How To Do A Gender Sensitive Budget Analysis

A usA I D
How to do a gender-sensitive budget analysis:

Contemporary research and practice
Debbie Budlender & Rhonda Sharp

with Kerri Allen
Commonwealth
Secretariat
contemporary research and analysis
About the authors

Debbie Budlender
Debbie is a Principal Researcher with the Community Agency for Social Enquiry, a
non-governmental organisation involved in social policy research in Cape Town,
South Africa. She is also seconded to South Africa's Statistical Bureau (Statistics SA)
where she works primarily on gender and employment issues. She is editor of three
South African Women’s Budgets (1996, 1997, 1998).
Rhonda Sharp
Rhonda is an Associate Professor of economics in the School of International
Business and Director of the Research Centre for Gender Studies, University of South
Australia, Adelaide, Australia. She has published widely in the field of feminist
economics, particularly in the area of women and economic policies.
Kerri Allen
Kerri is a Research Associate in the School of International Business, University of
South Australia, Adelaide, Australia. Her research and writing cover a broad range of
issues in relation to gender inequality and work.
1998 Debbie Budlender, Rhonda Sharp & the Commonwealth Secretariat
ISBN 0 86803 615 3
Published by Australian Agency for International Development,

Canberra and the Commonwealth Secretariat, London
Text design, typsetting, editorial assistance Barbara L Brougham,
Flexible Learning Centre,
University of South Australia, Underdale 5032
Printing Document Services, Underdale SA 5032
Distribution Commonwealth Secretariat
Marlborough House, Pall Mall,
London SW1Y 5HX
Telephone: (44) 01718393411
Fax: (44) 01719300827
Suggested retail price: £8•99
ii
Table of Contents
Section 1
Introduction 3
Section 2
What is a gender-sensitive budget 5

Gender-neut ral or gender blind? 5
Mainst reaming gender 8
Co unt ry examples 9
Init iat ives in o t her co unt ries 12
Section 3
Gender, sex, men and women 17

Sex and gender 17
Gender issues in government budget s 19
Women and men in different ro les 22
Different ro les and gender bias in budget ary
reso urce allo cat ions 24
Gender indicat ors 25
Ho w can we measure gener and gender relat ions? 28
Lo cal co nt ext 30
Section 4
What is a government budget 31

The funct io ns of a government budget 31
The budget funct ions and policy 33
Budget ary co nst raint s 34
Variat io ns bet ween count ries 34
The pro cess of drawing up t he budget 34
Section 5
Tools for a gender-sensitive analysis of budgets 37

Gender-aware policy apraisal 38
Gender-disaggregat ed beneficiary assessment s 41
Gender-disaggregat ed public expendit ure incidence
analysis 44
Gender-disaggregat ed t ax incidence analysis 47
Gender-disaggregat ed analysis of t he impact of t he
budget o n t ime use 48
Gender-aware medium t erm eco no mic po licy framewo rk 50
Gender-aware budget st at ement 50
iii
Section 6
Putting care into the economy 51

Ho usehold work in t he syst em of nat ional acco unt s 51
Time use 52
Engendering econo mic models 55
The care econo my 55
Section 7
Preparing a gender-sensitive budget statement 57

Gender budget s: A framework for public expendit ure 57
Gender budget s: A framework for public revenue 68
Section 8
Portfolio work 73
Int ro duct io n t o po rt fo lio work 73
Preliminary t asks 73
Dat a needs 73
Developing a port folio analysis 74
Developing an act ion plan 77
Section 9
References and bibliography 79
iv
List of comments
comme n t 1 What i s a ge nde r se nsi ti ve budge t? 5
comme n t 2 G e nde r bl i ndne ss 6
comme n t 3 G ove r nme nt budge ts and ge nde r 6
comme n t 4 M ai nstr e ami ng ge nde r budge ts i n

Fi nanc e mi ni str i e s 8
comme n t 5 Austr al i a: Intr oduc ti on to the 1995-96

Wome n’s B udge t State me nt 11
comme n t 6 M oz ambi que : A woman’s budge t 13
comme n t 7 The Tanz ani an G e nde r B udge t Ini ti ati ve 14
comme n t 8 Re c omme ndati ons to the UK G ove r nme nt 15
comme n t 9 A snapshot of ge nde r -r e l ate d Wor l d B ank

l e ndi ng 15
comme n t 1 0 Pr i vati sati on vs de -i nsti tuti onal i sm 16
comme n t 1 1 Se x and ge nde r 17
comme n t 1 2 G e nde r e d soc i al r e l ati ons 18
comme n t 1 3 Wome n and me n 18
com m en t 14 B asi c wate r suppl y 19
com m en t 15 The e ffi c i e nc y be ne fi ts of r e duc i ng ge nde r

i ne qui ty 21
v
comme n t 1 6 Wome n’s r ol e s 22
comme n t 1 7 H e al th se r vi c e r e str uc tur i ng: i nc r e ase i n

e ffi c i e nc y or tr ansfe r of c osts 25
comme n t 1 8 Indi c ator s 26
comme n t 1 9 Cate gor i si ng ge nde r stati sti c s 28
comme n t 2 0 The budge t and pol i c y 33
com m en t 21 Austr al i an Commonwe al th budge t ti me tabl e

1994-1995 35
comme n t 2 2 Tool s for ge nde r -se nsi ti ve anal ysi s of budge ts 37
comme n t 2 3 Land r e for m i n South Afr i c a 39
comme n t 2 4 Dome sti c Vi ol e nc e Pol i c y i n Tasmani a, Austr al i a 40
comme n t 2 5 B e ne fi c i ar y asse ssme nt of publ i c e xpe ndi tur e i n

Austr al i a 42
comme n t 2 6 B e ne fi c i ar y asse ssme nt of he l ath se r vi c e

de l i ve r y i n Sr i Lank a 43
comme n t 2 7 Di saggr e gati ng be ne fi t i nc i de nc e : G e nde r vs

publ i c soc i al spe ndi ng 44
comme n t 2 8 Sr i Lank a food r ati on and stamps pr ogr amme 46
comme n t 2 9 Val ue adde d tax i n South Afr i c a 48
com m en t 30 I s ti me an asse t? 49
com m en t 31 G e nde r di ffe r e nc e s i n ti me use 49
vi
com m en t 32 Austr al i an e sti mate s of the val ue of unpai d
house hol d wor k 53
comme n t 3 3 The e ffe c t of mar r i age on the ti me use of

Austr al i an me n and wome n 54
comme n t 3 4 The c ar e e c onomy 56
comme n t 3 5 South Austr al i a: Agr i c ul tur e

(Expe ndi tur e Cate gor i e s 1, 3 and 2) 58
comme n t 3 6 Two stor i e s about publ i c se c tor e mpl oyme nt 62
comme n t 3 7 Commi tte e s i n B ar bados

(Expe ndi tur e Cate gor y 2) 63
comme n t 3 8 O mbudsman’s O ffi c e

(Expe ndi tur e Cate gor y 3) 64
comme n t 3 9 The Austr al i an Wome n’s B udge t - Assi stanc e

for B usi ne ss (Expe ndi tur e Cate gor y 3) 65
comme n t 4 9 The Nati onal Publ i c Wor k s Pr ogr am i n South

Afr i c a (Expe ndi tur e Cate gor y 3) 67
comme n t 4 1 Pe r sonal i nc ome tax i n Austr al i a 69
com m en t 42 B r i tai n’s ne w tax-be ne fi t de al for wor k i ng

fami l i e s 71
com m en t 43 Data 74
vii
List of figures and tables
Figure 2.1 Percentage of projects with gender-related actions, 1988-94 15

Figure 6.1 His marriage and her marriage 56
Figure 6.2 The interdependence of the paid and unpaid economies 57
Table 3.1 Comparison of HDI, GDI and GEM ranks

for selected countries (1994) 27
Table 3.2 Unemployment rates (%) in South Africa, 1995 - gender and race 29
Table 3.3 Unemployment rates (%) in South Africa, 1995 - gender and
location 30
Table 5.1 Positive recognition of benefits from public expenditure (% of
respondents) 42
Table 5.2 Percentage of respondents wanting increased government

expenditure in selected portfolios 42
Table 5.3 Gendered incidence of public expenditure on education 45

Table 6.1 Average time in minutes spent daily on activities by Australian men
and women 53
viii
Acknowledgements
This document is the result of the efforts of many people. It is a concrete

expression of our involvement in the development of women’s budget exercises
in Australia and South Africa and of our work with policymakers, community
groups and workshop participants in several countries over a number of years.
We would like to thank these people for their ideas and their interest in, and
commitment to, developing gender-sensitive budgets.
This document has benefited from being able to draw on the emerging
international research on gender and economics. The work of Diane Elson,
University of Manchester, has been particularly important and the document
draws extensively on a range of her publications. The 1995 special issue of the
journal World Development, ‘Gender, Adjustment and Macroeconomics’, offers
a rich source of conceptual and theoretical possibilities for an economics that
makes gender issues visible. We have also drawn on our long association with
the practical and theoretical work of many people from both developing and
industrialised countries who have sought to engender public policy, public
sector institutions and economic development processes.
We would like to express our appreciation to those who have given us
permission to use their work. Your generosity has contributed to a better
document.
Several institutions have provided important assistance and resources. The
Australian Agency for International Development (AusAID) and the
Commonwealth Secretariat provided funding for the project. The Gender and
Youth Affairs Division of the Commonwealth Secretariat, by employing us as
consultants in their pilot project, ‘Integrating gender into national budgetary
policies and procedures within the context of economic reform’, provided us
the opportunity to work in, and learn from, different countries within the
Commonwealth.
We would like to thank our own institutions for their encouragement and
practical support - the Community Agency for Social Enquiry in Cape Town,
South Africa and the Research Centre for Gender Studies and the School of
International Business at the University of South Australia, Adelaide, Australia.
Finally, the International Association For Feminist Economics, through its
electronic forum, Femecon-L and annual conferences initially put us in touch
with each other, thus making this collaborative project possible.
Rhonda Sharp and Debbie Budlender

November 1998
ix
x
contemporary research and practice
white spot
Section 1
Introduction
This document grew out of the experiences of developing gender-sensitive

budgets in several countries. Australia was the first country to develop a
gender-sensitive budget with the Federal government publishing in 1984 the first
comprehensive audit of a government budget for its impact on women and girls.
Women’s budget exercises were also undertaken by each of the Australian State
and Territory governments at various times during the 1980s and 1990s. In
South Africa, parliamentarians, together with non-governmental organisations,
started working on a gender-sensitive analysis of budgets in 1995. In 1997 the
South African government, too, started doing a gender-sensitive budget
analysis. Today many other countries have joined these two in undertaking these
budget exercises. Our data has been drawn from those countries which already
have gender-sensitive budgets in place, or are in the process of initiating them.
There is no single approach or model of a gender sensitive budget exercise. In
some countries, for example, these exercises are implemented by the
government while in other countries individuals and groups outside government
undertake the budgetary analysis. This publication is primarily aimed at
informing those governments considering undertaking a gender-sensitive budget
exercise. It covers the issues, methods and strategies for the first year of
implementing such an exercise.
While the document has a strong practical orientation we recognise that good
practice needs a strong research base. Accordingly, we have sought to apply a
range of ideas from the emerging research on gender and government budgets
with the objective of pointing to new approaches and models of a gender-
sensitive budget analysis. However, we recognise that this is a rapidly
developing area and new ideas and practices are emerging even as we go to
press.
The material in this document has been organised into 7 sections following this
introduction. Section 2 addresses the meanings and terms applied to gender-
sensitive budgets and gives examples from countries undertaking or initiating
these exercises. Section 3 discusses how gender issues arise in a budgetary
context. Section 4 provides background information on government budgets and
introduces seven tools or methods which may be used in a gender-sensitive
analysis of budgets. Section 5 gives a range of applications of these tools, using
examples from countries which are engaged in gender-sensitive budget
exercises. Section 6 provides a brief introduction to new theoretical
developments in macroeconomics, which take into account unpaid care
responsibilities, and discusses their budgetary implications.
Section 1: Introduction 3
How to do a gender-sensitive budget analysis
Budlender, Sharp & Allen
Section 7 shows how to prepare a gender-sensitive budget statement. A system

of dividing up government expenditures and revenues is outlined. Section 8
provides a initial (in the first year) approach for those governments wishing to
undertake a gender-sensitive analysis of their budget.
Typically, this publication might form the basis of a series of structured
workshops which bring together civil servants from the Ministry of Finance and
a number of selected portfolios. Ideally, participants would include those with
responsibility for budgets, those responsible for policy, those with a knowledge
of information systems, and staff from gender units. At the beginning of the
workshop, all participants should introduce themselves and state their
knowledge of, and exposure to, budget issues and gender issues, and their
expectations of the workshop. These introductions should be followed by an
explanation by the facilitators of what is planned for the workshop, what will be
expected of participants after the workshop and to what extent participants’
expectations might be fulfilled.
The material in this document includes some theory, some examples (called
‘comments’) on what others have written on particular topics, as well as
discussion questions. Some of these questions refer to ‘this country’ to
encourage participants to explore the issues in relation to their own countries.
Unless the workshop is small, these exercises will be best performed in groups.
The exercises in the last sections of the document will work best if participants
are divided into teams according to the different portfolios. This will allow
participants to share knowledge, identify problems and provide the basis for the
further development of gender-sensitive budgets in their portfolio areas.
The following resources are needed in preparing for a workshop:
the Country’s budget papers;
the Country’s latest report on the Convention on the Elimination of All
Forms of Discrimination Against Women (CEDAW) and statistical reports
such as Women and Men publications; and
any other relevant policies and legislation on gender.
During the workshop participants will identify further resources, such as

statistics and other information relevant to their portfolio, necessary for a
successful gender-sensitive budget analysis.
4 Section 1: Introduction
Section 2
What is a gender-sensitive budget?
Different terms are used to describe gender-sensitive budgets. These are

explained in Comment 1.
c omme nt What is a gender-sensitive budget?

1
‘Gender-sensitive budgets’, ‘gender budgets’, ‘women’s budgets’ and ‘women’s budget
statements’ refer to a variety of processes and tools aimed at facilitating an assessment
of the gendered impacts of government budgets. In the evolution of these exercises, the
focus has been on auditing government budgets for their impact on women and girls.
This has meant that, to date, the term ‘women’s budget’ has gained widest use.
Recently, however, these budget exercises have begun using gender as a category of
analysis so the terminology ‘gender-sensitive budgets’ is increasingly being adopted.
It is important to recognise that ’women’s budgets’ or ‘gender-sensitive budgets’ are not
separate budgets for women, or for men. They are attempts to break down, or
disaggregate, the government’s mainstream budget according to its impact on women
and men, and different groups of women and men, with cognizance being given to the
society’s underpinning gender relations.
Rhonda Sharp (1999 forthcoming), ‘Women’s budgets’, in Meg Lewis and Janice Peterson (editors),
Dictionary of Feminist Economics, New York: Edward Elgar.
Gender-neutral or gender blind?

Few government revenue raising activities or expenditures are designed on the
basis of gender. Indeed, many countries have been replacing measures that were
regarded as discriminatory with gender-neutral approaches (Himmelweit
1998a: 6). For example, mother’s benefits have been replaced with child
allowances paid to the person who is primarily responsible for the care of the
child. Taxation rebates for dependant wives (sometimes referred to as ‘married
men’s allowances’) are being phased out or paid to a dependant spouse of either
sex.
However, the goal of removing discrimination or achieving ‘gender neutrality’
should not be confused with the fact that budgetary policies can have
significantly different impacts on women and men and on different groups of
women and men. Sometimes these differences can be justified to achieve certain
policy goals and sometimes it can undermine social and economic objectives.
However, to ignore the gendered impact of policy does not constitute gender
neutrality; rather, it describes ‘gender blindness’.
Good policy requires understanding both the impact of policy and how it might
be better designed to achieve outcomes which meet the needs of women and men
and girls and boys as well as different groups of women, men and children.
Section 2: What is a gender-sensitive budget? 5

Below are some observations of the link between government budgets and
gender.
c omme nt Gender blindness

2
A budget, on the face of it, appears to be a gender-neutral policy instrument. It is set
out in terms of financial aggregates - totals, and sub-totals of expenditure and revenue,
and the resulting budget surplus or deficit. As usually presented, there is no particular
mention of women, but no particular mention of men either. However, this appearance
of gender-neutrality is more accurately described as gender-blindness. The way in
which the national budget is usually formulated ignores the different, socially
determined roles, responsibilities and capabilities of men and women. These
differences are generally structured in such a way as to leave women in an unequal
position in relation to the men in their community, with less economic, social and
political power.
Source: Diane Elson (1997a), ‘Gender-neutral, gender-blind, or gender-sensitive budgets? Changing

the conceptual framework to include women’s empowerment and the economy of care’, Preparatory
Country Mission to Integrate Gender into National Budgetary Policies and Procedures, London:
Commonwealth Secretariat, p 1.
c omme nt Government budgets and gender

3
The budget reflects the values of a country - who it values, whose work it values and
who it rewards … and who and what and whose work it doesn’t. Past [South African]
budgets are clear reflections of the priorities of apartheid, capitalist and patriarchal
South Africa. The budget is the most important economic policy instrument of
government, and as such it can be a powerful tool in transforming our country to meet
the needs of the poorest.
Government budgets and policies are often assumed to affect everyone more or less
equally: to serve the ‘public interest’ and the needs of the ‘general person’. Until now
the average citizen targeted has been white, male, Afrikaans and middle class. Yet in
South Africa the average citizen is actually black, poor and a woman.
Disaggregated data is needed to demystify the apparent neutrality and, more
specifically, the gender neutrality of the budget. It will expose how tariffs, industrial
relations, taxation, education, employment or industrial policy impact on women due to
their different location in the family and in the economy.
Who gets the jobs and what is the nature of the jobs that are created? Who gets the
subsidies? Who gets the housing and what is the nature of the homes and communities
which are being developed? What are the traditional policy assumptions in the budgets,
for example, are women dependent breadwinners? The fact of the matter is that the
same rules and procedures can often reinforce existing inequalities and work against the
interests of women.
Source: Pregs Govender (1996), ‘Foreword’, in Debbie Budlender (editor), The Women’s Budget,
Cape Town: Institute for Democracy in South Africa, p 7.
6 Section 2: What is a gender-sensitive budget?

As the above quotes make clear, a women’s budget or a gender-sensitive budget

is not a separate budget for women. Rather, it is an analysis of the government
budget to establish the budget’s differential impact on women and men and on
different groups of women and men. A gender-sensitive budget is about taking
government’s gender commitments and translating them into budgetary
commitments.
Women’s budgets are fundamentally about mainstreaming gender issues - about
ensuring that gender issues are integrated into all national policies, plans and
programs rather than regarding women as a special ‘interest group’. The
objective of a gender-sensitive budget is to inform debate about policy and the
appropriate allocation of public expenditure and taxation.
Several countries have already produced gender-sensitive budgets and others are
in the process of doing so. In each country the exercise differs according to the
local situation. Some of the differences between the models can be categorised
according to the following criteria:
Location
Who initiates the exercise? Who does the exercise? Participants in these
exercises can include government, parliamentarians, non-governmental
organisations (NGOs) and others in civil society. If the exercise is conducted
within government, there are further questions as to who actually does the day to
day work and how the chain of responsibility will function. In practice, this may
involve the women’s ministry or other government gender machinery, the
finance department, or other staff within individual portfolios and contracted
consultants.
Scope
What is covered in the exercise? Will it be all levels of government (national,
state/provincial and local)? Will it cover both expenditure and revenue? In the
case of expenditure, will it cover all or only selected portfolios? If it covers
selected portfolios, will it be only the social sectors, or will it include a balance
of social and economic portfolios? If an analysis of revenue is part of the
exercise, will it cover only taxation or all sources of revenue? Which figures will
be examined - past budget, past actual expenditure, estimated current budget
allocations, future budgets as projected in medium-term expenditure
frameworks?
Reporting format
Will it be published as a separate budget document or integrated in existing
public documents? The choice depends to some extent on the existing budget
documents in a country. Australia and the United Kingdom, for example,
produce a series of budget-related papers which are tabled on the day of the
Budget.

South Africa has three documents - the White Book containing all the estimates
of expenditure, the Minister’s Budget Speech and the Budget Review which
contains discussion of the economy and the different portfolios. Mozambique
has a single document which contains both the estimates and the Minister’s
Budget Speech.
Politics
Who will be involved in the process at the different stages? Who will use the
products? Who will fund the exercise? Who has the power to enforce
accountablility for gender equity? Who will resist? In what ways will gender
become part of the public debate on budgetary policy?
Mainstreaming gender
The institutions of government are not always well endowed with a capacity to
integrate gender into budgetary policies. The Ministry of Finance/Treasury
plays a central role in the implementation of a gender-sensitive budget analysis
but institutional capacity may need to be systematically built. Some of these
challenges are discussed in the extract below.
c omme nt Mainstreaming gender in Finance ministries

4
Mainstreaming gender in Finance ministries is a major challenge… The average
Finance ministry views gender as peripheral to its own role and work. Changing this
mindset will require both political will to back up the mainstreaming effort, as well as
carefully crafted methods for drawing out the enthusiasm for cooperation of Finance
ministry officials and staff…
[There are] five main barriers to mainstreaming gender in Finance ministries:
insufficient analytic clarity regarding the Finance ministry’s work and, in particular,
its changing role during the recent period of globalization and liberalization of
economies;
absence of a clear understanding of how gender is linked to that role;
non-conducive institutional structures and ethos within which Finance ministries
function;
weak understanding of the attitudes prevalent among those who work within
Finance ministries, and how these have been changing over time; and
insufficient knowledge and capacity among women’s organizations to engage
effectively in macroeconomic policy debate.
Finance ministries in the past typically functioned with relatively short time horizons,
and played an essentially supportive role vis a vis Planning ministries or Planning
Commissions whose task was to set the strategic direction for the rate, structure and
pattern of economic growth, the distribution of income and employment, the creation of
physical infrastructure, human development, and poverty alleviation.

The Planning ministry in the past functioned as the apex body that coordinated and
balanced the strategic plans of the other ministries and took the lead.
These roles and the division of labour between Finance and Planning ministries have
changed during the current period of structural adjustment oriented economic
reforms… In a time of sharply increased fiscal stringency, the role of Finance ministries
has changed from a supportive one to a disciplining one. The changed role of the
Finance ministry vis a vis other ministries and especially relative to the Planning
ministry reflects a major change that has occurred worldwide in the meaning and
parameters of macroeconomic management.
In addition, maintaining fiscal and monetary discipline has become part of the larger
structural reform process in which Finance ministries have come to have a key role.
Fiscal stabilization and structural reforms directed at liberalizing and privatizing the
economy set the parameters today for any attempt to mainstream gender.
Source: Gita Sen (1997), Mainstreaming gender into Commonwealth government ministries and
related agencies responsible for finance, Bangalore: Indian Institute of Management Report prepared
for the Gender and Youth Affairs Division, Commonwealth Secretariat, pp 3-5.
a Discussion questions
List and discuss where you agree and disagree with the views of the above
author.
To what extent do the issues she raises apply in this country?
Country examples: Australia and South Africa

Australia
Australia was the first country to introduce a gender-sensitive budget analysis.
An assessment of the budget for its impact on women and girls was undertaken
by the Federal Government for 12 years between 1984-1996. Women’s budgets
were introduced in South Australia (1985), Victoria (1986), the Australian
Capital Territory (1989), Queensland (1991), Tasmania (1992) and the
Northern Territory (1993). New South Wales conducted a budget exercise
focussing only on new expenditure initiatives for women and girls in the early
1980s. It introduced a comprehensive women’s budget in 1991.
The form that women’s budgets took in Australia established a model whereby
each government agency was required to provide an audit of the annual
government budget of the government’s achievements in relation to women and
girls. The exercise was a comprehensive one with respect to government
expenditures. It was strongly emphasised to departments that all the agency’s
programs and expenditures were relevant, not just those expenditures directly
allocated to women and girls.

The South Australian women’s budget, introduced in 1985, was the first to
make this criterion explicit (Sharp & Broomhill 1998). Agencies were asked to
report on their global budget according to whether these expenditures and
programs were:
1 ‘specifically targeted to women and girls expenditures’ (for example,
Aboriginal women’s health initiatives and programs to increase young
women’s access to non-traditional job training);
2 ‘equal employment opportunity expenditures’ directly undertaken by the
agency on behalf of its women workers (for example, mentoring programs
for women public servants and rewriting base grade clerical job descriptions
for gender bias); or
3 ‘general or mainstream expenditures’ (for example, identifying the users of
legal aid and who accesses the export market industry assistance).
The total expenditure framework enabled a quantitative assessment of the

proportion of government expenditures which are targeted to women and girls
relative to non-targeted, indirect or general expenditures. An analysis of one of
the early Australian state budgets showed that the category ‘general or
mainstream’ or indirect expenditures averaged 99 per cent or more of the total
budgets of the 26 participating agencies (Sharp & Broomhill 1990: 3).
Specifically targeted expenditures to women and girls in the community and
equal opportunity expenditure within the public service, while strategically
important, were small, being less than 1% of the total budget of the selected
government agencies.
Another important feature of the Australian women’s budget model has been the
critical role the central women’s policy offices have played in coordinating and
driving the women’s budget exercises. This has been crucial in shaping the
politics of the Australian women’s budget model. The women’s policy offices
have worked closely with Treasury Departments because, in many cases, the
report was published as one of the government’s budget papers. The key role
played by women’s policy machinery within government led to Australian
women’s budgets being described as ‘an example par excellence’ of a
bureaucratic-led strategy rather than a community-based strategy (Summers
1986: 66). As a consequence, the published results of the women’s budget
exercises were presented to women in the community largely as a
communication exercise as to what the government had achieved.
The Australian women’s budget exercises have undergone various changes
since the first Federal Government women’s budget in 1984. While all States
and Territories had implemented women’s budgets in the 1980s and 1990s, by
the end of 1998 only three States and Territories continued to used women’s
budgets as an audit exercise in government.
A common theme with these exercises is to report on the implementation of
various policies which are important in achieving the government’s goals in
relation to women and girls. The following extract from the Australian Federal
Government’s Women’s Budget Statement 1995-6 illustrates the range of issues
covered by a gender budget.

c omme nt Australia: Introduction to 1995-96 Women’s

Budget Statement
5
Economic growth strengthened during 1994-5 and is expected to continue at strong and
sustainable levels in 1995-96 with low inflation. Over 544,000 new jobs have been
created since April 1993, surpassing the Government’s target of 500,000 new jobs. Just
under half of those jobs have gone to women.
The 1995-96 Budget includes measures to boost national savings and promote strong
investment growth, while limiting additions to foreign debt. The Budget strategy is designed
to maintain the economy on a sustainable growth path, while enabling the Government to
pursue its objectives of raising living standards and advancing a more equitable society.
In keeping with its National Agenda for Women, the Government continues to place a high
priority on the particular needs and concerns of women. This is especially significant in the
lead-up to the United Nations Fourth World Conference on Women in Beijing in
September 1995. Australia is taking a lead role to ensure that this Conference is a
‘Conference of Commitments’ by governments, including the Australian Government, to
achieve practical and focussed outcomes to advance the status of women.
This Budget includes a range of practical measures to advance the status of women.
It features reforms in assistance to families, in particular the introduction of a new
Maternity Allowance.
It includes initiatives to improve access to child care and provide greater practical
recognition of the contribution of carers, mostly women, of older or sick people with
disabilities.
Measures highlighted in the Budget are also designed to better target the Government’s
spending on health, community services and public housing programs: all important issues
for Australian women.
Most significantly, this Budget lays the groundwork for a new strategic approach to
addressing inequalities between indigenous and non-indigenous Australians. It provides
substantially increased funding for land acquisition, health care and infrastructure for
Aboriginal and Torres Strait Islander people, and measures to improve the effectiveness of
these programs.
The Budget sets out action to address the unacceptable health outcomes of indigenous
people and provide them with a better access to mainstream health and community
services. Community consultation and participation by women will be key factors in
developing and implementing these reforms.
The Budget also includes provision for major improvements in access to justice, a key issue
for Australian women...
Source: Department of Treasury, Commonwealth of Australia (1995), Women’s Budget Statement

1995-96, Budget Related Paper No 3, Canberra: Australian Government Publishing Service, p 1.
Commonwealth of Australia copyright reproduced by permission.

South Africa
South Africa has two different types of gender-sensitive budget analyses. The
Women’s Budget Initiative started in mid-1995 as a joint effort of
parliamentarians and non-government organisations. The Initiative focused on
the national and provincial budgets. Within the first three years the budgets of
all portfolios were analysed, as were taxation and public sector employment. It
drew on a wide range of researchers and advisors situated both in women-only
and in general organisations, universities, and government itself. South Africa
has three spheres of government – national, provincial and local. During 1998/9
the Initiative will focus on local government budgets. The parliamentary-NGO
initiative has provided information which has been used by parliamentarians and
civil society groupings for lobbying and advocacy.
In South Africa, the first three women’s budgets contained fairly detailed
academic analysis accessible only to a limited proportion of the population. In
1998, the parliamentary-NGO group published Money Matters: Women and
the Government Budget, a simplified version which looks at eight portfolios and
which is aimed at second language English speakers with at least ten years of
education. The group also plans to develop workshop material to be used to
reach an ever widening audience.
During 1997, the South African government agreed to initiate gender-sensitive
budget analysis from within the Department of Finance, as a pilot for the
Commonwealth initiative to engender macro-economic policy. The product of
the first year of the pilot was the inclusion of discussion of gender issues in
particular portfolios in the Budget Review, tabled on Budget Day in March
1998 with the estimates of expenditure. The pilot will be extended to other
portfolios during 1998, as well as to provincial level.
The parliamentary-NGO initiative has been funded by international donors. The
government’s initiative has received support from the Commonwealth
Secretariat.
Initiatives in other countries

Over the last few years many countries have expressed interest in initiating their
own gender-sensitive budgets. In some cases the interest has come from
government, in others from civil society. In several instances donors have
played a key role in exposing people to the idea and promoting it. Some of the
countries which have started work on gender-sensitive budgets are examined
below and on the following pages.
Barbados
Barbados is investigating the possibility of being the third country to join the
Commonwealth pilot project.

Canada
Members of the Centre for Policy Alternatives, a social justice research group,
are developing a gender-sensitive budget as part of the Centre’s annual
Alternative Federal Budget.
Mozambique
As shown in Comment 6 below, the government has clearly stated its intention
to have a Women’s Budget (Orcamento Mulher) in its orientation document for
the elaboration of the economic and social plan for 1999, the State Budget for
1999 and the triennial plan of public investment for 1999-2000. During 1998
the Ministry of Planning and Finance initiated a gender-sensitive budget
exercise, with financial assistance from the Swiss Development Corporation.
In the first year departmental staff, together with staff from the relevant
portfolios, will examine allocations for labour, health, education and agriculture
as well as public sector employment.
c omme nt Mozambique: A women’s budget

6
Another point to consider in the elaboration of the 1999 State Budget is the inclusion of
the gender dimension, realised in the ‘Women’s Budget’. By identifying staff
composition and salaries expenses per sex, and the level of public expenses on
women, it will be possible to verify the extent to which agreements of the 1994 Beijing
Conference are being affirmed.
Source: Ministerio do Plano e Financas, Direccao Nacional do Plano e Orcamento (1998),
‘Orientacoes para a elaboracao do: Plano Economico e Social 1999, Orcamento do Estado 1999,
Plano Trienal de Investimento Publico 1999-2001’, Maputo, May, p 2.
Namibia
Staff of the Ministry of Finance worked with consultants from the Swedish
International Development Agency (SIDA) to produce an analysis of selected
portfolios for the 1998 National Budget.
Sri Lanka
This country joined the Commonwealth countries pilot project on gender and
macroeconomic policy in 1997. Its initial gender-sensitive budget exercise was
coordinated by the Department of National Planning. It focussed on the
portfolios of health, education, public sector employment, agriculture, industry
and social services.
Tanzania
The NGO, Tanzania Gender Networking Program, is leading a coalition of
NGOs. In the first year analysis focused on health and education, two key
sectors deemed critical by the poor women and men who are the groups’
primary constituents.

In Tanzania the Ministry of Finance and the Planning Commission are also
undertaking a parallel gender-sensitive budget exercise, with assistance from the
Swedish International Development Agency (SIDA).
c omme nt The Tanzanian Gender Budget Initiative

7
The main objectives of the [Tanzanian] Gender Budget Initiative are:
to strengthen lobbying and advocacy skills for women and gender oriented lobby
groups
to effectively campaign for women’s rights and gender equality/equity especially in
relation to more resource allocation and women’s participation in policy-making
and public resource management structures
to examine the budgeting process and budgeting allocation in selected strategic
ministries starting with that of Health and Education for 1997-1998 and how they
impact on women and other disadvantaged groups in the society, later going to
other sectors of concern
to organize and carry out a lobbying campaign to influence technocrats and
legislatures to increase budget allocations to sectors which are sensitive to women
and other disadvantaged people
to provide popular information and an advocacy campaign to influence the public
to be informed about the budgetary process and allocations and its consequences
on community development highlighting the distribution of national resources by
gender, but also stressing the actual utilization [using] a gender perspective.
Source: Gemma Akilimali (1997), ‘Gender Budget Initiative: Tanzania’s Experience’ Tanzania Gender
Networking Project, Brief presented at a workshop on Mainstreaming a Gender Equality Perspective
into National Budgets, Stockholm, p 5.
Uganda
The strong parliamentary Women’s Caucus, and the affiliated NGO Forum for
Women in Democracy, have taken the lead in initiating a gender budget
exercise. In the first year the analysis will focus on selected portfolios.
United Kingdom
The Women’s Budget Group, an NGO consisting of researchers and activists, is
seeking to put a gender-sensitive budget analysis on the public policy agenda.
At the time of the 1998 Budget, the British Treasury produced an issues paper
which looked at the effect on women of the new deal for working families.

c omme nt Recommendations to the UK Government

8
The [Women’s Budget] Group’s core recommendations to the UK Government are:
There should be a comparative ‘gender impact statement’ published with every Budget
and Public Expenditure White Paper to:
display the differential impact and consequences of policies on women and men,
ensure that women are treated as individuals and not as dependants or subsumed
into the household,
cause the distribution between the sexes of financial burdens and benefits to be
more equitable in future,
begin to show the different forms of unpaid and unvalued work that require
amendments to Treasury models.
Source: Sarah Robinson (editor), The Purse or the Wallet? Proceedings of a seminar of The
Women’s Budget Group held on February 12, 1998, inside front cover.
It is not only governments which can disaggregate their spending. The World
Bank (see Comment 9) has also attempted to estimate the proportion of its
lending which includes special gender provisions.
c omme nt A snapshot of gender-related World Bank lending

9
It is difficult to estimate the share of projects that benefit women. All [World Bank]
projects are expected to benefit women, at least indirectly. Some projects have more
identifiable gender-related components than can be measured and tallied. Thus,
between 1988 and 1994, about 30 percent of projects approved by the World Bank
included specific gender-related actions, up from less than 10 percent during 1967-85.
Bank projects have tended to make the strongest efforts to improve women’s health
and education and increase their options in agriculture. [Projects have made the
weakest efforts in financial and ‘other’ areas].
100
gender related
80
PHN* = population,
health & nutrition
60
40
20
Figure 2.1 - Percentage of

0 projects with gender-related
agriculture
water & sanitation

PHN*
other
urban development
financial
education
actions, 1988-94
Source: The World Bank (1995),
Advancing Gender Equality: From
Concept to Action, Washington, p 3.

As can be seen, in some countries, governments are centrally involved in these

exercises, in others it is civil society, and in yet others it is both. While these
exercises aim to inform debate about the appropriate allocation of public
expenditure and taxation, government and non-government exercises usually
differ. In particular, government undertakings usually take the form of more
comprehensive audits intended to provide information, while civil society
products involve more critical discussion.
The extracts in Comment 10 illustrate two different stories when the exercises
are conducted inside and outside government.
c omme nt Privatisation vs de-institutionalisation

10
Story 1: Outside government
In the current neo-conservative international climate calls are heard for privatisation
and for ‘community care’, or more involvement by ‘the community’ in caring for
vulnerable groups. In fact privatisation in the social services is usually introduced as a
policy of ‘community care’.
The possibility for privatisation, and its costs and effects, will differ. Aspects of health
care can be privatised in a way in which aspects of welfare cannot. To put it bluntly, a
woman who collapses in the street with acute appendicitis will receive attention from
public or private health services, no matter what the standard of care. A woman who is
emotionally and physically battered by her partner cannot secure care from the state,
nor will the market provide care for profit.
An analysis of community care politics can be a useful demonstration of the absolute
interdependence of economic and social policy. ‘The community’ is not an agency
which can care: in fact a double equation is at work which assumed that community
care is equal to care by families which is equal to unpaid care by mostly women.
It rests on the assumption that there is an endless supply of women’s unpaid work, and
that this is not an economic variable. It shifts the costs of caring from the formal to the
informal sector. It is much more difficult to ‘do the accounting’ when this happens.
Source: Francie Lund (1996), ‘Welfare’, in Debbie Budlender (editor) The Women’s Budget, Cape
Town: Institute for Democracy in South Africa, pp 114-5.
Story 2: Inside government

The Department of Welfare has embarked on a process of de-institutionalising its
welfare services where it is deemed appropriate. The rationale underlying this policy
shift is that communities themselves are often able to provide more appropriate social
services than institutions. In these instances community care is also usually a more
cost efficient alternative to institutionalisation.
Source: Department of Finance, Republic of South Africa (1998), Budget Review 1998, p 6.60.
a Discussion question
What are the relative strengths and weaknesses of a government, compared to a
‘civil’ society or community-based, gender-sensitive budget analysis?

Section 3
Gender, sex, men and women
Government policies and resource allocations impact on us as people, whether

as individuals or as members of groups. There are characteristics which
differentiate us as individuals and which assign us to broad groups in society.
Sex and gender are two of such characteristics. Race, class, sexuality, age, rural
and urban location are others.
Sex and gender

Sex and gender are often assumed to mean the same thing. However, as
Comment 11 (below) shows, they have quite different meanings and these
differences are important for policy.
c omme nt Sex and gender

11
Biological Differences (Sex) Social Differences (Gender)
These are difficult to change because These are able to be changed because
we are born female or male. our gender identity is determined by
society.
Throughout history and across cultures In different societies and at different

sex differences exist. times in history gender roles have been
different.
Policies respond to sex differences in Policies can either respond to gender

any area to do with the physical body stereotypes and traditional gender roles
(eg childbearing and prostate disease). (eg paying benefits on the assumption
that only men are heads of households)
or attempt to change them (eg taking
into account the barriers to women and
girls participating in non-traditional
training courses).
Even though gender-sensitive budgets are often referred to as ‘women’s

budgets’ they are fundamentally, as suggested in Comment 12 on the next page,
about gender relations.
Section 3: Gender, sex, men and women 17

c omme nt Gendered social relations

12
We understand that women’s oppression arises out of a system of gendered social
relations which shape women’s (and men’s) position within society. We know that men
are sometimes also discriminated against or disadvantaged by those self-same social
relations, and we know also that gender oppression will not be successfully tackled
unless we address ourselves to both women and men. Nevertheless, we also want to
stress that it is overwhelmingly women who suffer most as a result of asymmetrical
gender relations.
Source: Debbie Budlender (1996), ‘Profile of South African Women’, in Debbie Budlender (editor),
The Women’s Budget, Cape Town: Institute for Democracy in South Africa, p 25.
The focus on gender relations is also occurring in policy and development areas.
It is being recognised increasingly, as shown in Comment 13 below, that the
status of both men and women needs to be considered to address the needs of
women and to identify the needs of particular groups of men.
c omme nt Women and men

13
Where previous efforts have sought to address women’s rights and needs in special
and separate development programmes, the gender and development approach seeks
to integrate women’s needs into the wider picture, calling for the different life courses of
men and women to be considered at an early stage and emphasising the need to
monitor the different impact of policies and programmes on women and men, girls and
boys. This shift in focus from women to gender recognises that the status of women
cannot be addressed as a separate issue: it can only be addressed by considering the
status of both sexes.
Source: Commonwealth Secretariat (1995), The 1995 Commonwealth Plan of Action on Gender and
Development: A Commonwealth Vision, Women’s and Youth Affairs Division, London, p 6.
When you were growing up, what were the incidents that made you aware that
you were a man/woman?
Do you believe these were due to sex or gender differences?
How have gender relations changed since you were a child, between your
grandparents’ time, parents’ time, and your time? Who has gained? Who has
lost? Is this the same for other groups of people in this country (eg based on
class, ethnicity, race, sexuality, age, urban/rural location)?
18 Section 3: Gender, sex, men and women

Gender issues in government budgets

Gender issues pervade the lives of women and men and have economic as well
as social implications. Gender shapes our ‘opportunities, access to resources
and needs. There is growing awareness around the world that policies and
measures affect women and men in different ways and that policies and plans
need to be designed accordingly’ (Hedman, Perucci and Sundstrom 1996: 13).
An example of how gender issues have been considered in one government
budget is provided in Comment 14.
c omme n t Basic water supply

14
In late 1995 only a third of African households and less than three-quarters of coloured
households had running tap water inside their dwellings, compared to 97 per cent of
Indian and white households. Outside urban areas only 12 per cent of African
households had an inside tap and 21 per cent had a tap on the site, while 28 per cent
were collecting water from a river, stream, dam or well and 16 per cent from a
borehole. One in six African households who had to fetch water were forced to travel at
least a kilometre to do so. In 1993 the average time spent on the task by members of
rural African households forced to collect water was 1 hour and 40 minutes. The
average time spent by individual women who collected water was over an hour, at 74
minutes. Just under half of rural African women over the age of 18 were spending part
of their time collecting water.
A programme to supply water is one of the Presidential Lead Projects of the
Reconstruction and Development Programme. One thousand days after the 1994
elections more than a million people had benefited from the ready availability of fresh,
safe water for the first time in their lives. By the end of October 1997 the programme
had provided basic water supply and sanitation to approximately 1.2 million people and
spent approximately R800 million on the more than 1,000 projects initiated since 1994.
Of the 195,878 jobs created by the twelve Presidential Lead Projects initiated in 1994/5,
25,750 (13 per cent) were reported to have been taken up by women and 12,516 (6 per
cent) by youths.
In late 1997, on the basis of available information, the estimated involvement of

women in various roles was as follows:
Role per cent women
Employees on schemes 14
Trainees on schemes 16
Contractors 0
Consultants 25
Steering committee members 20
continued over page

The Department has initiated the Working for Water Programme as part of its national
water conservation campaign. The programme provides for the employment of local
people in the clearing of invasive alien plants. By doing so it both enhances the
available water supply and provides much-needed employment.
In May 1997, 7,400 previously unemployed people, of whom more than half were
women, were working on the Working for Water Programme. Meanwhile the Forestry
division of the Department has encouraged the planting of trees in community
woodlots. Many of these are managed by local women’s groups.
Source: Department of Finance, Republic of South Africa (1998), Budget Review 1998, p 6.58.
What are the gender issues in this example?
How do they relate to the government budget?
Are there similar issues in this country?
More broadly, why should governments be concerned about gender issues? One
reason relates to equity. Many governments have a commitment to promote
equality among citizens, to reduce systemic disadvantage, and to redistribute
resources from the haves to the have-nots.
A second reason relates to efficiency. As Diane Elson suggests, gender
inequalities impact negatively on the overall performance of government and
impose costs on society as a whole. She points out ‘there is a growing
realisation in the Commonwealth and in the wider international community that
gender inequalities, apart from being unfair, are also costly - not only to women
but also to men and children. The costs can be measured in lower economic
efficiency, lower output, lower development of people’s capacities, and lower
well-being’ (Elson 1997b: 12). Examples of these costs are provided in
Comment 15.

c omme nt The efficiency benefits of reducing gender inequality

15
Research on agricultural productivity in Africa shows that reducing gender inequality
could significantly increase agricultural yields.
For instance, giving women farmers in Kenya the same level of agricultural inputs
and education as men farmers could increase yields obtained by women farmers
by more than 20 per cent…
Research on economic growth and education shows that failing to invest in education
lowers gross national product (GNP).
Everything else being equal, countries in which the ratio of female-to-male
enrolment in primary or secondary education is less than 0.75 can expect levels of
GNP that are roughly 25 per cent lower than countries in which there is less gender
disparity in education…
Research on gender inequality in the labour market shows that eliminating gender
discrimination in occupation and pay could increase not only women’s income, but
also national income.
For instance, if gender inequality in the labour market in Latin America were to be
eliminated, not only could women’s wages rise by about 50 per cent, but national
output could rise by 5 per cent…
Gender inequality also reduces the productivity of the next generation - the World
Bank reports mounting evidence that increases in women’s well-being yield
productivity gains in the future.
The probability of children being enrolled in school increases with their mother’s
educational level and extra income going to mothers has more positive impact on
household investments in nutrition, health and education of children than extra
income going to fathers…
Research shows that gender inequality hampers a positive supply response to
structural adjustment measures by:
reducing women’s incentives to produce tradable goods and increasing women’s
time budgets…
Women’s time burdens are an important constraint on growth and development -
women are an over-utilised, not an under-utilised resource. The benefits of reducing
this gender-based constraint can be considerable.
For instance, a study in Tanzania shows that reducing such constraints in a
community of smallholder coffee and banana growers increases household cash
incomes by 10 per cent, labour productivity by 15 per cent, and capital productivity
by 44 per cent…
Source: Diane Elson (1997c), ‘Integrating gender issues into national budgetary policies and
procedures within the context of economic reform: Some policy options’, Preparatory Country Mission
to Integrate Gender into National Budgetary Policies and Procedures, London: Commonwealth
Secretariat, pp 1-2.

Women and men in different roles

The examples above suggest that women and men tend to undertake different
roles in the economy and in society. In other words, a gendered division of
labour characterises many cultures. This division of labour is evident in the
multiple tasks and roles undertaken by men and women. Comment 16
categorises the main work roles of people in all societies as being reproductive,
productive and community. The reproductive role ‘encompasses tasks such as
rearing and bearing children and housework. Productive work covers the tasks
that society usually acknowledges as work and often sees as the sum total of
work. Community work encompasses the tasks women and men perform in
maintaining the life of the community outside their immediate households’
(Budlender 1996: 31).
Women and men make different contributions across and within these
categories. Furthermore, some work roles are valued more than others. The
areas valued less highly tend to be invisible to policy makers. For example,
women’s unpaid care labour in the reproductive sphere is not taken into account
because it is assumed to be always available. This needs to be taken into
account in the design of equitable and efficient policies and budgetary resource
allocations.
The Women’s Budget Initiative in South Africa listed the following examples of
roles relevant to policy-makers and those who draw up budgets:
c omme nt Women’s roles

16
Reproductive role Productive role Community role
service receiver breadwinner and political citizen
transfer receiver (child economic citizen service receiver
benefit, unemployment wage-earner member of school parent
insurance person requiring access committee
user of household to capital victim/survivor of violence
infrastructure direct tax payer occupant of (unequal)
consumer public service employee spaces (farm dwellers,
indirect tax payer worker on public works rural dwellers)
user of technology programme exerciser of control over
sexual being user of technology resources generated by
government expenditure
receiver of direct and
indirect incentives and
rebates
beneficiary of affirmative
access, restitution
farm worker
Source: Debbie Budlender (1996), ‘Introduction’, in Debbie Budlender (editor) The Women’s
Budget, Cape Town: Institute for Democracy in South Africa, p 31.

What are common reproductive roles of women in this country? And of men?
Are these paid or unpaid? What support does government provide to those who
perform the roles?
What are common productive roles of women? And of men? Are these in the
formal or informal sectors? Are they paid or unpaid? Are the actors self-
employed, employer, employee, family worker? What support does government
provide to those who perform the roles?
What are common community roles of women? And of men? Are they paid or
unpaid? What support does government provide to those who perform the roles?
Are these patterns changing? Does government policy support change and, if so,
to whose advantage are these changes? Are the boundaries between the different
roles clear-cut?
Note your response

Reproductive roles Productive roles
Community roles Changing patterns

Different roles and gender bias in budgetary resource

allocations
Differences in the roles of men and women, which are built into the economic
and social structures of a society, frequently lead to unintentional gender bias in
policies and programmes and their accompanying resource allocations. This can
lead to social injustices and to inefficiencies in the use of resources.
Traditionally, assessments of efficiency in resource allocation have focused on
minimising the cost of resource use in the paid, productive sector of the
economy. However, if government budgets were to take into account all
resources, in the paid and unpaid sectors, then a different calculation of cost
minimisation is required. The paid productive or market-oriented sector uses
prices to signal the cost of resource use. However, the activities of the
reproductive and community sectors, which need and use considerable resources
to function, do not readily lend themselves to the same type of financial
calculation.
Moreover, as outlined in Section 6 ‘Putting Care into the Economy’, the links
between the paid and unpaid sectors should not be overlooked in ensuring the
overall efficient use of resources. Reproductive and caring activities, such as
producing and maintaining the labour force and developing our human
capacities such as trust and citizenship are not only vital in building a society
but are important also for the functioning of the market-oriented or productive
sector. Government health, education and welfare expenditures are important in
supporting reproductive activities, as individuals and families are unable to do it
all themselves.
Government budgets and policies that do not account for transfers in resource
use between the market-oriented, paid productive sector, and the unpaid
reproductive sector cannot be guaranteed to be efficient. Instead, they are more
likely to transfer costs than minimise costs. They are in danger of being a false
economy. This can have feedback effects on other sectors in the economy.
Additionally, because women provide a disproportionate amount of resources in
the form of their unpaid labour in reproductive activities compared to men, such
budgetary actions are also gender-biased. This raises important social justice
questions about who bears the costs of budgetary changes.
Comment 17 contains a discussion of the restructuring of health services.

c omme nt Health service restructuring: increase in

efficiency or transfer of costs?
17
Many countries are restructuring health services to try to achieve greater efficiency.
Efficiency of health services is judged by indicators such as:
a reduction in costs per patient treated, and
the rate of return to capital invested.
Health services providers have incentives to:

reduce the provision of ancillary services (such as laundry) in hospitals, and
discharge patients from hospital more quickly.
Reductions in the provision of ancillary services transfer costs to the unpaid

reproductive economy, for example:
women doing laundry for hospitalised relatives and friends.
Reductions in the time patients spend in hospital transfers costs to the unpaid
productive economy when:
women look after convalescing relatives and friends.
Source: Diane Elson (1998), ‘Programme Assistance and Gender: A Logframe Synthesis’, Workshop
on Integrating Gender Issues in Programme Aid, Sector Investment Programmes, Market Reform and
Other Forms of Economic Policy Assistance, Workshop Document No 1, Paris, p 14.
What are the gender issues in the above example of the restucturing of health
services?
Are the main outcomes a transfer of costs or efficiency?
Gender indicators
In order to undertake a gender-sensitive budget exercise, data on how men and
women are positioned economically and socially is needed. For example, an
analysis of employment and training policies and expenditures might indicate
what percentage of men and women have undertaken different forms of training
and the rates of employment and unemployment between men and women as
well as between groups of men and women.
Indicators of living standards and participation in economic and political life
have been developed by the United Nations Development Programme (UNDP).
These indicators, while broad, can be a useful starting point. They are primarily
used for international comparative purposes. For policy purposes, more detailed
statistics relevant to particular portfolio areas are required.

Gender-sensitive budget exercises require a wide variety of data which is

usually obtained from a number of sources including officially published
statistics, research published by academics and unpublished data collected by
government agencies and community groups.
Three key United Nations Development Programme (UNDP) indicators are
provided in Comment 18. These are used to compare broadly levels of human
development in several countries in Table 1.
c omme nt Indicators
18
Three indicators were chosen to make an index for human development based on these
three objectives. They were:
life expectancy at birth, which was used as a measure for a long healthy life,
nutrition and freedom from disease;
adult literacy, which was used as an indicator of the formation of human
capabilities and skills; and
real GDP per capita adjusted for purchasing power, which was selected as a proxy
measure for access to control over resources.
The values for HDI range between 0 and 1 and a value of 1indicates the highest level of
well-being. Values closer to zero indicate more severe levels of deprivation.
The Gender Development Index (GDI)

The Gender Development Index (GDI) is a gender-sensitive adjustment to the HDI. It
uses the same variables as the HDI. The difference is that the GDI adjusts the average
achievement of each country in life expectancy, educational attainment and income in
accordance with the disparity in achievement between women and men. Like the HDI,
the values for GDI range between 0 and 1 with 1 indicating the highest attainable levels
of gender-adjusted well-being.
The Gender Empowerment Measure (GEM)

The GEM tries to measure the relative power of women and men in political and
economic life. The three variables which make up the index are:
women’s and men’s percentage share of administrative and managerial positions;
women’s and men’s percentage share of professional and technical jobs; and
women’s and men’s percentage share of parliamentary seats.
As with the other two measures, the value for the GEM varies between 0 and 1. Values
nearer 1 indicate higher levels of women’s empowerment while values nearer 0 indicate
lower levels.
A comparison of different countries’ ‘human development’ is provided in Table 1 using
the United Nations’ indicators.

Table 3.1 Comparison of HDI, GDI and GEM ranks for selected countries (1994)
Country HD1 GD1 GEM

value rank value rank value rank
/175 /146 /94
Canada 0.960 1 0.939 1 0.700 6

Norway 0.943 3 0.934 2 0.795 1
United States 0.942 4 0.928 5 0.671 7
Australia 0.931 13 0.917 9 0.659 11
United Kingdom 0.931 14 0.896 13 0.543 20
Barbados 0.907 25 0.885 17 0.602 14
Fiji 0.863 46 0.763 53 0.329 68
South Africa 0.716 90 0.681 71 0.531 22
Sri Lanka 0.711 91 0.694 70 0.307 70
Mozambique 0.281 166 0.262 139 0.430 43
Source: Compiled from United Nations Development Programme (1997), Human Development
Report 1997, New York: Oxford University Press, p 41. Copyright by the United Nations Development
Programme. Reprinted by permission of Oxford University Press, Inc.
Is the necessary data available to make these calculations for this country?
Does this country do better or worse on the gender indicators than on the HDI?
In which of the components of the gender measures does the country perform
relatively well, and in which is it lagging?
Are gender disaggregations at a more detailed level available for the country?
For example, how many women and men are:
in cabinet?
in top decision-making positions in the civil service?
on private and public boards and committees?
The GDI and GEM are at a very high level of abstraction and aggregation. In
some countries marked gender differences might exist in only one of the
components of the GDI (for example, income). The income component of the
GDI is a measure of women’s earnings and labour force participation relative to
men’s. However, changes in women’s earnings and labour force participation
ignore what might be happening to women’s unpaid work burdens. Also,
average female income may not be a reliable indicator of the resources actually
available to women and girls in households.
The HDI has been criticised also for a first-world bias. The use of GDP,
longevity and literacy levels in the base of the HDI can result in these measures
producing high rankings for developed countries even where there is significant
gender inequality in a country (Mohiuddin 1996).

What are the strengths and limitations of the HDI and GEM as indicators of
women’s and men’s relative economic and social positions in this country?
How can we measure gender and gender relations?

There is a growing variety of measures that can be used to indicate the socially
constructed relations that exist between the sexes. As a result, the choice of
statistics for analysing particular areas of policy and expenditure can be
facilitated by developing a series of categories of what it is that needs
measuring. For example, gender statistics may be needed to measure:
the underlying causes of the problem/s; and the
consequences and effects.
Using this approach the Swedish Statistics Agency has made the following
suggestions for the policy area of Economic Life.
c omme nt Categorising gender statistics

19
Underlying causes
Sex segregation in education
population 15 years and over by level of education and age
population with third level education by field of study and age
school enrolment at third level by field of study
population economically active by level of education and age
Unequal sharing of responsibilities within the family

time spent in paid and unpaid work by marital status
time spent in different activities of unpaid work by marital status, with and without
children
employed population by marital status and age
employed population by marital status, number and age of children
Women’s reproductive role

economically/not economically active population by marital status and number of
children
employed population leaving the labour force after birth of the first/second child
employed population on parental leave
continued over page

Consequences and effects

Different wages and salaries
wages/salaries by occupation
wages/salaries by industry
wages/salaries in public and private sector
Different career opportunities

employed population by career position in the workplace
employers by industry and size of enterprise
Different roles in decision-making

employed population by career position in the workplace
employed population by career position and public/private sector
members and elected officials in trade unions
high level officials in ministries
Source: Birgitta Hedman, Francesca Perucci and Pehr Sundstrom (1996), Engendering Statistics: A
tool for change., Stockholm: Statistics Sweden, pp 50-1.
Which of the indicators discussed in Comment 19 are relevant and useful for
policy-making purposes in this country?
Which are available?
What other indicators would you add?
What other categories of gender statistics might be needed?
Gender is not the only divide linked often with disadvantage. In most countries
there are also differences between people on the basis of class, race, location
and so on. Tables 2 and 3 below show the interplay of race and gender, and
location and gender in South Africa in terms of unemployment rates.
Table 3.2 Unemployment rates (%) in South Africa, 1995 - Gender and race
African Coloured Indian White

Women 47 28 20 8
Men 29 18 10 4
Total 37 22 13 6
Source: Compiled from Central Statistics, Republic of South Africa (1997), Statistics in
Brief 1996, Cape Town, p 10.4.

Table 3.3 Unemployment rates (%) in South Africa, 1995 - Gender and
location
Urban Rural
Women 33 46
Men 21 25
Source: Calculations on data from 1995 October household survey of Central

Statistical Service, Republic of South Africa.
What are the divides in this country?
How might these differences in unemployment rates relate to health status and
delivery?
Are there any programmes that already take this into account?
Are there any programmes that could address it?
Local context
Gender differences and issues vary across cultures, time and economies.
Therefore, a knowledge of the local context is crucial to a gender-sensitive
budget analysis. It is suggested that in a workshop context an analysis of the
local context be provided at this stage. If appropriate, the Women’s Ministry,
an academic or a local consultant can be brought in to present all or part of the
session. The session might cover the points listed below.
Past activities and commitments in the gender area. For example,
government and NGO participation in the international UN conferences,
ratification of the United Nations Convention on the Elimination of all
Forms of Discrimination Against Women (CEDAW), any relevant
constitutional provisions, the form of the national gender machinery (eg,
gender units, a women’s ministry, independent bodies, civil society
organisations).
A broad social and economic profile of the country in gender terms. For
example, data may be compiled in a Women and Men statistical booklet for
the country. This profile can be adapted if the intention is to focus on
specific portfolios in the gender-sensitive budget analysis.
Discussion as to whether the government adopts a mainstreaming approach
to gender and how this is actually done.
Clarification of who initiated the gender-sensitive budget exercise in the
country and who is funding it.
What are the advantages and disadvantages (pluses and minuses) of doing a
gender-sensitive budget analysis?

Section 4
What is a government budget?
The budget reflects the choices that government has to make, and is the tool it uses to
achieve its economic and development goals. The government has to balance a wide
range of legitimate demands with limited resources at its disposal. In the budget
government sets out what it is going to spend (expenditure) and the income it collects
through taxes (revenue), which it needs to finance expenditure.
Department of Finance, Republic of South Africa (1997), People’s Guide to the Budget, Pretoria, p 2.
The functions of the government budget

Government budgets have an essential role in the planning and control of the
economic activities of a nation. Gender-sensitive budget exercises recognise that
government budgets ‘command substantial resources and that the state is an
influential force through its budgets in shaping gender outcomes both directly
and indirectly’ (Sharp 1999, forthcoming).
Government budgets have many functions. Three central economic functions
are:
allocation of resources
This function relates to the provision of public goods and services by the
government. All the goods and services in a country are produced either by
government, by the formal and informal market sectors, or by the not-for-
profit community and unpaid household sectors. In allocating resources, the
government must decide both the relative size of public service provision, as
well as how available resources are divided among the various government
functions (e.g. Administration, health, defence), policies and programs.
distribution of income and wealth

This function refers to the use of budgetary policy to try to redress
inequalities in income and wealth distribution. Governments make decisions
about what constitutes a ‘fair’ distribution between different groups of
people.
stabilisation of the economy

Government budgets are used to promote a certain level of employment,
stability in prices, economic growth, environmental sustainability and
external balance. Stabilisation policy requires economic, political and social
judgements in determining, for example, which objective has priority at any
one time and what are acceptable levels of unemployment, debt, interest
rates and so on. Budgetary policy can encourage sustainable economic
growth through the planning potential of the budget.
Section 4: What is a government budget? 31

by Budlender, Sharp & Allen
In addition, government budgets are:

a means of ensuring that governments are accountable to Parliament for
their revenues and expenditures; and
a measure by which governments can maintain control over their finances.
There are four important points to note about these functions:

1 Each of the functions may have different implications for women and
men, particularly in the areas discussed below.
allocation of resources
Allocations to certain functions, policies and programmes will benefit
women more than men, and vice versa. Gender-sensitive budgets seek to
uncover the allocation consequences of budgets.
distribution of income and wealth
The government’s concern could be about inequalities between women and
men, as well as between and within households or other groupings.
Gender-sensitive budgets are premised on the assumption that a ‘fair’
distribution between women and men is an important goal.
stabilisation of the economy
A gender-sensitive budget analysis requires an understanding of the
macroeconomic constraints, assumptions and theories that underpin the
budget. The questioning of traditional assessments of efficiency by gender-
sensitive budget analyses means that problems in achieving
macroeconomic goals can be identified.
2 The three central economic functions can be shared between the different
levels of government. An examination of the budget/s of a single level of
government will give an incomplete picture of government provision or
lack of provision.
3 The objectives which are ‘ends’ in themselves and objectives which are
‘means’ to an end need to be distinguished. For example, external
balance is not a goal in itself, but rather it is important insofar that it
increases the welfare of citizens. Similarly with price stability, or even
with economic growth.
4 Budgets themselves are a means to achieve objectives. Therefore, the

first task in drawing up budgets is to determine objectives and the policies
that are likely to achieve these objectives. Budgets should follow policy,
rather than vice versa. If the policy is bad, then the budget cannot be a
good one.
32 Section 4: What is a government budget?

The budget functions and policy

Difficulties arise when the links between policy and budgets are broken. Some
of the problems which this may entail for implementation of coherent and
gender-sensitive policies and programs are discussed in Comment 20.
c omme nt The budget and policy

20
The division between policy and budget directorates reflects the use of the budget
primarily as a means for ensuring accountability, rather than as a tool for supporting
policy-making or monitoring. This emphasis has several other negative implications.
The concentration on accountability means that the budget process has aimed, first
and foremost, at delineating clear organisational and individual responsibility for
expenditures. The published budget seeks primarily to define which chief director must
manage which funds. The process has sought to define the procedures for using
resources rather than desired outputs or outcomes.
Establishing accountability by defining procedures rather than aims militates against
new programmes. Treasury rules seek to ensure that expenditures occur only when
budgets demonstrably reflect probable requirements and timetables. Yet in the nature
of things, the budgets for new programmes will be less well-defined and precise than
those for long-existing activities. Moreover, when they rely on untried delivery systems,
significant delays in implementation may occur. These factors mean that new
programmes run a higher risk of being ruled out or cut by the budgetary authorities.
The focus on accountability has also led to budgeting for a single year at a time.
Transforming government, however, requires planning phasing of changes, which an
annual budget cannot support…
Finally, defining accountability in terms of procedures rather than outcomes makes it
almost impossible to use the budget to monitor government activities. The published
budgets define expenditure, not in terms of specific aims, but in terms of the
responsibility of chief directors for programme budgets.
Each agency, then, must publish only a handful of programmes - generally less than
five… As the number of programmes varies relatively little between departments, the
largest spenders end up publishing the least detail… For instance, almost the entire
education budget goes to primary and secondary education, with no more precise
definition. The equalisation of current spending on all schools, meeting infrastructural
backlogs in school buildings in black communities and transforming the curriculum will
not lead to any major changes in the published budget.
Source: Julia de Bruyn and Neva Seidman Makgetla (1997), ‘Engendering the budget process’, in
Debbie Budlender (editor), The Second Women’s Budget, Cape Town: Institute for Democracy in
South Africa, p 71.

Budgetary constraints
Government budgets are not drawn up in a vacuum. All governments face some
constraint on the overall size of the budget. To some extent they are constrained
by the ability or willingness of taxpayers to contribute. In poorer countries,
because of the inability of taxpayers to contribute sufficiently, they are further
constrained by the willingness of donors and lenders to supplement local funds -
and by the acceptability of conditions imposed by the donors and lenders. Other
constraints include policy decisions to reduce budget deficits quickly or to
increase expenditures such as military funding. An understanding of the
budgetary constraints is important in assessing to what extent a gender-sensitive
budget analysis will be conducted in the context of a budget which can only
reallocate its resources, as opposed to increasing the overall levels of revenues
and expenditures.
What are the current and historical budgetary constraints in this country?
Variations between countries

All government budgets share some common characteristics. There are,
however, also many differences, even in the form in which they are presented.
There are also differences in the level of detail provided.
Some of the common variations between countries are whether:
there is a single budget for the whole country, or separate budgets for
different levels (national, provincial/state, local), the extent to which these
budgets are consolidated in terms of reporting, and the extent to which they
are dependent on each other (for example, where revenue raised at one level
is transferred to another);
outside funding is integrated into the government budget;
there is a separate development or investment (often multi-year) budget;
individual budgets are formulated in terms of a multi-year plan; and
the budget documents include previous budget figures and/or previous
actual expenditure.
The process of drawing up the budget

What is also important for anyone wanting to analyse government budgets is to
know the process by which they are drawn up. Again, this will differ from one
country to another. The timetable for the process provides clues as to the most
effective points for intervention. These points will not be the same for all
stakeholders.

For example, the budget timetable for the Australian Federal government in
Comment 21 shows that community groups have an opportunity to influence
decisions before the February meetings when the Finance minister sets the
budgetary framework.
In South Africa, the parliamentary Finance Committee as well as individual
portfolio committees hold public hearings after the tabling of the Budget, where
civil society stakeholders can put forward their views.
c omme nt Australian Commonwealth budget timetable 1994-1995

21
Nov/Dec 1993 Budget timetables and process considered by Cabinet and
Expenditure Review Committee (ERC)
Dec/Feb 1994 Commonwealth Portfolio Ministers and their departments
formulate summary submissions of new policies and savings
Jan 1994 Community and interest group submissions summarised and
circulated...
Feb 1994 Treasury gives the ERC and Cabinet a report on the economic
outlook and macro-economic policy implications.
Minister for Finance submits to the ERC and Cabinet the latest
forward estimates and proposals...
Each Portfolio Minister meets with the Treasurer and the Minister
for Finance for “trilateral discussions”...
A Fiscal Framework Paper is agreed to by Cabinet and discussed
by the Caucus...
Mar 1994 Ministers submit summaries of new policy proposals and any
offsetting savings. The ERC provides feedback on likely areas of
additional spending and priority...
Commonwealth convenes Premiers’ Conference for funding to
the States
Apr 1994 ERC deals with final submissions for Ministers and
Departments... The ERC Ministers meet with each of the Senate
parties for a final discussion of their proposals.
Revenue Committee reviews revenue collections in the last
financial year and revenue estimates for the next financial year
and new revenue measures.
May 1994 The 1994-95 Budget brought down.
Debate begins in House of Representatives.
Senate Estimates begin.
Key Revenue Bills introduced into the Senate
Jun 1994 Appropriation Bills introduced into the Senate.
Supplementary Estimates begin.
30 Jun 1994 Appropriations and Revenue Bills passed and assented to
Source: Glenn Withers, David Throsby and Kaye Johnston (1994), Public Expenditure in Australia,
Economic Planning and Advisory Commission Paper No 3, Canberra: Australian Government
Publishing Service, pp 43-4. Commonwealth of Australia copyright reproduced by permission.

a Discussion questions and calculations with budget figures

Table A Mozambique state budget for 1998
Expenses according to functional classification
Meticais m %
General public services 1300442.2 13
National defence 385301.4 4
Security and public order 579664.0 6
Education 1344460.4 14
Health 1356830.5 14
Security and social assistance 580102.4 6
Housing and community services 283226.7 3
Recreational, cultural & religious services 108874.4 1
Energy 426743.2 4
Agriculture, forestry, fishing etc 267476.9 3
Extractive industries 44568.1 0
Transport and communications 1465544.1 15
Other economic services 361818.7 4
Other expenses 930106.5 10
Provisao 212990.3 2
Total 9648149.8 100
Source: Compiled from Republic of Mozambique, Ministry of Planning and Finance (1998),
Orcamento do Estado para o ano economico de Maputo, p 51.
How does the allocation of this country’s budget compare, in percentage terms,
with that of Mozambique? How do you explain the differences? Can you say
anything about the budget allocations by sectors and what it might mean for
gender equality?
Is this country’s government budget increasing or decreasing? Is this happening
faster or slower than inflation?
What proportion of the budget goes on:
paying government debt?
paying personnel?
the social sectors?
the economic sectors?
the protection sectors?
the military?
What proportion of revenue comes from:
taxes?
international donors?
other sources?
What other issues are important in understanding the budget of this country?
What is the budget timetable of this country?

Section 5
Tools for a gender-sensitive

analysis of budgets
A gender-sensitive budget analysis requires the utilisation of a set of

procedures. Comment 22 identifies seven possible tools for such an analysis.
However, this list is not exhaustive. Our knowledge of possible tools is being
developed in the process of conducting a gender-sensitive budget analysis in
different countries. Some of the tools are used more intensively and frequently
than others.
For instance, to date, ‘gender-aware policy appraisal’ has been used much more
often than any of the other techniques. The choice of tools used in the analysis
also depends ultimately upon the availability of data, expertise of personnel,
time constraints and other resources such as access to computers.
c omme nt Tools for a gender-sensitive analysis of budgets

22
Diane Elson has suggested that different ‘tools’ can be utilised for a gender-sensitive
analysis of budgets. Her suggestions, with some adaptations, are:
1 gender-aware policy appraisal
This is an analytical approach which involves scrutinising the policies of different
portfolios and programmes by paying attention to the implicit and explicit gender
issues involved. It questions the assumption that policies are ‘gender-neutral’ in
their effects and asks instead: In what ways are the policies and their associated
resource allocations likely to reduce or increase gender inequalities?
2 gender-disaggregated beneficiary assessments
This research technique is used to ask actual or potential beneficiaries the extent
to which government policies and programmes match these peoples’ priorities.
3 gender-disaggregated public expenditure incidence analysis
This research technique compares public expenditure for a given programme,
usually with data from household surveys, to reveal the distribution of expenditure
between women and men, girls and boys.
4 gender-disaggregated tax incidence analysis
This research technique examines both direct and indirect taxes in order to
calculate how much taxation is paid by different individuals or households.
5 gender-disaggregated analysis of the impact of the budget on time use
This looks at the relationship between the national budget and the way time is used
in households. This ensures that the time spent by women in unpaid work is
accounted for in policy analysis.
continued over page
Section 5: Tools for a gender-sensitive analysis of budgets 37

6 gender-aware medium term economic policy framework

This attempts to incorporate gender into the economic models on which medium-
term economic frameworks are based.
7 gender-aware budget statement
This involves an accountability process which may utilise any of the above tools. It
requires a high degree of commitment and co-ordination throughout the public
sector as ministries or departments undertake an assessment of the gender impact
of their line budgets.
Source: Adapted from Diane Elson (1997b), ‘Tools for gender integration into macroeconomic policy’
in Link in to Gender and Development, 2, Summer, p 13.
1 Gender-aware policy appraisal

A gender-sensitive analysis of the budget begins with the assumption that
budgets must follow policy. If budgets follow policy then a gender-aware policy
appraisal offers a means of identifying policy gaps and limitations as well as the
adequacy of the allied resource allocations. Thus, a gender-aware policy
appraisal makes a direct and strong link between policy and resource allocation.
In contrast to some other tools (eg gender-disaggregated tax and expenditure
incidence analysis), this analytical approach operates at a relatively aggregated
level. It recognises, for example, that it is difficult to determine which
individuals or groups are directly affected or benefit from government resource
allocation. Government preventative health policies, for example, appear to
have a diffuse effect. However, it is still possible and appropriate for good
policy to analyse the possible gender impact of programmes or strategies within
the policy (eg the way in which the public health campaign targets the different
health issues of men and women).
A gender-aware policy appraisal involves the development of an analysis which
reflects an understanding of the policy’s gendered implications by:
identifying the implicit and explicit gender issues;
identifying the allied resource allocations; and
assessing whether the policy will continue or change existing inequalities
between men and women (and groups of men and women) and patterns of
gender relations.
The techniques that might be utilised to develop this analysis include:

a checklist of questions for assessing the policy, including checking the
gendered assumptions of the policy against the evidence;
a discussion of events, activities and associated budget allocations
generated by the policy; and
checking the policy against its stated aims and performance objectives.
38 Section 5: Tools for a gender-sensitive analysis of budgets

A weakness of the gender-aware policy appraisal tool, which relates to its level
of aggregation, is that the ‘nature and scale of the links in the causal chain
cannot be predicted with accuracy’ (Diane Elson, 1997d: 1). This tool can,
however, be used readily by analysts either inside or outside of government.
An example from South Africa in Comment 23 is the gender-aware appraisal of
its land reform policy.
c omme nt Land reform in South Africa

23
The Department of Land Affairs is responsible for creating and implementing a land
reform programme… The allocation to Land Affairs for 1998/99 is R685,4 million,
increasing to R967,4 million in 2000/01. The rapidly growing allocations reflect the
increasing pace of implementation of land reform…
The poorer provinces are disproportionately rural, and contain disproportionately many
women and children. Although this is a characteristic of many other countries, in South
Africa it has been exacerbated by influx control, pass laws, forced removals and the
migratory labour system.
In addition to the racial inequalities experienced by black men, women suffer
discrimination from social practices. Legal restrictions have impeded women’s access
both to land and to the financial resources to develop it. In addition to legislation which
designated women as minors who could not own property or conclude contracts in their
own right, customary law has, in many cases, deprived women of the right to own
property.
Women tend to have access to smaller plots, with less available irrigation, and a
smaller likelihood that they will have paid employment to augment their livelihoods.
Around 80 per cent of woman-headed households with access to land contain no wage
or salary earners.
The Department of Land Affairs is conscious that unless more attention is directed to
meeting women’s needs and concerns, the land reform programme could exacerbate
existing gender inequities in the allocation of land and its productive use. The
Department is providing gender training and is endeavouring to build gender concerns
into its monitoring and evaluation system.
Source: Department of Finance, Republic of South Africa (1998), Budget Review 1998, Pretoria,
pp 6.36-7.
Another example of a gender-aware policy appraisal is from an Australian State

government women’s budget in Comment 24. This example focuses on the links
between policy goals and implementation by responsible agencies.

c omme nt Domestic Violence Policy in Tasmania, Australia

24
Policy
Tasmanian Government Domestic Violence Policy Statement 1994 -
Strategies identified within the policy are:
emphasising that assault that occurs in a domestic setting is a crime;
providing immediate and follow up safety, protection and support for victims and
survivors;
increasing community awareness of the costs and consequences of domestic
violence; and
ensuring that services in the Government and non-government sectors respond
appropriately and effectively to domestic assault and violence issues.
Key stakeholder agencies identified include:

Department of Justice;
Office of the Status of Women;
Tasmania Police; and
Department of Community and Health Services.
Implementation
Department of Police and Public Safety, Tasmania
Output group
Policing support to the community...
Output description
This output is directed at maintaining and improving personal safety in the community,
reducing the incidence of property offences as well as promoting community
participation in managing public order, safety initiatives and crime prevention. Activities
include patrols; responding to requests from the public; responding to incidents;
licensing; dispute intervention; community partnerships; and crime prevention
education.
Elements of this output that specifically address the Tasmanian Government
Domestic Violence Policy Statement 1994
Tasmania Police actively embrace a pro-arrest policy in relation to domestic violence
and support the prosecution of offenders where evidence of domestic violence exists.
Users of these elements include Tasmania Police Officers; victims of domestic
violence; and, perpetrators of domestic violence. Action also acts as a general deterrent
and so benefits the broader community.
Objectives of these elements include:
prosecution of perpetrators of domestic violence;
the removal of victim from harm; and,
maximising the safety of victims.
Performance measures relevant to these elements are:

rate of arrests for Domestic Violence incidents attended;
number of referrals or complaints of police inaction received by the Domestic
Violence Crisis Service;
recorded number of complaints; and
recorded number of enquiries. continued over page

Major issues affecting delivery of these elements:

availability of evidence;
gaining the support of victims so that prosecutions may proceed with their cooperation;
inappropriate categorisation of certain incidents as ‘domestic violence’; and
public complacency.
Proportion of the budget allocated to this output that was directed toward
elements that specifically address the Tasmanian Government Domestic Violence
Policy Statement 1994:
A snapshot across uniform policing throughout Tasmania for a period of one week in
September 1966 indicated that 1.4% of uniformed officers’ time was spent responding
to domestic violence incidents. This result is yet to be validated by a further activity
survey, however the indicative figure would suggest that approximately $700,000 per
annum is expended responding to incidents of domestic violence.
Tasmania Police provided $17,000 toward the operating costs of the Sexual Assault
Support Service.
Source: Government of Tasmania (1998), Achievements for Women from the Budget 1997-98,
Hobart: Government Printing Service, pp 5-6.
What other issues might be examined in assessing the implementation of the
domestic violence policy in the following year?
2 Gender-disaggregated beneficiary assessments

This is a tool which allows the voice of the citizen to be heard. Potential and
actual beneficiaries of a government programme are asked, using a variety of
techniques, their views as to whether existing forms of public service delivery
meet their needs as they perceive them. These responses are analysed in order to
assess the extent to which a government’s current budget meets the priorities of
women and men. In essence, women and men participants in beneficiary studies
are being ‘asked how, if they were the Finance Minister, they would slice the
national budgetary pie’ (Diane Elson, 1997b: 13).
Data for disaggregated beneficiary assessments can be collected by using
quantitative surveys (eg opinion polls and attitude surveys) and qualitative
processes (eg focus groups, interviews, participant observation). Each technique
has particular strengths and limitations. The assessments can be initiated by
various agencies within government or by groups within civil society (see also
Diane Elson, 1997e).
An example of a beneficiary assessment of peoples’ preferences for public and
private provision of goods and services using a quantitative survey to collect the
data is given in Comment 25. Both sides of the budget – expenditures and
revenues – were included by asking if people would be prepared to pay more in
taxes in order to acquire increased public goods and services.

c omme nt Beneficiary assessment of public expenditure in Australia

25
The following two tables are based on a survey of 600 Australian adults in late 1992.
The survey examined peoples’ preferences for public (government) or private
provision of different goods and services. Table 5.1 below shows the percentage of
people who acknowledged the benefits of public expenditure in selected portfolios.
Table 5.1 Positive recognition of benefits from public expenditure (% of respondents)
Benefit community Benefit

individual/household
Public transport 68 29
Source: Extracted from
Police, law & order 69 53
Glenn Withers, David
Throsby and Kaye Johnston Roads 61 64
(1994), Public Expenditure
National defence 44 27
in Australia, Economic
Planning Advisory Medical & hospital 57 57
Commission Paper No. 3,
Education 64 40
Canberra: Australian
Government Publishing Housing 51 13
Service, p 31. Unemployment 74 12
(Commonwealth of Australia
copyright. Reproduced by Age pensions 71 17
permission.) Family assistance 56 13
Industry assistance 31 5
Table 5.2 shows the percentage of people who said they would like government
expenditure to increase in these same portfolios. The first column of figures gives
the percentage favouring an increase if they must pay for this. The second column
gives the percentage favouring an increase if there is no payment.
Table 5.2 Percentage of respondents wanting increased government

expenditure in selected portfolios
If must pay If don’t pay

Source: Extracted from
Public transport 50 59
Glenn Withers, David
Throsby and Kaye Johnston Police, law & order 71 77
(1994), Public Expenditure
Roads 67 68
in Australia, Economic
Planning Advisory National defence 38 34
Commission Paper No. 3,
Medical & hospital 85 83
Canberra: Australian
Government Publishing Education 74 83
Service, p 38. Housing 48 50
(Commonwealth of Australia
copyright. Reproduced by Unemployment 23 30
permission.) Age pensions 55 62
Family assistance 63 71
Industry assistance 52 51

The responses in the above tables are not disaggregated according to whether the
respondent was a woman or man.
How do you think responses would differ for women and men? If so, why?
A gender-disaggregated beneficiary assessment of urban and rural poor women

in Sri Lanka is provided in Comment 26, which uses the qualitative technique of
focus groups.
c omme nt Beneficiary assessment of health service delivery

in Sri Lanka
26
Small group sessions were held with a group of urban poor and a group of rural poor
women to discuss their problems and needs to see what impact nearly two decades of
Women in Development programmes have had.
Urban
Urban women were most articulate about social and family problems, particularly the
effect of drug abuse on the family, in the context of weak law enforcement.
Health hazards such as stagnant water in canals, festering garbage, dense mosquito
populations and lack of toilet facilities were also a major concern. Legal disabilities in
acquiring housing resulted in overcrowding. Education facilities were available but their
poverty denied them access to ‘ good schools’ and vocational training. Stable
employment was out of reach and their finances were inadequate for self employment.
Rape and sexual harassment in public places were also concerns.
Rural
Rape, child abuse and alcoholism were seen as requiring harsh punitive action.
Women said that public awareness should be raised on the consequences of marital
breakdown, the dowry system and the negative portrayal of women in the media. Drug
abuse was of less concern than among urban women.
They were acutely conscious of their poor nutritional status, high incidence of
morbidity, poor sanitation and lack of health education. Economic constraints
prevented them from making full use of the educational facilities available, though
education was seen as an avenue of economic development. The travails of women in
the Export Processing Zones, low wages and inadequate inputs for self employment
were seen to be barriers to upward mobility.
Both groups were fully aware of the issues that arise from inequitable social and gender
relations within and outside the family, but the more deprived urban women had to face
a harsher environment and were more concerned with immediate problems. Group
formation encouraged through social mobilisation programmes have developed a
sense of solidarity among these women and strengthened their capacity to analyse,
protest and suggest appropriate interventions to alleviate their problems.
Source: Extracted from Centre for Women’s Research, Sri Lanka (1996), ‘Perception of Women in
Economically Disadvantaged Families’, in Facets of Change: Women in Sri Lanka 1986-95, pp 408-414.

a Discussion exercise
Design a beneficiary assessment for selected programme/s of this government.
Discuss the methodology, logistics, costs, strengths and weaknesses of your chosen
approach.
3 Gender-disaggregated public expenditure

incidence analysis
This tool can be used to provide an assessment of the distribution of government
expenditure of a given programme between men and women and boys and girls.
For example, gender-disaggregated public expenditure incidence analysis could
be used to compare expenditure on public schools in different localities to the
number of boys and girls enrolled. It could also compare expenditures in
different areas with the levels of gender specific needs revealed in surveys.
This tool requires considerable amounts of quantitative data in order to estimate
both the unit cost of providing a particular government service and the
utilisation of public expenditures by households or individuals disaggregated by
gender. In respect of the cost of service provision, data must usually be obtained
from government agencies. Data to estimate programme utilisation can often be
derived from a variety of sources including household surveys conducted by the
national statistical agency. A summary of the gender-disaggregated approach to
public expenditure incidence analysis is provided in Comment 27.
c omme nt Disaggregating benefit incidence:

Gender and public social spending
27
Benefit incidence analysis involves a two-step methodology. First, estimates are
obtained on the unit cost of providing a particular service. These are usually based on
officially reported public spending on the service in question. Second, these unit costs
are then ‘paid’ to households which are identified (usually through a household survey)
as users of the service. Households which use a subsidized public service in effect gain
in in-kind transfer, which depends on the unit subsidy involved (say, the subsidy per
primary school enrolment) and the number of units consumed by the household (the
number of children currently enrolled in a state-subsidized primary school).
The distribution of these subsidies, therefore, is determined by two broad factors. First,
it depends on government spending itself, and how it is allocated within a sector. The
lower the spending, and the greater the effective cost recovery, the lower will be the
subsidy embodied in the service provided. Second, the distribution will depend on
household behavior - on who uses the services that the government provides.
continued over page

It is only by using the service (by sending a child to primary school, or visiting the
outpatient department at a hospital) that individuals and households can lay claim to
the in-kind transfer that is implicit in the subsidy. Expenditure incidence analysis
therefore brings together two sources of information: data on the government subsidy
(estimated as the unit cost of providing the service less any cost recovery to the
government) allocated to the different categories of service…and information on the
use of these services by individuals and households, which is usually obtained from
household surveys.
How important gender disaggregations are in benefit analysis will depend on the type of
sector disaggregations that are possible. At one extreme, it may be possible to identify
services that are entirely gender specific - for example, the provision of pre-natal care
in the health sector. The greater is the share of total health spending allocated to such
services, the greater will be the benefit incidence to females. In most cases, however, it
is not possible to obtain such disaggregations, and most services defined with a sector
are usually available to both gender groups. Usually education services are divided into
primary, secondary and tertiary levels, while health services are disaggregated into
health centers/clinics, outpatient hospital services, and in-patient hospital care.
Nevertheless, there remain gender differentials which emerge even at this level of
aggregation. Females are less likely than males in many developing countries to use
university schooling, so that the greater share of government spending allocated to
universities, the lower the share of education spending accruing to females…
Benefit incidence analysis is an exercise in accounting. It takes behavior - of both
households and governments - as given…[B]enefit incidence reveals the problems, but
only provides some indication of where the answers lie. It needs to be complemented
with more in-depth study of the underlying behavioral relationships.
Source: Lionel Demery (1996), Gender and Public Social Spending: Disaggregating Benefit
Incidence, Poverty and Social Policy Department, World Bank, Washington, pp 2-4.
Most benefit incidence analysis to date has been disaggregated according to

income quintiles, rather than by gender. When gender is introduced, it should be
added to income quintile analysis, as the existing studies have found that gender
effects often differ according to income or class. In Ghana, for example, benefit
incidence analysis found that poor women benefit as much as poor men from
health spending, but women in all income quintiles benefited less from spending
at all levels of education. Table 5.3 below indicates that this is the case also in
both Pakistan and Kenya.
Table 5.3 Gendered incidence of public expenditure on education
Country Spending per female Spending per male
Pakistan 26 rupees 56 rupees
Kenya 543 shillings 670 shillings

Source: World Bank, quoted in Diane Elson (1997a), ‘Gender-neutral, gender-blind, or gender-
sensitive budgets? Changing the conceptual framework to include women’s empowerment and the
economy of care’, Preparatory Country Mission to Integrate Gender into National Budgetary Policies
and Procedures, London: Commonwealth Secretariat, p 1.

Benefit incidence analysis is a powerful analytical method, but it has some

shortcomings. These include:
inherent problems in assigning collectively consumed benefits of
government provided goods and services to specific households and
individuals within households;
not accounting for the difference between measuring the financial cost of
providing a service and measuring its social impact (for example, not
acknowledging the social benefits (externalities) of educating women); and
not providing any indication of differences in women’s and men’s needs and
their respective requirements for particular services (see also Elson, 1997f).
An example from Sri Lanka of a gender-disaggregated public expenditure

incidence analysis is given in Comment 28.
c omme n t Sri Lanka food ration and stamps programmes

28
Under the ration programme, a quota of rice, wheat flour and sugar at varying levels of
subsidisation was provided by the Food Commissioner’s Department to the entire
population over one year of age (with the exception of income tax payers); subsidised
rice, flour, sugar, dhal and infant milk foods were also made available on the open
market for much of this period.
The food subsidy which emerged on account of these operations consisted of a
consumer subsidy on food distributed through the ration programme and on the open
market, and a producer subsidy on rice; government attempted to manage the food
subsidy by cross subsidisation, varying the quantity of subsidised items and the prices
at which they were made available.
In January 1978, the ration programme was restricted to 6,857 million persons in
households with annual incomes of less than Rs 3,600. In September 1979 it was
terminated and replaced by food and kerosene stamps for the same target group but
covering 7,259 million persons.
No additions were allowed to the list of recipients after 1980; all new-borns and families
experiencing subsequent loss of income were thereby excluded. In 1985 a review of
food stamp beneficiaries was done and the programme brought under the Poor Relief
Act, abandoning any semblance of a nutritional objective.
The value of the stamps was initially determined to provide the same basket of goods
as ration holders had been receiving at that time. However, in January 1980, prices of
subsidised essential food commodities were allowed to rise sharply and continued to
climb gradually thereafter, eroding the real value of the income transfer. A residual
subsidy on infant milk foods was maintained by government, but the subsidisation of
other food commodities practically ceased.
continued over page

Despite rapid economic growth between 1980/1-85/86 averaging 5.2% p.a., the
continuous erosion of the real value of the stamps and a decline in the real incomes of
the poor...resulted in serious calorie deficiencies being experienced by the poor.
Although the Consumer Finance Survey data shows that daily per capita consumption
had been maintained at adequate levels in general i.e. 2,283 calories in 1978/79 and
2,271 in 1981/82, the per capita calorie consumption of the poorest 40% of the
population feel to 1,834 calories in 1978/79 and 1,865 in 1981/82...the consumption of
the lowest decile had fallen to the unprecedented level of 1,181 calories per capita a
day...
In contrast, the upper 60% of the population had increased their daily intakes from
2,574 to 2,700 calories per capita during these two years... In 1969/70, when the ration
was in effect, the daily per capita calorie consumption of the bottom 40% was 2,064
and that of the rest of the population was 2,352...
The women and girls in these households probably took the brunt of the increasing
food deficit in the nineteen-eighties as indicated by higher levels of malnutrition among
pre-school and school girls and declining birth weights of babies born to low income
mothers.
Source: P. Alailima (1997), Poverty and Unemployment in Sri Lanka, Colombo: Department of
National Planning, Ministry of Finance and Planning, pp 6-7.
What would be the steps in doing a benefit incidence analysis of Sri Lanka’s
changing food assistance policy over the years?
4 Gender-disaggregated tax incidence analysis

This technique examines the taxation component of revenue. It looks at both
direct and indirect taxes and calculates how much taxation is paid by different
individuals or households. It requires data on income and expenditure patterns,
which can typically be obtained from household surveys and from revenue
collection agencies. A limitation of the analysis is that it usually assumes equal
sharing of income within households, whereas many studies have shown this
does not often occur.
An example of tax incidence analysis for a gender-sensitive budget is provided
in Comment 29.

c omme nt Value added tax in South Africa

29
The narrow tax base and the characteristics of the distribution of income and
employment mean that although women do not pay a large proportion of total tax
gathered, increasingly the burden of indirect taxation is falling disproportionately on
women….
Both general sales tax (GST) and value-added tax (VAT) are regressive taxes -
affecting the lower income earners more severely as the proportion of their income paid
in tax is much higher relatively speaking than that of higher income earners. At present
a selection of basic foodstuffs is zero-rated: brown bread, maize meal, dried beans,
milk powder, rice, vegetables and fruit are included…
Projections of revenue to be collected and its impact on households, as estimated by
the Department of Finance, reveal unequivocally the strong regressive nature of VAT.
The table [below] indicates the very poor (household income up to R10,250 per annum)
would be paying 9.0% of household income on VAT. The very high income category
(over R328,000 per annum) would be paying 5.4% in VAT, at its current rate plus zero-
ratings.
Estimated percentage of household income spent on value-added tax

very poor low income middle income high income very high income
9,0% 7,3% 7,4% 7,1% 5,4%
Source: Trudi Hartzenberg (1996), ‘Taxation’ in Debbie Budlender (editor), The Women’s Budget,
Cape Town: Institute for Democracy in South Africa, pp 218-233.
5 Gender-disaggregated analysis of the impact

of the budget on time use
This is a method of analysing the relationships between the national budget and
the way time is used in households. It examines the extent to which budgets rely
on unpaid work such as caring for young people, the aged and those who are ill,
collecting fuel and water, cooking, cleaning and so on. The analysis is
dependent on having available time use studies. Time use studies have rarely
been conducted at the national level in developing countries, but are sometimes
available in the form of case studies. A national time use study would need to be
done by the national statistical agency, but the subsequent analysis can be
conducted both inside and outside government.
A discussion of the gender division of time use and its implications for poverty
alleviation policies is provided in Comment 30.

c omme nt Is time an asset?

30
Time-use studies find that in almost every country women spend more hours than men in
work - paid and unpaid. [The United Nations Development Programme’s] Human
Development Report 1995 showed that of the total time spent in work, women on average
account for 53% and men for 47%. As demands on women’s time increase, they share their
work with their daughters and with other children, but very rarely are household
responsibilities transferred to men.
Some circumstances - such as having a water supply close to home and transport to the
workplace - increase the time available. Others - such as deforestation and pollution - reduce
the time available and increase the vulnerability of women.
The importance of time for poor people has policy implications. For example, in setting a
poverty line based on the income required for survival, policy-makers need to recognize that
survival income will vary depending on how much time is spent caring for children and the
sick and how much is available for income-earning work.
Policies that impose a financial burden on families, such as a reduction in health care
benefits, may also impose a time burden - with further repercussions for the ability to escape
poverty. And policies that reduce the time burden of poor people, such as by improving the
water supply, can remove a critical constraint on their ability to escape poverty.
Source: United Nations Development Programme (1997), Human Development Report, New York,
p 62. Copyright 1997 by the United Nations Development Programme. Reprinted by permission of
Oxford University Press, Inc.
An example of gender differences in time use in agricultural activities is

provided in Comment 31.
c omme nt Gender difference in time use

31
Table B Percentage of time spent by rural Sri Lankan women and men on
different activities
Activity Women Men
Food preparation 92 8
Winnowing and parboiling rice 100 0
Preserving food for the hungry season 80 20
Storing grain at harvest time 30 70
Production of fruits and vegetables for home consumption 80 20
Fetching water 98 2
Collecting firewood 65 35
Upkeep of house and yard 95 5
Bringing up children 90 10
Bathing children 80 20
Attending to the sick in the family 85 15
Source: Janet Henshall Momsen (1991), Women and Development in the Third World, London and
New York: Routledge, p 38.

What does the study on Sri Lankan agriculture indicate about the relationship
between gender and time use?
How could information on time use inform government policy?
How might time use studies be relevant for a gender-sensitive budget analysis in
this country?
6 Gender-aware medium term economic policy

framework
This incorporates gender into the economic models on which medium-term
economic frameworks are based, with the aim of affecting future budgets. Some
of the ways in which gender can be integrated into models is by:
disaggregating variables where gender is applicable;
incorporating both national income accounts and household income
accounts reflecting unpaid work; and
changing underlying assumptions about the social and institutional setup in
society, as well as about how the economy works.
Medium term economic modelling is done at two levels. At the more general
level, there is modelling of the total economy which involves variables such as
growth rates, budget deficits, inflation, interest and employment. This work is in
its very early stages (see Cagatay, Elson and Grown 1995).
At the more detailed level - and probably more immediately relevant for
gender-sensitive budget purposes - there are models which look at multi-year
budgetary allocations within the overall expenditure projected by the general
model. For instance, in education, the modelling would be based on the
projected number of pupils and the pupil:teacher ratios. In health, it would be
based on the projected number of patients to be treated by the public health
system and the average number of visits per patient. These variables can be
subjected to a gender analysis. The analysis, however, requires relatively high
level modelling skills and access to detailed economic data.
7 Gender-aware budget statement

This is a statement, or report, from each government portfolio on its audit of
policies, programmes and related budgets. In the process of developing such a
statement, the tools or methods discussed above are utilised. A gender-aware
budget statement is an accountability report by government in relation to its
gender equity objectives. Much of the rest of this document will focus on how to
develop a gender-sensitive budget statement.

Section 6
Putting care into the economy
A gender-sensitive budget analysis has to engage with the effects of government

policies and resource allocations on the unpaid activities of the household and
community sector.
Women and men tend to undertake different roles in economic and social life.
Nowhere are these differences more sharply drawn than in the work of
households. The productive nature of much household work is gradually gaining
recognition. Households are also the base from which much of the care activity
in relation to the young, old, sick and disabled takes place.
Female labour is intensively utilised in the work of caring, work which is vital
to, amongst other things, the reproduction of the future labour force. Many
government policies and resource allocations can directly and indirectly support
or fail to support the caring and productive activities of households. This
section provides a brief introduction to international developments in the field of
measuring, conceptualising and theorising the activities of households for the
economy.
Household work in the system of national accounts

In order to fulfil their functions, government budgets are formulated on
assumptions about what is of value and what constitutes output and work. In
short, government budgets and the policies they finance are based on particular
notions of what the economy is and how it is measured.
Traditionally, value, output and work, have been measured by gross national
product and other macroeconomic aggregates, such as investment and savings,
imports and exports along with government expenditures and revenues. These
goods and services are produced and exchanged in a market context. That is,
they are activities which involve payment by cash or credit. In this view, unpaid
activities and the work of households and communities are treated as being of
little or no consequence for economic and social policy. Increasingly, however,
this view is being questioned and shown to be erroneous. For example, in an
extensive analysis of the United Nations System of National Accounts, Marilyn
Waring looked at subsistence work, the environment and housework. She uses
an interesting example of a young Zimbabwean girl, Tendai, whose daily toil
begins at 4am and ends at 9pm. She ‘is considered unproductive, unoccupied,
economically inactive. According to the international economic system, Tendai
does not work and is not part of the labour force’ (Waring 1988: 13). Waring
compares this young girl’s labour with the counted work of a highly paid man
who sits all day in an underground military facility.
Section 6: Putting care into the economy 51

Since the mid-1980s the international women’s movement has drawn attention
to the problems of omitting unpaid work from a country’s statistical base. A
recommendation from the United Nations Second International Conference on
Women was that the unremunerated contributions of women in all areas of
development should be included in a country’s economic statistics including the
gross domestic product (GDP). This would mean, among other things,
quantifying the unpaid contributions undertaken largely by women in
agriculture, food production, reproduction and household activities.
Some of these ideas have been taken up at an international level with the 1993
extension of the UN System of National Accounts (SNA). The major change
has been to include estimates of the informal sector as part of the country’s
measured output and production. It is proposed that unpaid household work, in
the form of domestic and personal services produced by household members for
their own consumption, be measured in separate satellite accounts. A large part
of the unpaid work of households thus remains outside the official production
and output boundaries as determined by the United Nation’s SNA. This has
contributed to inconsistencies in how activities are seen for policy purposes.
The unpaid work of households has, however, been estimated using time use
studies in several countries. Many economists and policy makers are arguing
that the unpaid work of households is of economic value and policies which do
not take it into account are biased. Estimates in developed countries suggest
that, if unpaid work was included, GDP would be at least one and a half times
as large as currently measured. The traditional measures of output are thouoght
to involve even greater undercounting of productive activities in developing
countries.
Time use
The primary way of generating input data for satellite accounts is through a
time use study. In developing countries, the existing studies have generally been
small-scale, confined to certain areas or activities (eg agriculture) and have used
non-standard classification systems. There have been several significant studies
in OECD countries. Comment 32 illustrates some results of time use studies in
Australia.
52 Section 6: Putting care into the economy

c omme nt Australian estimates of the value of

unpaid household work
32
The industries of the household economy are collectively larger users of labour than the
combined sectors of the market economy. For example, in Australia in 1987, as
measured by weekly hours of labour input, market industries used 252 million hours
and household industries 282 million hours. Unpaid work exceeded paid work by 12 per
cent… 76 million hours per week are used in meal preparation, 63 million in cleaning
and laundry and 53 million in shopping… These three household industries compare
with the three largest market sector industries: wholesale and retail trade with 49
million hours per week, manufacturing with 43 million and community services with 41
million.
Source: Duncan Ironmonger (1994), ‘Why Measure and Value Unpaid Work?’, International
Conference on the Measurement and Valuation of Unpaid Work: Proceedings, Statistics Canada,
Cat No 89-532E, Ottawa, p 38.
Estimates of the differences in time use on different activities by Australian

women and men are given in Table 6.1.
Table 6.1 Average time in minutes spent daily on activities by Australian

women and men in 1992
Activity Women Men Average

Labour force 425 516 482
Domestic, childcare & 301 175 242
purchasing
Personal 628 621 624
Education 325 384 354
Community participation 99 125 110
Social and leisure 345 351 348
Source: Compiled from Australian Bureau of Statistics (1994), How Australians Use Their
Time, Revised Publication, Canberra: Australian Bureau of Statistics, p 20.
What is the average percentage of time spent on different activities by Australian
women and men?
What are the policy implications of this division of labour?
Time use studies show that marriage has a marked effect on the time
spent on unpaid household activities of women and men. See Comment
33.

c omme nt The effect of marriage on the time use of

Australian women and men
33
Despite many newly-weds’ hopes that marital life will involve a partnership of sharing
and mutual caring, the Time Use Survey shows that, when it comes to housework,
these hopes are not fulfilled. The effect of marriage is diametrically opposite for men
and women. Among married men, cooking, cleaning and laundry time are all reduced
whereas for women, the time devoted to these tasks is dramatically increased by
marriage. Compared to a single woman of equivalent age living alone, a married
woman spends 40 per cent more time in cooking, time spent cleaning increases by 17
per cent, and the time taken for laundry by 37 per cent. Women who live in shared
households would face after marriage a doubling up of time spent on laundry, a 73 per
cent increase in cleaning time and a 49 per cent increase in cooking time. The steepest
increase in indoor housework faces the bride who has come directly from her family of
origin. She would experience a fourfold increase in laundry and a doubling of her
previous cleaning and cooking times.
Source: Michael Bittman and Jocelyn Pixley (1997), The Double Life of the Family: Myth, Hope and
Experience, Sydney: Allen and Unwin, pp 105-6.
Figure 6.1 His marriage and her marriage

Source: Office for the Status of Women (1991), ‘Marriage: a bigger step for some’, Selected Findings
from Juggling Time, Canberra: OSW, p 14.

Engendering economic models

New ways of measuring the economy, such as time use studies, have made
many gender issues visible, which may have significant implications for policies
and their allied resource allocations. Making changes to how the economy is
measured means that we must also change our macroeconomic models. This, in
turn, results in an altered understanding of economic growth.
Traditionally, measures and models of the economy have treated the household
as being primarily a site of consumption. New thinking in macroeconomics
recognises that households also contribute to production. One of the central
activities of households is to produce goods and services, a large proportion of
which can be thought of as making up the care economy.
The care economy

The care economy can be incorporated into macroeconomic thinking by utilising
a simplified model of an economy’s output. In this circular flow model, shown
in Figure 6.2, national output is the product of the interaction of three sectors:
the private sector commodity economy; the public service economy; and the
household and community care economy. The creation of wealth in a country
depends on the output of all three sectors.
unpaid care economy paid economy
human capabilities and social framework
social and economic social and economic

household infrastructure public sector infrastructure private sector
and
community
care sector
consumption and investment commodities
Figure 6.2 The interdependence of the paid and unpaid economies

Sources: Diane Elson (1997a), ‘Gender-Neutral, Gender-Blind, or Gender-Sensitive Budgets?:
Changing the Conceptual Framework to Include Women’s Empowerment and the Economy of Care’,
Preparatory Country Mission to Integrate Gender into National Budgetary Policies and Procedures in
the Context of Economic reform, London: Commonwealth secretariat, p 9 and Susan Himmelweit
(1998b), ‘The need for gender impact analysis’, in Sarah Robinson (editor), The Purse or the Wallet?,
Proceedings of a seminar of The Women’s Budget Group held on 12th February, London, p 7.

The three sectors of the economy produce different types of goods and services
and are characterised by different values and motivations. The different sectors
are described in Comment 34.
c omme nt The care economy

34
The private sector commodity economy produces market-oriented goods and
services primarily in response to the profit motive…. The public service economy
produces a social and physical infrastructure, which is used for consumption and
investment in the commodity economy and the care economy… The public service
economy is market-oriented to the extent that its employees are paid wages and it is
financed through taxation, user charges and borrowing (and, sometimes, by increases
in the money supply). It is less market-oriented than the commodity economy because
it delivers many services free at the point of consumption….
The care economy produces family and community-oriented goods and services as
part of the process of caring for people. Work in the care economy is not paid, though it
may be supported by transfer payments from the government (such as pensions and
child benefit)… This economy is excluded, as a matter of principle, from the UN
System of National Accounts. Both men and women work in the care economy, but
overall it is relatively intensive in its use of female labour. The care economy
contributes to the welfare of the individuals receiving care, but it also contributes to the
activities of the commodity economy and the public service economy by supplying
human resources and by maintaining the social framework (supplying what some
economists call human capital and social capital to the commodity economy and public
service economy). The relative size of the three sectors varies depending on the level of
development and economic strategy pursued.
Source: Diane Elson (1997), ‘Gender-Neutral, Gender-Blind, or Gender-Sensitive Budgets?:

Changing the Conceptual Framework to Include Women’s Empowerment and the Economy of Care’,
Preparatory Country Mission to Integrate Gender into National Budgetary Policies and Procedures ,
London: Commonwealth Secretariat, pp 8-9.
The three sectors of the economy are interdependent, but the interdependencies
are not widely recognised. Though the dependence of the public sector on the
wealth generated by the private sector is often noted, the dependence of the
other two sectors on decisions taken in the unpaid care sector is frequently
overlooked. (For a more detailed discussion see Elson 1998, 1997a and
Himmelweit 1998b.)
Give examples of how a reduction in government expenditures in particular
portfolios might impact on the ‘care’ economy.
What factors might women take into account in their decisions to engage in
income-earning activities?

Section 7
Preparing a gender-sensitive budget

statement
A gender-sensitive budget analysis proceeds on the basis of being able to

classify government expenditures and revenues and then apply the tools and
analyses developed in Sections 5 and 6. The categorisation of expenditures and
revenues provides a conceptual framework for developing a gender-sensitive
budget statement.
Gender budgets: A framework for public expenditure

Government expenditures affect women and men directly, by design, or
indirectly as part of general policies. In order to identify the impact of budget
expenditures on women and girls, three categories of expenditures are
important. Together they add up to 100 per cent of budget expenditures.
Public expenditure categories
Category 1 Category 2
Specifically targeted Equal employment opportunity
expenditures by government expenditure by government
departments and authorities to agencies on their employees.
women or men in the
community intended to meet
their particular needs.
+ For example, training for lower
levels clerks (where women may
predominate), paid parental
For example, women’s health leave, crèche facilities for
programs, domestic violence children of employees.
counselling for men, special
programs for women with young
children.
Category 3
General or mainstream budget
expenditures by government
+ agencies which make goods or
services available to the whole
= Total
Expenditure
community, but which are
assessed for their gender
impact.
For example, who are the users of
primary health care? Who are the Source: Rhonda Sharp (1995), A Framework
learners in government-provided for Gathering Budget Information from
literacy classes? Who receives Government Departments and Authorities,
agricultural support services? Mimeo, Adelaide: Research Centre for Gender
Studies, University of South Australia.
Section 7: Preparing a gender-sensitive budget statement 57

Evidence suggests that the vast majority (more than 99 per cent) of government
expenditures fall into Category 3. While many governments have some targeted
programs for women and men (Categories 1 and 2) these are small in terms of
the total government budget, and are often designed to fulfil a short-term need.
The examples which follow illustrate how the three categories of expenditures
have been reported in gender-sensitive budgets. These examples are not intended
to be read as ‘best practice’, but as examples to stimulate discussion about how
to undertake a gender-sensitive budget exercise in a specific country.
Comment 35 illustrates the utilisation of the three categories of expenditure in
an early women’s budget statement on agriculture in South Australia.
c omme nt South Australia: Agriculture

(Expenditure Categories 1, 3 and 2)
35
The majority of agricultural properties are operated by partnerships which involved
women who also have an active role in the management and operation of many
properties. Recognising the contribution that women make to the rural community and
economy, the Department of Agriculture is committed to improving rural women’s
access to information and involvement in decision making. Furthermore, the
department promotes the principles and practices of equal opportunity in employment
and is actively working to improve career opportunities for women working in the
department. The departmental programs and budget allocations are as follows.
Program 1988-89 1989-90

$000s $000s
Agricultural Industries Policy 66,186 62,701
Agricultural Crop Industries 9,853 11,056
Horticultural Crop Industries 4,068 4,578
Animal Industries 15,372 16,549
Farm Management and Rural Community Support 1,961 2,450
Agricultural Resource Management 16,550 18,350
State Disaster Planning, Control and Relief 3,200 3,072
Provision of Advisory and Analytical Chemistry - 2,519
Services
Service and Supply 12,817 12,943
Total 130,007 134,218
continued
58 Section 7: Preparing a gender-sensitive budget statement

Allocations specifically targeted to women and girls (Expenditure Category 1)

Project Women’s Agricultural Bureau
Aim To encourage interest and participation in the development of rural
life, particularly in agriculture.
To encourage educational and cultural opportunities at all levels.
To influence policy making at all levels
Issues Most women on family farms are members of business
partnerships and have responsibility for financial record-keeping.
Many do not have any formal training in business management.
Activity 1 Financial Planning Seminars for Women coordinated by the
Women’s Agricultural Bureau
Indicator Twenty-five participants per seminar. In 1989-90 averaged 22
participants per seminar
Action 14 seminars in 1989-90
Expected Greater skill and confidence in financial management issues.
outcome
Activity 2 Computer workshops for rural women coordinated by Women’s
Agricultural Bureau.
Action 8 courses conducted in 1989-90
Expected Women on farm properties benefit from information about new
outcome technology relevant to farm tasks.
Expenditure 1988-89 $42,200 1989-90 $50,000

Full-time 1.0
equivalent
staff
General allocations - impact of key activities on women/girls

(Expenditure Category 3)
Program Animal Industries
Issues The current level of child infection from
dogs is a danger to the health of the
children, as well as a problem for women
who are the primary care givers.
Activity Study of worms and infectious skin
conditions in dogs in Aboriginal
communities, and cross infection rates of
humans, particularly children.
Indicator Level of infection
Action Survey the kinds of infections currently
present in dogs at Ernabella and design a
program treat them.
Expected outcome Reduction in infections in dogs and in
humans.
Expenditure 1988-89 $- 1989-90: $15,000
continued

Program Farm Management and Rural Community Support

Issues Women working and living on farm properties are
primary targets for farm safety campaigns, both for
themselves and the occupational health and safety
problems of other members of their families.
Activity Farm safety seminars offered statewide, coordinated
jointly by the Women’s Agricultural Bureau, Rural Youth
and the Agricultural Bureau.
Indicator Percentage of women attendees at seminars.
Action Seven seminars conducted in 1988-89. Number of
seminars for 1989-90 not yet finalised.
Expected outcome Greater awareness of potential farm accidents.
Expenditure 1988-89 $6,000 1989-90-$8,500
Full-time equivalent 0.1
staff
Program Farm Management and Rural Community Support
Issues The need for equal access to program in personal and
group development and opportunities to develop
leadership skills and enable women to play an equal role
in community activities.
Activity Rural Youth Movement of South Australia
Indicator Proportion of female membership is 50%.
Action Provide equal access to programs.
Expected outcome Women are now taking an equal role in organisational
leadership at state level. The aim is to increase this at
grassroots level.
Expenditure 1988-89 $33,500 1989-90 $38,500
Full-time equivalent 1.0
staff
Program All Programs

Issues Access to information and advice is essential for
production and management of agricultural properties.
Activity Advisory services to farmers and farm managers
Indicator Women make up a significant minority of all those
identified as farmers and farm managers in Australian
censuses. This proportion rose from 29% in 1981 to
34% in 1986, mainly through the decline in the absolute
number of males in these categories.
Action A number of programs, funded by commonwealth grants
and using the Women’s Agricultural Bureau and Rural
Youth, aimed at raising the confidence of women.
Expected outcome Further growth in the proportion of women who regard
themselves, and are regarded by others, as contributing
positively to agricultural production in South Australia.
continued

Equal employment opportunity evaluation for women employees

% Female 1983 % Female 1988
Classification Profile
C01 below the barrier 78 79
C01 above the barrier 77 92
A01 to A05 10 42
E01 and above 7 13
Occupation
Clerical and Administrative 51 62
Technical 22 20
Professional 5 7
Weekly paid 28 31
Total 25 26
Employment Part-time Part-time Full-time Full-time Temp Temp

status 1983 1988 1983 1988 1983 1988
Females as % 10 15 65 60 25 25
of total female
work force
Males as % of 1 - 79 86 20 14
total male work
force
Aboriginal employees: The department has no Aboriginal employees.
The department has 87 committees, with a total of 77 female and 632 male
members.
Equal Employment Opportunity Budget 1989-90 $45 000.
No specific allocation to women.
Sources: The Budget and Its Impact on Women 1989-90 (1989), Financial Information Paper No 5,
Adelaide: South Australian Government Printer, pp 50-52; The Budget and Its Impact on Women
1990-91 (1990), Financial Information Paper No 5, Adelaide: South Australian Government Printer, p
39; The Budget and Its Impact on Women 1991-92 (1991), Financial Information paper No 5,
Adelaide: South Australian Government Printer, pp 133-135.

a Discussion exercises
Discuss this extract from an Australian State women’s budget, pointing out
strengths and weaknesses.
Think about the three categories of expenditure.
Relate your discussion to the agricultural situation in this country.
Comment 36 provides examples from South Africa of statements for public

sector employment (Category 2) expenditure.
c omme nt Two stories about public sector employment

36
Story 1 The Public Service in South Africa (Expenditure Category 2)
[The South African] government aims to create a ‘genuinely representative public service
which reflects the major characteristics of South African demography’. The White Paper
on the Transformation of the Public Service sets targets to be achieved over the next four
years:
half of management should be black;
30 per cent of new recruits to middle and senior management should be women.
IIn September last year there were 1,164,843 public servants. Over seven in every ten
(828,494) were employed by provincial governments. More than half are women.
Whites, who make up 20 per cent of the public service, are over-represented compared to
their 13 per cent share of the population. Women are underrepresented among Africans
and over-represented among whites.
The employment patterns differ between national and provincial government, reflecting the
differences in functions. The three large social services - Education, Health and Welfare -
are largely provided by provinces and have traditionally employed more women. The
security functions, which have traditionally employed more men, are provided at the
national level.
Source: Department of Finance, Republic of South Africa (1998), Budget Review 1998, Pretoria, p 6.41.
Story 2 Educators in the South African public service (Expenditure Category 2)

Educators constitute the single largest occupational family, accounting for over a quarter
of all Public Service employees. Although women dominate this family (67% of the total),
they are rarely found at the top end of the salary scale. They account for 7% of those
earning R131,478 and above….
The table below gives the average wage for each race-gender grouping, first in absolute
terms, and then as a proportion of the average African woman’s wage for this
occupational family. The greatest variation occurs between white men (1,64 compared to
African women) and African women. These variations generally occur because of
differences in qualifications, length of service and rank... There are also substantial
differences between white women and African women...
continued over page

Average salaries: Educators

Total Total Total African African White White
All Women Men Women Men Women Men
Actual 46545 44233 50519 38365 41495 54761 65888
Index 1.21 1.15 1.32 1.00 1.08 1.43 1.72
Source: Julia de Bruyn (1996), ‘Public Service’, in Debbie Budlender (editor) The Women’s Budget,
Cape Town: Institute for Democracy in South Africa, pp 192-3.
Membership of government boards and committees is an indicator of both

government expenditures on equal employment opportunity and the
participation of women and men in a country’s economic life. Comment 37
outlines the membership of committees and selected boards in Barbados.
c omme nt Committees in Barbados (Expenditure Category 2)

37
Representation at the level of statutory boards in the political process reveal some
interesting trends. A study analysing the Gender Composition of Boards reveals, for
instance, that in 1994, 7% of members of boards that generate revenue are women.
Traditionally the economic boards are dominated by men whilst the social boards are
dominated by women. In recent times, however, there has been a change. More men
are sitting on the social boards, but there has not been a significant increase in the
number of women on boards that generate revenue.
Membership of selected boards in Barbados, 1996

Name of body Women Men Total % Women
Personnel Administration Division 1 5 6 17%
Judicial and Legal Service 0 5 5 0%
Town & Country Planning Advisory 0 10 10 0%
Committee
Electrical Wireman Licensing Board 0 5 5 0%
Police Service Commission 0 5 5 0%
Electoral and Boundaries 1 4 5 20%
Commission
Medical Council 2 7 9 22%
Paramedical Professions Council 1 6 7 14%
Professional Engineers Registration 0 3 3 0%
Board
Commission Board/Defence Board 0 4 4 0%
Defence Board 0 4 4 0%
Barbados Tourism Investment 0 6 6 0%
Corporation
Cinematograph Film Censorship 4 9 13 31%
Board
Source: Background research for a preliminary mission to consult with the government of Barbados on
the integration of gender into the national budget, 1998, p 15.

What is the gender issue in this example?
What indicators (other than those shown in Comments 35, 36 and 37) could be
used to show the gender impact of public sector employment resource allocation?
Is the gender membership of boards and committees in this country an example
of government expenditures on equal employment opportunity (Category 2) OR
an example of general government policies and expenditures (Category 3)?
In some countries, an official (called an ombudsman) is appointed to investigate

individual complaints against public authorities. The funding of this office,
which women and men may use differently, can provide an important service to
the community. Comment 38 provides a simple example of a gender-
disaggregated analysis of a ‘general’, or Category 3, expenditure.
c omme nt Ombudsman’s Office (Expenditure Category 3)

38
The role of the Ombudsman’s Office is to determine and resolve complaints made by
the public about administrative actions of Government agencies. Women access the
services of the Ombudsman at a lower rate than men. Over the past three years,
statistical data demonstrates the proportion of women accessing the Ombudsman
service has remained at a consistent level.
Initiatives in 1998-9 will target younger women across the Territory... and women living
on [Aboriginal] communities. Information presentations about the services of the
Ombudsman will be offered to high school students during visits to centres outside
Darwin. A survey of Aboriginal language groups is currently being conducted with a
view to developing aural translations of Ombudsman’s advisory information.
Share of Ombudsman’s services access by females

1995-6 1996-7 1997-8
Police Services 29% 30% 33%
All other 38% 41% 40%
Total 37% 38% 38%
Source: Northern Territory Government (Australia) (1998), Women in the Budget 1998-99, Budget
Related Paper. Darwin: Northern Territory Government Publications, p 72.
What does this example tell us about the impact of ‘general’ expenditures made
by the Government Ombudsman?
If you were doing an analysis of these government community services to the
community in this country, would you do it in a different way?

The links between gender and a ‘general’ expenditure category are not always
adequately recognised, even when the area of policy and expenditure might be
expected to have significant gender effects. Gender-sensitive budget analyses
undertaken by governments in Australia have often been motivated by a desire
to publicise government policies. As a result, the budgetary analysis of
expenditure and policy has been limited. In Comment 39 a Category 3
expenditure is presented using a limited gender-aware policy appraisal.
c omme nt The Australian Women’s Budget -

Assistance for Business (Expenditure Category 3)
39
Approximately one-third of Australia’s small- and medium-sized enterprises (SMEs) are
owned by women and another 28 per cent are owned jointly by women and men.
Recent studies have revealed that the biggest problems facing SMEs, especially those
owned by women and particularly those with the potential to export, are the availability
of finance and the lack of information on how to access and manage finance.
The Government’s industry measures in the White Paper on Employment and Growth
boost assistance to small business. The following are key measures, to be implemented
over four years commencing in the 1994-95 financial year, which particularly assist SMEs.
Enterprise Development Expansion of the enterprise development program

Program which will involve additional funding of $102.3m for
the National Industry Extension Service (NIES) to
reach SMEs.
Enterprise Networking Funds of $38.2m will be provided to encourage
small- and medium-sized enterprises to form
networks of three or more firms.
AusIndustry Creation of AusIndustry, to enhance the delivery of
business improvement programs. AusIndustry
will…deliver business programs and act as an
information and referral service for related business
programs offered by other Government agencies.
Improved Small- and Initiatives costing a net $19.9m include:
Medium-Sized
Activities to improve the skills of business managers
Enterprises’ Access to
and capital providers, development of a set of
Finance benchmarks for manufacturing and service industries
to assist financial institutions in their lending
decisions and encouragement to SMEs to seek expert
advice in obtaining commercial export finance;
Advance payment bonds will be introduced to assist
exporters whose customers require a guarantee of
advance payments on a contract;
Funding for an advertising campaign to promote
awareness of the Commonwealth Development Bank
as a specialist source of finance for SMEs;
An increase in the tax incentive and relaxation of
many restrictions on Pooled Development Funds
(PDFs);
continued

Improved Small- and Enhancing of the Industry Innovation Program to the

Medium-Sized Enterprises tune of $118m. A network of technology diffusion
Access to Finance centres will be established to demonstrate new
technology to SMEs... SMEs will also receive
assistance to take up new technology and for early
commercialisation of technological innovation by
small firms; and
Arrangements to facilitate Commonwealth
purchasing from SMEs. Access to SMEs to
purchasing opportunities within the Commonwealth
market will be enhanced.
Training and Research: The Government will provide further training support
to SMEs.
It will also address barriers to women’s take-up of the
New Enterprise Incentive Scheme (NEIS), designed
to facilitate self employment. The White Paper
increases places available under NEIS from 5 000 to
6 500 in 1994-95 rising to 9 000 in 1996-97.
In addition, the Government is working with the
Australian Federation of Business and Professional
Women on a national survey to determine the factors
that hinder or encourage the growth of enterprises
run by women. A clearer understanding of those
factors will help Government to develop effective
policies to encourage growth of the small business
sector in Australia.
Source: Women’s Budget Statement 1995-96 (1995), Budget Related Paper No 3, Canberra:
Australian Government Publishing Service, pp 19-20. Commonwealth of Australia copyright
reproduced by permission.
How could this gender-aware policy appraisal be improved?
What are the gender issues in relation to small business assistance in this
country?
What questions would you raise in undertaking a gender-sensitive analysis of
small business assistance in this country?
Gender-sensitive budget analyses are ultimately concerned with producing better

policies, programs and resource allocations. Comment 40 is an example of a
‘general’ (Category 3) expenditure for which a gender-sensitive budget analysis
contributed to a programme revision.

c omme nt The National Public Works Program in South Africa

40
Approximately R350 million has been allocated to the National Public Works Program
from national budgets since 1994. The Community Based Public Work Program
(CBPWP) accounts for R250 million of this. Some provinces have added funds from
their own budgets.
A 1997 evaluation of the program found that the quality of the assets produced was
exceptionally high compared to similar programs elsewhere in the world. It found the
599 projects to be well-distributed geographically with respect to the relative poverty
profiles of the nine provinces. KwaZulu-Natal, Eastern Cape and Northern Province
jointly received nearly two-thirds of the nationally allocated funds.
Of those employed on the projects, 41 per cent were women and 12 per cent were
youths. While the figure for women is lower than the female proportion of the
population in the rural areas in which the projects operate, it is almost certainly higher
than would have been the case without explicit targeting. Unfortunately, the evaluation
suggests that women were often assigned the more menial jobs, that their average
wages were lower, that they were generally employed for shorter periods than men, and
that they were less likely than men to receive training. 37 per cent of men who were
employed received training, compared to 32 per cent of women.
The evaluation results have formed the basis of the Department’s current plans to fine-
tune the program and further improve targeting.
Department of Finance, Republic of South Africa (1998), Budget Review 1998, Pretoria, p 6.44.
How could this gender-aware policy appraisal be improved?
How do these policy issues relate to this country?

Gender budgets: A framework for public revenue

Gender analysis of revenue is still relatively undeveloped. Australia, the United
Kingdom and South Africa have, however, done some initial work in the area.
The possible categories of revenue are as follows:
Government revenue categories
Taxation Donor funds

For example, direct (income) Loans and grants
tax, indirect taxes (eg VAT,
customs and excise) and tax
‘expenditures’ (eg tax
+
incentives and rebates)
Other revenue
+ For example, user pays fees,

asset sales and borrowings = Total Revenue
The next part of Section 7 provides some examples of issues that need to be
taken into account in the reporting of a gender-sensitive analysis of government
revenues. The gender-sensitive analysis of government revenues is more
difficult, and often politically more sensitive, than is the gender analysis of
government expenditures. Nevertheless, it is still possible and desirable to
undertake an analysis.
One area which is usually the easiest to analyse is that of personal income tax,
as it can be analysed on the basis of whether the taxpayer is female or male.
Possible problems are that the authorities do not collect disaggregated
information, or that the income taxation unit is married couples rather than
individuals. Another problem area is where individual taxpayers who earn
below a certain threshold do not submit individual tax returns, with their
contributions being paid as lump payments by the employer. A third problem -
particularly in poorer countries - is that individual taxpayers comprise a very
small proportion of the population, therefore personal income taxation provides
a relatively small proportion of government revenue.
In virtually all cases, analysis of personal income tax will reveal that men,
overall, contribute a larger share of this revenue than women. This is because
more men than women are usually earning cash, and, as illustrated in Comment
41, they generally earn more than women and therefore tend to pay more tax.

c omme nt Personal income tax in Australia

41
In the 1990-91 income year there were approximately 3.4 million women taxpayers.
Women constitute 43 per cent of all taxpayers and pay 31 per cent of total income tax.
Married women constitute 60 per cent of all women taxpayers and pay slightly more
income tax on average than single women.
Income year 1990-91
Female Male Total
Single
Number of taxpayers 1337152 1652305 2989457
Taxable income $m 27872 39148 67019
Net tax $m 5460 8666 14127
Married
Taxable income $m 41916 88437 130352
Net tax $m 8101 22012 30112
Total
Taxable income $m 69788 127584 197372
Net tax $m 13561 30678 44239
Source: Women’s Budget Statement 1992-3 (1992), Budget Related Paper No 5, Canberra: Australian
Government Publishing Service, p 296. Commonwealth of Australia copyright reproduced by permission.
What are the gender issues in personal income taxation in this country?
The differential effect of taxes

More sensitive analyses of personal income tax would need to look at the extent
to which different types of tax rebates and other incentives are differentially
available to women and men, perhaps because of the sector they work in, the
type of jobs they do, or their levels of earning. Other benefits, such as tax
concessions for contributory employment-related pensions (recorded as a tax
expenditure in most budgets), will also tend to benefit men more than women
because more men than women are usually employed in the type of formal
sector jobs where these pensions are standard.
Many other taxes are more complicated because they are paid by households
rather than individuals. (The example, in Comment 40, compares the burden of
the indirect value-added tax on poorer and richer households). Value-added
taxes are generally regressive, meaning that poorer households tend to pay a
larger share of their income on these taxes than richer households.

This will have gendered implications to the extent that women are more likely to
be members of poorer households, and/or are usually responsible in most
households for daily purchases.
Taxes and duties on goods and services such as alcohol, tobacco and gambling
will affect men more than women to the extent that men in many societies are
more likely than women to drink, smoke and gamble. There could, however, be
a negative effect on women - these taxes may mean that men withhold greater
amounts of money from the common household pot.
Donor funds are a significant source of revenue in many developing countries.
The recipient country may have little choice in how donor funds are spent. For
example, funds can be tied to particular programs and policies which may have
differential gender implications and impacts.
One of the non-tax sources of government revenue is user fees. These could
include school and university payments, payments for health services, payments
for use of local government facilities, service charges for water and electricity,
and so on. User fees generate income for government. They can change usage
patterns; for example, they could force poor parents to choose between sending
boy and girl children to school. They can also impose additional time burdens
on citizens (usually women) where, for example, people are unable to afford
government charges for water and electricity, and women must collect water
and fuelwood instead.
A second non-tax source of government revenue is sales of assets. While these
generate once-off sources of income for government, they can result in
individuals subsequently having to pay for services which are currently
provided free or on a subsidised basis.
What are the different amounts and proportions of government revenue, using
the suggested government revenue categories, in this country?
How might a change in revenue raising policy, such as an increase in user pays
charges, affect women’s time burdens in the ‘care economy’?
The tax-benefit system

Changes to revenue raising (eg reductions in income tax rates) are frequently
accompanied by changes in government expenditures or benefits (eg greater
targeting of child support). A gender-sensitive analysis of these tax-benefit
changes requires an analysis of the interactions between the taxation system and
the system of government benefit payments.
Recent changes to the tax-benefit system in Britain, which attracted attention
for their potential gender effects, are provided in Comment 42.

c omme nt Britain’s new tax-benefit deal for working families

42
The Chancellor has announced five key proposals in the Budget:
a new tax credit paid to working families with children: the Working Families Tax
Credit [WFTC];
a new childcare tax credit within the WFTC to help working families with the cost of
childcare;
a new tax credit to help sick and disabled people into work: The Disabled Person’s
Tax Credit;
the biggest reform of national insurance contributions [NIC] since 1975, which will
encourage job creation at the lower end of the labour market, remove distortions
and reduce administrative burdens on employers. The NIC burden on the lower
paid will be reduced both for employees and employers, reducing the combined
charge by 3.20 pounds a week at earnings of 64 pounds a week;
a 1.25 billion pounds package to increase support for children, raising child benefit
for the oldest child by 2.50 pounds a week from April 1999 and the child premia for
the under 11s within the income-related benefits by 2.50 pounds a week from
November 1998.
The WFTC [is designed to make work pay for]...low- and middle-income families with
children. A disproportionate number of the lowest-earning households are ones where
the main earner is a woman. Couples will be able to choose whether the mother or the
father receives the tax credit. The WFTC also poses no threat to independent taxation.
So the WFTC, especially in conjunction with the National Minimum Wage, will be of
particular benefit to women.
Sources: HM Treasury (1998a), ‘The Working Families Tax Credit and work incentives’, The
Modernisation of Britain’s Tax and Benefit System, Number Three, p 5; HM Treasury (1998b),
‘Government Launches New Deal for Working Families ‘Making Work Pay’, Press Release, p 2.
How do tax credits differ from government expenditures?
What are the gender issues in the above example?
Are there any gender issues that might arise in the tax-benefit system in this
country?


Section 8
Portfolio work
Introduction to portfolio work

An initial strategy for undertaking a gender-sensitive budget analysis is to select
a limited number of portfolios/ministries or sectors. The selection of portfolios
requires some thought. It is readily accepted that ‘social’ portfolios such as
education, health and welfare have gendered implications. However, so do other
areas such as industry, agriculture, land and water, examples of which have
been provided in earlier sections of this booklet.
The purpose of the portfolio work is to identify gender issues and problems
specific to given sectoral allocations. In Section 7, we presented three categories
of expenditure and three categories of revenue. The aim of this section is to
produce one statement for each of the three expenditure categories in a given
sector, plus a fourth statement for the revenue category. This is an initial
strategy in the development of a gender-sensitive budget statement.
Preliminary tasks
Portfolio work is best done using a team approach. Participants should divide
into portfolios/sectors/ministries (eg education, agriculture) and:
study the material on gender issues for the country (e.g. Women and Men
booklet, or CEDAW report);
extract from the local material three problems/concerns related to gender
issues and present supporting data;
discuss the underlying causes and effects of the problems;
identify any sectoral programmes that address/relate to these problems;
determine the allocations to these programmes; and
discuss any difficulties in answering these questions.
Data needs
The preliminary tasks raise many data issues. Comment 42 identifies three
different types of data needed to undertake a gender-sensitive budget analysis.
Section 8: Portfolio work 73

c omme nt Data
43
The types of data needed for gender budget analysis can be divided into three broad
categories:
Inputs: measure what is put into the process. For example, the amount of money
budgeted or the staff allocated for a particular programme or project.
Outputs: measure direct products of a particular programme or project. For
example, the number of beneficiaries receiving a particular good or service.
Outcomes: measure the results of the policy or programme. For example,
increased health, educational levels, availability of time.
All three types of data are necessary. A given change in policy or in a project will affect
inputs and outputs far quicker than it affects outcomes. It is also usually very difficult to
attribute a given outcome to a particular, or single, policy or project. Ultimately,
however, a policy or project must be judged on the basis of outcomes.
The tools of gender-sensitive budget analysis will be needed to identify outputs and
outcomes.
a Discussion exercises
What input, output and outcome data are needed for your chosen programmes?
What are the different sources of data in this country? For each one indicate
availability, regularity and up-to-date-ness.
(Examples might include: household surveys; censi-population and other;
enterprise surveys; administrative data; non-government sources eg universities,
other institutions, NGOs.)
Will the analysis of the programmes focus on inputs, outputs or outcomes?
Developing a portfolio analysis

Each of the portfolio analyses needs to begin with a general introduction of the
portfolio area and its gender implications. This includes:
an outline of the main purpose of the portfolio and the gender issues in
general terms; and
an identification of the chief programmes within the portfolio, indicating the
amounts allocated for each programme and the percentage of the total
portfolio budget.
74 Section 8: Portfolio work

Expenditure analysis
Using the categories of expenditure developed in Section 7, the following three
statements can be developed:
Statement
1
Gender specific allocations
Describe specifically gender-targeted allocations by referring to:
the aim of the programme or project;
an analysis of the programme’s impact which refers to output indicator/s
(this might be the number of women or men beneficiaries of the
programme);
the allocation of resources; and
changes planned in the coming year.
Statement
2
Public sector employment
Recall that, in a gender-sensitive budget analysis, gender issues in public sector
employment can be thought about in two ways. Firstly, public service
employment patterns can reflect (or not) principles of equal employment
opportunity between men and women and between groups of men and women.
Secondly, the gender-sensitive delivery of programmes can be affected by the
gender employment structure of the public service and other government
appointed personnel. Although there was some discussion earlier as to whether
these allocations were Category 2 or 3 expenditures, for the purposes of
developing a portfolio statement, matters relating to public service employment
will be treated as Category 2 expenditure.
Accordingly, analyse the equal opportunity issues in public sector employment

by:
describing employment patterns within particular portfolios or the
public sector as a whole.
Disaggregate by gender, level of employment, form of employment (full or
part-time, permanent or temporary), salaries and benefits. Include other
disaggregations such as race and disability.
describing special initiatives to promote equal employment opportunity -
amount spent and numbers reached (eg women in management training,
gender training for officials).

analysing the number of women and men in positions with a gender

focus or specialism (eg police, medical and welfare officials dealing with
rape and domestic violence; men in positions dealing with men and gender
violence).
analysing the membership of boards and committees established under
the portfolio
Distinguish between paid and unpaid appointments and levels of
remuneration.
describing any changes planned in the coming year.
Statement
3
General allocations
‘General’ budgetary allocations can be analysed using any of the tools discussed
in Section 7. Choose a programme which is significant in both budget terms and
in gender terms. Present this by describing:
the aim of the programme or chosen activity within the programme (that is,
what it seeks to do);
how much is spent on the programme/project and what percentage this is of
total budget;
gender policy issue/s in the programme
an analysis of the impact of the programme, which includes an output or
outcome indicator/s; and
Statement
4
Revenue analysis
Using the categories of revenue outlined in Section 7, choose a source of
revenue (or a tax/benefit/policy) which is important for the portfolio area and
significant in gender terms. A revenue issue can be analysed by using a variety
of tools and presented by describing:
the proportion of total revenue this constitutes and how this has changed
over time;
the gender issues raised;
an analysis of the revenue issue, referring to indicator/s of how the burden
(benefits) falls according to gender and other categories; and

Developing an action plan

At this point we need to check that the process can continue successfully after
the initial portfolio work.
The following checklist of tasks and decisions needs to be discussed both in a
portfolio team and amongst members of the wider group undertaking the
gender-sensitive budget analysis. At the group level, decisions need also to be
reached on the overall form the exercise will take in this country, at least for the
first year. In particular, there must be agreement on scope, location and politics.
Checklist
What are the gender issues in terms of the four selected categories of
expenditures and revenues for each portfolio/sector/ministry?
Which tools of analysis will be used?
What indicators will be developed?
What are the data sources – both the sources which have to be developed
and those which are available already?
Will the focus be on future, current and/or past budgets/expenditures (and
revenues)?
What are the structures of responsibility?
In what document/format will the gender-sensitive budget analysis be
presented?
How does the development of a gender-sensitive portfolio analysis fit into
the budget cycle?
How will the exercise be evaluated?

A schedule of tasks needs to be drawn up by each portfolio team as well as by the

coordinators of the overall exercise.
Schedule of tasks
Task Components of the task Person(s) Material Start End
responsible submitted to date date

Section 9
References and bibliography
All of the comments, figures and tables provided in these materials have been
thoroughly referenced on the pages where they are found. The following references
cite other academic works referred to in this text or materials which may assist the
reader in furthering their understanding of the topic.
Bakker, Isabella (1998), Unpaid Work and Macroeconomics: New Discussions,

New Tools for Action, Ottawa: Status of Women Canada.
Blades, Derek (1997), A Proposal for the Measurement of Non-market
Household Production, Paris, OECD.
Brown, L (1995), Gender and the Implementation of Structural Adjustment in
Africa: Examining the Micro-Meso-Macro Linkages, Washington DC:
International Food Policy Research Institute.
Budlender, Debbie (ed) (1998), The Third Women’s Budget, Cape Town: Institute for
Democracy in South Africa.
Budlender, Debbie (ed) (1997), The Second Women’s Budget, Cape Town: Institute
for Democracy in South Africa.
Budlender, Debbie (ed) (1996), The Women’s Budget, Cape Town: Institute for
Democracy in South Africa.
Cagatay, Nilufer, Elson, Diane and Grown, Caren (eds) (1995), ‘Gender,
Adjustment and Macroeconomics’, Special Issue. World Development, Vol 23,
No 11.
Australian Bureau of Statistics (1994), How Australians Use Their Time,
Canberra: Australian Bureau of Statistics, ABS Catalogue No 4153.0,
Canberra.
Department of Finance, Republic of South Africa (1997), People’s Guide to the
Budget, Pretoria.
Elson, Diane (1998), ‘The Economic, the Political and the Domestic: Businesses,
States and Households in the Organisation of Production’, New Political
Economy, Vol 3, No 2, pp 189-208.
Elson, Diane (1997a), ‘Gender-Neutral, Gender-Blind, or Gender-Sensitive
Budgets?: Changing the Conceptual Framework to Include Women’s
Empowerment and the Economy of Care’, Preparatory Country Mission to
Integrate Gender into National Budgetary Policies and Procedures, London:
Commonwealth Secretariat.
Elson, Diane (1997b), ‘Gender and Macroeconomic Policy’, Link in to Gender
and Development, Issue 2, Summer, pp 12-14.
Elson, Diane (1997c), ‘Integrating Gender Issues into National Budgetary
Policies and Procedures within the Context of Economic Reform: Some Policy
Options’, Preparatory Country Mission to Integrate Gender into National
Budgetary Policies and Procedures, London: Commonwealth Secretariat.
Section 9: References and bibliography 79

Elson, Diane (1997d), ‘The Development of Policy Options for Governments to

Integrate Gender into National Budgetary Policies and Procedures: Tool 3,
Gender-Aware Policy Evaluation of Public Expenditure by Sector’,
Preparatory Country Mission to Integrate Gender into National Budgetary
Policies and Procedures, London: Commonwealth Secretariat.
Elson, Diane (1997e), ‘The Development of Policy Options for Governments to
Gender-Disaggregated Beneficiary Assessment of Public Service Delivery and
Budget Priorities’, Preparatory Country Mission to Integrate Gender into
National Budgetary Policies and Procedures, London: Commonwealth
Secretariat.
Elson, Diane (1997f), ‘The Development of Policy Options for Governments to
Gender-Disaggregated Public Expenditure Incidence Analysis’, Preparatory
Country Mission to Integrate Gender into National Budgetary Policies and
Procedures, London: Commonwealth Secretariat.
Hedman, Birgitta; Perucci, Francisca and Sundstrom, Pehr (1996), Engendering
Statistics: A Tool for Change, Stockholm: Statistics Sweden.
Hill, Anne and King, Elizabeth (1995), ‘Women’s Education and Economic
Well-being’, Feminist Economics, Vol 1, No 2, pp 21-46.
Himmelweit, Susan (1998a), ‘Care and the Budgetary Process’, paper presented to
Out of the Margins 2: Feminist Approaches to Economics Conference, held at the
University of Amsterdam, The Netherlands, June 2-5.
Himmelweit, Susan (1998b), ‘The Need for Gender Impact Analysis’, in
Robinson, Sarah (ed), The Purse of the Wallet?, Proceedings of the Women’s
Budget Group Seminar, held at Church House Conference Centre,
Westminster, London, February 12, London: Fawcett Society.
Hurt, Karen and Debbie Budlender (eds) (1998), Money Matters: Women and the
Government Budget, Cape Town: Institute for Democracy in South Africa.
Ironmonger, Duncan (1996), ‘Counting Outputs, Capital Inputs and Caring
Labour: Estimating Gross Household Product'’ Feminist Economics, Vol 2,
No 3, pp 37-64.
Mohiuddin, Yasmeen (1996), Country Rankings by the Status of Women Index, Paper
presented at the 1996 Conference of the International Association of Feminist
Economics, American University, Washington D.C.
Saito, K and Spurling, D (1992), Developing Agricultural Extension for Women
Farmers, World Bank Discussion Paper 156, Washington DC.
Sawer, Marian (1996), Femocrats and Ecorats: Women’s Policy Machinery in
Australia, Canada and New Zealand, Geneva: United Nations Research
Institute for Social Development.
Sawer, Marian (1990), Sisters in Suits: Women and Public Policy in Australia,
Sydney: Allen and Unwin.
Sharp, Rhonda (1999 forthcoming), ‘Women’s Budgets’, in Lewis, Meg and
Petersen, Janice (eds), Feminist Dictionary of Economics, New York: Edward
Elgar Publishers.
Sharp, Rhonda and Broomhill, Ray (1998), ‘International Policy Developments in
Engendering Government Budgets’ in Shannon, Elizabeth (ed), Australian
Women’s Policy Structures, Hobart: Centre for Public Management and Policy,
University of Tasmania.
80 Section 9: References and bibliography

Sharp, Rhonda and Broomhill, Ray (1990), ‘Women and Government Budgets’,
Australian Journal of Social Issues, Vol 25, No 1, pp 1-14.
Sharp, Rhonda and Broomhill, Ray (1988), Short Changed: Women and
Economic Policies, Sydney: Allen and Unwin.
Summers, Anne (1986), ‘Mandarins or missionaries: Women in the federal
bureaucracy’, in Grieve, Norma and Burns, Ailsa (eds), Australian Women:
New Feminist Perspectives, Melbourne: Oxford University Press.
Tibaijuka, A (1994), ‘The Cost of Differential Gender Roles in African
Agriculture: A Case Study of Smallholder Banana-Coffee Farms in the
Kagera Region, Tanzania’, Journal of Agricultural Economics, Vol 45, No 1,
pp 69-81.
Tzannatos, Z (1991), ‘Potential Gains from the Elimination of Gender
Differentials in the Labour Market’ in George Psacharopoulos and Zafiris
Tzannatos (eds), Women’s Employment and Pay in Latin America, Report No
10, Latin America and Caribbean Technical Department, Washington DC:
World Bank
Waring, Marilyn (1988), Counting for Nothing: What Men Value and What
Women are Worth, Sydney: Allen and Unwin.
World Bank (1995), Toward Gender Equality: The Role of Public Policy,
Washington DC: World Bank.
Section 9: References and bibliography 81

82 Section 9: References and bibliography

How To Do A Gender Sensitive Budget Analysis

Uploaded by

Copyright:

Available Formats

How To Do A Gender Sensitive Budget Analysis

Uploaded by

Document Information

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

How To Do A Gender Sensitive Budget Analysis

Uploaded by

Copyright:

Available Formats

A usA I D

How to do a gender-sensitive budget analysis:

Debbie Budlender & Rhonda Sharp

About the authors

1998 Debbie Budlender, Rhonda Sharp & the Commonwealth Secretariat

ISBN 0 86803 615 3

Published by Australian Agency for International Development,

What is a gender-sensitive budget 5

Gender, sex, men and women 17

What is a government budget 31

Tools for a gender-sensitive analysis of budgets 37

Putting care into the economy 51

Preparing a gender-sensitive budget statement 57

References and bibliography 79

comme n t 1 What i s a ge nde r se nsi ti ve budge t? 5

comme n t 2 G e nde r bl i ndne ss 6

comme n t 3 G ove r nme nt budge ts and ge nde r 6

comme n t 4 M ai nstr e ami ng ge nde r budge ts i n

comme n t 5 Austr al i a: Intr oduc ti on to the 1995-96

comme n t 6 M oz ambi que : A woman’s budge t 13

comme n t 7 The Tanz ani an G e nde r B udge t Ini ti ati ve 14

comme n t 8 Re c omme ndati ons to the UK G ove r nme nt 15

comme n t 9 A snapshot of ge nde r -r e l ate d Wor l d B ank

comme n t 1 0 Pr i vati sati on vs de -i nsti tuti onal i sm 16

comme n t 1 1 Se x and ge nde r 17

comme n t 1 2 G e nde r e d soc i al r e l ati ons 18

comme n t 1 3 Wome n and me n 18

com m en t 14 B asi c wate r suppl y 19

com m en t 15 The e ffi c i e nc y be ne fi ts of r e duc i ng ge nde r

comme n t 1 7 H e al th se r vi c e r e str uc tur i ng: i nc r e ase i n

comme n t 1 8 Indi c ator s 26

comme n t 1 9 Cate gor i si ng ge nde r stati sti c s 28

comme n t 2 0 The budge t and pol i c y 33

com m en t 21 Austr al i an Commonwe al th budge t ti me tabl e

comme n t 2 2 Tool s for ge nde r -se nsi ti ve anal ysi s of budge ts 37

comme n t 2 3 Land r e for m i n South Afr i c a 39

comme n t 2 4 Dome sti c Vi ol e nc e Pol i c y i n Tasmani a, Austr al i a 40

comme n t 2 5 B e ne fi c i ar y asse ssme nt of publ i c e xpe ndi tur e i n

comme n t 2 6 B e ne fi c i ar y asse ssme nt of he l ath se r vi c e

comme n t 2 7 Di saggr e gati ng be ne fi t i nc i de nc e : G e nde r vs

comme n t 2 8 Sr i Lank a food r ati on and stamps pr ogr amme 46

comme n t 2 9 Val ue adde d tax i n South Afr i c a 48

com m en t 31 G e nde r di ffe r e nc e s i n ti me use 49

comme n t 3 3 The e ffe c t of mar r i age on the ti me use of

comme n t 3 4 The c ar e e c onomy 56

comme n t 3 5 South Austr al i a: Agr i c ul tur e

comme n t 3 6 Two stor i e s about publ i c se c tor e mpl oyme nt 62

comme n t 3 7 Commi tte e s i n B ar bados

comme n t 3 8 O mbudsman’s O ffi c e

comme n t 3 9 The Austr al i an Wome n’s B udge t - Assi stanc e

comme n t 4 9 The Nati onal Publ i c Wor k s Pr ogr am i n South

comme n t 4 1 Pe r sonal i nc ome tax i n Austr al i a 69

com m en t 42 B r i tai n’s ne w tax-be ne fi t de al for wor k i ng

Figure 2.1 Percentage of projects with gender-related actions, 1988-94 15

Table 3.1 Comparison of HDI, GDI and GEM ranks

Table 5.2 Percentage of respondents wanting increased government

Table 5.3 Gendered incidence of public expenditure on education 45

This document is the result of the efforts of many people. It is a concrete

Rhonda Sharp and Debbie Budlender