturing,
foreign competitors face a different labour code and the producer cannot as easily pass increases on.
The distinction between the tradable and non-tradable sectors is a central pillar of the panel’s
approach. Their recommendation is that jobs should be created in the tradable sector, because this
sector can create the kinds of jobs that use more of the human resources that the country has in
abundance (i.e. unskilled and semi-skilled workers). Furthermore, the country needs to earn more
foreign exchange in order to sustain economic growth over the long term.
Another important factor not always taken into account in the context of bargaining councils is their
impact on the likelihood of young or inexperienced workers finding employment. National Treasury
( : ) has for instance stated that South Africa’s system of sectoral minimum wages may have
contributed to the low levels of youth employment by pushing up the cost of entry-level workers. The
potential effects of minimum wages on youth employment and unemployment rates have been
examined in a number of international studies. The balance of this international empirical evidence
suggests that high minimum wages can have a negative impact on youth employment by driving a
wedge between youth labour costs and their expected productivity, thereby raising unemployment
and discouraging some youth from entering the labour market.
National Treasury ( ) quotes data from Andrew Levy, which show that the average minimum wage
across all sectors is about % of the average formal sector wage. This is very high by international
standards and far above the average in the OECD ( %), which is already elevated compared to
emerging and developing countries. Furthermore, whereas many countries differentiate minimum
wages by age through the inclusion of subminima for youths, this is not the case for South Africa. The
minimum wage in South Africa therefore does not account for the lower productivity of younger
workers. This exacerbates the implicit gap between entrylevel wages and productivity, and hinders the
hiring of younger workers. One way of negating some of the potential negative impact of bargaining
councils on youth unemployment would be to differentiate wage levels by age (OECD, : ).
Leaving aside the question of wage differentials for young workers, the general wage inefficiencies
caused by bargaining councils seem to be refuted by research undertaken by Bhorat and Van der
Westhuizen ( : ). They find that with the exception of the public sector bargaining council, there
is no evidence of a significant wage benefit for those employees covered by a bargaining council,
particularly in when compared with . As far as the bargaining council for the public sector is
concerned, another study by the researchers points to a wage premium of % in (Bhorat &
Cheadle, : ).
However, what they do not seem to take into account is that the outcome of public sector bargaining
often sets a precedent for other sectors through the demonstration effect or the threat effect, which
might be one reason why there is limited evidence of a significant wage benefit for those employees
covered by non-public sector bargaining councils. In any event, Bhorat’s research seems to be refuted
by other research, which shows evidence of a significant wage effect for bargaining councils, for
instance by Magruder ( ).
However, the conclusion that workers have not benefited from bargaining councils is different when
considering specific occupations. Unskilled elementary workers within a bargaining council earned
more in both and than workers not in a bargaining council. Elementary workers who are
not part of a bargaining council earned about % of the average wage of a bargaining council
member in , and about % of the average wage of a bargaining council member in (Bhorat
& Cheadle, : ). This is precisely where the unemployment problem in South Africa is the most
serious.
In , the only other occupation category that benefited from being covered by a bargaining
council agreement was service and sales workers, whose earnings were almost double those of non-
bargaining council service workers. However, this seems to be mainly because of the coverage of
some public sector workers being categorised as service workers, who were not covered by a
bargaining council in (Bhorat & Cheadle, : ).
. . Extension of agreements
As indicated, the Minister of Labour can be requested by the parties in a bargaining council to declare
an agreement concluded in it to be binding on all parties that fall within the council’s jurisdiction,
even though they are not parties to the agreement. Where the parties in the council are
representative of the majority of employers and employees, the minister is obliged to do so, and in
other cases he has some discretion.
The major rationale for the extension is to buttress sectoral bargaining – if it is cheaper to compete
from outside the council, employers might resign from their employer organisation. This will lead to
an increasingly unrepresentative employer organisation and to the eventual collapse of the council.
However, the extension of bargaining council agreements can also have a particularly negative effect
on non-party employers and particularly small employers. The wages agreed in bargaining councils
are usually based on what the party employers can afford, and these are often the large, profitable
and capital-intensive enterprises. The wages are consequently often higher than the smaller, more
labour-intensive and less profitable enterprises (who are usually not members) can manage.
The authors of the ILO Review ( : – ) refer to seven effects on the labour market caused by
the compulsory extension of agreements, including: making it more difficult for smaller enterprises to
expand employment; higher wage settlements; creating a barrier to entry for small enterprises; and
increasing the dualism in the South African economy.
The government’s Labour Market Commission ( : ) has expressed its “deep concern” that non-
parties are not afforded the opportunity to influence bargaining council agreements. The Labour
Market Commission ( : ) has recommended that the Minister should have greater discretion in
deciding whether to extend agreements, and should specifically take into account the effect of
agreements on non-parties and on job creation.
The extension of agreements, together with the fact that representation on councils is not always
properly regulated, has led to claims by some critics that bargaining councils are “minority rule
masquerading as majority rule” (as quoted by Madden, CEO of a bargaining council, : ).
In trying to determine how significant the impact of the extension of agreements is, consideration
could be given to the percentage of additional employees affected when the agreements are
extended. Godfrey et al. ( ) conducted an extensive survey of bargaining councils. As far as
bargaining councils in the private sector are concerned, employees working for employer parties of
the council form % of all employees (party and non-party employers). This means that an additional
% of employees are covered by the extension of agreements. However, in some sectors the
percentage of additional employees covered is quite a bit higher, for instance in the following sectors,
with the percentages showing the percentage of additional employees covered by the extension of
agreements in :
Building 41%
Metal and engineering 45%
Construction: Electrical 47%
Clothing 48%
Road freight 57%
However, these authors have also researched how representative bargaining councils are in terms of
party employers as a percentage of all employers in the particular sector. In this regard, Godfrey et al.
( ) found that bargaining councils are unrepresentative: party employers make up only % of all
registered employers. As many as councils (out of ) are unrepresentative when examining party
employers as a proportion of all employers, as this proportion is below % for these councils.
National councils and other large councils are even less representative on the third measure than the
system as a whole (seven of the nine councils in this category are unrepresentative on this measure).
Similarly, seven of these nine councils are unrepresentative in terms of the number of employees
belonging to party unions as a percentage of all employees.
Boccara and Moll (as quoted by Nattrass, b) have confirmed that wage levels agreed in
bargaining councils do have a significant effect on smaller enterprises. It is for these various reasons
that the government in its Growth, Employment and Redistribution (GEAR) policy suggested that the
Minister of Labour should have more discretion in the extension of agreements to permit the minister
to bring labour market considerations into play (RSA, : ). However, this was never fully
implemented: if parties to the council have majority representation, the minister has no option but to
extend the agreement if requested by the parties. This would also be the case if the employer parties
to the council are large employers and thus represent only a small proportion of employers in the
industry, but employ the majority of employees.
The OECD ( : ) has in fact stated that a weakening of the legal extension of sectoral bargains
would likely also help with wage moderation, since social partners would know that agreed wage
levels could be undercut by other firms.
Whenever a strike is settled, the consumer knows who will have to do the settling.
A
. . Making provision for special circumstances of employers

Bargaining councils introduce an element of inflexibility. They do not take into account that
circumstances can vary radically from one enterprise to the next, and from one region to another.
Bargaining councils cannot make provision for the special circumstances of individual employers, even
if they are members of the council. They could therefore be an important source of labour market
inflexibility (see section . ).
Two main mechanisms have been introduced to accommodate companies that are unable to afford
the wage rates set by bargaining councils. The first is that small and medium enterprises need to be
represented on bargaining councils. The second mechanism is the provision for companies to apply
for exemption.
The LRA requires the representation of small and medium enterprises on bargaining councils.
However, these businesses are not defined clearly. Furthermore, parties to the council can themselves
decide how seats are allocated among representatives, and the Department of Labour’s overview role
in this regard is thus very limited. The LRA does not regulate this important issue; apart from
prescribing that small and medium businesses should be “adequately” represented it gives no
indication of what this implies (see Madden, : ). There is thus no certainty that bargaining
councils do, in fact, adequately provide for these businesses. A survey among small businesses by
Godfrey et al. ( : viii) confirms that small businesses are not pursuing the option of representation
in councils. This finding is not surprising, knowing all the pressures on small business and the fact that
they would be a lone voice among the big boys.
In another study, Godfrey et al. ( ) state that employer bodies should become more representative
of small business in order to better accommodate their needs. In an extensive survey of bargaining
councils, they have found that employer organisations on large national councils were less
representative of small businesses than employer organisations at smaller regional and local councils.
They also found that the average size of party firms at most of the large national councils is
significantly bigger than the average size of non-party firms. However, their proposed solution is
equally unlikely to address the needs of small business. Small businesses by their very nature do not
have the capacity or knowledge to properly participate in the activities and caucuses of employer
bodies. The researchers also raise the possibility of blanket exemption for start-up businesses, and a
gradual phase-in period. However, this again assumes that small businesses will be able to afford the
types of wages paid by large businesses, which are often more mechanised and hence use less labour
that is more skilled.
The second mechanism in terms of which provision is made for the special circumstances of
businesses is the exemption system. The LRA requires bargaining councils to have a mechanism to
deal with applications for exemptions, and the legislation also provides that bargaining councils
should have an independent body to deal with exemption appeals.
Some researchers state that the exemption mechanism makes more than adequate provision for the
shortcomings of bargaining councils and especially their extension. Godfrey et al. ( : – ) for
instance, examined the data on exemptions from bargaining councils for , and .
They found that for those years, between and % of applications for exemptions were granted,
either in full, partially or conditionally. The majority of applications were granted in full. They
compared these results with data obtained from the Department of Labour (DoL) for and .
The DoL data cover councils in and councils in and show a slightly higher success
rate of about %. The evidence here is that almost % of exemptions are granted.
However, the reality is that in the clothing industry many clothing factories have either closed, moved
to neighbouring countries or refuse to pay bargaining council wage rates, which seems to indicate
that there might be some flaws in these findings.
The fact that bargaining councils grant the majority of applications for exemptions is not the full
picture. Firstly, it is not known how many are granted in full, except that one study states that the
“majority” of those that were granted were granted in full (Bhorat, : ). It would have been much
more useful to provide data on the percentage of the original applications that were granted in full,
rather than simply referring to the “majority” of those that were granted.
Secondly, and more importantly, many employers simply do not apply for exemption. Apart from the
fact that many small employers do not have the wherewithal to apply for exemption because the
procedures could be very cumbersome and time-consuming, employers often have to submit proof
that the majority of employees support the proposed exemption, which is unrealistic (see, for
example, clause ( )(g) of the NBC Main Agreement).
Furthermore, the system of exemption often caters only for unexpected, temporary calamities and not
for the long-term difficulties a company might have in competing on international markets –
exemptions are only granted for a temporary period (usually one year at a time) and an extension of
the exemption is not always easy. Under these circumstances, it is not surprising that many employers
experiencing difficulties do not apply for exemption.
The authors of the ILO Review ( : , ) list a number of criticisms against the system of
exemptions. More fundamentally, it could be argued that there would be few, if any, exemptions if
simpler, less onerous “framework” agreements were to be concluded at sectoral level.
Van Niekerk ( : ) also highlights the shortcomings of the exemption system. He has proposed
that there should be a single national, independent body to deal with applications for exemptions in
bargaining councils, and that this body should consist of experts who are knowledgeable about the
small business environment.
. . The link between wages and productivity

A further disadvantage of bargaining councils is that the link between wages and productivity
becomes tenuous. Wages and productivity cannot be effectively linked at industry level, and
productivity bargaining can really only take place at enterprise or shop-floor level, but then without
any direct link to wages. Even though productivity is usually bargained in the workplace and not in the
bargaining council, any employer would know that very few tradeoffs to achieve higher productivity
are possible if wage negotiations are not conducted in the workplace as well. Unions are less involved
in factors that influence the sustainability of an individual company because they do not negotiate
directly with that company.
Sector-level bargaining has also allowed employers to abdicate responsibility for their relationship to
employees and their productivity levels by blaming the sector-wide deliberations over which they
have little influence (Yudelowitz, as quoted in Business Day, August , ). Godfrey et al. ( :
iv) refer to the fact that there is a perceived serious decline in plant-level bargaining within the
jurisdiction of councils. This suggests that little or no productivity bargaining is taking place.
SUMMARY
The strongest criticism against bargaining councils is that agreements are extended to non-parties. This eliminates
competition among employers on the basis of labour costs, can lead to tacit collusion and may be to the disadvantage
of the non-participating employer, consumer and job-seeker.
There is significant evidence that highly centralised or highly decentralised bargaining systems produce far better
outcomes in terms of employment and inflation than South Africa’s intermediate system of bargaining councils.
. MINIMUM WAGES
Bargaining councils are one mechanism used in South Africa to ensure that a minimum wage applies
in the sector covered by the bargaining council. This is in line with international instruments of the
ILO, which provide for the setting of wages through collective bargaining, and for mechanisms to be
created by the state in those industries in which wages cannot be regulated effectively by collective
bargaining and where wages are exceptionally low. Conventions ( ), ( ) and ( ) of
the ILO deal with the implementation of mechanisms for the determination of minimum wages. These
mechanisms do not mandate the level of minimum wages as such, but rather the procedures for
arriving at this level (Shaheed, : ). South Africa ratified Convention in (National
Manpower Commission (NMC), : ).
. . Arguments in favour of minimum wages

The main argument in favour of a minimum wage is that every worker should be paid at a level that
enables a certain minimum standard of living to be maintained (NMC, : ). Society should have
an improvement in the well-being of all its citizens as one of its objectives. If wages were left entirely
to market forces, many workers (especially the unskilled) could be exploited by unscrupulous
employers and would not be able to maintain the minimum standard of living required to satisfy basic
needs. For competitive reasons, other employers would be forced to follow the same policy, i.e. pay
“starvation” wages. The government should therefore intervene by introducing a minimum wage to
ensure that society’s objectives in this regard are achieved.
Another argument in favour of a minimum wage is that the employer will use labour more efficiently
and therefore improve productivity because of the increased wage costs (NMC, : ). Employers
will be more careful to recruit the right worker for a particular job, and will be more inclined to train
and develop workers. Unprofitable, inefficient enterprises should not be “subsidised” by poorly paid
workers.
In addition, higher wages can, under certain circumstances, lead to improvements in the workers’
morale and nutrition, and reduce absenteeism, illness and labour turnover, all of which can bring about
productivity improvements (ILO Review, : ). This is part of the so-called “efficiency wage”
theory, and is discussed in more detail in the box in section . . .
A further argument in favour of a minimum wage is that the market mechanism does not operate
effectively. There are many distortions of the market as a result of, for instance, discrimination,
uncompetitive conditions in product markets, labour that is not particularly mobile between regions,
occupations or employers, and inadequate information. Furthermore, if there are monopsony
employers, they can exploit labour because of their strong bargaining power and their influence on
the price of labour. In such circumstances minimum wages are required to counterbalance some of
the distorting factors.
. . Arguments against minimum wages

An important argument against minimum wages is that they do not accommodate the specific
circumstances of an individual enterprise (and in the case of a national minimum wage, the
circumstances of a specific industry). The profit position of enterprises constantly changes and a
minimum-wage system does not provide for this eventuality. Minimum wages also do not take
account of the productivity of individuals, and workers should be remunerated in accordance with
their marginal productivity (as discussed in Chapter ).
The patrimony of a poor man lies in the strength and dexterity of his hands; and to hinder him from
exploiting this … is a plain violation of this most sacred property.
A S
Another argument against minimum wages is that entrepreneurs will usually be able to pass the
increased costs resulting from minimum wages on to consumers in the form of price increases. This
can create an inflationary effect. Consumer demand pressures can compound this effect. Because the
introduction of minimum wages may especially benefit many employees, the average consumer
demand level in the country will rise. However, if this is not accompanied by higher productivity (an
increase in the supply of goods and services), higher inflationary pressures will result.
The main argument against a system of minimum wages, especially if set at relatively high levels, is
that it will interfere with the proper and flexible operation of the market, reducing efficiencies and
eventually leading to a lower economic growth rate and higher unemployment than would otherwise
have been the case. Every employee should be free to sell his or her services for whatever wage he or
she can obtain. If a minimum wage is introduced, it distorts the price of labour compared to capital
and results in workers being replaced by capital equipment. In addition, productive investment
becomes less profitable and enterprises may close down or new enterprises not start up. These factors
will result in more unemployment, and society at large will not benefit from the introduction of
minimum wages.
Particularly in periods of economic contraction and declining demand, a minimum wage may imply
that, with no possibility of wages declining, employees will have less chance of keeping their jobs.
They will be dismissed and swell the ranks of the unemployed. In fact, workers who lose their jobs
owing to minimum wages in the protected sectors of the economy may move to unprotected sectors
(e.g. the informal sector) and further depress wages or earnings there. Thus, for the economy as a
whole, wage dispersion and inequality increase. This principle is also discussed in the context of the
impact of unions (section . . ). Even within the formal economy the market mechanism is distorted
as wage differentials between skilled and unskilled workers are reduced, and the incentive for
education and training and promotion is thereby also reduced.
It is necessary to consider these arguments more carefully.
For minimum wages to have these harmful effects on the economy they would have to comply with
the following conditions (compare Shaheed, : ):
To exceed the wage levels the market generates, or to increase all or most wage levels in the formal
sector.
To apply to a large proportion of the workforce.
To increase unemployment.
As regards the first condition, it is not unusual to find that developing countries accorded minimum
wages too ambitious a role in the postcolonial euphoria. However, in most countries governments
soon reduced real minimum wages to more realistic levels, with the result that their effect on wage
levels was relatively modest (Shaheed, : ).
Regarding the second condition, the proportion of the workforce covered varies from one country to
the next. The coverage depends on whether the minimum wage mechanism is intended to ensure
only a bare minimum for the poorly paid and unorganised or whether its role is to ensure a “fair”
wage for a greater proportion of workers. Shaheed ( : ) states that “if a large proportion of
workers received increases in money wages from the general minimum, it is believed that subsequent
inflation would lead to no improvement in real wages, and/or that unemployment would be adversely
affected”. These harmful effects will be avoided if the minimum wages directly affect only a relatively
small proportion of the workforce.
The final condition, i.e. the effect on unemployment, is obviously influenced in part by the first two
conditions. However, it also depends on the elasticity of the demand for labour, i.e. whether the
demand for labour is very sensitive or responsive to wage changes. This is illustrated in Figure . in
section . . , which indicates different demand elasticities.
The line DELASTIC represents a relatively elastic (responsive) demand for labour. If a minimum wage is
introduced at the level of W , then unemployment to the extent of (E – E ) will result. However, if the
demand for labour is relatively inelastic (such as at DINELASTIC), unemployment to the extent of only
(E – E ) will result.
Both Shaheed ( ) and Herr, Kazandziska & Mahnkopf-Praprotnik ( ) refer to various studies
relating to the impact of minimum wages on general employment; Shaheed ( : ) found that “the
general consensus seemed to be that the minimum wage has at best a modest or no impact on
employment”, while Herr et al. ( : ) concluded that “increases in minimum wages do not have
systematic employment effects, neither positive, nor negative ones”.
However, empirical studies also show that labour demand is much more elastic for disadvantaged
groups such as youths, Africans and women than for other workers. This means that even though they
probably are most in need of protection, they will be worst affected by a minimum wage. Various
studies in this regard are discussed in the NMC Report on Minimum Wages ( : ). Fearn ( :
) sums up various findings in respect of the US and concludes that it seems that an increase of %
in the minimum wage there reduced employment among all young people by between and %.
There are thus a number of arguments in favour of, and against, minimum wages. However, what
does seem critical is the purpose of setting minimum wages, which will again affect the level at which
the minimum wages are set. The Labour Market Commission ( : ) found that there was
surprisingly little research on the potential impact of a minimum wage in South Africa, and that recent
studies suggest that the evidence against minimum wages is “surprisingly fragile”, even in less
developed countries. The commission concluded that minimum wages: should be set at a realistic
level; should not attempt to regulate the conditions of too large a percentage of the workforce;
should not be seen as the ultimate solution to poverty in South Africa; and should not be used as a
substitute for collective bargaining.
Saget ( ) uses the ratio of the minimum wage to the average wage as an indication of the impact
of minimum wages. In her study she found that this ratio varied considerably between countries and
within countries over time. According to her, the level of minimum wages has had an insignificant
effect on the level of employment. She also considered the impact of minimum wages on poverty and
found that minimum wages in developing countries do not affect the poorest part of the population,
but rather the upper levels of the low-income population. However, a minimum wage may have
positive results in poverty alleviation by improving the living conditions of workers and their families,
in spite of it not directly affecting the poorest people.
In South Africa, on the basis of the LFSs, Hertz ( ) has found that the employment of domestic
workers fell after the introduction of a minimum wage. Using both the broad and the narrow
definition of domestic worker employment, employment levels fell by . % or . % respectively.
Female domestic workers have been on the receiving end of this decline as their employment
declined by about %. Interestingly, male employment increased by about %, albeit off a low base,
but the reasons for this are unclear.
Overall, Hertz found that the real wages, average monthly earnings and total earnings of all employed
domestic workers have risen since the minimum wage came into effect. The overall estimated
elasticities suggest that the regulations should have reduced poverty somewhat for domestic workers,
although this last conclusion is the least robust of the findings in light of the loss of some jobs.
In a study on the impact of minimum wages on farm workers, Conradie ( ) considered labour
trends on wine and grape farms in the Western Cape. She found that technological advancement far
overshadowed labour laws as the most important reason for job losses. She found that productivity
increased sharply, which was the major reason why there were no largescale job losses as a result of
higher minimum wages. However, she also found that labour laws (including the Extension of Security
of Tenure Act) had the following unintended consequences that need to be taken into account:
The number of permanent jobs has been reduced by half since – for the households involved,
the increase in wages does not compensate for the loss of jobs and benefits.
The wages of the lowest-paid workers did increase substantially, but that of higher-paid workers
increased far less. To some extent, there was an inclination for the minimum wage to become the
maximum wage. This reduced the incentive to improve productivity.
Basic benefits such as health care and subsidies related to housing costs were reduced to
compensate for the higher cash wages. However, these reductions were not sufficient to neutralise
the wage increases.
A recent study by Bhorat et al. ( ) examined the impact of minimum wages on youth ( – years)
and adult ( – years) employment in six sectors in – . The authors found a significant
increase in both youth and adult employment in the retail and taxi sectors, but a significant decline in
agricultural employment after the imposition of the minimum wages.
SUMMARY
Evidence shows that the negative effect of minimum wages on the economy will be negligible provided that they are set
at realistic economic levels, affect only a relatively small percentage of the workforce, are not used to combat poverty,
and are not substituted for collective bargaining.
. . Minimum wage-setting mechanisms in South Africa

Minimum wages in South Africa are determined by a dual system.
Bargaining councils
In those sectors where there is adequate collective bargaining between unions and employers,
bargaining councils normally determine wages and other conditions of employment through
negotiation. Agreements within bargaining councils are reached without government involvement.
However, the government gives statutory force to the agreement by promulgating it in the
Government Gazette. If the parties to the bargaining council are sufficiently representative, the
agreements can be extended to apply to all employers and employees in the specific sector.
Sectoral determinations
In those sectors ( in total) where proper collective bargaining does not take place, the Employment
Conditions Commission can be requested by the Minister of Labour to investigate conditions in the
industry and to make a recommendation in respect of wages and other conditions of employment.
The Employment Conditions Commission advises the Minister on the making of sectoral
determinations and any other matter concerning the BCEA (Basic Conditions of Employment Act).
Three members of the commission are appointed by the minister on the basis of their knowledge of
the conditions of vulnerable and non-organised workers. The minister appoints a further two
members nominated by organised labour and business in NEDLAC.
Table . shows the minimum wages determined by sectoral determination in selected sectors.
TABLE . Sectoral determination in selected sectors
Sector Minimum wage (per Period

hour)
Contract cleaning R17.66 2016 Dec 1 to 2017 Nov
30
Civil engineering – Task Grade 1 R20.50 2012 Sep 1 to 2013 Aug
31
Civil engineering – Task Grade 9 R43.16 2012 Sep 1 to 2013 Aug
31
Domestic workers – Area A (working more than 27 hours per R12.42 2016 Dec 1 to 2017 Nov
week) 30
Domestic workers – Area B (working more than 27 hours per R11.31 2016 Dec 1 to 2017 Nov
week) 30
Domestic workers – Area A (working 27 hours or less per week) R14.54 2016 Dec 1 to 2017 Nov
30
Domestic workers – Area B (working 27 hours or less per week) R13.35 2016 Dec 1 to 2017 Nov
30
Taxi drivers R15.47 2016 Aug 1 to 2017 Jun 30
Forestry R13.37 2015 Apr 1 to 2016 Feb 29
Farm workers R14.25 2016 Mar 1 to 2017 Feb
28
Hospitality (employers with 10, or fewer than 10 employees) R15.17 2016 Jul 1 to 2017 Jun 30
Hospitality (employers with more than 10 employees) R16.91 2016 Jul 1 to 2017 Jun 30
Sources: Department of Labour website ( )
The detailed areas can be found on the Department of Labour website.
When making recommendations on wages, the Employment Conditions Commission has to take into
account the ability of employers to carry on their undertakings successfully, the cost of living and the
operation of small and new enterprises. The commission should also take into account the likely
impact of any proposed condition of employment on current employment or the creation of
employment. After having considered the recommendation, the minister may make a sectoral
determination.
Policy issue: a national minimum wage for South Africa
At the time of writing, negotiations on the feasibility and the appropriate level of a national minimum wage for South
African workers were finalised. According to the results of the proposal presented by a panel of advisors appointed by
the National Economic Development and Labour Council (NEDLAC), it was suggested that the national minimum wage
would be set at about R per month from , with wages in the domestic work and agricultural sectors being set at
set at % and % of the proposed national minimum wage respectively. On the other hand, the study by Isaacs ( )
suggested a minimum wage which begins at levels between R and R , but gradually increases to between R
and R after five years.
. . Minimum subsistence levels
When one considers the issue of minimum wages, attention should also be given to the minimum
subsistence level that applies in South Africa, i.e. the minimum financial requirements of a family to
enable members to maintain their health and have acceptable standards of hygiene and sufficient
clothing. There is, however, no such thing as a general or universal minimum living standard to satisfy
certain basic needs. The concept of basic needs can, for instance, be defined in a variety of ways,
depending on the standard or quality of the clothing, food or whatever is required for subsistence. In
South Africa there are also considerable differences of opinion on how basic needs should be
measured. A further problem regarding the satisfaction of basic needs is whether the individual or the
family should be taken as the basis.
Other factors that should be taken into account when considering general living standards in a
country are the extent to which employees can rely on various fringe benefits provided by the
employer, the availability of state services, and transfers between households. This relates to the
concept of the social wage, which is the total sum of all social benefits received free or partially free,
such as state transfers, subsidies and services (unemployment insurance, state old-age pensions, free
public health services, etc.) and kinship and community transfers (compare ILO Review, : ;
Barker & Holtzhausen, : ). It therefore complements the direct and indirect income of an
employee, such as money wages, fringe benefits and employer-provided entitlements to cover
contingency risks (e.g. ill health, pension, etc.).
Money wages and employer-provided entitlements to cover contingency risks are relatively high in
South Africa. Neither the state nor rural kinship support systems have played a major role because of
the legacies of apartheid (ILO Review, : ). In fact, kinship support has tended to flow from
urban to rural communities, which has raised the wage required by the typical urban worker. Female
unemployment has also been high, which further increases the wage required by those in
employment. In addition, the enterprise share of the social wage has been raised because the state
provides relatively few transfers. A high proportion of the social wage has therefore been borne by
enterprises, which has placed them at a structural disadvantage as far as international competitiveness
is concerned.
In South Africa, the following three absolute poverty lines, proposed by Woolard and Leibbrandt, have
been used in various recent studies for deriving poverty estimates and trends. These lines (in
prices, per capita per month) were derived as follows:
Food poverty line (R ): the expenditure needed to purchase enough food to meet the basic daily
food-energy requirement of kilocalories for the average person over one month, as
recommended by the South African Medical Research Council (MRC).
Lower bound poverty line (R ): this was calculated by observing the essential nonfood expenditure
of households that spent approximately R on food. It was found that these households spent
R on essential non-food items. Hence, the lower bound poverty line was equal to R (= R
+R ).
Upper bound poverty line (R ): this was derived by observing the total non-food expenditure of
households that spent roughly R on food. It was found that these households spent R on all
non-food items. Therefore, the upper bound poverty line was equal to R (= R +R ). This
also implies that the expenditure on non-essential non-food items of the abovementioned
households was R (= R – R ).
SUMMARY
Minimum wages in South Africa are determined by a dual system. In those sectors where there is adequate collective
bargaining, bargaining councils normally determine minimum wages through negotiation. In other sectors, minimum
wages are often set by the Minister of Labour on the recommendation of the Employment Conditions Commission
(through sectoral determinations).
KEY CONCEPTS
Bargaining council
Bargaining power
Bilateral monopoly
Collective bargaining
Employer’s concession curve
Employers’ organisation
Minimum subsistence level
Minimum wage
Poverty line
Sectoral determination
Social wage
Spillover effect
Strike
Superior worker effect
Tacit collusion
Threat effect
Trade union
Union’s resistance curve
FOR STUDENTS
. Discuss any five factors that determine the relative power of the collective bargaining parties.
. Explain the main reasons why workers join a trade union.
. Explain the factors that would increase the strength and influence of trade unions.
. Explain, with the aid of two separate figures, how the wage increase in the unionised sector would have an
impact on wages and on employment in the unionised and non-unionised sectors, assuming an upward-
sloping labour supply curve in both sectors. In particular, explain why unemployment takes place in both
sectors.
. Explain, with the aid of a figure, the impact of labour demand elasticity on employment as a result of wage
increase in the unionised sector.
. Discuss the positive and negative influences of unions on flexibility and productivity.
. Explain, with the aid of a figure, why strikes are not Pareto optimal.
. Explain, with the aid of a figure, how the expected strike duration is determined in the Hick’s model.
. Discuss the arguments for and against the determination of minimum wages.
. Discuss South Africa’s dual system of minimum wage determination.
SUGGESTED READING*
Armstrong, P. & Steenkamp, J. . South African trade unions: An overview for to . Stellenbosch
Economic Working Papers / . Stellenbosch: University of Stellenbosch.
Bhorat, H., Goga, S. & Van der Westhuizen, C. . Institutional wage effects: revisiting union and bargaining
council wage premia in South Africa. South African Journal of Economics, ( ): September.
Bhorat, H., Naidoo, K. & Yu, D. . Trade unions in an emerging economy. WIDER Working Paper / .
Helsinki: United Nations University-World Institute for Development Economics Research (UNU-WIDER).
Bhorat, H., Van der Westhuizen, C. & Goga, S. . Analysing wage formation in the South African labour market:
the role of bargaining councils. DPRU Working Paper / . Cape Town: Development Policy Research Unit,
University of Cape Town.
Herr, J., Kazandziska, M. & Mahnkopf-Praprotnik, S. . The theoretical debate about minimum wages. GLU
Working Papers No. . Berlin: Global Labour University (GLU).
Madden, L. . Bargaining councils: Which way Gulliver? South African Labour Bulletin, ( ): March.
Shaheed, Z. . Does South Africa need minimum wages as part of a system of labour standards? Paper
presented at an NMC workshop on minimum wages. Pretoria.
International Labour Organisation (referred to as ILO Review).
Webster, E. . COSATU at the crossroads: serving the core or the working poor? South African Labour Bulletin,
( ): December /January .

Productivity and labour market flexibility

. Introduction
. The importance and measurement of productivity
. Productivity in South Africa
. Considering productivity and wages together: unit labour cost
. The importance and types of labour market flexibility
. General considerations regarding flexibility
Go to the ant, thou sluggard; consider her ways, and be wise: Which having no guide, overseer or ruler, provideth her
meat in the summer, and gathereth her food in the harvest.
T B ,P :
. INTRODUCTION
Wages cannot be discussed without reference to productivity – it is the joint outcome of wages and
productivity that determines whether South Africa is globally competitive. The joint impact of wages
and productivity is measured by a concept known as unit labour costs.
The importance and measurement of productivity, productivity issues in South Africa and factors that
influence productivity are therefore examined in this chapter, as well as developments with regard to
unit labour costs.
Productivity and flexibility are very closely linked – a company can usually increase productivity if it is
able to increase flexibility. Labour market flexibility is often regarded as an important policy
instrument for improving the functioning and outcome of labour market processes. This issue is the
subject of heated debate in South Africa, to such an extent that the ruling party, the ANC, released a
discussion document in this regard (ANC, ). Various research studies refer either to how inflexible
the country’s labour laws are, or that they are no more inflexible than those of other developed
countries. Flexibility is discussed particularly in the context of the critical imperative to create jobs in
the country, and therefore the perceptions of the owners of businesses cannot be ignored – after all,
they are the major job creators in the economy. However, perceptions are not always an accurate
reflection of the true state of affairs, and the issue of flexibility thus needs to be addressed both at the
level of fact and at the level of perception.
. THE IMPORTANCE AND MEASUREMENT OF PRODUCTIVITY
. . The importance of productivity

Productivity depends to a large extent on the flexibility that an employer enjoys or has engineered
(for instance through a skilled workforce). Productivity is of critical importance for the long-term
wellbeing of any country and deserves proper consideration and understanding.
Nothing is more important for the longrun economic welfare of a country than improving
productivity.
M F
An increase in productivity, and especially labour productivity, is important for the following reasons:
The productivity of human resources determines their wages, while the productivity of capital
determines the return of its holders. High productivity not only supports high levels of income, but
also allows citizens the option of choosing more leisure instead of longer working hours; it also
creates the national income that is taxed to pay for public services, which in turn boosts standards
of living. The capacity to be highly productive also allows a nation’s companies to meet stringent
social standards, which improve standards such as health and safety, equal opportunity and
environmental impact.
It is the basic source of improvements in real wages and thus living standards (see Fedderke &
Mariotti, : ). With regard to the economy as a whole and in the long term, real income per
worker can increase only at the same rate as real output per worker. If income increases more
rapidly than real output, the resulting inflation will reduce the changes in real income to levels
consistent with the output improvements. In the same way, income redistribution cannot increase
the income of some workers at the expense of others as this will at most have only a shortor
medium-term influence on the living standards of the workers of a country as a whole. Productivity
is therefore the real link between output and reward in production. An increase in the productivity
of individuals over the longer term therefore enables these individuals to consume more.
An improvement in productivity is important to attain a higher economic growth rate. Productivity
improvements, rather than low wage levels, should provide countries with an international
competitive advantage.
Productivity growth is an anti-inflationary force in that it tends to offset or absorb increases in
money wages. In this respect it can help to restrain increases in unit labour costs (see section .
below) and, in so doing, help to maintain a country’s international competitiveness.
. . Definitions and measurement of productivity

Traditionally, productivity has been defined as the relationship between real output (i.e. the quantity
of goods and services produced) and the quantity of input used to produce that output (McConnell et
al., : ). It is therefore a measure of input efficiency, which is expressed in terms of a ratio:
quantity of output units of production
Productivity = e.g.
quantity of input units of capital or labour or both
There are various forms of productivity:
Labour productivity is defined as the number of units of output obtained from a unit of labour
input, that is, production divided by number of workers. At a macroeconomic level, it is usually
expressed as real GDP per worker.
Capital productivity is the number of units of output per unit of capital input.
Multifactor productivity, also called total factor productivity (TFP), takes account of both labour and
capital inputs, and is the number of units of output obtained from the combined inputs of labour
and capital. It measures the efficiency of all inputs to a production process. Increases in multifactor
productivity are usually driven by technological improvements or innovations.
The traditional approach to defining productivity suffers from two shortcomings. First, it takes no
account of the quality of the product produced or service provided. By concentrating only on
quantities, quality may suffer, and this may affect the market share of the company and reduce profits.
This leads to the second shortcoming of the traditional approach, namely, that productivity cannot be
defined without reference to the consumer. Productivity should thus compare sold outputs with
inputs.
As a result of these shortcomings, the modern approach is increasingly used to see productivity at the
workplace as a holistic concept, where account is taken of quality, quantities of inputs and outputs,
consumer satisfaction, etc. A useful approach is to concentrate on value added, i.e. the value that
labour and capital add to the total output of the enterprise for the satisfaction of the consumer. This
is measured by subtracting the total cost of inputs from the total value of sold outputs.
It is true hard work never killed anybody, but I figured why take the chance.
R R
SUMMARY
Productivity is normally regarded as the relationship between the quantity of output and the quantity of input, but it is
more appropriate to regard the concept holistically, i.e. also to have regard for sold output, consumer satisfaction, quality
and value added.
. . Shortcomings of productivity data

Apart from the shortcomings of the productivity concept, the following shortcomings exist with regard
to productivity data:
Most productivity studies tend to focus on labour productivity, but on closer inspection “labour
productivity” appears to be a misnomer. In the first instance, it creates the impression that labour is
fully responsible for labour productivity. This is only partly true. As will be pointed out below, most
of the factors that govern productivity can largely be influenced by management, thus labour
productivity is as much the responsibility of management as of workers. Even decisions made at
sectoral or national levels can have a direct bearing on labour productivity. In situations where
workers have very limited influence over decision making at plant or other levels, it is difficult for
them to have more than a partial effect on productivity.
Labour productivity can increase without workers having any role in it. Labour productivity can, for
instance, increase if more capital equipment is utilised (production can be increased by the
utilisation of more capital equipment, whereas the number of workers may remain the same or
even be less). Labour productivity will then increase at the expense of capital productivity. It can be
argued that in these circumstances the employer (who made the investment in capital equipment),
and not the worker, should receive the benefit of the labour productivity increase. On the other
hand, it can also be argued that as the workers carry more responsibility and their skills levels
increase as a result of the capital equipment they use, they should also share in the benefits of the
increase in labour productivity. Overall, however, there may be no increase in total factor
productivity, and neither the employer nor the employee will then be able to share in the benefit of
any productivity increase.
Similarly, labour productivity can also decrease as a result of factors not under the control of workers.
This will happen, for instance, if there is a cyclical downswing in the economy (less production,
same workforce), or if there are interruptions in the production process due to a lack of raw
materials or poor equipment. Workers should not be expected to carry the burden of such
productivity decreases alone.
Another problem with the concept of productivity that should by now be apparent is that it is a very
elastic concept – it changes with the business cycle. During an economic upswing, output may
increase faster than employment, and this will be reflected in improved labour productivity. During
a downswing, output will decline more rapidly than employment (employers try to stabilise their
workforce or are discouraged by legislation from retrenching), which results in poor labour
productivity data even though the workers may not be less “productive”. The same is true for
capital productivity.
The quality of output is not always taken into account when productivity is measured. A better-
quality product may, for instance, be produced without increasing its price, and this will imply a
productivity improvement, but will not be reflected in the data. Improvements in the quality of
output while maintaining unchanged inputs are obviously an improvement in productivity, but
normal productivity instruments will understate productivity growth.
Any measurement of productivity should reflect the combined influence of a number of factors,
including changes in technology, capital investment, utilisation of production capacity,
management skills, the quality of labour, economies of scale and the structure of the economy. Due
to the methodological shortcomings in measuring labour productivity as such, it is important to
give attention not only to labour productivity, but also to capital and multifactor productivity.
SUMMARY
Productivity data should take account of both labour and capital productivity, because labour productivity often increases
at the expense of capital productivity. For instance, if production processes are suddenly mechanised, capital will replace
labour, resulting in an increase in labour productivity and a decline in capital productivity.
. . Productivity and employment
A general misconception regarding productivity is that it necessarily leads to a reduction in
employment, as such arguments ignore society’s desire for additional output and the fact that rising
productivity increases incomes and aggregate demand (see McConnell et al., : ). Increases in
productivity enable society to achieve higher levels of output and, as a result, of employment and of
living standards (see section . . ). The Labour Market Commission ( : ) states that there is no
causal link between increased productivity and job losses – as long as output increases more rapidly
than productivity, the overall employment level will be maintained or increased.
However, these arguments apply to the economy as a whole and it cannot be denied that
improvements in labour productivity have in some instances led to labour redundancies at the level of
the individual enterprise. Under such conditions it is understandable that the workers or unions
involved do not view favourably arguments that the country as a whole benefits from improvements
in productivity. If these misconceptions and suspicions regarding improvements in productivity are to
be allayed, employers and the government will need to give much greater consideration to the
following two aspects:
The benefits of productivity improvements should be shared more equitably with workers. All
stakeholders should gain from productivity improvements.
Workers who are negatively affected by productivity improvements should, as far as possible, be
accommodated elsewhere, for instance by being retrained and assigned to other jobs.
One of the greatest labor-saving inventions of today is tomorrow.

V T. F
Meth ( : ) has also found some evidence that job losses associated with productivity
improvements often occur during phases of economic “distress”. Frequently, productivity
improvement results from both output and employment declining, with the latter falling faster than
the former. This leads to a simultaneous increase in productivity and unemployment (see the
accompanying box).
McConnell et al. ( : ) have found that there is no systematic relationship between productivity
and employment growth in industries in the US. They found that productivity increased in of
industries over the period – , while employment increased in industries and declined in
others. Their conclusion is that one cannot generalise on the relationship between productivity
growth and employment growth.
. . Some factors influencing productivity

Productivity can be improved in a number of ways, and the labour market consequences are not
always the same. For instance, if improved productivity is achieved by improving the skills of the
workforce, it will benefit both the enterprise and its employees. However, this will primarily be a
management responsibility, whereas productivity improvement through greater union–management
cooperation will require a great deal of trust between management and unions.
Is it possible that productivity increase is “harmful” to the economy?
As discussed above, productivity data should always be interpreted with great caution due to their various shortcomings,
and productivity increase does not necessarily lead to employment reduction. Consider the following example on the
labour productivity of five hypothetical companies between and using the traditional approach, with reference
to the data in Table . :
TABLE . Output, employment and labour productivity in five hypothetical companies, –
Company 2016 2017

Output Employment Labour productivity Output Employment Labour productivity
A 1 000 100 10 1 200 100 12.00
B 1 000 100 10 1 600 80 20.00
C 1 000 100 10 2 000 150 13.33
D 1 000 100 10 1 000 70 14.29
E 1 000 100 10 800 50 16.00
Looking at the results of each company, it can be seen that:
Company A: labour productivity increases since output increases while labour remains constant. Unfortunately, such
productivity increase is not accompanied by job creation. It is possible that the labour productivity increase is driven
by utilising more capital.
Company B: labour productivity increases since output increases while labour declines. This can be defined as
jobless growth – an increase in output associated with a decrease in employment.
Company C: labour productivity increases since output increase is more rapid than labour increase. This would be the
most ideal scenario, as it is associated with both economic boom and job creation.
Company D: labour productivity increases since output remains constant while labour declines. Although this outcome
implies an increase in unemployment, it also suggests that the production method adopted in was inefficient as
the company may have hired more workers than actually needed to produce the same output level.
Company E: labour productivity increases since output declines at a slower rate than labour. This is the most
unwanted type of labour productivity increase, as it is associated with both economic recession and job loss.
Greater workplace and workforce flexibility might also increase productivity, but might also result in
less favourable working or employment conditions, and might not always be sustainable without the
support of the union. If productivity is improved by discouraging labour-intensive firms and shedding
low-wage and low-productivity jobs, it will be accompanied by significant readjustment, which might
lead to instability. Labour policies will have to be geared to facilitate the creation of replacement jobs
to avoid high unemployment.
An efficient employee is one who keeps on his toes but never steps on the other fellow’s.
A
Productivity improvements can thus have different outcomes depending how they are achieved. It is
important to take these broader implications into account when considering the various factors that
influence productivity.
Quality of the labour force

The quality of the labour force probably has the biggest impact on productivity; thus it is listed as the
first factor that has an impact on productivity. As indicated in Chapter , the skills level of South
Africa’s workforce leaves much to be desired. There is a high rate of illiteracy and employers invest
less in training than do those in many other countries. In addition, there appears to be a shortage of
skills in most phases of the business cycle. Obviously, these factors have a negative influence on
South Africa’s productivity performance.
Furthermore, if the Employment Equity Act and especially the BBBEE Codes exacerbate the shortage
of skills by compelling enterprises to become equitably representative of the population while the
supply of skilled black people is insufficient, the impact on productivity will be negative (see section
. . ). However, the Employment Equity Act might have a positive effect if it leads to the better
utilisation of women and black people in a manner that retains flexibility and takes into account the
circumstances of individual enterprises and sectors.
Labour market flexibility

Numerous aspects of labour market flexibility can have a critical impact on productivity. Among these
are flexible working practices and working time, multitasking and extended job descriptions, transfers
within and between workplaces, and appropriate job security.
However, because this is a very topical issue and also quite complex, it will be discussed in section
. , with reference to the various types of flexibility, and whether the South African labour market can
be described as “flexible” or not.
Management responsibilities
The Labour Market Commission ( : ) has indicated that a lack of management skills in
leadership, mentoring, work reorganisation and decision-making abilities is a major constraint on
productivity improvement. Many of the factors that influence productivity can be found in the
organisation itself and can largely be influenced by management. The remaining factors consist of
external influences. It is generally accepted that South Africa’s lack of management skills and absence
of productivity awareness are important contributing factors to poor productivity performance.
The lack of competition in the South African market and insufficient investment in research and
development are among the many factors contributing to poor productivity that can regarded as a
management shortfall.
An important part of the responsibility of management is work restructuring and reorganisation to

create an environment conducive to productivity (ILO Review, : ).
Sociopolitical circumstances
Social circumstances in South Africa are an important factor contributing to low productivity growth.
In particular, the lack of proper housing, long travelling distances, unsatisfactory health care and
nutrition, and various other socioeconomic factors play an important role. HIV and AIDS is also likely
to have a negative effect on productivity, as outlined in section . . .
Productivity incentive schemes

There are various schemes whereby remuneration and productivity can be linked or productivity
promoted. However, such schemes should not be seen as a panacea for solving productivity problems
and should comply with a number of conditions. There should, for instance, be a direct link between
personal or team performance and remuneration. Workers should have control or influence over the
factors that drive both performance (output) and payment. The system should not be used as a
method for reducing the fixed or guaranteed portion of pay and should not be implemented in an
autocratic, top-down way. Furthermore, there should be joint development of the common purpose
of such a scheme, the method by which productivity is to be measured and the targets. The mining
industry attempted such initiatives in , and various lessons can be learnt from this experience (see
for instance, Barker, b).
By working faithfully eight hours every day, you may eventually get to be a boss and work twelve
hours a day.
R F
Management–union cooperation
The labour market is often characterised by a confrontational relationship between employers and
workers. In many cases there is mutual distrust, and this makes productivity bargaining much more
difficult. However, employers in the past have also not given enough attention to sharing the benefits
of productivity improvements with employees, or addressing any possible negative effects of
productivity improvements on employees. This has resulted in opposition from unions to productivity
bargaining. Employers are also often unwilling to disclose financial and other information during
collective bargaining negotiations, which makes a joint approach to productivity improvement and
the sharing of the benefits of such improvement difficult.
There is an English proverb that says “there are no bad students, only bad teachers”. I believe it also
applies to a company. There are no bad employees, only bad managers.
TS L
The Labour Market Commission ( : ) has identified the following elements that have a negative
impact on productivity: authoritarian and discriminatory supervision of the shop floor; an adversarial
and hostile relationship between management and unions at shop-floor level; a general absence of
workplace democracy and inadequate worker-participation structures; poor identification by the
workforce with productivity issues: and volatile industrial relations.
In the new LRA, the government has attempted to improve employer and employee cooperation by
making provision for workplace forums (see Appendix B). These forums do not seem to have
garnered much support, in contrast to the position on workplace cooperation in Germany.
An important question is whether unions on their own have a positive or negative influence in so far
as productivity is concerned. Section . . should be consulted in this regard.
Productivity SA
Productivity SA, previously known as the National Productivity Institute (NPI), plays an important role
in increasing productivity in South Africa (website: https://www.productivitysa.co.za). Among other
things, Productivity SA is responsible for the promotion of productivity awareness, and for education
and training. It provides consulting services to both the private and public sectors.
SUMMARY
There are various factors that influence productivity, among others the quality of the labour force, management
responsibilities, sociopolitical circumstances, productivity-linked pay, labour market flexibility and management–union
cooperation.
. PRODUCTIVITY IN SOUTH AFRICA
After increasing only marginally in the s, labour productivity increased rapidly from the beginning
of the s, and has generally increased more rapidly than capital and multifactor productivity, as
shown in Figure . . In , Productivity SA estimated the labour, capital and multifactor productivity
average annual growth rates by sector in – , and Table . shows that six sectors experienced
an increase in labour productivity (the increase was the greatest in the construction and finance
sectors), five sectors enjoyed an increase in capital productivity (the increase was the greatest in the
finance sector), while multifactor productivity increased in seven sectors (the greatest increase took
place in agriculture). The table also shows that all three productivity measures declined in –
for the mining and utilities sectors.
FIGURE . Productivity in South Africa, –
Source: National Productivity Institute ( )
TABLE . Average annual growth rates (%) of labour, capital and multifactor productivity by sector,
–
Sector Labour productivity Capital productivity Multifactor productivity

Agriculture –6.4 0.1 6.7
Mining –1.8 –4.8 –3.6
Manufacturing 4.3 1.3 3.0
Utilities –0.7 –6.1 –4.2
Construction 4.6 –1.0 1.7
Trade 2.4 0.8 1.4
Transport 2.1 –0.5 0.4
Finance 4.6 2.7 3.3
Government services 1.0 0.3 0.6
Table . shows that the increase in labour productivity averaged . % and . % per annum for the
periods – and – respectively, compared to an increase of . % per annum during
the period – , and even lower before that. However, when one considers the decline in
employment in the early to mid- s (Hodge, : ), it becomes apparent that the improvement
in labour productivity was probably partly achieved through the shedding of labour. The economy
continued to experience a sharp increase in capital intensity during this period, which enabled
employers to reduce their workforce and maintain production. This changed only after , when
both labour productivity and employment increases were achieved at the same time.
Table . also shows that capital productivity increased at an average rate of . % per annum in
– compared to a decline in the six years before that. This probably indicates that producers
started to utilise capital more effectively in the economy, without blindly replacing labour with capital
irrespective of productivity considerations. However, the increase in capital productivity still lags
behind that of labour productivity, which suggests that the replacement of labour by capital has not
ceased, as indicated in section . . (see also Edwards & Golub, : ). Capital productivity
decreased by . % in – ; interestingly, real earnings per employee decreased by . %, while
there was zero growth in multifactor productivity during the same period.
TABLE . Productivity changes in the private economy
Average % change per annum

1981–1984 1984–1989 1989–1995 1996–2004 2012–2013
Labour productivity 1.45 1.14 2.56 3.76 1.90
Capital productivity –2.88 0.40 –0.28 2.46 –1.40
Multifactor productivity –0.45 0.88 1.19 3.17 0.00
Real earnings per employee 1.47 0.13 2.22 3.20 –2.90
Source: National Productivity Institute & Productivity SA
Numerous studies (see the summary of their findings in Table . ) calculated to what extent
multifactor (total factor) productivity as opposed to capital and labour inputs contributed to growth in
real GDP from onwards. Overall, there was an encouraging increase in total factor productivity in
the s, after very low increases up to . In fact, the higher total factor productivity after
was driven by trade liberalisation, openness to capital flows, greater private sector participation in
total investment and lower interest rates (Arora & Bhundia, ; Arora, ; Du Plessis & Smit,
). Nonetheless, total factor productivity growth became less rapid and was even negative at times
after the global financial crisis in .
TABLE . Sources of output growth in South Africa, –
Study Period Average in percentage points (%)

Real GDP growth Labour contribution Capital contribution Total factor productivity/technology/productivity
Fedderke (2002) 1970–1979 3.21 1.17 2.54 –0.49
1980–1989 2.20 0.62 1.24 0.34
1990–1999 0.94 –0.58 0.44 1.07
Arora & Bhundia (2003) 1980–1993 1.0 0.1 0.9 0.0
Study Period Average in percentage points (%)
Real GDP growth Labour contribution Capital contribution Total factor productivity/technology/productivity
1994–2001 2.8 –0.9 0.6 3.1
Arora (2005) 1980–1994 1.2 0.7 0.8 –0.4
1995–2003 2.9 0.9 0.7 1.3
Du Plessis & Smit (2007) 1985–1994 0.80 1.11 0.45 –0.76
1995–2004 3.10 0.62 0.62 1.86
Anvari et al. (2014) 1990–1999 1.4 0.4 0.5 0.4
2000–2008 4.2 0.9 2.0 1.3
2009–2013 1.9 –0.1 1.8 0.2
Kemp & Smit (2015) 2006–2007 4.3 1.2 1.6 1.4
2008–2010 2.9 1.4 1.7 –0.2
2011–2014 2.2 0.8 1.5 0.0
Source: Author’s own compilations of the results from the above studies
Fedderke ( ) found that the s and s saw growth that was mostly determined by the
growth in capital inputs and to a lesser extent labour inputs, with very little contribution by total
factor productivity. In the s the growth in labour inputs contributed negatively (with formal sector
employment declining) and the contribution of capital inputs also declined. Instead, the greatest
contributor to GDP growth during this period was the contribution by total factor productivity.
Efficiency gains thus became the main driving force of the economy during the s and s,
whereas increases in factor inputs (especially capital) were the main drivers in earlier years.
Productivity cannot be considered in isolation from wage increases. This relationship is addressed by
using a concept known as unit labour cost. Unit labour cost developments in various sectors are
discussed in the next section. A comparison with other countries is important from South Africa’s
competitiveness position. This is discussed in Appendix A.
SUMMARY
Labour productivity in South Africa has improved quite significantly in recent years. This may be related to the substitution
of capital for labour and probably accounts for developments prior to , as indicated by low or negative increases in
capital productivity. However, in – , both capital and labour productivity increased.
. CONSIDERING PRODUCTIVITY AND WAGES TOGETHER: UNIT LABOUR COST
. . The concept of unit labour cost

The above discussion of the importance of productivity improvement gives three reasons why
productivity and wages should be considered together:
Productivity is the basic source of improvements in real wages. For the economy as a whole and in
the long term, real income per worker can increase only at the same rate as real output per worker.
Productivity is therefore the real link between output and reward in production.
Productivity improvements, rather than low wage levels, should provide countries with an
international competitive advantage.
Productivity growth is an anti-inflationary force in that it tends to offset or absorb increases in
money wages.
Therefore, changes in labour costs must be considered in relation to the changes in productivity that
may have taken place at the same time.
The extent to which changes in productivity and wages have moved together is determined by unit
labour cost, as previously mentioned.
Unit labour cost takes account of changes in remuneration costs as well as labour productivity. Unit
labour cost is the cost of labour to produce one unit of output. It is also referred to as remuneration cost
per unit produced. This can be explained as follows (Mohr et al., : ):
Labour productivity = Q/E, where
Q = output
E = number of workers
Unit labour cost = WE/Q, where
W = wage rate or earnings per worker

E = number of workers
WE = cost of labour
Q = output
Unit labour cost (ULC) can in turn be expressed as follows in equation terms:
WE W earnings per worker

ULC = = =
Q (Q/E) labour productivity
The change in unit labour cost is approximately equal to the growth rate of earnings per worker
minus the growth rate of output per worker. That is,
Δ(W E) Q
Δ( )
Q E
ΔW
ΔULC = ≈ –
WE W Q
( ) ( )
Q E
The above equation highlights the important point that unit labour cost will remain unchanged if the
percentage increase in money or nominal wages is matched by a similar percentage increase in labour
productivity. If the wage increase is greater than the productivity increase, the unit labour cost will
increase; similarly, if the wage increase is smaller than the productivity increase, the unit labour cost
will decrease. The latter point suggests that rapid increases in nominal wages are not harmful,
provided they are accompanied by proportional – and hopefully greater – increases in labour
productivity (for example, nominal wages increase by % but labour productivity increases by %).
By the same token, low labour productivity growth need not be a problem from an inflation or
competitiveness perspective, as long as nominal wages are kept in check.
Figure . shows the nominal and real unit labour cost indices in – . The former index
increased continuously during the period (the increase became more rapid from ), while the
latter index initially stagnated in – , before a clearer upward trend took place. Unit labour
cost in South Africa is compared to that of other countries in Appendix A.
FIGURE . Nominal and real unit labour cost, –

(index = in )
Source: Author’s own calculations using South African Reserve Bank data
on nominal unit labour cost (code: ) and Statistics South Africa’s
consumer price index
Workers cannot be expected to bear the brunt of inflationary pressures alone – wage increases should
in some way make up for inflation, especially if inflation originates from external sources such as
sharp declines in the exchange rate. Figure . consequently compares labour productivity and real
wage increases, i.e. after allowing for inflation. It is clear from this graph that the real wage increase
was higher than labour productivity increases in – . Klein ( : ) argues that the
misalignment between labour productivity and real wages “may reflect the outcomes of the collective
bargaining framework in South Africa, which not only contributes to the weak link between pay and
productivity, but also reduces the responsiveness of real wage to business cycle fluctuations”.
FIGURE . Wages and productivity, –
Source: IMF, : (Real earnings and productivity indices = in )
. . Unit labour cost in various sectors

Figure . indicates the increase in unit labour cost in various sectors for the period – . All
sectors experienced an increase in unit labour cost, with the construction and utilities sectors being
the highest. This was as a result of labour productivity not keeping up with relatively high wage
increases.
FIGURE . Average annual growth rates (%) of unit labour
cost by sector, –
There is a common misconception that blue-collar workers alone are responsible for the increase in
unit labour costs. However, bluecollar workers earn on average only a third of what their white-collar
colleagues earn. Depending on the relative numbers of workers involved and their productivity, blue-
collar workers will therefore not necessarily have the greatest influence on labour costs. This is
probably why the Labour Market Commission ( : ) concluded that depressing the wages of
unskilled workers may have a relatively small impact on average unit labour costs. At the same time,
the commission also argued that, because of the high burden of adjustment international competition
places on South African enterprises with a high unskilled labour component, the wages of unskilled
workers in certain segments of the South African economy may well be comparatively high relative to
productivity.
. . Impact of wage increases on productivity: efficiency wages

Thus far wage increases have simply been compared to productivity increases, with the implicit
assumption that there should be some relationship between high productivity increases and high
wage increases – if workers can increase their productivity, they should benefit through higher
remuneration. However, it is possible that this can work the other way – under certain circumstances a
wage increase can lead to an equivalent productivity increase. This is called the efficiency wage
hypothesis.
Most discussions on wages and productivity explicitly or implicitly assume that wage increases should
be linked to productivity increases. However, higher wage rates (e.g. as a result of collective
bargaining) can conceivably also lead to higher productivity (Ehrenberg & Smith, : ;
McConnell et al., : ). A direct comparison between wage and productivity data is even more
difficult – how do you separate cause and effect? Productivity increases are not always the origin of
improvements in remuneration levels – high wage increases could, in fact, result in job-replacing
mechanisation, which often means higher productivity. So productivity, wage increases and job losses
are intertwined in various ways:
Higher wages might enable (or induce) the employer to recruit employees more carefully, which
means better job matching, and the employer might also be induced to employ better-quality
workers.
A higher wage might make employers more willing to invest more in their workers, for example
through training and development, in order to increase their productivity and make the wage
increases affordable.
Workers’ morale might improve due to their higher wages, and their rate of absenteeism might be
lower because their opportunity cost of not working (i.e. the wages they lose when not working) is
higher.
Workers’ nutritional and health levels might improve, positively affecting their physical vigour and
mental alertness, and therefore increasing their productivity (McConnell et al., : ).
Workers might, at least in the short term, be motivated by the higher wages to improve their
productive effort. They might even do so over the long term if they fear being dismissed from a
well-paying job. Labour turnover, i.e. the rate at which workers quit their jobs, might also reduce,
which would be to the advantage of the employer.
In a less positive vein, labour productivity might also increase as a result of wage increases if the
employer attempts to reduce the wage bill by becoming more capital-intensive, i.e. replacing
workers with machinery. Labour productivity will then increase at the cost of employment. The
employer might also pass on higher wage costs to the consumer. The extent to which this might
happen will depend in part on the monopolistic power of the employer (i.e. on the extent of
competition prevalent in the marketplace) and on the sensitivity of consumer demand to price
increases (the elasticity of demand – see section . . ).
There is therefore a possibility that labour productivity will increase as a result of a wage increase.
However, what is important is not whether productivity increases, but whether it increases by more
than the wage increase, i.e. whether the productivity improvement fully compensates for the labour
cost increase. If this does indeed happen, it might be more efficient for the employer to pay higher
wages, as this might reduce the unit labour cost.
SUMMARY
Unit labour cost measures the joint impact of changes in remuneration costs and labour productivity on the cost of
output. It is defined as the cost of labour to produce one unit of output.
Unit labour cost in South Africa has been increasing quite rapidly, which indicates that nominal wages have increased
without concomitant increases in productivity. In addition, the increase in real wages in recent years has been more
rapid than the labour productivity increase.
Wage increases can conceivably lead to higher productivity due to factors such as improved recruitment, training,
higher morale and better nutrition. This is called the efficiency wage hypothesis. However, the increase in productivity
will not necessarily be sufficient to compensate fully for the higher labour costs.
. THE IMPORTANCE AND TYPES OF LABOUR MARKET FLEXIBILITY
Low productivity and unemployment could be caused by labour markets not adjusting as quickly as
they should to changing circumstances. If markets could adjust to be more in line with demand and
supply conditions, unemployment might, according to supporters of flexibility theory, be reduced
(Organization for Economic Cooperation and Development (OECD), ). For an excellent discussion
of the flexibility debate, see Gladstone, Wheeler, Rojot, Eyraud and Ben-Israel ( ).
Labour market flexibility refers to the extent to which an enterprise can alter various aspects of its work
and workforce to meet the demands of the business, for example, the size of the workforce, the content
of jobs, working time, etc.
The argument is that labour market rigidities prevent enterprises from being as productive as they
might have been because enterprises are unable to adjust to technological transformation, changing
economic circumstances, external shocks (e.g. an oil price rise), and more intense international
competition. The labour market therefore does not operate as efficiently as it should and productivity
outcomes are poor. Employers are also said to be reluctant to employ labour on a permanent basis
because of its high cost. They therefore switch to capitalintensive technology, or lose market share.
The end result is reduced employment. It has been shown that in OECD countries with flexible labour
laws, the employment rate is . to . % higher than in other OECD countries (see sources quoted by
World Bank, : ).
Flexibility and labour rights are very closely interrelated. In South Africa, the protection of the rights of
labour has been an important element of the democratic order since , and some labour rights are
enshrined in the Constitution of South Africa, for instance the rights to fair labour practices, freedom
of association and collective bargaining (section of the Constitution of the Republic of South
Africa). Moreover, the protection of labour rights is in line with South Africa’s obligations as part of
the international community. As such, these rights are sacrosanct and any debate about the
appropriateness or otherwise of such rights is superfluous.
The debate about labour market flexibility should not take place as binary policy alternatives, i.e. a
stark choice between flexibility and regulation. This is conceptually and empirically wrong (Bhorat &
Cheadle, : ). The issue is much more complex and this is not always recognised by the
protagonists of either of the two extreme positions. Deakin and Wilkinson (as quoted by Benjamin &
Theron, : ) in this context refer to the sterile opposition of rights and efficiency. The alternative
to this sterile opposition, these authors suggest, is for economic insights to shift policy away from
deregulation towards the design of labour legislation with economic policy goals in mind.
Numerous types of flexibility have been identified by the ILO (Ozaki, : ; Rodgers, : ; Venter
& Levy, : – ), and the following four types are discussed here:
1. Contracts of employment flexibility. This refers first of all to the employer’s ability to terminate
contracts of employment, and whether the employer is restricted in terms of his or her hiring
decisions. Secondly, it refers to the ability of the employer to have recourse to the so-called
“atypical employment contracts”, for example part-time work, temporary work, seasonal
employment and fixed-term contracts.
2. Wage flexibility. This relates to the ability of the firm to vary workers’ wages in response to the
supply and demand conditions of labour. A variety of regulations and institutions may limit wage
variation, including the imposition of a minimum wage, trade union activity, the extent to which
wages can be linked to performance, and bargaining structures (i.e. whether bargaining takes
place at a centralised or decentralised level).
3. Working time flexibility. This represents the power of the firm to set and change working hours
and patterns, and relates to normal or overtime working time, or working weekends or at night.
4. Work organisation flexibility. This is determined by the extent to which new forms of work
organisation are introduced, for instance through the multiskilling and multitasking of workers,
and the extent to which teamwork is encouraged.
Some of the most important elements of flexibility in the workplace are now discussed in the context
of these different types of flexibility.
SUMMARY
Labour market flexibility refers to the extent to which different elements of the labour market can adjust to changing
circumstances and is said to be an important element of the battle against unemployment.
. . Flexibility in contracts of employment: job security and hiring decisions

The ability of an enterprise to adjust the size of its workforce to changes in the demand for the
product (i.e. the ability to hire or fire) and to use various types of employment contract, such as short-
term contracts and subcontracting, is also referred to as external or numerical flexibility.
Job security
The increase in job security in South Africa since the advent of democracy, through the measures in
the LRA preventing unfair dismissals and discouraging retrenchments, can be described as an
important social advance.
However, job security is far from cost free (Le Roux, : – ). Not only do legal and other
processes incur high costs, but job security regulations are blamed for the reluctance of employers to
engage new workers on indefinite-period contracts. In addition, employers may substitute labour with
capital or utilise subcontractors, casual and temporary workers, or they may employ workers on fixed-
term contracts only. In the end, unemployment may simply be increased.
Britain has invented a new missile. It’s called the civil servant – it doesn’t work and it can’t be fired.
G S W W
Job security for those workers already in employment may thus be to the detriment of labour
intensity and therefore to the disadvantage of workers trying to enter the labour market for the first
time (see sources quoted by Le Roux, ).
Apart from anecdotal evidence, there is little clear-cut information to prove either the negative or the
positive effect of high job security. The ILO (Ozaki, : ) also states that even where legislation was
relaxed, not all employers seemed to be taking advantage of the new possibilities of flexibility in
employment. This is because employers often realise the advantages of job security for themselves.
This conclusion is confirmed by other ILO research, which considered a number of factors that may
have an impact on the employment intensity of economic growth. The conclusion is the following:
“There was, however, one noteworthy surprise in the study: stricter employment protection was not
found to be significantly related to the employment intensity of growth” (ILO, b).
Many countries have nevertheless relaxed job security regulations, for example by extending the
authorised probationary period, moving away from government approval for retrenchments and
making temporary employment contracts possible (Gladstone et al., : ). Other possibilities are
broadening the scope of exclusions from unfair dismissal legislation and easing or abolishing
administrative controls such as trade union consultation requirements (Ozaki, : ), as well as
allowing new businesses to issue four-year-term contracts for employees, and excluding small
businesses (with fewer than employees) from onerous rules on dismissals (World Bank, ).
In , Australia, for instance, excluded employees working for small and medium-sized
undertakings (with fewer than employees) from unfair dismissal protection to encourage such
businesses to “take a chance about employing new people”. However, termination for a discriminatory
reason remained prohibited. For larger businesses, i.e. with more than employees, an employee
would have had to have been employed for more than six months before he or she can lodge an
unfair dismissal application (see Appendix A and http://www.australianbusiness.com.au).
Even in countries that have had high job security in the past, this was often balanced by other factors.
Japan, for example, had a system of lifelong employment before the recent economic woes resulted in
decreased job security. This was balanced by a high degree of internal flexibility – workers were well
trained and, in addition, were rotated within the enterprise as a rule every three to five years. Such
rotation was not always upwards or even sideways, and could result in less favourable conditions of
employment. This led to the multiskilling of workers, and ensured that they were adaptable to
changing circumstances (Hanami, ).
There is fairly extensive protection of employees against unfair dismissals in South Africa in terms of
the LRA, with regard to both procedure and reasons for dismissal. However, Van Niekerk ( )
suggests that the requirements of procedural fairness have contributed more than any other factor to
perceptions that the South African labour law is inflexible. The Tokiso Report on Dispute Resolution
(Tokiso Dispute Settlement, ) found that % of those cases in which findings are made against
employers incorporate an element of procedural unfairness. The study concludes that the most likely
interpretation of these figures is an intense focus on the minutiae of procedure that currently besets
allegations of unfair dismissal. Bhorat ( : ) considers a number of research reports and also
comes to the conclusion that there is an overly strict interpretation of the requirement for procedural
fairness by employers, labour consultants and lawyers, the CCMA and the Labour Court. This was
never the intention of the legislation.
Apart from fairly extensive procedures with regard to individual dismissals, there are also various
procedures relating to retrenchments. Such procedures in effect introduce a moratorium of days
on retrenchments. On the one hand, time limits can, in principle, serve a useful purpose by promoting
certainty and the finalisation of disputes, and can also ensure that employers are absolutely sure
about the need for retrenchment before they dispatch workers into the practically inevitable abyss of
unemployment. The moratorium also assists parties in the consultation process. On the other hand,
such moratoriums do not help to bring about more certainty and finality with regard to the
retrenchment process, and any eventual decisions taken by the employer can be overturned by
judicial processes.
There are also other mechanisms to protect the job security of employees, for instance a right to
strike about retrenchment. This does not apply to employees in enterprises with fewer than
employees. Excluding small business is a step in the right direction, but the cut-off point is probably
too low. In addition, the size of a retrenchment that can trigger the right to strike (i.e. a retrenchment
of between and % of the workforce in any one year) is so small that, practically speaking, there
will often be very little distinction between a small-scale retrenchment and a large-scale
retrenchment. The rationale for covering only large-scale retrenchments by such provisions is that
these retrenchments can have severe and substantial socioeconomic consequences. However, this
motivation does not apply when retrenchments are relatively small.
It is crucial for enterprises to be able to adjust quickly, though fairly, to changing market demands
and thus remain competitive, both locally and globally. An inability to do this will have a negative
impact on an enterprise’s competitiveness and will put other jobs at risk. Furthermore, if enterprises
can restructure only with difficulty and after relatively long time delays, it will serve as a further
disincentive to job creation. At the same time, excessive freedom to hire and fire will not be in the
interests of either the employee or the employer.
Bhorat and Cheadle ( : ) found, within the global distribution of labour legislation, that there
was an increase in hiring and firing rigidity between and , but a decline in the cost of hiring
and firing. (The cost of hiring and firing includes all the social security and health costs associated with
hiring a worker, whereas the firing costs include costs such as severance pay and those associated
with the notice period. The cost and rigidity associated with the employment equity legislation are not
included.) Their conclusion is that in the case of legislative provision for firing workers, South Africa
possesses a “particularly high level of regulation” ( : ). The South African labour market, in an
international context, appears to have become more regulated over the last decade as far as hiring
and firing rigidity is concerned. A more recent study by Bhorat et al. ( : ) also found that South
Africa’s hiring and firing rigidity in was higher, compared with other upper-middle-income
countries.
Bhorat et al. ( ) examined whether an increase in employment protection has reduced

employment by acting as a tax on firms to constrain hiring and firing decisions. Using the database of
the CCMA, the authors investigated the relationship between the number of industrial disputes and
the level of employment protection in the labour, and found that regions characterised by a lower
number of disputes but greater efficiency of the regional offices to handle the disputes (i.e. offices
associated with shorter conciliation and arbitration mean time) are associated with higher
employment levels.
Van Niekerk ( : – ) also makes the point that many of the flexibilities that are inherent in the
relevant international standards were never considered for incorporation in the appropriate legislation
in South Africa, and that this is particularly so in relation to work security rights. Even though the
model underpinning South African labour legislation is sound, he refers to the following potential
flexibilities that should be considered:
The possibility should be considered of limiting the application of laws by referencing the number
of employees engaged in the enterprise, particularly the exclusion from unfair dismissal laws of
small enterprises, as contemplated by Convention of the ILO.
Workers serving a qualifying period of employment should be excluded from certain forms of
employment protection.
International labour standards do not require a formal hearing prior to dismissal – this relates to the
fact that the law does not prevent an overly strict interpretation by the CCMA of the dismissal
requirements in the legislation (as referred to above).
International practice recognises the desirability and legitimacy of excluding small businesses (e.g.
those employing fewer than employees) from the ambit of consultation and information
obligations relating to retrenchments.
The unrestricted access to the CCMA should be addressed by making cost orders the norm rather
than the exception, and there should be a requirement to pay a filing fee.
SUMMARY
Job security is an important social advance, but is far from cost free, as it often incurs costs and may make employers
reluctant to engage new workers.
The rules on hiring and firing in South Africa are said to be too procedurally complicated and overly prescriptive, which
was never the intention of the legislation.
Restrictions on hiring
If an employer’s ability to recruit and appoint the best person for the job is inhibited, it will have a
negative impact not only on flexibility, but also on the search for excellence. There are several
examples of restrictions on hiring in South African legislation, for example the provision for closed-
shop agreements, in terms of which all employees must join a union or face dismissal.
Another restriction is the duty on employers to implement employment equity (EE) measures to give
preference in hiring and promotion to so-called “disadvantaged groups” in terms of the Employment
Equity Act (see Summary box in section . . ). On the other hand, the act might, over the longer
term, lead to the better utilisation of human resources, which will increase productivity. This positive
effect is likely in the light of the many years of discrimination and underutilisation of an important
part of the country’s human resources.
Another important limitation on the employers’ right to hire is the introduction of indirect “racial
quotas” by the BBBEE Codes. This is discussed in more detail in section . . .
The provisions for probationary employment in the LRA might reduce some of the disadvantages of
substantial job security and might thereby increase flexibility, provided this right is clear and is not
undermined by rash decisions by CCMA commissioners. This will allow the employer to assess the
competence and suitability of the employee for a few months before that employee obtains job
security rights.
. . Flexibility in contracts of employment: atypical employment

Employers increasingly appear to be less inclined to employ workers on a full-time, permanent basis.
There thus appear to be two categories of workers. The regular, full-time employees who are part of
the corporate family and who receive fringe benefits are referred to as the core workers. They have a
long-term relationship with their employer. Atypical, non-standard or contingent employment is a
more flexible type of employment, where the workers are not part of the corporate family and do not
have a longterm relationship with an employer. This can include, for instance, casual or part-time
workers, or some type of contracting arrangement (see accompanying box).
What is atypical employment?
Temporary or casual employment (employment for a short period of time)

Part-time employment (in which daily hours of work are less than the norm)
Outsourcing (where an outside enterprise runs a non-core activity of the enterprise, e.g. catering, cleaning and
security, but this increasingly also includes core activities)
Dependent contractor (a person who is nominally independent but economically dependent on the “employer”)
Specialised subcontracting (the performance of specialised functions by the subcontractors)
Labour broking or labour-only subcontracting (where the labour broker recruits and supplies labour or remains the
employer)
Homeworkers or outworkers (working from home for an employer)
The development of atypical employment is normally through one or more of the following processes
(Department of Labour, : par . . ):
(i) Casualisation. This refers to the process whereby standard employment is displaced by
employment that is temporary or parttime (or both). However, this mostly takes place at the
workplace of the employer and employment continues to be regulated by a contract of
employment.
(ii) Externalisation. This refers to a process of economic restructuring, in terms of which employment
regulated by a contract of employment is displaced by employment that is regulated by a
commercial contract. This can include the employment of an independent contractor for a
specific task and/or where workers are employed by an intermediary to work for someone else
(so-called “labour broking”).
(iii) Informalisation. This is the process whereby employment is increasingly unregulated, in part or
altogether. This is closely related to the other two processes.
In most parts of the world there has been a trend away from core workers or the standard
employment relationship towards atypical workers (Belous, : ). It has been estimated that
between a third and a half of all jobs created in the US, for instance, in the period – , could be
described as atypical employment. Furthermore, this sector is growing considerably faster than the
entire labour force, and at least a quarter of the American labour force is probably now made up of
atypical workers (Belous, : ).
The labour market in the OECD countries has become more segmented in the sense that employers
explicitly differentiate between employees who are career orientated and those with little career
orientation and training (Strydom, : ).
Australia has one of the highest levels of non-standard employment of developed countries. It is
estimated that to % of all newly created jobs in that country can be classified as precarious
(Department of Labour, : par . ). Similarly, part-time workers are used extensively in the
Netherlands, to such an extent that it has become known as a part-time economy. This resulted from
policy changes to enhance flexibility.
Even though the nature of new jobs might be more non-standard, it seems as if the bulk of jobs are
still arranged in terms of the standard employment contract. Studies of the advanced industrial
economies of the world revealed that in no less than nine-tenths of employees were still
employed on open-ended contracts, and another study even found that there has recently been a rise
of % in the standard employment relationship (quoted by Maree, ).
South Africa
In South Africa, the proportion of formal-sector employees hired on a permanent basis dropped from
% in to % in , as shown in Figure . . Interestingly, this proportion was much higher in
the case for unionised workers – above % during the whole period.
FIGURE . Percentage of formal-sector employees hired
permanently, –
Source: Author’s own calculations using the October Household

Survey, and Labour Force Survey September, and ,
and Quarterly Labour Force Survey quarter data
Various surveys in South Africa have shown that the resort to temporary or casual labour has been
quite widespread, and a total of . % of enterprises in an ILO survey had employed temporary or
casual labour (ILO Review, : ). This increased to % in (Department of Labour, : par
. ). Outsourcing and subcontracting is the second most prevalent form of atypical employment ( %
of the firms in ). Part-time employment was used much less (only % of the firms), as was
flexitime (only % of the firms) (Department of Labour : par . ). The ratio of subcontracted
labour to full-time employment increased from % in to % in .
The study by Bhorat et al. specifically looked at employment contribution by the temporary
employment services (TES) or the labour broker sector; it was found that employment in this sector
more than doubled from . million in to . million in . The majority of them worked as
protective services workers, such as security guards, bodyguards and cleaners.
Increasing use of atypical employment in certain sectors

Construction. Labour-only subcontracting is used extensively in this sector, where firms provide unskilled or semi-
skilled workers on construction sites, usually to perform a specific task. The contractor is normally not required to
supply any of the materials or even equipment. According to the LFS, % of workers in this sector are in atypical
employment (Department of Labour, : par . ).
Retail. The majority of employees in this sector can be classified as non-standard (Department of Labour, : par
. ). This is related to the efforts of retailers to staff their stores during peak periods, holidays and weekends, and for
late-night shopping.
Reasons for the use of atypical employment

Atypical employment has several advantages:
Young, inexperienced people make use of atypical employment to get into the labour market for
the first time. With high unemployment among young workers, and many of the unemployed
persons being without a job for lengthy periods of time, atypical employment is thus an important
element of any strategy to address youth unemployment.
It reduces labour costs, in particular because fringe benefits are usually not provided, smaller
enterprises providing the services may pay lower wages, and because contingents are paid only for
the actual work required by the enterprise. The authors of the ILO Review ( ) state that reduced
costs seem to be the most important reason for the increased use of such labour.
In the past, employers often resorted to labour-broking services to sidestep burdensome labour
regulations, but changes to the LRA and the BCEA have made this more difficult. Research
conducted for the Department of Labour suggests that labour legislation has been the impetus for
externalisation, both in that the avoidance of labour legislation provides a motive to externalise,
and that the legislative provisions regarding temporary employment services provide the
opportunity (Department of Labour, : par . ).
Atypical employment allows the employer more flexibility in the utilisation of labour, especially if
there are fluctuating levels in the demand for labour, and to avoid the hassle factor often associated
with employing workers, especially those who are highly unionised.
It can mean more freedom and flexibility for the workers. Parents with children at school may, for
instance, prefer to have some freedom to perform certain tasks with regard to their children
without having to ask the employer for time off.
Disadvantages of atypical employment

Even though the utilisation of atypical employees has a number of advantages, the increase in atypical
employment also has several disadvantages:
Research in South Africa suggests that the increasing use of non-standard employment is eroding
labour standards in the following manner (Department of Labour, : par ):
– Wages. The wages of temporary workers employed by a temporary employment service (TES)
have been found to be lower than those of other employees. However, where a TES is not
involved, this does not seem to be the case.
– Conditions of employment. In theory, all part-time and temporary workers are protected by most
provisions of the BCEA, but this does not always apply in practice.
– Health and safety. In mining, the mine is responsible for the health and safety of all workers on
the mine, but in other sectors, workers placed by a TES have to work under conditions
determined by the client.
– Training. There might be little incentive to train temporary workers. Training and human
resources development might thus be neglected, and equal opportunity goals might not to be
met.
– Social security. There is evidence that TESs do not register all the workers with bargaining council
social security funds in order to save costs. If they are able to, employers are less likely to have
such employees covered by retirement provision and health care because of their temporary
nature.
– Employment security. The temporary employee is in a more precarious position with regard to
unfair dismissals and the BCEA notice periods.
– Trade union rights and collective bargaining. Even though such employees enjoy the same rights
in theory, in practice it is very difficult for a union to organise workers whose workplace and
hours of work are frequently changed.
For the employer, atypical employment can also be dysfunctional. Heavily flexible workplaces lose
their ability to attract, bind and retain skilled workers. Long-term, secure employment relations are
reportedly seen “as a necessary condition to develop sustainable competitive advantages based on
the tacit and intangible knowledge embedded in the heads and hands of the workforce” (Dickens,
as quoted by Maree, : ).
Because of these contradicting advantages and disadvantages of atypical employment for different
stakeholders, policy makers face an important policy dilemma (see accompanying box).
Policy issue: what to do about the increasing use of atypical employees?
Research undertaken for the Department of Labour concludes that policy makers in South Africa probably face the same
kind of strategic choice that the European Union is facing. On the one hand, there is the defence of standard
employment at all costs, and on the other hand, the need to adapt to address the challenges of unemployment and
international competitiveness (Department of Labour, : par . ). This is related to the relative priority given to the
constitutional right to fair labour practices as opposed to the constitutional right to work.
The priority area for policy intervention should be labour broking because it has clearly played a major part in
accelerating a process of externalisation. Some of the policy proposals made by the researchers are the following:
All temporary employment services (TESs) should be registered.
TESs should not be used by clients to employ temporary workers on an indefinite basis.
There should be a regulatory body for TESs.
Employees should also be able to exercise rights against the employer controlling the workplace.
There should be a Code of Conduct on atypical employment.
COSATU is strongly opposed to one specific type of atypical employment, namely employment by labour brokers, i.e.
where workers work (mostly on temporary contracts) for one person, but are in the employment of another person, the
labour broker. COSATU argued that it was not opposed to temporary employment, but that the practice of labour broking
amounted to slavery, and the federation wanted labour brokers to be banned.
The amended Labour Relations Act (LRA) in introduced provisions dealing with TESs. In section A of the act,
there is a distinction between what can be regarded as an “acceptable” use of the TESs and the “unacceptable” use of
its services. In addition, “temporary service” is defined as follows (Le Roux, ): ( ) a TES employee assigned to a
client for less than three months; ( ) a TES employee assigned to a client to replace an employee who is temporarily
absent from work; ( ) a TES employee assigned to a client to perform work, which is determined to be a temporary
service by a collective agreement concluded in a bargaining council, a sectoral determination or a notice published in the
Government Gazette by the Minister of Labour. This provision clearly aims at discouraging the long-term use of TES
employees to avoid the costs of permanent employment.
SUMMARY
In most parts of the world, including South Africa, there has been a trend away from employing regular, full-time workers
towards employing so-called “atypical workers”. The imperative of reducing the cost of labour seems to be an important
reason for this development.
. . Wage flexibility
Wage flexibility means that wages should be responsive to economic and market conditions, whether
in the country as a whole, a sector of the economy, or an individual enterprise. The question of wage
flexibility is usually more relevant to the downward responsiveness of wages during difficult economic
times, as unions play an important role in ensuring that wages are responsive in an upward direction
under favourable economic circumstances or high inflation. Wages are therefore usually rigid
downwards and flexible upwards, i.e. responsive to price increases and unresponsive to
unemployment and recession.
Wage flexibility also relates to the extent to which wages are linked to the productive effort of the
individual worker or a group of workers, and to the productivity and profitability record of the sector
or the enterprise. This is referred to as performance-related pay. The ILO (Ozaki, : ) has found
that the ability to do this is much more limited in countries where union density is high. Employers
have also experienced certain disadvantages in linking pay to performance, for example the high
administrative costs involved, the difficulties in measuring performance and the high probability that
bonuses will become a permanent feature of pay, even when performance no longer warrants them.
Wage flexibility is important for job security because it has been found (see, for instance, Elvira, )
that where flexible wages constitute a large proportion of labour costs, employees are less likely to be
laid off.
The levels of collective bargaining, and the extent to which provision is made at such levels for
flexibility arrangements, directly influence wage flexibility. Especially where bargaining takes place at a
more centralised level, for example in bargaining councils, there has been a tendency in the past
towards the rigid application of agreements, subject only to inefficient and cumbersome exemption
mechanisms. As discussed in section . , the authors of the ILO Review ( : – ; – )
question the system of bargaining council agreements, including the application of such agreements
to parties who have not been part of the negotiations.
The ILO points out elsewhere (Ozaki, : ) that “there has been a clear trend over the past
decade towards decentralizing wage bargaining, although sectoral and central wage bargaining
continue to play an important role in many Western European countries”. However, whether or not
decentralisation actually leads to better economic performance is a subject of considerable
controversy. It seems as if there is a so-called “U-shaped relationship” between a country’s economic
performance and the level of collective bargaining – meaning that both highly centralised and
decentralised bargaining perform better than intermediate forms of bargaining (such as the
bargaining councils in South Africa). It also seems as if decentralised but highly coordinated
bargaining has a positive effect on economic performance. This issue is discussed in more detail in
section . .
Minimum wages seem to be one of the major causes of unemployment among those groups they set
out to protect, for example young people (ILO, : ). If firms can get more educated or
experienced workers at the minimum wage they are required to pay, they are unlikely to try out
inexperienced workers, especially if they find it difficult and costly to undo the employment
relationship with such inexperienced workers.
Whether minimum wages would have an overall negative impact on unemployment would depend to
a large extent on the system of minimum wage fixing, as discussed in section . . The ILO (Ozaki,
: ) states that several countries have taken measures to reduce the effect of minimum wages.
These range from abolishing minimum wages to suspending minimum wages or postponing periodic
adjustments. Certain categories of workers have also been excluded, for example, young people.
In South Africa minimum wages laid down by the former Wage Board have in the past been
particularly conservative and would have had little, if any, negative influence on labour market
flexibility (see also ILO Review, : ). However, minimum wages determined by bargaining
councils might, in some instances, introduce rigidities in the labour market and can therefore reduce
employment creation among lower-skilled and younger workers.
In a recent study by Von Fintel ( ), he argues that wage inflexibility, particularly in the case of
unionised workers, may result in shortrun job losses. However, he finds that wages are flexible in
response to local unemployment rates in the long run, upon analysing the labour data of district
councils. Hence, he concludes that wages appear to be flexible in the long run, as wage agreements
take local labour market conditions (that are unfavourable to higher wage growth) into account at a
later stage. In other words, structural factors (to be discussed in section . . ) explain more of the
unemployment puzzle in the long run.
SUMMARY
As a result of the influence of trade unions and minimum wage regulation, wages are usually rigid downwards and flexible
upwards. Bargaining councils and minimum wages increase wage inflexibility.
. . Working time flexibility

Working time flexibility concerns the adjustment of total working time within a given period and the
possibility of rearranging working time within a given period (Van Niekerk, a: ). Time flexibility
can take various forms, such as
overtime
flexitime (flexible starting and finishing times)
the averaging of hours over a certain period, for example over months
shift work
a shortened work week.
Some countries have introduced a system whereby overtime is paid only if the total number of hours
during a year exceeds a certain prescribed limit (Ozaki, : ), or the sum paid for overtime is fixed
below a certain limit. Enterprises are also allowed to provide time off in lieu of payment for overtime,
and in some countries the overtime premium has been reduced.
Many enterprises in South Africa and abroad have introduced flexitime arrangements. In this manner
the needs of both the employer and the employee can be met to a greater degree than in the past,
and employers can more fully utilise their employees’ productive capability.
We need to break a mindset that sees workers simply as hands and legs to be supervised, rather than
brains and emotions to be motivated.
T M
Many overseas enterprises also apply a system of averaging the hours of work (Gailen, : ).
Workers have to work a specified number of hours over a certain “reference period”, for example one
year (this is referred to as the “annualisation” of working time). How the number of hours is worked
through the year is subject to agreement between the employer and employees. This is especially
useful during seasonal peak activities, but is at present possible only to a limited extent in South
Africa in terms of the averaging provisions in legislation.
Whether the BCEA increases the flexibility of working arrangements in South Africa is a moot point.
On the one hand, it can be said that the BCEA does increase flexibility of working time by making
provision for the averaging of ordinary working hours over a four-month period, and for a compressed
working week (to work up to ordinary hours per day, over a five-day, -hour working week).
On the other hand, however, the BCEA also imposes various limits on hours of work, which negatively
affects flexibility and which might or might not increase productivity. The BCEA allows only limited
scope for collective agreements to deviate from the limitations in the Act.
Some countries have introduced legislative changes authorising collective bargaining partners to
agree upon innovative working time arrangements (Ozaki, : ), for instance Australia and New
Zealand.
Furthermore, with regard to the limitation on working time, most countries have a limitation on hours
of work in order to prevent the exploitation of employees through excessive working hours. However,
some important industrialised countries have more lenient restrictions in this regard. For instance,
Australia and the US do not have a limit on working hours, and the European Union has a generic
limit of hours per week, compared to hours in South Africa. Individual European Union
countries may have more stringent requirements, and collective agreements in many sectors in the EU
also provide for a -hour or even shorter working week (see “Working Time Developments ” on
the website http://www.eurofound.europa.eu/eiro).
SUMMARY
The Basic Conditions of Employment Act (BCEA) is a blend of flexibility and inflexibility as far as the flexibility of working
hours is concerned.
. . Work organisation flexibility

The extent to which workers or conditions of employment within an enterprise can adjust to changing
circumstances is also called internal or functional flexibility. Rules and agreements relating to the
workplace, conditions of employment and the skills level of the workforce can affect this. Other
measures that can inhibit internal flexibility have their origins in legislation (e.g. the prohibition in the
mining industry against working on certain days), bargaining council or collective agreements on work
rules, union rules and work practices. The authors of the ILO Review ( : ) also refer to work-
process flexibility, which means that workers can be shifted between tasks relatively easily and working
practices can similarly be changed easily and at low cost.
Bargaining councils established in terms of the LRA might also have a negative impact on work
organisation flexibility. There has been a tendency in the past to apply agreements rigidly, subject
only to inefficient and cumbersome exemption mechanisms. As discussed in section . , the authors
of the ILO Review ( : – , – ) question the system of bargaining council agreements,
including the exemptions from such agreements. The link between wages and productivity
improvements is also severed by bargaining councils. Attempts to increase flexibility and productivity
at enterprise level will be very difficult for as long as wages are determined elsewhere.
. . A skilled workforce
New patterns of work often require flexible and highly skilled workers. Workers are required, for
instance, to function without direct supervision and to be able to correct small interruptions in the
production process on their own initiative. These new patterns, as well as job descriptions that have
become much broader and with less strict demarcation between occupations, necessitate the
multiskilling of workers. Highly capital-intensive production techniques and equipment also require
skilled workers to operate such equipment successfully.
Competitiveness is not and should not simply be based on cost, but should more importantly be
based on the effective and efficient utilisation of all productive resources. One of the most important
resources is the workforce, and a highly skilled workforce can do much to increase the flexibility and
competitiveness of an organisation. Competitiveness is also enhanced by quality, good design, on-
time delivery and rapid adjustments to consumers’ requirements. All these elements require workers
who are skilled, provide quality work and can rapidly adjust to changing conditions. This can be
achieved only by skilled employees who are properly trained.
Trade unions organised strictly on a craftunion basis often resist the trend towards multiskilling and
broader job descriptions, as these are regarded as a threat to the existence of the union itself.
Industrial unions, i.e. unions that recruit workers irrespective of their craft, occupation or grade, have a
certain advantage from a flexibility point of view over craft unions.
Education and training have been neglected in the past in South Africa and this has inhibited
functional flexibility. Hopefully the Skills Development Act and other education and training initiatives
will in time increase the skills of the labour force and thereby enhance flexibility. (This issue is
discussed in greater depth in Chapter .)
. GENERAL CONSIDERATIONS REGARDING FLEXIBILITY
. . Achieving flexibility through collective bargaining

The BCEA allows relatively little scope for the variation of its provisions through collective bargaining
(unless within the confines of a bargaining council) and even less through individual contracts of
employment between employer and employee. The BCEA introduces so many standards of best
rather than minimum conditions of employment and allows so little scope for collective bargaining
outside bargaining councils that collective bargaining between a union and an employer is relegated
to a relatively minor role (see Barker, ).
This contrasts with the LRA, which was implemented less than a year before the BCEA and which has
as one of its primary aims the promotion and facilitation of collective bargaining. Even the
government’s own Labour Market Commission ( : ) and the authors of the ILO Review ( : )
propound so-called “voice regulation”, i.e. regulating minimum standards through bargaining
between employees (and their unions) and employers (and their associations). The BCEA also contains
certain “core rights” that may not be varied under any circumstances, for instance a maximum
working week of ordinary hours.
In contrast to the limited scope for variation through collective and individual agreements, substantial
powers are given to the Minister of Labour to vary the conditions of the BCEA upwards or downwards.
These go even further than the powers of bargaining councils. Although some of these powers might
arguably enhance flexibility, they might also be used in such a way that the opposite result is actually
achieved. Such wide powers will certainly not help to create more legal certainty.
Another method of introducing some flexibility is the making of sectoral determinations for a
particular sector and area, although some might argue that such determinations are more likely to
introduce more onerous or additional provisions than allow for greater flexibility. This is partly
because the new BCEA allows the Minister of Labour more discretion in relation to the powers of the
Employment Conditions Commission than was the case with the former Wage Board, and because the
scope of sectoral determinations is much wider than wage determinations under the previous act.
A final issue in this regard is whether employers can change conditions of employment unilaterally for
operational reasons. For instance, if an employer is experiencing financial difficulties, the question is
whether the employer will be able to change certain conditions of employment in spite of the
employees or the union opposing such changes. The LRA states that it will automatically be unfair to
dismiss workers in order to compel them to accept a change in working conditions. This was the
subject of a case in the Labour Appeal Court (so-called “Fry’s Metals”). In this case, the workers
refused to work in terms of the new conditions and were retrenched.
The judgment found in favour of the employer, saying that in this case the retrenchment was not used
as a threat to get employees to accept the changes, but as final in order to hire new workers who
would agree to the new conditions. One could argue that the judgment allows flexibility for
employers to adapt work practices to suit an evolving environment, but on the other hand it could be
seen to undermine collective bargaining, even though this is subject to onerous conditions.
Considerable management time is required to comply with these conditions, which are to ensure that
all alternatives are considered, that the only possible option has been chosen, and that the option
makes the best possible business sense (Mischke, : ).
SUMMARY
The Basic Conditions of Employment Act (BCEA) allows little scope for variation of minimum conditions of employment
through individual or even collective agreements.
. . Labour market flexibility compared internationally

Employers in South Africa often argue that they are at a disadvantage vis-à-vis employers in other
countries because those employers (especially in developing countries) face less onerous labour
standards. The impact of labour laws on the cost of labour and on flexibility in South Africa is
discussed in sections . and . respectively. A survey of large and small businesses in by
the Small Business Project (SBP) ( ), for instance, gives an indication of the extent to which
employers state that labour issues have an impact on job creation and business decisions. The
following were indicated as the most important constraints with regard to increased employment, in
order of priority:
Lack of confidence or demand in the economy

Labour laws/government regulations in general
High labour costs
Lack of skills in the labour market
Productivity and reliability
The SBP project also indicated that after VAT, labour laws, the CCMA and bargaining councils are the
most time-consuming and troublesome issues.
On the other hand, a survey on flexibility in South Africa initiated in by the ILO (Macun, )
concluded that the labour market in manufacturing is reasonably flexible and that there are signs that
enterprises are able to adapt and to innovate in certain areas. But the ability of enterprises to adapt is
severely constrained by the past, in particular by the disadvantages and discrimination experienced by
African workers, and by the resultant legacy of low trust between employees and employers. Other
key constraints are inadequate managerial skills and training and, in the external environment,
inadequate transport, violence and a poor education system.
The question then is whether South African labour laws are more complex and onerous than those of
“other countries”, however defined. A comparison of labour laws in South Africa with those in other
countries is very difficult, if not impossible. A few points need to be made.
Firstly, the measures of rigidity, especially in so far as firing rigidity is concerned, do not measure
the interpretation of labour legislation by the relevant courts of law. An examination of case law
shows changing opinions and views of Labour Court judges in the interpretation of legislation
relating to job security (Bhorat & Cheadle, : ). This might increase or ameliorate the practical
impact of the relevant legislative provisions.
The trouble about a free market economy is that it requires so many policemen to make it work.
N A
Another factor that is relevant to the practical impact of labour legislation is the strength of
employer and employee organisations. Strong trade unions, for instance, will be swift in ensuring
that the maximum benefit of legislative provisions and court rulings accrue to their members
(Bhorat & Cheadle, : ). South Africa has experienced a sharp increase in the level of
unionisation, and has one of the highest levels of union density (see section . . and Siebert,
: ). In addition, bargaining councils give unions additional power, as discussed in section . .
This has an important effect – not only do unions increase the general level of wages, but they
ensure far better compliance with labour laws than might be the case in countries (especially
developing countries) with similar levels of regulation. Furthermore, as indicated in section . . ,
the wage differentials between unionised and non-unionised sectors in South Africa are higher than
in other countries.
Siebert ( : ) has found a high correlation between high levels of unionisation and high
unemployment. He refers to Australia, Ireland, Korea and Poland as examples of countries that have
had real wage increases but where unemployment has declined because, according to his analysis,
trade union density has fallen.
The quality of human resources is also important. If a country has a legacy of poor education and
training, as is the case with South Africa, coupled with historic discrimination that aggravated skills
shortages, it is very difficult for employers to be able to cope with the same level of labour
legislation as countries that operate with a highly skilled workforce.
It also needs to be borne in mind that there will always be one or more countries that have a
provision in labour legislation that is similar to, or even more onerous than, that in South Africa.
However, the burden of a single legislative provision is not very significant – it is more important to
consider the whole range of labour laws. Therefore, the cumulative effect of all labour and labour-
related policies (against the background of issues such as the management of HIV and AIDS, the
risk profile of the country, crime and other factors) needs to be considered. This is practically
impossible to do.
It is not always the level of regulation that is important, but the change in that level. For instance,
for South Africa to have come from a system of relatively little regulation of dismissals to a
regulatory system found in some European countries will cause some shock to the system for a
number of years, at least until the role players have adjusted. Siebert ( ) says that we are at
midpoint in the OECD rankings in so far as job security is concerned, but we have moved from a
relatively unregulated system to this level, which inevitably has caused an adjustment shock to
many employers.
Research findings
Any firm conclusions about whether South Africa is flexible or not compared to other countries is thus
rather difficult. Research findings present divergent outcomes, depending on whether the research
was based on the perceptions of employers or on a comparison of complex legislative regimes, taking
into account other factors.
Benjamin and Theron ( ) have a useful discussion of the difficulties of comparing labour legislation
across countries, and also have some suggestions as to how better indicators to measure the impact
of various types of labour regulation could be developed, albeit under conditions of optimum data
availability. The most important comparisons of labour regulation will be referred to briefly.
The World Bank does a comprehensive and regular survey of the cost of doing business. This is based
on the inputs received by respondents with significant and routine experience of the particular
regulatory issue. These are professionals or government officials who routinely administer or advise
on the legal and regulatory requirements covered in each Doing Business topic. Because of the focus
on legal and regulatory arrangements, most of the respondents are lawyers (see
http://www.doingbusiness.org). For a critical discussion of the shortcomings of the World Bank study,
see Benjamin and Theron ( ).
Bhorat and Cheadle ( : ), considering the World Bank data and some other research findings,
found that in most measures of labour regulation South Africa is not extraordinarily over-regulated or
under-regulated. However, as stated above, when comparing and , there is an increase in
rigidity in terms of hiring and firing but a decline in the cost of hiring and firing. According to these
authors, South Africa has relatively low levels of regulation with regard to individual employment
relations, which is counterpoised by relatively higher levels of regulation when it comes to collective
rights. In the case of legislative provision for firing workers, South Africa possesses a particularly high
level of regulation. They also refer to the fact that union labour power in South Africa is “extremely
high” (Bhorat & Cheadle, : ).
The World Economic Forum (WEF) ( ), for instance, found that South Africa ranks th (out of
countries) in relation to the ease of hiring and firing practices, th in relation to flexibility of wage
determination and th (ranked last) with regard to labour–employer relations (indicating a
confrontational relationship). On several other indicators on the labour market efficiency pillar, South
Africa also does not fare too well (see Table . ).
TABLE . South Africa’s score and ranking in each indicator on labour market efficiency
Indicator Score (out of 7) Ranking (out of 138)

Cooperation in labour–employer relations 2.5 138
Flexibility of wage determination 2.8 135
Hiring and firing practices 2.3 135
Effect of taxation on incentives to work 4.0 59
Pay and productivity 3.6 98
Reliance on professional management 5.5 32
Country capacity to retain talent 3.5 69
Country capacity to attract talent 3.6 53
Labour market efficiency: aggregate score 3.9 97
Source: World Economic Forum ( : )
The International Monetary Fund (IMF) ( : ) also referred to the difficulties with regard to hiring
and firing, but found in an empirical study that, compared to other industrial and emerging market
countries, South Africa was not more rigid than the average for these countries when measured by: (i)
existence of minimum wages; (ii) level of employment protection and benefits; and (iii) trade union
density. However, the IMF could not consider the difficulties with regard to hiring and dismissal
procedures in the study owing to “measurement difficulties”.
Finally, by comparing the South African unfair dismissal regime with OECD data ( : Chapter ), Van
Niekerk ( ) calls into question the conclusions reached in the World Bank and other studies.
Benchmarking South African labour legislation against indices utilised in the OECD study, he
concludes that South African unfair dismissal laws are not significantly less flexible than those
applicable in most OECD countries, specifically with regard to severance pay, notice periods,
probation periods, fairness criteria, reinstatement and temporary employment. However, he adds that
this does not deny that there are significant problems in relation to the implementation of dismissal
laws in this country, particularly in relation to dispute-resolution structures and procedures. As stated
above, he argues that inappropriate procedures in particular may be the biggest cause of perceptions
of labour market inflexibility.
SUMMARY
An international comparison of labour laws is difficult because factors such as the cumulative effect of policies, a
change in the level of regulation, the level of unionisation and the skills level of the population will be important to
determine the relative impact of labour laws on flexibility.
Various studies on labour market flexibility reach divergent conclusions with regard to whether the South African labour
market is less or more flexible than that of other countries.
. . Balancing the social costs and economic benefits of labour market flexibility
Apart from the supposed or real effects of labour market rigidities on unemployment, there is the
danger that a distinct lack of flexibility might result in an increased differential between the solid core
of workers and those on the periphery, i.e. it might result in increased labour market duality (OECD,
: ). The solid core of workers will be the privileged élite (usually urban union members) who are
employed permanently and who enjoy the full spectrum of labour rights, while those on the periphery
will be non-unionised and with little protection (e.g. in the informal sector) or the unemployed.
A delicate balance is therefore required between protecting worker rights and ensuring adequate
labour market flexibility. Very sensitive issues are involved: not only basic human and worker rights,
but also the survival of enterprises and the wellbeing of the economy. As stated above, it is not simply
a matter of a binary policy choice between flexibility and regulation (Bhorat & Cheadle, ).
Every measure that affects flexibility needs to be considered carefully, with regard to both its social
value and economic cost. On the one hand, it might be found that the social value of a measure,
especially in a country such as South Africa with its historical legacies, is such that there is no
alternative but to retain it. The economic and social cost of trying to remove the measure might be
higher than the cost of retaining it. On the other hand, an objective evaluation might show that the
economic cost is unjustifiably high in relation to the social value, in which case there is a responsibility
on the government, employers and unions to achieve a better cost-benefit relationship.
Addressing this issue at a time when it has never been more necessary to ensure basic worker rights
while simultaneously creating employment opportunities is not easy. There is a great deal of mutual
suspicion between employers and trade unions, and many labour market practices have become
entrenched. Practices that are the norm in developed countries might not be suitable in a developing
country such as South Africa. However, it will not be appropriate to ignore such norms as well as
recent developments in these countries, especially as South Africa usually uses the norms of the most
developed countries as important guidelines for its own policies.
It should be firmly emphasised that labour market inflexibility is not the only, or even the major,
reason for unemployment. Other factors contributing to unemployment include low economic growth
rates and high population growth rates. There is a need for structural economic adjustment (e.g. in
respect of international trade, the development of entrepreneurship, research and development) and
other appropriate macroeconomic policies. Labour market flexibility should be seen within this set of
structural and active labour market policies, and not as an isolated measure to address
unemployment. (Unemployment is discussed in greater depth in Chapter .)
KEY CONCEPTS
atypical or contingent employment

contracts of employment flexibility
efficiency wage hypothesis
efficiency wages
external or numerical flexibility
internal or functional flexibility
job security
labour brokers or temporary employment services
labour market flexibility
labour productivity
multifactor productivity
productivity
remuneration cost per unit produced
total factor productivity
unit labour cost
wage flexibility
working time flexibility
work organisation flexibility
FOR STUDENTS
. Why is productivity improvement important?

. How is productivity traditionally measured and what are some of the shortcomings of this method of
measurement?
. Discuss the main shortcomings of labour productivity data.
. Discuss the factors influencing productivity, focusing on the South African labour context.
. Use the information from the following table to calculate and compare the unit labour cost between and
of two hypothetical firms:
Firm Year Earnings per worker Labour productivity

Firm A 2016 R100 10
2017 R130 13
Firm B 2016 R100 10
2017 R150 12
. Use the efficiency wage hypothesis to discuss the statement: “Under certain circumstances a wage increase
could lead to an increase in productivity.”
. Define labour market flexibility and discuss its main types.
. Define atypical employment and explain its various types.
. Discuss the advantages and disadvantages of atypical employment.
. Discuss how working time can be made more flexible.
SUGGESTED READING*
Bhorat, H., Jacobs, E. & Van der Westhuizen, C. . Do industrial disputes reduce employment? Evidence from
South Africa. African Growth Initiative Working Paper . Washington DC: Brookings Institution Press.
Du Plessis, A. & Smit, B. . South Africa’s growth revival after . Journal of African Economies, ( ): August.
Fedderke, J. . The cost of rigidity: the case of the South African labour market. Comparative Economic Studies,
( ): December.
McCarthy, C. . Productivity performance in developing countries. Country case studies: South Africa. Geneva:
United Nations Industrial Development Organization.
Productivity SA. . Productivity Statistics . Midrand: Productivity SA.
National Productivity Institute (Productivity SA). . Productivity, economic efficiency and equity. A NEDLAC
occasional paper / . Johannesburg: National Economic Development and Labour Council.
Rodgers, G. Labour market flexibility and decent work. DESA Working Paper No. . New York: Department of
Economic and Social Affairs (DESA), United Nations.
Von Fintel, D. . Wage flexibility in a high unemployment regime: spatial heterogeneity and the size of local
labour markets. Stellenbosch Economic Working Papers / . Stellenbosch: University of Stellenbosch.

The authors gratefully acknowledge the assistance by Veliswa Dywili and Mmabatho Mackay of Productivity SA in
providing the latest available labour productivity data for this chapter.
Unemployment in South Africa

. Introduction
. Definition, types and measurement of unemployment
. Unemployment in South Africa
. Reasons for high unemployment
. A strategy to create jobs
. Special employment and income maintenance programmes
Those with jobs and property cannot afford a rebellion of the marginalized against a system that guarantees the property
owners their property rights and comfortable lives, while condemning the excluded millions to lives of misery.
T M
. INTRODUCTION
Unemployment is probably the most severe problem in South Africa and is conceivably the root cause
of many other problems such as crime, violence, poverty and inequality. In the Southern Africa Labour
and Development Research Unit (SALDRU) Living Standards Survey, African, coloured and Asian
respondents put jobs as the first priority (quoted in the ILO Review, : ). Prominent leaders inside
and outside government have also stated that South Africa cannot be governed effectively if this
problem is not solved. These are only some of the reasons why employment creation should become
the highest policy priority (see accompanying box).
As discussed in Chapter , the supply of labour depends on a number of factors, such as the available
wage rates, whether jobs are available, social circumstances (determining the participation rates of
women) and so forth. It is estimated that about people enter the labour market every year on
a net basis after allowing for retirements and deaths. Jobs therefore need to be created for all these
people. For many years, the South African economy could not create sufficient jobs for new entrants,
and the backlog of unemployed people increased year by year. Only in recent years have a greater
number of jobs been created, but still not a sufficient number to make a substantial dent in the
unemployment problem.
Levinsohn ( : ) details a few trends when considering unemployment in South Africa: firstly,
employment rates in South Africa are low, and secondly, unemployment has risen over time. This is
particularly an issue for those individuals who have Matric or less, and for those who are young.
Thirdly, although labour participation rates are low by international standards, they have shown an
upward trend over time. This increase is particularly striking for women.
This chapter looks at the definition of unemployment and related concepts. Various methods for
measuring unemployment are also discussed. The extent of unemployment in South Africa, as
measured by these various methods, is then dealt with. The most important elements of an
employment strategy are discussed in some depth.
A man willing to work, and unable to find work, is perhaps the saddest sight that Fortune’s inequality
exhibits under this sun.
T C
Policy issue: why job creation should be the highest policy priority (or should it?)
Job creation can be an explicit or implied policy priority. It is explicit if the government, for example, incentivises the
creation of jobs. It is implicit if the government focuses on, for instance, making it easier to open and operate a business
and improves investor confidence, which in turn creates an environment that is conducive to job creation. Both policy
avenues would ultimately seek to grow employment. Although the difference may seem slight, the difference in outcome
can be significant. A low economic but high population growth economy that focuses overwhelmingly on job creation will
find that, since the economy is not growing at the same rate as the population, there is simply less wealth or national
income to spread. Conversely, a fast-growing economy might require incentives to place interns and skills training to
avoid the so-called “jobless growth” phenomenon.
While the interaction between fiscal, labour and several other sector policies of a given economy might often contradict
what the overarching policy priority is, it is best to seek consensus on what the priority should be in order to guide policy
negotiations. To say job creation is a policy priority is not really sufficient; the question is how those jobs will be created.
The need for precision when articulating a jobs policy is all the more pertinent precisely because the impact of
unemployment is so severe:
Unemployment has grave consequences for any country. It not only affects an individual’s dignity and self-respect and
erodes his or her standard of living, but also affects society as a whole, because of high crime rates and frustration
leading to unrest and lawlessness. Knowledge and skills that are acquired at great cost are lost quickly through
disuse.
Employment in the wider sense of the word (i.e. including self-employment and employment in the informal sector) is
the main bridge between economic growth and higher living standards. It provides people with incomes that enable
them to establish command over a whole range of goods and services that may enhance their and their dependants’
standard of living (NEDLAC, : ). Employment helps empower women, leading to lower fertility rates and
consequently higher living standards.
Higher job creation also addresses other important social priorities. A major reason for inequality in this country is the
gap between employed and unemployed people. With South Africa’s high unemployment, generating jobs, even with
low wages, will dramatically improve overall income distribution.
Affirmative action policies are much more difficult to implement successfully without employment growth. An important
reason for the success of these policies in countries such as Malaysia was that the number of jobs increased rapidly,
making it easier to change the structure of employment without compromising the rights of existing workers and
without losing their skills and experience.
Unemployment confines people to poverty, especially in rural areas (ILO Review, : ). NEDLAC ( : ) points
out that poverty and employment status are closely linked, with half of all poor people of working age being outside
the labour market altogether. The South Africa Foundation ( ) has estimated that . million new low-wage jobs
(paying R per month in prices) could lift . million people out of poverty. The recent / Income and
Expenditure Survey data also found that two-thirds of poor households were headed by those who were not
employed.
If a free society cannot help the many who are poor, it cannot save the few who are rich.
J F. K
In South Africa the consequences of unemployment could damage the prospects of a stable and successful
democracy, one that is characterised by a peaceful, just, equitable, prosperous and relatively crime-free society.
High unemployment is having a serious effect on perceptions in society of the success or failure of the market
economy. There is a fairly widespread perception among disadvantaged communities that the market economy is at
best not able to address the problem, or is at worst responsible for unemployment.
. DEFINITION, TYPES AND MEASUREMENT OF UNEMPLOYMENT
. . Definition of unemployment
An unemployed person is a person who is without work, is currently available for work, and is seeking
work or wanting to work. Two definitions are generally used to derive statistics on unemployment,
namely the official, narrow (strict) and the broad (expanded) definition.
According to the narrow (strict) definition used in the OHS, LFS and QLFS, unemployed persons
represented those in the labour force who:
did not work during the seven days prior to the interview
wanted work and were available to start work during the reference period (one week in the OHS
and QLFS; two weeks in the LFS)
had taken active steps to look for work or to try to start a business in the four weeks prior to the
interview.
In the OHS and LFS, the broad (expanded) definition of unemployment excluded the third criterion
above. That is, that the expanded definition only required a desire to find employment. The difference
between the number of unemployed derived from the expanded definition and the number of
unemployed derived from the strict definition is then equal to the number of discouraged work-
seekers.
However, the discouraged work-seekers are derived very differently in the QLFS. For someone who is
not classified as employed, if his or her answer to the question “What was the main reason why you
did not try to find work or start a business in the last four weeks?” is “no jobs available in the area” or
“unable to find work requiring my skills” or “lost hope of finding any kind of work”, the person would
be classified as a discouraged work-seeker (Yu, : ). The expanded unemployed in the QLFS is
then equal to the sum of strictly defined unemployed and discouraged work-seekers.
Since the discouraged work-seekers are defined more strictly in the QLFS by considering why the
person did not find work or start a business in the past four weeks, the number of discouraged work-
seekers (and subsequently the number of expanded unemployed) is lower in the QLFS compared to
the OHS and LFS. This will be shown later in Figure . .
There is a great difference between wanting a job and wanting work.

A
The unemployment rate is expressed as the number of unemployed persons as a proportion of the
total labour force (LF) or the economically active population (EAP), which includes both the employed
and the unemployed. The unemployment rate is thus:
number of unemployed persons

Unemployment rate = × 100
total labour force
Table . summarises the derivation of the labour force participation rate (LFPR) and the
unemployment rate (according to the strict and expanded definitions respectively) as well as the
employment rate; it can be seen that discouraged work-seekers are included as part of both the
labour force and unemployed according to the expanded definition, but are regarded as inactive (not
part of the labour force) under the strict definition.
TABLE . Derivation of labour force participation, unemployment and employment rates
Labour market status

(1) Employed
(2) Unemployed
(3) Discouraged work-seeker
(4) Inactive
Narrow (strict) labour force participation rate

= Labour force/Working-age population
(1)+(2)
=
(1)+(2)+(3)+(4)
Broad (expanded) labour force participation rate

= Labour force/Working-age population
(1)+(2)+(3)
=
(1)+(2)+(3)+(4)
Narrow (strict) unemployment rate

= Unemployed/Labour force
(2)
=
(1)+(2)
Broad (expanded) unemployment rate

= Unemployed/Labour force
(2)+(3)
=
(1)+(2)+(3)
Employment rate
= Employed/Working-age population
(1)
=
(1)+(2)+(3)+(4)
Work is a drug that dull people take to avoid the pangs of unmitigated boredom.
W. S M
Natural rate of unemployment. There is no such thing as full employment, even under the most
favourable economic conditions. At any one stage, some workers will be unemployed, either moving
between jobs, or moving from outside the labour market to inside (e.g. moving from being a full-time
student to trying to find a job), or younger workers trying out different job opportunities and so on.
This constant rate of unemployment is called the natural rate of unemployment, and exists where
there is neither excess demand nor excess supply in the overall labour market.
Borjas ( : ) as well as Ehrenberg and Smith ( : ) argue that the natural rate of
unemployment could also be defined in the following ways:
the rate at which all unemployment is voluntary (frictional and perhaps seasonal – to be discussed
in section . . )
the rate at which the number of vacant jobs equal the number of unemployed
the rate at which any increases in aggregate demand causes no further reduction in unemployment
the rate that persists regardless of the inflation rate (this relates to the Phillips curve – to be
discussed in section . . ).
Some factors may cause the natural rate of unemployment to change, such as: changes in the number
of young workers who are comfortable to be between jobs; changes to the reservation wage of
workers (the wage at which they will be prepared to accept a job); changes in the availability of labour
market information and the effectiveness of policies that bring work-seekers and employers together
more quickly; and changes in the proportion of women in the labour market (who may also wait
longer for the “right job” to enable them to balance family and work life). The concept of a natural
rate of unemployment is theoretical as far as South Africa is concerned, and will remain so for the
foreseeable future, because there is an inadequate demand for labour.
. . Types of unemployment
In order to address the problem of unemployment successfully, a distinction should be drawn
between different types of unemployment. This gives an indication of the possible reasons for
unemployment, and therefore also some idea of how the problem should be addressed (see
McConnell et al., : Chapter for a full discussion):
Frictional unemployment arises as a result of the normal labour turnover that occurs in any dynamic
economy and the time lags involved in the reemployment of labour. Since there are people moving
between jobs and new entrants to the labour market at any given time, there are both unemployed
persons and vacancies that can be filled by them, and it usually takes time for those seeking work
to find and fill these vacancies. Frictional unemployment is usually of relatively short duration and
can be reduced even further by improving labour market information and placement services so
that the employer and job-seeker can find each other sooner and more effectively. Frictional
unemployment is one of the factors that determine the natural rate of unemployment.
Seasonal unemployment is due to normal and expected changes in economic activity during the
course of a single year. It is found in many sectors, with agriculture and the retail trade probably the
best examples. Persons working during peak periods and unemployed in offpeak periods are
described as seasonal workers or seasonally unemployed. This type of unemployment occurs on a
regular and predictable basis.
Cyclical unemployment (or demand-deficient unemployment) arises during recessionary periods,
when aggregate demand, and therefore also the demand for labour, is low. During recessionary
periods few or no jobs are created for new entrants to the labour market, and even existing workers
might lose their jobs through retrenchments. Once the economy improves, however, the cyclically
unemployed again find jobs. In South Africa, cyclical unemployment has a dimension that makes it
difficult to address successfully: it is superimposed on large-scale structural unemployment. As a
result, the unemployment problem is severe, complex and difficult to alleviate.
Structural unemployment is more difficult to define, but generally refers to the overall inability of
the economy, owing to structural imbalances, to provide employment for the total labour force
even at the peak of the business cycle. Even during periods of high economic growth, job
opportunities do not increase fast enough to absorb those already unemployed and those newly
entering the labour market. There are various reasons for this, for example the rapid growth of the
labour force, the use of capital- or skills-intensive technology, or an inflexible labour market.
Structural unemployment could also refer to a skills mismatch, i.e. between the skills that employers
require and those that employees offer, or a geographic mismatch, i.e. between the locations of job
vacancies and those of job-seekers. Ehrenberg and Smith ( : ) argue that had wages been
completely flexible and the costs of skills and geographical mobility been low, market adjustments
would have quickly eliminated structural unemployment. However, these conditions are unlikely to
hold in reality, and hence this type of unemployment exists. Lastly, the major proportion of
unemployment in South Africa is structural rather than cyclical (Chadha, : ; Banerjee, :
).
SUMMARY
A person is said to be unemployed if he or she is without work, is currently available for work, and is seeking work or
wanting to work.
The official, narrow (strict) definition of unemployment is that a person should not be in paid employment or self-
employment, should want to work and be available for employment, and should have taken active steps to find
employment.
The broad (expanded) definition requires that, instead of actively taking steps to find a job, the person must simply
have the desire to work and to take up employment. There is much controversy over whether the official
unemployment rate is a true reflection of unemployment in the country because of the requirement that a person
actively needs to seek a job before being counted as unemployed.
Discouraged work-seekers are included as part of the unemployed (and labour force) according to the broad definition.
They are defined more strictly with the introduction of the QLFS.
. . Measurement of unemployment
Although there are various ways of measuring unemployment, there are three broad approaches:
The sample survey method. A survey is undertaken among a number of households to determine
the economic status of the members of the household.
The census method. The economic status of the whole population is determined by asking
individuals what their economic status is.
The registration method. Unemployed persons register as such.
The most dependable source of unemployment statistics in South Africa is probably regular sample
surveys of the labour force. Different labour surveys have been conducted over time, namely the
– October Household Surveys (OHSs), – Labour Force Surveys (LFSs) and the
Quarterly Labour Force Survey (QLFS) introduced since . There are important caveats to the
comparison of these surveys over time and, in this regard, careful note should be taken of the
discussion in sections . . and . . .
As far as the population census is concerned, it is subject to two shortcomings:
( ) Only a limited number of questions are asked to capture labour market status and these
questions are not detailed enough to pinpoint the complex phenomenon of unemployment,
compared to the above-mentioned labour surveys.
( ) The census is conducted at most every five or years and there is a substantial lag between the
data collection and the release of the results.
In South Africa, statistics on registered unemployment are obtained from returns submitted on a
monthly basis by the different placement centres of the Department of Labour. These returns reflect
the numbers of people who have registered at the offices of the Department as unemployed.
However, the data on registered unemployment do not relate in any way to the unemployment
figures calculated according to other methods.
The two reasons for this are:
( ) Not all unemployed persons register, as registration is voluntary and the chances of being placed
in employment are relatively small – the Department succeeds in placing only a small percentage
of persons registered as unemployed.
( ) Even if they register, many unemployed are still not eligible for unemployment benefits, for
example those unemployed for longer than six months, new entrants to the labour market and
the unemployed individuals who have not worked long enough to qualify.
. UNEMPLOYMENT IN SOUTH AFRICA
. . Total unemployment
The total unemployed in South Africa according to the official, narrow (strict) definition nearly tripled
from two million in to . million in , as shown in Figure . . This means that on average the
number of unemployed increased by approximately per annum. The figure shows that the
total unemployed according to the expanded definition increased from . million to . million
during the same period. The abrupt decline of unemployment under the broad (expanded) definition
between and (a decrease of nearly two million) was attributed to the changes in the
methodology to define the discouraged work-seekers discussed in section . . (The difference
between the broad labour force and narrow labour force gives the number of discouraged work-
seekers). For the remainder of section . , the focus is on unemployment according to the narrow
(strict) definition.
FIGURE . Total number of unemployed, –
. . Unemployment by demographic and education characteristics

Table . shows that both male and female unemployment increased from one million in to
about three million in , with the female share accounting for % in . As far as
unemployment by population group is concerned, Bannerjee et al. ( : ) point out that it is far
higher among Africans than other groups, and this is confirmed by the results shown in Table . :
African unemployment trebled from . million to . million, and the African share of total
unemployed increased from . to . % between and .
TABLE . Unemployment in South Africa by gender (’ )
1995 2005 2015 2016

Male 922 2 052 2 737 3 017
Female 1 104 2 424 2 678 2 853
1995 2005 2015 2016
Total 2 027 4 479 5 415 5 869
Female unemployed 54.5 54.1 49.5 48.6
as proportion of total
Note: Unemployed with unspecified gender are not shown separately in this table but are included as part of the total.
TABLE . Unemployment in South Africa by population group (’ )
Africans Coloureds Indians Whites Total Proportion of total

/Asians
1995
Total 1 692 217 42 75 2 027 100.0
2005
Total 3 902 384 82 106 4 479 100.0
2015
Total 4 735 483 72 124 5 415 100.0
2016
Total 5 123 488 79 150 5 869 100.0
Note: Unemployed with unspecified population group are not shown separately in this table but are included as part of the
total.
Table . shows that approximately two-thirds of the unemployed were aged – years at the time
of the survey. In other words, the youth accounts for a high share of unemployed people. Most of
them have never worked before, which means that they do not qualify for unemployment insurance
benefits. Such persons, therefore, have no sources of support other than their families, and may thus
easily turn to crime or begging.
TABLE . Unemployment in South Africa by age cohort (’ )
15–24 25–34 35–44 years 45–54 years 55–64 years Total Proportion of total
years years
1995
Total 646 821 366 153 41 2 027 100.0
Proportion of 31.9 40.5 18.1 7.5 2.0 100.0
total
15–24 25–34 35–44 years 45–54 years 55–64 years Total Proportion of total
years years
2005
Total 1 499 1 807 721 355 97 4 479 100.0
Proportion of 33.5 40.3 16.1 7.9 2.2 100.0
total
2015
Total 1 378 2 157 1 201 521 157 5 415 100.0
Proportion of 25.5 39.8 22.2 9.6 2.2 100.0
total
2016
Total 1 496 2 317 1 347 568 141 5 869 100.0
Proportion of 25.5 39.5 23.0 9.7 2.4 100.0
total
This also explains the introduction of the Employment Tax Incentives Bill (ETIB) or youth wage subsidy
in South Africa since (to be discussed in section . ), with specific focus on boosting youth
employment.
Table . shows that the number of unemployed without secondary education remained constant at
between and , but its share of total unemployed decreased drastically from % to
%. This result is not surprising, as the labour force (and unemployed) have become more educated
over the years, as shown in section . . . A worrying finding is that unemployment among the
educated is increasingly becoming a phenomenon in South Africa. The number of unemployed
individuals with Grade increased from . to . million during the period – , while the
number of unemployed with post-Matric qualifications increased from to . It generally
takes a completed university degree to escape unemployment in South Africa (Banerjee et al., :
).
Although graduate unemployment remains small in relative terms – only . % and . % of all
unemployed persons in and respectively – the growing incidence of graduate
unemployment at a time when serious skills shortages are experienced is worrying. This also implies a
high cost to the country in terms of expenditure on these people’s education because the economy is
not using this education. Furthermore, people with higher educational qualifications become
frustrated more easily if they do not find a job, and this increases the potential for crime and civil
unrest. Strydom ( ) has, for instance, pointed out that the potential for political unrest increases as
the educational level of the population increases.
Although not shown in Table . , in nearly % of unemployed people with post-Matric

qualifications had a certificate or diploma (that is, the remaining % had a Bachelor’s degree),
indicating problems with the country’s Further Education and Training (FET) system. The majority of
these unemployed came from the fields of commerce, education, information technology and health.
These results imply that graduates do not possess the right qualifications and often these
qualifications are not of a standard or quality that is required by employers.
TABLE . Unemployment in South Africa by education (’ )
None–Grade 7 Grade 8–11 Grade 12 Grade 12 + certificate/diploma/degree Total Proportion of total

1995
Total 689 825 439 62 2 027 100.0
2005
Total 1 014 1 977 1 313 158 4 479 100.0
2015
Total 640 2 504 1 841 408 5 415 100.0
2016
Total 656 2 847 1 869 467 5 869 100.0
Note: Those with certificates or diplomas but did not have Grade fall under the “Grade – ” category. Also, labour force with
unspecified education are not shown separately in this table but are included as part of the total.
Finally, the increase in unemployment among people with secondary or even tertiary education seems
to have had a significant impact on African people. This may relate to perceived differences in the
quality of tertiary education, especially at previously disadvantaged tertiary educational institutions
(Bhorat & Oosthuizen, : ).
Many fellows nowadays have a B.A., M.D. or Ph.D. Unfortunately, they don’t have a J.O.B.
A “F ”D
How do the unemployed cope?
The high level of unemployment indicated by the labour surveys raises the question of how unemployed people survive.
As is indicated below, unemployment insurance covers only those who have worked before, and only for a certain period
of time. The following coping strategies have been identified (Erasmus, ):
The majority of the unemployed (between and %) receive remittances from employed family members or friends
and/or family members or friends who receive a pension.
Others are involved in some form of income generation from which they seem to make a living (from % up to as
high as % of those who actually consider themselves as “unemployed”).
Some unemployed people do benefit from the government’s special job creation and training programmes,
employment services, and small, medium and microenterprise (SMME) support and development services, although
the impact is relatively small. For example, the September LFS found that about people had been
involved in the government’s expanded public works programme (EPWP) during the previous six months.
A substantial portion (at least %) is not reached by any of the above services, does not have a supportive network
and does not earn a living in the informal sector.
Social security benefits apply to only a small portion ( %) of this group of people (at least, this was the case in ,
but it might have changed since then owing to the expanded nature of such benefits).
. . Unemployment rates by demographic and education characteristics

Figure . indicates that the official, narrow (strict) unemployment rate initially increased from . %
in to an all-time high of . % in , before declining to as low as . % in . Another
upward trend took place since , and the unemployment rate rose to . % in . A similar trend
could be observed for the broad (expanded) unemployment rate, except that the abrupt -
percentage-point decline of this rate between and was mainly attributed to the adoption
of a different methodology to identify the discouraged work-seekers with the introduction of the
QLFS. For the remainder of this section, narrow (strict) unemployment rates by demographic and
education characteristics are looked at.
FIGURE . Unemployment rate, –
The unemployment likelihood has always been particularly higher among women, as shown in Table
. . However, the difference between male and female unemployment rates has diminished from .
to . percentage points between and . This result suggests the increasing feminisation of
the South African labour market (see sections . . and . . ). On the other hand, Table . indicates
that the unemployment rate increased in all four race groups, but the African unemployment rate
remains the highest – exceeding % in .
TABLE . Unemployment rates (%) by gender
Male Female
1995 13.8 23.0
2000 22.3 29.2
2005 22.6 31.7
2010 23.5 27.9
2015 23.5 27.9
TABLE . Unemployment rates (%) by population group
Africans Coloureds Indians/Asians Whites

1995 21.7 15.9 10.6 3.9
2000 30.4 18.5 15.8 5.9
2005 31.6 22.5 15.7 5.1
2010 27.9 21.5 7.7 5.6
2015 28.8 22.8 12.5 5.9
2016 30.5 22.9 13.2 7.3
Unemployment probability decreases across the older age cohorts, as indicated in Table . . In
particular, more than % of labour force aged – years were unemployed in ; this proportion
was also high at nearly one-third for those aged – years. Table . shows that the
unemployment rates of those without Matric were above % in , while the unemployment rate
of those with Matric ( . %) was twice the likelihood of those with post-Matric qualifications ( . %).
This finding indicates that having post-Matric qualification would significantly boost a labour force
entrant’s likelihood of finding employment.
TABLE . Unemployment rates (%) by age cohort
1995 36.5 20.0 11.4 8.8 6.2
2000 46.6 29.9 17.7 12.3 6.9
2005 51.4 30.3 18.2 13.0 8.5
2010 52.3 29.7 18.2 13.0 7.7
2015 49.9 30.3 19.7 14.1 10.1
2016 54.2 32.1 21.5 14.6 9.0
TABLE . Unemployment rates (%) by education
None–Grade7 Grade 8–11 Grade 12 Grade 12 + certificate/diploma/degree

1995 19.0 21.4 17.4 4.5
2000 23.5 30.8 29.8 10.3
2005 25.1 33.6 28.2 7.8
2010 24.0 32.0 26.6 9.3
2015 21.7 32.5 27.2 11.3
2016 23.4 34.4 27.0 13.2
. . Other characteristics of the unemployed

Urban areas. In , the official unemployment rate in urban areas ( . %) was only slightly lower
than that in traditional or rural areas ( . %).
Province. Unemployment rate increased across all nine provinces between and , as shown in
Table . . This rate is the lowest in the Western Cape ( . % in ), while the Free State, North
West and Mpumalanga are associated with the highest rates of unemployment.
TABLE . Unemployment rates (%) by province
1995 2000 2005 2010 2015 2016

Western Cape 13.8 16.5 18.9 22.5 20.7 21.7
Eastern Cape 24.4 25.4 30.1 26.7 29.2 28.2
Northern Cape 20.4 18.6 24.8 25.3 34.8 29.6
Free State 12.4 23.5 30.4 29.7 31.5 34.2
KwaZulu-Natal 20.8 27.8 32.9 19.6 20.5 23.5
North West 17.2 28.6 27.4 27.7 25.4 30.5
Gauteng 15.8 26.5 22.8 27.4 28.7 29.1
Mpumalanga 16.5 26.6 27.0 28.4 26.2 30.4
1995 2000 2005 2010 2015 2016
Limpopo 20.1 28.00 30.2 24.7 18.8 21.9
South Africa 17.6 25.4 26.8 25.4 25.5 27.1
Long-term unemployment. In , by far the largest proportion of the unemployed ( . %) had been
searching for work for longer than a year, and nearly half of these people had been looking for work
for five years or more. This has extremely serious implications for those involved, as unemployment
insurance benefits (see section . . ) are normally paid for six months only.
Previous employment. In , about % of the unemployed had never held down a job. They would
therefore not have contributed to an unemployment insurance fund, and would not qualify for such
benefits. For the remaining % who had worked before, % of them last worked more than a year
before, with half of them last having worked more than five years before. They could also be defined
as chronically or long-term unemployed. Finally, for those with past work experience, %, % and
% of them worked in skilled, semi-skilled and unskilled occupations respectively, while the
proportions of them working in the primary, secondary and tertiary sectors were equal to %, %
and % respectively in .
SUMMARY
The unemployment rate according to the official definition of unemployment was . % in the third quarter of , with
. million people unemployed. Most at risk of unemployment are African women without post-Matric qualifications and
living in the rural areas of the Free State, North West and Mpumalanga.
Unemployment among educated people is increasing.
If discouraged work-seekers, i.e. those who have stopped taking active steps to find work, are included as
unemployed, then the number of unemployed people increases to . million and the unemployment rate rises to
. % in .
. REASONS FOR HIGH UNEMPLOYMENT
The high unemployment that is prevalent in South Africa can be ascribed to factors on both the side
of the supply of labour and the demand for labour.
As far as the supply of labour is concerned, there was a rapid increase of the labour force in the late
s (Banerjee et al., ; Bhorat, ), as shown in Figure . . Burger et al. ( ) argue that such
a rapid increase could be attributed to the two policies implemented by the Department of Education
to reduce the number of over-age learners in the school system: learners who were more than two
years older than the correct grade-age were not allowed to be accepted, and students could not be
held back more than once in each of the four schooling phases (Grade – , – , – , – ). Hence,
these lowly-educated learners had no alternative but were “forced” to enter the labour market to seek
work. There has also been an influx of unskilled, illegal immigrant labour, which is largely unmeasured
in surveys. Since the likelihood of finding work is low for these people, this may explain their long-
term unemployment status, and subsequently the rising unemployment of the country.
Regarding the demand for labour, technological advancements and capital deepening lead to a
structural change of the economy, resulting in a growing demand for skilled labour but a decreasing
demand for unskilled labour (Bhorat, ) as shown in Table . and Figure . . The demand for
unskilled labour fell in the two sectors that would normally be a large employer of unskilled labour,
namely agriculture and mining. At the same time there was a relative shrinking of manufacturing,
especially export-orientated manufacturing, resulting in a collapse in the demand for relatively
unskilled labour. Hence, a skills mismatch took place as there was an “excess demand for skilled
labour and oversupply of unskilled labour” at prevailing wage rates (Arora & Ricci, : ).
FIGURE . Proportion of skilled workers in each sector, selected years
Source: Author’s own calculations using the October Household Survey, Labour Force Survey September and
It is also argued that the provisions in the labour legislation have made it time-consuming, costly and
difficult for employers to retrench non-performing workers and hire new workers (Mahadea, ;
Arora & Ricci, ; Yu, ). The constraints on the employers include, among others: high
procedural costs associated with the dismissal of unproductive workers; the time-consuming set of
procedures that employers have to follow during the hearings with the CCMA; and additional direct
costs to the employers that relate to minimum standards that apply for workers’ benefits when hiring
workers, such as unemployment insurance and skills development funds. The ultimate outcome is
unnecessary employment rigidity, discouraging employers from hiring labour.
While the impact of minimum wages on employment have been discussed in section . , critics have
suggested that the existing process of collective bargaining may stifle the growth of new business
(particularly job creation in smaller enterprises), as larger firms agree to minimum wages that are
frequently unaffordable to smaller firms. The latter businesses, typically more labour intensive, are
forced to become more capital intensive, retrench workers or completely shut down (Nattrass, ;
Bhorat, ). Although it was discussed in section . . that unit labour cost remains constant if the
wage increase is complemented by a labour productivity increase, this does not seem to be the case
in South Africa (see Figure . ).
There is also an inherent asymmetry between the desires of the currently employed (i.e. insiders) and the
unemployed (i.e. outsiders). The insiders prefer wage levels to be high, while some of the outsiders
may be willing to accept wage levels below the minimum wage. However, the views of the outsiders
are not well represented when trade unions and employers negotiate over worker wages, and wages
are too sticky and slow to fall even during times of low labour productivity and labour demand. Thus,
the outsiders (with most of them being youngsters desperately seeking their first job) remain
chronically unemployed (Von Fintel & Burger, ; Paton, ).
In many other countries, the decline in formal employment would to some extent have been
compensated for by an increase in the size of the informal sector. However, as pointed out in section
. , the informal-sector employment in South Africa has been stagnating at the . – . million range
since . In other words, there is no indication that the informal sector is rapidly absorbing the
surplus labour from the formal sector. There are numerous barriers to entry to the informal sector,
ranging from crime, lack of access to formal or even informal credit and inadequate government
promotion of microenterprises and informal enterprises, to weak infrastructure, underdevelopment of
entrepreneurial skills and social networks (Kingdon & Knight, ; ; Rogerson, ; Burger &
Woolard, ; Devey, Skinner & Valodia, ).
Finally, in a few recent studies Burger and Jafta ( ; ) found that affirmative action might not be
a success to rapidly reduce employment discrimination by race and gender, even after controlling for
differences in demographic and education characteristics between groups. Labour market
discrimination is the focus of Chapter .
. A STRATEGY TO CREATE JOBS
Numerous official documents addressing the problem of unemployment in South Africa have
appeared. The ANC alliance produced the Reconstruction and Development Programme (RDP) ( ),
and the Government of National Unity published an RDP White Paper ( ), both of which relate
directly to employment creation. In , the ANC-led government released its GEAR policy, which has
job creation as one of its most important objectives (RSA, ).
In , general concern about South Africa’s unemployment problem culminated in the holding of a
job summit. The most comprehensive document published on this issue by government to date is its
input to the job summit (RSA, (a), http://www.polity.org.za or
http://www.gov.za/documents/combsubl.htm). In the government released its Accelerated and
Shared Growth Initiative of South Africa (ASGI-SA) (RSA, ). Two more initiatives were launched by
the government in the s, namely the New Growth Path (NGP) (RSA, ) and the National
Development Plan (NDP) (RSA, ).
These economic initiatives (RDP, GEAR, ASGI-SA, NGP and NDP) have job creation as one of their
primary policy goals by implementing various labour market policies that included, among others,
public works programmes, education and skills development, the promotion of small and medium
enterprises and sectoral investment strategies.
FIGURE . South African economic policy initiatives since
The primary labour market objective of each policy initiative is as follows:
RDP: to create . million jobs between and

GEAR: to create two million jobs between and
ASGI-SA: to halve unemployment and reduce the official unemployment rate to % by the end of
NGP: to create five million jobs and reduce the official unemployment rate to % by
NDP: to create million jobs in years from and reduce the official unemployment rate to
% and % by and respectively.
In the discussion below, attention is given to a number of aspects of employment that are raised in
these strategies, as well as some of those raised in the strategies of other stakeholders.
. . Economic growth and exports

One of the reasons for South Africa’s high unemployment is the relatively slow growth of the
economy when compared to other upper-middle-income countries such as China, Malaysia, Mauritius
and a number of other industrialising countries. However, even though economic growth is a major
determinant of expanding employment opportunities, higher growth does not invariably and
automatically translate into rapid increase in employment (see section . . ). The degree to which
employment is created also depends on the pattern of growth and the labour intensity of the
economic growth. This is discussed in more detail in the sections that follow.
Most serious observers would agree that economic growth should form the main component of an
employment creation strategy. The elements of such a growth strategy fall outside the scope of this
book, save to say that South Africa has been depending on the savings of others to finance
investment, which is a critical driver of economic growth. This is reflected in recurrent deficits on the
current account of the balance of payments. A large capital inflow is required to finance this deficit,
and if government policies are such that investors have less confidence in the country, this inflow of
foreign capital will rapidly cease and might even reverse.
However, even with the capital inflow, and in spite of a commodities boom, South Africa’s economic
performance has been rather unimpressive by world standards, as shown in Figure . . The Centre for
Development and Enterprise and Business Leadership South Africa ( ) state that badly coordinated
policies were an important reason for this state of affairs. As stated by the authors:
FIGURE . Annualised real GDP growth rates (%), selected upper-middle-income

countries
This lack of co-ordination was the result of a political compromise which saw organised labour’s
agenda implemented on labour market regulation, and business’s agenda implemented on
inflation and interest rates, while the government itself pursued a programme of trade
liberalisation. Looked at in isolation, each of these policies has positive features. But implemented
together, they meant that our industries kept on shedding unskilled labour while the economy
grew ( : ).
So even if the positive trends of the past decade could continue – which is not possible because of
the global economic crisis – by more than % of working-age labour force in South Africa are
still out of work.
However, economic growth cannot be relied on to fully address the country’s unemployment
problems. This is because the so-called trickle-down effect of economic growth (i.e. the extent to
which people at the bottom of the economic order will benefit from economic growth, e.g. through
the creation of jobs) is relatively low. Some other measures that relate more directly to the labour
market are thus discussed below.
. . Allowing higher inflation in order to create jobs: the Phillips curve

In , the economist, Prof. A.W. Phillips, documented the negative correlation between the rate of
inflation and the rate of unemployment in the UK over a relatively long period of time. As inflation
increased, unemployment seemed to decline. During the s, the experience of the US economy
seemed to confirm that there is a trade-off between inflation and unemployment. This is known as the
Phillips curve. Figure . (a) indicates that if the unemployment rate is % and inflation is % (at
point A), the government can pursue expansionary (monetary or/and fiscal) policies that will move the
economy to point B, which will reduce unemployment to % at the cost of inflation increasing to %.
Inflation rate increases since the expansionary policies would lead to a rightward shift of the AD curve,
causing the general price level to go up from P to P while real GDP increases from Q to Q , as
shown in Figure . (b). For instance, unions in South Africa have at times pressured the government
to pursue monetary and fiscal policies that would reduce unemployment at the cost of a “reasonable”
increase in inflation.
FIGURE . The Phillips curve trade-off between inflation and
unemployment
However, the experience of the US during the s shows the folly of this approach. The assumed
relationship between inflation and unemployment disappeared, and both unemployment and
inflation seemed to be moving to constantly higher points. There was thus no longer a trade-off
between inflation and unemployment – the unemployment rate had increased to higher levels, but
the inflation rate had also increased.
The explanation offered by economists for this phenomenon is also provided in Figure . (a). If the
government follows expansionary policies, there is a greater demand for workers, and inflation is at
the same time pushed up to %. Unemployment is reduced to % and the economy moves to point B.
However, after a while workers adjust their wage demands to the higher inflation rates. The higher
real wages cause the quantity demanded of labour to drop, and unemployment increases to the
previous rate of % again. It is also possible that with the higher inflation rate of %, input costs
increase, so some firms withdraw from the market, causing the aggregate supply (AS) to shift
leftwards, real GDP to drop from Q to Q (which in turn leads to job losses) and price level to
increase to P , as shown in Figure . (b). At the end, the economy moves to point C in Figure . (a).
Comparing points A and C, unemployment remains unchanged, but inflation has increased. There is
thus no long-run trade-off between inflation and unemployment and the long-run Phillips curve is
vertical – a particular unemployment rate can be compatible with various inflation rates. Finally, it is
argued that the unemployment rate at which inflation persists ( % in Figure . (a)) stands for the
natural rate of unemployment, as discussed in section . . .
. . Improving the labour absorption capacity of the economy

Although economic growth is important for job creation, a country such as South Africa also needs to
create more employment opportunities for every percentage point increase in economic growth than
is the case at the moment. GEAR, RDP, NGP and NDP refer to the importance of improving labour
absorption (RSA, : ; Government of National Unity, : ; RSA, ; RSA, : ; RSA, :
). Unfortunately, the opposite has happened over the past three decades, at least until the mid-
s. As indicated in section . . , fewer formal employment opportunities were created for every
percentage point of economic growth in the late s and early s than was the case in earlier
years.
To increase labour absorption and the labour intensity of growth, attention must be given, among
other things, to employment-enhancing policy shifts with regard to the cost of labour and labour
policies. These are discussed below (see sections . . and . . ). Appropriate technology that takes
into account the relative shortage of capital and the comparative abundance of semi-skilled and
unskilled labour in South Africa can play an equally important role here. Section . . should be
consulted with regard to the issue of capital and labour intensity, and the factors that encourage the
utilisation of capital as opposed to the use of labour. Small business development can also play a
significant role in this regard. This is also referred to later.
In ASGI-SA, three labour-intensive sectors were identified for special priority attention, namely
business outsourcing, tourism and biofuels. In contrast, the NGP framework rather acknowledged that
while it is important to promote large-scale employment in labour-absorbing sectors such as
agricultural value chain, light manufacturing and services, it is undeniable that for full employment to
be achieved in the long run, the state must increasingly support capitaland knowledge-intensive
sectors.
Inward industrialisation
Inward industrialisation can help to turn around the trend of low labour absorption. Inward
industrialisation was first defined by Lombard ( ) and refers to a process whereby the effective
demand for locally produced basic goods and products increases because of various factors operating
together. These factors include rising real income among lower-income categories, falling birth rates
in such communities, a higher urbanisation rate and more government spending on less privileged
communities.
The combination of these factors can unleash a meaningful demand for basic consumer goods (e.g.
clothing, shoes, furniture, food) and other facilities (e.g. low-cost housing). Most of these goods and
services can be produced in a labour-intensive manner, they have low import content and require a
less skilled labour component to produce. Relatively little additional pressure will therefore be placed
on the balance of payments or the supply of highly skilled labour. This is in line with one of the key
programmes of the RDP White Paper, namely, meeting basic needs (Government of National Unity,
: ). In its employment strategy, the government also emphasised the importance of fast-tracking
new labour-absorbing industries such as tourism, and sectors producing goods for domestic markets
(RSA, : par. . ).
COSATU ( ) and the National Labour and Economic Development Institute (Naledi) ( ) have
also identified a number of measures that will improve the ability of the South African economy to
create employment. These include: supporting job-creating sectors; expanding the demand for
domestically produced goods and services; encouraging import substitution (particularly of
intermediate and capital goods); awarding government contracts on a basis that will encourage job
creation; and reforming the credit and financial system to support job creation. FEDUSA ( ) lists
the following sectors in rank order in terms of their job-creating ability:
Tourism
Construction/infrastructure
Agriculture
Furniture manufacturing
Labour-intensive export manufacturing
Information service industries
Preconditions
There are important preconditions for the success of inward industrialisation. First, real wage increases
should not exceed productivity increases so that there will be an increase both in real incomes (i.e.
effective demand) and concomitantly in production. A further precondition is that the small-enterprise
and informal sectors be stimulated so that the benefits of the inward industrialisation process can be
spread widely among the population.
SUMMARY
The process whereby the economy is creating fewer jobs for every percentage point of economic growth needs to be
reversed by employment-enhancing policy shifts and moderation in the labour cost increases.
. . The cost of labour

As illustrated in section . , a substantial body of research shows that increases in the cost of labour
without concomitant increases in productivity have a negative impact on employment. The cost of
labour includes not only money wages, but also indirect costs, such as fringe benefits and employer-
provided entitlements to cover contingency risks. The authors of the ILO Review ( : ) state that
“reducing labour costs will help boost employment”, but point out that there are many ways of
reducing labour costs besides focusing on the wage rate. The government’s macroeconomic strategy,
GEAR (RSA, : ), also refers to the importance of “wage moderation” and this is echoed in a
working paper by Chadha ( : ), an IMF researcher. GEAR (RSA, : ) calls for a national social
agreement that ensures, among other things, that wage and salary increases do not rise more than
productivity growth. NDP ( : ) mentions that income of low-income and working-class
households is raised through productivity growth, while NGP ( : ) even calls for increasing
multifactor productivity “on the basis of fair rewards to workers plus greater employment creation”.
In the government’s employment strategy, labour organisations are challenged to commit themselves
to wage bargaining appropriate to productivity gains and rates of inflation, and which do not
contribute to job losses (RSA, : par. . . . ). The strategy also proposes a youth learnership wage
to encourage the creation of more jobs for young people (ibid.: par. . . . ).
It should be noted that the structure of bargaining in the country, and the approach and philosophy
of unions, have a significant effect on wage levels and other labour costs, as pointed out in sections
. and . . As far as the impact of unions is concerned, it seems as if they have caused a higher
differential between unionised sectors and non-unionised sectors in South Africa than in other
countries, a differential that seems to be in the region of % (see section . . ). This is the result of
strong unions in this country (a very high rate of unionisation), the system of centralised bargaining at
sectoral level, i.e. bargaining councils, and high unemployment. There has been much criticism against
bargaining councils in South Africa, among others from the ILO, the Presidential Labour Market
Commission and various researchers. International evidence suggests that highly centralised or highly
decentralised bargaining structures deliver the best economic outcomes, and South Africa’s system of
sectoral bargaining delivers the worst.
It is therefore not unexpected that the government’s employment strategy calls for amendments to
the provisions regarding the extension of bargaining council agreements to non-parties in order also
to take into account the sensitivity of such agreements to job creation (RSA, : par. . . . . ), and
agreement between unions and employers on entry-level wages to reduce the initial cost of hiring
young workers (RSA, : ).
However, other aspects of the government’s labour policies and legislation have also resulted in an
increase in the direct and indirect cost of labour. This is discussed in more detail in section . .
The authors of the ILO Review ( : ) have pointed out that money wages and employer-provided
entitlements to cover contingency risks are relatively high in South Africa. This is so because neither
the state nor rural kinship support systems have played a major role in providing some elements of
the so-called “social wage” (see section . . ). The cost of labour to the employer in this country
therefore appears to be relatively high, and thus has a negative impact on employment.
The sections referred to above contain important information related to the cost of labour, and should be consulted as an
integral part of the discussion in this section.
Reservation wages
An important question that is often asked in the context of unemployment is whether many persons are not unemployed
because they have high wage expectations and are unwilling to accept a job (for instance, in the informal sector) that
does not meet these expectations. The Centre for Development and Enterprise ( a: ) points out that the evidence
in this regard is mixed. Whereas some studies have found the presence of a reservation wage inhibits employment,
other studies again have found no such evidence. The fact that the special employment projects in South Africa do not
seem to have difficulty in attracting participants might show the absence of a reservation wage, as would the fact that
Statistics South Africa’s surveys show that an overwhelming percentage of persons are unemployed because they have
not been offered any job, not because the wages are inadequate (in Statistics South Africa parlance, because they could
not find “suitable work”).
However, Banerjee et al. ( : ) refer to studies that find that the social pensions being paid in South Africa do result
in people becoming more particular about the job they take. They have also found that the social pensions in South
Africa are quite generous, with the payment amounting to at least twice the per capita income of the country. There is no
comparable country that does the same.
The Centre for Development and Enterprise ( a: ) also refers to another possibility, namely that a sizeable
percentage of unemployed persons might not try to find work in certain enterprises, for instance small businesses. This
is because small businesses generally pay a lower wage for the same job than do larger businesses. The fact that
employees with the same skills and doing the same job receive different salaries because they work in different size
businesses, in different geographic areas or in different sectors, means that so-called “wage cliffs” exist.
Rankin and Roberts ( ), focusing on the youth aged – years, found that reservation wages decrease with age, as
young people become more realistic about what they can expect to earn and the number of available vacancies as they
age. Also, those having previous work experience in small firms have lower reservation wages.
Everything depends on everything else!

E ’
(Authors’ note: after having come across the umpteenth cross-reference, readers might appreciate
this lament.)
SUMMARY
Moderation in labour cost increases seems to be an important element of successful economic and labour market policies
in several other countries.
. . Labour market flexibility and productivity

As discussed in section . , labour market flexibility enhances labour absorption directly. This is
because employers will be more willing to take on additional labour in good times if they know they
can adjust their labour force relatively easily in bad times. In addition, flexibility with regard to the
utilisation of labour, among others, encourages labour-intensive production techniques, whereas
inflexibility encourages higher capital intensity.
Labour flexibility is also important for increased productivity, which in turn could neutralise the
negative impact of high increases in labour costs on employment. The achievement of a high-wage-
high-employment economy is thus possible in an environment where the total productivity
performance of the country is positive. However, total productivity performance in South Africa is
unsatisfactory. Unit labour cost is increasing at a relatively high rate, as pointed out in section . . .
This is damaging the country’s international competitiveness, and thus the job-creating capacity of
the economy.
Employer: “John, I wish you wouldn’t whistle while you’re working.”

Employee: “I wasn’t working, sir, only whistling.”
A
As indicated in section . . , productivity can be improved in many ways. Among the most important
of these is labour market flexibility, but attention should also be given to issues such as: efficient
education and training; making a significant part of remuneration dependent on individual and team
performance; and increasing managerial efficiency and enterprise innovation, through, among other
things, low trade barriers and enhanced product competition. To increase productivity by reducing the
workforce, as has happened in South Africa in recent years, simply increases unemployment in the
country, and the reasons for this phenomenon should be further considered.
The sections referred to above contain important information related to this issue, and should be
consulted as an integral part of the discussion in this section.
SUMMARY
Labour market flexibility and productivity are important to increase the labour absorption capacity of the economy and
allow significant real wage increases without harming competitiveness and employment.
. . Active labour market policies

Active labour market policies (ALMPs) are those policies that aim to improve the operation and results
of labour markets so as to maximise quality employment and minimise unemployment and
underemployment (ILO, : ), for example, employment subsidies, training schemes and public
employment services. In contrast, passive labour market policies (PLMPs) typically involve spending on
income transfers to lessen the financial hardships of the unemployed, without necessarily improving
their skills and employment prospects (Mlatsheni, : ). While unemployment insurance benefits
(an example of PLMP) can help alleviate the hardship of job losses, active labour market policies aim
at preventing unemployment or remedying it by returning displaced workers to productive jobs.
An important element of active labour market policies is job creation initiatives at the local level,
particularly those focusing on activities to meet the needs of the community (e.g. household,
environment, culture, infrastructure and social care). In addition, a special dispensation should be
introduced to make the employment of young workers and high-unemployment groups more
attractive.
In the government’s employment strategies (RSA, : par. . . ; RSA, : ), the introduction of a

social plan (discussed in section . . ), improved housing and transport, and a special focus on youth
unemployment (e.g. through a youth training subsidy) are emphasised. Improved access to labour
market information programmes is also mentioned, where the objective is to ensure that adequate
information regarding the labour market is available to all players in the labour field. This relates to
short- and long-term market conditions, skills requirements, and so on. Other elements could include:
work sharing; rural development programmes and programmes to improve the mobility of workers
(e.g. job placement services) both between occupations and between regions; rescue operations for
weak enterprises and sectors; and broadening the expanded public works programme.
Several activities and programmes that are discussed elsewhere in this chapter can be classified as
active labour market policies. Such policies can normally be classified under three main headings (ILO,
):
Programmes intended to enhance the quantity or quality of the labour supply or to alter its
distribution across occupations or locations (e.g. training and retraining, geographic labour mobility
assistance so that work-seekers can move to areas where there are job vacancies, early retirement
schemes and schemes providing work experience).
Policies that act primarily to influence the structure, composition or level of the demand for labour,
for instance: wage subsidies (see section . . ); promoting self-employment and supporting small
enterprises (section . . ); and public works programmes (see section . . ).
Programmes that can improve labour market processes and institutions, such as: placement
services; skills and aptitude testing; vocational guidance and counselling; provision of job search
skills; labour market information on vacancies and hiring requirements; restructuring
unemployment benefit programmes (see section . . ); placing the long-term unemployed in
short-term jobs to avoid deskilling and the loss of work habits; and promoting labour market
flexibility (discussed more fully in section . ).
It should be noted that in their review of the South African labour market, the authors of the ILO
Review ( : ; ) raise two points of concern with regard to active labour market policies:
They are strongly opposed to the concept because of the directional (in the sense of
interventionist) tone of the phrase and the pejorative tone of the opposite, namely a passive labour
market policy.
They warn that there is limited capacity to implement an interventionist labour market approach in
this country, whether through labour market training, employment services, public works or
subsidised job-creation efforts. They conclude that policy should rather err on the side of the direct
provision of enhanced income security, i.e. direct transfers that reach people with minimal
transaction costs.
There are also some other problems that need to be considered when designing and implementing
active labour market policies:
The first is known as dead weight loss, which is when some action is subsidised, which would have
occurred in any event. Employers could, for instance, receive government subsidies to hire workers
they would have hired anyway, or workers would have moved to a different geographic area even if
removal expenses had not been subsidised.
Another problem is displacement, which occurs when the programme participant displaces
someone who was in a particular job or would have obtained that job. Active labour market policies
could thus redistribute unemployment from targeted groups to non-targeted groups.
A third possible problem is creaming, which occurs when employers employ the best of the target
population, or when the providers of training or job-search assistance provide such services only to
selected individuals in the target group. If, for instance, placement or training services are
subsidised on the basis of the number of people placed in employment, there will be a bias against
applicants who are more difficult to place.
A final danger is that active labour market policies can stigmatise participants. If a particular
programme is targeted at the unskilled or long-term unemployed, graduates of such a programme
may face even greater difficulties in finding regular employment than their peers who have not
participated in the programme.
SUMMARY
Active labour market policies are policies that aim to improve the operation and results of labour markets and include
policies to enrich labour supply, enhance the demand for labour and improve labour market processes.
. . A strong skills base

The availability of a strong skills base is an essential element of any strategy to relieve unemployment.
Efficient education and training not only play a major role in supporting higher economic growth, but
can also contribute to finding a better match between the availability of skills and the demands of the
labour market. Education and training can therefore reduce structural unemployment. Chapter
gives more attention to this aspect.
Training programmes in some countries are increasingly targeting the long-term unemployed and are
aimed at promoting their employability, as evidence has shown that the chances of getting a job fall
significantly with the duration of unemployment (European Commission, : ). Initiatives in the
educational field include adapting education to the needs of working life, and making the links
between education, training and the workplace more direct and effective. These initiatives, as well as
improved training and apprenticeship systems especially for persons who have left school
prematurely, assist in the transition from school to work and have shown some success in addressing
unemployment among young people (European Commission, : ; ).
Other policies emphasised in this regard in the government’s employment strategy (RSA, : par.
. ; RSA, : ) include: a more open immigration policy to expand the supply of high-level skills; a
learning allowance for youths; and improved interface between the labour market and the education
system.
. . Formal and informal small-enterprise sector

In many parts of the world it has been found that small enterprises and the informal sector represent
a significant part of total employment, or have been responsible for a major portion of the total
increase in employment opportunities in recent years. Until recently, the small-enterprise sector in
South Africa had not shown great employment creation potential, as shown in Figure . – the
number of formal sector employees who reported to be working at small firms with fewer than
employees remains stable at about . million, but the number of employees working in large firms
with at least workers rapidly increased from . million in to . million in .
FIGURE . Number of non-agricultural formal sector

employees by firm size
Source: Author’s own calculations using the and Labour Force

Survey September and and Quarterly Labour Force Survey
quarter data
The promotion of the small-enterprise sector is a key element of the government’s strategy for
employment creation and it proposes strategies such as improving access to land, finance and
support services. The strategy highlights linkages between enterprises, an appropriate legal and
regulatory framework and improved access to managerial support and capital as important for the
development of this sector. Tourism is identified as an important sector where smallenterprise
development should be encouraged. Barriers to entry appear to be poor market information, a weak
skills base and poor access to finance. Small-business enterprises require a high degree of flexibility to
ensure their survival. In particular, they are hardest hit by labour market rigidities, as discussed in
section . and by bargaining councils, as discussed in section . .
If someone comes to me and asks, “What are you doing for us small businessmen?”, I’d say, “The only
thing I’m going to do for you is to make you freer to do things for yourselves.”
M T
However, the current state of entrepreneurship in South Africa does not seem very positive. The
Global Entrepreneurship Monitor (GEM) (http://www.gemconsortium.org), in its GEM survey, has
reached the following conclusions about entrepreneurship in South Africa:
South Africa had the lowest entrepreneurial activity of all developing countries surveyed.
Apart from in Mexico, a new business in South Africa has the least chance of surviving its first three
months.
Only . % of South African adults are owner-managers of businesses older than . years, which is
the lowest of all the developing countries.
South African Indian business owners are the most likely to create jobs, and Africans the least likely.
Survivalist businesses are likely to create the fewest jobs, and the more sophisticated the business
and the higher the educational qualification of the owner, the more jobs are likely to be created.
South Africa is ranked th overall (out of countries) in the World Bank Ease of Doing
Business Index (http://www.doingbusiness.org/). In particular, the country is ranked quite low in the
following indicators: getting electricity ( th); enforcing contracts ( th); starting a business ( st);
and trading across borders ( th).
. SPECIAL EMPLOYMENT AND INCOME MAINTENANCE PROGRAMMES
In a developing country such as South Africa, unemployment is likely to be a permanent

phenomenon, although hopefully at much lower levels than is currently the case. There have been
various programmes in various countries over the years to assist people who are unlikely to find a job
in the formal sector, or who may have lost their job and are in the process of finding a new job. These
various programmes are discussed in this section.
Work does more than get us our living; it gets us our life.
H F
. . Public works and special employment programmes

South Africa urgently requires a programme to assist unemployed persons who do not receive
unemployment insurance benefits. The RDP states that a system to provide “hand-outs” to the
unemployed should be avoided (ANC, : ). One method of giving unemployed persons the
opportunity to participate in the economic life of the country is to launch special employment
programmes that are labour intensive, for example public works programmes (PWPs). PWPs have
been introduced with varying levels of success in South Africa and in other countries.
History of public works programmes

In spite of their possible shortcomings, PWPs have been used worldwide, by various countries and
under different circumstances, to correct pressing social and economic problems. It was in
industrialised countries, during the depression years of the s, that PWPs first emerged on a
significant scale as a remedy for cyclical unemployment. The most extensive PWPs were pioneered in
the US, but similar schemes were also launched in other industrialised countries.
In later years their importance declined in such countries, but increased in less developed countries. In
contrast to the position in industrialised countries, where PWPs were used to remedy cyclical
unemployment, they were used as a long-term development strategy to combat structural
unemployment in less developed countries. Special emphasis was also placed on meeting basic needs
and providing community assets through them (Urban Foundation, : ).
The history of PWPs in South Africa started in the s, when the government launched ambitious
initiatives to combat unemployment among unskilled white workers. A special employment-creation
programme was again launched in the mid- s to address unemployment in general. However, the
implementation was very sudden and on an ad hoc basis. This resulted in various shortcomings. The
first of these was that the objectives of projects were not clearly defined. They were initially aimed at
drought relief and the eradication of cyclical unemployment. Thereafter the objective was to provide
relief to the unemployed through temporary employment.
In the late s the objective became longer-term job creation, skills generation and infrastructure
provision. The changing of objectives was done at relatively short notice, which made proper planning
and coordination difficult, if not impossible. The state apparatus for the implementation of projects
was also inappropriate and poorly geared to deal with such programmes. The issue of labour
standards, and specifically the wage rate, caused numerous problems. This employment programme is
discussed in more detail in Barker ( ), Abedian and Standish ( ) and Development Bank of
Southern Africa ( ).
Renewed attention was given to PWPs after the transition to a new government and with the
implementation of the RDP. A number of poverty relief and special employment programmes have
since been announced, focused on labour-intensive public works, water and welfare projects. The
projects include: clearing invading alien trees; labour-intensive construction of infrastructure; funding
to kick-start microenterprises in rural areas; and income and nutritional provision through agricultural
production.
According to the RDP, the key area in which special measures to create jobs can lead to building the
economy and meeting basic needs is in redressing infrastructural disparities (ANC, : ).
Amenities such as water supply, sanitation, clinics and childcare facilities should, according to the RDP,
receive special attention. Furthermore, environmental damage should be repaired. The government’s
employment strategy (RSA, : par. . ) refers to the following:
Clean Cities Campaign (the delivery of waste services to poorly serviced areas)
Working for Water (clearing invasive alien vegetation)
Land Care Campaign (rehabilitation and conservation of natural resources)
Municipal Infrastructure Programme (for low-income areas)
Welfare programmes (which offer training, education and other opportunities for the destitute)
Community-based public works programmes (primarily in rural areas)
In , the Expanded Public Works Programme (EPWP) (http://www.epwp.gov.za), arguably South

Africa’s largest ALMP, was launched and is still currently being implemented. The project hires workers
on a temporary basis either by government, contractors or other non-governmental organisations
under the Ministerial Conditions of Employment or leadership employment conditions. The
participants of the programme receive an allowance for any work they have completed, and the
programme focuses on four sectors, namely infrastructure, nonstate, environmental and culture, and
social. At the time of writing, the programme is in its Phase , with a goal of creating six million work
opportunities by .
Special employment programmes are, however, subject to much controversy among economists (see
sources quoted by Barker, ; Abedian & Standish, ). Attitudes towards such programmes vary
from complete opposition to conditional support.
Arguments against
People opposed to PWPs use, among others, the following arguments:
They can play only a temporary (i.e. cyclical) role and South Africa’s structural unemployment
problem cannot be addressed through such programmes.
If workers do not find a job in a relatively short time after training, this can lead to a higher level of
frustration for the job-seeker and concomitant greater social problems. It can also result in newly
acquired skills being quickly lost.
Such programmes usually involve the denial of resources to some other programme. By launching
such programmes, either government expenditure will be increased or other important
programmes will be neglected. The method of financing such programmes is also important.
Increasing taxes will probably result in the private sector investing and consuming less (Urban
Foundation, : ). Increasing domestic borrowing will raise interest rates and have much the
same effect. Increased local and foreign borrowing will impose the future cost of repayment on the
economy.
Such programmes might conflict with the fiscal and monetary policy required at that specific time
from the point of view of stabilisation.
Arguments in favour
Supporters, on the other hand, argue as follows:
Unemployment can be reduced through the creation of productive labour-absorbing jobs through
labour-intensive approaches.
Programmes can be targeted specifically at long-term unemployed persons, which cannot happen
through general stimulatory measures.
Workers on such programmes can gain marketable skills, life skills or exposure to the world of work,
thereby increasing their employability.
The programmes can be linked to specific development initiatives and might therefore fit in well
with the government’s role in uplifting certain communities, for example by the provision of
infrastructure (Naledi, : ).
The creation of dependence and indigence commonly associated with anti-poverty welfare
programmes is avoided (National Economic Forum (NEF), : ).
In contrast to general stimulatory measures, these programmes usually have a negligible effect on
the balance of payments, owing to the minimal import content of the projects themselves and the
fact that any consumption expenditure by the participants will also have a very small import
content. They will also have only a small effect on inflation because of relatively low wage rates, and
on demandled or supply-push inflation.
The capacity of communities to manage their own affairs can be increased, local government and
other institutions strengthened and sustainable economic development generated.
Two-pronged approach
PWPs to create employment should ideally entail the reorienting of normal public sector
infrastructure provision towards the increased use of labour and the provision of training, as well as
provide opportunities and training for emerging entrepreneurs. Such PWPs can be targeted at
structural unemployment. However, it often takes quite some time to reorient programmes in this
manner.
Short-term, temporary relief measures for the unemployed are therefore also required as a second
prong of PWPs. Programmes for this purpose can be financed from a specially designated fund and
should be phased out over a period of, say, five years (NEF, : ).
Targeting
Special attention should be given to targeting the beneficiaries of a PWP. Communities and sectors
most in need should be targeted. Job programmes should be targeted especially at the long-term
unemployed and single-parent households. Single women with children and the rural unemployed
should be key target groups, as well as the urban youth and residents of informal settlements (NEF,
: ). It is important, however, to emphasise that, given the widespread nature of unemployment
and existing budgetary constraints, a PWP can hope to employ only a fraction of the unemployed at
any one time. The ideal would be for the PWP simply to facilitate the entry of the unemployed into
the formal economy, but this would depend on the performance of the broader economy. Targeting
could also reduce the impact of the dead weight effect, i.e. jobs created under the scheme that would
have been created in any case (ILO Review, : ).
One way to avoid expensive selection methods is to use self-selection techniques. This involves
setting wages at a level at which only those in dire need will participate in the programme. However,
this raises the broader question of the level of labour standards that should be applicable in such
programmes.
Labour standards
One of the most critical issues for the successful implementation of a PWP is labour standards,
especially wages. These will significantly affect the overall cost of a programme, as well as the number
of beneficiaries. It is useful to identify two types of PWP in the discussion of labour standards:
Labour-intensive public works. Labour standards set at a lower level than those for conventional
public works should apply (as outlined below).
Community self-help initiatives. These are small-scale projects that create infrastructure for specific
communities, and where members of that community contribute labour to the project. Very few
labour standards, if any, should apply, as such projects do not normally entail employment
relationships.
No one can worship God or love his neighbour on an empty stomach.

W W
The wage rate to be paid to persons accommodated by PWPs presents special problems (see
COSATU, : , Centre for Development and Enterprise, : ):
A high wage rate will mean that fewer persons can be accommodated, and in addition will divert
many people from other employment opportunities, however inadequate they might be, for
example from subsistence agriculture or the informal sector. In the long term this will aggravate the
unemployment problem. Not only should PWPs not compete with permanent employment
opportunities, but participants should also be encouraged to seek other employment or become
self-employed.
Too low a wage rate on the other hand can result in unemployed persons opting to make a living by
unacceptable methods (e.g. crime) rather than through the PWP. It will also lead to very low
productivity levels on the programmes. If wage rates are too low, the poverty problem is not really
alleviated and the programme may simply be seen as the exploitative use of labour to provide
infrastructure that the state should provide in any event (Barker, : ).
The wages paid in such programmes should at least differentiate between urban and rural areas.
Furthermore, the direct benefits provided to the communities involved in the PWPs should also be
taken into account when determining wages, especially when such benefits entail an income-
generating infrastructure, for example access roads, land reclamation or dams (Urban Foundation,
: ). Wages should also not be considered in isolation, but account should be taken of the
complete benefits package. This package includes the very important benefits of employment
experience and training for the unemployed. Finally, any remuneration system should make provision
for a task-based payment system rather than a fixed daily wage. This is especially important on
labour-intensive projects, as the supervisory element of such projects usually causes many headaches.
In January , the government issued regulations that permit workers to be paid at below the
minimum wage for unskilled workers in the area, and per sector and task completed (Centre for
Development and Enterprise, : ). PWP contractors are also exempted from some of the
conditions required in terms of the BCEA.
From time to time, Statistics South Africa also measures the impact of South Africa’s PWPs. For
instance in the September LFS, only % of the population of working age had even heard of
PWPs. The percentage was the highest in the Eastern Cape, where % of the respondents had heard
of the programme. Of these, only (or %) had been involved in such activities in the six
months prior to the survey. Even though this indicates a relatively low awareness of the programme,
the people involved in the programme would have made a difference to unemployment in
the country – if all these people had been involved in the programme for the full six months, and if
they had all been looking for work before (which are both quite strong assumptions to make), the
PWP would have made a . % difference in the unemployment rate. On the other hand, the results of
the Labour Market Dynamics survey conducted by Statistics South Africa indicate that % of the
EPWP participants and other government job-creation programmes were eventually employed – up
from % in .
SUMMARY
The wages paid to workers participating in PWPs should be set at a level that will discourage even low-income earners
in the rest of the economy from participating in such programmes, because such persons already have a means of
earning a living.
PWPs can, at most, provide temporary relief for the unemployed and should not be seen as a solution to the
unemployment problem.
. . Programmes focused on young people

As discussed earlier in this chapter, the unemployment rate among young persons (ages – years
of age) is particularly high – nearly three-quarters of young persons have never worked before.
There are many reasons for employers being less likely to employ young people. The most important
reason is probably that young people have not yet been properly “tested” in the labour market. They
also do not have the job experience of older workers, which is often more important than a formal
academic qualification. Also, employers are often unwilling to carry the cost of training school leavers
(which could be relatively high given the problems with South Africa’s educational system and the
generally poor output the schools are said to produce – see Chapter ). The employers might spend a
significant amount on training, simply to find that the newly trained person, especially the better
ones, leave soon after training for a better job.
The structure of wage determination is also likely to contribute to high unemployment among young
people. Bargaining councils and other mechanisms for determining wages generally do not make
provision for young, inexperienced workers. Therefore, if an employer is facing a choice of employing
a young, inexperienced worker or a more experienced worker at the same wage rate, that employer
would rather employ the more experienced worker.
The Centre for Development and Enterprise and Business Leadership South Africa ( : ) proposed
a series of measures to encourage the employment of young people, for example:
A combination of tax breaks and year-long exemptions from laws on hiring and firing for employers
employing first-time employees between the ages of and .
Creating special economic zones (SEZs) where businesses are exempted from a number of
legislative measures increasing their costs, and where the employment of young people can be
encouraged.
A much larger and more effective vocational education stream should be started in high schools.
Vocational education should be piloted in the relatively manageable context of some of South
Africa’s fast-growing medium-sized towns, where experience can be gained in linking local
companies with the schools that should be providing their workforce.
Large and simple temporary employment schemes for young people should be launched in the
poorest provinces such as Limpopo and the Eastern Cape.
In , the government also mooted the possibility of a wage subsidy scheme for young people
(Minister of Finance, ), leading to the implementation of the Employment Tax Incentives Bill on
January . The general principles underlying a wage subsidy scheme are discussed in the next
section.
. . Wage subsidies
Given the high unemployment among young persons, Levinsohn ( : ) has pointed out that even
though there is significant turnover in the labour market, it seems that once workers get a job in the
formal sector, they tend to stay employed, albeit not necessarily in the same job. This means that it is
important to get young workers into their first jobs.
The school-to-work transition can therefore be facilitated by encouraging employers to experiment

with younger workers by a once-and-for-all wage subsidy for such workers.
Wage subsidies have been used in several countries as a means of encouraging employers to create
jobs, although these may simply be temporary jobs that will last for as long as the subsidies are made
available. In this sense, the jobs are also dependent on the continued availability of public funding, as
is the case with special employment programmes.
A wage subsidy reduces the price of labour relative to the price of capital or that of non-targeted
workers, and in this way encourages enterprises to substitute targeted workers for unsubsidised
capital or non-targeted workers. It also reduces the cost of hiring, training and dismissing employees,
which could also encourage employment. It therefore reduces the enterprise’s costs, encouraging it to
expand its output and demand for labour. The wage subsidy thus leads to a shift up the firm’s labour
demand curve, in terms of the demand models discussed in Chapter . Specifically, it reduces the cost
of labour for the employer, thus increasing the demand for labour.
However, as also explained in Chapter , the extent to which a wage subsidy increases the
employment of targeted groups will depend on the elasticity (sensitivity) of the demand for such
workers. If a reduced wage does not result in a greatly increased demand for workers (the demand for
labour is thus relatively inelastic), the employment effect of subsidies will be negligible, and may
simply increase the profits of employers. A further general disadvantage of subsidies is that there may
be no long-term positive impact on employment creation, or that it may lead to the inefficient
utilisation of labour, because employers do not carry the full cost of employing the person. In
addition, if the employer has to overcome numerous administrative obstacles in order to access the
subsidy, the participation of employers in such a scheme is likely to be quite low.
There are various types of wage subsidy:
The first is a general subsidy (sometimes also called a stock subsidy), which is simply paid as a
percentage of the enterprise’s total wage bill. Although the aim is to reduce the employer’s wage
bill, it will not necessarily result in increased employment, as it may simply result in increased wage
rates for existing workers or higher profits for the employer. This is called the deadweight loss of
subsidies. It would also be very expensive.
The second type of wage subsidy is paid only in respect of workers about to be retrenched. This is
one type of marginal subsidy. In addition to being very difficult to administer (thus increasing the
administration costs), the disadvantage of subsidies aimed specifically at averting redundancies is
that ailing industries or unsuccessful enterprises may be kept going and the structural adjustments
that are required may not be affected. In this sense subsidies may impede structural change.
Another type of marginal subsidy is the recruitment subsidy, which is paid in respect of new
recruitments. The subsidy may be wasted if “churning” takes place, for example if enterprises
dismiss existing employees and replace them with subsidised workers (called the substitution effect).
However, the likelihood of this happening in South Africa is very slight due to the strict dismissal
rules in this country. Another possibility is that firms that receive employment subsidies might be
able to produce at lower prices than firms that are unable to receive subsidies. The former firms will
then potentially crowd out firms not getting subsidies (the displacement effect).
This problem may be partly solved by making the subsidy incremental, i.e. only on the basis of
workers being appointed to new, additional jobs created in the enterprise’s labour force.
Incremental subsidies have the disadvantage that the funds may be wasted if paid to enterprises
that would have expanded employment anyway (this is called the deadweight loss).
Subsidies could also be targeted, i.e. limited to certain areas or certain sectors (e.g. sectors
experiencing economic woes) or aimed at certain unemployed groups such as youths or the long-
term unemployed. The big advantage of introducing targeted subsidies is that it creates
opportunities for unemployed persons with no work experience to gain such experience, which
makes such persons more employable. However, if targeted subsidies are not also incremental (only
paid in respect of a net increase in an employer’s employment), it might lead to the substitution
effect referred to above, namely, that subsidised employees might replace non-subsidised
employees. However, if the subsidies are properly targeted (for instance, only paid in respect of
persons who have been unemployed for longer than a certain minimum period of time, or for
workers with no experience), such substitution is unlikely to take place because of the reluctance of
employers to employ the longterm unemployed or workers with no experience – employers are
concerned about their productivity in the working environment. The payment of a subsidy might
equalise the value for the employer of inexperienced and experienced workers, thereby having a
net positive impact on the operation of the labour market. However, there is the opposite danger
that targeted workers might, in fact, be stigmatised by them being eligible for a subsidy and this
might increase, rather than reduce, the difficulty of their finding a job.
The various potential disadvantages of subsidies can therefore be summarised as follows (National
Treasury, : ):
There might be a deadweight loss if the subsidy is paid without it resulting in an increase in
employment, but instead it simply leads to higher profits or higher wages for workers in
employment. This is more likely to happen if the subsidy is a general subsidy rather than a targeted
subsidy.
The substitution effect takes place when workers who are not eligible for the subsidy are replaced
by targeted workers.
There might be a displacement effect if firms that receive the subsidies crowd out firms that do not
receive subsidies. Employment in the subsidised firms then simply displaces employment in the
non-subsidised firms.
There is also the danger that a subsidy targeted at school leavers might induce some persons to
leave school for a subsidised job, or might discourage further tertiary study by that person. With
the high rate of unemployment in South Africa, it is unlikely that this will take place on a large scale,
because the number of school leavers being employed due to the subsidy would be relatively
limited.
Finally, there is the danger that targeted workers might be stigmatised by a subsidy, and this might
increase the reluctance of employers to employ such workers. In this sense, the subsidy might
achieve the opposite of what was intended. However, if the targeted group is big enough, for
instance all school leavers, there is unlikely to be a stigma effect.
However, a properly designed subsidy system can overcome most of these disadvantages. A perfect
system should also not be the aim, because as can be imagined, the more complicated the subsidy,
the more difficult to administer and the more opportunities for employers to abuse the system or for
various types of fraud to take place. The infrastructure required to administer such a complicated
system could be very extensive, which pushes up the costs of the subsidy. Furthermore, should the
targeting result in unfair competition, with the unsubsidised employer losing markets to the
subsidised employer, the former may reduce employment and there will thus be no net increase in
employment overall.
However, the biggest concern about a wage subsidy is that it simply delays critical adjustments in
South Africa’s labour regulatory regime. As discussed in Chapter , there is significant concern that
the country’s labour legislation, or at least the manner in which it is implemented, is having a negative
impact on the willingness of employers to increase employment.
In general, unemployment in South Africa is so high that there is only a small likelihood of a person
with little or no experience finding a job. A targeted wage subsidy is only likely to encourage
employers to take a risk by employing inexperienced persons if the payment of a subsidy is directly
tied to revised rules for dismissal for those receiving the subsidy. For this reason, it is essential that the
targeted wage subsidy entail a probationary period during which a “noquestions-asked” dismissal
policy is in effect (Levinsohn, : ).
There is concern that this might result in employers terminating the employment of subsidised
employees just before the end of their probationary period. Employers might replace targeted
workers at the end of their subsidised period of employment with newly subsidised workers. Firstly, it
is not clear whether the worker is worse off than he or she would have been without any subsidy
linked to probationary employment, because the person did receive some work experience, whereas
in the absence of such policies, the person would probably still be unemployed. Secondly, if the
worker has been able to prove himself or herself during the period of employment, the employer is
unlikely to replace a newly employed productive worker with a person with questionable productivity.
On the other hand, if the worker has not proven himself or herself in the work situation, the person
might still suffer from a lack of proper training or motivation, or might be unsuitable for the particular
position.
Employment Tax Incentives Bill

The Employment Tax Incentives Bill (ETIB) became effective in , targeting the youth aged –
years earning up to R per month working at registered companies. It is a firm-side subsidy
provided to the employer for two years. In the first year that the employer hires a qualifying
employee, the monthly subsidy amount paid to the employer is:
% of the monthly remuneration of the qualifying employee, if the remuneration is R or less

R if the remuneration is more than R but less than R
R – % × (remuneration – R ) if the remuneration is more than R but less than R
.
The subsidy payable to the employer is calculated in the same way in the second year, except that the
amount halves.
At the time of writing, Ranchhod and Finn ( ) is the only study empirically estimating the impact of
the ETIB on youth employment, and the authors find no significant evidence that it has a positive
impact on youth employment likelihood. Figure . also shows similar findings, as youth employment
only increased marginally from . million to . million between the first quarter of and third
quarter of . There is also no indication of a rapid increase in the youth labour force participation
rate and employment rate, while the youth unemployment rate actually increased from % to %.
FIGURE . Labour market indicators of the youth aged – years, Q – Q
Source: Author’s own calculations using the – Quarterly Labour Force Survey data
SUMMARY
If a wage subsidy system is applied, it should be targeted at those who are difficult to employ, and should be incremental
to encourage the expansion of job opportunities. Such an approach is likely to reduce or eliminate many of the potential
disadvantages of an employment subsidy.
. . The social plan

A “social plan” refers to policies designed to deal with the negative social implications associated with
economic restructuring and, more specifically, large-scale job losses (Barker & Holtzhausen, : ).
Interventions can be specially aimed at alleviating the consequences of largescale redundancies,
which occur relatively rapidly and which are concentrated in particular geographical locations. They
are not intended to arrest the necessary process of restructuring. The types of intervention normally
initiated as part of a social plan are: training and retraining; job searches and counselling for affected
workers; assistance in securing alternative employment; public works programmes; small enterprise
promotion; and the design of appropriate social security nets.
The Labour Market Commission ( : ) recommended the establishment of a “social plan fund” to
support programmes and strategies negotiated between employers and workers facing structural
employment decline. The social plan fund should be deployed only to augment social funds
established at plant, sector, regional or local level for the purpose of finding alternative employment
or income-earning opportunities for those rendered redundant.
The social plan guidelines

The Presidential Jobs Summit in October gave rise to a social plan that applies when an
organisation or industry is in trouble or in the event of threatening large-scale retrenchments
(Department of Labour, a). It makes provision for three phases in dealing with retrenchments:
Phase : Saving jobs. The social plan encourages employers and workers to be proactively involved in
preventing job losses. It promotes the establishment of future forums, which can analyse problems in
an organisation and assist in identifying and implementing solutions. A future forum is simply a
strategic association between management, workers and their representatives. A future forum may
also request technical support from the National Productivity Institute (Productivity SA), including the
analysis of problems, facilitating access to government assistance programmes, providing information
on trends in various industries and recommending solutions.
Phase : Managing retrenchments. This phase is triggered when an organisation contemplates large-
scale retrenchments. It involves the Department of Labour, which may open a job advice centre on, or
close to, the premises of the organisation. The services provided by such centres can involve the
following:
Counselling (e.g. how to cope, managing money matters and job-hunting skills)
Placement in a vacant position or assistance with starting a business
Assistance with obtaining unemployment insurance benefits
Discussing the training needs of those who have been retrenched
Assistance with retraining
Labour relations
Phase : Creating jobs for local economies. Provincial and local government may launch initiatives to
develop opportunities in the local or provincial economy, especially in areas that have been negatively
affected by large-scale retrenchments. Financial assistance may also be obtained from the central
government to assist with such activities (Department of Labour, a).
In the mining industry, companies will be unable to obtain a mining licence unless they are able to
present a full social and labour plan relevant to the particular mine, which, among other things,
outlines the plan of action when the mine reaches the end of its life.
. . Unemployment insurance
Most countries provide some system of income maintenance for persons experiencing a period of
unemployment. In South Africa, the Unemployment Insurance Fund (UIF) provides shortterm relief to
contributors when they become unemployed or are unable to work because of illness or maternity, or
the adoption of small children, and also provides relief to dependants of deceased contributors. The
UIF is an insurance scheme in the full sense of the word and not social assistance (see Strydom, :
for a description of terms).
Persons excluded from the UIF include some categories of civil servants, persons working less than
hours per month, and persons in learnerships. Persons who have never contributed to the UIF, such as
new entrants to the labour market, and persons whose benefit period has run out, do not enjoy
benefits in terms of the Unemployment Insurance Act. Domestic workers have also been covered
since .
Benefits are funded from a payroll levy of . % on employers, and a levy of . % on the employee’s
income. Also, from October , the maximum earnings ceiling is R per month. Therefore the
maximum contribution that can be deducted for workers who earn more than R is R . per
month (R × %).
An unemployed worker who has contributed to the UIF is entitled to a benefit that varies according to
income, from % of previous remuneration for highly paid workers to % for the lowest paid
workers. To qualify, the worker and his or her employer must have contributed to the UIF for at least
weeks in the months before he or she became unemployed. Statistics and other particulars
regarding the UIF are contained in the annual reports of the Unemployment Insurance Commissioner.
The positive influence of unemployment insurance. Unemployment insurance obviously entails a

benefit to persons experiencing hardship owing to unemployment. It is also beneficial for society in
that the costs of unemployment, whether direct (e.g. higher taxes, welfare contributions or family
support) or indirect (e.g. increased unrest, crime), are reduced. It may also be argued that labour
market mobility and flexibility are increased, because unemployment benefits reduce the pressure on
employers to keep redundant workers in employment, and on unions to protect jobs. Any production
adjustments required can therefore be effected more readily. (This flexibility is, however, reduced
through statutory limitations on dismissals.)
The negative influence of unemployment insurance. Most observers tend to agree that labour market
flexibility is, in fact, reduced by unemployment benefits, because they can increase the wage demands
of unemployed workers – the so-called “reservation wage” at which they will accept employment.
Fearn ( : ) states that “a substantial body of literature finds that unemployment compensation
has significant effects on the level and/or duration of unemployment in the United States”.
(Unemployment benefits in the US amount to about % of the wage that had been earned.)
Unemployment insurance benefits increase the minimum wage at which an unemployed worker will
accept employment, because he or she can “afford” to look for a job that suits him or her better, i.e.
one that pays a higher wage. Fearn adds that the availability of unemployment insurance may lower
the weekly probability of a person taking a job by as much as to %. If, however, these longer
unemployment spells mean that the job-matching process is improved, then it can have an overall
positive economic effect because of reduced job turnover and increased productivity. As already
stated, unemployment insurance is very necessary from the point of view of social responsibility
towards persons who have lost their jobs.
Restructuring unemployment insurance policy is an important element of the so-called “active labour
market policies” that many countries have introduced. It could entail programmes such as the
following:
Actively encouraging the unemployed into training or into a job, even if it is temporary public
employment or a subsidised private sector job, as a precondition to receiving unemployment
insurance benefits
Allowing persons to capitalise their stream of unemployment insurance benefits to use as start-up
capital for self-employment
Providing “workfare”, in which the unemployed are obliged to take low-paying jobs or places in
labour market training schemes under threat of losing entitlement to unemployment benefits
Ensuring that such programmes do not discourage job-seekers from taking up employment
Linking benefit payments with recipients actively seeking employment
Subsidising recipients in new jobs by paying a certain sum to the employer
Financing retraining programmes
Only paying benefits for a limited time period, after which benefits are available only if the person
participates in some labour market programme such as training.
SUMMARY
The Unemployment Insurance Fund is an insurance scheme that provides temporary income maintenance for persons
with previous work experience who have lost their jobs.
. . Basic income grant

The Committee of Inquiry into a Comprehensive System of Social Security (the Taylor Committee,
) recommended the introduction of a basic income grant (BIG) to alleviate absolute levels of
poverty.
At the time of investigation, the Committee argued that at least million people in South Africa fell
below the poverty line. On average, they survived on less than R per month (equivalent to about
R in prices). Many of the country’s poorest households fell through the social security net
because they did not receive UIF, or qualify for a state old-age pension, a disability grant or a child
maintenance grant.
The introduction of a national basic income grant was therefore seen as a method to provide all
households with a minimum level of income to enable the nation’s poorest households to meet their
basic needs more adequately. The Committee recommended that the grant be paid to all individuals
irrespective of employment status and not be subject to a means test. It was argued that this would
diminish the administrative burden, the disincentive to work and opportunities for corruption that are
often associated with means-tested grants.
The Taylor Report recommended the gradual introduction of comprehensive income support. As
current institutional arrangements did not favour the introduction of a basic income grant, it
recommended a two-phased process. In the first phase, the child support grant was to be extended to
all children up to the age of years over the period – (this only happened in / , as
found by Beukes, Jansen, Moses & Yu, ). In the second phase, the income support grant was to be
extended to everyone over the period – .
Even though the Committee did not recommend a level at which the grant should be set, it based its
analysis of the impact of such a grant on R (about R in prices) per month for all South
Africans. This level reflected the level proposed by a community-based grouping, which included
COSATU. This grouping also indicated that the grant should be inflation indexed.
The additional cost of a R monthly grant was estimated to be R billion, but the net cost of the
grant after tax offsets would be roughly R billion annually, according to the proponents’
calculations. This was because most of the cost of a basic income grant would be recovered through
progressive taxation, according to the Committee. New measures to expand the fiscus, including the
restructuring of the government employee pension fund and a modest increase in deficit spending,
were suggested to allow for the financing of the remaining costs. However, even if the R billion was
a good estimate of the costs involved, it would require an increase in VAT from to %.
There are reports that the claims as to the net cost of BIG have not been properly researched. Even
the original research documentation on which the Commiittee based its findings did not indicate what
research had been done to come to these conclusions. The Minister of Finance also stated that the
cost could be as high as R million and could bankrupt the fiscus (see Barker, : ).
Possible benefits of a basic income grant
Poverty alleviation. A R grant would double the amount available for consumption by people in
the poorest % of the population, and would promote improved nutrition especially for children.
There are two distinct classes of men … those who pay taxes and those who receive and live upon
taxes.
T P
Response to the HIV and AIDS pandemic. The current social assistance system is insufficient to
absorb the additional burden that households affected by HIV and AIDS have to carry. The BIG
would fill this gap and give HIV-affected households additional resources to help them to cope.
Stimulus to increased consumer spending, job creation, investment and economic growth. Cash
transfers into households increase and stabilise demand, consumption and savings. Spending
would likely be concentrated on basic, locally produced commodities, thus benefiting local markets
and stimulating job creation. Increased consumption would be likely to have a particular impact on
rural areas where it has the potential to kick-start the economy.
Possible disadvantages of a basic income grant
Corruption. The nature of a benefit paid to all persons is such that there would be many more
possibilities for corruption than with most other social security schemes. People could try to claim
twice, might claim on behalf of people who have died or are non-existent, or the administration
itself might exploit the numerous opportunities for corruption. Administratively it might be easier to
run than a scheme that has a means test, but would remain very difficult to manage.
Regional mobility of labour. Although the grant would be restricted to South Africans, the
introduction of this scheme would be likely to increase the illegal flow of poverty-stricken people
from neighbouring countries. These people are often able to obtain residence documents from
corrupt officials, thereby increasing the cost of the scheme. Even good administration would not
eliminate a proportion of such people from benefiting from the scheme.
Funding the scheme. Taking into account the probability of high administration costs and the influx
of labour from neighbouring countries, the proponents of the scheme seem to have seriously
underestimated its cost. High tax rates and increased deficits to fund it would have a negative effect
on the competitiveness and creditworthiness of South Africa in a highly globalised world, and might
also increase the outflow of skilled people from the country. The country would gain unskilled
poverty-stricken people and lose highly skilled people. Even with the country’s social grant system,
which has a far smaller coverage than a basic income grant to everyone, there are only . million
individual taxpayers, whereas there were . million people drawing social grant benefits in .
SUMMARY
The basic income grant (BIG) has been proposed in some quarters as a method of providing a basic income to the
poverty stricken. However, the cost of BIG is likely to be high.
KEY CONCEPTS
active labour market policy (ALMP)

ASGI-SA
basic income grant
cyclical or demand-deficient unemployment
discouraged work-seekers
employment tax incentives bill (ETIB)
frictional unemployment
GEAR
jobless growth
narrow (strict) and broad (expanded) definitions of unemployment
natural rate of unemployment
NDP
NGP
official definition of unemployment
passive labour market policy (PLMP)
public works programmes
RDP
seasonal unemployment
skills mismatch
social plan
structural unemployment
unemployment
unemployment insurance
unemployment rate
wage subsidies
FOR STUDENTS
. Why should South Africa regard the unemployment problem as its most important social priority?
. Describe how unemployment is generally defined and compare this definition with the two definitions used in
the Quarterly Labour Force Survey.
. Define the four types of unemployment, with the aid of an example in each case.
. Discuss the general characteristics of unemployment in South Africa.
. Use the information from the following table to calculate and compare the narrow (strict) and broad
(expanded) unemployment rates of a hypothetical country between and .
2016 2017
Employed 5 500 000 6 250 000
Unemployed 2 250 000 2 500 000
Discouraged work-seekers 1 750 000 2 250 000
Inactive 3 500 000 3 000 000
Working-age population 13 000 000 14 000 000
. Discuss the main causes of unemployment in South Africa.

. Explain, with the aid of a figure, the derivation of the short-run and long-run Phillips curves.
. Define active and passive labour market policies and discuss their most important elements, with specific
reference to South Africa.
. Show, by means of a graphic illustration, why a wage subsidy could increase employment, and use the graphs
to show under what circumstances more jobs would be created and under what other circumstances fewer
jobs would be created.
. Discuss the pros and cons of unemployment insurance, and South Africa’s system of unemployment
insurance.
SUGGESTED READING*
Banerjee, A., Galiani, S., Levinsohn, J., McLaren, Z. & Woolard, I. . Why has unemployment risen in the new
South Africa? Economics of Transition, ( ).
Bhorat, H. . Unemployment in South Africa: Descriptors and determinants. A paper presented at the Fourth
IZA/World Bank Conference on Employment and Development, Bonn, Germany.
Kingdon, G.G. & Knight, J. . Unemployment in South Africa, – : causes, problems and policies.
Journal of African Economies, ( ).
International Labour Organisation. . Active labour market policies in a wider policy context. Geneva:
International Labour Organisation.
Mahadea, D. Employment and growth in South Africa – hope or despair? South African Journal of Economics, ( ):
March.
National Treasury. . Confronting youth unemployment: policy options for South Africa. Pretoria: National
Treasury, February.
Ranchhod, V. & Finn, A. . Estimating the short run effects of South Africa’s employment tax incentives on
youth employment probabilities using a difference-in-differences approach. South African Journal of
Economics, ( ): June.
Special Economic Zones (SEZs) have recently been renamed Industrial Development Zones (IDZs)
Human capital and the demand for skilled workers

. Introduction
. Human capital theory
. Some criticisms of human capital theory
. Human capital in South Africa
. Education in South Africa
. Skills development in South Africa
. Skilled labour shortages
Give a man a fish and you feed him for one day. Teach a man to fish and you feed him for a lifetime.
C
. INTRODUCTION
The ILO is unambiguous in stating ( b: ) that education, together with learning and training, are
essential prerequisites for gainful participation in the new economy and thus for economic growth.
Education cannot be leapfrogged in the development or growth process. This is probably even more
important as the economy becomes more technologically orientated. The ILO states: “It is probable
that inadequacies in education and training have always been brakes on economic growth; they have
arguably become more so in the digital age” ( b: ). In the new economy, work is increasingly
knowledge based, and those who are knowledge enabled hold the greatest advantage in the
networking economy (Strydom, : ).
Work by the University of Stellenbosch’s Department of Economics points to the fact that South
Africa’s acceleration in economic growth was made possible by improved use of black human
resources (Van der Berg, ). However, the authors point out that the growth of welleducated
workers has fallen behind the growth of the economy and that in order to reduce income inequality,
higher black wages will need to be built on improved education and skills. Without this, labour market
interventions to improve equity will remain counterproductive.
The position with regard to the skilled and unskilled labour forces in South Africa is very similar to
that in other developing countries: there is an oversupply of unskilled labour force (unemployment)
and a shortage of skilled labour force. The supply of skilled work-seekers in South Africa did, however,
to some extent keep up with demand because of low economic growth in the s and s and
through net immigration gains at least up to . With South Africa having moved to higher levels of
economic growth since (before slow growth or even recession began in ), the shortage of
skills has become much more pronounced and is likely to become the most important factor
inhibiting higher long-term economic growth.
South Africa’s poor showing in the utilisation and development of its human resources is already
damaging its international competitiveness. The World Competitiveness Report has ranked South
Africa relatively low on the competitiveness scale as a result of, among others, this factor (see
Appendix A).
Another factor that will also affect the future availability of skills is HIV and AIDS (see also section
. ). The epidemic will, among others, have an impact on the availability of educators, and
furthermore result in the personal and economic returns from years of investment in education being
cut off before any returns have accrued to the individual, family or society (Simkins, ).
This chapter addresses some of the issues relating to human capital and skills development, while the
aspect of discrimination and employment equity will be dealt with in Chapter . As an introduction to
the discussion there is an explanation of human capital theory and its shortcomings.
. HUMAN CAPITAL THEORY
People with higher educational qualifications consistently earn higher salaries. The Quarterly
Labour Force Survey earnings data, for instance, indicate that the education premium is particularly
high in finance and electricity. The mean hourly wage of workers without Matric is only about % of
that of their more educated colleagues.
The reason why a person with a higher educational qualification earns a higher salary is that skills
development enhances a person’s stock of human capital and therefore increases that person’s
productive potential. The person is more valuable to the employer and this, in turn, leads to higher
earnings. There is thus a progression from education and training, to higher productivity, to higher
earnings. This can be illustrated by paths A and B as indicated in Figure . .
FIGURE . Human capital theory

However, education and training can also provide immediate benefits (such as subjective satisfaction
and status) in addition to long-term monetary rewards through higher earnings.
The question now is whether there is any reason why a person would not educate himself or herself to
the highest possible level. The answer is that there is a cost involved in improving education. This cost
consists of the following (Ehrenberg & Smith, : ):
Direct cost or out-of-pocket cost: this includes the cost of tuition, books and related expenses.
Indirect cost or foregone earnings: this arises because during the time the individual spends on
improving his or her education, that person has decided to study instead of working and earning a
wage.
Psychic losses: this occurs as learning is often tedious and difficult (since it is not easy to measure
this cost in monetary terms, psychic losses will not be taken into consideration for the human
capital theory from this point onwards).
The decision to invest in education thus depends on the monetary return on education and training.
According to human capital theory, this is the net return, which is arrived at by taking into account
both the costs and the benefits of education and training. It is thus a cost-benefit analysis of
education. Most individuals do not necessarily calculate the sums directly, but similar factors come
into play when an individual decides what to do after school – to go to university or to start working.
If you think education is expensive, try ignorance.

R B. L
Calculating rates of return

The total direct and indirect cost of education should thus be compared with the increased earnings
of the individual over the working life of that individual, in order to determine the “profitability” or
rate of return of an investment in education and training.
This is reflected in Figure . . If the individual (A) starts working immediately after school (at age ),
he or she immediately earns a salary, and his or her earnings profile is HH. Another individual (B)
decides to go to university, which entails a direct and indirect cost. However, upon entering the labour
market three or four years later (it is four years in the case of Figure . – the person starts his or her
studies at a university at years and graduates when he or she is years), that individual can
immediately enjoy a higher earnings level than individual A. Individual B’s earnings profile will thus be
line CC. Individual B’s earnings profile (CC) is steeper than individual A’s (HH) from years until
retirement at years – the gap between CC and HH becomes greater as time goes by – because
individual B is associated with a greater likelihood of promotion and being involved in skilled work
with higher pay due to his higher educational attainment.
FIGURE . Potential earnings stream of a Matriculant compared with a graduate
Source: McConnell et al. ( : )
If the area of the incremental earnings (area ) is greater than the total cost (the sum of areas and ),
then the investment in education will have been worthwhile from an economic point of view.
However, for various reasons, future money has less value than present money, for example because
present money can be invested and will be worth more in future. Therefore, the costs and benefits
should be translated to some common point in time, the most obvious being the present value. The
net present value of future costs and earnings is calculated by using certain econometric techniques to
discount the future values to the present. The result will be the rate of return (“profitability”) of
education (see McConnell et al., : ).
Private and social rates of return

It should be noted that the rate of return could be a private rate, in which only the benefits and costs
to the individual would be taken into account. However, and even more importantly from a policy
point of view, it could also be the public or social rate of return, in which case the costs and benefits of
education and training from a societal point of view would be considered. The costs to calculate the
social rate of return would be all public subsidies to education, whereas for the individual these costs
would be excluded simply because the individual does not carry these costs. Similarly for the
individual, the after-tax incremental earnings would be calculated, whereas for society as a whole it
would be before taxes. The benefits for society as a whole would include societal objectives such as
economic growth, reducing unemployment, eliminating poverty or reducing inequalities. However,
the benefits could also include the potentially lower crime rates and lower law enforcement costs due
to lower unemployment, and more intergenerational benefits because the children of educated
parents would more naturally be inclined to have a high educational standard and have a better home
environment in which to pursue educational objectives.
The distinction between the private and social rates of return is important because society may decide
that the benefits of education are so important to achieve societal objectives (the social rate of return
is relatively high) that society should allocate more resources to education than an individual would,
based on the individual rate of return.
The rate of return can be calculated for any educational level and will differ over time, depending on
demand and supply conditions in the labour market (which influence future earnings).
Predictions of the theory

Four general predictions can be made regarding the demand for university education, based on
Figure . (Ehrenberg & Smith, : – ):
Forward-looking people are more likely to go to university, as they weigh future events (higher
earnings profile upon graduating from a university) heavily, compared with present-oriented
people.
Most university students are young, as they prefer to have a longer remaining work life ahead.
University attendance increases if the direct and indirect costs decrease. That is, individuals are
likely to go to universities if they receive bursaries or subsidies (meaning lower direct cost) and/or
realise that they would not earn too much in the labour market if they only have Grade
certificates (meaning lower indirect cost).
University attendance increases if the gap between the earnings of university graduates and high
school graduates widens. If this does not happen, the individuals become less motivated to spend
three or four years to pursue tertiary education.
Research evidence
One of the foremost exponents of human capital theory, Gary Becker ( ), found that college
graduates in the US earned a private rate of return of about to %. Other estimates have put the
private rate of return to college education at between . and % (McConnell et al., : ).
However, private rates as high as % for primary education in certain developing countries and as
low as minus % to plus % for certain types of graduate education in the US have been reported.
Surprisingly, social rates of return have been found to be quite comparable to the private rates of
return (McConnell et al., : ).
Keswell and Poswell ( : ) found that the marginal rate of return in South Africa is “extremely
high” for tertiary levels of education and small (approaching zero) for lower levels of education. What
was striking was that the rate of return remained more or less constant at very low rates for the first
few years of schooling, up until approximately years, whereafter it increased substantially. One
could thus invest in education for years with practically zero return, or invest or more years and
receive a high return.
In respect of South Africa, Gustafsson and Mabogoane ( ) found an average increase in earnings
of around % for every additional year of schooling possessed in the range of two to eleven years of
schooling, and a large increase of around % associated with the difference between eleven and
twelve years of schooling, in other words with having attained Grade . Branson, Garlick, Lam and
Leibbrandt ( ) examined returns on education in , and , and found that returns on
Matric and post-Matric education increased over time but returns on education levels below Matric
remained constant. However, both studies relate only to the private rate of return of education and
thus suffer from the shortcoming that the public cost of education is not taken into account. It
appears that no recent analysis of the public rate of return has been done for South Africa.
Another way in which the human capital theory has been applied is to estimate to what extent
economic growth can be ascribed to an improvement in human capital, i.e. the quality of labour.
Denison estimated that % of the increase in output in the US economy from to could be
credited to improvements in the quality of labour (quoted by Marshall et al., : ).
Fedderke ( ) has shown that even though human capital is important in explaining economic
growth in South Africa, it is only very specific types of investment in human capital that contribute
positively to productivity growth. It is thus the production of quality human capital that is important,
not simply the production of human capital. However, both the quantity and quality of human capital
raise political instability. Since instability lowers output, an increase in human capital may at the same
time both increase and reduce output, which is why studies about the impact of human capital in
economies in transition have shown inconclusive results.
It’s a shame to waste a college education on a freshman who already knows everything.
A
The most controversial aspect of human capital theory is that it has been used to motivate and justify
massive investment in education (Marshall et al., : ). If the theory is valid, such investment on
its own will result in sharp improvements in the earnings and therefore the standards of living of
workers, without causing inflation, because of higher productivity on the part of those workers. The
theory also concludes that income inequality in society could be reduced by ensuring a more equal
distribution of human capital, by concentrating education and training efforts on the poorer and
disadvantaged sections of society. However, these conclusions have been severely criticised, and this
is discussed in the next section.
SUMMARY
The rate of return (or “profitability”) of education can be calculated by comparing the present value of the benefits (e.g.
the higher earnings of people with higher educational qualifications) with the costs of education (e.g. the cost of tuition
and foregone earnings).
The social rate of return takes into account the broader costs of education, e.g. all educational subsidies and broader
societal objectives, such as lower unemployment among people with a higher qualification, lower poverty levels and
higher economic growth.
. SOME CRITICISMS OF HUMAN CAPITAL THEORY
The various criticisms that have been levelled against human capital theory are briefly discussed in
this section.
. . Measurement
To calculate the effect education and training have on productivity, the education and training as well
as the productivity of the individual have to be determined. However, this in itself presents a problem.
How is the “amount” of education and training measured? Years of schooling and even degrees
obtained are not necessarily an indication of the quality of education and training (Marshall et al.,
: ). Even though there are measures that could potentially be used to measure the quality of
education, for instance, teacher qualifications and teacher-to-pupil ratios, these are not necessarily
related to the quality of education.
Furthermore, it is in many cases practically impossible to measure the productivity of an individual

worker, and this needs to be done to determine the impact of education on productivity. How is the
productivity of a managing director measured, or that of a teacher?
In many studies the progression from education and training to productivity to earnings, i.e. paths A
and B in Figure . , was simply ignored because of measurement problems, and a simple link
between education and earnings was proved, i.e. path C. This link could, however, simply indicate a
bias in favour of university graduates on the part of employers, which is not necessarily justified by
considerations of productivity. Measurement and calculation problems are also one of the main
reasons why Hosking ( ) doubts the value of this approach for South Africa.
. . Ability or educational level?

Other factors besides education influence earnings, and here it is especially important to consider the
influence of ability. Different educational qualifications may, in fact, be an indication of pre-existing
differences in ability (Marshall et al., : ). Workers with the necessary abilities attain higher
educational qualifications and it is therefore ability, not educational qualifications, which is rewarded
with higher earnings. If this is indeed the case, then sharp increases in educational spending will not
necessarily lead to increases in productivity and, consequently, earnings. Attempts to quantify ability
biases have, for instance, used data on twins to control for ability differences. Others have used test
scores such as IQ as a measure of cognitive ability (Keswell & Poswell, : ).
A young man, having just received his degree from university, rushed out saying, “Here I am world; I
have a BA.”
The world replied: “Sit down, son, and I’ll teach you the rest of the alphabet.”
A
. . The screening hypothesis

In terms of the screening hypothesis, employers apply educational attainment as an inexpensive
screening device to determine the “ability” of workers and thus their job placement. Workers of a
higher educational level are regarded, rightly or wrongly, as having more ability. This leads to their
being appointed or promoted to better jobs relatively soon, which leads to higher earnings. However,
higher education levels might not be related to higher ability, and the link between education and
productivity is again short-circuited via line C in Figure . . Although it makes sense for the individual
to obtain an educational qualification (because it leads to higher earnings), the social rate of return of
education may be overstated.
. . Family background and financial status

Family background and financial status can lead to both higher educational qualifications and also
higher earnings, without there necessarily being a direct relationship between education and
earnings. Family background and financial status can, for instance, facilitate entry to better
educational facilities and to certain occupations, for example in the business world, which might
otherwise have been much more difficult to access. At the same time, this background and financial
status can make the attainment of educational qualifications possible and more probable.
. . Dual labour market theory

A serious challenge to human capital theory, and one that is based on theoretical premises, is the dual
or segmented labour market theory described in section . . (Marshall et al., : ). According
to this theory, workers in the internal labour market segment enjoy good jobs, high earnings, high
upward mobility and job security. Jobs “on the outside” are characterised by low wages, little job
security and limited upward mobility (Fearn, : ). Internal promotions and similar restrictive
mechanisms normally deter access to the internal labour market.
This theory’s criticism of the human capital approach is that wages are not necessarily dependent on
education and training or even productivity, but on the operation of the internal labour market. The
wages of the workers on the outside, in the secondary labour market, remain low irrespective of
educational attainments, because access to the internal labour market and high wages is severely
restricted, and because there is little incentive for employers to utilise workers better.
Earnings inequalities cannot therefore be adequately reduced by investment in education and

training. Attention should rather be given to the integration of the secondary labour market into the
primary labour market.
. . The radical approach

There is another view of the working of labour markets and income inequalities called the radical
approach (Marshall et al., : ). According to this approach, neither education and training
programmes nor the integration programmes of the dualists will significantly alter relative income
structures in the labour market. It is in the interests of capitalists to have different classes of workers,
and they will ensure that this structure is maintained. The government is the instrument of the
capitalist class and will not address the problem. According to this approach, inequalities can be
changed only if the power relations in society are changed, i.e. the power of the capitalist class should
be eliminated.
The radical approach, however, can also be criticised, because class structures, inequality and so forth
are found in all economic systems and are not unique to capitalism. In addition, industrialisation in
various economic systems has also resulted in income and hierarchical structures. Efficient production
methods often demand that work follows a routine and that jobs be ordered hierarchically.
Due to these and other criticisms of the human capital approach, many economists and others have
objected to large spending on education and training that is simply directed at those areas with the
highest rate of return. According to these critics, this is not necessarily in the interests of society as a
whole, and will not necessarily reduce income inequalities unless attention is also given to other
reasons for these inequalities. In spite of the criticisms of human capital theory, there is little doubt
that the neglect of education and training in South Africa in the past, combined with apartheid
policies, did result in severe income inequalities between Africans and whites, as indicated in section
. . .
SUMMARY
Human capital theory has been used to justify huge investments in education and training, but it has also been severely
criticised. One such criticism is that there may be other factors that result in individuals with higher educational
qualifications also earning higher salaries, without there necessarily being a direct link between qualifications and
earnings. These factors include, among others, the innate ability of the individual, the family background that gives
access to better jobs, or employers using education as a screening tool.
. HUMAN CAPITAL IN SOUTH AFRICA
. . Educational level of the labour force

Figure . gives an indication of the educational level of the EAP in , and . The source
of the statistics is not the same ( and from the population census and from the QLFS),
but was the last census with complete coverage of the whole of current South Africa and is
comparable in that respect. It is apparent that the proportion of the labour force with no or very little
education has substantially decreased, from about % having less than a Grade qualification in
to % having no education or only primary-level schooling (which is up to and including Grade
) in . Literacy levels have also improved substantially (see accompanying box for a discussion of
the literacy rate). The percentage of the labour force with a Grade qualification or a degree or
diploma increased from % in to % in .
Literacy
Illiteracy is usually defined as an educational level representing less than seven years of formal schooling (i.e. less than
a Grade ) or its equivalent (Schindler, : ).
According to the QLFS data of Statistics South Africa, the literacy rate in was %. There were . million adults
(age years and older) considered to be illiterate in the sense of having less than seven years of schooling. The
literacy rate of % was a substantial improvement on the rate of % according to the census.
The literacy rate among Africans was lower than the overall rate at %, and among whites very high at %. Greater
access to education in recent years has resulted in younger people having higher literacy levels – % among those
aged – compared to only % among those aged more than years.
The rate also varies greatly between provinces, with Gauteng and the Western Cape the highest at % and %
respectively, and the Northern Cape the lowest at just above %.
FIGURE . Educational level of the labour force

Sources: : Statistics South Africa, ; & : Author’s own
calculations using the census and Quarterly Labour Force
Survey quarter data
Illiteracy is bad, but it’s not as bad as being able to read all the daily news.
J B
What is disconcerting is the high number of unemployed graduates. In there were

graduates without work and a further with a diploma or certificate. These included
persons with teaching qualifications and persons with commerce qualifications.
Further particulars on the educational qualifications of different population groups are provided in
section . . .
SUMMARY
The educational level of the labour force has improved substantially over the last two decades, but there are still large
numbers of illiterate people in the country, with an illiteracy rate of about a quarter of the adult population.
. . Occupational structure of the labour force

The classification of occupations
Occupations are usually classified according to the type of work done, although in certain cases
attention is also paid to the industrial sector in which an occupation is usually exercised, as well as to
the training required for a specific occupation. For example, a clerk can be classified as a filing clerk or
a stock clerk, referring to the type of work done. Artisans, however, are classified by industrial sector
(e.g. motor mechanic), but are also distinguished from apprentices or other occupations by the
training undertaken. A chartered accountant and an engineer, on the other hand, are usually classified
on the basis of their educational qualifications.
Qualifications are not always a good indicator of skills level as the classification should be based on
work done and not necessarily on qualifications. People may also have acquired their skills in a non-
formal way and in these instances experience will be a more important factor than qualifications.
Occupational structure of workers in employment

Table . shows the occupational structure of people in employment in the formal sector and
informal sector. In both sectors, the largest component of the employed consists of the elementary
occupations. The table also shows that nearly % of formal-sector workers are involved in skilled
occupations, compared to less than % for informal-sector workers. On the contrary, more than half
of the informal-sector workers are involved in unskilled occupations, compared to only about % in
the case of formal-sector workers.
TABLE . Occupation of workers in the formal and informal sectors, and
Formal sector Informal sector

2006 2016 2006 2016
1 000s % 1 000s % 1 000s % 1 000s %
Managers 799 8.9 1 178 10.0 70 1.9 173 4.3
Professionals 579 6.5 811 6.9 18 0.5 23 0.6
Technicians 1 112 12.4 1 334 11.3 112 3.0 131 3.2
Skilled 2 490 27.8 3 323 28.2 200 5.4 327 8.1
Clerks 1 217 13.6 1 577 13.4 22 0.6 73 1.8
Service and sales workers 1 250 13.9 1 890 16.0 375 10.1 582 14.5
Skilled agricultural workers 74 0.8 54 0.5 348 9.4 18 0.5
Craft and related trades 1 228 13.7 1 297 11.0 668 18.0 648 16.1
Operators and assemblers 982 11.0 1 080 9.2 127 3.4 232 5.8
Formal sector Informal sector
2006 2016 2006 2016
1 000s % 1 000s % 1 000s % 1 000s %
Semi-skilled 4 751 53.0 5 898 50.0 1 539 41.5 1 553 38.6
Elementary occupations 1 720 19.2 2 572 21.8 1 083 29.2 1 120 27.8
Domestic workers 0 0.0 4 0.0 882 23.8 1 027 25.5
Unskilled 1 720 19.2 2 576 21.8 1 966 53.1 2 147 53.3
Total 8 963 100.0 11 797 100.0 3 705 100.0 4 028 100.0
Source: Author’s own calculations using the Labour Force Survey September and Quarterly Labour Force Survey
quarter data
Note: Domestic workers and agricultural workers are included; employed people with unspecified occupations are not shown
separately in this table but are included as part of the total.
Work is accomplished by those employees who have not yet reached their level of incompetence.
L J. P
. . Skills levels of workers in formal employment

Focusing on permanently employed formal sector employees (the informal sector employees are not
looked at, as a very low proportion of these employees enjoy permanent employment), Table .
indicates that approximately % of them are involved in skilled occupations, with one-third of them
working as managers in .
TABLE . Skills levels of permanent employees in the formal sector, and
2006 2016
1 000s % 1 000s %
Skilled – total 1 878 29.1 2 447 31.7
Skilled – managers 501 7.8 772 10.0
Skilled – professionals 458 7.1 643 8.3
Skilled – technicians 919 14.2 1 032 13.4
Semi-skilled 3 527 54.7 4 007 51.9
Unskilled 1 046 16.2 1 262 16.4
Total 6 451 100.0 7 716 100.0
quarter data
. EDUCATION IN SOUTH AFRICA

. . Educational shortcomings in South Africa
Formal education is categorised into three levels, as indicated in Table . in section . below. The
General Education and Training (GET) band incorporates a reception year and learners up to Grade ,
as well as an equivalent adult basic education and training (ABET) qualification. The Further Education
and Training (FET) band comprises Grades – in school education, out-of-school youth and adult
learners. Technical, youth and community colleges, as well as a range of other industry-based and
non-formal providers also fall into the FET band. The Higher Education (HE) band incorporates a
range of national diplomas and certificates up to and including doctoral degrees.
The South African education system accommodates more than . million learners and students. The
system encompasses established public and registered independent educational institutions,
and educators and lecturers (Department of Education, ).
It appears, however, as if the increase in school and tertiary enrolments has recently levelled off as a
result of falling fertility rates and an increase in infant mortality owing to HIV and AIDS. The average
annual rate of increase in school enrolments between and was . %, and this rate fell to
just . % between and (Simkins, ), and . % between and (Department of
Education, ).
Despite great progress in education in South Africa (especially with regard to access to resources),
“the real goal is equity of educational outcomes as a means to equity of social outcomes” (Van der Berg,
).
A man should be educated enough to know that education alone is not enough.
A
Low progression of learners

When one compares the number of learners who progress from one grade to the next the following
year, there has been an improvement in the early grades. This indicates that in the lower grades, the
dropout rate of learners is relatively low. However, this is not the case in the middle or higher grades.
In the higher grades in particular, the flow-through rate has in fact declined quite sharply. This
suggests that there is probably gatekeeping (not allowing learners who are unlikely to pass Matric to
progress to Grade ) or “culling” of learners (removal of over-age learners, or discouraging them
from registering for the senior certificate examination).
Figure . shows the proportion of African adults aged – years who completed each schooling
grade, and the results indicate that the proportion completing Grade increased from a mere . %
in to . % in , but the latter proportion is still much lower when compared to the white
adults (increasing from . % in to . % in ). Hence, far too many learners, particularly the
Africans, do not make it to Grade , which points to a significant wastage of resources. Bot ( : )
calculates that this wastage could amount to % based on the number of learner years required to
get one learner to Grade .
FIGURE . Proportion (%) of African adults aged – years completing each schooling grade
Source: Author’s own calculations using the censuses from , , , , , and , and Community
Survey and data
Shortage of quality teachers

There has been an improvement in the number of the learners per teacher, and there has also been a
more equal distribution of learner-to-teacher ratios between provinces. In , this ratio ranged
between and across the provinces, and this improved to a more equal range of between and
in (Bot, : ; Motshekga, ).
The number of underqualified educators dropped from in to in , or from

to % of the total and has further dropped to . % of the total (eNCA, ). Whereas there are still
significant differences between provinces, Bot ( : ) points out that the provinces with the best
qualified teachers do not necessarily have the highest pass rates.
There is grave concern about the supply of teachers in the foreseeable future (see, for instance,
Crouch, as quoted by Simkins, ). The supply of new teachers appeared to be excessive at one
stage ( students in ), but when a number of teacher training colleges were closed in the
s, no account was taken of the increasing mortality among teachers owing to AIDS. It is estimated
that about teachers leave the profession each year through death, disease or resignation. To
this number should be added the additional teachers required owing to increased numbers of
learners and the need to reduce learner-to-teacher ratios. The total number of new teachers required
each year ranges from to (Van Broekhuizen, ), whereas the total teacher training
capacity is only per year (Bot, : ).
Further cause for concern is the fact that the number of African students training to be teachers,
especially in the Foundation (mother-tongue) Phase, has “collapsed”. Data (although incomplete)
show that less than % of students training to be teachers in were African (Financial Mail,
January ).
Some factors that have been responsible for this state of affairs are the following:
The reduction in the number of training institutions

Severe restructuring in the profession and the consequent large-scale retrenchments damaging the
image of the profession as a long-term career
The impact of HIV and AIDS
Better career and salary opportunities for skilled African workers elsewhere
Racism, violence and other manifestations of antisocial values or behaviour in some schools
discouraging teachers from entering the profession
The lack of bursaries and/or loans
Backlogs in infrastructure
Considerable progress has been made since in decreasing backlogs. For instance, overcrowding
has reduced with a decline in the average number of learners to a classroom from (in ) to
(in ). The number of schools without water, electricity or telephones has also reduced quite
substantially.
However, by , % of schools still did not have water and % did not have electricity (Bot, :
). There was also a shortage of classrooms, but this had been reduced from about in to
in . There was also a severe lack of libraries, with only one in five schools having a library.
The results from the Southern and East Africa Consortium for Monitoring Educational Quality
(SACMEQ) data also indicate that . % of Grade learners were in schools with buildings in good
condition – up from . % in . Availability of electricity to learners increased from . % in
to . % in , yet this proportion was relatively low with regard to Eastern Cape learners – . %
in and . % in (Department of Basic Education, ).
Disparity in incomes and high poverty levels have a negative effect in that family life in many African
urban areas has been seriously damaged and inadequate living conditions and dietary deficiencies
affect learning ability and learning opportunities.
Poor quality in spite of increased educational expenditure

An analysis of the reading ability of South African learners in comparison with those from other
countries clearly indicates the poor quality of South African education. An international comparison of
reading achievement in primary schools is regularly undertaken by the International Association for
the Evaluation of Educational Achievement (IEA). The survey on Progress in International
Reading Literacy Study (PIRLS) was conducted in countries.
The worst-performing participant was South Africa, with an average score of . Even though South
Africa is classified as an upper middle-income country, this score is far below the average of even the
lower middle-income countries, which was . Furthermore, South Africa had the highest proportion
of students ( . %) failing to reach the Low International Benchmark (Taylor & Yu, ).
Taylor and Yu’s ( : ) conclusion is that the South African school system is unlikely to make a
significant contribution to social mobility, and that it may, in fact, be acting as a constraint on South
Africa’s economic development. Meaningful change will require well-targeted, innovative solutions
and a strong political commitment sustained over a long period of time.
Education is the largest functional area of government expenditure and the estimated total spending
was about R billion in / . This is more than social protection (R billion) and health (R
billion). South Africa’s expenditure on education as a proportion of total government expenditure has
been very stable at about % in recent years.
South Africa’s expenditure on education as a percentage of GDP ( . %), is higher than that of other
countries (World Bank, ):
Australia 5.3%
Germany 4.9%
India 4.1%
Ireland 5.7%
Japan 3.8%
Russia 4.2%
Switzerland 5.0%
Thailand 4.0%
United Kingdom 5.7%
US 5.2%
After , the government addressed the earlier discrimination in the allocation of resources by
reallocating resources. The disparities between the provinces were also addressed as part of this
initiative – whereas the per-learner expenditure ranged between and % of the national average
in , this improved to a range of between and % of the national average (Bot, : ). This
shift in fiscal incidence was the most dominant resource shift in the overall budget (Van der Berg, as
quoted by Strydom, : ), and mainly took the form of more teachers and higher salaries for
teachers in historically black schools, coupled with cutbacks in teacher numbers in other schools.
However, real resource shifts in learner-to-teacher ratios were not substantial, with the result that
educational outcomes did not improve substantially. Therefore, even though there was more equity in
educational resource allocation, there was not much improvement in the inequities with regard to
educational outcomes (Strydom, : ; Van der Berg, ).
Fedderke ( ) has shown that even though human capital is important in explaining economic
growth in South Africa, it is only very specific types of investment in human capital that contribute
positively to productivity growth. As stated earlier, it is thus the production of quality human capital
that is important, not simply the production of human capital. After , the South African
government was very intent on widening access to education, and the deepening of quality was of
secondary importance. Fedderke points out that attention needs to turn to the provision of quality
education. If this is not done, the quality of education is likely to be an important inhibiting factor in
innovation and economic growth in South Africa. To quote Strydom ( : ): “The South African
education system does not shape up effectively to the demands of the twenty-first century.”
Poor quality mathematics and science education

Mathematics and science education is a requirement for many tertiary qualifications. If the skills level
of South African workers is to be improved, the country needs to produce substantial numbers of
learners with quality mathematics and science education. However, the opposite is happening (CDE,
a: ).
The numbers of learners writing the National Senior Certificate examinations in mathematics and
mathematical literacy were and in , with both numbers dropping by about
from the previous year. What is alarming is that the proportions of these learners achieving % and
above were only . % and . %. There were only learners writing the physical science exam
in , with only . % achieving at least %.
In addition, fewer matriculants are enrolling for teacher qualification courses in these subjects, and the
entry of newly qualified mathematics and science teachers is not even keeping up with retirements
and resignations.
When mathematics and science teaching is compared with that in other countries, the situation in
South Africa appears dismal, as the Trends in International Mathematics and Science Study (TIMSS)
(https://timssandpirls.bc.edu) conducted in countries in found that South African learners
performed second worst of all the countries in both Grade and Grade mathematics, and was the
worst-performing country in Grade science (South Africa did not take part in Grade science).
Countries such as Botswana, Egypt and Morocco all did better than South Africa.
Many of these countries performing better than South Africa have also suffered from social problems
such as multiplicity of languages, massive income differentials and recent histories of conflict. These
factors therefore cannot be used as an excuse for the poor performance of the country.
Some of the reasons offered by the CDE ( a: ) for South Africa’s poor performance are the
following:
South Africa’s school year is substantially shorter than that of other countries.
South Africa has higher absentee rates for both educators and learners.
The time devoted to mathematics and science education is the least.
Significantly more time is spent on repetition and homework.
The government has tended to be overambitious with new policy changes, but at the same time
there has been a discontinuity in policy.
Consider maths and science to be the launching pad of life.
M S
Problems with tertiary education

Table . shows that qualifications in some of the humanities (such as languages, social sciences and
the like) declined between and . Over the same period, there was a sharp increase in the
number of qualifications awarded in business, commerce and management sciences, as well as
computer sciences and communication/media studies.
TABLE . Tertiary qualifications awarded by field of study, and
Field of study 1996 2012 Percentage change (%) per annum 1996 to 2012
Agriculture and related sciences 1 215 2 650 5.0
Architecture and environmental design 1 322 2 182 3.2
Business, commerce and management 13 913 35 043 5.9
Communication, journalism and related studies 768 2 980 8.8
Computer science and data processing 1 697 4 854 6.8
Education 19 005 23 185 1.3
Engineering 5 110 9 974 4.3
Family ecology and consumer sciences 728 452 –2.9
Health professions and clinical sciences 6 772 7 612 0.7
Language, linguistics and literature 5 722 2 224 –5.7
Law 5 097 4 767 –0.4
Life and physical sciences 3 577 6 366 3.7
Mathematics and statistics 1 010 1 496 2.5
Philosophy, religion and theology 1 681 587 –6.4
Psychology 4 020 3 421 –1.0
Public management and services 5 896 4 785 –1.3
Social sciences and studies 6 467 5 051 –1.5
Visual and performing arts 1 290 2 763 4.9
Total 85 989 120 397 2.1
Source: DHET, http://www.dhet.gov.za, Table . for all institutions to nd order CESM, accessed July , supplied by the
Institute for Race Relations
It would appear that the National Plan for Higher Education, which uses planning and financial
incentives to shift the balance of enrolments in universities between (a) the humanities, (b) business
and commerce and (c) science, engineering and technology, is bearing fruit.
Growth in university exemptions has slowed down dramatically, from % per annum from – ,
to only . % per annum from – (Van der Berg, ). This potentially holds back economic
growth and black social mobility.
The impact of HIV and AIDS on the education system

HIV and AIDS is likely to have a severe effect on education. Firstly, the expenditure of households and
the government alike will probably be redirected from education and training to medicine and health,
or for basic necessities where the breadwinner has died because of AIDS. Mortality among
breadwinners and other parents will weaken families, which will damage the acquisition of human
capital by children. Children are likely to be absent from school more often to tend the sick. There will
also be many more orphans who will have to fend for themselves, with the result that they will never
attend school and will not receive material or emotional support to attend school. Such persons
probably face a life of unemployment and perhaps crime because they will be largely untrainable. The
expensive investment in education will also be lost since a substantial number of secondary school
learners – perhaps in the order of – will be HIV positive and likely to die before they are able
to apply their knowledge in the work situation properly (Simkins, ). The personal and economic
returns from years of investment in schooling and higher education will be cut off before any returns
are able to accrue to the individual, family or society (Simkins, ).
It was estimated (see Institute for Futures Research, b: ) that in more than half the teachers
being trained would replace AIDS victims. By , this increased to nearly %. The country will thus
have to train many more teachers than is the case currently.
As pointed out above, the growth in school enrolments has dropped quite sharply since , and an
absolute drop occurred in the period – . Although this has been partly due to lower fertility
rates, increased infant mortality owing to HIV and AIDS has also been a significant contributing factor
(Simkins, ).
. . Starting to address the educational shortcomings

The Centre for Development and Enterprise ( ) has found that even though schooling reform is
difficult, measurable improvement can be achieved in as little as six years.
The quality of teaching and teachers is a central determinant of student performance. Teacher quality
cannot be reduced to formal qualifications, which often have little impact on student results. Teacher
absence and a significant loss of instructional time are key impediments to learner performance.
School leadership, notably by principals, plays a key role, especially in motivating teachers and
creating a culture of learning. The rights of principals to hire and fire, and of their superiors to fire
principals if necessary, should be reinforced.
The key to improving education in Africa is to strengthen accountability. In the case of schools, three
levels of interventions are needed to achieve this: information; school-based management; and
teacher incentives. Schooling reform in Africa is most effective when it starts from the ground up, and
empowers those who are closest to learners, namely, parents and communities.
Sustained schooling reform requires a new approach to the teaching profession. Society needs to
value the importance of teachers more highly, and teachers need to see themselves as professionals
and behave accordingly. The CDE makes the critical point that incentive-based pay is in this respect
essential.
What is required is a strategy that places school leadership and effective, professional teaching at the
heart of educational reform. The performance of school managers and teachers, in turn, should be
judged in terms of improved learner performance.
SUMMARY
There are numerous reasons for the educational problems that are experienced in South Africa, among them: legacies of
past ideological divisions; unequal distribution of resources; inadequate resources; unfavourable educational environment
(unstable family life, inadequate living conditions, no job prospects); and a failure to supply the skills needed by the South
African economy.
. SKILLS DEVELOPMENT IN SOUTH AFRICA
. . Skills development system and structures

South Africa’s national skills framework consists of the National Skills Authority (NSA), the South
African Qualifications Authority (SAQA) and a number of Sectoral Education and Training Authorities
(SETAs). Workplaces are linked to the sectoral level by the requirement to submit workplace skills
plans to the SETAs and, in turn, the SETAs need to develop sector skills plans, which feed into the
national level.
The NSA advises the Minister of Labour on policies and strategies related to skills development and
reports to him or her on progress made in their implementation. To ensure that its objectives related
to skills development are realised, the NSA has suggested a number of “success indicators”, for
example: the number of completed learnerships; the number of people in learnerships who find jobs;
the percentage of workers who progress up the National Qualifications Framework; and the
percentage of enterprises that claim skills development grants.
The SAQA is appointed by the Minister of Education in consultation with the Minister of Labour
(website: http://www.saqa.org.za). SAQA is responsible for:
establishing a single unified system of education and training qualifications in the country by
overseeing and implementing the National Qualifications Framework (NQF)
generating unit standards and qualifications for registration on the NQF
creating and accrediting the institutions necessary to ensure that these qualifications are of a high
quality, which means that the quality of training provided by training providers has to be
monitored.
National Qualifications Framework

The NQF is based on the principles of access, i.e. everyone in South Africa should have access to
learning of quality and portability so that the skills learnt are transferable across the different sectors
of the economy.
The NQF is a national register of education and training unit standards and qualifications. It is a
means by which such qualifications (a learning outcome) will be recognised nationally, whether
achieved by full-time, part-time or distance education or training, whether formally (in an education
or training institution) or informally (at work). It provides the basis for achieving an integrated
approach to education and training, and for facilitating mobility and progression in education,
training and career paths. This helps to create an integrated system that encourages lifelong learning.
SAQA administers the NQF.
The NQF consists of levels along a continuum of learning, as indicated in Table . . Each level
represents a point at which national qualifications are awarded. The levels measure how difficult the
learning for different qualifications is and allow for comparisons between courses. For example, Level
of the NQF comes at the end of ordinary compulsory schooling up to Grade , but can also be
reached through adult basic education and training (ABET) for adults who did not have the
opportunity to study at school. Level includes doctoral and postdoctoral research.
TABLE . National Qualifications Framework structure
NQF Bands Grade Types of qualification and certificate

Level
10 Higher Education and Training Doctoral degree
band
9 Master’s degree
8 Honours degree, postgraduate diploma or professional
qualifications
7 Bachelor’s degree and advanced diplomas
6 Diploma and advanced certificates
5 Higher certificates and advanced national (vocational)
certificates
4 Further Education and Training 12 Grade 12 or National Vocational Certificate level 4
band
3 11 Grade 11 or National Vocational Certificate level 3
2 10 Grade 10 or National Vocational Certificate level 2
1 General Education and Training 8–9 Senior Phase ABET Level 4
band
6–7 Intermediate Phase ABET Level 3
1–5 Foundational Phase ABET Level 2
0 Preschool/Early Childhood Development ABET Level 1
A qualification is made up of different standards. The generation and adoption of the qualifications
and standards are undertaken by various bodies established for this purpose. These are firstly
Standards Generating Bodies (SGBs) where everyone with a direct interest in a standard gets together
to agree what the learning outcome should be. Once they have agreed, the qualifications and
standards are sent to the National Standards Body (NSB), which makes sure that all standards and
qualifications fit into the NQF.
Framework for quality assurance

SAQA accredits bodies to monitor the quality of training given by providers, according to the
standards and qualifications of the NQF. These bodies are called Education and Training Quality
Assurers (ETQAs). The ETQAs accredit education and training providers. They also check the courses or
providers and accredit them if they can show that they have the staff and systems needed to support
successful learning. Quality assurance includes ensuring that there are assessors who can see that the
standards are met even when they have been learnt outside the course. The ETQA also awards
certificates to learners who achieve these standards and qualifications.
Learnerships
A learnership agreement is a training agreement entered into by a learner, an employer and possibly
a training provider to train the learner to a set of agreed standards in a particular field. A learnership
is a para-professional and vocational education and training programme that incorporates, but also
builds and improves on, traditional apprenticeships. It is a mechanism through which qualifications
that are registered on the NQF can be achieved. Like apprenticeships, learnerships combine practical
work experience (in normal workplace conditions) and structured learning (which may be taught by an
educational institution). Learnerships will assist young unemployed people to enter employment, as
well as help existing workers to improve their skills levels. A learner, an employer or a group of
employers and an accredited training provider or a group of training providers enter into a
learnership agreement for a specified period. The agreement confers specific responsibilities on each
party to the agreement, and an allowance is paid to the learner.
Don’t learn the tricks of the trade. Learn the trade.

A
However, there are also differences between learnerships and apprenticeships, for example:
Learnerships apply to all parts of the economy. They go beyond “blue-collar” trades and also
prepare people for jobs in the new services sector and for higher para-professional occupations.
Learnerships fit into the NQF. They give the learner a qualification registered by SAQA. They cover
more levels than the old apprenticeships so that people who qualify are able to move on later to
professional and other qualifications.
Sector Education and Training Authorities

One of the most important vehicles for implementing the skills development strategy is the Sector
Education and Training Authority (SETA) system. SETAs have been created in different sectors of the
economy ( in total) by representatives from employers and employees in that sector, as well as
government departments. They are concerned with education and training and their job is to help
implement the national skills development strategy and to increase the skills of people in their sector.
They may establish chambers to ensure that the special needs within a particular industry are met.
In order to achieve these objectives, SETAs have, among others, the following duties:
To develop and implement a sector skills plan

To develop and administer learnerships
To undertake quality assurance
To disburse levies collected from employers in their sector
A sector skills plan is a plan to describe the trends in each sector, the supply of and demand for skills,
and to identify priorities for skills development. It therefore provides the framework and background
for the SETA’s activities. The sector skills plans are presented to the NSA and approved by the Minister
of Labour.
The experience to date has shown that the SETA system is mostly dysfunctional, as many beneficiaries
wait for lengthy periods to receive funding or do not receive the funding at all. A survey by the
South African Chamber of Commerce and Industry (SACCI) found that the cash balances for the
/ financial year, i.e. unpaid grants, of SETAs accumulated to over R . billion, an increase of
just below a billion rand from the previous year.
Funding of skills development

Skills development in South Africa is funded through a compulsory skills development levy, in terms
of the Skills Development Levies Act. Every employer with an annual payroll in excess of R
pays a skills levy of % of total payroll to the South African Revenue Service. Of the funds collected,
% goes to the National Skills Fund (NSF). Each SETA may get the remaining % of the money paid
by the employers in its sector. The SETA uses a certain percentage of this funding (not more than %)
to cover its administrative costs. The rest of the money can be paid back as grants to companies that
comply with criteria that are mainly nationally set and to a lesser degree determined by the SETA in
terms of its sector skills plan. Such criteria are, for instance, the submission of a workplace skills plan
and the subsequent implementation report on the training provided.
The National Skills Fund

The NSF receives % of all the levies paid by employers. This fund is used to train unemployed and
other vulnerable people in areas where there is a strong potential for growth and employment. This
includes training people for jobs created by new investment or the expansion of existing enterprises,
as well as jobs in the construction of new infrastructure. The funds can also be used for youth
development, development of small, micro and medium enterprises (SMMEs) and learnerships of
national priority.
. . Joint Initiative on Priority Skills Acquisition

The Joint Initiative on Priority Skills Acquisition (JIPSA) was launched as a three-year project in March
to identify priority skills in all sectors of society. It seeks to develop skills, from boosting primary
school education to improving higher education learning. It plans to recruit retired experts and expert
mentors in certain fields. It aims to recruit those South Africans who have gone abroad, to import
scarce skills, establish a skills database and make sure that underutilised training facilities are used.
Skills that JIPSA has identified as having priority include those of artisans, technicians, managers,
biotechnology professionals, information and communication professionals as well as skills to extend
basic public services.
A number of challenges have been identified:
Finding a way to bring the over people who enter the labour market without a Matric into
the training system
Finding jobs for, or retraining, unemployed graduates
Developing mid-level skills (craft and artisan skills)
Establishing learnerships for unemployed learners
Defining the role of the National Skills Fund
Encouraging mathematics, science and IT in schools
Revitalising career guidance in schools
Clarifying terminology – such as scarce and critical skills – to ensure that misunderstandings do not
hamper progress
Even if we study to old age we shall not finish learning.

A
. . The extent of in-service training activities in South Africa

Structured learning programmes
The first phase of the – National Skills Development Strategy (NSDS) set a number of skills
targets to be achieved through the SETAs (South African Institute for Race Relations, : ).
One of these was to ensure that % of all employees of SETA-affiliated employers had an NQF level
qualification – equivalent to Grade – by March . This was equivalent to . million workers. At
the time the goals were established, . million or % of the . million SETA-affiliated workers
already had such a qualification. By April , . million or % had such a qualification. SETAs had
therefore contributed to workers gaining such a qualification in / (see Table . ).
A second target was for % or . million employees to enrol in a structured training programme and
for % (or ) to complete such a programme by March (South African Institute for Race
Relations, : ). By April , . million employees had embarked on structured training, of
whom . million had completed their programmes – more than double the number targeted to
enrol and three times the number targeted to complete the programme (see Table . ).
TABLE . Workers enrolled in and completing structured learning programmes
Workers completing NQF level 1 Structured learning programmes

Workers participating Workers completing
2001/02 5 079 760 916 570 615
2002/03 111 367 1 398 461 330 192
2003/04 433 437 3 041 753 2 147 040
2004/05 43 597 n/a 3 112 461
NQF target 904 000 1 398 033 699 017
Sources: Department of Labour ( b); Department of Labour ( )
In the second phase of the – NSDS (Department of Labour, ), five broad objectives were
set:
1. Prioritising and communicating critical skills for sustainable growth, development and equity: this
is achieved via the training of sector specialists, skills development facilitators and career guidance
councillors. The target number of people trained was achieved in / , / , / and
/ (the / results are not available yet).
2. Promoting and accelerating quality training for all in the workplace: the Department of Labour set
a goal that by March , %, % and % of large, medium and small firms, respectively,
would be assisted through skills development. All three targets were achieved before .
3. Promoting employability and sustainability livelihoods through skills development: the
Department set numerous goals for March : ( ) at least unemployed are trained (this
number reached in / ); ( ) at least non-levy paying firms, non-
governmental organisations (NGOs) and community-based cooperatives are supported by skills
development (this number exceeded in / ); ( ) at least unemployed have
participated in ABET (this number was in / ).
4. Assisting designated groups to participate in accredited work, integrated learning and work-based
programmes to acquire critical skills to enter the labour market and for self-employment: the
Department hoped to achieve the following goals by March :( ) unemployed people
successfully complete the programmes (this was achieved in / ); ( ) at least young
people are trained and mentored to form sustainable new ventures (this was also achieved in
/ ).
5. Improving the quality and relevance of training and learning provision, by ensuring that
institutions integral to the implementation of the strategy perform to or beyond a minimum set
standard. This objective also focuses on the need to improve institutional capacity to develop and
implement learning programmes of increased relevance to the individuals’ needs as well as the
economy’s development.
Skills development levy

Employers are required to pay % of their wage bill as a skills development levy to the South African
Revenue Service. Such a levy may be partly reclaimed if the employer has satisfied certain of its
industry SETA requirements regarding the training of its employees.
For large employers (employing more than employees) the NSDS set the target of % of levy-
paying employers claiming back part of their levy as grants (South African Institute for Race Relations,
: ). By March , this target had almost been achieved (see Table . ).
TABLE . Payment and claiming of skills development levy

Paying levy Claiming levy % claiming the levy
A. Large employers (employing more than 150 workers)
2002/03 3 589 2 591 72.2
2003/04 4 461 2 989 67.0
B. Employers (employing between 50 and 150 workers)
2002/03 7 397 3 699 50.0
2003/04 7 882 4 177 53.0
C. Small employers (employing fewer than 50 workers)
2002/03 135 328 13 162 9.7
2003/04 181 842 17 511 9.6
Source: Department of Labour, b
A smaller percentage of medium-sized employers (about half) also claimed back their levy. This was
much higher than the NSDS target of %.
The percentage of smaller employers reclaiming their levy was less than %, which was much lower
than the NSDS target of %.
Learnerships
As far as learnerships are concerned, the percentage of active learnerships increased sharply after
/ , which shows that workers are being trained by way of such learnerships.
At the Growth and Development Summit (GDS) in , SETAs committed themselves to enrolling
approximately learners in learnership agreements by March . The NSDS set the goal of
learners being enrolled by March . By March the GDS target had been exceeded (see
Table . ), and this number has exceeded since / . Certain SETAs did better than
others; the Services SETA exceeded its GDS commitment by almost %, while other SETAs such as
the Public Services SETA had only enrolled learners despite a GDS commitment of (South
African Institute for Race Relations, : ).
TABLE . Progress with learnerships
Number of learnerships Active Workers in Unemployed persons in

registered learnerships learnerships learnerships
2002/03 478 136 n/a 26 493
2003/04 666 335 28 529 69 306
2004/05 n/a n/a 45 813 84 429
2005/06 n/a n/a 70 362 n/a
2006/07 n/a n/a 58 462 n/a
2007/08 n/a n/a 110 870 n/a
2008/09 n/a n/a 103 296 n/a
Sources: Department of Labour ( b; b; )

Overall state of training
The LFS included some questions with regard to in-service training. According to this survey,
more than % of those in employment did not receive any form of in-service training. Thus, only %
of workers have work-specific training (South African Institute for Race Relations, : ).
According to a survey undertaken in in the Greater Johannesburg Metropolitan Area, % of the

firms surveyed spent nothing on in-house training, and % indicated that they did not spend any
money on outside training (Bhorat & Lundall, : ). Thus, a fairly significant portion of the
sample indicated that no resources were dedicated to internal or outside training. At the median,
firms were spending the equivalent of % of total labour costs on training. This low level is probably
due to a relatively high proportion of firms not spending any money on training. The mean spending
is also relevant, and indicated that . % of total labour costs was spent on training. The proportion
spent on training by small firms was significantly lower than that of medium-sized firms, which also
spent relatively more than large firms.
Skills development targets in terms of the Broad-based Black Economic Empowerment Codes
The BBBEE Codes (Department of Trade and Industry, ) require that black employees in
learnerships should comprise at least % of the total number of employees. Since learnerships are
more or less full-time, enterprises would have to withdraw % of their employees from productive
employment to achieve the learnership objectives of the code (and would probably have to replace
them with temporary employees). This is likely to have an impact on productivity and labour costs. In
addition, learnerships are not the only formal training programmes – there are also internships, cadet
schemes and apprenticeships, which are not catered for in the codes.
SUMMARY
The National Qualifications Framework (NQF) is a national single unified system of education and training
qualifications and is administered by the South African Qualifications Authority (SAQA).
Learnerships are vocational education and training programmes that incorporate but also build and improve on
traditional apprenticeships.
Sector Education and Training Authorities (SETAs) have been created in different sectors of the economy, and are
responsible for skills development in their sectors.
Skills development in South Africa is funded through a compulsory skills development levy.
. SKILLED LABOUR SHORTAGES
The International Panel on ASGI-SA (Hausmann, : ) makes the valuable point that all empirical
studies of labour demand show that highly skilled and low-skilled workers are strongly
complementary, not substitutes. Coffee and tea are substitutes, while coffee and sugar are
complements. The implication of complementarity is that the greater the supply of one, the greater
the demand for the other. The shortage of highly skilled workers causes a lower demand for low-
skilled workers: the lack of engineers may cause the loss of hundreds of blue-collar jobs. If the rate of
substitution is low, the constraint on highly skilled workers may cause such a low demand for low-
skilled workers that the wage at which they would be employed is unacceptably low. Since the shared
growth strategy involves maximising the job opportunities of the less skilled, it is fundamental that
the high-skill constraint be relaxed, especially in tradables. An obvious policy to quickly relax the
constraint is through the liberalisation and encouragement of high-skilled immigration. We shall
propose such a policy below. Greater immigration of highly skilled workers will help limit the increase
in wage inequality that is already happening and that will get worse as the economy keeps growing at
the current rate.
. . Skills shortages in South Africa

With the exception of deep recessionary periods, labour shortages in certain skilled occupations are
experienced during most phases of the economic cycle in South Africa. As indicated by Barker ( :
), skills shortages are not easy to measure, although vacancy data are usually used as a proxy for
skills shortages. The vacancy rate measures vacancies as a percentage of the total number of posts:
vacancies
Vacancy rate = × 100
employment + vacancies
Vacancy data are, however, subject to serious shortcomings, for example:
Quality deficiencies are not measured, for instance, vacancy data do not reflect workers being
appointed to skilled occupations without proper training or the ability to do the job.
Employers might indicate in the statistical returns that they have vacancies, but do not actively try
to fill them or do not offer realistic remuneration (Barker, : ). In these instances there are
vacancies, but strictly speaking, there is no real shortage of labour.
Many employers do not formally record vacancies, even though they might wish to employ
additional skilled workers in certain occupations. In these instances a labour shortage does exist,
although it is not recorded as a vacancy.
Frictional vacancies can also exist, i.e. vacancies as a result of normal labour turnover, which is a
phenomenon comparable to frictional unemployment (see section . . ).
A number of years ago, the authors of the ILO Review were of the opinion that there was little
evidence of substantial skills shortages in South Africa ( : ). They also contended that there was
not enough evidence to show that the shortages were constraining economic development. However,
as indicated by Barker ( : – ), it is practically impossible to statistically isolate the impact that
skills shortages might have on economic growth.
There are relatively few sources of information on skills shortages. One such source is surveys, for
instance the survey undertaken in by the World Bank in conjunction with the Greater
Johannesburg Metropolitan Area (Lewis, : ; Bhorat & Lundall, : ). This survey was
conducted among large firms across eight manufacturing sectors. Approximately % of large
firms experienced extreme to moderate difficulty in finding managerial and professional staff; %
reported the same with regard to service and craft skills. However, as Fallon and Lucas ( ) have
pointed out, the issue is not so much whether there is in fact persistent “excess demand” for higher-
skilled workers, but rather whether or not the skills base is sufficient to support higher economic
growth.
What is interesting from the survey in the Greater Johannesburg Metropolitan Area is that even semi-
skilled workers were in fairly short supply, since % of firms experienced difficulties in finding
“operators”. Only with regard to labourers were there practically no difficulties in finding workers. This
highlights the importance of focusing the skills development programme on skilled as well as semi-
skilled occupations.
Another interesting development that might indicate that the government is concerned about skills
shortages is that the Department of Home Affairs announced (in the Government Gazette, February
(RSA, )) quotas for specific occupations, in terms of which immigrants in these occupations
will gain easier access to the South African labour market. Some specific occupations that were listed
were: engineering, mathematics and science teachers; the information technology professions; health
and medical science professions; agricultural professions; and management and commerce
professions (specifically actuaries and financial analysts).
SUMMARY
Skilled labour shortages in certain skilled occupations are experienced during most phases of the economic cycle in
South Africa and, according to some observers, may indicate that the existing skills base is insufficient to support higher
economic growth.
. . Projection of the future demand for, and supply of, labour by skills category
Shortcomings of labour forecasts and labour-planning models
Labour forecasts should be treated with great circumspection. There are so many factors that
influence the demand for skilled and unskilled workers that blindly accepting the results of detailed
labour market forecasts for educational and career-planning purposes will result in a significant over-
and/or under-supply situation in many occupations. One of the most important factors influencing
the demand for labour is economic growth, which is already difficult enough to project into the
future, but impossible when it comes to the differential economic growth rates in various sectors.
Higher growth rates in sectors intensively utilising skilled labour will result in a greater demand for
such workers than high growth rates in other sectors. It is simply not possible to forecast with any
degree of certainty the economic growth rates and structural changes in various sectors, as these are
influenced by numerous developments in and outside the country.
God made economists to make weather forecasters look good.

A
A centrally planned labour programme cannot work because decision making regarding actions to be
taken is too centralised. This greatly increases the possibilities of major errors. In a market-oriented
system, where there are a great many small decision makers, the chances of any single decision maker
committing an error large enough to influence the market very negatively are much more remote.
Some prominent international experts have come out strongly against central planning approaches.
Psacharopoulos ( : – ), for instance, says, “Planning is becoming a word to avoid … why then
has manpower planning failed? The reason has to be sought in the inability of human beings to
anticipate future developments accurately”. Moura Castro ( : ), head of the Training Policies
Branch of the ILO, says: “We know full well that most orthodox manpower plans were disastrous in
market economies and we suspect that they were possibly even more disastrous in command
economies, although the evidence is less adequate.” Adams, Middleton and Ziderman ( : )
simply state that they “reject(s) manpower requirements forecasting”.
All these experts call for a more active analysis of past and present labour market developments, so
that trends in the balance of skills demand can be better identified, rather than statistical forecasts.
This approach is called labour market signalling. In contrast to the emphasis on occupation – a
feature of labour requirements forecasting – the labour market signalling technique is concerned both
with economic outcomes, measured in terms of wages and employment, and with the cost of specific
education and training programmes (Adams et al., : ). These provide a basis for determining
the desirable level of public and private expenditure on such programmes. Adams et al. refer to
various signals that can be utilised, several of which have been dealt with in this book.
Apart from the movements in wages and the employment of workers with specific levels of schooling
and training, some of the labour market signals that can and should be utilised are: unemployment
rates for particular skills levels; job vacancy rates; enrolment and graduation trends; tracer studies that
follow graduates into the labour market for a year or two; and household surveys.
Sharp increase in the demand for skilled workers

If one surveys the various labour market signals highlighted in this book, it is quite clear that the
demand for skilled workers will continue to increase much more rapidly than that for semiskilled
workers. Some of the professional categories where high growth rates are expected are law,
accountancy, engineering, information technology and finance. Little change is expected at the level
of clerical, sales and service workers, while a decline is forecast in the formal employment of unskilled
workers.
This is in line with international trends (see accompanying box) and relates to two important
changes:
First, there is a movement away from primary (i.e. mining and agriculture) and secondary (e.g.
manufacturing) industries to tertiary or service sectors. In general, there are fewer employment
opportunities for the semiskilled and unskilled worker (the labourer) in the latter sectors and certain
basic skills are required.
Second, there is also a movement within individual sectors away from unskilled and semi-skilled
occupations to those that require a higher level of skill. This relates especially to technological
innovation. Furthermore, in recent years there has been a movement away from mass production to
“personalised” production, in which the consumer is seen as an individual with personal tastes that
differ from one consumer to the next. Due to constantly increasing international competition,
production units have to adjust very rapidly to changing demand conditions. This requires each
worker to be more innovative and flexible.
In South Africa these trends will be quite apparent in the First World part of the economy, which has
to be competitive in international markets. An acute need for highly skilled workers will be
experienced. The less sophisticated part of the economy will provide for domestic demand, especially
basic consumer goods, and a need for workers with certain basic skills will be experienced. Skilled
workers will also be required to plan and lead the development process, which will have a high
priority in this country over the next two to three decades.
HIV and AIDS is likely to have an impact on the future demand and supply of labour (see also section
. ). Firstly, as far as the demand for labour is concerned, HIV and AIDS will increase the cost of
labour, and will therefore discourage employment creation. Unemployment might thus increase, but
at the same time employers will fill positions that might have become vacant owing to people falling
ill or dying of HIV and AIDS. There will also be a loss of skills due to people becoming sick with full-
blown AIDS, and less will be spent on education and training as expenditure of households will be
increasingly redirected to medical care and even basic necessities if the breadwinner falls ill with AIDS.
The supply of skills is therefore likely to reduce, whereas the demand might increase as employers
attempt to fill vacancies left by those who are sick or dying.
The basic skills required to live and work in the st century
The following have been identified in studies in the US as the basic skills required to live and work in the st century
(Packer & Christensen, ): teamwork; managing information; interpreting and communicating information through
mathematics and science; using technological and scientific systems; problem solving; thinking and learning in different
ways; and acquiring fundamental personal qualities such as integrity and the ability to make wise choices.
As a result of the rapid introduction of new production technologies as well as constant and rapid changes in consumer
tastes, there will be fewer routine or repetitive jobs. This, in turn, will require a more skilled workforce and employees
who are more adaptable and flexible, more innovative and creative. Multiskilling, constant reskilling and multitasking will
become the norm. Tasks and hierarchical structures will be less uniform and less clearly defined, and workers will be
expected to be more self-directed.
This also implies changes to remuneration systems. Pay systems will be linked to a much greater extent to performance,
and this could be either the performance of the individual or of the team. Collective agreements will increasingly be
restricted to framework agreements, and will become outdated in many growth sectors as these sectors generally have
low rates of unionisation (Blanpain : ).
In South Africa, the Services SETA has identified specific scarce skills, which should be addressed when deciding on
training interventions (Raath, ):
. Management and leadership
. Team leader and supervisory skills
. Project management
. Business-related skills
. Plant operators, e.g. bulldozers, graders
. Call centre operators
. Marketing skills
. Event management skills
. Property-related skills
. Embalming skills
. Hairdressing multiskills
The Department of Higher Education and Training ( : – ) also identified the occupations in high demand in
and, as expected, the majority of them come from the three skilled occupation categories, namely managers,
professionals, as well as technicians and associated professionals.
SUMMARY
The size and structure of labour demand cannot be forecast with any reasonable accuracy because it is not possible to
forecast economic growth rates and structural changes in various sectors. Labour planning models are therefore
regarded with great scepticism and are opposed by many observers because of the damage that planning based on
incorrect forecasts can do.
The basic skills required for work in the st century are likely to include teamwork, adaptability, initiative, creativity,
and managing information, communication and technology.
KEY CONCEPTS
direct costs
human capital
illiteracy
incremental earnings
indirect costs
labour market signalling
learnerships
National Skills Authority (NSA)
National Skills Fund (NSF)
private rate of return
rate of return
screening hypothesis
social rate of return
Sector Education and Training Authority (SETA)
South African Qualifications Authority (SAQA)
vacancies
vacancy rate
FOR STUDENTS
. Discuss human capital theory as a possible explanation for higher earnings being associated with increased
education and training, and investment in education and training being “profitable”.
. Human capital theory has been used to justify huge educational expenditure, but there have also been
significant criticisms of this theory. Discuss these criticisms.
. Discuss the most important problems facing the education system in South Africa.
. Discuss the most important institutions and mechanisms with regard to the skills development strategy in
South Africa.
. Discuss data on vacancies as a measure of skilled labour shortages and the shortcomings of such an indicator.
SUGGESTED READING*
Adams, A.V., Middleton, J. & Ziderman, A. . Market-based manpower planning with labour market signals.
International Labour Review, ( ).
Bot, M. . School education in South Africa: tracking change over ten years. Edusource Data News, : June.
Department of Labour. . National skills development strategy implementation report. April – March
. Pretoria: Department of Labour.
Department of Labour. b. State of skills in South Africa, . Pretoria: Department of Labour.
Reddy, V., Bhorat, B., Powell, M., Visser, M. & Arends, A. . Skills supply and demand in South Africa. Pretoria:
Human Sciences Research Council.

Labour market inequalities and discrimination

. Introduction
. Definitions and types of discrimination
. Theories of discrimination
. Inequalities in the South African labour market
. Addressing inequalities
. Equal pay for work of equal value
. Affirmative action
The boundaries of democracy have to be widened so as to include economic equality also. This is the great revolution
through which we are all passing.
J N
. INTRODUCTION
The issue of labour market inequalities and discrimination is a wide-ranging one, which requires a
study of its own. This issue, which is particularly relevant in South Africa, is thus discussed relatively
briefly in this chapter. It should be accepted that justice cannot be done to this important subject in a
single chapter.
Because of South Africa’s history of discrimination, inequalities between different race groups are
especially pertinent. However, this should not obscure other inequalities, in particular, those between
men and women and between those with a job and those without. The issue of racial categorisation is
also a very sensitive one.
Most of the statutory discriminatory measures were abolished in the s. The effect of these
measures will, however, be felt for a long time to come. The fact that African workers are still
underrepresented in several skilled positions can probably be ascribed to these measures.
This chapter starts with some definitions of discrimination. It is important to distinguish between
different types of discrimination: those that have their origin in the labour market and those that start
even before entry to the labour market, because measures to address discrimination differ, depending
on the source of the discrimination.
South Africa has advanced politically by disasters and economically by windfalls.

C.W. D K
Attention is then given to various theories that have been put forward to explain discrimination, as
market forces are supposed to eliminate discrimination in a perfectly competitive labour market.
Inequalities arising from different types of discrimination in the South African labour market are
discussed. These include inequalities in education, employment, incomes and occupations. Finally,
there is a discussion of various measures that can be, or are, used to address inequalities and
inequities.
. DEFINITIONS AND TYPES OF DISCRIMINATION
Labour market discrimination refers to the inferior treatment of groups of workers with respect to
employment policy or practice, through factors not related to the labour market, such as being female
or black (Barker & Holtzhausen, ). McConnell et al. ( : ) describe economic discrimination
as follows: “Economic discrimination exists when female or minority (i.e. black) workers – who have
the same abilities, education, training and experience as white male workers – are accorded inferior
treatment with respect to hiring, occupational access, promotion, wage rate, or working conditions.”
Another definition, by Fearn ( : ), is that discrimination in an economic sense refers to
differences in treatment (e.g. wages, employment, schooling) that are not based on cost or expected
cost differences or differences in productivity or expected productivity.
I tremble for my country when I reflect that God is just; that his justice cannot sleep forever.
T sJ
Convention of the ILO defines discrimination as any distinction, exclusion or preference based on
one of the grounds, which this instrument enumerates (e.g. race, national extraction, social origin,
gender, political opinion or religion) that has the effect of nullifying or impairing equality of
opportunity or treatment in employment or occupation (ILO : ).
The ILO Convention refers to certain measures that are not deemed to be discrimination, of which the
most important is any distinction, exclusion or preference in respect of a particular job based on its
inherent requirements. Considerations of nationality and gender have, for example, been regarded as
legitimate grounds for distinction in certain types of employment. Affirmative action measures would
also not be deemed to be discrimination. In terms of South African law, these distinctions would not
be “unfair” discrimination.
There are four types of possible labour market discrimination (McConnell et al., : – ):
Human capital discrimination is discrimination in respect of education and training, which may
include impaired access to education or training, or inferior-quality education.
Employment discrimination occurs when some groups bear a disproportionate share of the burden
of unemployment.
Wage discrimination exists when workers belonging to certain groups are paid less than workers
belonging to other groups for doing the same work.
Occupational discrimination occurs when specific groups are underrepresented in skilled
occupations, although they are as capable as other groups and also have the necessary
qualifications.
The first category is referred to as before-the-market discrimination, i.e. it occurs before the individual
seeks employment or, in the case of on-the-job training, before he or she is fully productive in the
labour market. The other three categories are referred to as within-the-market discrimination, because
they are encountered after the individual has entered the labour market (even if still unemployed).
SUMMARY
Before-the-market discrimination is discrimination before employment commences (e.g. education), while within-the-
market discrimination includes employment, wage and occupational discrimination.
. THEORIES OF DISCRIMINATION
In a perfectly competitive labour market, there are no wage differentials for reasons other than those
caused by market factors (e.g. one worker being more productive than the other). Any discrimination
in wages is eliminated by competition between workers or between employers. Clearly, this is not how
the real world operates. Even without laws and policies that discriminate, groups of workers in many
countries experience discrimination of one kind or another. Various theories have been formulated to
explain discrimination; these are discussed below.
. . Taste-for-discrimination model
In his attempt to explain discrimination, Gary Becker originally presented the taste-for-discrimination
theory in his doctoral thesis in . The economic analysis of discrimination is thus relatively recent
when compared to the analysis of other economic phenomena.
According to Becker, some employers prefer certain groups of workers, for example whites or males,
even if this preference or prejudice means that they have to pay for their prejudice in terms of higher
wages for such workers. Such employers are thus in effect saying that it imposes an indirect cost on
them to employ black workers, for example. The cost of employing a black worker will then for that
particular employer be the wage of the black worker plus the psychic “disutility” of hiring black
workers. The strength of the psychic cost of employing a black worker by the discriminating employer
is called the discrimination coefficient (d), and varies from one employer to the next. Such employers
will only be willing to employ black workers if their wage rate is below that of white workers, and then
only if black wages are lower than white wages by the amount of the discrimination coefficient.
If the wage rate of white workers is WW and that of black workers WB, then the discriminating
employer will act as if the cost of employing a black worker is WB + d, even though the actual
monetary cost of employing this black worker is WB. If d equals zero (that is, the employer does not
have any preference or prejudice over any group of workers) and assuming white and black workers
are equally productive and perfect substitutes, it is expected that WW equals WB. However, if d is
positive and the employer discriminates against black workers, the employer’s hiring decision would
be as follows:
Hire only white workers if WW < WB + d, as it is relatively cheaper to hire white workers, after taking
the psychic disutility of hiring black workers into consideration.
Hire only black workers if WW > WB + d; although the employer discriminates against black workers
(as indicated by the positive d), even after taking the psychic disutility of hiring black workers into
account, it is still relatively cheaper to hire black workers.
Be indifferent with regard to black and white workers if WW = WB + d, after considering the psychic
disutility of hiring black workers. Racial prejudice thus blinds the discriminating employer to the
true monetary cost of employing a black worker, and he or she will be willing to pay a premium for
white workers. There is a mixed workforce in the firm, as the employer would hire some black
workers and some white workers.
Let’s elaborate these possible hiring decisions further with the following mathematical examples:
WB = R ,d=R and WW = R . Since WW = < = WB + d, the employer would only hire

white workers, despite the fact that in actual monetary terms it is cheaper to hire black workers
(R vs. R ). Only white workers are hired as the employer acts as if it costs R (WB + d) to hire
a black worker, after taking d into consideration.
WB = R ,d=R and WW = R . Since WW = > = WB + d, the employer would only hire
black workers. Although the employer discriminates against black workers (d > ), it is still relatively
cheaper to hire black workers, even after the employer acts as if it costs R (WB + d) to hire a
black worker by taking d into account, compared to the more expensive cost of hiring a white
worker (WW = R ).
WB = R ,d=R and WW = R . The employer is indifferent in terms of hiring black and white
workers, because WW = = WB + d. Even though in actual monetary terms the black wage is
cheaper (R vs. R ), the discriminating employer acts as if it costs R to hire a black worker, so
at the end he feels as if it costs the same to hire the two groups of workers.
The taste-for-discrimination model can also be reflected graphically. The vertical axis of Figure . is
different from the usual labour market analyses in that it measures the ratio of black wages to white
wages – the higher along the axis, the higher the black wage relative to the white wage. At the level of
. , white and black wages are equal. The horizontal axis indicates the number of black workers
employed in a specific occupation by all employers.
FIGURE . Wage discrimination in the labour market
Source: Adapted from McConnell et al. ( : )
The non-discriminating employers will pay white and black workers the same, which means that their
discrimination coefficients (d) are zero. Their employment of black workers is reflected by the portion
AB on the black labour demand curve DB. After point B, black employment will increase only if
discriminating employers employ black workers, and they will do so only if the wages of black workers
are lower relative to those of white workers (i.e. WB/WW is less than ). A movement along the curve
from B to point C indicates discriminating employers whose discriminating coefficients (d) are
increasingly higher. At point C, employers are pure racists and will not appoint black workers at any
cost.
The supply of black workers can now be added to the graph – SB. It is upward-sloping because the
quantity of black labour supply increases if WB/WW increases (i.e. the black wage is closer and closer
to the white wage). The intersection of the two curves establishes the ratio of black wages relative to
white wages, i.e. . , which means that black workers will earn % of that of white workers owing to
discrimination.
It should be clear that non-discriminatory employers will be at a competitive advantage because they
need not pay a premium to employ white workers. By employing equally productive black workers,
they thus gain in terms of profits. This is illustrated in Figure . . At point AB, the non-discriminating
employers (with discrimination coefficient zero) make maximum profits. Thereafter, and as the
discrimination coefficient increases, employers make less and less profit. At some point, racists refuse
to employ any black workers, and their profits take a plunge due to the excessive premium they pay
to employ white workers only.
FIGURE . Profits and the discrimination coefficient
Taste-for-discrimination model: a mathematical example
Let’s assume a hypothetical country consists of two population groups, with % being blacks and % being whites.
Affirmative action policy is implemented in this country which recommends that employers have a workforce composition
similar to the racial profile of the population.
Now we have two firms. Firm A, owned by a non-discriminating employer, earns a total revenue of R . It is assumed
that labour is the only input cost. WB = WW = R and d = , and the employer hires eight workers. Since the employer
does not discriminate against any group of workers, and the white wage equals black wage, he finally hires six black
workers and two white workers to meet the affirmative action requirement. Total cost equals R (R × ) and hence
firm A earns a profit of R (R –R ).
Firm B, owned by a discriminating employer, also earns a total revenue of R . It is once again assumed that labour
is the only input cost and the firm needs to hire eight workers. The information in Table . indicates that the employer of
this firm treats white and black workers as the same (d = ) up to the point that three black workers are hired. However,
from hiring the fourth black worker onwards, the employer starts to discriminate against the black workers (d > ), and he
will only hire the additional black workers if the disutility-adjusted black wage equals the white wage (WB + d = WW). For
example, he is indifferent to hiring either a white worker or the fourth black worker if WB = R and WW = R as d
equals R . That is, the employer pays a wage premium of R to hire a white worker as WW increases from R
(when d = ) to R (d = ). Similarly, the employer is indifferent to hiring a white worker or the fifth black worker if WB
= and WW = R . as d equals R . .
TABLE . Taste-for-discrimination model: a hypothetical example
WB d WW WB/WW QD QS
100.00 0.00 100.00 1.0 1 5

100.00 0.00 100.00 1.0 2 5
100.00 0.00 100.00 1.0 3 5
100.00 25.00 125.00 0.8 4 4
100.00 66.67 166.67 0.6 5 3
100.00 150.00 250.00 0.4 6 2
100.00 400.00 500.00 0.2 7 1
Using the information in the table to plot the labour demand and labour supply curves, Figure . indicates that at
equilibrium, four black workers are hired. That is, out of the eight workers hired by firm B, % of them are blacks and
% are whites (or it can be said that firm B does not follow the recommendation of affirmative action as the employer
discriminates against blacks). Total cost equals R (R × +R × ) and hence firm B earns a profit of R
(R –R ).
FIGURE . Taste-for-discrimination model: a hypothetical

graphical illustration
Firm B’s profit is lower when compared to firm A, because the white workers enjoy a wage premium of R (R –
R ) due to the fact that firm B discriminates against black workers (d = R ).
In a perfectly competitive market, the discriminating employer will have to charge higher prices to
cover the wage premium paid to white workers. The discriminating employer will thus either
disappear or will have to end discriminating practices. According to the theory, in a perfectly
competitive market, the wages of black and white workers will equalise, and discrimination should be
eliminated in the long term.
The fundamental criticism of this model has been directed at the conclusion that discrimination will
disappear by itself in a competitive market. Evidence in the US, for instance, does not indicate
declining discrimination (McConnell et al., : ; Faundez, : ), and this has led to further
theories to explain discrimination.
Is it possible for the discrimination coefficient to be negative?
The above discussion assumes that d is positive, with the employer discriminating against black workers. However, it is
possible that some employers prefer to hire black workers as they enjoy utility from hiring them. This type of behaviour is
known as nepotism, which is the opposite of discrimination. The employer’s utility-adjusted cost of hiring a favoured
worker (blacks in this case) equals WB – n, where n stands for nepotism coefficient. This essentially means that the
discrimination coefficient (d) is negative, as it is essentially equal to the nepotism coefficient in negative terms, i.e. d = –
n.
For example, assuming WB = WW = R but n = R , then WW = < = WB – n. Therefore, even though in actual
monetary terms it costs the same to hire black and white workers (R ), the employer would only hire black workers
because he acts as if it only costs R to hire a black worker.
. . Statistical discrimination
This theory has been put forward to explain why discrimination persists in competitive labour markets.
According to the theory, employers discriminate not because of a taste for discrimination (they are
thus, in fact, not prejudiced against blacks), but because of their perception that, on average, workers
from certain groups are less productive, based on the employment histories of workers of these
groups hired in the past. Statistically, therefore, there is (at least according to these employers) a good
chance that an individual from a certain group (e.g. black or female) will have certain personal
characteristics that might disadvantage the employer. This “statistical generalisation” of individual
workers leads the employer to discriminate against those individuals. The characteristic of race or
gender, for instance, is then used (incorrectly) as an inexpensive substitute for determining the actual
production-oriented attributes of workers. The proper selection of personnel is expensive. Employers
therefore employ these cheaper “methods”, which are not accurate (Hamermesh & Rees, : ).
The cheapest method of selecting personnel is to focus on educational qualifications, because there is
the belief that educational qualifications have something to do with workers’ productivity. Similarly,
many employers use a criterion like skin colour or gender in the belief that, for various reasons (such
as the quality of schooling, family background, nutrition or cultural background), one group is
somehow less productive than another. This selection criterion is then used regardless of whether the
individual might have better characteristics than the group as a whole. He or she is therefore
discriminated against on the basis of belonging to that group.
For example, let’s assume statistical records in a hypothetical country reveal that many women leave
the firm when they reach their early thirties due to pregnancy and family commitments. Now two
persons aged years, one male and one female, possessing the same characteristics and potential
productivity, apply for the same job at a firm. Although the employer does not know if the female
applicant might be pregnant and resign in a few years’ time, he or she infers from the statistical
records that the female applicant has a higher likelihood of resigning within the next few years. The
employer thus thinks it may be a waste of money to invest in the female applicant (such as further
education and training) and hence offers the job to the male applicant.
Statistical discrimination may have an impact on the wages of different groups, just as what happens
using the taste-for-discrimination model, but this falls beyond the scope of this chapter and is not
discussed further.
. . Market imperfections
There are specific circumstances where an employer may find it profitable to practise wage
discrimination, i.e. to pay differential wages to equally productive groups of workers, for example
whites and African workers, in cases where perfectly competitive conditions do not exist.
Labour supply less elastic

The employer will, for instance, find it profitable to pay lower wages to groups of workers whose
labour supply is less elastic, i.e. less responsive to a wage reduction. This is particularly relevant in
South Africa, as the labour supply of African workers in the past might have been less elastic than that
of white workers – there was an adequate labour supply at wages for which white workers were not
prepared to work. This might have been because African workers were less mobile, they feared
unemployment more, and there were fewer employment alternatives for them. They were also in
general less skilled than white employees because of discrimination in education, training and
occupational mobility. Furthermore, as they were less likely to emigrate than white workers, they were
also less responsive to lower wages.
However, this might have changed in recent years as black workers have increased their mobility and
skills in the labour market. Currently there are probably a number of occupations where the elasticity
of the supply of black workers might in fact be higher than that of white workers, or where they are
actually paid more than white workers owing to the pressure on employers for their workforce to
become more representative, combined with the fact that the educational and training system cannot
provide sufficient numbers of black skilled workers.
Dual labour market

The dual labour market theory is another possible explanation for continued discrimination (see
section . . ). African workers are often found in the secondary labour market, i.e. where wages are
relatively low and there is little training or job security. They are therefore tagged as unstable and
undesirable workers, although this is because they are unable to enter the primary labour market
(where employment conditions are better, proper training is provided and there is more job security,
which leads to a more stable and dependable workforce).
. . The crowding model
According to one variant of this model, the employer will, in the interests of labour peace, segregate
white and African workers (for instance) into different occupational categories and thereby avoid
racial tensions in the workplace (McConnell et al., : ). White customers might also prefer white
workers and this results in African workers being crowded into less attractive occupations (with
African people becoming the predominant consumers in South Africa, this will rapidly change).
Another variant relates to the fact that women in particular are overrepresented in a small number of
traditionally women’s occupations (such as secretary, sales worker, primary-school teacher and nurse).
Being crowded into a limited number of occupations drives down the group’s wage rates and results
in inequalities. This theory does not, however, fully explain why employers do not employ more
women (at lower wages, and attain higher profitability) in traditional male occupations.
. . The transaction costs model

African people have their own languages and culture, and often lower educational qualifications than
whites. The cost to the employer of restructuring the workplace to accommodate these differences, or
to provide additional training or bridging courses, can be perceived to be substantial. If an employer
employs African workers, he or she is forced to offset these higher transaction costs by paying lower
wages. If this theory is valid, then education and other differences (and not necessarily discriminatory
practices by the employer) can be the most important reason for discrimination.
Another example can be women workers in certain occupations such as mining. Additional ablution
and other facilities need to be provided both on the surface and underground to cater for female
workers, and the safety and other equipment also needs to be re-engineered to provide for the
different physique of women. This increases the transaction cost of employing women workers, which
leads to discrimination.
. . Legislation and government policy

In South Africa, a further explanation for inequalities is that for many years African people were
discriminated against through official policy. Legislation such as: job reservation, which reserved
certain jobs for whites only; influx control, which limited access to urban areas; the Group Areas Act,
which limited the availability of housing for African people and restricted social interaction; and
several other pieces of legislation, resulted in the entrenchment of discriminatory practices
(apartheid). In addition, until the early s African workers were excluded from the provisions of the
LRA, which meant that they could not bargain collectively to improve their wages and conditions of
employment.
The scrapping of discriminatory measures by the previous government during the s reduced
statutory discrimination in the labour market although, if several of the above theories are correct,
discrimination will not be totally eliminated. Many of these theories offer an explanation of why
discrimination will persist, especially in conditions where labour markets are not fully competitive and
where educational backlogs will be felt for many years to come. Much literature has been published in
South Africa in recent years regarding methods to reduce or eliminate inequalities.
It should also be noted that labour markets that were discriminatory in the past but no longer are
might appear to perpetuate discrimination where the influence of the dual labour market is quite
marked. This is because new entrants (i.e. those discriminated against) can usually only enter the
primary or internal labour market from the bottom, and it will take time for them to move through the
ranks to higher positions.
SUMMARY
In a perfectly competitive labour market, there will be no differentials (e.g. in wages) for reasons other than those caused
by marshortket factors. However, wage and other forms of discrimination continue to exist, and various theories have
been put forward to explain discrimination.
. INEQUALITIES IN THE SOUTH AFRICAN LABOUR MARKET
Due to various kinds of discrimination, the South African labour market is characterised by severe
inequalities, of which the most important are discussed in this section. In many cases the inequalities
may amount to unfair discrimination, but this is not necessarily always the case (see discussion on
wage differentials in section . ). To determine whether discriminatory practices are actually evident
in every instance would require rigorous analysis, for which adequate statistical data on a macro basis
are not always available. For example, Hausmann ( : ) reports in the final findings of the
International Panel on Growth that South Africa was “able to allocate about percent of GDP to social
transfers to its previously disadvantaged peoples and significantly expand the allocations for health,
education and low-income housing in the budget, suggesting that income inequality after taxes and
transfers may show a substantially different picture”.
. . Educational level
An important indicator of the divergent effects of the discriminatory educational system of the past is
the educational levels of white and African people. Only % of the African labour force had a Matric
or higher qualification in , whereas the percentage for the other population groups was % (see
Figure . ). There has, however, been an encouraging improvement in educational levels over time.
Whereas less than % of the African labour force had at least a Grade qualification in , %
had at least a Grade qualification in .
FIGURE . Educational level of the labour force by
population group,
Source: Author’s own calculations using the Quarterly Labour Force
Survey quarter data
It’s only in the cemetery that all men are equal.

A
The shortcomings of South Africa’s education system are discussed in Chapter . These shortcomings
have compounded inequalities in South Africa.
For instance, in the past there were wide differences in the per capita education expenditure for white
and African learners, with that for Africans comprising about a third of that for whites. Teachers of
mainly African learners were furthermore often unqualified or underqualified (and therefore earned
less than those teaching mainly white learners). Unqualified or underqualified teachers are an
important reason for inferior education, which is described above as one of the types of before-the-
market discrimination.
The ratio of learners to teachers in mainly African schools was also much higher than in mainly white
schools, which resulted in a lower quality of education. This is partly due to discrimination and partly
due to the higher proportion of African learners in primary schools (worldwide bigger classes are the
norm in primary schools). The declining birth rate among whites and the sharply increasing number of
African learners further contributed to the differences in learner-to-teacher ratios in racially based
classes in the past.
In addition, the education system could not deal with the high rate of increase in the number of
learners in black schools, and the quality of education declined sharply. The increasing incidence of
unrest and disruption in the education system is also relevant, with the unrest being probably both
the cause and consequence of the declining quality of education.
SUMMARY
There are a number of indicators of educational inequalities in South Africa: the per capita expenditure per learner; the
learner-to-teacher ratios; the pass rates of Grade learners; and the education level of the labour force.
. . Employment and unemployment inequalities

The unemployment rate, according to the QLFS, is much higher among the African labour force than
among whites. For African people, the official unemployment rate was approximately % in (see
Figure . ), whereas it was only about % among whites. The expanded rate among African workers
was much higher ( %), due to the high proportion of people in especially rural areas who have
become discouraged job-seekers and are thus not counted as being officially unemployed.
Furthermore, % of the unemployed African people had been looking for a job for more than three
years, while among whites the percentage was about % in .
FIGURE . Strict and expanded unemployment rates (%) by population

group,
Source: Author’s own calculations using the Quarterly Labour Force Survey
quarter data
Although discrimination is an important reason for differential unemployment rates between African
and white people, it should be remembered that inequalities in employment relate very closely to
other factors such as inequalities in education, differential population growth rates and the level of
urbanisation. Since African workers on average have lower educational qualifications than whites, and
at the same time their numbers have increased sharply, it is expected that unemployment among
them will be higher than among whites, who have higher educational qualifications and experience
low or declining population growth rates. In addition, if African workers are the least urbanised group,
it is to be expected that employment among them will also be lower because of fewer employment
opportunities in rural areas. If discrimination in employment is to be eliminated, the causes of much of
the discrimination will have to be addressed.
Rospabé ( : ) looked at the period to , and found that the first effects of affirmative
action had already become apparent in the s, because African workers benefited most from the
jobs that were created in South Africa during that time. However, he also found that if the
white/African gap in employment is analysed, two-thirds of the gap is accounted for by differences in
employment-enhancing characteristics such as education, better family background and location in
areas of lower employment. Obviously, a number of these factors point to before-the-market
discrimination.
One-third of the racial gap in employment remains unexplained, and can “cautiously” be attributed to
discrimination. At the same time, there may be other (unmeasured) characteristics that may explain
part of the difference, and the omission of these variables might lead to an overestimation of within-
the-market discrimination.
The two recent studies by Burger and Jafta ( ; ) examined the period – , and found
that affirmative action policies have no observable impact on reducing employment discrimination by
race and gender.
A significant recent development that is described in section . . is that the employment of whites
has been stagnant for many years, whereas that of Africans has recently been increasing. Between
and , for instance, African employment increased by . % per annum, whereas the
employment of white people increased by only . % per annum. This is even more significant if one
takes into account that the increase in the employment of African persons was off a much higher base
than that of whites.
There was no significant increase in unemployment among whites (see Table . ), which means that
many white workers either left the country or retired. Whites are the key skills reservoir of the country,
and this loss has been described by Bhorat ( b) as “extremely serious”.
It is also clear that women bear a disproportionate burden of the unemployment problem in South
Africa. African women, for instance, constituted half of total African unemployment, but formed only
% of the African labour force in . Both the strict and expanded unemployment rates among
African women ( % and % respectively) are consistently much higher than those of African men
( % and % respectively). The extent of the hardship experienced by women as a result of high
unemployment is, however, difficult to determine. This can only be done if information is available on
the alternative incomes and size of the family affected and the number of unemployed women who
are heads of a household.
SUMMARY
Unemployment rates among African workers are much higher than among whites, and higher among women than
among men.
Recently, there has been a significant increase in the employment of Africans, whereas the employment of whites has
been stagnating for several years.
. . Occupational inequalities
Inequalities by race
Africans comprised about % of all workers in the formal sector in . However, they make up a
much smaller percentage of skilled workers. Figure . indicates that Africans occupy only about %
of senior official and management positions. The percentages for the other occupational groups are
higher, for instance % of professionals (many of these would be teaching and nursing-related
occupations) and % of technical and related occupations. The percentage of Africans in elementary
occupations (under the unskilled category) is very high at %.
FIGURE . Africans as proportion of all formal-sector employees in different occupational categories,
Source: Author’s own calculations using the Quarterly Labour Force Survey quarter data
Table . contains information about the skills levels of African employees in and . There
has been an improvement in the share of African employees in the skilled occupations, rising from
. % in to . % in . Nonetheless, the African share also increased when it comes to the
semi-skilled and unskilled occupations.
TABLE . African employees by skills level, and
2006 2016
Number % of total Number % of total
Skilled – total 935 491 45.2 1 513 916 54.3
Skilled – managers 144 202 28.0 311 330 38.7
Skilled – professionals 264 137 51.2 409 293 55.6
Skilled – technicians 527 152 50.7 793 293 63.6
Semi-skilled 2 915 409 65.5 4 167 243 74.4
Unskilled 1 039 992 81.1 1 732 011 87.1
Total 4 890 892 62.7 7 413 170 71.4
quarter data
Rospabé ( : ) has found that about two-thirds of the occupational differences between white
and African workers can be explained by objective factors such as educational levels. Although these
might be due to before-the-market discrimination, it appears that only a third of the difference in
occupational differentiation can be ascribed to discrimination. One of the unexplained factors may be
the choice of particular types of work, which might be influenced by social norms in a particular
group. Burger and Jafta ( ), after controlling for differences in characteristics across the racial
groups, found that there is some indication of discrimination against Africans and Coloureds
(compared to whites) when it comes to skilled occupational attainment.
Inequalities by gender
African women comprised only % of all skilled workers in but this share nearly trebled to %
in , as shown in Table . . In particular, the female share of workers working as technicians rose
to nearly % in .
TABLE . African women employees by skills level, and
2006 2016
Skilled – total 448 708 10.9 812 917 29.1
Skilled – managers 56 301 10.9 108 066 13.4
Skilled – professionals 115 116 22.3 221 502 30.1
Skilled – technicians 277 291 26.7 483 349 38.7
Semi-skilled 855 477 19.2 1 497 670 26.7
2006 2016
Unskilled 377 780 29.5 778 430 39.1
Total 1 681 965 21.6 3 089 017 29.8
quarter data
Women have no need to prove their manhood.

W S H
SUMMARY
Africans represent more than half of the skilled occupations.

Women account for about % of skilled occupations.
. . Income and wage inequalities

It is well known that South Africa is ranked as one of the most unequal societies in the world, and that
the level of inequality has not reduced since the advent of democracy in the country, nor as a result of
the positive economic growth rates sustained over relatively long periods of time.
The extent of income inequality in South Africa compared to that of other countries can be
determined in various ways, but the most wellknown instrument is the so-called Gini coefficient. If the
perfect distribution of income is a situation where % of all earners receive % of all earnings, %
of all earners get %, and so forth, the Gini coefficient measures the extent to which the actual
distribution of earnings in a country deviates from the perfect distribution. The coefficient can range
from , which represents perfect equality (i.e. no deviation from the perfect distribution), to . So the
higher the number, the greater the level of inequality.
Another instrument is the Theil index, which has the advantage that it can decompose overall
inequality into a proportion originating between subgroups and a proportion originating within
subgroups. For example, overall inequality can be decomposed by race, with a certain proportion of
overall inequality being explained by inequality between the race groups, and the remainder being
explained by inequality within race groups (Bhorat, Van der Westhuizen & Jacobs, : ).
The Gini coefficient for the South African economy as a whole increased from an already high . in
to . in (Mohr et al., : ). The level of inequality among Africans is the highest, at
. , compared to . among whites in . All race groups have experienced a significant increase
in inequality since .
Some of the reasons for this increase in inequality within the African group are the following:
The high unemployment rate among Africans (i.e. a large proportion of Africans with no or very low
levels of income)
Affirmative action and the sharply increasing wage and income levels of Africans at the top end of
the income spectrum
The high and rising premium being paid to highly skilled workers from all race groups
Improvement in schooling for some and not for others, and presumably also differences in the
quality of education
Some African employees being employed as atypical workers (which erodes “rate-for-thejob” wage
setting)
Centralised bargaining (which benefits urban, organised workers)
Increasing duality between the primary and secondary labour markets – see section . . ). The
second economy has for all practical purposes not benefited significantly from the economic
growth trajectory in South Africa, and there are high barriers to entry into the primary or formal
part of the economy.
Earlier studies found that the rising level of inequality among Africans was an important reason for the
increase in total inequality in South Africa (see sources quoted by Bhorat et al., : ; ).
However, if inequality is measured according to the Theil index, the contribution of inequality within
race groups (and Africans in particular) to overall inequality seems to have declined between and
, while the inequality between race groups has become a more important factor (Bhorat et al.,
: ). It is therefore primarily income differentials between race groups rather than within race
groups that have been the driving force behind increasing overall inequality in South Africa.
It should be noted that Fedderke, Manga and Pirouz ( ) have grave reservations about whether
inequalities in the South African labour market have in fact increased. They state that there are
contradictions depending on whether income data or expenditure data are utilised. In addition, they
contend that there are clear signs of declining inequalities among the bottom third of African
households, which might be related to the increased welfare transfers to these households in recent
years.
As indicated above, one of the sources of inequality in South Africa is the high proportion of people
outside the formal economy. South Africa has one of the highest unemployment rates in the world,
and it is therefore not surprising that income inequality is also among the highest. A relatively old
study has shown that when the unemployed and the informal sector are excluded from the
calculations, the Gini coefficient was only . for the period – , compared to a coefficient of
. for the whole economy during that period (Central Economic Advisory Service, : ). This
lower level is similar to the income distribution pattern found in Western economies, and it is
generally reckoned that a Gini coefficient of around . is the lowest practical figure obtainable for
any country. This is confirmed by another study that shows that even though the distribution of
income for the population as a whole has not improved, there has been an improvement among
formal-sector income earners (McGrath & Whiteford : ). This is especially as a result of sharp
increases in the earnings of qualified African workers.
Although the combination of increasing unemployment and the sharply increased earnings of a select
group of Africans have in the past contributed to an increase in inequality in South Africa, this seems
to be a less important contributing factor according to recent studies.
There have also been studies to determine how the different income sources have contributed to
income inequality in South Africa. The main income sources are wage income, income from self-
employment, state grants, capital income, private pensions and remittances. Bhorat et al. ( : )
have found that the contribution of wage income to income inequality has increased by .
percentage points over the period – , from % in to almost % in . The most
important reason for this phenomenon was the rising premium being paid to highly skilled workers.
On the other hand, Leibbrandt et al. ( : – ) found that % of income equality was driven by
wage income in , while the Theil indices indicate that the contribution of within-race inequality to
overall inequality increased from % in to % in .
The Gini coefficient increases significantly when grant income is excluded (Bhorat et al., : ). The
social security system in the country has thus contributed significantly to lowering the overall levels of
inequality in the country. In this regard, Fedderke et al. ( ) found declining inequalities among the
bottom third of African households, which might be related to the increased welfare transfers to these
households in recent years.
We cannot help the poor by destroying the rich.

A
Wage differentials and discrimination

The South African Reserve Bank ( : ) states that the wage compression policies of trade unions
have resulted in lower inequality in the unionised sectors than in the non-unionised sectors.
Furthermore, since unions raise the average wage in the unionised sectors relative to the non-
unionised sectors, they may well increase the inequality between these sectors.
An important reason for wage differentials is that average remuneration is used, and as indicated in
section . , there is a host of factors that might be responsible for wage differentials, for example
differences in occupational levels, education, experience, seniority and numerous other factors. White
workers often occupy more senior positions than African workers and it is thus not unexpected that
their average wage is higher. Although this may be one of various other types of discrimination, it is
not necessarily wage discrimination by the employer.
Wage differentials can thus only be properly determined by considering the wages of employees of
the same gender and job grade. McGrath and Whiteford ( ) conclude that racial wage differentials
had already declined considerably in the s. It is likely that the “rate for the job” is more often than
not currently the norm in South Africa. Moll (as quoted by Van der Berg & Bhorat, : ) comes to
a similar conclusion, namely that while total discrimination accounted for % of the difference in the
African wage in , it fell to % in . Thus, only % of the difference in wages between African
and white people can be ascribed to wage discrimination rather than other characteristics such as
differences in age, education, etc.
Rospabé ( : ) has also found that discrimination does not account for a substantial part of the
African/white wage gap – less than % in . However, he further found that the proportion of the
wage gap that could probably be ascribed to discrimination increased between and . If his
findings are correct, there appears to be an increase in wage discrimination rather than a reduction, as
would have been expected.
Bhorat ( a: ) maintains that there are many reasons for these continuing differences in wages,
but the most significant explanation for higher earnings is education, especially Matric or tertiary
education. However, he also found that whites were relatively better remunerated than African
workers for the same level of education. He provides three possible reasons for this phenomenon:
A higher proportion of whites are educated in fields such as computing and engineering, which are
in much greater demand than fields such as the humanities, in which most African workers with
tertiary degrees are educated
A degree from a historically white university is perceived as being of a higher quality than one from
a historically black university (and thus graduates of the former are remunerated better)
Employers continue to discriminate.
Bhorat ( a: ) refers to some policy implications of these findings. The first is that skills
development should receive high priority to reduce the skills shortages, which is an important reason
for the wage differentials between and within population groups. He also emphasises that a simplistic
legislative approach to reducing the wage gap between different skills levels (see section . for a
discussion on the wage gap) would be “foolhardy”. The wage gap has arisen owing to a fairly intricate
interplay of factors and cannot simply be reduced or eliminated by a blunt instrument such as
legislation.
Differences in wages of the different population groups are more severe than those between men and
women (Bhorat, a: ; Burger & Jafta, : ).
A few words of caution

As should be apparent from the discussion above, the issues of income inequalities and the
redistribution of income or wealth are very complex and cannot be dealt with satisfactorily in a few
paragraphs.
All animals are equal, but some animals are more equal than others.
G O
SUMMARY
Intraracial earnings inequality among African workers, and African men in particular, increased considerably, and the
intraracial inequality in fact exceeds the interracial inequality.
South Africa’s Gini coefficient increased from . in to . in .
Inequality in South Africa is driven to a significant extent by the gap between those with jobs and those without.
. ADDRESSING INEQUALITIES
The major inequalities in South Africa today are those between African and white people, between
men and women, and between those with and without a job, irrespective of race. As indicated above,
this is also quite evident in the labour market, and instead of referring to the reduction of inequalities,
numerous authors refer to programmes for black advancement or black economic empowerment.
Some measures to address inequalities and inequities are discussed in this section. The discussion can
by no means cover the whole field.
It should be emphasised from the outset that any action to address inequalities in South Africa will
confront major obstacles. The most important is probably that one of the disadvantaged groups
(African people) is by far the majority of the population. Programmes to address inequalities should
be aimed at the broader population and not only at the urban or unionised and employed élite. If
expectations outpace reality, then even more instability could result from programmes to reduce
inequalities.
The forces of capitalist society, if left unchecked, tend to make the rich richer and the poor poorer.
J N
Care should also be taken to reduce and not increase racial tension. For many decades into the future,
whites will form a critical component of the labour force and any action aimed at reducing inequalities
should not result in skilled whites leaving the country. The country is already losing skills to a
significant extent (see section . . ) and this brain drain will seriously damage the availability of
human capital and consequently economic growth. It will thus have a negative impact on the ability of
the economy to rapidly reduce inequalities and increase living standards.
. . Higher economic growth and employment opportunities

Although a number of legal and other methods of reducing inequalities are considered below, these
remedies can be effective only under conditions of higher economic growth and increasing
employment opportunities.
The cure for capitalism’s failing would require that a government would have to rise above the
interests of one class alone.
R L. H
To a significant extent, inequality is driven by the gap between those without jobs and those in wage
employment. Work done by Bhorat et al. ( ) shows that wage inequality accounts for . % of
total inequality. Of this wage inequality, % can be ascribed directly to unemployment, in particular
because one-third of South African households have no wage income. The remaining % is due to
inequality within wage-earning households, but even then unemployment might still play a role
because low household income might be caused by some members being unemployed or in the
informal sector.
Van der Berg ( ) has made some projections to determine what the effect would be of a high
economic growth rate (and therefore high employment growth) in comparison to the effect of a rapid
reduction in the wage gap, and his conclusion is that even a very rapid reduction in the wage gap will
have a smaller impact on per capita racial income distribution than economic growth and job
creation. Through economic growth more employment opportunities will be created, especially for
African workers, while it will also further assist in rapidly reducing the wage gap. With South Africa’s
high unemployment, generating jobs, even at low wages, will dramatically improve the overall income
distribution.
These findings are confirmed by Simkins ( ) – at a % economic growth rate, African per capita
income would have increased by . % per annum from to , while that of whites would have
declined by . % per annum.
A further consideration is that the white male labour force is no longer increasing – as indicated in
section . , there is, in fact, stagnation in the employment of whites. Higher economic growth will
therefore lead to increased employment opportunities for disadvantaged people such as black people
and women. For many years unemployment will remain at very high levels in South Africa, and only a
limited number of people will benefit from programmes to reduce inequalities if there is not a
substantial increase in employment opportunities through much higher economic growth.
The fear that positive economic growth on its own will have a systematic negative effect on the
distribution of income is unfounded (Deininger & Squire, : ). There is a strong systematic
relationship between overall growth and growth in the income of the poorest % of the population:
economic growth has a strong positive effect on the poor (Deininger & Squire, : ). Economic
growth will thus reduce inequalities.
It should be emphasised that the analysis of inequalities above suffers from an important
shortcoming. It does not measure the dynamics of the labour market from the point of view of the
underprivileged. The extent of inequality is one thing, but even more important is the extent to which
the labour market opens up opportunities for the underprivileged and increases their social mobility.
If there are enough measures to assist the lower classes to improve their standard of living (among
others, through active labour market policies) and there is an adequate so-called “social wage”, then
the problem of inequality will be less serious than in the absence of these factors. The term “social
wage” refers, among other things, to the aggregate of social benefits available wholly or partly free to
individuals and families, and which are paid for by the community through various forms of taxation.
They may include free education, free medical services, housing assistance, etc. (see section . . ).
The impact of inequalities on economic growth

One final point needs to be made on the relationship between economic growth and inequality. The
findings of Van der Berg and Simkins above show that reduced economic growth has a negative effect
on reducing inequalities. However, inequality also seems to have a negative impact on economic
growth (see accompanying box).
Policy issue: does income inequality hinder economic growth?
International evidence shows that economic growth is slowed by a high degree of income inequality (see sources quoted
in the ILO Review, : ). A reduction in inequality will therefore probably promote higher economic growth, to the
benefit of the country. However, Deininger and Squire ( ) found that the negative impact of income inequality on
growth is not very strong. In contrast, inequality of assets, as measured by the distribution of land, exerts a significant
negative effect on economic growth.
What are the mechanisms through which an unequal distribution of income or assets might affect growth? The theory is
(or was, at least until recently) that a low level of income inequality stimulates economic growth by: inducing large
increases in savings and investment; contributing to political and macroeconomic stability; increasing the efficiency of
low-income workers; and increasing the demand for domestic goods. In societies where distributional conflict is severe
because of unequal income distribution, political, fiscal and regulatory decisions are likely to result in populist policies
that limit private appropriation across a broad spectrum of society. Furthermore, crime is probably more endemic in
unequal societies, resulting in assets that could have been used for productive purposes being used for protection
against criminal activity. In addition, education quality in unequal societies is also likely to vary sharply, entrenching racial
and ethnic prejudices, lack of access to labour markets, and the continuation of inequalities.
The combination of factors found in unequal societies is thus likely to be damaging for economic growth because
incentives for productive growth depend on the general ability of individuals to appropriate the returns of their effort, and
this hinges on tax, regulatory and other policies, as well as the skills of the population. In addition, such policies are likely
to reduce investment incentives.
However, Deininger and Squire ( : ) have found that evidence does not support this theory. They have found rather
that access to credit is the mechanism through which inequalities in assets affect economic growth. Access to credit
markets, which allows growth-generating investment in physical and human capital, is conditional on the ownership of
assets. However, if assets are distributed very unequally, fewer individuals will have access to credit and be able to
make investments in physical and human capital, resulting in lower growth. The researchers therefore call for
redistributive policies that enhance people’s access to credit markets. However, they warn that the pursuit of a
redistributive strategy at the expense of investment could actually decrease the income of the poor (Deininger & Squire,
: ).
SUMMARY
Even a rapid reduction in the wage gap will have a relatively small impact on per capita racial income distribution,
especially between white and African workers. The main instrument for changing primary income distribution is
economic growth, because of its impact on job creation.
Evidence suggests that a high level of inequalities, particularly in the distribution of assets, has a negative effect on
economic growth.
. . Education and training

South Africa is experiencing serious backlogs in respect of its educational system, and these will have
to be addressed if inequalities in society are to be reduced. Various educational programmes are at
issue: literacy and numeracy training; vocational guidance; career education; and an integrated
educational system that enjoys credibility among the majority of learners, to name but a few. In
respect of training, it is clear that South African enterprises do not provide enough, and attention
should be given to, for instance, management training and accelerated training in various
occupations. Education and training is discussed in more detail in Chapter .
. . Entrepreneurship
Disadvantaged groups are underrepresented among owners of the wealth-creating instruments of the
economy (see section . . ). Attention needs to be given to the accelerated training and
development of entrepreneurs, for example through special training and financing programmes.
Large contracts could also be broken up into smaller ones, which could be allocated to enterprises
owned by disadvantaged groups. State tenders, subcontracting, franchising, and special housing
programmes in which enterprises owned by disadvantaged groups are involved could also be
considered. Joint ventures between white-owned and African-owned enterprises are now a popular
method of involving previously disadvantaged groups in wealth creation.
. . Government employment and tender practices

In some countries, including South Africa, the government prescribes codes of conduct or guidelines
for reducing inequalities and implementing black-advancement programmes in respect of companies
tendering for government contracts (Ehrenberg & Smith, : ). Such measures also have
disadvantages. The government, for instance, should give higher priority to the proper performance
of the contract than to the enforcement of non-discriminatory or black-advancement provisions.
Employers might also attempt to comply with the requirements by moving disadvantaged persons
from non-governmental to governmental projects, which on balance would not benefit the
disadvantaged groups.
The Employment Equity Act contains measures to ensure that enterprises tendering for state contracts
have affirmative action plans in place. The BBBEE Codes discussed later in this chapter are also a
method of ensuring that companies that tender for state contracts comply with certain objectives to
eliminate inequalities and promote black-advancement programmes.
Real equality is not to be decreed by law. It cannot be given and it cannot be forced.
R M
. . Legal remedies
Legal remedies could consist simply of negative remedies, such as placing a prohibition on any type
of discrimination, or could be enhanced by positive actions, such as affirmative action programmes.
The constitution prohibits unfair discrimination, while it allows positive measures to promote equality.
This approach is echoed in several laws, for example the Employment Equity Act as discussed below.
A shortcoming that relates to legal remedies in general is that the courts are usually not very
accessible. In addition, legal procedures are very slow, expensive and uncertain. Some countries (e.g.
the US) have attempted to address this shortcoming by introducing special bodies such as the Equal
Employment Opportunities Commission (Marshall et al., : – ). Such bodies normally have
powers to investigate complaints, conduct public hearings, seek court orders in cases of
discrimination and publish studies on discrimination. They can also assist with legal actions against
specific employers, launch programmes to educate and reorient employers, and conciliate in cases of
possible discrimination. They can even launch special programmes, for instance equal pay for work of
equal value or affirmative action.
It should be noted that equal opportunities are not the same as equal treatment. Equal treatment
simply means that there is no apparent discrimination in respect of employment or promotion of
workers. However, especially in the light of South Africa’s history of apartheid, disadvantaged groups
do not have the tools to compete on an equal basis with advantaged groups. It would require
additional training, development, mentorship, bridging programmes and various other remedial
actions in order to “level the playing field”. Only once this has been done will there be a greater
equalisation of opportunity.
Employment Equity Act

South Africa’s Employment Equity Act (Act No. of ) consists of the following main elements:
The act prohibits direct or indirect unfair discrimination by any employer on any arbitrary ground,
and lists a total of grounds, among them race, gender, age and sexual orientation. Excluded
from the definition of unfair discrimination are affirmative action measures and preferential
treatment on the basis of the inherent requirements of a job. Applicants for employment are also
protected by this part of the act.
Affirmative action measures apply only to so-called “designated employers” (mainly excluding small
employers). This aspect of the act is discussed in section . . .
The act also compels employers to reduce so-called “disproportionate income differentials” (see
section . for a criticism of this approach).
The act also makes provision for a special body, the Commission for Employment Equity. It advises the
minister on issues relating to employment equity, relevant codes of conduct and regulations. It may
also do research and make awards recognising achievements of employers in furthering the purpose
of the act. It consists of a chair and eight other members representing the state, organised business,
organised labour, and organisations representing community interests. This body does not deal with
any workplace disputes. These are dealt with by the CCMA, while complaints may also be lodged with
the labour inspector.
The act has been criticised on a number of grounds, for example that it is reracialising South African
society by categorising racial groups that must receive preferential treatment, rather than individuals
who have been discriminated against. Furthermore, the act is said to deny employers the opportunity
to apply merit in appointing or promoting persons, in that it will no longer be the best person for the
job who is appointed or promoted, but a person from a designated group who is no more than
suitably qualified.
Jefferies ( : ) has, for instance, objected to the act on three grounds:
It mandates state intervention in the private domain and this exceeds the proper role of
government in a liberal democracy.
Such intervention on a racial basis is counter to the constitution, which permits rather than requires
race discrimination in the form of affirmative action.
It fails to address the country’s key problem – the poverty attendant on widespread unemployment
– and instead adopts an approach likely to provide a disincentive to the job creation so urgently
required.
What is clear is that legislative mechanisms alone will not be sufficient to achieve employment equity
in South Africa. Special programmes of affirmative action and black advancement will be required and
these can include literacy training, quality education, accelerated training, mentorship, targeted
recruitment practices and various other programmes. These must be aimed at ensuring that the
composition of the labour force, and especially of skilled occupations, is more fully representative of
South Africa’s population, but without reducing standards and efficiency in the long term.
As these programmes will assist only those already in employment, special attention needs to be
given to reducing the high level of unemployment that is being experienced. Policies to implement
affirmative action and those to increase economic growth will most likely conflict rather than be
complementary, unless they are very delicately fine-tuned. The electorate, however, is unlikely to
accept this, and might insist on more robust equality measures, which will not be in the interests of
economic growth and the unemployed, either in the short or long term.
Two important legislative interventions are discussed below, namely, equal pay for work of equal
value and affirmative action.
SUMMARY
The Employment Equity Act prohibits discrimination and imposes a duty on all except quite small employers to ensure the
equitable representation of black people, women and disabled persons.
. EQUAL PAY FOR WORK OF EQUAL VALUE
Equal pay for work of equal value is an approach that is often legislated, but not always directly, as
will be seen below.
The phrase “equal pay for work of equal value” has replaced the older terminology “equal pay for
equal work”. Although it is relatively easy to determine if work is equal (i.e. the same), this terminology
was regarded as relatively restrictive. In the US, for example, one of the first pieces of legislation
dealing with gender discrimination was the Equal Pay Act of . This act prohibited different wage
rates for men and women if they did “equal work on jobs, the performance of which requires equal
skill, effort and responsibility, and which are performed under similar working conditions”. However,
as discussed above, one reason for discrimination is the crowding model, where most of the
employees doing one particular job (such as clerical, sales, typing, etc.) are women, and those doing
another are men. A discriminating employer can thus relatively easily bypass the provisions of this
legislation by not employing men and women to do the same job.
The inherent vice of capitalism is the unequal sharing of blessings; the inherent virtue of socialism is
the equal sharing of miseries.
W C
Equal pay for work of equal value (or work of comparable value, or work of comparable worth) is
therefore a much wider term and is a principle whereby remuneration rates are established on the
basis of job content (Barker & Holtzhausen, ). “Value” is a concept that evades adequate
definition and could, for instance, relate to the “value” of the work performed or the “value” of the
employee to the employer. In terms of the wider concept, a female secretary or clerk should, for
instance, receive the same remuneration as a male truck driver or construction worker if the
knowledge, skill, value to the employer, and so on, of the two workers are equal (Hamermesh & Rees,
: ). An attempt must therefore be made to look at the inherent characteristics of the job, and
value is often determined by job evaluations.
Arguments for and against
The main argument in favour of enforcing the principle of equal pay for work of equal value is that
the labour market is neither efficient nor fully competitive. There are many distorting factors in
terms of the remuneration of workers. A system of job evaluation provides a better indication of
worth than existing wages or job content. Wages are distorted by institutional factors such as past
discriminatory practices or the preponderance of discriminated groups in some occupations
compared to others, which drives down wages in those occupations. The differences in wages in
these instances do not reflect differences in productivity and this can be rectified by basing all
remuneration on job evaluation.
A further argument in favour of the introduction of this principle is that it helps enhance overall
gender equality in the world of work by addressing gender pay differentials (Oelz, Olney & Tomei,
: ), as this principle helps: improve women’s time allocation between paid and unpaid work
over their life cycle; reduce women’s financial dependence; reduce the likelihood of low-income
female-headed households remaining poor; decrease recourse to child labour; lower labour
turnover; increase productivity; and grow the enterprises’ capacity to attract and retain the best
workers.
The main theoretical argument against the enforcement of the principle of equal pay for work of
equal value is that even though there are distorting factors in the labour market, it is nevertheless
reasonably competitive and wages are paid more or less according to productivity. Job evaluations,
on the other hand, are subjective and arbitrary, and different systems are in operation, which often
do not have the same outcomes. Furthermore, the introduction of this principle might even reduce
the number of female or black workers in certain underrepresented occupations if it results in
substantial pay hikes for these workers. If higher wages have to be paid, for instance in traditionally
female occupations because their value has become equal to that of some male occupations, it will
attract even more females to these occupations and not necessarily to the male occupations. It
might, therefore, reinforce the sexist or racial pattern of occupational structures. It might also
encourage mechanisation.
A further argument against the introduction of this principle is that there will be an efficiency loss
for the whole economy. Wages are basically set according to productivity levels, and job evaluations
do not necessarily measure productivity. In addition, the economy will not be able to adjust rapidly
to changing circumstances. There are always undersupply or oversupply situations in occupations
owing to changes in, for instance, the demand for products or in technology. This means that
workers have to be recruited for undersupplied occupations and they must be trained. However, if
employers are forced to remunerate workers at equal levels in different occupations, there is no
incentive for workers to want to undergo retraining for the undersupplied occupations because the
wage rates cannot be increased.
A major consideration in respect of the concept of equal pay for equal work or even equal pay for
work of equal value is that it will apply to only a minority of skilled black workers. This will not
address the problem of before-the-market discrimination, i.e. that black workers do not have the
necessary education, training and opportunities for upward mobility.
Van Niekerk ( b) maintains that substantially more black employees may benefit from equal value
claims than is suggested in the last point above. Existing occupational structures might tend to
undervalue the work traditionally performed by black people, while overvaluing work predominantly
performed by whites. The fact that the work may be rated as equivalent by the employer’s job
evaluation scheme is not definitive – equal value claims may question the assumptions that may be
embodied in such a job evaluation scheme, and this may affect the relative valuations of different
occupations.
The Employment Equity Act does not directly provide for equal pay for equal work. However,
discrimination in pay is covered by the prohibition on direct and indirect discrimination. In addition,
employers are obliged to reduce what is referred to as “disproportionate” income differentials, an
issue which is discussed in more detail in section . .
SUMMARY
Equal pay for work of equal value is an important mechanism to address inequalities, but is also subject to a number of
shortcomings, among others that job evaluations might be subjective and arbitrary, and that it further inhibits the efficient
operation of the labour market.
. AFFIRMATIVE ACTION
Affirmative action refers to policies and practices aimed at redressing social, economic or educational
imbalances or inequalities arising out of unfair discrimination against certain groups (Barker &
Holtzhausen, ). Relying only on general policies of economic growth with equity and a prohibition
on discrimination for removing such group inequalities will not necessarily show results (at least not in
the short to medium term), leading to resentment or even civil conflict.
The South African Law Commission, in its report on group and human rights ( : ), gave this
opinion on affirmative action:
Affirmative action is a recognised way of promoting the principle of equality of opportunities in

societies where this principle has suffered as a result of discrimination or where less developed
persons have to compete with more developed persons. It is recognised in international law as
being non-discriminatory, so long as it is temporary and is not enforced against the will of the
minority. In appropriate cases it does not amount to discrimination, since the fact that those who
have lost their positions in a programme of affirmative action properly executed do not actually
lose anything they would have had in a fair race. Programmes of affirmative action must be
implemented with great care if reverse discrimination is to be prevented.
Some countries have successfully implemented a programme of affirmative action, and Malaysia is
one such example.
The UK, on the other hand, has avoided the issue of affirmative action and has opted for a prohibition
on direct AND indirect discrimination. Indirect discrimination refers to certain rules or procedures that
appear neutral but may disproportionately affect certain groups (i.e. women or black people) (Barker
& Holtzhausen, ). If an employer sets a requirement for employment or promotion that has a
differential impact on, for example, African or female workers, it is indirect discrimination, even
though it does not directly discriminate. An employer is allowed to present good business reasons for
the particular requirement set. It should be noted, however, that the problem of inequities and
discrimination (and even more importantly, the legacies thereof) is probably smaller in the UK than in
certain other countries such as South Africa and Malaysia.
Even though affirmative action policies have generally been “quite successful in achieving their goals”
(United Nations Development Programme, : ), one worrying aspect is that most countries that
have adopted such policies have also experienced an increase in overall inequalities in personal
income (accompanied by growing inequality within the underprivileged group). This is also the case in
South Africa, as discussed in section . . above.
The final recommendations of the International Panel on ASGISA (Accelerated and Shared Growth
Initiative South Africa) (Hausmann, : ) included comment that BEE rules in South Africa
are increasing the demand for high-skilled previously disadvantaged South Africans at a time
when they are already facing very high and rising demand. To the extent that this tightens the
skills constraint at the top it lowers the demand for lesser skilled workers and thus widens income
and opportunity disparities among the previously disadvantaged. Encouraging the retention of all
high skilled South Africans and the attraction of foreign high skilled persons will be crucial to
limit wage inequality and facilitate the creation of jobs for the less skilled and thus achieve shared
growth.
The report suggests that
there is ample room to improve empowerment through job creation, training and supplier
development for people currently at the bottom of the income distribution. We shall propose
that the scorecard be rebalanced to encourage these latter types of activities. It would also be
useful to define sunset clauses for BEE: if the policy is successful, it should become redundant
(Hausmann, : ).
It is the mark of the cultured man that he is aware of the fact that equality is an ethical and not a
biological principle.
A M
. . The case for affirmative action

Arguments in favour of introducing programmes of affirmative action are as follows:
(a) The dismantling of discrimination cannot take place instantly, simply by getting rid of
discriminatory legislation. After many years of such legislation, the people who were
discriminated against are far behind in terms of ownership of the wealth of the country and the
filling of skilled occupations (Marshall et al., : ). Goldstone ( : ) says
I have no doubt that it will take many decades to ameliorate, let alone to obliterate, the many
negative consequences of centuries of racial oppression in South Africa. If positive steps are not
taken to bridge the racial gap, then South Africans will never live in harmony with each other and
the economy will not have the manpower so necessary to drive it efficiently.
(b) African people should be provided “with the same boots and straps as you do whites before you
shout the platitude ‘pull yourself up by your bootstraps’ to those who have neither boots nor
straps” (Khoza, : ). Affirmative action is not a form of reverse discrimination, but a method
of providing reparation to benefit members of a group that has been discriminated against on
account of race (Faundez, : ).
(c) From an economic point of view, the argument is that markets do not function properly. If
discrimination is the result of, for instance, statistical discrimination (see section . . ),
intervention by government in the form of affirmative action might bring about a change in
employers’ perceptions that individuals from certain groups are less productive than those from
others. If discrimination is caused by market imperfections (see section . . ), it will not be
eliminated unless there is intervention by government.
An example of the first type of discrimination and the positive impact of government intervention is
provided in Figure . (a). This is an adaptation of the isoquant and isocost curves discussed in
section . . . In this case, the option that the employer has of choosing either capital or labour as
production factors is now replaced with the option of white or black labour. These are reflected along
the vertical and horizontal axes respectively. The isoquant indicates the various combinations of white
and black labour to produce a specific level of output. The isocost curve shows the cost combinations
of using black and white labour. As will be recalled from Chapter , cost minimisation or profit
maximisation takes place where the isoquant and isocost curves just touch, i.e. are tangent.
The discriminating employer will employ more white than black workers, even if this is more
expensive. This is indicated by the point P in Figure . (a). This provides less profit than point Q,
because it lies on a higher cost curve. In fact, the isoquant and the isocost curves are not tangent to
each other at point P. Government policy to force employers to employ more black labour will bring
about a movement from P to Q, the profit-maximising position.
FIGURE . Profit maximisation with affirmative action
. . The case against affirmative action

The arguments often put forward in opposition to affirmative action are the following:
(a) Affirmative action programmes reduce economic efficiency because the best person for the job is
not necessarily appointed. This implies that recruitment standards are reduced and that
competition, the basic principle of markets, is negated. However, if these standards are reduced
only in the short term to enable the victims of discrimination to compete on an equal basis, it will
not necessarily be detrimental to the economy. Depending on the period and manner in which
standards are relaxed, it might even be beneficial. This is because the productive potential of the
victims of discrimination, which has not been fully developed or utilised before, will be unlocked
by affirmative action.
An example of the inefficiency that can be caused by government intervention can be illustrated by
using Figure . (b). Let us assume that the employer is in this instance not discriminating. He is
therefore at the profit-maximising position R, where the isoquant and isocost curves are tangent. If
the government now forces the employer to employ more black workers, the employer will move to
position S, where more black and fewer white workers are employed. However, this point moves off
the profit maximising point R, because it falls on a higher isocost curve.
(b) If standards are reduced over the long term or permanently, a positive economic outcome is
much less likely. In fact, affirmative action will tend to push up the wages of the beneficiaries of
affirmative action and, if this group forms the majority (as in South Africa), this could be especially
harmful to employers and the international competitiveness of the country.
Affirmative action programmes are unlikely to be temporary. Sowell ( : ) says, “preferential

programmes, even when explicitly defined as temporary, have tended not only to persist but to
expand in scope”. Even though a country such as India decided to end “reserved legislative districts”
ten years after the constitution of , these districts have been extended since at ten-year intervals
(United Nations Development Programme, : ). Affirmative action programmes are therefore
very difficult to end, except if linked to specific discriminatory practices of the past.
An interesting proposal in this regard was that of Judge Goldstone at the Labour Law
Conference. He suggested that affirmative action programmes should not be based on racial criteria,
but on other temporary and non-racial criteria, for example be aimed at persons who were educated
under the segregated educational system. Once education departments were integrated and quality
improved, persons educated in an integrated system should not be entitled to benefit from
affirmative action programmes, and such programmes should gradually and automatically disappear
as the disadvantaged groups move through the labour force. Other groups that could qualify for
selection to affirmative action programmes would be the homeless and persons registered as
unemployed.
(c) A further possible disadvantage of affirmative action is that the victims of discrimination may use
affirmative action as a source of entitlement, in the same way as whites used the system of
apartheid as a source of privilege. This could stigmatise those who do not need help by
suggesting, for instance, that a person has been appointed to a position because he or she is
black, and not because of merit. This will undermine his or her self-esteem and reinforce the myth
of racial inferiority. Racism will therefore not be eradicated and racial hostilities will not be
reduced. Affirmative action could amount to front-office tokenism and could negate the
considerable strides made by many victims of discrimination in terms of moving into highly
skilled occupations.
(d) Affirmative action could also be problematic if it is applied in such a way that individuals are
deprived of existing rights, although there may be ways of circumventing this problem (Marshall
et al., : ). For example, preference may be given to a black applicant for a job if that
applicant and a white applicant are nearly equally qualified. In this case, the white applicant will
not be deprived of preexisting rights.
Quota systems, where the victims of discrimination must fill a certain number of skilled jobs, do
not, however, fulfil this requirement, as discussed below.
. . Affirmative action in the Employment Equity Act

The Employment Equity Act stipulates a duty to apply affirmative action to so-called “designated
employers” – those with more than employees and those with fewer than employees but with
an annual turnover above a defined amount. It also requires that such employers introduce a process
of ensuring that, over time, there is progress towards employment equity.
Designated employers must introduce affirmative action measures to ensure that suitably qualified
people from designated groups (black people, women and disabled people) have equal employment
opportunities and are equitably represented in all occupational categories and levels in the workforce.
The act requires designated employers to prepare plans for progress towards employment equity, and
to submit these plans to the Department of Labour for assessment. This approach aims to encourage
workplace transformation through self-regulation by employer and employee parties. The eventual
success of the act will, however, largely depend on how the Department of Labour applies the act and
its approach to the employment equity plans submitted by employers.
The act specifically excludes the imposition of a quota system. However, the employer must set
numerical goals to achieve the equitable representation of suitably qualified persons from designated
groups and a timetable for achieving the goals.
The director-general of the Department of Labour can review an employment equity plan to
determine if an employer is complying with the act. Factors that the director-general (or for that
matter any other person or body applying this act) must take into account include:
Whether suitably qualified persons from designated groups are “equitably represented” within each
occupational category and level, and the following factors will determine this:
– The profile of the national and regional EAP (i.e. demographics)
– The pool of suitably qualified persons from designated groups from which the employer can
appoint or promote persons
– The economic and financial circumstances of the employer and of the particular sector in which
the employer operates
– Present and planned vacancies and the labour turnover of the employer
Progress made by other employers in the same sector and operating under similar conditions
Reasonable efforts made by the employer in achieving employment equity
The extent to which the employer has eliminated employment barriers that adversely affect persons
from designated groups.
SUMMARY
Affirmative action is not a form of reverse discrimination, but a method of providing reparation to benefit members of a
group that has been discriminated against on account of race.
Some countries have been reasonably successful in implementing affirmative action, although they have been
assisted by high economic growth rates.
. . Quotas versus Employment Equity targets: the BBBEE Codes

The Employment Equity Act requires an employer to set numerical goals to achieve equitable
representation of suitably qualified black persons, women and disabled persons and a timetable for
achieving these goals. It specifically excludes the imposition of a quota system.
However, the codes issued by the Department of Trade and Industry in in terms of the Broad-
based Black Economic Empowerment Act (BBBEEA) (No. of ) introduce the concept of what is
called compliance targets, in effect externally imposed “quotas”, aimed at affirmative action for black
people and black women in particular (Department of Trade and Industry, ). The codes must be
applied by government departments and other agencies when awarding contracts, licences,
partnership arrangements and the like. The black economic empowerment (BEE) status of companies
has to be verified by verification agencies; these agencies award points depending on the extent to
which the companies comply with the targets. There are probably very few companies in South Africa
that do not feel the impact of the codes – not only does the BEE status of a company apply when
licences are awarded, but companies must also ensure that other companies supplying goods and
services to them comply with the codes in order to obtain procurement points.
The origin of these codes was the sectoral charters introduced in the early s. The Mining Charter,
for instance, required a % participation of historically disadvantaged South Africans (HDSAs) in
management within five years (RSA, ). The Financial Sector Charter has various five-year targets,
ranging between and % at various management levels.
The draft codes released for comment are quite detailed, and set compliance targets for ownership,
management control, procurement, training, enterprise development and corporate social investment.
The codes also set targets for employment equity, as outlined in Table . . This table also shows the
current skilled labour structure according to the latest available statistics. These apply to medium-
sized and large enterprises, as the codes have lesser requirements for small enterprises, and
microenterprises are totally excluded.
TABLE . BBBEE compliance targets compared to level of representation of black workers
Black people Black women

Compliance CCE Report Compliance CCE Report
targets 2013/14 targets 2013/14
Top management 50% 33% 25% 10%
Black people Black women
Compliance CCE Report Compliance CCE Report
targets 2013/14 targets 2013/14
Senior management 60% 40% 30% 14%
Professional and middle 75% 57% 38% 27%
management
Source: Republic of South Africa, Amended BBBEE Codes of Good Practice ( ); Commission for Employment Equity,
Annual Report –
The education and training system will probably not be able to adjust rapidly enough to provide the
skills required by the labour market over the short term. An estimated highly skilled black
people will be required to meet these targets over the ten-year period referred to in the codes. A
significant number of them will have to be skilled in occupations where mathematics and science are
core competencies. However, the education system produces fewer than African learners per
annum with higher-grade mathematics. Many of the degrees awarded are not in the disciplines that
the economy requires. For instance, less than % of degrees awarded are in engineering, life
sciences, and physical and computer sciences.
The number of skilled jobs will have to increase significantly over this period to change the
percentage composition. However, most new jobs are created in new businesses and in smaller
companies. Larger companies seldom increase their employment substantially. If they do not increase
employment, it will be exceedingly difficult for them to change their employment profile substantially
without actually retrenching current skills.
Even if it were possible for all companies to increase their employment of skilled people at the same
(high) rate as the total number of skilled people increased over the last few years, the targets still
appear to be unachievable. The number of skilled jobs has increased by . % per annum between
and (and a lower . % between and ). If it is assumed that this rate of increase
continues over the next years, and it is moreover assumed that the education and training system
provides the required number of black people with the right qualifications, and that only black people
are appointed to the new positions, then at the end of the -year period, black people will occupy
only about % of senior management, and just over % of professional and middle management,
and skilled technical and junior management positions.
The current targets thus ignore the country’s current dire skills shortage. The codes also ignore the
long lead times it takes for professionals to qualify, and thereafter the additional years needed to gain
experience, specialise and rise up the corporate ladder. The country should also guard against the
emigration of skills, even if these are the skills of white persons. Emigration of any skill is a very costly
haemorrhage. Not only is it an erosion of the country’s human capital in the form of the “free” export
of education, training and experience, but it also entails a loss of tax income for the country.
The codes impose external targets to be achieved irrespective of the circumstances of the enterprise
or sector, the wishes of the union(s) within the enterprise, or the outcome of negotiations about the
achievement of affirmative action. For instance, a sector might be in economic decline; very few new
positions are thus created, and with low labour turnover, affirmative action is much more difficult to
achieve.
The central challenge is creating a flexible framework under which realistic and sustainable black
economic empowerment can be facilitated. Setting targets and benchmarks has to take into account
the practical realities of implementation and must recognise that BEE is a process and not a once-off
event.
SUMMARY
The Broad-Based Black Economic Empowerment Codes set unrealistic targets for employment equity. These could
exacerbate the skills shortages and harm the country’s international competitiveness.
KEY CONCEPTS
affirmative action
before-the-market discrimination
crowding model
discrimination
discrimination coefficient
employment discrimination
equal pay for work of equal value
Gini coefficient
human capital discrimination
nepotism
nepotism coefficient
occupational discrimination
social wage
statistical discrimination
taste-for-discrimination
transaction costs model
wage discrimination
within-the-market discrimination
FOR STUDENTS
. Discuss the various types of discrimination often found in labour markets, with the aid of an example in each
case.
. Assume WM, WF and d stand for male wage, female wage and discrimination coefficient respectively, and a
hypothetical firm discriminates against male workers. Explain, in each of the following cases, which group of
workers would be hired.
. WM = R , WF = R ,d=R
. WM = R , WF = R ,d=R
. WM = R , WF = R ,d=R
. WM = R , WF = R ,d=R
. WM = R , WF = R ,d=R
. WM = R , WF = R ,d=R
. Assume WA, WW and n stand for Asian wage, white wage and nepotism coefficient respectively, and a
hypothetical firm prefers to hire white workers. Explain, in each of the following cases, which group of workers
would be hired.
. WA = R , WW = R ,n=R
. WA = R , WW = R ,n=R
. WA = R , WW = R ,n=R
. Explain statistical discrimination with the aid of an example.

. Explain the transaction costs model with the aid of an example.
. Explain, with the aid of a graphical illustration, how the differences in the elasticity of labour supply across two
population groups would lead to wage discrimination, assuming there is a decrease in labour demand in both
groups.
. Discuss the inequalities and discrimination in the South African labour market.
. Discuss the arguments for and against policies to ensure equal pay for work of equal value.
. What are the possible advantages and disadvantages of affirmative action?
. Explain, with the aid of a graphical illustration, that affirmative action does not always lead to cost
minimisation or profit maximisation.
SUGGESTED READING*
Bhorat, H., Van der Westhuizen, C. & Jacobs, T. . Income and non-income inequality in post-apartheid South
Africa: what are the drivers and possible policy interventions? Development Policy Research Unit, Working
Paper, / . Cape Town: University of Cape Town.
Burger, R.P. & Jafta, R. . Returns to race: labour market discrimination in post-apartheid South Africa.
Stellenbosch Economic Working Papers / . Stellenbosch: University of Stellenbosch.
Burger, R.P. & Jafta, R. . Affirmative action in South Africa: an empirical assessment of the impact on labour
market outcomes. CRISE Working Paper No. . Oxford: Centre for Research on Inequality, Human Security
and Ethnicity.
Finn, A., Leibbrandt, M. & Oosthuizen, M. . Poverty, inequality and prices in post-apartheid South Africa.
WIDER Working Paper / . Helsinki: United Nations University-World Institute for Development
Economics Research (UNU-WIDER).
Hodge, D. . Growth, employment and unemployment in South Africa. South African Journal of Economics,
( ): December.
Holzer, H. & Neukark, D. . Assessing affirmative action. NBER Working Paper . Cambridge: National
Bureau of Economic Research (NBER).
Leibbrandt, M., Finn, A. & Woolard, I. . Describing and decomposing post-apartheid income inequality in
South Africa. Development Southern Africa, ( ): March.
Oelz, M., Olney, S. & Tomei, M. . Equal pay: an introductory guide. Geneva: International Labour Organization.
Van der Berg, S. . Current poverty and income distribution in the context of South African history. Stellenbosch
Van der Berg, S., Louw, M. & Du Toit, L. . Poverty trends since the transition: what we know. Stellenbosch

Appendices
A Globalisation and the labour market
Outline of this appendix

A. Introduction
A. South Africa’s links with the world
A. South Africa’s competitive position
A. The impact of international trade on the labour market
A. The impact of trade liberalisation on the South African labour market
A. Addressing the challenges of globalisation
I’ve always had a weakness for foreign affairs.

M W
A. INTRODUCTION
New technological advances have led to a rapid increase in globalisation since the s and
especially the s. These advances include, for instance, greatly reduced transport,
telecommunications and computation costs, which have helped to overcome the natural barriers of
space and time that separate national markets (El Toukhy, ). Globalisation has also been spurred
on by large developing countries choosing to improve their investment climates in order to attract
overseas investors and open up to foreign trade.
Globalisation refers to the widening and deepening of international trade, finance, information and
culture (e.g. movies) in a single integrated world market (El Toukhy, ). According to another
description, globalisation is characterised, in particular, by an intensification of cross-border trade and
increased financial and foreign direct investment flows promoted by rapid liberalisation and advances
in information technology (Daouas, ). It thus refers to the increasing integration of economies
around the world.
Globalisation has an important effect on the labour market. Research points to a sharp increase in the
effective global pool of labour (International Monetary Fund, ). This global pool of labour is
accessed (especially by advanced economies) through imports and immigration. The imports of final
goods (and, to a lesser extent, intermediary goods) is the more important and faster-expanding
channel, in large part because immigration remains very restricted in many countries.
Countries are also becoming more closely connected as trade barriers are dismantled and import
tariffs reduced. This puts workers in one country in competition with workers in another country and
increases the danger of a “levelling down” in wages and working conditions. The flow of money and
goods between countries has also increased substantially. Numerous multinational corporations have
annual sales totalling more than the GDP of many countries, including South Africa.
In pursuit of a more outward-looking economy, South Africa introduced a far-reaching trade

liberalisation programme in the s, especially after the democratic elections in . The country
reduced its average import tariffs on manufactured goods from % in to . % in (ILO, as
quoted by Bezuidenhout, : ) to encourage export-led growth and to reap the benefits that
globalisation provided for many countries. This formed an important element of the government’s
economic policies to create an environment conducive to economic growth.
In some cases the liberalisation programme even exceeded the requirements of the General
Agreement on Trade and Tariffs (GATT) and the World Trade Organisation (WTO) negotiations. These
included tariffs being simplified and reduced and other trade-related measures introduced to abolish
practices that contravene WTO rules, such as local content requirements and export incentives.
It should be mentioned from the outset that the government has been criticised sharply for reducing
tariffs more rapidly than required by our international obligations (see, for instance, the discussion in
Nattrass, b: ). This would, it has been said, contribute to a contraction in employment because of
the structural adjustments required to compete with the products of other countries, both within and
outside South Africa. This issue is discussed later in this appendix.
However, as pointed out by Bhorat et al. ( : ), there is disagreement about the degree to which
trade has been liberalised when considering the effective rate of protection (ERP). They quote studies
that maintain that even though the average ERP has fallen in line with the fall in nominal tariffs, it still
remains fairly high, particularly in manufacturing, where it averaged about . % in . Despite
significant declines, ERP appears to remain quite high for the tobacco, textiles, clothing, footwear and
furniture sectors.
In this appendix, attention is first given to the openness and competitiveness of the South African
economy. The impact of globalisation is then discussed, with specific reference to the labour market.
Many countries, including South Africa, face the question of how the challenges of globalisation
should be addressed. Some case studies and a few other approaches in this regard are outlined.
A. SOUTH AFRICA’S LINKS WITH THE WORLD
South Africa has a relatively open economy with regard to trade, investment and the mobility of
labour. Exports and imports form a large part of total domestic production, and foreign capital is also
important because South Africa does not save enough to finance the required level of investment.
Together, South African exports and imports form on average about % of its GDP. Some countries
such as Malaysia, Singapore, Hong Kong and the Netherlands are far more trade-dependent than
South Africa, whereas the least trade-dependent countries are Japan, the US, Argentina and Brazil
(Mohr et al., : ).
South Africa remains an important exporter of mineral products, such as platinum, coal, iron ore and
diamonds. However, exports of gold have declined substantially since the early s. The largest
imported products are machinery, vehicles, aircraft and vessels, as well as oil and defence equipment
(Mohr et al., : ).
However, manufactured exports from South Africa are generally capital and technology intensive
compared to other middle-income countries. Furthermore, export growth has been mediocre
compared to other dynamic emerging economies (Edwards & Schoer, : ). The share of
unskilled labour-intensive products in total exports has risen but remains low. This shows that South
Africa has failed to make effective use of the comparative advantage that an abundant supply of
labour provides. Edwards and Schoer state that a combination of supply-side policies (lack of skills,
lack of infrastructure, etc.) and labour market institutions (unions, bargaining councils, role of labour
laws) played a role in this regard.
Exports increased relatively sharply in the s, with the volume of exports (excluding gold)
increasing on average by . % per annum during this period. Some factors contributing to this
growth were the favourable political climate and the depreciation of the rand, although over the
medium to longer term this latter advantage has been negated by the increase in domestic labour
costs (Bureau for Economic Policy and Analysis, ). This is probably an important reason for South
Africa failing to increase its stake in world dollar exports, with its exports still forming less than % of
total world exports.
Owing to a shortage of domestic savings to finance investment, South Africa has had to rely heavily
on foreign investment to close the gap between savings and investment. However, much of the inflow
of foreign investment has been in the form of short-term capital, for instance investing in shares on
the JSE Limited, rather than direct investment (in productive capacity), which is likely to be more
stable.
Through the interlinking of product markets and the high mobility of skills, labour markets too have
become closely linked with those of other countries. This applies to both skilled and unskilled labour.
As far as skilled labour is concerned, remuneration and taxation practices in South Africa relative to
other countries have an impact on the country’s ability to attract or retain skills in high demand the
world over. Section . . shows that in the past South Africa could depend on a net immigration gain
to complement its skills. However, this has turned into a net loss over the last couple of decades,
which means that the country is losing skills that have been developed at great cost to society.
With regard to the lower skills levels, the southern African region is already rapidly integrating with
regard to labour, with substantial numbers of workers from neighbouring countries working in South
Africa legally and illegally. In addition to the mobility of labour between countries, jobs may in many
cases also be quite mobile. Employers may, for instance, move their businesses to neighbouring
countries should the prospect of profits be better in light of wage levels, taxation, political stability,
etc.
SUMMARY
South Africa has a relatively open economy with regard to trade, investment and labour markets. Exports and imports
form a large part of total domestic production. Foreign capital is also important because South Africa does not save
enough to finance the required level of investment. There is an outflow of skilled labour and an influx of legal and illegal
unskilled labour.
A. SOUTH AFRICA’S COMPETITIVE POSITION
A country’s international competitive position depends on a number of factors. Apart from the price
of the product, factors such as the quality of the product, the ability to produce customised products,
on-time delivery, general good service to the customer and movements in the exchange rate all play a
role. With regard to the latter, a depreciating currency, which South Africa has often experienced,
makes exported products less expensive and therefore more attractive to foreign buyers. The Labour
Market Commission ( : ) argues that the depreciation of the rand has lowered South Africa’s unit
labour costs relative to those of its major trading partners.
However, a depreciating currency does not have advantages only; it also poses certain threats. For
instance, it leads to higher import prices that might increase local inflation and wipe out any benefit
of a depreciating currency.
Taking account of various factors, it seems as if South Africa’s competitive position does not always
compare favourably with that of other countries, as is discussed below.
A. . World competitiveness: benchmarking South Africa against other countries

According to the World Economic Forum, South Africa fares poorly in terms of its international
competitiveness (website: http://www.weforum.org). The competitiveness position of a country is
determined according to nearly criteria of hard data as well as a survey of the opinions of
executives in different countries. The survey is given a weighting of one-third in the final result.
The difficulty is that we have an industrial base with so many characteristics of an industrial museum
or of an industrial hospital.
B J
In / South Africa occupied the th position on a list of countries. Some of the countries
that regularly occupy the top spots are Switzerland, Finland, the US, Germany, Japan, Singapore,
Sweden and Denmark, while Malawi, Mozambique and Burundi are among the lowest. Other
countries in sub-Saharan Africa that did reasonably well included Mauritius (ranked th), Rwanda
( th) and Botswana ( th).
South Africa performed particularly poorly as far as its labour market is concerned, occupying the th
spot. Some of the areas where South Africa suffered the most serious competitive disadvantage
according to the survey were the following:
Cooperation in labour–employer relations (ranked last)

Flexibility of wage determination ( th)
Hiring and firing practices ( th)
Pay and productivity ( th)
Comparing labour laws internationally
Employers often state that labour legislation is an important reason for the lack of flexibility and thus the low productivity
referred to above. However, it is important to emphasise that international comparisons of labour laws are always
difficult. As discussed in greater depth in section . . , such comparisons are influenced by a number of other factors:
It does not make much sense to compare individual labour law provisions, as it is often the cumulative impact of laws
that have an effect.
The impact of legislation greatly depends on union strength: unions often bargain to increase benefits over and above
the minima required by law, thus they cause high wage differentials.
It is also often not only the level of legislative burden, but the change in the level that results in negative perceptions
about the burden of labour laws.
The quality of human resources, including the quality of education, has an impact on the extent to which the employer
can improve productivity to compensate for labour laws.
SUMMARY
South Africa’s competitive position is weakened by factors in connection with labour market efficiency, such as
cooperation in labour-employer relations, flexibility of wage determination, hiring and firing practices, as well as pay and
productivity.
A. . Unit labour cost compared internationally

With rapidly increasing globalisation, producers face much stiffer competition than before from
producers in other countries in both domestic and international markets. The prices of products
therefore play a particularly important role in determining competitiveness. Prices are determined by
the interplay between cost and productivity, and this section is dedicated to the joint impact of cost
and productivity.
In general, an increase in the cost of labour will negatively affect a country’s international
competitiveness, unless compensated for by an increase in productivity. The combined impact of
wages and productivity is measured by unit labour cost. As will be recalled from section . , unit
labour cost is the cost of labour of one unit of output. As capital moves more freely around the globe,
labour costs are becoming increasingly important in determining comparative advantage (Wood,
).
An increase in wages and an increase in productivity might take place in tandem, and unit labour cost
(and thus the competitive position) will remain unchanged. In some cases, an increase in wages might
actually improve a country’s international competitiveness and thereby promote economic growth
and employment. This will happen where a situation of efficiency wages exists, i.e. where wage
increases result in productivity increases (see section . . ). However, productivity does not normally
improve sufficiently (if at all) to fully compensate for wage increases, and this leads to an increase in
unit labour cost.
The research results regarding South African wages and productivity show mixed results. Golub (as
quoted by Edwards & Schoer, : ), for instance, found that South African labour costs in the
s were competitive compared to industrial countries. Even though labour productivity was low,
the relative wages were even lower. However, he also found that South African labour costs were not
competitive compared to almost all developing countries that were major exporters of manufactured
goods. Even though labour productivity in South Africa was relatively high compared to these
countries, the relative wages were even higher.
Edwards and Golub ( : ) also compared both the levels and the changes in productivity levels
in South Africa and a large number of other countries, both developed and developing. They found
that the South African labour and total factor productivity were well below that of developed
countries, and below that of some developing countries, but above that of others. South African
relative productivity growth in the s and s was also weak, especially when compared to other
newly industrialising countries in Asia and South America. These findings are relevant to most
manufacturing industries.
Van Dijk ( ) found that there is a considerable gap in labour productivity between South Africa
and the US, which is widening continuously over time. In , labour productivity in South Africa
stood at % of the US level, while it was only % in . When one also takes into account labour
costs by comparing unit labour costs, South Africa is competitive vis-àvis the US, although some
manufacturing industries show higher unit labour costs than the US.
Productivity SA has indicated that unit labour costs in South Africa are increasing at a much more
rapid rate than in its major trading partners. Figure A. compares the increase in unit labour costs in
South Africa with that of the country’s main trading partners.
FIGURE A. Unit labour cost in manufacturing in various countries, – (indices: = )
The graph shows how much more rapidly unit labour costs at current prices in manufacturing in South
Africa increased compared with those of its most important trading partners. The poor wage–
productivity ratio in South Africa probably contains an element of socioeconomic correction. This
should be taken into account when making direct comparisons with other countries, although this
does not help to make the country more competitive.
However, to compare unit labour cost at current prices when inflation in South Africa is higher than in
its major trading partners can be misleading to some extent. One can expect increases in labour
remuneration to keep up with inflation, i.e. if South Africa’s inflation is higher than that of its major
trading partners, wages will also increase more rapidly.
By comparing increases in unit labour cost in dollar terms, one takes account of inflation, because the
effect of inflation will eventually be reflected in a depreciation of the national currency. This is
precisely what has happened in South Africa – there was a long-term depreciation of the currency due
to internal inflation. If unit labour cost is consequently compared in US dollar terms, a totally different
picture emerges from that outlined above – unit labour costs in South Africa increase by about the
same rate as in other countries. However, this approach has its own shortcomings since the
depreciation may have been caused by high increases in nominal unit labour costs, and this increase
in costs is not taken into account when all comparisons are made in US dollars. In addition, the
depreciation of the national currency will in itself have a negative effect on the economy, because
imports become more expensive and the consequent higher cost of fuel, machinery and other
necessities further increases inflation, which again has a negative impact on competitiveness.
The Labour Market Commission ( : ) illustrates the difficulty of international comparisons by
referring to two studies. One study shows that by international standards South African wages
(especially unskilled wages) are high relative to productivity, and the other study draws more
favourable conclusions with respect to the country’s unskilled and semi-skilled wage rates.
SUMMARY
Labour costs are becoming increasingly important in determining international comparative advantage.
Unit labour cost at current prices in manufacturing in South Africa increased much more rapidly than those of its most
important trading partners, which damaged the country’s international competitiveness. However, international
comparisons should be approached with caution owing to the complex interaction of many factors.
A. THE IMPACT OF INTERNATIONAL TRADE ON THE LABOUR MARKET
A. . Introductory remarks
Trade liberalisation can affect the economy and the labour market in various ways, including the
following:
Trade liberalisation can increase national welfare if countries succeed in exploiting their comparative
advantage. Increased welfare and economic growth can lead to an increase in employment.
However, countries may also fail to compete successfully, leading to higher unemployment and
poverty.
Trade liberalisation may affect labour standards. On the one hand, labour standards may improve if
higher production leads to a higher demand for labour and higher productivity leads to higher
wages. On the other hand, labour standards will be negatively affected if trade liberalisation leads
to pressure on enterprises to reduce labour costs in order to remain competitive.
Trade liberalisation has a distributional effect because it increases the demand for some production
factors (those used more intensively in domestic production) and reduces the demand for others.
The demand for skilled labour may, for instance, increase, while the demand for unskilled labour
may decrease. This can increase inequalities. On the other hand, inequalities may be reduced by the
positive effects of higher economic growth on job creation and the living standards of lowerskilled
employees.
Trade helps spread knowledge and this contributes to productivity, partly through access to
imported inputs. It also increases innovation. Productivity in enterprises producing for the local
market increases because of import competition, and similarly in those producing for foreign
markets because of the preference for high-quality production often found among foreign (more
discerning) consumers.
Because an open economy increases competition among countries, it can also have a positive
impact by reducing monopoly rents and the influence of bureaucratic connections and political
power. Furthermore, it provides powerful feedback on the effectiveness of productivity and growth
policies, and exposes inefficient policies.
This section is a summary of the most important research findings regarding the various
methodological problems that must be considered when trying to determine the impact of trade
liberalisation on the labour market.
A. . Methodological problems and issues

Measurement of openness. The openness of an economy is the degree to which foreigners and
nationals can transact without governmentimposed costs (including delays and uncertainty) that are
not levied on a transaction between two domestic citizens (Berg & Krueger, ). Measures such as
trade tariffs, additional health requirements for foreign products and so forth raise the cost of buying
abroad. Some of these are very difficult to assess accurately, with the result that it is difficult to
determine changes in openness over time.
Dynamic or static analysis. There is a difference between opening and openness – a country may be
quite closed, but by opening up trade it may increase economic growth for a time. On the other hand,
a country may have been open for a very long time, so it will be very difficult to detect the impact of
“openness” on economic growth. What may be possible to measure is whether the openness has had
an impact on income levels, equality and so on compared to relatively closed countries.
Cross-country or time analysis. A further problem is whether different countries should be compared
with each other, or whether a specific country should be compared over time as it becomes more or
less open. Studies have been done using both methods.
Correlation or causation. Openness and economic growth may show a correlation, but this correlation
may be caused by a third factor that has an impact on both openness and economic growth. In such a
case it would be difficult to separate the effect of openness and of this other factor on economic
growth. One such possible factor, as is pointed out below (section A. . ), is institutional quality, for
instance the effectiveness of government. Good institutional quality may have a positive impact on
both openness and economic performance. There may also be other factors that correlate with both
economic growth and openness, for instance education and skills development. Heavy investments in
human capital may promote both growth and greater openness, but not because greater openness
caused more growth.
Feedback loop. An increase in income or economic growth will increase disposable income, and this
may result in an increase in trade. This will create the impression that there is a correlation between
growth and openness, whereas the direction of causality will actually have been in the opposite
direction. Several studies incorporate controls for this “feedback loop”.
Population size. As is pointed out below, if different countries are compared without taking into
account the size of their populations, then the results are quite different from when account is taken
of population size. To determine whether human beings are benefiting from trade liberalisation, it is
more appropriate to determine if the world population as a whole is benefiting, rather than individual
countries.
Measuring consumption. A final measurement problem raised by Ravallion in The Economist ( March
) is that different methods of measuring consumption show consistently different results. To
measure the impact of globalisation on living standards, it is necessary to determine what happens to
consumption patterns. However, national accounts data tend nearly always to give a much more
optimistic view of trends in poverty than do household survey data. According to The Economist,
much of the “frequently acrimonious debate” about the impact of globalisation revolves around this
technical issue and also explains why some studies show a rapid decline in poverty across the world
over the last few decades and others show little, if any, decline.
A. . Impact on economic growth and income per capita
Cross-country comparisons
Berg and Krueger ( ) refer to various research studies that have found that the huge differences
between countries in the level of income per capita are systematically and importantly related to
openness.
However, the research has also found that institutional quality is another important determinant of
the differences in income per capita between countries. Institutional quality is defined broadly as
factors such as the rule of law, the effectiveness of government and so forth. If this institutional
quality is better, it is likely that per capita income will also be higher. It may be that institutional
quality and openness are also related. This may happen if, as part of improving institutional quality,
the country has implemented a package of beneficial reforms, such as opening up the economy to
achieve objectives such as higher efficiency, technological development, etc. It would then be difficult
to separate the impact of openness from the impact of institutional quality on income per capita.
Changes over time within particular countries

This type of research examines the relationship between changes in openness and changes in income
per capita over time within countries. The advantage of this approach is that the impact of
institutional quality on the research results is reduced – institutional quality within a country normally
changes only very gradually, which will more effectively isolate the impact of openness.
The general conclusion from the research studies is that changes in trade volumes are important
determinants of changes in economic growth. Dollar and Kraay (see Berg & Krueger, ) estimate
that, if the share of trade in GDP increases from to % over years, it will increase real GDP per
capita by %.
China and India are probably the most wellknown examples of countries that have benefited hugely
from opening up their markets. At the beginning of the s these countries were among the
poorest in the world. Once they opened up their markets – both externally and internally – their per
capita incomes increased sharply. Between and , India’s real GDP per capita more than
doubled, whereas China experienced a rise in real income per head of well over % during this
period (Wolf, as quoted by the CDE, b: ). They are no longer among the world’s poorest
countries, and have reduced poverty dramatically.
Dollar and Kraay ( ) considered the experience of developing countries (with a combined
population of three billion people) that have doubled their ratio of trade to GDP since . This
group of post- globalisers excludes the East Asian tigers. The combined per capita GDP growth
rate of these countries increased from below % in the s and s to . % in the s and to
% in the s. This was not only due to the growth in China and India – a total of of the
countries experienced strong economic growth. One can compare these rates with the growth
experience of the non-globalisers, where the annual economic growth rates dropped from . % in the
s to only . % in the s. Dollar and Kraay also undertook timeline studies of the same
countries to reduce the impact of country-specific factors, and still found a statistically significant and
economically meaningful effect of trade on growth: an increase in trade as a share of GDP of
percentage points increased economic growth by between . and one percentage point a year.
However, the rest of the developing world, representing some two million people, actually trades less
today than years ago. This has resulted in them becoming marginalised, showing a decline in
average GDP per capita.
In the early s, the ILO gathered a group of individuals from diverse backgrounds and
affiliations to set up an independent World Commission on the Social Dimension of Globalisation. This
commission recognised in a positive way that the global market had generated many significant
benefits and had great productive potential in terms of economic, political and social development.
However, it also found (World Commission, : ) that the income gap between the richest and
poorest countries had increased significantly. One reason was that the richest countries were well
placed to gain substantial benefits from the increasing globalisation of the world economy. Only a
minority of developing countries were able to be highly successful in terms of increasing their exports
and attracting large inflows of foreign direct investment (FDI). The foremost among these were the
newly industrialised countries (NICs) of East Asia, some other middle-income countries in Asia, the
European Union accession countries, and some Latin American countries. Most of the sub-Saharan
African region was excluded from the benefits of globalisation.
A. . Impact on inequalities
There is a substantial body of evidence that shows that inequalities between countries have become
more skewed over the last five decades. Furthermore, there is also evidence that inequalities within
countries have similarly increased. These two effects (inequalities across countries and inequalities
within countries) have resulted in greater global inequality throughout much of the post-Second
World War period (although there are signs that this may have stabilised since about ) (see
Masson, : ).
Inequalities between countries

To determine whether the greater inequalities in the world can, in fact, be ascribed to globalisation,
one should compare like with like as a result of trade liberalisation, i.e. what has happened between
the globalising countries as a group. Masson ( : ) has found that if only globalising countries are
considered, globalisation has probably led to a greater convergence in per capita income of rich and
poor countries, and not greater inequalities. The per capita income of countries that opened their
economies grew at a much faster rate than those that did not open up to international trade – in fact,
the per capita income in countries that did not globalise barely increased at all. The increasing
inequality in the world is thus not directly linked to globalisation, although it is important to
emphasise that it may be indirectly linked if some countries’ economies are disadvantaged because
they have been unable to open up to international trade.
Subject to this proviso, the general conclusion seems to be that globalisation has reduced inequalities
between globalising countries. There seems to have been a reduction in the extent of inequalities
between rich and poor globalising countries. One reason for this phenomenon is that the globalising
developing countries have been growing more rapidly than the developed countries (for instance, an
average of % per annum in the globalising developing countries compared to . % in the rich
countries). Another reason is that the per capita incomes in some of the developing globalisers were
initially very low, meaning that the increase from the low base has contributed substantially to the
reduction in inequalities between countries (Dollar & Kraay, ). This relates especially to India and
China.
The developments in these two huge countries, which together account for a third of the world’s
population, have certainly skewed the results. The massive increase in economic growth in China and
India means that even if inequality within each country in the world is increasing, and even if
inequality between the richest and poorest countries in the world is increasing, global inequality, as
measured across all the world’s individuals, is not necessarily increasing. Because these two countries,
both of which have a high proportion of poor people and which also form a very large part of the
world population, are growing so rapidly, the sharp increase in the average incomes in these two
countries will result in inequality as measured across all individuals in the world actually falling.
The success of India and China can probably be ascribed directly to their commitment to move
towards a market economy, one in which private property rights, free enterprise and competition
have increasingly taken the place of state ownership, planning and protection. They chose, however
haltingly, the path of economic liberalisation and international integration (Wolf, quoted by Business
Day, September ).
Even though globalisation can and has greatly benefited numerous developing countries, the fact
remains that there are still many poverty-stricken countries. However, many of these seem to have
failed because, as the World Commission on the Social Aspects of Globalisation puts it ( : ), they
are “dysfunctional states torn apart by civil strife, authoritarian governments of various hues, and
states with democratic governments but severe inadequacies in terms of policies and institutions
required to support a well-functioning market economy”. Many African countries, furthermore, do not
have the prerequisites to be able to trade successfully, such as a well-developed transport
infrastructure, functioning ports, functioning financial and credit markets, and skilled workers (see
Bhorat, Mail & Guardian, – July ).
At the same time, these countries also face further obstacles – in particular their current unfavourable
trade position vis-à-vis developed countries. Developed countries protect their labour-intensive
industries, especially their agriculture, against competition from developing countries. In particular,
subsidies to European farmers undercut imports from developing countries. This means that
developing countries cannot fully exploit their competitive advantage. Developing countries have
since the beginning of the st century begun to stand together to address this injustice (Wolf, as
quoted by the CDE b: ).
Inequalities within countries

When it comes to research about inequality within countries, the results are inconclusive. Some
studies have found that inequality has increased as a result of trade liberalisation, and others that it
has either reduced or not changed substantially (see sources quoted by Kohler & Bruce-Brand, ).
In some countries, such as the Philippines and Malaysia, inequality has in fact declined. In Latin
America, trade liberalisation has further widened wage inequalities, although this is probably due to
earlier extreme inequalities in educational attainment (World Bank, as quoted by the Institute for
Futures Research, a). Also, in China inequalities have increased, but this may also have been due
to other factors such as a general liberalisation of the economy, neglect of the rural areas and so forth
(Dollar & Kraay, ).
Dollar and Kraay ( ) researched a total of developing and developed countries. They looked at
changes in trade and in inequality measured over periods of at least five years in order to capture the
medium to long-run relationship between trade and inequality. Their conclusion is that there is little
evidence of a systematic tendency for inequality within countries to either increase or decrease with
increased trade.
However, the ILO ( a) found that the impact of globalisation was not always positive for all
workers. It states that the positive impact is biased towards the high-skilled occupations and has less
impact on the unskilled.
Whether this is purely the result of globalisation or only due to technological change is open to
debate. There are many potential reasons for those already in low-skilled jobs not benefiting much.
The ILO ( a) speculates that even though low-skilled wages may not have increased, the initial
impact of globalisation and growth may have resulted in more jobs being created; in other words,
rather than wages increasing, previously unemployed or underemployed people will have obtained
jobs in the formal labour market. According to this hypothesis, formal employment will thus have
increased, as has been the case in South Africa from the start of the s. If this were the case,
individual workers might not be better off, but the population as a whole would be better off because
employment has increased and unemployment declined.
The impact of trade liberalisation on equality objectives also depends on the manner in which the
affected country deals with the adjustment process. In the US, for instance, labour standards were
flexible. The adjustment process consequently took the form of a significant downward movement in
unskilled wages. This resulted in rising inequality but also a sharp increase in employment. In Europe,
on the other hand, unskilled wages were relatively better protected. The adjustment process
consequently took the form of a sharp contraction in employment but no increase in inequality (see
section A. . ).
The ILO ( a) has found that globalisation has reduced the gender pay gap in lower-skilled
occupations, especially in those occupations where women are widely represented. However, because
skilled male workers probably benefit more from globalisation than female workers, this again
increases the overall gender pay gap.
In the study by Meschi and Vivarelli, the impact of trade on within-country income inequality of
developing countries is examined, and the authors have found that aggregate trade flows are
weakly correlated with inequality. Nonetheless, once the authors had disaggregated total trade flows
according to their areas of origin and destination, they found that trade with high-income countries
worsens income distribution in developing countries, through both imports and exports. This finding
suggests that technological differentials between trading partners plays a role in shaping the
distributive effects of trade openness.
A. . Employment and wages

After analysing a number of studies, the ILO ( a) concluded that “globalisation initially has a
negative impact on wages, but this effect dissipates over time and eventually the impact of
globalisation on wages becomes positive”. This positive impact generally occurs within three to four
years of openness. The initial negative impact can be attributed to import penetration and a move of
producers to more profitable sectors, with resultant job losses in some sectors.
The World Commission on the Social Aspects of Globalisation ( : ) refers to studies that show
sharply contrasting results of the impact of trade on employment and wages, and concludes that the
relationship between trade liberalisation and employment is likely to be a contingent one, dependent
on a host of country and external characteristics. With regard to the impact of FDI, the conclusion is
that, on the whole, FDI does increase economic growth. However, empirical evidence on the
employment impact of FDI is sparse and does not permit simple generalisation.
The International Monetary Fund has found ( ) that, contrary to fears that globalisation is driving
down wages, total labour compensation has grown by a cumulative % on average since . This
is in part due to globalisation as export opportunities have risen, while productivity and output have
benefited from lower input costs and better production efficiencies arising from globalisation.
Manufacturing wages in emerging economies, especially in Asia, have also increased and often
converged to the United States levels.
However, the share of income accruing to labour (as opposed to capital) has fallen due to the
negative impact of globalisation on the demand for unskilled labour in many countries (International
Monetary Fund, ). The point of impact has varied between countries, with some countries
experiencing a fall in the employment of unskilled workers, and others an increase in inequality
between skilled and unskilled labour.
Countries adopting reforms to lower the cost of labour to business (by lowering the tax wedge – the
difference between the payroll cost to a firm and the net take-home pay of workers) and improve
labour market flexibility have generally had a smaller decline in labour share. Other policy priorities
are to strengthen access to education and training, and to adopt adequate social safety nets.
A. . Reduction in poverty
Poverty reduction and reducing inequalities are not the same and, even if inequalities have increased,
it does not necessarily mean that the poorest people have become even poorer as a result of
globalisation.
The World Commission on the Social Aspects of Globalisation ( : ) has found that the number
of people living in absolute poverty worldwide declined significantly between and .
However, it found that most of this decline can be accounted for by the sharp reduction in poverty in
two very large globalising economies, India and China.
Having considered the vast microeconomic literature in this regard, Berg and Krueger ( ) found
that there is no systematic relationship between openness and the income of the poorest, beyond the
positive effect of openness on overall growth. The evidence shows that on average the income of the
poorest tends to grow one-for-one with average income, although in some cases the poverty stricken
may do better, and in other cases worse.
The number of poverty-stricken people may thus decline at the same time as inequalities increase;
this will happen if there is a general increase in incomes but the incomes of the rich increase more
than those of the poor.
This seems to have been the case in China and India. In both these countries there has been a massive
increase in incomes as a result of globalisation. This has significantly reduced poverty, even though
inequalities have increased. Because the populations of these countries form such a large part of the
world population, it affects average trends in the world.
Dollar and Kraay ( ) have found that the combination of increases in growth and little systematic
change in inequality in the globalisers has considerably boosted efforts to reduce poverty. In
Malaysia, for example, the average income of the poorest fifth of the population grew at a robust
. % annually. Even in China, where inequality did increase sharply and the income growth rate of the
poorest fifth lagged behind average income growth, incomes of the poorest fifth still grew at . %
annually.
Neutel and Heshmeti ( ) constructed a globalisation index (which focuses on four key areas,
namely economic integration, personal contacts, technological connections and political engagement)
before comparing its relationship with poverty in developing countries, and found that
globalisation leads to significant poverty reduction. In contrast, the study of Harrison ( ) suggests
that the poor are more likely to share in the gains from globalisation when there are complementary
policies in place, including: policies to promote credit and technical assistance to farmers; policies to
promote macroeconomic stability; as well as investments in human capital and infrastructure. Harrison
has also found that trade and foreign investment reforms benefit the poor in the exporting sectors
and sectors receiving foreign investment.
Protectionism is the sacrifice of the consumer to the producer – of the end to the means.
F B
A. . Conclusion
The World Bank’s conclusion is that workers seem to gain from global integration in the long term (as
quoted by the Institute for Futures Research, a). Wages have grown twice as fast in more
globalised developing countries than in less globalised countries. There is also evidence that an open
economy, especially in so far as foreign investment is concerned, increases the return on education,
and thereby encourages more investment in education. However, in the short term, formal-sector
workers (especially older ones) in protected industries may suffer job losses and more job insecurity.
Government social protection is therefore very important for the renewed welfare of affected workers.
A good education system that provides opportunities for all is critical for success in the globalising
world.
Dollar and Kraay ( ) say the following:
The experiences of the post- globalizers show that the process [of globalization] can have
great benefits, contributing to rising incomes and falling poverty and enabling some of the
poorest countries in the world to catch up with richer countries. The real losers from globalization
are those developing countries that have not been able to seize the opportunities to participate
in this process.
The latter remark applies in particular to sub-Saharan Africa, which for the most part has not seen any
benefits of globalisation.
SUMMARY
Developing countries that have been globalising (the majority of the world population) have benefited significantly from
trade liberalisation in terms of per capita income. Non-globalising countries are trading less today than before, and
have experienced low growth in per capita GDP.
The impact of trade liberalisation on inequalities within countries is indeterminate – there is no systematic increase or
decrease in inequalities. In some cases it seems to have had a negative effect on the demand for unskilled labour.
Trade liberalisation has reduced inequalities between the rich and poor globalising countries, with sharp increases in
per capita incomes in China and India contributing substantially to this reduction.
Trade liberalisation has, because of its positive impact on per capita income, also contributed to an overall reduction in
poverty.
A. THE IMPACT OF TRADE LIBERALISATION ON THE SOUTH AFRICAN LABOUR

MARKET
Several decades of relative economic isolation meant that enterprises in South Africa were ill-
prepared to take advantage of the opportunities arising from trade liberalisation, while the potential
adjustment costs were correspondingly high.
Trade liberalisation has certainly resulted in a sharp increase in exports, especially in the
manufacturing industry. Manufactured exports increased by an average annual nominal rate of more
than % between and , and this already high growth rate was much higher in the period
thereafter, i.e. after trade sanctions were lifted (Tsikata, : iv). However, as far as employment in
particular is concerned, the available research shows that trade liberalisation did not have the positive
effect that might have been expected on the basis of the good export performance (Nattrass, a:
).
Rangasamy and Blignaut ( ) point to the potential positive impact of trade. Even though the study
was based on data up to , it found that those sectors that became more externally oriented had
lower inflation rates and higher growth rates than the other sectors in the economy for the period
– . In particular, the average growth rate for the externally oriented sectors was . %
compared to . % for the other sectors. The difference in the inflation rate was less marked, i.e. . %
for the external sectors and . % for the others.
Jenkins ( ) found that globalisation has a negative impact on employment in South Africa, in
particular the unskilled workers, but has not had any negative effects on wages. Jenkins also found
that globalisation has a negative impact on the quality of employment, yet she argues that it is
difficult to isolate the extent to which globalisation has been a cause of the growth of atypical forms
of employment, as other factors such as the promotion of a more flexible labour market also play a
role.
Bhorat et al. ( : ) point out that the impact of trade on the demand for labour has been
relatively small. Even though changes in relative demand for labour due to trade flows in the period
– had a positive impact on all skill levels, the positive impact was smaller for African and
coloured workers and for male workers. Since , these categories of employees were again the
losers (being over-represented in the production categories of employment), with highly skilled and
certain semi-skilled employees benefiting most from international trade. Overall, the impact of
international trade on employment has been relatively benign, with only (semiskilled) production
workers experiencing actual job losses between and . The impact on unskilled elementary
workers was positive but almost negligible in terms of magnitude.
As pointed out by the authors, there were other factors that had a much greater impact on the labour
market than international trade, such as certain within-sector developments (such as the increasing
preference of employers for higher-skilled employees). In addition, the between-sector developments
also played a significant role, in particular the relatively higher growth of the tertiary sectors, which
resulted in a higher demand for skilled labour, whereas the primary sectors, which employ most of the
production workers, experienced a relative decline.
Globalisation therefore seems to have shifted production in favour of capital- and skillsintensive
sectors to the detriment of labourintensive sectors (Nattrass, a: ; Jenkins, : ). Those
subsectors that are classified as most capital intensive account for more than % of all
manufacturing exports. Within manufacturing imports, by contrast, less than one quarter falls into this
category (Hayter, Reinecke & Torres, : ). The largest export expansion therefore occurred in the
relatively capital-intensive sectors, whereas the unskilled labour-intensive sectors have performed
poorly in terms of exports.
This reflects the fact that South Africa specialises in capital-intensive products, which is a legacy of
past industrial policies (see section . . ). Another reason is that South Africa’s competitiveness may
be hampered by high wages, especially those of low-skilled workers, which exceed corresponding
productivity levels (Tsikata, : vi). This puts the country at a competitive disadvantage in low-wage,
unskilled labourintensive activities. The destruction of lowproductivity, low-wage jobs is related to the
fact that South African manufacturing wages are high in relation to productivity, as Nattrass ( b: )
found in a study of countries. All developing countries in the Nattrass sample had a lower wage-to-
productivity ratio than South Africa. South Africa is thus rapidly moving away from a labour-intensive
growth path as envisaged in GEAR, and high wages certainly play a role in this regard.
There seem to be a number of reasons why the employment of less skilled workers might have been
negatively affected by trade liberalisation, and these can be summarised as follows:
Exporting firms have been found to be relatively skills intensive. A large proportion of South Africa’s
exports, for instance, originate from natural resource-based production, which has become
relatively capital and skills intensive.
There is increased competition from lowwage, labour-intensive export production in other
developing countries. Firms that experience high tariffs are less skills intensive, and they will thus
lose more market share from trade liberalisation. This, combined with high unskilled wages relative
to productivity in South Africa, has had a negative impact on the employment of unskilled workers
in this country.
Firms that import a large percentage of their raw materials are more skills intensive. As they will
benefit more from trade liberalisation, trade liberalisation increases the skills structure of the
economy.
Trade liberalisation often induces “defensive innovation”, which is more skills intensive. The drive to
increase productivity has resulted in less demand for unskilled workers and a greater demand for
skilled workers.
South Africa, as a middle-income country, has in effect been squeezed from both ends – on the one
hand it cannot really compete with low-wage countries such as China or India because its wage levels
are too high; on the other, it can also not successfully compete with the high-wage, high-productivity
countries because of skills shortages and lack of technological know-how and the latest equipment.
(There are some significant exceptions to this generalisation.)
So even though there have been sectoral shifts away from unskilled labour, this is not necessarily due
to trade liberalisation, but may be due to other factors such as changes in domestic demand,
technology or relative wages. Edwards ( : ) also refers to the adverse effect of labour regulation
making adjustment more difficult, with the result that adverse demand or other shocks show up in the
form of unemployment rather than in other forms of labour market adjustment.
Impact on wages and inequalities

Edwards and Behar ( ) have found that trade liberalisation has had the biggest negative impact on
the wages of semi-skilled workers and artisanal workers, with the wages of unskilled workers being
affected least. However, because of wage rigidities, these wage effects have most likely been
translated into employment effects. What this means is that in a freely functioning labour market,
wages of lower-skilled workers have declined owing to trade liberalisation. However, because
bargaining councils and collective wage agreements ensure that wages cannot really decline, trade
liberalisation then impacts negatively on employment levels.
Although there is relatively little empirical evidence, trade liberalisation in South Africa might thus
have increased inequalities, at least over the short term. This is mostly because the employment of
lower-skilled workers might have been negatively affected, as outlined above, and an increase in such
unemployment will have increased inequalities. However, once the positive effects of trade
liberalisation on economic growth starts filtering through to the economy (as might have started to
happen from the beginning of the s), employment and also equality objectives may be more
favourably affected.
SUMMARY
Trade liberalisation has resulted in a sharp increase in South African exports, especially in the manufacturing industry.
However, it did not initially have the positive effect on employment that might have been expected, even though this
might have changed after .
Trade liberalisation has shifted production in favour of capital- and skills-intensive sectors to the detriment of labour-
intensive ones, and has thereby had a negative impact on the employment of unskilled labour.
The employment losses in export-orientated sectors have been relatively greater than in import-competing ones, at
least in the initial years after trade liberalisation.
A. ADDRESSING THE CHALLENGES OF GLOBALISATION
South Africa has moved from being a relatively closed economy to a relatively open one. Some other
countries have also made this transition, but apparently with more success than South Africa in terms
of addressing the unemployment consequences of such a transition. This is discussed below. These
countries have adjusted their labour market policies in various ways. However, since South Africa
seems to have gone in the opposite direction by firstly increasing non-wage labour costs; secondly by
introducing various measures that increase wage costs; and thirdly by introducing measures that
often reduce rather than increase flexibility. These are discussed in sections . and . .
The opening up of South Africa’s borders to international competition has had a significant impact on
several industries that are covered by bargaining councils. Many of these industries suddenly faced
competitive forces that did not exist before and the pressure on employers to reduce costs and
increase productivity increased greatly. This is an example of tariff reform outpacing labour market
reform, with negative consequences for employment creation and unemployment.
A. . General principles
Before considering some of the success stories of countries that have managed globalisation, it may
be useful to recall what the World Bank has said about failures. It identified the following reasons for
some countries, especially in Africa and the former Soviet Union, not being able to improve living
conditions and alleviate poverty as a result of globalisation:
These countries rely heavily on the export of a narrow range of primary commodities, which makes
them very vulnerable to declining export prices relative to import prices.
Some countries have poor policies and infrastructure, weak institutions and corrupt governments.
Good government is particularly important to ensure correct macroeconomic and microeconomic
policies, a predictable environment, and the absence of crony capitalism (i.e. where friends and
family are bestowed benefits such as state contracts, licences and so forth).
Some countries suffer from the intrinsic disadvantages of adverse geography and climate.
Some countries may have permanently missed the opportunity to industrialise as a result of a
temporary phase of poor policies.
The World Commission on the Social Dimension of Globalisation ( ) strongly emphasises that the
manner in which countries manage their internal affairs will influence the extent to which people will
benefit from globalisation and be protected from its negative effects. “The response to globalisation
can be said to begin at home” ( : xi). In particular, the report states that there is wide international
agreement on the essentials which all countries must urgently strive towards:
Good political governance based on a democratic political system, respect for human rights, the
rule of law and social equity
An effective state that ensures high and stable economic growth and raises the capabilities of
people through education
A vibrant civil society, with strong representative organisations of workers, employers and other
interest groups
The report calls for a fair and inclusive process of globalisation. There should be more institutions to
ensure the smooth and equitable functioning of global markets and address the asymmetric effect on
rich and poor countries. Among others, core labour standards should provide a minimum set of
global rules in the global economy. However, demanding that poor countries adopt the standards of
the rich countries may also be counterproductive and may increase poverty in these countries.
A. . Some case studies of how countries have succeeded in a globalised world

Nattrass ( c) points out that countries undergoing trade liberalisation must be particularly
concerned about their labour market policies. Where high tariff barriers provided the space for both
profitable investment and labour market protection, trade liberalisation reduces that space
dramatically. Reforming countries thus face a dilemma if they are to avoid the possible negative
employment consequences of globalisation: either the labour market must be made more flexible by
reducing non-wage labour costs and/or allowing wages to fall, or productivity must rise dramatically if
enterprises are to remain competitive. The latter, however, takes time and may even involve job losses
over the short and medium term. Greater openness makes the economy increasingly subject to global
competitive forces. Competition on the basis of labour costs seems to have become increasingly
prevalent in international trade, which means that appropriate labour policies play a critical role in a
country’s ability to adjust to and face international competition.
There is nothing Japan really wants to buy from foreign countries except, possibly, neckties with
unusual designs.
Y I
More jobs or better jobs: the US versus Europe

If the experience of other countries is anything to go by, countries face a choice in addressing the
challenges of globalisation – either to allow greater wage flexibility, which will increase inequalities, or
to protect the wages of the unskilled, which will increase unskilled unemployment. At the risk of
oversimplifying through generalisation, the US and Europe are examples of countries that opted for
opposing policy choices.
In the five years to some million jobs were created in the US, while there was almost no net
increase in jobs in continental Europe over the same period. In , unemployment fell to a -year
low in the US, while it reached a postwar high in Germany, for example.
Most observers agree that the fundamental reason for this has been more flexible labour markets in
the US, and wage flexibility in particular, as opposed to highly regulated labour markets in Europe.
The European Commission ( : ) maintains that social security systems and taxation policies in
European Union member states have not yet been fully adapted to flexible forms of work. High direct
and indirect labour costs have meant that European employers have been unable or unwilling to
expand employment, whereas wages in the US have been sufficiently flexible to “clear” the labour
market, i.e. to keep unemployment to a minimum.
The European Commission ( : – ) has highlighted the increasing burden of tax on labour in
member states, and the negative effect this has had on employment. Social security contributions, for
instance, now represent more than % of total labour costs. This has substantially increased the
indirect cost of labour, while the opposite has happened to the indirect cost of other factors of
production, such as capital, energy and natural resources. Some member states have reduced the
starting wages and social security contributions of young workers to encourage their recruitment.
However, the lower wages in the US occurred at the lower end of the skills spectrum, with the result
that wage differentials, and consequently inequality, increased significantly (Wood, : – ).
Furthermore, the real wages of unskilled workers often actually declined, resulting in a growing
phenomenon of the working poor (Buckberg & Thomas, ). In Europe, there was practically no
increase in inequality.
The growing inequality in the US and increasing European unemployment have generated much
controversy. For instance, it has been argued (Wolf, ) that Europe has been able to generate
remunerative employment for the male breadwinner and that most unemployment is found among
what Wolf calls “outsiders”, i.e. women, the young and older workers. Furthermore, wage flexibility
alone is not sufficient to create more jobs, although it seems to be a necessary condition for this. It is
also said that even though the newly created jobs in the US may not be quality jobs, the wages paid
in these jobs are mostly higher than the average wage and allow people to acquire skills and
experience.
Some observers state that the reason for the US’s performance does not lie with wage flexibility, but
with less rigid regulations governing output-related matters such as working hours, planning and
zoning, combined with the more effective channelling of capital to start-up enterprises (Baker, ).
Overall, however, these findings suggest that if countries respond to competition from lowwage
countries by allowing greater wage flexibility, then wage differentials will increase (as in the US),
whereas if they respond by protecting the wages of unskilled labour, then unemployment will rise (as
in Europe). However, one country in Europe seems to have succeeded in creating jobs while retaining
the basic principles of social democracy and the protection of working standards, and that is the
Netherlands. Another apparently successful example is Ireland.
More and better jobs: the Netherlands

The experience in the Netherlands has been different from that of both the rest of continental Europe
and the US. Employment in the Netherlands increased quite rapidly from the latter part of the s
and into the s. Unemployment was halved to . % in , and fell to even less thereafter. It
occurred with little conflict and without sacrificing the national goal of redistributing money from the
rich to the poor (Mayer & Grillet-Aubert, ). The country also succeeded in increasing the
employment of young people.
With the support of social partners, the Netherlands introduced the following measures to improve its
economic and labour market performance (Mayer & Grillett-Aubert, ; European Commission,
: ; Auer, ; Ozaki, ):
Wage restraint was introduced through national centralised bargaining. Auer ( : , ) refers to
wage moderation as a “crucial part of macroeconomic policy” and “the core target of the social
pact”.
Minimum wages were initially cut and thereafter frozen.
Public spending was cut sharply.
The social security contributions of employers were reduced from to . %.
Part-time work was made easier (by permitting part-timers to be paid less than full-timers for the
same job). This resulted in the proportion of part-time workers increasing sharply, by over %
between and , which meant that two-thirds of the country’s employment growth was in
parttime jobs.
Unemployment benefits were reduced and it was made harder to qualify for such benefits. This
reduced the reliance of the unemployed on unemployment benefits during a spell without work
and served as an incentive for them to find a job.
Unemployment benefits were actively used to promote work experience.
There was a radical reform of sickness benefits to ensure that they were more cost-effective and
prevention of illness was emphasised.
The number of young people combining work and training was sharply increased through the
availability of part-time work and a reduction in study grants.
However, the Dutch economy, which is highly dependent on an international financial sector and
international trade, did not escape the ravages of the financial crisis. The Dutch financial sector
in particular suffered, due in part to the high exposure of some Dutch banks to US mortgage-backed
securities, and the government had to pump massive amounts of money into the financial system to
prevent a collapse. Unemployment increased, but after three years of recession, the unemployment
rate still stands at only . %.
From rural backwater to Celtic tiger: Ireland

Ireland is one of the economic success stories of the s. Economic growth and employment
increases have far exceeded those achieved elsewhere. For instance, the Irish economy expanded by
almost % in the ten years from , compared to an expansion of only % in the US (Barry, :
). The number of jobs increased by % over the same period, outstripping even the US, the “job
creation dynamo”, which created % new jobs in this time. The result was that the unemployment
rate dropped from % in to % in .
It all started in the mid- s when the government, labour and business realised that they had to
develop new ways of working or face a continuing spiral of unemployment, debt and economic
decline. This led to a number of social pacts that were negotiated in a national collective bargaining
forum. They addressed a wide range of economic and social policy issues covering tax reform, wages
and a commitment to health, education, job creation and social security issues.
The ILO (Auer, : ) selects only three factors to explain the relative success in the labour markets
in both Ireland and the Netherlands. These are macroeconomic policy, social dialogue and labour
market policy (including labour market reforms).
With regard to the first, macroeconomic policy, Auer refers to price, interest and exchange rate
stability, as well as to providing fiscal stimuli to the economy against a general background of fiscal
consolidation.
The government agreed to create an enabling environment for job creation by, among other things,
bringing government expenditure under control. This, together with financial support from the
European Union, allowed for tax reductions, which were also traded in wage negotiations for wage
moderation on the part of trade unions. One-third of the increase in Irish real after-tax wages since
has therefore come in the form of tax reductions, especially employment taxes (Barry, : ). In
addition, the Irish government cut corporate taxes to a mere . % compared with % or more in
European Union and OECD countries.
Another important factor that contributed to Ireland’s success story is that the economy was
substantially opened up and the protectionist barriers dismantled. Dependence on primary product
exports (agricultural products) was reduced, partly by opening up the economy and partly by policy
initiatives. Ireland also channelled substantial resources into education, by among other things
providing free secondary education to everyone. Tille and Yi (as quoted by Strydom, : )
maintain that Ireland’s economic boom was directly related to investment in people. About % of
the population have an upper secondary level of education. The increase in the supply of skilled
people through reforms in education was probably one of the most important reasons for the sharp
increase in FDI in Ireland.
The second factor, social dialogue, has facilitated wage moderation, and has also assisted with reforms
of social security, the labour market and labour market policy. Auer refers to the fact that these
reforms have not always been easy for the partners, especially trade unions, to accept because they
were not always in the short-term interests of their members.
The parties agreed to various trade-offs. Labour agreed to a form of wage restraint with wage
increases equal to or less than inflation. Increases were also to be linked to productivity improvements
with the approach that there would be a “change for pay rather than pay for change”. This simply
meant that labour agreed to changes in work practices, which could in turn lead to productivity and
wage increases, rather than employers granting the unions wage increases to “buy” changes in work
practices. In exchange, employers and the government agreed to reduce weekly working hours from
to . Employers also agreed to become more competitive and create jobs. The social partnership
approach since the mid- s also resulted in industrial peace, which together with other policy
initiatives attracted foreign multinationals into Ireland.
The third factor is the introduction of labour market reforms aimed at making the labour market more
flexible (Tille & Yi, as quoted by Strydom, : ). This included passive, incomereplacement schemes
as well as active labour market policies such as labour market training, adjustment flexibility for
enterprises and improved delivery of employment services. The ILO (Auer, : ) as well as Barry
( : ) emphasise the importance of the centralisation of wage bargaining at national level for the
success of both Ireland and the Netherlands. Three bargaining models seem to produce positive
outcomes, i.e. highly decentralised bargaining, highly centralised bargaining, and moderately
decentralised but highly coordinated bargaining (Calmfors & Driffill, ; sources quoted by Auer,
: ). Negative outcomes seem to be produced by a position in between, which is the model in
South Africa (see section . ).
Labour productivity increases in Ireland in the s were nothing short of astounding – at the same
time the capital-to-labour ratio was declining (in contrast to sharp increases in South Africa), and
labour productivity and the total factor productivity were increasing very rapidly. This has resulted in
Ireland currently being the richest country in the European Union after Luxemburg in terms of per
capita GDP.
However, there has also been some criticism of the “Irish model”. Allen ( ), for instance, argues
that the country’s success was not the result of a social partnership, but a particular set of
circumstances that would be difficult to repeat. For instance, US companies needed a platform inside
the European Union to gain access to that market, and Ireland offered an English-speaking, highly
educated workforce that was relatively cheap. The counter-argument is that although access to the
European Union might have been a unique factor, many other factors such as the highly educated
workforce, labour at low cost, low corporate taxes and the elimination of industrial strife were very
much part of the strategy for building the economy.
It is true that the Irish success story also has a darker side – real wages lagged behind and the
economy has more inequalities than many other European countries. The environment has also
become less union friendly, and there is currently an absence of legislative compulsion to recognise
unions.
The strongest criticism against the success of the Irish model is that the global recession that started
in had a massive negative impact on Ireland. The high economic growth rates up to that stage
had resulted in increasing inflation, and this led to a property price bubble. The Irish economy was
skewed towards the construction industry, and the high property prices were financed by the banks.
The result was that when the property bubble burst, the banks were at the forefront in experiencing
the impact of the recession, resulting in the government having to step in to support the banks.
Ireland was consequently the first EU country to enter a recession. At the time of writing, the Irish
economy had not yet showed firm signs of escaping from the recession that commenced in .
Moreover, unemployment is relatively high at a rate of %.
Freeing up the labour market: Australia

In the s the Australian government introduced a wide-ranging set of policies to help foster the
internationalisation of that economy. This related in particular to liberalising the economy by
substantially reducing protectionist measures, abolishing exchange control, and introducing
significant tax reforms and a more liberal foreign investment policy (Kyloh, : ). However, far-
reaching labour market reforms were also implemented, for instance with regard to wage-setting
mechanisms, trade union structures, the social welfare system, skills development and significant
wage moderation (Kyloh ).
It is important to note that most of these changes were initially implemented with the implicit or
explicit support of the trade unions in terms of a accord between the government and the
unions (the Prices and Incomes Accord). Initially, the system of wage determination was radically
centralised. This allowed the accord to be given force by the Arbitration Commission (later called the
Industrial Relations Commission) by fully indexing wages to inflation. Two years later, however, wage
increases below the rate of inflation were granted and the outcome of the accord was a significant fall
in real wages – by anything between and % (Bramble & Kuhn, : ). In , another
requirement was added, in terms of which wage increases would be granted in exchange for
productivity increases.
In , these changes had the eventual effect of all parties supporting enterprise bargaining rather
than the awards system of the Industrial Relations Commission. This resulted in a significant
liberalisation of the labour market, among others by encouraging even individual contracts of
employment. In , the Minister for Employment and Workplace Relations again referred to the
significant legislative reform of the system, which had the objective to make the primary focus of the
system as agreement making at the workplace.
The advantage of enterprise bargaining and workplace agreements (i.e. contracts between an
employer and a single employee) is that they provide for much more specific linkages between
productivity and pay. The government argues that the outcome of these agreements has been
positive for all stakeholders, with the average weekly pay for workers covered by a workplace
agreement being about % higher than for workers on collective agreements. However, not all
analysts are as positive about the impact of the changes to the Australian labour policies on workers
(see, for instance, Bramble & Kuhn, ).
Further significant labour policy changes were implemented in , this time without the support of
the unions. The changes included making agreements much simpler by eliminating overly
prescriptive provisions; allowing agreements between employers and individual employees to vary;
federal awards containing minimum employment standards; excluding small and medium-sized
businesses (with fewer than employees) from unfair dismissal laws, and requiring employees in
enterprises with more than employees to have been employed for six months before they can
enjoy protection against unfair dismissals.
These policy changes were intended to further enhance the ability of the Australian economy to
adjust rapidly to international competition. Australia has created nearly one million new, permanent,
full-time jobs since , and the unemployment rate in dropped to a -year low of %. Labour
productivity has been increasing at an average of % per annum since the early s. Industrial
unrest has been greatly reduced – from working days lost on average per employees in the
s to the current level of less than .
There were further significant changes in , with the introduction of the National Employment
Standards, the Fair Work Act, a national minimum wage order and the modern awards system (see
http://www.fwa.gov.au). The National Employment Standards introduced minimum standards in
respect of employment conditions such as working hours, leave, notice periods, severance pay and
the like. These apply to all employees and cannot be displaced, even if an enterprise agreement
includes terms that have substantially the same effect. In fact, enterprise agreements are subject to
approval, and provision is no longer made for individual agreements.
SUMMARY
Countries undergoing trade liberalisation must be particularly concerned about their labour market policies, because
the reduction of high tariff barriers will eliminate the space previously available for both profitable investment and high
labour market standards.
Countries seem to face a choice in addressing the challenges of globalisation either to allow greater wage flexibility,
which will increase inequalities (as in the US), or to protect the wages of the unskilled, which will increase unskilled
unemployment (as in Europe).
However, two countries, Ireland and the Netherlands, achieved relative success with their policies by giving attention
to macroeconomic policy, social dialogue and labour market policy, particularly wage moderation and active labour
market policies.
A. . Other measures to address the challenges of globalisation

Many other labour-orientated measures that countries may consider as part of their strategy to
address the challenges of globalisation are discussed in other chapters, particularly Chapter where
an employment strategy is discussed. These strategies are thus not repeated here, although they are
equally relevant.
KEY CONCEPTS
globalisation
institutional quality
trade liberalisation
unit labour cost
world competitiveness
FOR STUDENTS
. How competitive is South Africa internationally, and what factors should be taken into account when
considering South Africa’s competitiveness?
. “South Africa’s unit labour cost is increasing much more rapidly than that of its most important trading
partners as a result of the high wage demands of South African unions.” Critically discuss this statement with
regard to wage and productivity trends, exchange rates, inflation and the impact of unions on wages and
productivity, among others.
. Describe some of the difficulties in determining the impact of trade liberalisation on the labour market.
. Describe what the effect of globalisation is likely to be on the labour market, including per capita income,
inequalities and poverty.
. Describe the possible impact that globalisation has had on the South Africa labour market, taking into account
recent developments.
SUGGESTED READING*
Auer, P. . Employment revival in Europe: labour market successes in Austria, Denmark, Ireland and the
Netherlands. Geneva: International Labour Organization.
Berg, A. & Krueger, A. . Lifting all boats: why openness helps to curb poverty. Finance and Development,
September.
Bhorat, H., Lundall, P. & Rospabé, S. . The South African labour market in a globalizing world: economic and
legislative considerations. Employment Paper / . Geneva: International Labour Organization (ILO).
Centre for Development and Enterprise. b. Why globalisation works. Johannesburg: Centre for Development
and Enterprise.
Hayter, S., Reinecke, G. & Torres, R. . Studies on the social dimensions of globalization: South Africa. Geneva:
International Labour Organization.
Institute for Futures Research. a. The World Bank’s latest take on globalisation. Economic Issues, ( ),
February.
Jenkins, R. . Globalisation and the labour market in South Africa. Journal of International Development, ( ),
March.
Nattrass, N. a. Globalisation and the South African labour market. Trade and Industry Monitor, .
Nattrass, N. c. Labour-demanding growth: lessons from international experience. CDE Focus.
World Economic Forum. . The Global Competitiveness Report – . Geneva: World Economic Forum.

B Social dialogue and codetermination
Outline of this appendix

B. Introduction
B. Corporatism as a form of social dialogue
B. National Economic Development and Labour Council
B. Other national consensus-seeking bodies
B. Codetermination in the workplace
B. International consensus seeking in the labour arena: the International Labour Organization
You cannot lift the wage earner by pulling the wage payer down.
A L
B. INTRODUCTION
There are several examples of countries achieving positive economic and labour market outcomes
through collaboration between the government, employer and union bodies at national level.
Appendix A, for example, refers to the Netherlands and Ireland as two examples of countries that
significantly increased economic growth and reduced unemployment and inflation by means of social
accords at national level. Chapter cites Malaysia as an example of where affirmative action was
successfully implemented because the government of national unity that existed from to
brought about political stability. This enhanced investment confidence, and the incorporation of
opposition viewpoints was important to ensure that the groups that were not beneficiaries of
affirmative action were left with sufficient economic space to make cooperation preferable to
opposition.
At the time of the transition to democracy in South Africa, the RDP White Paper stated that the RDP
would be implemented through the widest possible consultation with, and participation of, the
citizens of South Africa. Structured consultation processes at all levels of government would be
introduced to ensure the participation of stakeholders in policy making and planning, as well as
project implementation (Government of National Unity, : ).
Furthermore, the National Economic Development and Labour Council (NEDLAC) was also formed at
that time. The objective of NEDLAC is to reach consensus and make agreements on matters
pertaining to social and economic policy, including macroeconomic policy, labour policy and
developmental issues.
Apart from NEDLAC, a number of other consensus-seeking and advisory bodies were also formed to
deal with specific labour market issues, such as employment equity, conditions of employment, and
the like. They are all characterised by the representation of employer and union interests in some
form or another, and some of the bodies represent other interests as well.
Has South Africa achieved its stated objective of the widest possible consultation? Is the government
still as committed to social dialogue as it was in ? These are the issues discussed in this appendix.
There are, however, some people who maintain that social dialogue is not enough and that the
country will really only be successful if there is much more meaningful codetermination, especially at
the level of the workplace. Germany is the most well-known example of significant codetermination in
the workplace, and South Africa attempted to introduce some semblance of the German model
through workplace forums. Codetermination will thus also be discussed.
B. CORPORATISM AS A FORM OF SOCIAL DIALOGUE
The involvement of employer and employee stakeholders in negotiation and consultation on national
policy issues beyond parliament is sometimes referred to as corporatism. Corporatism is an
institutional framework that incorporates to varying degrees the pinnacles of employer and employee
organisations in the economic and social decision making of society (Barker & Holtzhausen, ). It
therefore refers to the institutionalised process of negotiation between sectors of the state and
powerful organisations, such as employer organisations and trade unions movements, whose
cooperation is indispensable if public policies are to be credible and capable of implementation.
Two main forms of corporatism can be distinguished. The first is state corporatism, which is
characterised by the institutional incorporation of the leadership of capital and labour organisations
within state apparatuses and state administration to represent such class interests. Recognition by the
state is a necessary condition for association and the continued operation of the organisations. This is
done in return for greater restraint on the part of such organisations and greater governmental
control of the economy. This form of corporatism is often associated with authoritarian regimes.
Societal or bargained corporatism, the second form of corporatism, refers to a situation where labour
and employer organisations retain their autonomy and have greater independent influence on
government policy. The state grants recognition to these organisations as a matter of political
necessity. The latter situation is sometimes also referred to as neo-corporatism. This form of
corporatism is evident in several democratic countries.
However, as already referred to, there are some countries that have in recent years very successfully
involved the pinnacles of labour and business in important economic and labour policy decisions to
improve the economic performance of those countries. Examples are the Netherlands, Ireland and, at
least until a few years ago, Australia. The economic experiences of these countries are discussed in
Appendix A.
Webster ( : ) points out that corporatism has been in retreat in many countries since the s,
and refers specifically to Britain and Sweden as examples. In this regard, South Africa appears to have
defied the trend during the s, when it created various forums for bargained corporatism,
particularly NEDLAC. However, this might have changed more recently, as will be discussed in the
section dealing with NEDLAC (section B. ).
Some people might argue that there might even be a few elements of state corporatism in South
Africa, in the sense that the trade union federation, COSATU, is in alliance with the governing party. In
that role it enjoys many privileges not available to other stakeholders, for instance involvement in
policy planning sessions with the governing party, even in parliamentary committees. In the past,
COSATU also had the right to nominate a certain proportion of the ANC representatives in Parliament.
This may reflect some elements of state corporatism, although not to the extent of COSATU fully
sacrificing its independence. The second type of corporatism is more characteristic of the various
negotiating bodies found in the country, the most important of which is probably NEDLAC.
However, Baskin, as quoted by Friedman ( ), argues that the classic corporatist decision-making
mode, in which policy is shaped by highly structured bargains between equally highly structured
parties (with a high capacity to enforce agreements), is impossible in South Africa. The most that can
be expected is “concertation”, where the parties reach broad understandings though less formal and
binding interchange than the classic corporatist setting. For a full discussion of the corporatist debate,
particularly in the South African context, consult Schreiner ( ).
The RDP White Paper refers to NEDLAC as a mechanism of consultation, coordination, engagement
and negotiation by key stakeholders. This is set out more fully in the RDP source document, which
states that multipartite policy forums constitute important opportunities for organs of civil society to
participate effectively in and influence policy making. Similarly, they provide the democratic
government with an important mechanism for broad consultation on policy matters (ANC, : ).
The document also states that this should not undermine the authority and responsibilities of elected
representative bodies (ANC, : ).
Institutions for building consensus are important determinants of the level of labour market stability
and, therefore, economic viability. The importance of creating democratic institutions capable of
managing this restructuring is vital, especially in South Africa where the process of transformation
requires major economic restructuring. Such bodies reduce the transaction costs of solving conflict
between social partners (NEDLAC, : ). They also help in finding the balance between equity and
efficiency, as perceived by different interest groups, and in reducing tensions.
These institutions not only deepen democracy and involve more stakeholders in policy formulation,
but can also have positive effects in terms of managing the process of transformation. First, they can
help reduce unrealistic expectations, whether from the underprivileged regarding the speed with
which their circumstances will be improved, or from the economically privileged concerning the
extent to which they will have a say in the effects of transformation. Second, such institutions can
foster a better understanding of divergent viewpoints, and therefore help to break down resistance to
change. Finally, such institutions can conceivably provide information to ensure better government
decisions.
The stakeholders (and union leaders in particular – see Webster, : ) are thus expected to play a
dualistic role: first, that of sacrificing their narrow interests to the overall demands of national
development; and second, the representation of the job interests of the rank and file members.
Webster also points out that economic reform is likely to lead to negative consequences that in turn
will lead to real tensions within the trade union movement. This could result in a widening gap
between the leadership that is drawn into compromises in NEDLAC and the rank and file membership
on the shop floor.
Not only the government but also the social partners should thus be fully aware of the shortcomings
of tripartite or multipartite forums. Business and labour represented in such institutions by definition
represent vested interests, albeit important ones, but they do not represent the interests of all the
people or interest groups in the country. It is practically impossible to have certain interest groups
represented on such bodies, for example consumers, the unemployed, rural workers and small
enterprises. The democratically elected representatives of government represent all people and
should give special consideration to the interests of groups not represented on these forums.
Furthermore, consensus achieved in such forums is not necessarily the best solution for a particular
problem, even though consultation such as this normally improves the quality of decisions. By its very
nature such consensus will entail compromises based on the specific interests of the represented
parties and their relative bargaining power. Such consensus solutions are therefore not necessarily the
best from an objective point of view (although even the definition of “objective” might be debatable).
Parties do not necessarily have the organisational capacity and willingness, especially labour and the
community, to make the kinds of sacrifices that corporatism requires (Webster, : ). Often,
however, even the second best solution to a problem might be better than the ideal because, among
other things, it implies broader acceptance of the solution and consequent commitment to its
implementation.
These and other criticisms of NEDLAC, in particular, are discussed below. As will be indicated, there is
increasing criticism of NEDLAC, particularly from within the government. There certainly appears to be
disillusionment with the role that NEDLAC can potentially play, and with the fact that it more often
than not delays the implementation of new policies, without playing a significant role in building
greater consensus. This has led to fears that the corporatist model in South Africa is under threat.
However, Friedman ( ) argues that it is not corporatism that is under threat in South Africa,
because the country never really had corporatism, but the lesser form of “concertation” that exists in
South Africa according to some analysts (see above).
SUMMARY
Tripartite institutions not only deepen democracy and involve more people in policy formulation, but can also have
positive effects in terms of managing the process of transformation. However, they are also subject to shortcomings, for
instance that business and labour represent vested interests, and that the compromises reached might not necessarily
always be in the country’s best interests.
B. NATIONAL ECONOMIC DEVELOPMENT AND LABOUR COUNCIL

The National Economic Development and Labour Council (NEDLAC) was launched on February
to facilitate cooperation and seek consensus between organised business, organised labour and
government on economic, labour and development policy issues (website: http://www.nedlac.org.za).
It is a statutory body, constituted by legislation in , incorporating the National Manpower
Commission (NMC) and the National Economic Forum (NEF). It has taken over all the functions of
these two bodies and is also the constitutional successor to the NMC in terms of the legislative
provision that all labour legislation should be referred to the NMC or its successor before being
submitted to Parliament. The Minister of Labour is the link between government and NEDLAC, but
other ministers and departments are also represented. Both NEDLAC and the Secretariat are financed
by the government. All agreements of NEDLAC are made public and all reports are tabled in
Parliament, unless NEDLAC itself decides otherwise.
The main objectives of NEDLAC are as follows (Nel et al., : ):
To promote economic growth, participation and economic decision making and social equity
To reach consensus and make agreements on matters pertaining to social and economic policy,
including macroeconomic policy, labour policy and developmental issues
To consider proposed labour legislation, as well as significant legislation affecting economic and
development policy, prior to its being introduced in Parliament
To encourage and promote the formulation of coordinated policy on economic and social matters.
NEDLAC is a multipartite body, with the principal participants being business, labour and government.
However, during the deliberations on its establishment, the government indicated that it wanted to
prevent a narrow corporatist arrangement and involve the so-called “community and development
constituency”. Deciding who to appoint to represent the community was quite a challenge – these
had to be organisations that represented a significant community interest on a national basis, had a
direct interest in reconstruction and development, were constituted democratically and were able to
obtain mandates. Only three of the organisations that applied for membership were accepted; they
represent women’s, youth and civic organisations.
Business and labour allocate seats among the various employer and employee organisations. They
also decide on their own representatives in accordance with the allocation of seats. The aim of criteria
and procedures for the admission to NEDLAC of organisations that represent labour and business is
maximum inclusivity of all major national coordinating organisations.
Initially, there were significant problems in representing business, because business at that stage was
divided between Business South Africa and the National African Federated Chamber of Commerce
and Industry (NAFCOC). However, business has since united into a single organisation, Business Unity
South Africa (BUSA).
Labour is still divided among a number of organisations, principally COSATU, FEDUSA and NACTU.
Structures of NEDLAC are:

National summit. It takes place annually, attended by delegates from a wide range of relevant
groups. The function of the summit is to involve as many of these groups as possible to provide
feedback on NEDLAC’s activities, seek input from organisations and individuals normally not
involved in NEDLAC, and obtain broader consultation on the issues being discussed in NEDLAC.
Executive Council. It meets quarterly with the aim of reaching consensus and making agreements
between the parties. It consists of no more than members, with government, labour, business
and community having equal representation. The development constituency is also represented.
Chambers. Four have been established: the Public Finance and Monetary Policy Chamber, the Trade
and Industry Chamber, the Labour Market Chamber and the Development Chamber. They draft
reports and make recommendations for agreement in the Executive Council. Between five and
representatives of labour, business and government form the core membership of all chambers,
except that in the Development Chamber there is a fourth element of representation from
organisations having a direct interest in the issues being debated (the so-called “community
constituency”).
A committee is a cul-de-sac to which ideas are lured and then quietly strangled.
J A. L
Management Committee. It oversees and coordinates the work of NEDLAC. It consists of a

maximum of members from the various constituent organisations, with an additional four
members if deemed necessary. The members meet approximately once a month.
Disillusionment with NEDLAC

When NEDLAC was launched in , it was intended to inaugurate a new era of inclusive decision
making and consensus seeking in the economic and social arenas (Parsons, : ). It was the most
representative policy body that South Africa has ever had, since it included government, labour,
business and the community sector. It was not merely an advisory body, but was supposed to be an
agreement-making body of broadly equal partners.
However, the government’s stance in that macroeconomic policy cannot be negotiated in

NEDLAC seems to have been the beginning of the curtailment of NEDLAC’s influence. It appears that
the government’s philosophy with regard to the role of NEDLAC has gradually changed.
A committee meeting is an event at which the minutes are kept and the hours are lost.
A
The first public signs of the government’s discomfort with tripartism surfaced in , when the then
Director-General of Labour complained in public about NEDLAC’s excessive rights to consider laws
(Friedman, ). In particular, he wanted NEDLAC’s status to be reduced to that of an advisory body.
After the Jobs Summit in , widespread disenchantment set in about the efficacy of the NEDLAC
process. This started with the deadlocked negotiations about the Basic Conditions of Employment
legislation. In particular, the government believed that NEDLAC was obstructing, rather than
facilitating, its policy outcomes (Parsons, : ). The following are the main points of criticism that
have been raised against NEDLAC:
NEDLAC has been unable to provide a shared economic vision, and policy measures have been
considered in an ad hoc fashion, without being fused into a broader economic strategy.
There is much criticism against the unrepresentative nature of the participants in NEDLAC.
Organised labour, organised business (mostly dominated by big business) and government make
compromises, many of which directly affect the vast number of interests not represented at
NEDLAC negotiations, such as small business, the informal sector, unemployed workers, consumers
and provincial and local governments. Many of these parties are those that are the least able to
afford the cost of trade-offs involved in the NEDLAC deal making (CDE, : ).
Viewing NEDLAC as an agreement-making body rather than an advisory body often imposed
pressures on the structure that it could not easily bear in the light of the inadequate levels of trust
between the participating parties. It also meant that NEDLAC’s objective was seen as securing
agreement, rather than achieving a given set of objectives efficiently. Many of the bargains were
simply compromises born of the balance of battling forces, rather than agreements to work
together (Friedman, ). This was probably the reason for the government’s increasing
discomfort about the role of NEDLAC, in particular that it might be inhibiting the government’s
ability to govern (see Friedman, ).
This, of course, raises the question of whether NEDLAC does indeed play a role in ensuring both
equity and efficiency. Lambsdorff ( ) argues that social partnership has turned Germany into a
country that is increasingly incapable of reform. Although the system did, to some extent, spare
Germany the problems that Britain experienced in the s and s, it has also resulted in the
structural problems that caused high unemployment, and a solution seems impossible. According to
Lambsdorff, the economic success of Germany lies in economic freedom and competition, and not the
social partnership of today.
Also, as stated by Bezuidenhout ( : ), labour’s participation in NEDLAC has been hampered by a

lack of capacity to engage consistently in complex negotiations around issues such as trade
agreements. He refers to arguments that labour representatives have not succeeded in moving from a
reactive mode of operation to a proactive mode. This relates not only to the undermining role of the
non-negotiability of the macroeconomic framework, but also to labour not succeeding in setting up
functional structures or a coherent framework of mandating. There are also difficulties in mobilising
workers around the very complex issues under negotiation.
NEDLAC has often been involved in negotiating legislation clause by clause, and presenting
Parliament with a fait accompli. Parliamentarians understandably have felt aggrieved that their
prerogatives as the democratically elected representatives are being challenged. In this regard, it is
important to note the constitutional obligation on Parliament to provide meaningful opportunities
for public participation in parliamentary processes (see for instance, Jenkins ). This includes
proactive steps for stakeholders to be informed of processes of legislature and information on how
to participate in these processes.
There have been excessive expectations as to what the social dialogue in NEDLAC could achieve. It
has been impossible for NEDLAC to meet these expectations in light of the high levels of mistrust
arising from the legacies of the past. Towards the late s and early s, the level of tension
between the alliance partners and particularly between COSATU and the ANC started to increase.
Many of these tensions were often projected into the NEDLAC agenda and debates. This prevented
NEDLAC from dealing with certain issues successfully.
Many other tripartite bodies have been created by NEDLAC agreements or by government policy,
for instance the CCMA, the Employment Conditions Commission (ECC), the Commission for
Employment Equity (CEE) and others. This has detracted from NEDLAC’s role as a central policy-
coordinating forum, and created the danger of duplicating structures that frustrate cohesive policy
making. It has also put increasing pressure on the capacity of both the private and the public
sectors to support a multitude of structures, with the result that important policy issues have not
always been given thorough consideration. Some bilateral bodies have also come into being,
specifically the Millennium Labour Council (MLC), and questions are increasingly being asked about
the relationship between NEDLAC and the MLC (see section B. . ).
There are various signs of NEDLAC slowly being downgraded. One example is its role in drafting and
issuing codes of conduct. In the LRA, drafted in , NEDLAC is permitted to prepare and even issue
such codes. The next labour law, the BCEA ( ), provides for the minister to issue such codes after
having consulted NEDLAC. The Employment Equity Act (promulgated in ) makes no mention of
NEDLAC, but provides for the minister to issue the codes on the advice of the CEE. The Skills
Development Act of similarly refers only to the NSA being consulted when the minister makes
regulations. Another example of NEDLAC’s downgraded role is that some issues are not referred to
NEDLAC at all, for instance the GEAR policy and the Medical Schemes Bill, whereas in many other
cases NEDLAC is given an unreasonably short period of time to find consensus on very complicated
and controversial issues.
In spite of the criticism, South Africa needed an institution such as NEDLAC in the transformation to a
new political regime in the s. NEDLAC had very important conflict management mechanisms,
which helped with a smoother transition to democracy than would otherwise have been the case. In
addition, it was comforting to all parties to be involved in policy making in one way or another, and
this also helped ensure investor confidence. The fact that it was an agreement-making body helped to
entrench stability, and prevented the new government, which had relatively little experience at that
time, from experimenting with populist policies that would have been very damaging to the economy
(Parsons, ).
Grawitsky ( : ) lists the following achievements of NEDLAC:
It has created a forum where the social partners have been able to work on their relationship and
build some level of trust. It is through informal relations that the consolidation of democracy has
been strengthened.
It has been an important instrument to strengthen democratic governance and transparency in the
decision-making process.
It has given birth to a range of tripartite bodies and has thereby extended the concept of social
dialogue.
It has created a central place where greater consensus has been achieved among key stakeholders
on a diverse range of issues.
Friedman’s conclusion ( ) is that NEDLAC will continue to face severe obstacles and wield limited
influence. Parties will continue to attempt to impose their will on the other parties, and will repeatedly
be constrained by the need to take the others’ views into account. However, he remains optimistic
that, in time, parties might find a constructive form of “concertation” more attractive than grudging
recognition of the others as an unpalatable reality.
Impact of business and labour on government policy

As stated by Bezuidenhout ( : ), the very advent of political democratisation opened up space
for the labour movement to influence policy making. Also, its position in society is recognised through
the formalisation of workers’ rights as human rights that are protected in the Bill of Rights enshrined
in the constitution. However, with the establishment of NEDLAC at the same time as the advent of
democracy, commentators differ as to whether institutions such as NEDLAC enable the labour
movement to shape the direction of policy, or whether they merely “institutionalise” the labour
movement to accept, for instance, the government’s macroeconomic strategy. In other words,
whether the situation can be described as state corporatism rather than bargained corporatism.
The labour movement initially achieved several successes through its involvement in NEDLAC. These
include (see Bezuidenhout, : ):
Achieving a relatively progressive Labour Relations Act, despite certain weaknesses

A degree of success in putting human and labour rights on South Africa’s trade agenda, by
convincing the government to require trade partners to sign a “side letter” to trade agreements in
which they commit themselves to respect human rights and to commit themselves to work towards
the ratification of core international labour conventions
Compelling, through the Social Clause Framework Agreement, the South African government to
ratify certain core international labour conventions.
However, the refusal of the government to discuss the GEAR macroeconomic policy in NEDLAC has
curtailed the extent to which the labour movement has been able to use the council to influence the
national developmental policy framework. Instead, according to some observers, NEDLAC has become
an institution in which the implementation of liberalisation can be negotiated (Bezuidenhout, :
).
As far as the business community is concerned, Handley ( ) finds that the South African business
community is probably the largest and most independent-minded private sector on the continent, yet
its ability to lobby government successfully does not reflect this.
Handley ( ) also examines the way in which business in some other African countries has
influenced its respective governments, concluding that while corporate South Africa has done better
than its counterparts in Ghana or Zambia, the business community in Mauritius has been more
successful than business in South Africa.
She concludes that the major weakness of the South African business community is that it has put too
much emphasis on building consensus with organised labour. This approach caused big business,
time and time again, to concede to organised labour on SA’s industrial relations framework,
instituting a rigid system that prices the unemployed out of the job market. It is an approach to policy
making that attempts to minimise potentially damaging conflict, and this had been of great value. It
may be, however, that there are some issues important enough to risk a scrap over (Handley ).
She points out that in spite of concentrating its efforts on labour policy, most of business appears
unhappy with the outcome.
While the government has, since the elections in , radically shifted its macroeconomic policies,
and has adopted a macroeconomic policy that has pleased business, Handley argues that this has
been mainly because of the influence of “that amorphous entity the market”. The business community
does not seem to have had much influence in formulating economic policy.
The CDE ( : ) has also found that the private sector has had to adapt its lobbying strategies to a
new democratic environment, and its response has been slow and uncertain in many areas. A number
of mainly left-of-centre organisations have been much quicker off the mark and are having an impact
on the legislator. These groups are often funded with international money.
SUMMARY
NEDLAC was formed early in to facilitate cooperation and seek consensus between organised business,
organised labour, the development constituency and the government on economic, labour and development policy
issues.
Even though NEDLAC is much maligned, it does remain the only realistic alternative to build consensus on policy
issues among major stakeholders in the country.
B. OTHER NATIONAL CONSENSUS-SEEKING BODIES
B. . Millennium Labour Council

The Millennium Labour Council (MLC) was formed in as a bilateral structure between business
and labour. The objective of the MLC is to develop a shared analysis of the crisis of unemployment
and poverty in the country and to pursue potential solutions with the government and NEDLAC.
In the infancy of societies, the chiefs of the state shape its institutions; later the institutions shape the
chiefs of state.
B M
The MLC comprises members each from business and from trade union constituencies. The trade
unions are generally represented by the leadership of the three main trade union federations in the
country, i.e. COSATU, FEDUSA and NACTU. The business representation is more ad hoc, and consists
of a number of individuals, as well as the chief executives of some business associations, specifically
the South African Chamber of Commerce and Industry (SACCI), the Afrikaanse Handelsinstituut (AHI),
NAFCOC and the Foundation for African Business and Consumer Services (FABCOS). More
fundamentally, the business representatives on the MLC are not elected or appointed by any business
organisation, not least the overarching business lobbying body, BUSA. The business representatives
on the MLC thus not only function without a mandate from business, but also without the implicit
support of a large part of business.
In his address at the launch of the MLC, then President Thabo Mbeki referred to the MLC as a forum
to explore innovative solutions at a bilateral level, a think tank. Mbeki added that the significance of
the MLC did not lie in it being an institution of negotiation, because for that there is NEDLAC.
One of the most important reasons for the MLC coming into being was probably the fact that some
key leaders, especially senior business leaders, were not actively participating in the social dialogue
taking place in the country. The labour leadership corrected this by involving some individual decision
makers from both the white and the black business communities in negotiations with them in this
new body. However, the fact that these business leaders are not specifically elected nor mandated by
organised business is a serious shortcoming. When the MLC became involved in negotiations
regarding the labour law amendments in , it resulted in a deal that was not supported by a large
part of the business community.
It was also never envisaged that the MLC would become involved in the nitty-gritty of negotiations on
labour legislation. Its objective was to develop a shared vision to address current unemployment, job
losses and lack of job creation, together with alleviating the current levels of poverty and inequality.
The MLC and the outcome of the labour law negotiations were also sharply criticised in labour circles.
Jansen ( ) for instance criticises the MLC processes on the following grounds:
Compromising trade union independence. By agreeing to a joint vision about the economic future of
the country, he accuses the trade union representatives of having taken on board the agenda of
“monopoly capital”. The MLC vision refers to capital and labour acting together to make South
Africa the leading emerging market and the destination of first choice for both domestic and
foreign investment. Jansen argues that this can be achieved only by lowering wages, shedding jobs
and allowing working conditions to deteriorate in order to compete with countries such as China,
Indonesia and Vietnam, which, according to him, is the capitalist agenda.
Damaging trade union unity. The MLC agreement on the labour law amendments entrenches a two-
tiered labour market in legislation. Depending on workers’ remuneration, the MLC provides
different rights and remuneration in the areas of retrenchments, Sunday work payments and
whether or not these workers are on probation.
Ignoring principles of trade union democracy. Because the MLC consists of “members” and not
delegates or formal constituency representatives, it undermines the principle of worker control. In
addition, the MLC discussions are confidential and unmandated, which further undermines trade
union democracy.
Jansen ( ) also makes the telling point that the rationale for the MLC’s existence in relation to
NEDLAC has never been fully explained. The Millennium Agreement simply states that the MLC will be
“associated with” NEDLAC as a bilateral council and will operate with full policy autonomy.
SUMMARY
The Millennium Labour Council (MLC) is a non-statutory bilateral structure consisting of individuals from the business and
labour constituencies. The objective of the MLC is to develop a shared analysis of the crisis of unemployment and
poverty in the country and to pursue potential solutions with the government and NEDLAC.
Information deteriorates upward through bureaucracies.

Rule of defactualization.
A B
B. . Other bodies with a more focused role
The National Skills Authority (NSA) (website: http://www.nationalskillsauthority.org.za) was

established in to advise the Minister of Labour on policies and strategies related to skills
development and to report to him on progress made in their implementation. The members of
the NSA represent: organised labour; organised business; the community; government; education
and training providers; experts on employment services; and SAQA (website:
http://www.saqa.org.za) (see section . . ).
The Unemployment Insurance Board advises the minister on matters concerning the Unemployment
Insurance Act, hears and determines appeals regarding unemployment insurance from benefit
committees, and considers applications for the payment of unemployment benefits for periods in
excess of the prescribed period (see section . . ). Benefit committees operate in all the provinces
and consist of equal numbers of employer and union members.
The Compensation Board for Occupational Injuries and Diseases advises the Minister of Labour on
matters related to the compensation of injured employees. It must also be consulted on the
maximum and minimum compensation for temporary and total disability, as well as increases in
benefits.
The Advisory Council for Occupational Health and Safety advises the Minister of Labour on matters
relating to occupational safety and health in terms of the Occupational Health and Safety Act. It
consists of members, among others six representatives, each from business and labour.
The Mines Health and Safety Council was established in terms of the Mines Health and Safety Act. It
has a number of standing committees dealing with safety research, mining regulations and mining
occupational health.
The Employment Equity Act makes provision for the Commission for Employment Equity (CEE). It
advises the minister on issues related to employment equity, relevant codes of conduct and
regulations. It may also do research and make awards recognising the achievements of employers
in furthering the purpose of the act. It consists of a chair and eight other members representing the
state, organised business, organised labour and organisations representing community interests,
although these are sometimes found to be labour representatives in another guise.
The Commission for Conciliation, Mediation and Arbitration (CCMA) (website:
http://www.ccma.org.za) was formed in terms of the LRA to deal with disputes in the world of work
and other matters incidental to the LRA. The CCMA will conciliate most disputes and arbitrate
certain types of dispute if they remain unresolved. It also has other functions, for example to assist
with the setting up of workplace forums, conduct ballots and facilitate retrenchment consultations.
This body is governed by a tripartite body and operates with funds allocated by Parliament. The
governing body consists of members (chair and three representatives each from the state,
organised business and organised labour).
The LRA also makes provision for the establishment of an Essential Services Committee under the
auspices of the CCMA. It determines whether a service is an essential service or a maintenance
service, and deals with related disputes, hence it has an effect on the right of employees in such
services to strike, as well as the right of employers to appoint replacement labour. Labour, business
and government representatives serve on this committee.
Certain disputes (usually of major import), if not resolved by the CCMA in conciliation, may be
referred to the Labour Court. Cases can be taken on appeal to the Labour Appeal Court. Rules to
regulate the conduct of proceedings in the court are made by a multipartite Rules Board.
The Employment Conditions Commission (ECC) is involved in setting minimum wages for
unorganised sectors, and is discussed in section . . .
The principal participants in all the bodies referred to above, with the exception of the Labour Court
and the Labour Appeal Court, are business, labour and government. Other constituencies or individual
experts are also sometimes involved to a greater or lesser extent, for example education authorities,
the providers of education and training, experts on health and safety matters, attorneys and
advocates, etc.
B. CODETERMINATION IN THE WORKPLACE
Whereas corporatism in some form or another aims to improve cooperation and even joint decision
making at national policy level, mechanisms have also been proposed to improve cooperation at the
level of the workplace. Analysts often distinguish between collective bargaining relations, which are
more antagonistic and adversarial, and participative relations, in terms of which conflict can be
reduced to the minimum through regular consultation, joint decision making and compromises by
both parties – this is a more collaborative approach.
However, a more sophisticated view recognises that the relationship between collective bargaining
and participatory structures is not a simple one (see, for instance, Klerck, ). Four different models
have been identified in this regard:
1. Participative structures that are regarded as an alternative to collective bargaining and are often
used by employers to avoid negotiations with unions.
2. Participative structures that are marginal to collective bargaining – most of the activities and
developments are determined through collective bargaining, and there is little commitment to
voluntary participation in other structures, and low levels of trust between the parties.
3. Participative structures that compete with collective bargaining, for instance where union shop
stewards are opposed to such structures because of the role that non-union members might be
playing in such structures or where management attempts to displace issues from collective
bargaining to such “common interest” forums in the hope that the commercial interests of the
enterprise can become an important focus.
4. Participative structures that are regarded as an adjunct to collective bargaining – the two
approaches complement each other rather than compete with each other. Integrative and
distributive processes are each recognised as playing an important role for both parties, and are
pursued in different, separated forums.
One of the most successful and also well-known forms of linking collective bargaining and
participative structures is found in Germany, in the form of its dual structure of workplace forums and
supervisory structures. These are discussed first, after which South Africa’s attempts to introduce
similar structures in its legislation will be discussed.
B. . Codetermination in Germany
Germany has a dualistic labour relations system: one part consists of collective bargaining that is
highly centralised with large industrial unions; the other has codetermination playing the principal
role. The German model of codetermination consists of two parts (see Müller-Jentsch, ; Halbach,
Paland, Schwedes & Wlotzke, ):
(a) At the production unit level, workers elect a works council, which has its legal basis in the Labour
Management Relations Act originally enacted in and extensively amended in . Works
councils are meant to be established at every enterprise with five or more employees. They are
elected by all the employees and have a range of legal rights that force employers to consult or
to reach consensus with the works council on numerous matters affecting labour in the
enterprise.
(b) The workforce in medium-sized or large companies can influence company policy through their
representatives on supervisory boards. All labour members on supervisory boards are elected by
direct ballot or, in some cases, by delegates. Workers in companies with more than workers
have a % representation on the supervisory boards, but the chairperson of the board, who is
normally an employer representative, has a second, casting vote. In companies with between
and workers, workers have a % representation. This codetermination in supervisory
boards extends to all company activities. Thus the supervisory board, for instance, appoints the
members of the management board. It may also revoke their appointment, demand information
on all company matters, and have the last word on important business decisions, for example
with regard to major investments or rationalisation measures.
In large enterprises, the two pillars of codetermination, that of the enterprise and that of the
production unit, have a complementary relationship. Codetermination at enterprise level contributes
to transparency and to an orientation seeking the long-term sustainability of the enterprise, while the
works council is increasingly recognised as a positive contribution to the production process, as a
source of trust, and even as a sort of social-political comanager.
The political implementation goes back to the measures decided upon by the Allies, particularly the
British, with a view to breaking up and reconstituting the formerly highly concentrated coal and steel
sector of the German economy after the Second World War. About years later, in ,
codetermination was extended, with some modifications, to the rest of the economy as part of the
social-liberal reform programme.
Published opinion concerning codetermination is beginning to veer in a negative direction. For

decades it was considered a success story, but now it is beginning to be questioned. Along with area-
wide wage agreements, supplementary wage costs and protection against wrongful dismissal, it is
allegedly responsible for Germany’s current economic weakness. In addition, it is said to be an
obstacle to the harmonisation of corporate law in the European Union. With every new step to extend
codetermination, business has formulated massive criticism of the legal arrangements related to it.
One of the main thrusts of the criticism concerns the limitations imposed on the rights of business to
dispose freely of the assets of the enterprise (Müller-Jentsch, ).
However, practice and research results do not seem to support this criticism. Employers have certainly
also benefited from codetermination – the great crises in mining, in the steel industry, the periods of
recession in the motor car industry, the profound restructuring in any number of large enterprises
would have involved much greater conflict and would have been much less consistent with social
welfare had it not been for the moderating influences exercised by the institutions of codetermination
and the atmosphere of trust that they promote (Müller-Jentsch, ).
Recent surveys carried out among business leaders and managers also show that only a minority
advocates a cutting back or the abolition of the codetermination institutions.
The fact that the German economy has become static of late is in part due to the trade unions that
have remained too inflexible in an age where flexibility has become essential (see Maree, ). The
works councils, on the other hand, have come to play an important role in the improvement of
enterprise performance and the codetermination system has become what can be described as a
system of regulated flexibility. The situation has been reached where works councils are the least
controversial institution of labour relations in Germany. This is due to four factors:
1. The works councils represent all employees (not only union members).
2. They are bound by a peace obligation (they cannot call a strike).
3. They have a problem-solving approach rather than a distributive bargaining role.
4. They do not engage in wage conflicts.
As a result, works councils have probably become the most stable institutions in the German labour
relations system.
Available research literature also shows that in past decades the German economy has had a positive
experience with codetermination (see Müller-Jentsch, ). Empirical studies of a possible negative
effect of codetermination in the enterprise on profits or on the price of shares do, however, yield
ambiguous results (studies pointing to negative effects are contradicted by other studies tending to
show the opposite). Nevertheless, there is considerable evidence that codetermination does tend to
slow down the decision-making process. But there is a positive side to this: although the decisions are
reached more slowly, they are more stable, because they provoke less resistance from the employees.
Experience shows that the employees’ representatives on supervisory boards tend to resist the
temptation to orient their decisions to short-term income considerations at the cost of the longterm
existence of the enterprise.
Above and beyond the questions related to the economic efficiency of codetermination, there are the
social and political aspects. These include issues such as social recognition and democratic culture,
with employees being recognised in their role as participants in the economy and in the production
unit. Participation is thus not only the right of a citizen in the political sphere, but also a right that can
be claimed within the enterprise. The enterprise is not considered simply as a private institution with a
profit motive, but rather as a production cooperative with a range of stakeholders. This is in contrast
to the Anglo-Saxon “shareholders’ model”.
B. . Workplace forums in South Africa

The LRA introduced a new concept, the workplace forum, in an attempt to facilitate a movement away
from distributive collective bargaining towards integrative bargaining (Bezuidenhout, : ). It was
hoped that these forums would provide for codetermination in the workplace and that bargaining
over wages and conditions would gravitate towards the centre. The LRA envisaged a transformation
of adversarial industrial relations in the workplace into a regime of codetermination, where unions
would actively take part in efforts to improve productivity through their involvement in workplace
forums.
The broad objectives of workplace forums were to promote the interests of all employees, irrespective
of union membership, to enhance efficiency in the workplace, and to consult and participate in joint
decision making with the employer. Workplace forums were designed to perform functions that
collective bargaining could not easily achieve, which included the joint solution of problems and the
resolution of conflicts over production.
The workplace forums are intended to encourage joint problem solving and participation on matters
related to productivity and production. Essentially these forums are structures for meetings between
elected representatives of workers and the employers, with the intention of “increasing efficiency in
the workplace” (section of the LRA). For this purpose the workplace forum is entitled to
be consulted on a number of workplace issues (e.g. the introduction of new technology, job
grading, changes in the organisation of work, education and training, partial or total plant closures,
mergers and transfers of ownership, retrenchments and job grading)
participate in joint decision making on a limited number of identified issues (e.g. disciplinary codes,
affirmative action and changes to social benefit schemes).
Workplace forums are not entitled to negotiate wages and conditions of employment, which are both
seen as part of collective bargaining.
Although some of the principles embodied in the act are sound (e.g. that the workplace forum should
seek to enhance efficiency in the workplace), other approaches in the legislation are questionable. The
biggest shortcoming is that a majority union initiates the establishment of a forum. The Labour
Market Commission ( : – ) has expressed its concern on this issue. While the act intended to
prevent unions from being bypassed, it has resulted in the forums being union based, which will
probably reintroduce hostility in these forums.
There is also the question of whether the distinction between collective bargaining and the
organisation of work is not simply artificial (i.e. whether wage bargaining and productivity
enhancement can, in fact, be institutionally separated), especially if unions initiate workplace forums.
The forums might also result in factionalism between unions and other groups of employees, or
between employees themselves, because the whole workforce must elect representatives to the
forum. Furthermore, the employer is obliged to provide resources for the forum, which might result in
these forums having more resources than collective bargaining forums, and possibly undermining the
union (ILO Review, : ).
Klerck ( : ) refers to the concern in union circles that workplace forums could undermine the
alternative of militant unionism that has been quite successful in South Africa. Because of the
extensive requirements regarding consultation, joint decision making and disclosure of information,
the concern for employers is that the forums could inhibit and delay decision-making processes,
which might negatively affect productivity.
It is probably thus not surprising that since the LRA was adopted only a very limited number of
workplace forums have been established. Their net effect is thus indeterminable. Instead of
establishing such forums, there is evidence of a trend towards lean production based on the
casualisation of work and attempts to bypass unions, instead of involving them in restructuring
initiatives.
SUMMARY
Cooperation and joint problem solving in Germany is compulsory, whereas in South Africa the statutory mechanisms to
promote this can only be triggered by a majority union.
B. INTERNATIONAL CONSENSUS SEEKING IN THE LABOUR ARENA: THE

INTERNATIONAL LABOUR ORGANIZATION
The last institution that needs to be briefly referred to in the context of social dialogue is the
International Labour Organization (ILO) (website: http://www.ilo.org). The ILO is a tripartite body and
its purpose is to promote social justice and internationally recognised human and labour rights in the
context of the world of work. It has its headquarters in Geneva, Switzerland. The ILO adopts
conventions and recommendations, which can be regarded as international standards on labour
matters. Through ratification by member states, conventions are intended to create an obligation on
these states to put the provisions of the conventions into effect. South Africa lost its membership of
the ILO in , but rejoined in .
One of the obligations on members is to submit all conventions adopted by the ILO to their
appropriate legislative authority (which is the Parliament in South Africa) within a specified time. It is
in this sense that the ILO will affect the South African labour market. There were numerous
conventions in the past that had a bearing on the labour market and it is probable that the ILO will
adopt similar conventions in future.
At the time of writing, the ILO has member states and employs approximately people in its
Secretariat. Each member country has two government delegates and one employer and one worker
delegate to the Annual International Labour Conference. The governing body of the ILO is elected
every three years and consists of members ( government, employer and worker
representatives) and deputy members ( government, employers and worker
representatives). It meets three times a year.
The ILO also conducts an extensive programme of international technical cooperation to help
promote national economic and social development. This is another way in which the ILO might
influence the South African labour market and assist the country in overcoming some of the labour
market problems highlighted in this book.
SUMMARY
The ILO is a tripartite body and its purpose is to promote social justice in the context of the world of work. The ILO adopts
conventions and recommendations, which can be regarded as international standards on labour matters. A country
subscribes to conventions through ratification.
KEY CONCEPTS
bargained corporatism
codetermination
corporatism
International Labour Organization
Millennium Labour Council
NEDLAC
state corporatism
workplace forums
FOR STUDENTS
. Discuss corporatism as a model for involving stakeholders in decision making.

. Discuss the merits and demerits of involving tripartite consensus-seeking bodies in policy formulation on
labour matters.
. Discuss the structure, functions and effectiveness of the National Economic Development and Labour Council
(NEDLAC).
. Discuss the pros and cons of workplace forums in the context of the codetermination model used in Germany.
SUGGESTED READING*
Calmfors, L. & Driffel, J. . Bargaining structure, corporatism and macroeconomic performance. Economic
Policy: April.
Centre for Development and Enterprise. . Policy-making in a new democracy: South Africa’s challenges for the
st century (abridged). Johannesburg: Centre for Development and Enterprise.
Department of Labour. Annual. Annual Reports. Pretoria: Government Printer.
Friedman, S. . Tripartite policy negotiations in South Africa. Stalemate or productive compromise? Indicator
SA, ( ).
Handley, A. . Business and economic policy: South Africa and three other African cases. Johannesburg: South
Africa Foundation.
Nel, P.S., Kirsten, M., Swanepoel, B.J., Erasmus, B.J. & Poisat, M. . South African employment relations, th ed.
Pretoria: Van Schaik.
Parsons, R. . Steps towards social dialogue and the development of NEDLAC in a democratic South Africa
– . The South African Journal of Economic History, ( & ), September.

C South Africa’s labour market trends, –
NOTE
All tables are calculations done by the author using the – October Household Survey, –
Labour Force Survey September and – Quarterly Labour Force Survey quarter data.
Those with unspecified gender are not shown separately but are included as part of the total, when
the share of each gender is calculated, for tables C. , C. and C. .
Those with unspecified population group are not shown separately but are included as part of the
total, when the share of each population group is calculated, for tables C. , C. and C. .
Those with certificates or diplomas but who did not have Grade fall under the “Grade – ”
category.
Those with unspecified education are not shown separately but are included as part of the total,
when the share of each educational attainment category is calculated, for tables C. , C. and C. .
Formal- and informal-sector employment could not be derived in and (see Table C. ) as
employees were not asked in OHS ( ) and OHS ( ) to declare whether they worked in the
formal or informal sector.
Those with an unspecified occupation are not shown separately in Table C. but are included as
part of the total, when the share of each occupation category is calculated.
Those working within an unspecified industry are not shown separately in Table C. but are
included as part of the total, when the share of each industry category is calculated.
TABLE C. Number of working-age population, labour force, unemployed and discouraged work-seekers
( s)
Working-age Labour force Labour force Employed Unemployed Unemployed Discouraged work-
population (narrow) (broad) (narrow) (broad) seekers
1995 23 930 11 497 13 698 9 470 2 027 4 227 2 200
1996 24 717 11 164 13 505 8 940 2 224 4 564 2 340
1997 25 307 11 520 14 270 9 070 2 450 5 200 2 750
1998 25 471 12 501 14 968 9 346 3 155 5 622 2 467
1999 26 062 13 488 16 208 10 335 3 153 5 873 2 720
2000 27 629 16 325 18 538 12 170 4 155 6 368 2 214
2001 27 848 15 781 18 770 11 132 4 650 7 638 2 988
2002 28 259 16 167 19 353 11 237 4 930 8 116 3 186
2003 28 670 15 802 19 570 11 374 4 428 8 196 3 768
2004 29 027 15 725 19 665 11 596 4 130 8 070 3 940
2005 29 352 16 711 20 017 12 232 4 479 7 786 3 307
2006 29 733 17 124 20 335 12 738 4 386 7 597 3 211
2007 30 087 17 136 20 567 13 238 3 898 7 329 3 431
2008 31 810 18 832 19 923 14 536 4 296 5 387 1 092
2009 32 408 18 292 19 936 13 819 4 473 6 117 1 645
2010 33 009 18 290 20 367 13 638 4 652 6 729 2 077
2011 33 614 18 804 21 015 14 108 4 696 6 907 2 211
2012 34 226 19 449 21 661 14 551 4 898 7 110 2 213
2013 34 841 19 902 22 197 15 025 4 877 7 172 2 295
Working-age Labour force Labour force Employed Unemployed Unemployed Discouraged work-
population (narrow) (broad) (narrow) (broad) seekers
2014 35 463 20 253 22 765 15 106 5 147 7 659 2 512
2015 36 091 21 234 23 459 15 819 5 415 7 640 2 225
2016 36 728 21 694 23 983 15 824 5 869 8 159 2 289
TABLE C. Labour force participation rates, employment rates and unemployment rates (%)
Labour force participation Labour force participation Employment Unemployment rate Unemployment rate
rate (narrow) rate (broad) rate (narrow) (broad)
1995 48.0 57.2 39.6 17.6 30.9
1996 45.2 54.6 36.2 19.9 33.8
1997 45.5 56.4 35.8 21.3 36.4
1998 49.1 58.8 36.7 25.2 37.6
1999 51.8 62.2 39.7 23.4 36.2
2000 59.1 67.1 44.1 25.5 34.4
2001 56.7 67.4 40.0 29.5 40.7
2002 57.2 68.5 39.8 30.5 41.9
2003 55.1 68.3 39.7 28.0 41.9
2004 54.2 67.8 40.0 26.3 41.0
2005 56.9 68.2 41.7 26.8 38.9
2006 57.6 68.4 42.8 25.6 37.4
2007 57.0 68.4 44.0 22.8 35.6
2008 59.2 62.6 45.7 22.8 27.0
2009 56.4 61.5 42.6 24.5 30.7
2010 55.4 61.7 41.3 25.4 33.0
2011 55.9 62.5 42.0 25.0 32.9
2012 56.8 63.3 42.5 25.2 32.8
2013 57.1 63.7 43.1 24.5 32.3
2014 57.1 64.2 42.6 25.4 33.6
2015 58.8 65.0 43.8 25.5 32.6
2016 59.1 65.3 43.1 27.1 34.0
TABLE C. Share of narrow labour force by gender and population group (%)
Gender Race
Male Female African Coloured Indian White
1995 58.2 41.8 67.9 11.8 3.5 16.8
1996 56.8 43.2 66.5 12.4 3.4 17.8
1997 58.1 41.9 67.9 11.9 3.5 16.7
1998 57.3 42.7 69.5 11.1 3.2 16.1
1999 55.4 44.6 69.7 11.2 3.4 15.5
2000 54.4 45.6 73.3 10.0 3.0 13.6
2001 54.8 45.2 72.2 10.3 3.3 14.1
2002 55.0 45.0 72.8 10.3 3.3 13.4
2003 55.3 44.7 72.4 10.5 3.3 13.8
2004 55.8 44.2 72.7 10.5 3.1 13.5
2005 54.3 45.7 73.9 10.2 3.1 12.5
2006 54.0 46.0 74.4 10.2 2.9 12.2
Gender Race
2007 54.5 45.4 74.3 10.0 3.0 12.5
2008 55.1 44.9 74.7 10.2 3.0 12.1
2009 55.1 44.9 74.3 10.7 2.9 12.1
2010 55.6 44.4 74.2 10.8 3.1 11.8
2011 55.1 44.9 74.8 10.8 2.9 11.6
2013 54.4 45.6 75.8 10.4 2.9 10.9
2014 55.0 45.0 76.2 10.7 2.8 10.3
2015 54.9 45.1 77.4 10.0 2.7 9.9
2016 55.2 44.8 78.0 9.8 2.8 9.5
TABLE C. Share of narrow labour force by age cohort (%)
1995 15.4 35.6 28.0 15.1 5.8
1996 15.4 34.6 28.8 15.3 6.0
1997 14.3 35.5 28.8 15.8 5.7
1998 16.0 36.2 27.7 14.8 5.4
1999 16.7 35.6 27.5 14.6 5.6
2000 18.0 34.1 25.0 15.9 7.1
2001 17.9 35.2 25.1 15.4 6.4
2002 18.0 35.4 24.8 15.5 6.3
2003 17.4 35.8 24.5 16.0 6.3
2004 17.0 35.7 24.3 16.4 6.6
2005 17.4 35.7 23.8 16.3 6.8
2006 17.1 35.6 23.9 16.5 7.0
2007 16.7 35.6 24.1 16.5 7.0
2008 15.8 34.8 26.2 16.7 6.6
2009 14.3 34.7 27.3 17.1 6.6
2010 14.2 34.5 27.4 17.3 6.6
2011 13.7 34.2 27.8 17.6 6.7
2012 13.5 34.2 28.2 17.2 6.8
2013 13.3 33.6 28.6 17.6 7.0
2014 12.7 33.9 28.5 17.5 7.5
2015 13.0 33.5 28.7 17.5 7.3
2016 12.7 33.2 28.9 17.9 7.2
TABLE C. Share of narrow labour force by education (%)

1995 31.5 33.6 22.0 12.1
1996 30.4 33.4 23.2 11.9
1997 30.3 34.2 23.2 12.0
1998 32.4 32.4 23.7 11.3
1999 30.7 32.1 23.6 11.5
2000 32.6 33.3 20.8 12.4
2001 29.9 33.1 24.3 11.6
2002 28.3 33.9 24.9 12.0
2003 26.1 33.4 27.4 12.6
2004 25.1 34.4 27.6 12.0
2005 24.2 35.2 27.9 12.1
2006 22.9 35.8 28.2 12.6
2007 22.2 35.1 27.1 14.9
2008 19.5 35.8 29.1 14.5
2009 17.4 35.5 29.9 16.0
2010 16.1 36.7 30.8 15.2
2011 15.3 36.8 30.4 16.4
2012 14.3 38.0 31.0 16.0
2013 14.0 36.4 31.1 17.7
2014 13.2 37.2 31.8 16.9
2015 13.9 36.3 31.9 17.0
2016 12.9 38.2 31.9 16.3
TABLE C. Narrow labour force participation rates by gender and population group (%)
Gender Race
1995 58.6 38.4 43.5 60.4 56.8 64.0
1996 54.5 36.9 39.9 59.3 53.3 64.5
1997 55.1 36.7 40.9 57.5 54.7 62.5
1998 58.6 40.3 45.1 58.6 54.8 65.6
1999 59.6 44.5 47.5 62.9 60.9 68.3
2000 66.3 52.3 56.7 65.2 61.9 68.9
2001 64.0 49.7 53.5 64.4 63.5 69.7
2002 64.5 50.3 54.4 64.9 65.2 68.4
2003 63.0 47.7 51.7 63.6 63.1 70.2
2004 62.3 46.5 50.9 62.3 59.3 69.7
2005 64.0 50.3 54.3 64.3 63.4 68.7
2006 64.2 51.4 55.1 64.7 60.6 68.8
2007 64.3 50.1 54.3 62.2 61.9 71.7
2009 63.9 49.4 53.9 64.3 58.7 68.2
2010 63.1 48.1 52.7 63.8 62.8 67.2
2011 62.9 49.2 53.4 64.3 57.9 68.5
2012 64.1 49.8 54.6 64.5 59.1 67.7
2013 63.3 51.2 54.9 63.9 60.0 69.1
Gender Race
2014 63.9 50.6 54.9 65.6 59.4 67.2
2015 65.5 52.4 57.2 63.6 59.0 68.5
2016 66.0 52.3 57.7 63.1 60.9 67.5
TABLE C. Narrow labour force participation rates by age cohort (%)
1995 21.7 63.9 69.6 62.7 34.4
1996 20.3 59.9 66.9 57.8 32.2
1997 19.6 60.4 66.2 58.2 32.3
1998 22.9 66.4 70.2 62.6 32.7
1999 25.4 68.8 73.4 65.1 35.9
2000 31.9 76.1 79.2 72.7 48.4
2001 30.5 75.1 76.8 67.2 41.6
2002 31.3 75.4 77.4 67.6 41.1
2003 29.3 73.1 74.5 66.4 39.2
2004 28.2 71.3 73.3 66.3 40.1
2005 30.7 74.8 75.4 69.1 42.7
2006 30.6 75.1 77.4 70.1 43.7
2007 29.7 74.6 76.9 69.3 42.7
2008 30.5 76.6 78.7 70.4 44.5
2009 26.7 73.4 77.1 68.5 41.7
2010 26.2 72.0 75.1 67.6 40.8
2011 25.6 72.5 76.0 69.4 41.4
2012 25.9 73.8 77.7 69.1 42.0
2013 26.0 72.8 78.3 71.0 42.5
2014 25.1 73.6 77.4 70.2 44.7
2015 26.8 75.0 79.8 71.7 44.3
2016 26.8 74.6 80.2 73.0 43.3
TABLE C. Narrow labour force participation rates by education (%)

1995 44.8 39.5 61.5 79.6
1996 38.9 38.6 59.6 81.0
1997 39.3 38.1 61.3 82.9
1998 44.7 39.9 65.6 83.4
1999 45.6 43.5 66.7 84.6
2000 55.7 50.7 70.9 87.4
2001 49.8 48.8 72.5 86.4
2002 50.7 48.6 72.8 86.9
2003 47.1 45.9 71.6 88.9
2004 46.0 45.6 70.0 87.0
2005 48.8 48.7 72.1 85.7
2006 49.4 49.1 72.1 87.6
2007 48.3 47.5 71.4 88.8
2008 48.5 49.8 74.7 89.8
2009 44.7 46.9 70.3 88.5
2010 41.8 46.9 69.5 86.8
2011 42.8 46.9 69.1 87.6
2012 43.1 48.2 69.4 87.7
2013 44.6 47.1 69.2 87.8
2014 44.3 47.8 68.2 86.5
2015 46.4 49.1 70.8 86.4
2016 45.8 50.5 70.5 86.7
TABLE C. Share of employed by gender and population group (%)
Gender Race
1995 60.9 39.1 64.6 12.1 3.8 19.6
1996 59.4 40.6 61.3 13.6 3.8 21.4
1997 60.9 39.1 62.8 12.8 4.0 20.4
1998 60.1 39.9 63.1 12.5 3.7 20.6
1999 58.0 42.0 64.3 12.4 3.8 19.3
2000 56.7 43.3 68.4 10.9 3.3 17.2
2001 57.6 42.4 65.8 11.5 3.8 18.8
2002 58.5 41.5 66.5 11.5 3.8 18.1
2003 57.9 42.1 66.4 11.5 3.8 18.3
2005 57.3 42.6 69.1 10.8 3.6 16.2
2006 57.2 42.8 69.4 11.1 3.5 15.7
2007 56.6 43.4 70.3 10.3 3.5 15.6
2008 56.8 43.2 70.8 10.8 3.4 15.1
2009 56.2 43.8 70.3 11.1 3.3 15.2
2010 57.1 42.9 69.9 11.2 3.9 15.1
2011 56.6 43.4 71.1 10.9 3.4 14.6
2012 56.7 43.3 72.1 10.7 3.4 13.8
2013 55.4 44.6 72.8 10.4 3.4 13.4
2014 56.5 43.5 73.0 10.8 3.4 12.8
2015 56.3 43.7 73.9 10.4 3.2 12.5
2016 56.6 43.4 74.3 10.4 3.3 12.0
TABLE C. Share of employed by age cohort (%)

1995 11.9 34.6 30.2 16.8 6.6
1996 12.3 33.4 30.6 16.9 6.8
1997 10.9 33.7 31.0 17.8 6.7
1998 11.7 34.1 30.6 17.1 6.5
1999 12.6 34.0 29.8 16.9 6.8
2000 12.9 32.0 27.6 18.8 8.8
2001 11.8 32.8 28.6 18.8 8.0
2002 11.4 33.5 28.2 18.7 8.1
2003 10.8 34.4 27.7 19.2 8.0
2004 11.1 34.0 27.0 19.5 8.4
2005 11.6 33.9 26.6 19.4 8.6
2006 11.4 34.2 26.2 19.5 8.7
2007 11.5 34.4 26.7 19.0 8.4
2008 10.9 33.4 28.2 19.5 8.0
2009 9.7 32.6 29.7 19.9 8.1
2010 9.1 32.5 30.0 20.2 8.2
2011 8.9 32.0 30.1 20.6 8.4
2012 8.6 32.1 30.8 20.0 8.5
2013 8.8 31.6 30.7 20.3 8.6
2014 8.3 31.3 30.9 20.3 9.2
2015 8.7 31.4 30.9 20.1 8.8
2016 8.0 30.9 31.1 21.0 9.0
TABLE C. Share of employed by education (%)

1995 30.9 32.1 22.1 14.1
1996 28.7 32.1 23.9 14.3
1997 29.3 32.9 23.3 14.2
1998 31.5 30.4 23.9 13.8
1999 30.8 29.9 23.2 13.7
2000 33.4 30.9 19.6 14.9
2001 30.7 29.7 23.8 14.6
2002 29.2 30.0 24.4 15.4
2003 26.8 29.9 26.9 15.8
2004 25.9 30.8 27.0 15.1
2005 24.8 32.0 27.3 15.2
2006 23.4 32.6 27.8 15.6
2007 22.4 31.9 27.0 18.0
2008 20.0 32.8 28.7 17.4
2009 17.6 32.3 29.5 19.3
2010 16.4 33.4 30.3 18.5
2011 15.9 33.5 29.5 19.9
2012 14.7 34.7 30.4 19.4
2013 14.7 32.7 30.6 21.1
2014 13.5 33.6 31.6 20.2
2015 14.6 32.8 31.2 20.2
2016 13.6 34.4 31.9 19.4
TABLE C. Formal- and informal-sector employment ( s)
Formal-sector Informal-sector employment Formal-sector Informal-sector employment

employment (excluding agriculture and employment (including agriculture and
(excluding domestic workers) (including domestic workers)
agriculture) agriculture)
1997 6 421 1 039 6 946 2 051
1998 6 493 1 075 7 217 2 019
1999 6 784 1 567 7 582 2 661
2000 7 061 2 015 7 823 4 009
2001 7 007 1 954 7 767 3 209
2002 7 154 1 767 7 998 3 152
2003 7 347 1 891 8 175 3 146
2004 7 672 1 937 8 293 3 231
2005 7 958 2 442 8 533 3 624
2006 8 360 2 358 8 963 3 705
2007 9 001 2 075 9 664 3 459
2008 10 105 2 276 10 797 3 739
2009 9 779 2 106 10 382 3 437
2010 9 474 2 275 10 060 3 578
2011 9 994 2 262 10 562 3 546
2012 10 303 2 325 10 915 3 636
2013 10 702 2 321 11 329 3 696
2014 10 836 2 406 11 418 3 689
2015 10 924 2 719 11 700 4 119
2016 11 023 2 640 11 797 4 028
TABLE C. Share of employed by occupation (%)
[A] [B] [C] [D] [E] [F] [G] [H] [I] [J]
1995 5.2 3.4 11.2 12.0 11.4 1.2 11.8 11.7 24.7 7.3
1996 4.9 4.1 13.8 9.7 11.6 2.8 13.0 8.7 16.8 8.6
1997 7.3 8.8 8.3 8.8 10.3 3.0 14.4 10.3 16.6 9.1
1998 7.7 5.4 9.7 10.0 12.3 2.4 14.0 10.1 17.8 8.0
1999 6.6 5.4 10.1 10.3 11.8 4.5 13.1 10.6 18.2 7.9
2000 4.7 4.8 9.3 8.7 12.1 9.7 13.0 10.0 19.7 7.7
2001 5.9 4.4 10.5 9.8 12.8 4.6 13.7 10.1 20.1 7.9
[A] [B] [C] [D] [E] [F] [G] [H] [I] [J]
2002 6.4 4.5 10.7 9.9 11.0 6.2 12.9 10.2 20.3 7.5
2003 7.2 4.8 10.1 10.1 11.9 3.0 12.8 10.0 22.1 7.8
2004 7.8 4.0 9.9 10.1 12.5 2.8 13.2 9.6 22.5 7.6
2005 7.0 4.8 9.7 9.7 13.1 2.4 14.2 9.2 22.9 7.0
2006 6.8 4.7 9.6 9.8 12.8 3.3 15.0 8.8 22.2 6.9
2007 7.6 7.6 10.1 8.9 12.1 2.5 13.8 9.1 20.2 7.7
2008 7.8 5.4 10.9 10.7 12.9 0.7 14.0 9.0 21.4 7.3
2009 7.7 5.5 11.7 11.0 13.8 0.6 12.3 8.7 21.6 7.2
2010 8.1 5.2 10.8 10.9 14.3 0.6 12.4 9.2 21.5 7.1
2011 8.4 5.7 10.8 10.7 14.7 0.5 12.5 8.3 21.5 6.9
2012 8.3 5.9 11.2 10.1 15.0 0.5 12.2 8.4 21.8 6.7
2013 8.2 6.4 11.1 11.3 14.5 0.4 11.2 8.3 21.7 6.9
2014 9.0 6.1 10.4 10.6 15.2 0.6 12.0 8.4 21.4 6.4
2015 8.1 5.1 9.3 10.6 15.2 0.6 12.6 8.1 24.0 6.5
2016 8.5 5.3 9.3 10.4 15.6 0.5 12.3 8.3 23.3 6.5
[A]: Managers
[B]: Professionals
[C]: Technicians
[D]: Clerks
[E]: Service and sales workers
[F]: Skilled agricultural workers
[G]: Craft and related trades
[H]: Operators and assemblers
[I]: Elementary occupations
[J]: Domestic workers
TABLE C. Share of employed by industry (%)
[A] [B] [C] [D] [E] [F] [G] [H] [I] [J]
1995 13.0 4.7 15.1 0.9 4.7 17.5 5.0 6.1 22.9 8.4
1996 8.5 2.8 15.5 1.4 4.7 15.3 5.4 8.3 22.5 9.0
1997 8.3 4.3 16.7 1.3 5.6 17.3 5.8 8.0 20.6 8.3
1998 10.0 4.7 14.8 1.2 5.8 19.0 5.9 9.1 19.7 8.2
1999 10.6 4.6 14.5 0.8 5.5 20.0 5.2 9.0 19.1 9.3
2000 15.6 4.9 12.9 0.8 5.6 20.3 4.8 8.0 17.1 9.3
2001 10.5 5.0 14.5 0.9 5.7 22.0 4.9 9.3 17.8 9.2
2002 12.5 5.0 14.5 0.7 5.4 19.4 5.1 9.6 18.1 9.1
2003 10.6 4.9 13.6 0.8 5.8 21.3 4.7 9.6 19.1 9.4
2004 9.1 3.5 14.7 0.9 7.1 21.9 4.9 9.9 18.8 9.2
2005 7.4 3.4 13.9 0.8 7.6 24.6 5.0 10.5 17.8 8.7
2006 8.4 3.1 13.6 0.9 8.0 23.9 4.8 10.2 18.2 8.6
2007 7.8 3.3 13.2 0.7 7.9 22.1 5.2 11.1 19.2 9.0
[A] [B] [C] [D] [E] [F] [G] [H] [I] [J]
2008 5.6 2.3 14.1 0.7 8.1 23.0 5.7 12.2 19.1 9.3
2009 4.9 2.4 13.5 0.7 8.3 22.0 5.8 13.2 20.2 9.1
2010 4.9 2.4 13.3 0.7 8.2 22.6 5.9 12.4 20.6 8.9
2011 4.6 2.5 13.0 0.6 8.1 22.5 5.7 13.3 21.3 8.5
2012 4.8 2.6 12.6 0.7 7.7 21.5 6.2 13.3 22.3 8.4
2013 4.9 2.8 11.8 0.9 7.6 21.2 6.2 13.7 22.4 8.4
2014 4.5 2.9 11.5 0.8 8.5 21.2 6.2 13.4 23.2 7.8
2015 5.7 2.8 11.2 0.8 9.2 20.2 5.7 13.7 22.6 8.1
2016 5.6 2.8 10.6 0.8 9.4 20.2 5.8 14.7 22.1 8.1
[A]: Agriculture
[B]: Mining
[C]: Manufacturing
[D]: Utilities
[E]: Construction
[F]: Trade
[G]: Transport
[H]: Finance
[I]: Community and social services
[J]: Private households
TABLE C. Share of narrow unemployed by gender and population group (%)
Gender Race
1995 45.5 54.5 83.5 10.7 2.1 3.7
1996 46.3 53.7 87.5 7.4 1.8 3.3
1997 47.7 52.3 86.8 8.5 1.6 3.1
1998 49.0 51.1 88.4 6.9 1.9 2.8
1999 46.9 53.1 87.1 7.3 2.3 3.2
2000 47.7 52.3 87.6 7.3 1.8 3.2
2001 48.0 52.0 87.7 7.4 2.1 2.8
2002 46.9 53.1 87.2 7.8 2.2 2.6
2003 48.8 51.2 87.7 7.9 2.0 2.5
2004 49.0 50.9 86.9 8.7 1.6 2.8
2005 45.8 54.1 87.1 8.6 1.8 2.4
2006 44.8 55.2 88.8 7.8 1.1 2.2
2007 47.7 52.2 87.7 9.1 1.1 2.1
2008 49.2 50.8 87.9 8.5 1.5 2.2
2009 51.7 48.3 86.7 9.4 1.5 2.4
2010 51.4 48.7 87.0 9.7 1.0 2.4
2011 50.5 49.5 85.9 10.3 1.2 2.6
2012 51.2 48.8 85.7 10.4 1.3 2.6
Gender Race
2013 51.3 48.7 85.2 10.4 1.4 3.0
2014 50.8 49.2 85.7 10.1 1.3 3.0
2015 50.5 49.5 87.5 8.9 1.3 2.3
2016 51.4 48.6 87.8 8.3 1.4 2.6
TABLE C. Share of narrow unemployed by age cohort (%)
1995 31.9 40.5 18.1 7.5 2.0
1996 27.7 39.3 21.2 9.1 2.7
1997 26.9 42.1 20.6 8.5 2.0
1998 28.5 42.2 19.0 8.2 2.1
1999 30.4 41.0 19.8 7.0 1.7
2000 33.0 40.0 17.4 7.7 1.9
2001 32.4 41.1 16.9 7.3 2.4
2002 33.0 39.6 17.1 8.2 2.2
2003 34.4 39.6 16.4 7.7 2.1
2004 33.5 40.5 16.8 7.4 1.9
2005 33.5 40.4 16.1 7.9 2.2
2006 33.5 39.6 16.9 8.0 2.0
2007 34.5 40.0 15.4 8.1 2.0
2008 32.2 39.4 19.4 7.1 1.9
2009 28.6 41.1 19.8 8.8 1.8
2010 29.3 40.3 19.6 8.8 2.0
2011 28.0 41.0 20.8 8.6 1.7
2012 28.0 40.6 20.7 8.9 1.8
2013 27.3 39.6 21.9 9.3 2.0
2014 25.6 41.8 21.2 9.1 2.3
2015 25.5 39.8 22.2 9.6 2.9
2016 25.5 39.5 23.0 9.7 2.4
TABLE C. Share of narrow unemployed by education (%)

1995 34.0 40.7 21.7 3.1
1996 37.6 38.9 20.0 2.2
1997 34.1 39.0 23.1 3.6
1998 34.8 38.4 23.1 3.7
1999 30.5 39.6 24.8 4.3
2000 30.0 40.3 24.4 5.0
2001 28.3 41.3 25.4 4.5
2002 26.2 42.8 26.2 4.3
2003 24.5 42.2 28.7 4.3
2004 22.7 44.4 29.1 3.3
2005 22.6 44.1 29.3 3.5
2006 21.4 45.1 29.4 3.9
2007 21.5 46.0 27.7 4.6
2008 17.9 45.9 30.6 4.9
2009 16.8 45.5 31.3 5.7
2010 15.2 46.2 32.2 5.6
2011 13.6 46.8 33.0 5.9
2012 13.1 47.9 32.8 5.9
2013 11.9 47.8 32.8 7.2
2014 12.2 47.8 32.4 7.1
2015 11.8 46.3 34.0 7.5
2016 11.2 48.5 31.9 8.0
TABLE C. Narrow unemployment rates by gender and population group (%)
Gender Race
1995 13.8 23.0 21.7 15.9 10.6 3.9
1996 16.2 24.8 26.2 11.9 10.8 3.7
1997 17.5 26.5 27.2 15.3 9.8 3.9
1998 21.6 30.2 32.1 15.8 14.8 4.4
1999 19.8 27.9 29.2 15.3 15.6 4.7
2000 22.2 29.1 30.4 18.5 15.7 5.8
2001 25.8 33.9 35.8 21.2 18.9 5.9
2002 26.0 36.0 36.5 23.0 20.4 6.0
2003 24.7 32.1 34.0 21.1 16.7 5.0
2004 23.1 30.2 31.4 21.7 13.5 5.4
2005 22.6 31.7 31.6 22.5 15.7 5.1
2006 21.2 30.7 30.6 19.5 9.6 4.5
2007 19.9 26.2 26.9 20.7 8.3 3.8
2008 20.4 25.8 26.8 19.0 11.3 4.1
2009 23.0 26.3 28.5 21.6 12.6 4.9
2010 23.5 27.9 29.8 22.7 8.2 5.2
2011 22.9 27.5 28.7 23.9 10.8 5.6
2012 23.3 27.5 28.6 24.7 11.5 6.0
2013 23.1 26.2 27.5 24.5 11.5 6.8
2014 23.4 27.8 28.6 24.1 11.6 7.3
2015 23.5 27.9 28.8 22.8 12.5 5.9
2016 25.2 29.3 30.5 22.9 13.2 7.3
TABLE C. Narrow unemployment rates by age cohort (%)
1995 36.5 20.0 11.4 8.8 6.2
1996 35.9 22.6 14.7 11.8 9.0
1997 39.9 25.2 15.2 11.5 7.3
1998 45.0 29.5 17.3 13.9 9.9
1999 42.5 26.9 16.9 11.2 7.3
2000 46.3 29.9 17.7 12.2 6.9
2001 53.4 34.4 19.8 14.0 10.9
2002 55.9 34.1 21.0 16.1 10.4
2003 55.4 30.9 18.7 13.5 9.1
2004 51.8 29.8 18.2 11.9 7.4
2005 51.5 30.3 18.2 13.0 8.5
2006 50.2 28.5 18.2 12.4 7.2
2007 46.9 25.5 14.5 11.1 6.6
2008 46.6 25.9 16.9 9.7 6.5
2009 48.7 28.9 17.7 12.5 6.9
2010 52.3 29.7 18.2 13.0 7.7
2011 51.0 29.9 18.7 12.1 6.3
2012 52.2 29.9 18.5 13.0 6.8
2013 50.3 28.9 18.8 12.9 7.0
2014 51.3 31.3 18.9 13.2 7.9
2015 49.9 30.3 19.7 14.1 10.1
2016 54.2 32.1 21.5 14.6 9.0
TABLE C. Narrow unemployment rates by education (%)

1995 19.0 21.4 17.4 4.5
1996 24.6 23.2 17.2 3.7
1997 23.9 24.2 21.2 6.4
1998 27.1 29.9 24.6 8.3
1999 23.2 28.8 24.6 8.6
2000 23.3 30.7 29.7 10.2
2001 27.8 36.8 30.8 11.4
2002 28.3 38.5 32.1 10.8
2003 26.2 35.4 29.4 9.5
2004 23.8 33.9 27.7 7.2
2005 25.1 33.6 28.2 7.8
2006 23.9 32.3 26.7 7.9
2007 22.1 29.8 23.2 7.0
2008 20.9 29.3 24.0 7.6
2009 23.5 31.3 25.6 8.8
2010 24.0 32.0 26.6 9.3
2011 22.1 31.8 27.1 8.9
2012 23.2 31.7 26.7 9.3
2013 20.7 32.2 25.8 9.9
2014 23.5 32.7 25.9 10.7
2015 21.7 32.5 27.2 11.3
2016 23.4 34.4 27.0 13.2
Bibliography
Abedian, I. & Standish, B. . Public works programme in South Africa: coming to terms with reality.
Development Southern Africa, ( ): May.
Absa. . South Africa faces growing international competition for human capital. Absa Economic Perspective.
First Quarter.
Adams, A.V., Middleton, J. & Ziderman, A. . Market-based manpower planning with labour market signals.
African National Congress (ANC). . Development and underdevelopment: learning from experience to
overcome the two-economy divide. Johannesburg: African National Congress.
African National Congress (ANC). . The Reconstruction and Development Programme. Johannesburg:
Umanyano Publications.
Allen, K. . Irish success story: a social partnership myth. Business Day, May.
Altman, M. . Revisiting South African employment trends in the s. South African Journal of Economics,
( ): August.
Anvari, V., Ehlers, N. & Steinbach, R. . A semi-structural approach to estimate South Africa’s potential output.
South African Reserve Bank Working Paper Series WP/ / . Pretoria: South African Reserve Bank.
Armstrong, P. & Steenkamp, J. . South African trade unions: An overview for to . Stellenbosch
Arndt, C. & Lewis, J.D. . The macro-economic implications of HIV/Aids in South Africa: a preliminary
assessment. South African Journal of Economics, ( ): December.
Arora, V. . Economic growth in post-apartheid South Africa: a growth-accounting analysis. In Nowak, M. &
Ricci, L.A. (Eds). Post-apartheid South Africa: the first ten years. Washington DC: International Monetary Fund:
– .
Arora, V. & Bhundia, A. . Potential output and total factor productivity growth in post-apartheid South Africa.
IMF Working Paper WP/ / . Washington DC: International Monetary Fund.
Arora, V. & Ricci, L.A. . Unemployment and the labour market. In Nowak, M. & Ricci, L.A. (Eds). Post-apartheid
South Africa: the first ten years. Washington DC: International Monetary Fund: – .
Auer, P. . Employment revival in Europe: labour market successes in Austria, Denmark, Ireland and the
Netherlands. Geneva: International Labour Organization.
Baker, A. . “Euro-sclerosis” permeates labour hierarchy. Business Day, October.
Banerjee, A., Galiani, S., Levinsohn, J., McLaren, Z. & Woolard, I. . Why has unemployment risen in the new
South Africa? Economics of Transition, ( ).
Banerjee, A., Galiani, S., Levinsohn, J. & Woolard, I. . Why has unemployment risen in the new South Africa?
CID Working Paper No. . US: Center for International Development, Harvard University.
Barker, F.S. . ’n Ontleding van die problematiek van mannekragtekorte in Suid-Afrika. Unpublished Doctorate
thesis. Pretoria: University of South Africa.
Barker, F.S. . South Africa’s special employment programme of R million. Development Southern Africa,
( ), May.
Barker, F.S. . The implications of a reduced working week. Unpublished report.
Barker, F.S. a. Best conditions in the Basic Conditions of Employment Act. Contemporary Labour Law, ( ):
December.
Barker, F.S. b. Wage initiatives in mining: a new mould for industrial relations? Address at the Annual
Conference of the Industrial Relations Association of South Africa. Midrand. October.
Barker, F.S. . On South African labour policies. The South African Journal of Economics, ( ): March.
Barker, F.S. . Financing social security. In Strydom, E.M.L. (Ed.). Essential social security law, nd ed. Cape
Town: Juta.
Barker, F.S. & Holtzhausen, M.M.E. . South African labour glossary. Cape Town: Juta.
Barry, F. . How Ireland’s dramatic economic transformation was achieved. Trade and Industry Monitor, :
December.
Becker, G.S. . Human capital, nd ed. New York: National Bureau of Economic Research.
Behar, A. . Would cheaper capital replace labour? South African Journal of Economics, ( ): June.
Belous, R.S. . The rise of the contingent workforce: growth of temporary, part-time, and sub-contracted
employment. Looking Ahead, XIX( ).
Benjamin, P. & Theron, J. . Costing, comparing and competing: developing an approach to the benchmarking
of labour market regulation. DPRU Working Paper / . Cape Town: Development Policy Research Unit,
Berg, A. & Krueger, A. . Lifting all boats: why openness helps to curb poverty. Finance and Development,
September.
Beukes, R., Fransman, T., Murozvi, S. & Yu, D. . Underemployment in South Africa. ERSA Working Paper .
Claremont: Economic Research Southern Africa.
Beukes, R., Jansen, A., Moses, M. & Yu, D. . Exploring the eligibility criteria of the child support grant and its
impact on poverty. Social Indicators Research. Available at:
https://www.link.springer.com/article/ . /s - - -z (accessed on March ).
Bezuidenhout, A. . Towards global social movement unionism? Trade union responses to globalization in South
Africa. Geneva: International Institute for Labour Studies.
Bhorat, H. a. Wage premia and wage differentials in the South African labour market. Submission to the
Commission for Employment Equity. Pretoria: Department of Labour.
Bhorat, H. b. Labour markets and the challenge of poverty and inequality in South Africa. Cape Town:
Development Policy Research Unit, University of Cape Town.
Bhorat, H. a. Wage differences: their origins and policy implications. South African Labour Bulletin, ( ):
February.
Bhorat, H. b. Employment trends in South Africa. Johannesburg: Friedrich Ebert Stiftung.
( ): December.
Bhorat, H. . Unemployment in South Africa: Descriptors and determinants. A paper presented at the Fourth
IZA/World Bank Conference on Employment and Development. Bonn, Germany.
Bhorat, H., Cassim, A. & Yu, D. . Temporary employment services in South Africa: assessing the industry’s
economic contribution. An unpublished report prepared for the Confederation of Associations in the Private
Employment Sector (CAPES). Available at: http://www.capes.org.za/Research/TES-Economic-Impact-on-SA/
(accessed on January ).
Bhorat, H. & Cheadle, H. . Labour reform in South Africa: measuring regulation and a synthesis of policy
suggestions. DPRU Working Paper / . Cape Town: Development Policy Research Unit, University of Cape
Town.
Bhorat H., Goga, S. & Van der Westhuizen, C. . Institutional wage effects: revisiting union and bargaining
council wage premia in South Africa. South African Journal of Economics, ( ): September.
Bhorat, H., Jacobs, E. & Van der Westhuizen, C. . Do industrial disputes reduce employment? Evidence from
South Africa. African Growth Initiative Working Paper . Washington DC: Brookings Institution Press.
Bhorat, H., Leibbrant, M. & Woolard, I. . Towards an understanding of South Africa’s inequality. Paper
delivered at the African Research Consortium Conference. Johannesburg, December.
Bhorat, H. & Lundall, P. . Employment, wages and skills development: firm-specific effects: evidence from a
firm survey in South Africa. South African Journal of Economics, ( ): December.
Bhorat, H., Naidoo, K. & Yu, D. . Trade unions in an emerging economy. WIDER Working Paper / .
Helsinki: United Nations University-World Institute for Development Economics Research (UNU-WIDER).
Bhorat, H. & Oosthuizen, M.J. . Evolution of the labour market. In: Bhorat, H. & Kanbur, R. (Eds). Poverty and
policy in post-apartheid South Africa. Pretoria: HSRC Press: – .
Bhorat, H. & Van der Westhuizen, C. . A synthesis of current issues in the labour regulatory environment. DPRU
Working Paper / .
Cape Town: Development Policy Research Unit, University of Cape Town.
Bhorat, H., Van der Westhuizen, C. & Goga, S. . The role of international trade, technology and structural
change in shifting labour demands in South Africa. Issue Paper No. . Geneva: International Centre for Trade
and Sustainable Development.
Bhorat, H., Van der Westhuizen, C. & Goga, S. . Analysing wage formation in the South African labour market:
the role of bargaining councils. DPRU Working Paper / . Cape Town: Development Policy Research Unit,
Bhorat, H., Van der Westhuizen, C. & Jacobs, T. . Income and non-income inequality in post-apartheid South
Africa: What are the drivers and possible policy interventions? DPRU Working Paper / . Cape Town:
Development Policy Research Unit, University of Cape Town.
Blanchflower, D.G. . Changes over time in union relative wage effects in Great Britain and the United States.
NBER Working Paper . Cambridge, MA: National Bureau of Economic Research.
Blanpain, R. . Work in the st century. Industrial Law Journal, ( ).
Borjas, G.J. . Labor economics, rd ed. New York: McGraw-Hill Higher Education.
Borjas, G.J. . Labor economics, th ed. New York: McGraw-Hill Higher Education.
Borjas, G.J. . Labor economics, th ed. New York: McGraw-Hill Higher Education.
Börsch-Supan, A. . Reduction of working time: does it decrease unemployment? MEA discussion paper -
. Munich: Munich Center for the Economics of Aging.
Bot, M. . School education in South Africa: tracking change over ten years. Edusource Data News, : June.
Bramble, T. & Kuhn, R. . Can corporatism survive? A comparative study of Australia and South Africa. South
African Journal of Labour Relations, ( ): Autumn.
Branson, N., Garlick, J., Lam, D. & Leibbrandt, M. . Education and inequality: the South African case. SALDRU
Working Paper Series . Cape Town: Southern Africa Labour and Development Research Unit, University of
Cape Town.
Braude, J.M. . Braude’s handbook of stories for toastmasters and speakers. Englewood Cliffs, NJ: Prentice-Hall.
Buckberg, E. & Thomas, A. . Wage dispersion and job growth in the United States. Finance and Development:
June.
Bureau for Economic Policy and Analysis. . Economic growth and development constraints in South Africa:
focus on key economic issues. Pretoria: University of Pretoria.
Bureau for Economic Research (BER). . The macro-economic impact of HIV/Aids in South Africa. Stellenbosch:
University of Stellenbosch.
Burger, R.P. & De Villiers, P. . The macroeconomic impact of HID/AIDS in South Africa: a supply-side analysis.
Studies in Economics and Econometrics, ( ): March.
Burger, R.P. & Jafta, R. . Returns to race: labour market discrimination in post-apartheid South Africa.
Stellenbosch Economic Working Papers / . Stellenbosch: University of Stellenbosch.
Burger, R.P. & Jafta, R. . Affirmative action in South Africa: an empirical assessment of the impact on labour
market outcomes. CRISE Working Paper No. . Oxford: Centre for Research on Inequality, Human Security
and Ethnicity.
Burger, R.P., Van der Berg, S. & Von Fintel, D. . The unintended consequences of education policies on South
African participation and unemployment. Stellenbosch Economic Working Papers / . Stellenbosch:
University of Stellenbosch.
Burger, R.P. & Woolard, I. . The state of the labour market in South Africa after the first decade of
democracy. Journal of Vocational Education and Training, ( ): December.
Burger, R.P. & Yu, D. . Wage trends in post-apartheid South Africa: constructing an earnings series from
household survey data. DPRU Working Paper / . Cape Town: Development Policy Research Unit, University
of Cape Town.
Cahuc, P. & Zylberberg, A. . Reduction of working time and unemployment. In Boeri, T., Burda, M. & Kramarz,
F. (Eds). Working hours and job sharing in the EU and USA. Oxford: Oxford University Press: – .
Calmfors, L. & Driffel, J. . Bargaining structure, corporatism and macroeconomic performance. Economic
Policy. April.
Casale, D., Muller, C. & Posel, D. . “Two million net new jobs”: a reconsideration of the rise in employment in
South Africa, – . South African Journal of Economics, ( ): December.
Casale, D.M. & Posel, D.R. . The continued feminisation of the labour force in South Africa: an analysis of
recent data and trends. South African Journal of Economics, ( ): March.
Central Economic Advisory Service. . The restructuring of the South African economy: a normative model
approach. Pretoria: Government Printer.
Centre for Development and Enterprise (CDE) . Policy-making in a new democracy: South Africa’s challenges
for the st century (abridged). Johannesburg: Centre for Development and Enterprise.
Centre for Development and Enterprise (CDE). . Labour-intensive public works: towards providing employment
for all South Africans willing to work. Johannesburg: Centre for Development and Enterprise.
Centre for Development and Enterprise (CDE). a. From laggard to world class: reforming maths and science
education in South Africa’s schools. Johannesburg: Centre for Development and Enterprise.
Centre for Development and Enterprise (CDE). b. Why globalisation works. Johannesburg: Centre for
Development and Enterprise.
Centre for Development and Enterprise and Business Leadership South Africa. . Five million jobs: How to add
five million new jobs to the South African economy over the next five years: Summary. Johannesburg: Centre for
Centre for Development and Enterprise (CDE). a. A fresh look at unemployment. Johannesburg: Centre for
Centre for Development and Enterprise (CDE). b. Skills, growth and borders. Managing migration in South
Africa’s national interest. Johannesburg: Centre for Development and Enterprise.
Centre for Development and Enterprise (CDE). . Schooling reform is possible: lessons for South Africa from
international experience. Johannesburg: Centre for Development and Enterprise.
Chadha, B. . Disequilibrium in the labour market in South Africa. IMF Staff Papers, ( ).
Clarke, O. . Labour market flexibility: two sides of the coin. Social and Labour Bulletin, – .
Commission for Employment Equity. . th Commission for Employment Equity: Annual Report / .
Congress of South African Trade Unions (COSATU). . Creating jobs in South Africa: key issues and strategies.
Labour input to the Presidential Jobs Summit. Johannesburg: Congress of South African Trade Unions.
Conradie, B. . Arbeidstendense op wyn- en tafeldruifplase na die instelling van minimum lone. Cape Town:
Centre for Social Science Research.
Daouas, M. . Africa faces challenges of globalisation. Finance and Development, December.
Deininger, K. & Squire, L. . Economic growth and income inequality: re-examining the links. Finance and
Development, March.
Democratic Alliance. a. From the horse’s mouth: why we need labour brokers. Available at:
https://www.da.org.za/archive/from-thehorses-mouth-why-we-need-labour-brokers/ (accessed on March
).
Democratic Alliance. b. Labour brokers generated R billion in taxes in / . Available at:
https://www.da.org.za/archive/labour-brokers-generated-r -billion-in-taxes-in- / (accessed on
March ).
Department of Basic Education. . The SACMEQ III project in South Africa: a study of the conditions of schooling
and the quality of education. Pretoria: Department of Basic Education.
Department of Education. . Education statistics . Pretoria: Department of Education.
Department of Higher Education and Training. . List of occupations in high demand: . Pretoria:
Department of Higher Education and Training.
Department of Labour. . Green Paper: Policy proposals for a new employment standards statute. Government
Gazette. No. , February.
Department of Labour. a. The social plan guidelines. Pretoria: Department of Labour.
Department of Labour. b. Towards a -hour work week: report of investigation as to how the reduction of
weekly working hours to a level of hours per week may be achieved. Pretoria: Department of Labour.
Department of Labour. . Changing nature of work and “atypical” forms of employment in South Africa.
Department of Labour. a. Labour market review. Pretoria: Department of Labour.
Department of Labour. b. National skills development strategy implementation report. April – March
Department of Labour. . State of skills in South Africa, . Pretoria: Department of Labour.
Department of Labour. . National skills development strategy implementation report. April – March
Department of Labour. . Annual Labour Market Bulletin / . Pretoria: Department of Labour.
Department of Labour. . List of registered trade unions in South Africa on December . Available at:
http://www.labour.gov.za/DOL/downloads/documents/useful-documents/labour-
relations/regtradeunions_dec .pdf (accessed on January ).
Department of Labour. . Sectoral determinations. Available at:
http://www.labour.gov.za/DOL/legislation/sectoral-determinations(accessed on January ).
Department of Labour. Annual. Annual Reports. Pretoria: Department of Labour.
Development Bank of Southern Africa (DBSA). . Report on a study of the effectiveness and efficiency of the
special programmes for creating employment, with special reference to the financial year / .
Johannesburg: Development Bank of Southern Africa.
Devey, R., Skinner, C. & Valodia, I. . Second best? Trends and linkages in the informal economy in South Africa.
DPRU Working Paper / . Cape Town: Development Policy Research Unit, University of Cape Town.
Dollar, D. & Kraay, A. . Trade, growth, and poverty. Finance and Development, ( ): September.
Du Plessis, A. & Smit, B. . South Africa’s growth revival after . Journal of African Economies, ( ): August.
Edgren, G. . How different is the South African labour market? International perspectives and parallels.
Pretoria: Human Sciences Research Council.
Edwards, L. . Globalisation and the skills bias of occupational employment in South Africa. South African
Journal of Economics, ( ): March.
Edwards, L. & Behar, A. . Trade liberalisation and labour demand within South African manufacturing firms.
Working Paper No. . Claremont: Economic Research Southern Africa (ERSA).
Edwards, L. & Golub, S.S. . South African productivity and capital accumulation in manufacturing: an
international comparative analysis. South African Journal of Economics, ( ): December.
Edwards, L. & Schoer, V. . Measures of competitiveness: a dynamic approach to South Africa’s trade
performance in the s. South African Journal of Economics, ( ): September.
Ehrenberg, R.G. & Smith, R.S. . Modern labour economics, th ed. Boston, MA: Pearson.
Ehrenberg, R.G. & Smith, R.S. . Modern labour economics, nd ed. London: Scott Foresman.
El Toukhy, M.M. . Globalisation and developing countries. South African Journal of Economics, ( ):
December.
Elvira, M.M. . Incentives for flexibility? The relationship between incentive pay and employment security.
Proceedings of the th World Congress of the IIRA, . Geneva: International Industrial Relations Association.
eNews Channel Africa (eNCA). . Thousands of unqualified teachers are teaching SA children. eNCA,
September . Available at: http://www.enca.com/south-africa/thousands-unqualified-teachers-sa-schools
(accessed on December ).
Erasmus, J. . Coping strategies of the unemployed. Pretoria: Human Sciences Research Council (HSRC).
Essers, D. . South African labour market transitions during the global financial and economic crisis: micro-level
evidence. WIDER Working Paper / . Helsinki: World Institute for Development Economics Research.
Essop, H. & Yu, D. . The South African informal sector ( - ). Stellenbosch Economic Working Papers
/ . Stellenbosch: Stellenbosch University.
European Commission. . Commission draft for the joint employment report . Brussels: European
Commission.
European Foundation for the Improvement of Living and Working Conditions. . Industrial relations
developments in Europe in . Ireland: European Foundation for the Improvement of Living and Working
Conditions.
Eurostat. December . Statistics explained: wages and labour costs. Available at:
http://www.ec.europa.eu/eurostat/statistics-explained/index.php/Wages_and_labour_costs (accessed on
February ).
Fallon, P. & De Silva, L.A.P. . South Africa: economic performance and policies. Washington DC: World Bank.
Fallon, P. & Lucas, R. . South African labor markets: adjustment and inequalities. Washington DC: World Bank.
Faundez, J. . Affirmative action: international perspectives. Geneva: International Labour Organization.
Fearn, R.M. . Labour economics: the emerging synthesis. Cambridge, MA: Winthrop Publishers.
Fedderke, J.W. . The structure of growth in the South African economy: factor accumulation and total factor
productivity growth – . South African Journal of Economics, ( ): March.
Fedderke, J. . Technology, human capital and growth. Cape Town: University of Cape Town.
Fedderke, J.W. . The cost of rigidity: the case of the South African labour market. Comparative Economic
Studies, ( ): December.
Fedderke, J.W., Manga, J.W. & Pirouz, J. . Challenging Cassandra: household and per capita household income
distribution in the October Household Surveys – , Income and expenditure surveys & , and the
Labour Force Survey . Working Paper No. . Claremont: Economic Research Southern Africa.
Fedderke, J.W. & Mariotti, M. . Changing labour market conditions in South Africa: a sectoral analysis of the
period – . South African Journal of Economics, ( ): June.
Federation of Unions of South Africa (FEDUSA). . FEDUSA debate. Special job creation edition. Johannesburg:
Federation of Unions of South Africa.
Fields, G.S. . Labor market analysis for developing countries. Labour Economics, (Supplement ): December.
Filer, R.K., Hamermesh, D.S. & Rees, A.E. . The economics of work and pay. New York: HarperCollins Publishers.
Finn, A., Leibbrandt, M. & Oosthuizen, M. . Poverty, inequality and prices in post-apartheid South Africa.
WIDER Working Paper / . Helsinki: United Nations University-World Institute for Development
Economics Research (UNU-WIDER).
Finnemore, M. . Introduction to labour relations in South Africa, th ed. Cape Town: LexisNexis.
Fourie, F.C.V.N. . The South African unemployment debate: three worlds, three discourses? SALDRU Working
Paper Series Number . Cape Town: Southern African Labour and Development Research Unit.
Friedman, S. . Tripartite policy negotiations in South Africa: stalemate or productive compromise? Indicator
SA, ( ).
Gailen, A. . Flexible work patterns: why, how, when and where? Bulletin of Comparative Labour Relations.
Bulletin – .
Gladstone, A., Wheeler, H., Rojot, J., Eyraud, F. & Ben-Israel, R. . Labour relations in a changing environment.
New York: Walter de Gruyter.
Global Entrepreneurship Monitor (GEM). . GEM summary results. Available at:
http://www.gemconsortium.org/report (accessed on June ).
Godfrey, S., Maree, J. & Theron, J. . Conditions of employment and small business: coverage, compliance and
exemptions. DPRU Working Paper / . Cape Town: Development Policy Research Unit, University of Cape
Town.
Godfrey, S., Theron, J. & Visser, M. . The state of collective bargaining in South Africa: an empirical and
conceptual study of collective bargaining. DPRU Working Paper / . Cape Town: Development Policy
Research Unit, University of Cape Town.
Goldstone, R.J. . Affirmative action. Cambridge, MA: Centre for International Affairs, Harvard University.
Goliber, T.J. . Population and reproductive health in sub-Saharan Africa. Population Bulletin, ( ): December.
Government of National Unity. . RDP White Paper: Discussion Document. Pretoria: Government Printer.
Grawitsky, R. . NEDLAC: a bonny child or bothersome brat? South African Labour Bulletin, ( ): March.
Gustafsson, M. & Mabogoane, T. . South Africa’s economics of education: a stocktaking and an agenda for the
way forward. Stellenbosch Economic Working Paper / . Stellenbosch: University of Stellenbosch.
Halbach, G., Paland, N., Schwedes, R. & Wlotzke, O. . Labour law in Germany: an overview. Bonn: Der
Bundesminister für Arbeit und Sozialordnung.
Hamermesh, D.S. & Rees, A. . The economics of work and pay. New York: Harper & Row.
Hanami, T. . Individual security of employment in Japan. Fourth Judges’ Labour Law Seminar. Johannesburg,
November.
Handley, A. . Business and economic policy: South Africa and three other African cases. Johannesburg: South
Africa Foundation.
Harrison, A. . Globalisation and poverty. NBER Working Paper . Cambridge, MA: National Bureau of
Economic Research (NBER).
Hausmann, R. . Final recommendations of the International Panel on ASGISA. CID Working Paper No. , May
. Harvard: Center for International Development, Harvard University.
Hayter, S., Reinecke, G. & Torres, R. . Studies on the social dimensions of globalization: South Africa. Geneva:
Herr, J., Kazandziska, M. & Mahnkopf-Praprotnik, S. . The theoretical debate about minimum wages. GLU
Working Paper No. . Berlin: Global Labour University (GLU).
Hertz, T. . The effect of minimum wages on the employment and earnings of South Africa’s domestic service
workers. Pretoria: Trade and Industrial Policy Strategies (TIPS).
Hodge, D. . Growth, employment and unemployment in South Africa. South African Journal of Economics,
( ): December.
Hofmeyr, J.F. . African wage movement in the s. South African Journal of Economics, ( ): December.
Hofmeyr, J.F. . Segmentation in the South African labour market. Potchefstroom: South African Network for
Economic Research.
Holzer, H. & Neukark, D. . Assessing affirmative action. NBER Working Paper . Cambridge, MA: National
Bureau of Economic Research (NBER).
Hosking, S.G. . On social rates of return to investment in education. South African Journal of Economics, ( ):
June.
ILO Review. See Standing et al. .
Ingram, P. & Lindop, E. . Can unions and productivity ever be compatible? Personnel Management, July.
Institute for Futures Research. a. The World Bank’s latest take on globalisation. Economic Issues, ( ):
February.
Institute for Futures Research. b. The socio-economic impact of HIV/AIDS in Southern Africa in the public
sector. Stellenbosch: Institute for Futures Research.
International Labour Organization (ILO). . Amended draft resolution concerning statistics of the economically
active population, employment, unemployment and underemployment.
Thirteenth International Conference of Labour Statisticians. Geneva, – October.
International Labour Organization (ILO). . Job flexibility seen as a solution to unemployment. Social and
Labour Bulletin, / .
International Labour Organization (ILO). . Active labour market policies in a wider policy context. Geneva:
International Labour Organization (ILO). . Press release: ILO highlights global challenge to trade unions.
Geneva: International Labour Organization.
International Labour Organization (ILO). a. An ILO code of practice on HIV/Aids and the world of work. Geneva:
International Labour Organization (ILO). b. World employment report : life at work in the information
economy. Geneva: International Labour Organization.
International Labour Organization (ILO). a. Global trends in wages by sector and occupation. Key indicators of
the labour market. Geneva: International Labour Organization.
International Labour Organization (ILO). b. Trends in the employment intensity of economic growth. Key
indicators of the labour market. Geneva: International Labour Organization.
International Labour Organization (ILO). . Global Jobs Pact Country Scan, South Africa. Geneva: International
Labour Organization.
International Monetary Fund (IMF). . South Africa country report: Article IV consultations. Washington DC:
International Monetary Fund.
International Monetary Fund (IMF). . Finance and Development, ( ): June.
Isaacs, G. A national minimum wage for South Africa. National Minimum Wage Research Institute Summary
Report No. . Johannesburg: National Minimum Wage Research Institute, University of the Witwatersrand.
Jansen, M. . The Millennium Labour Council. South African Labour Bulletin, ( ): April.
Jefferies, A. . The Employment Equity Bill of : a briefing to business. Johannesburg: South African Institute
of Race Relations.
Jenkins, F.S. . A note on public participation in Parliament. Business Unity South Africa. Parliamentary
Newswatch, ( ): August.
Jenkins, R. . Globalisation and the labour market in South Africa. Journal of International Development, ( ):
March.
Jose, A.V. . The future of the labour movement: some observations on developing countries. Geneva:
International Institute for Labour Studies.
Kemp, J. & Smit, B. . Estimating and explaining changes in potential growth in South Africa. Stellenbosch
Keswell, M. & Poswell, L. . Returns to education in South Africa: a retrospective sensitivity analysis of the
available evidence. The South African Journal of Economics, ( ): September.
Khoza, R. . A black management perspective of equal opportunity programmes. IPM Journal, June.
Kingdon, G.G. & Knight, J. . Unemployment in South Africa: the nature of the beast. World Development,
( ): March.
Kingdon, G.G. & Knight, J. . Unemployment in South Africa, – : causes, problems and policies.
Journal of African Economies, ( ): November.
Klein, N. . Real wage, labour productivity, and employment trends in South Africa: a closer look. IMF Working
Paper WP/ / . Washington DC: International Monetary Fund.
Klerck, G. . Mapping the terrain of participation: workplace forums and shopfloor industrial relations. South
African Journal of Labour Relations, , Autumn/Winter.
Knight, J.B. . Labour allocation and unemployment in South Africa. Oxford Bulletin of Economics and Statistics,
May.
Kohler, M.R.A.R. & Bruce-Brand, J.O. . Should South Africa be using trade policy to promote higher labour
standards and prevent the exploitation of unskilled labour? Paper presented at the International Jubilee
Conference on the Economic Empowerment of Southern Africa. Johannesburg, – September.
Kyloh, R. . Flexibility and structural adjustment through consensus: some lessons from Australia. International
Labour Review, ( ).
Labour Market Commission. . See Presidential Commission to Investigate Labour Market Policy.
Labour Research Service. . Bargaining Monitor, ( ): June.
Lambsdorff, O. . The devil’s in the detail. Frontiers of Freedom, nd quarter.
Lawrence, D. & Van der Westhuizen, G. . Trade elasticities for South Africa: an empirical model. South African
Journal of Economics, ( ): September.
Le Roux, P. . Temporary employment services: an important decision. ENSafrica. Available at:
https://www.ensafrica.com/news/temporary-employment-services-an-important-decision?
Id= &STitle=employment% ENSight (accessed on: January ).
Le Roux, P.A.K. . South African unfair dismissal law: time for a reassessment? Contemporary Labour Law, ( ):
July.
Leibbrandt, M., Finn, A. & Woolard, I. . Describing and decomposing post-apartheid income inequality in
South Africa. Development Southern Africa, ( ): March.
Levinsohn, J. . Two policies to alleviate unemployment in South Africa. Ford School of Public Policy, University
of Michigan and NBER, August.
Lewis, J.D. . Promoting growth and employment in South Africa. South African Journal of Economics, ( ):
March.
Lombard, J. . Fokus op ekonomiese kernvrae. Fokus, .
Loots, A.E. . Unemployment in South Africa: a case of jobless growth? Paper delivered at the Economics,
Business Economics and Management conference. Port Elizabeth, November.
Macun, I. . Restructuring and flexibility in manufacturing. Industrial Democracy Review, ( ), February–April.
Madden, L. . Bargaining councils: which way Gulliver? South African Labour Bulletin, ( ): March.
Magruder, J.R. . High unemployment yet few small firms: the role of centralized bargaining in South Africa.
American Economic Journal: Applied Economics, ( ): July.
Mahadea, D. . Employment and growth in South Africa – hope or despair? South African Journal of
Economics, ( ): March.
Mail & Guardian (MG). . Marikana was not just about migrant labour. Mail & Guardian, September.
Available at: https://www.mg.co.za/article/ - - - -marikana-was-not-justabout-migrant-labour
(accessed on March ).
Maree, J. . Beyond traditional employment. Report on the th World Congress of the International Industrial
Relations Association (IIRA) held in Berlin, – September . Industrial Relations Association of South
Africa (IRASA) Newsletter. Summer.
Marshall, F.R., King, A.G. & Briggs, C.M. . Labor economics. Homewood, IL: Richard D. Irwin.
Masson, P. . Globalisation: facts and figures. IMF Policy Discussion Paper PDP/ / . Washington DC:
International Monetary Fund.
Mayer, T. & Grillett-Aubert, L. . The new Dutch model. Goldman-Sachs.
Mazumdar, D. & Van Seventer, D.E. . A decomposition of the real wage rate for South Africa: – .
South African Journal of Economics, ( ), September.
McCarthy, C. . Productivity performance in developing countries. Country case studies: South Africa. Geneva:
United Nations Industrial Development Organization.
McConnell, C.R., Brue, S.L. & Macpherson, D.A. . Contemporary labor economics. New York: McGraw-Hill.
McGrath, M. & Whiteford, A. . Inequality in the size distribution of income in South Africa. Occasional Papers
No. . Stellenbosch: University of Stellenbosch.
Meschi, E. & Vavarelli, M. . Globalisation and income inequality. IZA Discussion Paper No. . Bonn:
Institute for the Study of Labour (IZA).
Meth, C. . Productivity growth and job loss: a temporary inconvenience? Cape Town: South African Network for
Economic Research.
Mischke, C. . Changing terms and conditions of employment. Contemporary Labour Law, ( ): November.
Mlatsheni, C. . International experience in assisting youths to find jobs and lessons for South Africa. In Black,
A. (Ed.). Towards employment-intensive growth in South Africa. Cape Town: UCT Press, – .
Mohr, P., Yu, D. & Mollentze, S. . Economic indicators, th ed. Pretoria: Van Schaik.
Mohr, P. & Seymore, R. . Understanding microeconomics. Pretoria: Van Schaik.
Motshekga, A. . National Assembly for written reply question . Date of publication of internal question
paper: / / (internal question paper: / ). Available at:
http://www.politicsweb.co.za/politicsweb/view/politicsweb/en/page ?
oid= &sn=Detail&pid= (accessed on March ).
Moura Castro, C. . In defence of planning (but only up to a point). International Labour Review, ( ).
Mpedi, L.G. . Labour movement within the Southern African Development Community in the light of the
South African White Paper on international migration. SA Mercantile Law Journal, ( ).
Müller-Jentsch, W. . Codetermination in the enterprise: a proven institution in a consensual society. Article
published in German in ifo Schnelldienst. No. / .
Myburgh, A. . Explaining emigration from South Africa. South African Journal of Economics, ( ): March.
National Economic Forum (NEF). . A framework for implementation of a National Public Works Programme.
Johannesburg: National Economic Forum.
National Labour and Economic Development Institute (Naledi). . Focus: job creation: a framework for job
creation. S.A. Labour Bulletin, ( ): April.
National Labour and Economic Development Institute (Naledi). a. The State of COSATU: Phase one report.
Johannesburg: Naledi.
National Labour and Economic Development Institute (Naledi). b. The workers survey for COSATU.
Johannesburg: Naledi.
National Economic Development and Labour Council (NEDLAC). . Report on the state of social and economic
matters in South Africa – . Johannesburg: NEDLAC.
National Manpower Commission (NMC). . Report on the principle and application of a national minimum
wage with specific reference to the Republic of South Africa. Pretoria: Government Printer.
National Manpower Commission (NMC). . High-level and middle-level manpower in South Africa: recent
developments. Report RP / . Pretoria: Government Printer.
National Productivity Institute (NPI). . Productivity, economic efficiency and equity. A NEDLAC Occasional
Paper / . Johannesburg: National Economic Development and Labour Council.
National Productivity Institute (NPI). . Productivity Statistics . Pretoria: National Productivity Institute.
National Treasury. . Confronting youth unemployment: policy options for South Africa. Pretoria: National
Treasury, February.
National Treasury. . Budget Review. Pretoria: National Treasury.
Nattrass, N. a. Globalisation and the South African labour market. Trade and Industry Monitor, .
Nattrass, N. b. Growth, employment and economic policy in South Africa: a critical review. CDE Focus.
Johannesburg: Centre for Development and Enterprise.
Nattrass, N. c. Labour-demanding growth: lessons from international experience. CDE Focus. Johannesburg:
Centre for Development and Enterprise.
Nattrass, N. . Inequality, unemployment and wage-setting institutions in South Africa. Studies in Economics
and Econometrics, ( ): December.
NEDLAC. See National Economic Development and Labour Council.
Nel, P.S., Kirsten, M., Swanepoel, B.J., Erasmus, B.J. & Poisat, M. . South African employment relations, th ed.
Pretoria: Van Schaik Publishers.
Neutel, M. & Meshmati, A. . Globalisation, inequality and poverty relationships: a cross country evidence. IZA
Discussion Paper No. . Bonn: Institute for the Study of Labour (IZA).
Oelz, M., Olney, S. & Tomei, M. . Equal pay: an introductory guide. Geneva: International Labour Organization.
Organization for Economic Cooperation and Development (OECD). . Labour market flexibility: a controversial
issue. OECD Observer, July.
Organization for Economic Cooperation and Development (OECD). . The Employment Outlook :
reassessing the OECD jobs strategy. Paris: Organization for Economic Cooperation and Development.
Organization for Economic Cooperation and Development (OECD). . Economic Surveys: South Africa:
Overview. July.
Ozaki, M. (Ed.). . Negotiating flexibility: the role of the social partners and the state. Geneva: International
Labour Organization (ILO).
Packer, A. & Christensen, P. . SA and US on education cliff face. Business Day, April.
Parsons, R. . Steps towards social dialogue and the development of NEDLAC in a democratic South Africa
– . The South African Journal of Economic History, ( – ): September.
Paton, C. . Trade union strategies are not helping the poor and unemployed. In Institute for Justice and
Reconciliation (Eds). Transformation Audit : from inequality to inclusive growth. Cape Town: Institute for
Justice and Reconciliation, – .
Pauw, K., Oosthuizen, M. & Van der Westhuizen, C. . Graduate unemployment in the face of skills shortage: a
labour market paradox. South African Journal of Economics, ( ): March.
Penny, N.J.H. . Migrant labour and the South African gold mining industry: a study of remittances. South
African Journal of Economics, ( ): September.
Peter, L. . gems of wit and wisdom. London: Treasure Press.
Population Reference Bureau. . World population data sheet. Washington DC: Population Reference Bureau.
Posel, D. & Casale, D. . What has been happening to internal labour migration in South Africa, – ?
South African Journal of Economics, ( ): September.
Presidential Jobs Summit. . Declaration of the Presidential Jobs Summit. Pretoria: Republic of South Africa.
Prochnow, H.V. & Prochnow, H.V., Jr. . Jokes, quotes and one-liners. London: Thorsons Publishing Group.
Productivity SA. . Productivity Statistics . Midrand: Productivity SA.
Raath, Y. . The dream. People Dynamics, October.
Ranchhod, V. & Finn, A. . Estimating the short run effects of South Africa’s employment tax incentives on
youth employment probabilities using a difference-in-differences approach. South African Journal of
Economics, ( ): June.
Rangasamy, L. & Blignaut, Z. . How exposed is the South African economy to international trade? South
African Journal of Economics, ( ): September.
Rankin, N. & Roberts, R. . Youth unemployment, firm size and reservation wages in South Africa. South African
Journal of Economics, ( ): June.
Ravallion, M. . Pessimistic on poverty? The Economist, March. Available at:
http://www.economist.com/node/ (accessed on June ).
Reddy, V., Bhorat, B., Powell, M., Visser, M. & Arends, A. . Skills supply and demand in South Africa. Pretoria:
Human Sciences Research Council.
Republic of South Africa (RSA). . Growth, employment and redistribution: a macro-economic strategy (GEAR
strategy). Pretoria: Government Printer.
Republic of South Africa (RSA). . Creating jobs, fighting poverty: an employment strategy framework.
Government Gazette, July .
Republic of South Africa (RSA). . Government Gazette. Volume Number . Pretoria: Government
Printer.
Republic of South Africa (RSA). . New Growth Path: the framework. Pretoria: Government Printer.
Republic of South Africa (RSA). . National Development Plan . Our future – make it work. Executive
Summary. Pretoria: Government Printer.
Republic of South Africa (RSA). . General notice: Broad-Based Black Economic Empowerment Act, :
Amended Broad-Based Black Economic Empowerment Codes of Good Practice: For public comments. Pretoria:
Government Printer.
Roche, W.K., Fynes, B. & Morrissey, T. . Working time and employment: a review of international evidence.
Rodgers, G. . Labour market flexibility and decent work. DESA Working Paper No. . New York: Department
of Economic and Social Affairs (DESA), United Nations.
Rodrik, D. . Understanding South Africa’s economic puzzles. CID Working Paper No. . Harvard: Center for
International Development, Harvard University.
Rogerson, C.M. . The impact of the South African government’s SMME programmes: a ten-year review
( – ). Development Southern Africa, ( ).
Rospabé, S. . How did labour market racial discrimination evolve after the end of apartheid? An analysis of
the evolution of employment, occupational and wage discrimination in South Africa between and .
South African Journal of Economics, ( ): March.
Saget, C. . Is the minimum wage an effective tool to promote decent work and reduce poverty? The experience
of selected developing countries. Geneva: International Labour Organization (ILO).
Saunders, S.G. . Estimates of the informal economy in South Africa: some macroeconomic policy implications.
Unpublished Doctoral thesis. Johannesburg: University of Johannesburg.
Schindler, J. . An update of literacy in South Africa. Edusource Data News, : June.
Schlemmer, L. & Levitz, C. . Unemployment in South Africa: the facts, the prospects and an exploration of
solutions. Johannesburg: South African Institute of Race Relations.
Schreiner, G. . Forums and public policy making in a post-apartheid South Africa. Durban: Centre for Socio-
legal Studies.
Shaheed, Z. . Does South Africa need minimum wages as part of a system of labour standards? Paper
presented at a National Manpower Commission workshop on minimum wages. Pretoria.
Shisana, O., Rehle, T., Simbayi, L.C., Parker, W., Zuma, K., Bhana, A., Connoly, C., Jooste, S. & Pillay, V. . South
African national HIV prevalence, HIV incidence, behaviour and communication. Pretoria: Human Sciences
Research Council.
Siebert, W.S. . Investment, employment and South African labour laws: an international comparison. Free
Market Foundation Monograph No. . Johannesburg: Free Market Foundation.
Simkins, C. . The income distribution model. Johannesburg: Urban Foundation.
Simkins, C. . The jagged tear. Human capital, education, and Aids in South Africa, – . CDE Focus.
Johannesburg: Centre for Development and Enterprise.
Small Business Project. . Counting the cost of red tape for business in South Africa. Johannesburg: Small
Business Project.
South Africa Foundation. . Growth for all. An economic strategy for South Africa. Johannesburg: South Africa
Foundation.
South African Business Coalition against HIV/AIDS (SABCOHA) & Bureau of Economic Research (BER). . The
impact of HIV/AIDS on selected business sectors in South Africa. Randburg: South African Business Coalition
against HIV/AIDS & Stellenbosch: Bureau of Economic Research.
South African Chamber of Commerce and Industry (SACCI). . Escalation in strike activity is further harming
vulnerable economy. Available at: http://www.sacci.org.za/index.php?
option=com_content&view=article&id= :escalation-in-strike-activity-is-further-harming-vulnerable-
economy-press-release&catid= :press-office&Itemid= (accessed on March ).
South African Law Commission. . Project : group and human rights. Pretoria: South African Law
Commission.
South African Reserve Bank. . Labour markets and social frontiers. Pretoria: South African Reserve Bank.
Sowell, T. . What’s wrong with affirmative action? Washington DC: American Enterprise Institute for Public
Policy Research.
International Labour Organization (abbreviated as ILO Review).
Statistics South Africa. – . Labour Force Survey: historical revision, September series. Pretoria: Statistics
South Africa.
Statistics South Africa. a. Report on the response by Statistics South Africa to recommendations of the
International Monetary Fund on improvements to the Labour Force Survey. Pretoria: Statistics South Africa.
Statistics South Africa. b. Guide to the Quarterly Labour Force Survey. Report No. - - . Pretoria: Statistics
South Africa.
Statistics South Africa. c. Housing in the Consumer Price Index. Pretoria: Statistics South Africa.
Statistics South Africa. – . Quarterly Labour Force Survey: historical revisions of the QLFS. Pretoria:
Statistics South Africa.
Statistics South Africa. . Estimation of fertility from the Community Survey of South Africa. Report No. -
- . Pretoria: Statistics South Africa.
Statistics South Africa. . Mid-year population estimates, South Africa. Pretoria: Statistics South Africa.
Statistics South Africa. a. Mid-year population estimates . Pretoria: Statistics South Africa.
Statistics South Africa. b. Labour market dynamics in South Africa, . Pretoria: Statistics South Africa.
Steyn, F.G., Meintjies, C.J. & Haasbroek, P.J. . ’n Kostevoordeelstudie van immigrasie in die RSA. Pretoria:
Human Sciences Research Council (HSRC).
Strydom, E. (Ed.). . Essential social security law. Landsdowne: Juta.
Strydom, P.D.F. . Work and employment in the information economy. South African Journal of Economics,
( ): March.
Taylor Committee. . Report of the Committee of Inquiry into a Comprehensive System of Social Security for
South Africa. Pretoria: Department of Social Development.
Taylor, S. & Yu, D. . The importance of socio-economic status in determining educational achievement in South
Africa. Stellenbosch Economic Working Papers / . Stellenbosch: University of Stellenbosch.
Todaro, M.P. & Smith, S.C. . Economic Development, th ed. Boston, MA: Pearson.
Tokiso Dispute Settlement. . Tokiso Review / . Johannesburg: Tokiso Dispute Settlement. Available at:
http://www.tokiso.com (accessed on March ).
Tsikata, Y.M. . Liberalisation and trade performance in South Africa. Informal discussion papers on aspects of
the South African economy, . Washington DC: World Bank.
United Nations. . World Economic and Social Survey : International Migration. New York: United Nations.
United Nations Development Programme (UNDP). . Cultural liberty in today’s diverse world. Human
development report. New York: United Nations.
Urban Foundation. . Public works programmes in South Africa: a key element in employment policy?
Johannesburg: Urban Foundation.
Van Aardt, C. . The impact of HIV/Aids on the South African labour market from a critical perspective. SA
Journal of Demography, ( ).
Van Broekhuizen, H. . Teacher supply in South Africa: a focus on Initial Teacher Education (ITE) graduate
production. ReSEP Conference on Quantitative Applications in Education. Stellenbosch Economic Working
Paper WP / . Stellenbosch: University of Stellenbosch.
Van der Berg, S. . An interracial income distribution in South Africa to the end of the century. South African
Van der Berg, S. . The informal sector: shadow boxing over size. Indicator S.A., ( ).
Van der Berg, S. . Poor educational quality in SA: issues of equity and efficiency. HSRC Seminar Series, Pretoria,
May.
Van der Berg, S. . Current poverty and income distribution in the context of South African history. Stellenbosch
Economic Working Paper / . Stellenbosch: University of Stellenbosch.
Van der Berg, S. & Bhorat, H. . The present as legacy of the past: the labour market, inequality and poverty in
South Africa. DPRU Working Paper / . Cape Town: Development Policy Research Unit, University of Cape
Town.
Van der Berg, S., Louw, M. & Du Toit, L. . Poverty trends since the transition: what we know. Stellenbosch
Van der Westhuizen, C., Goga, S. & Oosthuizen, M. . Women in the South African labour market – .
DPRU Working Paper / . Cape Town: Development Policy Research Unit, University of Cape Town.
Van Dijk, M. . South African manufacturing performance in international perspective. South African Journal of
Economics, ( ): March.
Van Heerden, J.H.P. & Van Tonder, J.J. . Ekonomiese oorsake van vakbondgroei in Suid-Afrika. South African
Van Niekerk, A. a. Flexibility and centralised bargaining. Paper delivered at the Institute for Personnel
Management Conference. Johannesburg, October.
Van Niekerk, A. b. Equal pay: principles, policies and structures. Contemporary Labour Law, ( ): August.
Van Niekerk, A. . Regulating flexibility and small business: revisiting the LRA and BCEA. An unpublished paper
prepared as part of the Employment Promotion Programme. Johannesburg.
Van Niekerk, A. . Regulating flexibility and small business: revisiting the LRA and BCE. A response to Halton
Cheadle’s concept paper. DPRU Working Paper / . Cape Town: Development Policy Research Unit,
Van Tonder, J.J. . Algemene prys- en loonbeleid as ekonomiese instrumente vir die bekamping van die
werkloosheidinflasieprobleem in Suid-Afrika. Unpublished Doctoral thesis. Port Elizabeth: University of Port
Elizabeth.
Venter, R. & Levy, A. . Labour relations in South Africa, th ed. Cape Town: Oxford University Press.
Von Fintel, D. . Wage flexibility in a high unemployment regime: spatial heterogeneity and the size of local
labour markets. Stellenbosch Economic Working Paper / . Stellenbosch: University of Stellenbosch.
Von Fintel, D. & Burger, R.P. . The South African labour market in the global financial crisis: recovering lost
gains. In Institute for Justice and Reconciliation (Ed.). Transformation Audit : recession and recovery. Cape
Town: Institute for Justice and Reconciliation, – .
Webster, E. . NEDLAC: corporatism of a special type? South African Labour Bulletin, ( ): March (reprint of an
article first printed in ).
Webster, E. . COSATU at the crossroads: serving the core or the working poor? South African Labour Bulletin,
( ): December /January .
Whiteside, A.W. . Some aspects of labour relationships between the Republic of South Africa and neighbouring
states. Pretoria: Human Sciences Research Council (HSRC).
Whiteside, A.L. & Sunter, C. . AIDS: the challenge for South Africa. Cape Town: Human & Rousseau.
Wittenberg, M. . The mystery of South Africa’s ghost workers in : measurement and mismeasurement in
the manufacturing census, population census and October Household Surveys. South African Journal of
Economics, ( ): December.
Wolf, M. . million unemployed do not provoke EU panic. Business Day, November.
Wood, A. . North–south trade, employment and inequality. Oxford: Oxford University Press.
Woolard, I. & Leibbrandt, M. . Towards a poverty line for South Africa: background note. Cape Town: Southern
Africa Labour and Development Research Unit, University of Cape Town.
World Bank. . Unions and collective bargaining: economic effects in a global environment. Washington DC:
World Bank.
World Bank. . Doing business in : creating jobs. Washington DC: World Bank and International Finance
Corporation.
World Bank. . South Africa economic update: jobs and South Africa’s changing demographics. Washington DC:
World Bank.
World Bank. . World Bank open data. Available at: http://www.data.worldbank.org (accessed on March
).
World Commission on the Social Dimension of Globalisation. . A fair globalisation: creating opportunities for
all. Geneva: International Labour Organization.
World Economic Forum. . The Global Competitiveness Report – . Geneva: World Economic Forum.
Yudelowitz, J. . ‘One size fits all’ law robs SA of a future. Business Day, August.
Yu, D. . The comparability of the Statistics South Africa October Household Surveys and Labour Force Surveys.
Stellenbosch Economic Working Paper / . Stellenbosch: University of Stellenbosch.
Yu, D. . The South African labour market: – . Stellenbosch Economic Working Paper / .
Stellenbosch: University of Stellenbosch.
Yu, D. . The comparability of Labour Force Survey (LFS) and Quarterly Labour Force Survey (QLFS).
Yu, D. . Defining and measuring informal employment in South Africa: a review of recent approaches.
Yu, D. . An overview of real earnings trends of employed in post-apartheid South Africa. In Hofmeyr, J. &
Nyoka, A. (Eds). Transformation Audit : confronting exclusion. Cape Town: Institute for Justice and
Reconciliation: – .
Index
A
Accelerated and Shared Growth
Initiative of South Africa see
ASGI-SA
active labour market policies –
Advisory Council for Occupational
Health and Safety
affirmative action –
Employment Equity Act of
–
age/experience
AIDS see HIV and AIDS
ASGI-SA –
atypical employment , –
definition ,
Australia –
B
bargaining councils , – , –
agreements, and –
definition
employers, special circumstances
of –
impact of –
impact on labour costs –
importance of –
wages/productivity, and
bargaining power see collective
bargaining
Basic Conditions of Employment Act
of (BCEA) see BCEA
basic income grant –
BBBEE –
BCEA –
BCEA hours of work –
BCEA minimum wages –
birth rate see fertility rate
brain drain see
immigration/emigration
Broad-based Black Economic
Empowerment Act of
(BBBEE) see BBBEE
budget constraint –
C
Calmfors and Driffel model
capital intensity –
capital/labour ratio
definition
CCMA ,
CEE ,
Census see population census
classification of occupations –
codetermination –
Germany, in –
workplace forums –
collective bargaining , – ,
–
flexibility of work force, and –
outcome, factors that determine

–
Commission for Conciliation,
Mediation and Arbitration
(CCMA) see CCMA
Commission for Employment Equity
(CEE) see CEE
communes/communism
Compensation Board for
Occupational Injuries and
Diseases
competitive labour market see
perfectly competitive
labour market
Confederation of South African
Workers’ Unions (CONSAWU)
see CONSAWU
Congress of South African Trade
Unions (COSATU) see COSATU
CONSAWU
contingent employment see atypical
employment
corner solution
corporatism –
definition
COSATU –
cost of labour – , – , –
fixed cost of labour

job creation, strategy –
non-wage labour costs
trade unions –
South Africa, in –
variable cost of labour
cyclical unemployment
definition
D
definitions
atypical employment ,
bargaining councils
capital to labour ratio
corporatism
cyclical unemployment
discrimination
dual labour market
EAP
globalisation
income effect
indifference curve
inflation
informal sector –
invisible underemployment
isoquant curve
labour force
labour market
law of diminishing marginal
product
LFPR
literacy
marginal product of labour
marginal utility of income
marginal utility of leisure
migrant
minimum subsistence level –
monopsony labour market

perfectly competitive labour
market
primary segment
productivity –
reservation wage
scale effect
secondary segment
social plan –
spillover effect
substitution effect ,
superior worker effect
TFR
threat effect
unemployed person
unemployment and expanded
definition of –
unemployment rate
unit labour cost –
value of marginal product (VMP)
visible underemployment
demand for labour –
absorption capacity –
atypical employment
capital/labour intensity –
changes over time –
deriving the demand for labour
–
economic growth, and –
elasticity of labour demand –
elasticity of substitution –
factors influencing –
formal/informal sectors –
importance of –
in South Africa –
informal sector –
long-run demand of –
market –
Marshall’s rules of derived
demand –
measurement of –
measurement of employment
–
occupation/industry, by –
population
group/gender/age/education
–
short-run demand of –
skills, demand for –
total employment –
wages, and –
demand-deficient unemployment
derived demand ,
discouraged work-seekers –
discrimination , –
addressing –
definition
legislation/government policy
theories of –
types
discrimination coefficient
dual labour market – , ,
definition
primary segment
secondary segment –
E
EAP definition
earnings see wages
ECC ,
economic growth , , – ,
–
demand for labour
exports
globalisation, impact of –
rate of
unemployment, and
economically active population (EAP)
see EAP
education – ,
Education and Training Quality
Assurers (ETQAs) see ETQAs
education in South Africa –
shortcomings –
addressing
educational level –
EEA , – , – ,
EEA affirmative action, and –
efficiency wages –
elasticity of labour demand –
elasticity of labour supply –
elasticity of substitution –
emigration see
immigration/emigration
employees
employment – , – , – ,
– , , –
absorption capacity –
atypical employment
changes in over time –
coefficient –
discrimination
economic growth, and –
in South Africa –
informal sector –
job creation strategy –
labour legislation
measurement of –
population
group/gender/age/education
–
sector, by –
skills, demand for –
wages, and –
definition
Employment Conditions Commission
(ECC) see ECC
Employment Equity Act of
(EEA) see EEA
employment programmes –
Employment Tax Incentives Bill
public works –
wage subsidies –
employment/unemployment
inequalities –
enterprise, size of
equal pay –
Essential Services Committee
ETQAs
Europe/USA
experience/age
F
Federation of Unions of South Africa
(Fedusa) see Fedusa
Fedusa
female
employment
labour force – , –
unemployment
fertility rate –
factors influencing
fixed cost of labour
flexible labour market see labour
market flexibility
formal sector, average salaries and
wages –
formal/informal sector employment
–
definition
G
GEAR , ,
Germany –
Globalisation –
challenges –
definition
economic growth –
employment/wages –
inequalities –
labour market, and –
labour market, impact on –
poverty reduction –
South Africa –
South Africa, impact on – ,
–
government involvement/influence
–
Growth, Employment and
Redistribution strategy (GEAR)
see GEAR
H
Hicks strike models –
hiring, restrictions on see restrictions
on hiring
HIV and AIDS – , –
consequences of –
incidence/causes –
hours of work –
human capital –
criticisms of –
education –
future demand –
occupational structure –
skilled labour shortages –
skills
skills development –
South Africa, in – , –
theory –
human resource development see
skills development
I
ILO –
immigration, cost-benefit analysis
–
immigration/emigration –
factors causing
income see wages
income effect
definition
income grant see basic income grant
income/wage inequalities –
indifference curve –
definition
properties –
individual labour supply curves –
industrial action see strike action

inequalities – , –
addressing –
affirmative action –
Broad-based Black Economic
Empowerment see BBBEE
employment/unemployment
–
Employment Equity Act see EEA
equal pay for work of equal value
–
income/wage –
occupational –
South African labour market, in
–
inflation –
definition
informal employment
informal sector –
definition –
importance of –
measurement of –
in-service training –
learnerships
skills development levy
interior solution
international competitiveness –
International Labour
Organization (ILO) see ILO
invisible underemployment
definition
Ireland –
isocost curve
isoquant curve
definition
properties
J
JIPSA
job security –
Joint Initiative on Priority Skills
Acquisition (JIPSA) see JIPSA
L
labour cost – , –
international comparisons –
labour force – , –
age, by – , –
EAP
education, by – ,
female – , – ,
gender, by – , –
occupational structure
participation rate –
race, by – ,
skills
South African –
labour force participation rate (LFPR)
see LFPR
Labour Force Survey (LFS) see LFS
labour intensity –
labour legislation
negative impact of
labour market – , – , , –
, , –
characteristics –
collective bargaining – ,
–
definition
discrimination –
flexibility , – ,
functions of
globalisation, and –
graphic analysis of –
impact of –
inequalities –
international comparison –
objectives of
perfectly competitive –
productivity
social costs/economic benefits
South African, in –
theories of –
types – ,
Labour Relations Act (LRA) see LRA
labour supply see supply of labour
–
definition
learnerships , –
legislation/government policy
discrimination
inequalities, addressing –
LFPR –
definition
female –
male
LFS – , ,
literacy
long-run demand for labour –
LRA , , –
M
market demand for labour –
derivation –
factors affecting it –
marginal product of labour
definition
marginal utility of income
definition
marginal utility of leisure
definition
market supply of labour –
derivation –
factors affecting it –
Marshall’s rules of derived demand
–
measurement of employment –
business
comparing
household –
population census
surveys –
migrants – , –
benefits and costs –
definition
internal –
neighbouring countries –
South/southern Africa –
Millennium Labour Council (MLC) see
MLC
Mines Health and Safety Council
minimum subsistence level –
arguments against –
arguments for
definition
minimum wages – , –
MLC –
monopolist –
monopsony labour market , –
definition
discriminating –
non-discriminating –
N
NACTU
nepotism coefficient
National Council of Trade Unions
(Nactu) see Nactu
National Development Plan see NDP
National Economic Development
and Labour Council see
NEDLAC
National Qualifications Framework
(NQF) see NQF
National Skills Authority (NSA) see
NSA
National Skills Development Strategy
(NSDS) see NSDS
National Skills Fund (NSF) see NSF
NDP –
NEDLAC – , –
Netherlands –
New Growth Path see NGP
NGP –
NQF
NSA
NSDS
NSF
O
occupational differentials –
occupational inequalities –
October Household Survey (OHS)
see OHS
OHS – , ,
P
definition – , , –
Phillips curve –
population –
actors determining –
age –
factors influencing
fertility rate –
growth rates
size –
population census ,
primary segment definition
productivity – , – ,
capital
definition –
efficiency wages, and –
employment, and
importance/measurement of –
job creation
labour
labour cost –
multifactor
trade unions, and –
Q
QES
QLFS – , , , –
Quarterly Employment Statistics
(QES) see QES
Quarterly Labour Force Survey
(QLFS) see QLFS
R
recession,
registered unemployment
reservation wage –
restrictions on hiring
S
SACOTU
SALC
SAQA ,
scale effect
definition
seasonal unemployment
secondary segment
definition
sectoral determination
Sectoral Education and Training
Authorities (SETA) see SETA
SEE
SETA , ,
short-run demand for labour –
skilled labour shortages –
future demand –
skilled workforce – ,
skills – , , –
future demand –
skills base
skills development in South Africa
–
funding –
in-service training –
Joint Initiative on Priority Skills
Acquisition see JIPSA
learnerships
National Qualifications
Framework see NQF
National Skills Fund see NSF
quality assurance –
Sectoral Education and Training
Authorities see SETA
targets
skills development levy
small enterprise sector –
social dialogue –
consensus-seeking bodies –
corporatism as a form of –
International Labour
Organization see ILO
Millennium Labour Council see
MLC
National Economic Development
and Labour Council see
NEDLAC
social plan –
definition –
Solidarity
South African Confederation of
Trade Unions (SACOTU) see
SACOTU
South African Qualifications
Authority (SAQA) see SAQA
spillover effect
definition
strike action –
definition
substitution effect ,
definition ,
superior worker effect
definition
supply of labour –
benefits and costs –
budget constraint –
consequences of –
elasticity of labour supply –
factors affecting –
factors causing
fertility rate –
HIV/AIDS, impact of –
immigration/emigration –
incidence/causes –
income effect –
indifference curves –
individual labour supply curve
–
labour force –
market labour supply –
migrant – , –
population –
reservation wage – ,
South/southern Africa –
substitution effect –
worker preference –
Survey of Employment and Earnings
(SEE) see SEE
Survey of Total Employment and
Earnings
Surveys –
business –
comparing
household –
T
TFR –
definition
factors influencing
theory of demographic transition
theories of discrimination –
crowding model
statistical discrimination
taste-for-discrimination model
–
transaction costs model –
threat effect
definition
total fertility rate (TFR) see TFR
trade liberalisation –
trade unions – , –
agreements, and –
bargaining councils – , –
Calmfors and Driffel model

collective bargaining –
cost of labour, on , –
employers, special circumstances
of –
equality, on
federations –
flexibility/productivity, on –
wages, on –
Hicks strike models –
impact of – , –
importance of –
international –
management-union cooperation
minimum subsistence levels, and

–
minimum wages, and –
movement –
reasons for joining
South Africa, in – , –
strength/influence –
strike action –
types of
wages/productivity, and
training see skills development
training authorities
U
UIF –
Underemployment
unemployed person definition
unemployment , –
active policies –
basic income grant –
characteristics of –
cost of labour –
cyclical
definition –
economic growth/exports
employment programmes –
expanded definition –
frictional
job creation, strategy –
labour absorption –
labour flexibility/productivity
measurement of
Phillips curve –
public works –
reasons for –
registered
seasonal
skills base
small enterprise sector –
social implications –
structural
total –
types
wage subsidies –
young people, for
Unemployment Insurance Board
unemployment insurance –
Unemployment Insurance Fund (UIF)

see UIF
unemployment rate –
definition –
unionisation, degree of
unions see trade unions
unit labour cost – , –
definition –
internationally –
USA/Europe
utility maximisation –
V
vacancy rate
value of marginal product (VMP)
definition
variable cost of labour
visible underemployment
definition
W
wage flexibility –
wage increases –
wage subsidies –
wages – , – , – ,
–
Basic Conditions of Employment
Act see BCEA
collective bargaining
communes/communism
competitive labour market ,
–
determination of – , –
differentials –
economic growth –
efficiency, and –
employment, and –
government
involvement/influence
–
graduates income –
inflation, and –
Labour Relations Act see LRA
monopsony labour market ,
–
part of national income –
segmented labour market –
trade unions, and –
unemployment
wage increases –
work organisation flexibility
working environment
working time flexibility –
workplace forums –

ECS2604 Textbook

Uploaded by

Copyright:

Available Formats

ECS2604 Textbook

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ECS2604 Textbook

Uploaded by

Copyright:

Available Formats

Frans Barker’s

THE SOUTH AFRICAN LABOUR MARKET

Tel: DALRO (from within South Africa) or + ( )

Commissioning editor Marike Visagie

Website addresses and links were correct at time of publication.

This book has been reviewed by independent peer reviewers.

Non-economists. Non-economists may be disheartened to see some complicated graphs in parts of

Quotations. The sometimes inspiring, sometimes light-hearted and sometimes tongue-in-the-cheek

Preface to the sixth edition

Finally, my gratitude to Our Father, who assists with most things.

As always, comments and suggestions will be valued and carefully considered.

In memory of Frans Barker –

ABET Adult basic education and training

Business Unity South Africa http://www.busa.org.za

active labour market policies

atypical or “non-standard” employment

dual labour market

index number; index

internal labour market

labour force participation rate

labour market flexibility

law of diminishing marginal product

marginal revenue product

minimum living level

segmented labour market

total fertility rate

unit labour cost

value of marginal product

Chapter The supply of labour

Chapter The demand for labour

Chapter Unions, collective bargaining and minimum wages

Chapter Productivity and labour market flexibility

Chapter Unemployment in South Africa

Chapter Human capital and the demand for skilled workers

Chapter Labour market inequalities and discrimination

Outline of this chapter

. WHY STUDY THE LABOUR MARKET?

Positive economics versus normative economics

. UNIQUE CHARACTERISTICS OF THE LABOUR MARKET

Labour markets have important functions:

Labour markets should, by their functioning, contribute to the following objectives:

. WHY A THEORY WITH UNREALISTIC ASSUMPTIONS?

. . Perfectly competitive labour market

I don’t meet competition, I crush it.

. . Graphic analysis of the labour market

FIGURE . Equilibrium between supply and demand

There can be the following changes in respect of these curves:

. IMPORTANT CHARACTERISTICS OF THE LABOUR MARKET IN SOUTH AFRICA

. STRUCTURE OF THE BOOK

The structure of the book is presented in Figure . .

* The bibliography contains the full list of references.

Outline of this chapter

. THE LABOUR FORCE AS THE SUPPLY OF LABOUR

The supply of labour is determined by a number of factors, as graphically illustrated in Figure . .

Source: Adapted from McConnell et al. ( : )

. . The South African labour force

. . Labour force by gender

TABLE . The labour force of South Africa by gender (’ )