Adolfo Figueroa - Capitalist Development and The Peasant Economy in Perú (1984) PDF

CAMBRIDGE LATIN AMERICAN STUDIES
GENERAL EDITOR
MALCOLM DEAS
ADVISORY COMMITTEE
WERNER BAER, MARVIN BERNSTEIN,
AL STEPAN, BRYAN ROBERTS
47
CAPITALIST DEVELOPMENT AND THE

PEASANT ECONOMY IN PERU
For a list of other books in the
Cambridge Latin American Studies series,
please see page 141
CAPITALIST
DEVELOPMENT AND THE
PEASANT ECONOMY
IN PERU
ADOLFO FIGUEROA
Professor of Economics
Catholic University of Peru
The right of the

University of Cambridge
to print and sell
all manner of books
was granted by
Henry VIII in 1534.
The University has printed
and published continuously
since 1584.
CAMBRIDGE UNIVERSITY PRESS

Cambridge
London New York New Rochelle
Melbourne Sydney
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi
Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521253970
© Cambridge University Press 1984
This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 1984

This digitally printed version 2008
A catalogue record for this publication is available from the British Library
Library of Congress Catalogue Card Number: 83—18861
ISBN 978-0-521-25397-0 hardback

ISBN 978-0-521-10160-8 paperback
To IVAN and ROCIO
Contents
List of tables and figures Page viii

Acknowledgments xi
1 Introduction 1
2 Scope and method 3
3 The economic unit and economic organization 13
4 Production and exchange 24
5 The level and structure of peasant income 41
6 The economic behavior of the peasant family 58
7 Stagnation in the peasant economy and the role of demand 82
8 Economic crisis and the peasant economy, 1975-1980 102
9 Conclusions: reality, theory and policy 114
Appendixes
I The sample 126
II Methodological notes concerning the calculation of
peasant income 129
III Andean food plants 131
IV Methodology used in the calculation of Table 7.1 132
Notes 134
Bib liography 137
Index 139
vn
Tables
2.1 Intersectoral relations in the economy of a peasant

community 6
2.2 Peru: total and rural population by ecological region,
1972 9
2.3 The sample communities 12
3.1 Size of family and labor force 14
3.2 Distribution of cultivable parcels between cultivated and
fallow land 16
3.3 Number of parcels cultivated by number of crops 18
3.4 Animal stocks 19
3.5 Stocks of implements per family 20
4.1 Intersectoral relations in peasant communities 25
4.2 Average annual expenditure on modern inputs 33
4.3 Structure of monetary exports 35
4.4 Structure of monetary imports 38
4.5 Ratio of non-monetary exchange in peasant communities 40
5.1 Structure of income of peasant communities: own-
consumption and exchange 43
5.2 Productive structure of the communities 45
5.3 Structure of income of peasant communities: wages, rents
and self-employment 46
5.4 Money transfers 47
5.5 Structure of monetary income of peasant families 49
5.6 Structure of monetary gross expenditure of the peasant
family 50
5.7 Inequality in family monetary incomes 53
5.8 Summary of the regression analysis 55
6.1 Temporary migrations 68
6.2 Destination of temporary migrants to external labor
markets 69
viii
List of tables and figures ix
6.3 Employment sectors of temporary migrants 71
6.4 Residence of the children who left parents' homes 77
6.5 Residence and occupation of permanent migrants 78
6.6 Returns of members in peasant families 79
6.7 Reasons for returning 80
7.1 Peru: growth, inflation, distribution and pattern of
growth, 1950-80 84
7.2 Input-output table for food production-consumption
analysis 91
7.3 Peru: production structure for food, 1969 94
7.4 Lima: structure of family spending, 1969 95
7.5 Peru: imports/domestic output ratios for basic imported
food 98
7.6 Changes in world export structure 100
8.1 Peru: indicators of economic crisis, 1970-81 103
8.2 Peru: indicators of income distribution in the modern
sector, 1970-80 104
8.3 Lima: minimum wage, unemployment and underemploy-
ment rates (percentages) 105
8.4 Food subsidy incidence by families, 1980 111
A.I Expansion of the sample 127
Figures
6.1 Seasonality in peasant communities of the sierra 62
7.1 Engel curve and derived rural income curve 93
Acknowledgments
This book summarizes much of my research work of the last five

years. I initiated the study of the peasant economy in Peru in 1976,
thanks to a grant from the Social Science Research Council. The bulk
of the study on peasant communities was done within the ECIEL
Program (Estudios Con juntos de Integration Economica Latino-
americana) and financed by the Inter American Development Bank
through a convenio with ECIEL, during 1978-9. I must also
acknowledge the financial sources utilized during 1977-9 from the
grants given to the Economics Department, Catholic University, by
the Ford Foundation and the Canadian International Development
Agency. The University of Illinois at Urbana-Champaign provided
me with computational assistance during my stay at the Economics
Department as visiting professor in 1980.
Juan Ccamapaza, Wilfredo Ccori, Oscar Chaquilla, Jorge Diaz,
Matilde Ladron de Guevara, Felix Olaguivel, Ramiro Oregon, Cirilo
Quispitupa, Zoilo Quispitupa and Aurelio Succa, participated in the
fieldwork. In the stage of computational work I was assisted by
Bruno Barletti, Augusto Caceres, Jorge Rojas, Mario Tello and
Edgar Norton (at the University of Illinois). They all did an excellent
job and I am very grateful to them. Special thanks must go to Maria
Gabriela Vega and Daniel Cotlear who helped me efficiently in all the
stages of the research.
Many friends gave me valuable comments and suggestions at
different stages of the study. I received great intellectual support from
the ECIEL Program. Philip Musgrove, the technical coordinator of
the group, checked the analytics of the study on communities, and the
other members of the group, Alberto Petrecolla, Hector Dieguez,
Jorge Rodriguez, Mauricio Carrizosa and John Elac (from the Inter
American Development Bank), were always interested in the econ-
omic and social problems of the peasantry of my country. Werner
XI
xii Acknowledgments
Baer and the late Robert Ferber, at the University of Illinois,
contributed valuable comments to my work.
David Lehmann, Thomas Reardon, Alain Threffersen, Melissa
Birch and Russ Smith were generous enough to help me not only with
ideas and comments but also with translations and improvements in
the presentation of my ideas in English. Chapters 2-6 were published
in Spanish by the Catholic University as La Economia Campesina de la
Sierra del Peru (Lima, 1981). I also thank the two anonymous referees
provided by Cambridge University Press for their comments.
I would also like to thank my colleagues and my students of the
Economics Department, at the Catholic University in Lima, for their
encouragement. Javier Iguiniz, Head of the Department, helped me
in many ways to get this book published. Carmen Rosa Polo made an
excellent job of typing the manuscript.
This study could never have been done without the help of the
peasant communities studied. My gratitude to them is enormous; and
my commitment to the solution of peasant poverty is even greater. I
only hope that this book can contribute to the correct understanding
of the problem.
Lima
November 1982
1
Introduction
Peasant families represent approximately 25% of the population in

Latin America. They get their family income mostly from their small
plots of land. Despite the undoubted importance of peasant econ-
omies, there are as yet no satisfactory accounts or explanations of
their economic functioning and dynamics. The peasant economy
constitutes a 'reality without a theory'.
The importance of the peasantry comes not only from the number
of people but also from the fact that they represent the poorest social
group in Latin America. Several studies have shown that poverty in
Latin America is concentrated in rural areas and particularly in the
peasant families. Inequality and poverty will continue to be an
unfortunate characteristic of this region unless the real incomes of
peasant families are increased substantially. Economic policy to
reduce poverty must have the peasantry as one of the most important
target groups. Policies without theory however will not be success-
ful, except by accident. Again, our understanding of this particular
reality becomes a necessity.
This study is concerned with the peasant economy of Peru. This
country presents one of the extreme cases of inequality and rural
poverty in Latin America. Also the peasant population in Peru is a
large fraction of the total population (around 30%) and of the rural
population (around 66%). The historical process has, on the other
hand, generated in Peru one of the most extreme cases of cultural
duality, if one compares the sierra peasants and the upper and middle
classes living in Lima. Nowhere else does the notion of economic
duality seem more applicable than in Peru.
The scope of the study includes the analysis of the functioning of
the peasant economy in its present historical form in Peru, that is, in
the context of a predominantly capitalist system. The specific forms
that the economic and social relations between the peasantry and the
1
2 Capitalist development and peasant economy in Peru
rest of the economy take are the central issues in this study. From the
understanding of these relations, the role of the peasant economy in
the entire economic system can be better analyzed. The changes in
that role over time are also part of the scope of the study.
The method of investigation consists of studying in great detail a
sample of peasant communities in the most traditional and backward
region of Peru, the southern sierra. The fieldwork lasted about four
years (1976-9) and, in addition to questionnaires to collect quanti-
tative data, an anthropological approach was used in order to
understand processes and collect more qualitative data, particularly
on economic rationality. The results of this study of communities
constitute the core of the empirical data used in the analysis.
However, secondary sources have also been utilized at several points
in the book.
Investigating the case of Peru has a great methodological value for
the understanding of the peasant economy in Latin America. Because
this study refers to a case of extreme dualism in Latin America, some
results will apply with greater force to the general case in Latin
America. For instance, if our empirical research shows that the degree
of economic integration is significant in the case of Peru, the
conclusion to be drawn is that the peasant economy is an integral part
of the economic system in Peru, but this degree of integration is much
higher in the general case of Latin America. Other results of our study
will enhance the understanding of the peasant economy in other
individual countries.
The book is divided into nine chapters. Chapter 2 presents the
hypotheses about rural poverty in Peru and Latin America, which set
the questions to be answered in the study. A framework of analysis is
then constructed. This framework has been applied to the study of the
peasant communities in the southern sierra region of Peru. As
indicated above, this region is one of the poorest and most
'traditional' rural areas in Peru and in Latin America, and the study of
this reality has also a methodological value. Chapters 3-6 present the
results of the research in those peasant communities. Chapters 7 and 8
place the peasant economy in dynamic context, within the Peruvian
experience, with economic growth since 1950 and economic crisis
since 1975 respectively. Chapter 9 contains some conclusions in
terms of the hypotheses on rural underdevelopment, the dynamics of
the peasant economy under capitalist development in Peru, and the
economic policies that the case studies suggest.
2
Scope and method
Hypotheses on peasant economies

The persistence of poverty in the peasant economy in Latin America
calls for an explanation. For this, the study of the functioning of the
peasant economy and its relation to the rest of the economic system is
required. In fact, the relevant hypotheses can be classified according
to the emphasis given either to the production process or to the
process of exchange as the main cause of the economic backwardness
of the peasantry.
The production process

There are two common views concerning peasant economies: (1)
they are inefficient in their use of resources; (2) they are over-
populated, due to the absence of capitalist rules of production and
distribution. Against these two hypotheses, Professor Theodore
Schultz (1964) has developed a new proposition in terms of
'traditional agriculture'. His hypothesis is basically the following:
there is no significant inefficiency or overpopulation in peasant
economies. Poverty here can be explained by the poor quality of
resources and the traditional technology in use. Peasant families are
poor but efficient. As he put it: 'the community is poor because the
factors on which the economy is dependent are not capable of
producing more under existing circumstances. Conversely, ... the
observed poverty is not a consequence of any significant inefficiencies
in factor allocation' (p. 48).
In a dynamic sense, Schultz argues that 'the factors of production
on which a community depends are expensive sources of economic
growth' (p. 97). This is to say that the rate of return of traditional
capital is very low. The peasant economy is highly endowed with
traditional factors of production; the idea that there is scarcity of
capital in this economy and that the rate of return should then be very
high has no empirical base: the stock of traditional capital is very high
and, consequently, the rate of return very low. Moreover, this low
rate of return does not create incentives to save and invest in those
factors. The policy implication of this hypothesis is that the peasant
economy should shift to the use of modern factors of production.
Another hypothesis refers to the land tenure system. According to
this view, peasant families are exploited through non-capitalist social
relations, like share-cropping systems. These systems of production
and distribution would also have negative effects on the propensities
to introduce innovations and technical change in the production
process. A land reform program would be the policy to follow based
on this hypothesis.
The process of exchange

With regard to the exchange process, there are two conflicting
hypotheses. The first says that the peasant economy is not sufficiently
integrated into the rest of the economic system. This is the well-
known hypothesis of 'economic dualism'. The peasant economy is
basically a self-sufficient economy, outside the market system.
Therefore, the economic growth that takes place in the capitalist
sector does not spread to the peasantry due to the lack of linkages
between these two sub-systems.
The second hypothesis suggests exactly the contrary. The peasant
economy is well integrated into the capitalist economy through the
market mechanism and, therefore, the peasant family is subject to the
system of exploitation of capitalism. It is due to this integration that
the peasant economy becomes an underdeveloped economy (de
Janvry, 1974).
In order to test empirically the explanatory power of these
hypotheses, the process of production and exchange in peasant
economies must be studied. The particular reality which will be
studied for this purpose are the peasant communities of the sierra
region of Peru. The required analytical framework, which in-
corporates the basic elements of this reality, is presented below.
The analytical framework

Technological and market relations
The peasant community is not only an aggregate of families but a
social context which establishes certain economic relationships
Scope and method 5
among its members, and in which certain economic decisions are
reached collectively and certain economic activities are carried out
collectively. It is, indeed, the macroeconomic framework of the
peasant family which is the basic economic unit. As such, it comprises
three sectors of production: agricultural goods (A), livestock (P),
and a wide range of non-agricultural goods such as crafts, processed
foods, and construction which we shall call 'Z-goods' (Hymer and
Resnick, 1969). This mix of activities is found in most cases,
although, to be sure, there are also cases of specialization in potato
production, in pastoral activity, and in weaving.
In order to undertake production, the community possesses two
primary factors of production - land (T) and labor (H) - and three
types of initial stocks of products which correspond to the three
sectors mentioned above: seed (A), animals (P) and tools (Z). These
goods are produced, but since 'one needs commodities to produce
commodities', they must be already available as initial stocks if
production is to be undertaken. Other inputs enter into the
production of A, P and Z-goods, as flows, and they come either from
the community's annual production or from its imports. The
community as a whole exchanges with two 'external sectors': the rest
of the rural economy (JV) (other communities and haciendas) and the
urban economy (M).
A table of 'intersectoral relations' (Table 2.1) offers an analytical
approach to the relationships between primary factors of production,
initial stocks of goods and inputs on the one hand, and the annual
product and its allocation on the other. In Table 2.1 the agricultural
production of a particular year is represented by X±. This product is
net in the sense that it excludes the seed used, and it is allocated as
inputs to the same year's cattle production (e.g. as oats, barley and
other animal feeds), and to the production of Z-goods (the inputs
being processed into food, such as potatoes to be dehydrated for
chunoy and maize for beer, chicha). The remainder of the agricultural
production is for consumption (Cx), for the accumulation of further
stocks for the subsequent agricultural production cycle (71) and for
export or sale outside the community (Nx and Mx).
A year's livestock production is represented by X g . It comprises:
the number of animals produced, including chickens and guinea-
pigs; derivative products, such as milk, eggs, wool and skins; and
dung, which is used as fertilizer. Part of the production from this
sector is dedicated to the productive process as inputs: fertilizer for
agriculture; inputs for Z-good production (e.g. milk for cheese, wool
Table 2.1 Inter sectoral relations in the economy of a peasant community
Consump- Invest-
Sectors tion ment Exports
A P Z C I N M Total
Agricultural
goods A h M
i X,
y A y
Livestock P : |
Z-goods Z *:; x
*2 01 J^ M X2
Imports
in kind N h 0 0 Xn
Imports
(monetized) M Im 0 0 Xm
Labor H 0 Mh Xh
Land T in X,\ l\: % 0 0 0 X,
Agricultural
goods A ^11 ^12 ^li
Animals P
Tools Z 3
for blankets and clothes). Part is consumed as meat, milk, eggs and
skins; part is invested, as when the stock of animals is increased; and
part is exported.
Z-goods constitute a varied list, of which the main elements are the
following:
(a) processed food: chuno, chicha, cheese and dried meat;
(b) textiles: clothes, blankets, ponchos, sweaters, cloaks, sacks,
straps, harnesses and lassos;
(c) tools, and repair thereof;
(d) construction: houses, public buildings, yards and enclosures,
roads, canals, and construction materials such as adobe bricks and
roof-tiles;
(e) fuel: firewood;
(f) trade;
(g) transport;
(h) other crafts: ceramics, leatherwork, carpentry.
Some of these Z-goods contribute to agricultural and cattle pro-
Scope and method 7
duction, others to final demand. They contribute to agriculture as
sacks and other containers and in the repair of tools; they contribute
to cattle production as ropes, harnesses and lassos; they are consumed
as woolen textiles; they are invested in construction; and they are
exported as cheese.
Imported products from the rural economy (Xn) go into pro-
duction as agricultural inputs, as inputs for livestock production
(black salt fed to animals) or as inputs for the production of Z-goods.
They also are consumed, as in the case of maize imported by upper-
highland communities, and invested. The same can be said of urban
imports: fertilizers, pesticides, medicines for livestock, and dyes for
handicrafts are intermediate production goods; salt, sugar and
kerosene are for consumption; and they go to investment (above all
as steel implements).
Annual labor, measured in person-days, is used in all three sectors
of production, although production may often be in two sectors at
once, for example when someone tends cattle and spins wool at the
same time. Ch refers to 'consumption of labor', in rest-days or fiestas.
Nh and Mh denote exported labor (migrating for temporary work
outside the community).
The stock of land (Xt) is measured by area, and is allocated to
agricultural production, to livestock production when sown to
alfalfa, and to production of Z-goods when used for building. Ct
refers to land set aside for homes, plazas or sportsfields.
The last three rows represent stocks of capital goods which are used
in the current year's production — as opposed to the top three lines
which representflows,that is, quantities produced during the year, j*
represents agricultural stocks required for agricultural production
(such as seed, S ). S is the stock of animals used in all three sectors,
e.g. for motive power (i.e. oxen), reproduction (expansion of the
stock), and transportation. Finally, S3 represents the stock of tools
used in the production of goods in all three sectors.
It should be clear that each column of the table represents the
combination of elements of production necessary to produce the
corresponding good: it is a vector representing a technological
process. The four right-hand columns represent vectors of final
demand: the composition of goods consumed, accumulated and
exchanged. It will also be noticed that the community produces four
types of commodities: agricultural goods, livestock, Z-goods, and
labor for export to work temporarily elsewhere in town or country.
An empirical estimate of Table 2.1 would summarize the tech-
nological relations, the productive structure and exchange relations
in peasant communities. It would also show the specific ways in
which the peasant economy is connected to the market economy.
This estimate is presented in the next section. Before that, the
particular geographical setting of peasant communities must be
specified, for the production structure of any community will be
influenced by the characteristics of ecological levels under control,
micro-climates and so on.
Ecological levels and micro-climates

It is customary to divide Peru into three major regions: coast, sierra
and jungle (selva). This differentiation is based on the Andean
cordillera - the sierra - and coast and jungle are merely the regions
situated on either side thereof. But some geographers have proposed
alternative divisions. In particular, Pulgar Vidal (1972) insists on the
existence of eight ecological 'levels' in Peru. Taking into account the
greatest possible number of differentiating factors in the natural
environment, Pulgar Vidal distinguishes: coast,yunga, quechua, suni,
puna, cordillera, upper jungle and lower jungle. In this classification,
the region usually known as the sierra is divided into five parts.
The sierra, then, is not homogeneous. It includes a vast range of
altitudinal zones, often in close proximity to one another, which
comprise numerous types of soil and climate. As pointed out by Julio
C. Tello, 'the climatic conditions of the sierra vary from the tropical
climates of the valleys and ravines, to the arctic conditions of the high
cordillera' (quoted in Pulgar Vidal, p. 13). The clear zonal distinc-
tions made by the inhabitants of the sierra in their immediate
environment do in fact correspond to a large extent to the regional
classification proposed by Pulgar Vidal.
The five regions mentioned for the sierra define a series of
ecological levels. In terms of altitude, the junga ranges up to 2,300
meters and corresponds to the warm valleys; the quechua, the temperate
region, ranges from 2,300 to 3,500 meters; the suni is a cold region
ranging from 3,500 to 4,000 meters and the puna, the 'treeless zone',
rises up to 4,800 meters. Thereafter we reach the cordillera where
there is practically no agricultural or pastoral activity of any kind.
Table 2.2 shows the distribution of population among the eight
ecological levels in 1972. Although the coast contains 44% of the
Scope and method 9
Table 2.2 Peru: total and rural population by ecological region, 1972
Total Rural
population population
Altitude
Regions (meters) '000 0/
'000 /o
/o
Coast Under 500 5,929 43.8 1,208 17.0
Yunga 500-2,000 926 6.9 725 10.2
(500-1,000) (228) (1.7) (122) (1.7)
(1,000-1,500) (258) (1.9) (212) (3.0)
(1,500-2,000) (440) (3.3) (391) (5.5)
Quechua 2,000-3,500 4,073 30.1 3,215 45.3
(2,000-2,500) (892) (6.6) (625) (8.8)
(2,500-3,000) (1,171) (8.7) (1,012) (14.3)
(3,000-3,500) (2,010) (14.8) (1,578) (22.2)
Suni 3,500-4,000 1,325 9.8 1,101 15.5
Puna 4,000-4,500 192 1.4 124 1.7
Cordillera 4,500
and above 13 0.1 9 0.1
High jungle 500-1,000 371 2.7 280 3.9
Low jungle under 500 709 5.2 444 6.3
13,538 100.0 7,106 100.0
Note: The estimates were calculated on the basis of district population figures in the
National Population Census of 1972. However, they remain approximations because
the altitude of each district was assumed to be the same as the altitude of the district
capital. The altitudes were obtained from the Anuario Estadistico del Peru, Vol. 3,
Table 2.2.6 (Lima, Oficina Nacional de Estadisticas y Censos, n.d.) and from
inquiries at the Instituto Geografico Militar.
According to Pulgar Vidal the quechua begins at 2,300 meters and the Cordillera at
4,800. Since the intervals in our sources were 500 meters, a small and insignificant
part of the area covered by theyunga had to be included in the quechua, while the Cordillera
was defined as starting at 4,500 meters, thus impinging slightly on the puna. However,
these errors are marginal.
nation's population, 25% live in Lima. The next most important

region is the quechua, with 30% of the population. Rising from coast
to quechua, only 7% of the population are to be found in the yunga
because of its harsh environment and very steep terrain. Above the
quechua the population again declines until one passes the Cordillera
and descends the eastern slopes of the Andes to the upper jungle.
Taking the rural population alone - 53% of the nation - the heaviest
concentration is found in the quechua (45%), followed by the coast
(17%), the suni (16%) and theyunga (10%). Thus the five ecological
levels of the sierra comprise 73% of the rural and 39% of the national
population.
Each ecological level evidently exhibits different production possibi-
lities. For each zone Pulgar Vidal lists 'limit products' which cannot
be produced at higher altitudes. For thcyunga these are sugar cane and
certain varieties of fruit (avocado, lucumo, chirimoya (prickly pear),
guayava and citrus fruits). As far as the upper limits of the quechua one
finds maize, wheat, and other fruit (plums, peaches, apples and
quinces); and in the case of the suni the 'limit products' are beans,
small tubers (oca, olluco, mashua), high-altitude cereals (quinua, canihua)
and vegetables such as tarhui. These products of the suni also have a
lower limit of cultivation at the border of the suni and quechua. Finally,
only potatoes and barley can still be cultivated on the puna, and even
then only in some places and with lengthy fallow periods. Barley
cannot be cultivated lower than the quechua, but potatoes grow in all
the sierra regions.
The ecological levels are also marked by variations in the
possibilities of livestock production. For example, the cameloids
(llamas, alpacas and the like) pasture mostly on the puna, while equine
and caprine animals are most suited to the lower regions. Given the
variation in both flora and fauna, the production possibilities of non-
agricultural goods (Z-goods) are also influenced heavily by access to
ecological levels. Finally, it should be remembered that even within
each ecological level a wide variety of micro-climates is to be found
and, as a result, individual communities possess a wide variety of
resources. On the other hand, these resources differ between
communities for they have access to different ecological levels and
micro-climates, which implies that communities have different
production possibility sets.
The sample
The next chapters will present the results of research carried out in the
southern sierra of Peru, the poorest area of the sierra region in income
terms. The peasant families studied are minifundistas, possessing for
the most part fewer than five hectares of land; and they are organized
Scope and method 11
in peasant comunidades. The family (or household) and the comunidad
are therefore the units of analysis throughout the study.
Eight communities were selected from a sample universe which
contained the most important areas with peasant communities in the
southern sierra. The communities vary in their ecological characteris-
tics: Jacantaya, Culta and Ninamarca are typical of the suni zone;
Acobamba of theyunga, and the remainder are in the quechua. The only
one of the five departments in the area which is not represented is
Ayacucho, but this is not of importance since differences between
departments are minor.
The data were obtained by uniform questionnaires administered to
peasant families in all the communities. Other informants were also
interviewed, in particular persons who occupied or had occupied
positions of responsibility in the communities; these provided
invaluable information on the social and physical context of the study
which shed further light on the information provided by the families
interviewed, and also enabled us to detect when the information
given by the families was clearly beyond the bounds of local
possibilities. The research in these communities was initiated in 1976
with several visits and preliminary data collection (Figueroa, 1978).
The final stage in data collection, which was initiated in September
1978, lasted two weeks in each community and was carried out by an
average of four persons, three of whom worked in all the localities
and one of whom was usually a person from the community itself. In
cases where the family concerned possessed a faulty command of
Spanish the interviews were conducted in Quechua or Aymara. The
data collected on flows refer to the previous year ending before the
interview. The fieldwork ended in April 1979.
Table 2.3 shows the distribution of the sample families among the
eight communities. It was decided to sample about 40 families in each
community, in order to be able to carry out a statistical analysis of the
results by communities. All the families interviewed, including
replacements, were selected by random sampling. Therefore, each
sample is representative of its community. On the basis of population
data, weights were applied to the sample of families to make it
representative of the eight communities and the southern sierra. The
reader will find in each table the average values for these levels of
sample expansion as 'sample total' and 'southern sierra' respectively.
(For further details on the sample the reader is referred to Appendix I.)
Table 2.3 The sample communities
Number of
Date of families
Name study interviewed Department
Accha-Sihuina (SIH) Sept. 1978 41 Cuzco

Ninamarca (NIN) Oct. 1978 31 Cuzco
Acobamba-San Marcos (ACO) Nov. 1978 40 Huancavelica
Ancobamba (ANC) Dec. 1978 40 Apurimac
Ttiomayo (TTI) Jan. 1979 36 Cuzco
Huando (HUA) Feb. 1979 42 Huancavelica
Culta (CUL) March 1979 40 Puno
Jacantaya (JAC) April 1979 36 Puno
Total 306
3
The economic unit and economic organization
In order to study the functioning of the peasant economy the basic

economic unit of such a system must be clearly established. Who has
the control of resources ? Who makes the decisions on the allocation
of these resources ? Who makes the decisions on consumption and
accumulation? This chapter is devoted to answering such questions.
Size of family and labor force

In this study, the family is defined as all those persons who habitually
live in the same house. The criterion of residence poses some
difficulties on account of permanent and temporary migration, and
the rule adopted was to consider as resident any person who lived in
the house a total of six months or more in the preceding year.
The average family size varies between 4.2 and five members in the
communities (Table 3.1). The variation in these average sizes seems
to bear no particular relation to the community's ecological level.
Although the value of the standard deviation between family size
within each community is high, the differences between communities
seem to be small. These data show that it is difficult to identify a
'typical' peasant family size, at least without taking into account the
stage of the household cycle reached by each one.
This study uses two definitions of labor force. The first, designed
to take into account the role of children in production, and thus to
show the family's productive capacity, includes all members of six
years or more of age. The second includes persons of 18 years or more
of age and measures the 'adult labor force'. Starting from 18 years of
age, people can carry out adult tasks and they can also obtain identity
cards which enable them to move about more easily in the labor
market. Neither definition excludes elderly people since there is no
formal 'retirement age' - like the children, they too participate in
13
Table 3.1 Si^e of family and labor force (number of persons)
Labor force
Size of family Total Adult
Community S S S
Jacantaya 4.6 (2.6) 3.9 (2.2) 2.6 (1.3)

Culta 4.5 (2.0) 3.6 (1.7) 2.2 (0.7)
Ninamarca 4.7 (1.4) 3.6 (1.3) 2.2 (0.5)
Ancobamba 4.7 (2.5) 3.7 (1.7) 2.1 (0.7)
Ttiomayo 5.0 (2.2) 3.9 (1.6) 2.3 (0.9)
Sihuina 4.3 (1.9) 3.6 (1.5) 2.3 (0.8)
Huando 4.8 (2.6) 3.7 (1.8) 2.1 (0.8)
Acobamba 4.2 (1.7) 3.4 (1.4) 2.1 (0.6)
Sample total 4.6 3.7 2.2
Southern sierra 4.5 3.6 2.2
Note: X= average; S = standard deviation.
production. Table 3.1 also shows that, under the first definition, 'total
labor force' varies between 3.4 and 3.9 persons per family, while
'adult labor force' varies between 2.1 and 2.6. Once again, there is a
wide dispersion in the measure of total labor force, but this dispersion
is reduced for the measure of adult labor force: the vast majority of
the families have between two and three adult members, and in this
respect one can speak of a typical peasant family.
A further, equally important, observation is that the peasant family
is typically a nuclear family. It is extremely rare to find a household
with more than one married couple, and even in Culta, where 20% of
the households had four or more adult members, only a very small
proportion of these included two married couples. Hence, the
concepts of 'family' and 'household' coincide in the peasant com-
munities and can be used interchangeably.
Land: uses, tenure and fragmentation

In the peasant communities two types of land may be distinguished:
cultivable and natural pasture. Whereas pasture land can be used for
only one purpose (livestock production), cultivable land can have
alternative uses: it can be planted to crops, it can be used to cultivate
The economic unit and economic organisation 15
pastures, or it can be left uncultivated. If it is uncultivated it can still
serve as natural pasture. In this sense, uncultivated land does not
imply idle land; it means that land leaves the agricultural process and
enters into livestock production.
Pasture land is always communal property and is collectively
utilized. Each family has free access to it and derives economic benefit
in proportion to the family's holding of livestock. The tenure system
of arable land varies from community to community; in some there is
overall private ownership, and each family uses the land according to
its own interest; whereas in others, part of the arable land is used
under a system of collective rotation. This is the case for the low
quality, marginal land that must be left uncultivated after two to three
years of cultivation. If the families' individual plots are not protected
with fences, the areas for cultivation must be decided collectively so as
to keep animals off that area. The utilization of marginal lands for
crop and livestock production requires a collective rotation of land.
The land where rotation is done collectively is called laymi and is
divided into several areas for the annual rotation.1 Needless to say,
once the area of cultivation is decided, the decision as to what to
produce and how to produce is made by each family individually.
Table 3.2 shows the distribution of arable land between pasture
and crops. The unit of measurement is the number of plots rather than
hectares because of the imprecision of replies to questions concerning
the size of plots; moreover, the enormous number of plots belonging
to each family made the estimation of their aggregate size impossible.
Estimation in terms of number of plots is an adequate approximation
because the use of the plots is independent of their size. The
proportion of land assigned to agriculture, rather than pasture, varies
among the communities between 46% and 90%.
In six of the eight communities there exists the laymi system.
Jacantaya does not possess this system because it is located on the
shores of Lake Titicaca and its land is cultivated in terraces. The
terraces are small and possess clear boundaries, so that a family can
leave a plot fallow and use it for pasture even while the neighboring
parcel is being cultivated. The other community which does not have
a laymi system is Huando. This community did have such a system
until some 20 years ago, and its disappearance is mainly due to the
introduction of modern inputs, mainly fertilizers, which have raised
the productivity of the land. Table 3.2 shows the importance of the
laymi system in the other communities: between 21% and 71% of the
Table 3.2 Distribution of cultivable parcels between cultivated and fallow
land (number of parcels per family)
Community Type of tenure Cultivated Fallow Total
Jacantaya Private 19.5 15.1 34.6
Culta Private 11.3 0.6 11.9

Laymi 27.6 44.9 72.5
Total 38.9 45.5 84.4
Ninamarca Private 2.3 1.4 3.7

Laymi 4.7 3.8 8.5
Total 7.0 5.2 12.2
Ancobamba Private 6.5 0.9 7.4

Laymi 1.8 n.a. n.a.
Total 8.3 n.a. n.a.
Ttiomayo Private 2.9 0 2.9

Laymi 3.1 2.7 5.8
Total 6.0 2.7 8.7
Huando Private 8.5 1.0 9.5
Acobamba Private 2.5 0.3 2.8

Laymi 1.6 4.1 5.7
Total 4.1 4.4 8.5
Note: Data for Sihuina were not available.
land cultivated during the years of our research was under this system.
Its qualitative importance is further underlined by the observation
that it was in these lands that basic food products such as potatoes
were cultivated.
Table 3.2 also shows the extreme degree of fragmentation of
cultivable holdings, the number of holdings per family ranging from
nine to 84. In the extreme cases of Culta and Jacantaya parcels are
measured in furrows, which are at most 20 meters long. One
frequently hears a peasant saying that he has 'one plot of four furrows,
another of six furrows' and so on. A family's plots are generally
distributed across the different ecological levels to which the
community has access; however, there are often several plots even
within a given ecological level belonging to the same family.
This fragmentation can be explained in part by the desire of
producers to avoid erosion, when the land is sloped. But the existence
of fragmentation on flat land shows that there are also demographic
and economic factors involved. Families acquire land upon their
formation via inter-generational transfer, and these new families must
also gain access to land on diflferent ecological levels if they are to
produce a range of products. The need of a Vertical control of
ecology' by each family implies a process of land fragmentation in the
inter-generational transfer. Although this need would seem to reflect
a strategy aiming at a certain degree of family self-sufficiency, such a
hypothesis does not take into account the wide variations in micro-
climates, even within the same ecological level, which create
variations in yields as a result of frost, hail and floods. Thus the spread
of parcels would seem to reflect also a form of risk-averse economic
behavior, which would certainly be consistent with the low level of
the producers' incomes.
The number of crops sown and parcels cultivated by each family is
shown in Table 3.3. The crops involved are: potatoes, barley, oca,
olluco, mashua, quinoa, canihua, tarhui, broad beans, wheat, peas
and kidney beans. The average number of parcels cultivated per
family varies between communities from four to 39, and in all the
communities there are more parcels than crops, implying that families
grow one crop in more than one parcel. Thus in Jacantaya eight crops
are sown on an average of 20 parcels. Only in Acobamba is there near-
equality between the average number of parcels and the average
number of crops. The diversification of crop-location is even greater
than this indicates, since there is much inter-cropping. This offers
further support to the risk-aversion hypothesis.
Capital goods
The capital stock of the peasant economy consists of three types of
goods which are used in the production process: agricultural goods
(mainly seed), animals and tools. Seed is generally obtained from the
same crops; they are rarely bought by the peasant families. Consistent
with its portfolio of crops, each family must hold a portfolio of seeds.
The stock of seed per family was not estimated but it clearly varies
with the amount of cultivated land.
Animals can be divided into three types, depending on their place
Table 3.3 Number ofparcels cultivated by number of crops (% of families)
Communities
of crops JAC CUL NIN ANC TTI HUA ACO
0 7.1 2.6
1 5.0 0 0
2 3.2 0 2.4 7.7
3 9.7 2.5 27.8 2.4 28.2
4 2.4 5.1 9.7 10.0 36.1 7.1 25.6
5 2.8 7.7 19.4 7.5 19.4 11.9 20.5
6 13.9 12.8 29.0 22.5 11.1 26.2 2.6
7 22.2 17.9 19.4 12.5 5.6 16.7 7.7
8 16.7 20.5 9.7 20.0 19.0 2.6
9 11.1 20.5 15.0 2.4 0
10 13.9 10.3 2.5 2.4 2.6
11 13.9 5.1 0 2.4
12 2.8 2.5
Total (%) 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Number of
crops
Average 8.2 7.7 5.6 6.6 4.3 6.0 4.2
Standard
dev. 2.0 1.8 1.6 2.3 1.2 2.4 1.8
Number of
parcels
Average 19.5 38.9 7.1 8.2 6.1 8.5 4.1
Standard
dev. 18.3 19.3 2.7 3.5 2.4 5.0 2.0
Note: Data for Sihuina were not available.
in the production process. One type is used for consumption of its

products, in the form of meat, wool, milk and eggs. This includes
sheep, alpaca, pigs, goats, and minor domestic animals such as
chickens and guinea-pigs. The second type of animal is used in
productive services, such as horses, donkeys, mules and llamas. Cattle
are considered the third type for they provide both consumption and
productive services. Each family's mix of these types will obviously
influence the overall structure of its activities.
It will be seen from Table 3.4 that, on average, the peasant family
has very few animals: the typical family possesses 7 sheep, 1 pig, 3
chickens, 4 guinea-pigs and 2 head of cattle. (These figures appear as
weighted averages in the last column of the table.) However, the
composition of the stock varies substantially between communities,
largely on account of ecological differences. Goats are not found in
the higher communities, nor are llamas or alpacas found in the lower
ones. But there remain other differences even in those animals which
can be kept at all these altitudes, and these have an economic
explanation. The communities with most cattle and sheep are
Ancobamba, Culta, Ttiomayo and Ninamarca, and those with the
least are Huando and Acobamba. In Acobamba the land slopes
steeply, and there is thus the constant risk of falls, while in Huando
there is an ecological explanation as well as an economic one: the
community has no high pasture lands, and therefore does not control
enough ecological levels to practice both agriculture and cattle-
raising at the same time.
Variation in the endowment of horses is largely explained by
distance from a road. Ancobamba and Sihuina are the communities
furthest from the road in the sample and they have the higher average
of horses per family. The use of oxen in ploughing is limited when
deeper ploughing, beyond their capacity, is required, and when land
Table 3.4 Animal stocks: number of animals per family
Communities
Sample
Animals JAC CUL NIN ANC TTI SIH HUA ACO total
Sheep 6.39 13.08 12.32 4.85 6.97 5.56 3.74 4.03 6.95
Pigs 0.44 1.74 2.42 1.35 1.92 0.61 0.71 0.67 1.09
Goats 0.00 0.00 0.00 1.85 0.00 0.10 0.91 0.15
Alpaca 0.00 2.67 0.00 0.00 0.00 0.00 0.00 0.00
Chickens 1.64 2.92 1.94 1.78 2.67 2.93 2.17 3.85 2.51
Guinea-
pigs 0.89 0.59 5.45 5.95 8.62 4.22 2.95 3.36 3.94
Horses 0.14 1.40 1.13 2.68 0.45 1.71 1.05 0.63 1.17
Llamas 0.00 0.05 0.05 0.00 0.00 0.00 0.00 0.00
Oxen 0.14 0.75 0.94 0.53 1.03 0.54 0.57 0.10 0.56
Total
cattle 1.33 2.08 3.03 3.65 3.47 2.39 1.00 1.59 2.29
is too steep or is cultivated in terraces. It is for these latter reasons that
there are few oxen in Jacantaya and Acobamba. But in general, it can
be seen that animal draught power is limited, and that to a significant
degree the peasant economy depends on human energy. This brings
us to its implements. These are of two types: those used in
agriculture, such as oxploughs, foot-ploughs (chakita/lla), hoes and
spades; and those used in the production of non-agricultural goods
(Z-goods), especially in construction, such as picks and shovels, and
sewing-machines.
Table 3.5 shows that the typical peasant family uses a very small
range of tools requiring human energy: between 1 and 2 foot-
ploughs, 3 hoes and 2 picks. The variation in the number of foot-
ploughs per family as between different communities is largely
explained by differences in the quality of land: in Jacantaya oxen
cannot be used because of terrace cultivation, and in Ancobamba the
poor quality of the soil requires deeper ploughing than can be
obtained with oxen. The variations in hoes and spades seem to be
related to differences between families of varying wealth, while
sewing-machines are a preliminary indication of the wealth of a
community, and above all of its modernization. Huando thus appears
as the most modern of these communities, followed by the two
communities in the department of Puno: Culta and Jacantaya.
The economic unit and economic organization in

peasant communities
The exposition hitherto has shown that, typically, the peasant of the
southern sierra family has an adult labor force of about two persons;
Table 3.5 Stocks of implements per family
Communities
Sample
Implements JAC CUL NIN ANC TTI SIH HUA ACO total
Foot-
ploughs 2.71 0.87 1.13 1.90 1.14 1.56 1.95 1.53 1.57
Hoes 2.83 4.08 1.71 2.63 1.94 2.44 3.71 3.85 3.12
Sewing-
machines 0.31 0.41 0.14 0.08 0.17 0.12 0.55 0.10 0.28
Picks and
shovels 2.29 2.39 2.19 2.19 3.06 1.86 1.44 3.22 2.40
two or three hectares of cultivable land divided into a number of
parcels (between nine and 84) and access to a variable amount of
pasture land under collective use; its animal stock consists of 2 head
of cattle, 7 sheep, 1 pig, 1 horse, 2 or 3 chickens and 4 guinea-pigs; its
stock of tools consists of 1 foot-plough, 3 hoes and 2 picks. The
family's resource endowment is used in agricultural and livestock
production and in the production of non-agricultural goods and
services (Z-goods). In addition to the self-employment of its
resources, the family exchanges part of them particularly family labor.
Apart from the adult labor force, resource endowment per family
varies significantly between communities, especially in the case of
land. The land endowment depends on the community's access to
different ecological levels and on the topography of its land.
Ecological characteristics define production possibilities, and topog-
raphy is an important determinant of the technology in use.
In terms of control of resources it has been shown that individual
families own and control the arable land and the capital goods in the
peasant communities. Pastures are collectively owned and used but
animals belong to individual families. On the other hand, the
allocation of resources is also decided by individual families. The
allocation of the family labor to agriculture, livestock production, Z-
goods and labor exchange is a family decision. What to produce, in
which plots and with what mix of inputs are also family decisions.
Moreover, the risk, gains and losses in production are assumed by the
individual family. In terms of control of resources, capacity to decide
on the allocation of those resources to alternative uses, and
appropriation of the net output produced, the unit of production is,
clearly, the peasant family.
Nevertheless, the peasant family does not operate in isolation from
the rest of the community. In the productive process the family
establishes relations with the rest of the families of the community.
When a family has an excess demand or supply of a good or of labor
(which is frequent), there are up to three possible ways of achieving
equilibrium: reciprocity, barter and monetary exchange. Reciprocity
is a form of exchange in which the lending of a service is paid back
with the same service in a not very distant time period. It is a special
form of loan, without interest. If three days of labor are received,
three days of labor are returned; if two days' use of an ox are received,
the same amount of services must be returned. It is clear that this
exchange has a technological justification, since with this reciprocity
it is possible to obtain optimum factor proportions in production.
The most common form of exchange by reciprocity is the exchange
of labor. Family labor may not be sufficient to put a roof on a house,
or to carry out the labors of planting or harvesting. This system,
referred to as ayni or minka, adopts in many cases festive forms where
goods (determined culturally) are consumed. In this case production
and consumption are inseparable in the production process, and the
criterion of efficiency, that is, of producing at the minimum economic
cost, is not applicable. The criterion of efficiency establishes, it
appears, the lower limit to the factor proportions required tech-
nologically in an activity.
The existence of barter, that is, the exchange of a good or service
for another good or service, indicates that the excess demands and
supplies of the families are of different magnitudes. The case of
reciprocity is given when these quantities are of the same magnitude;
thus in a community with few differences in the ownership of
resources, reciprocity would dominate. Many families derive income
from the exchange of the services of their work animals, tools, land
and labor. To this we would have to add credit. These productive
services are paid in different ways, sometimes with food or non-food
goods, sometimes with other types of productive services. Finally,
there is exchange of productive services for money. This form of
exchange clearly becomes necessary given the limitations of barter-
ing, that is to say, its dependence on the complementarity between
two parties' supply and demand of goods and services. In all of the
communities studied there exists a market for productive services,
with transparent monetary prices.
Once it has been shown that the family is the unit of production in
the peasant communities, it will be much easier to show that the
family is also the unit of consumption. In fact, the basket of
consumption of goods and services is determined by the family. The
actions taken to acquire that basket, i.e. consumption of own-
production or exchange, are also family decisions. The fact that
capital goods are owned by individual families is sufficient to show
that capital accumulation is a family decision too. Thus the family is
the unit of production and consumption - that is, the economic
unit - in the peasant communities.
However, not everything is under the control and decision of
individual families in the peasant communities. Some resources are
subject to collective control — the case of pasture land has already
been mentioned. This case also includes the use of natural pastures
grown in the rotated land (laymi) when it is in the fallow period. In
some communities, even the stubble in the arable land of individual
families is open to all the community as collective pastures. Water is
another resource which is under collective control and is managed
through collective decisions.
As far as land resources are concerned, only in Ninamarca and
Sihuina was found communal arable land. In Ninamarca there exists a
plot of land which is cultivated collectively; the net income it
generates is used to finance the construction of public buildings (like
schools) or other public works. In Sihuina the communal land is
given mainly to families in charge of the communal fiesta in order to
help them with the expenses of such celebrations. There were no
communal lands that were used to distribute to new families.2
The peasant community also imposes some economic obligations
on individual families. One of these is the labor time that every family
must supply for public works. It is through this system that a
community can increase and maintain its social infrastructure (roads,
canals, buildings). In economic terms, this is equivalent to a local tax
system that is collected in labor time levied equally upon everyone. In
the eight communities studied the number of man-days per year that a
family had to supply to the community was between four and 20.
The existence of resources under collective control and decision
sets some limits to the economic decisions of the individual family.
However, those constraints do not imply that the peasant family
could not be the economic unit; they do imply that these units operate
within particular economic and institutional frameworks. To eluc-
idate the particular circumstances in which the peasant economy
operates in Peru today will be the object of the following chapters.
Thus, the next chapter explores the quantitative aspects of the
production and exchange processes in peasant communities.
4
Production and exchange
Intersectoral flows
Table 4.1 shows the intersectoral flows in six of the eight com-
munities studied; that is, it replaces the symbols of Table 2.1 with
empirical data, with one modification: the production of each
separate sector of intermediate goods — the first three columns of
Table 2.1 -has been consolidated into one column. This has been
done for the sake of simplicity and also because it is often difficult to
separate the production of intermediate goods by sector. The N
column refers to non-monetary barter transactions, and the M
column to monetary transactions. The former principally measures
transactions with other communities, the latter transactions with the
rest of the economy, above all the urban economy. Two of the
communities — Ttiomayo and Sihuina — have been omitted on ac-
count of the incompleteness of the data, but the other six are
sufficiently well distributed among ecological levels for the repre-
sentativeness of the sample not to be affected significantly. Of course,
whenever only these six communities have been used in tabulations,
the weights of the sampled families have been changed accordingly to
get the expanded values for 'sample total' and 'southern sierra'.
The reader will note that Table 4.1 contains certain identities. The
figure at the bottom of column 1 is the same as the A + P + Zin the
last column. The former represents the value of production viewed
from the side of the cost of production whereas the latter is the value
of production by destination (deliveries), and both must be equal.
Because of this identity, self-employment (row 20) was estimated by
difference. The transactions in kind have been transformed into
monetary values by applying average market prices observed in the
monetary transactions. Obviously, the valuation of imports to the
community must be equal to the valuation of exports when exchange
24
Table 4.1 Intersectoral relations in peasant communities (soles per yeary per family)
Jacantaya
A +P+Z
Inter- C I N M
mediate Consump- Invest- Exports Exports
goods tion ment in kind monetized Total
(1) (2) (3) (4) (5) (6)
A (Agriculture) (1) 3,700 14,896 0 2,854 5,401 26,851

P1 (Derived products) (2) 990 1,680 0 0 107 2,777
P2 (Consumption) (3) 0 1,151 0 7 325 1,483
P3 (Services) (4) 0 0 0 0 0 0
P4 (Cattle) (5) 0 594 0 0 5,603 6,197
P5 (Other) (6) 0 0 0 0 0 0
P (Livestock) (7) 990 3,425 0 7 6,035 10,457
Z (Manuf.) (8) 0 3,825 0 0 482 4,307

Z2 (Trade) (9) 0 0 0 0 3,213 3,213
Z3 (Other) (10) 0 5,685 0 0 9,682 15,367
Z (Z-goods) (11) 0 9,510 0 0 13,377 22,887
A + P+ Z (12) 4,690 27,831 0 2,861 24,813 60,195
N (Imports in kind) (13) 1,000 1,861 0 2.861
M (Imports in money) (14) 4,836 36,638 2,584 44,058
N+M+A + P + Z (15) 10,526 66,330 2,584

Table 4.1 (cont.)
Jacantaya
A +P+Z C / N M Total
(1) (2) (3) (4) (5) (6)
Wages - N (16) 41 0 0
Wages - M (17) 4,644 0 12,772
Rents - N (18) 0
Rents - M (19) 72
Self-employment (20) 44,912
Total 60,195 2,861 37,585
Net transfer (21) 0 6,473
Cult a
A (Agriculture) (1) 4,966 19,866 0 287 429 25,548

P (Derived products) (2) 2,145 3,066 0 0 1,502 6,713
P2 (Consumption) (3) 0 3,291 0 0 2,541 5,832
P3 (Services) (4) 0 0 0 0 316 316
P4 (Cattle) (5) 0 594 0 25 20,100 20,719
P5 (Other) (6) 0 0 0 0 408 408
P (Livestock) (7) 2,145 6,951 0 25 24,867 33,988
Z1 (Manuf.) (8) 0 4,358 0 0 0 4,358
Z2 (Trade) (9) 0 0 0 0 3,357 3,357
Z 3 (Other) (10) 0 0 0 0 1,715 1,715
Z (Z-goods) (11) 0 4,358 0 0 5,072 9,430
A +P+Z (12) 7,111 31,175 0 312 30,368 68,966
N (Imports in
kind) (13) 44 243 25 312
M (Imports in
money) (14) 2,707 33,812 4,867 41,386
N+M+A+P+Z (15) 9,862 65,230 4,892
Wages - N (16) 1,000 0 0
Wages - M (17) 2,052 0 9,403
Rents - N (18) 19
Rents - M (19) 298
Total 68,966 312 39,771
Net transfer (21) 1,615
Ninamarca
A (Agriculture) (1) 2,345 9,380 0 837 12,685 25,247

P± (Derived products) (2) 1,980 2,100 0 13 0 4,093
P 2 (Consumption) (3) 0 9,644 0 77 2,594 12,315
P 3 (Services) (4) 0 0 0 0 0 0
P 4 (Cattle) (5) 0 2,560 0 742 3,000 6,302
P 5 (Other) (6) 0 0 0 0 680 680
P (Livestock) 0) 1,980 14,304 0 832 6,274 23,390
Zx (Manuf.) (8) 0 2,900 0 116 448 3,464
Z o (Trade) (9) 0 0 0 0 1,232 1,232
Table 4.1 {cont.)
Ninamarca
A+ P+ Z C I N M Total
(1) (2) (3) (4) (5) (6)
Z3 (Other) (10) 0 0 0 0 32 32
Z (Z-goods) (11) 0 2,900 0 116 1,712 4,728
A + P+ Z (12) 4,325 26,584 0 1,785 20,671 53,365
N (Imports in
kind) (13) 125 601 1,059 1,785
M (Imports in
money) (14) 1,755 17,597 1,461 20,813
N+M+A+P+Z (15) 6,205 44,782 2,520

Wages - N (16) 137 0
Wages - M (17) 908 142
Rents - N (18) 0
Rents - M (19) 0
Total 53,365 1,785 20,813
Net transfer (21) 0

Ancobamba
A (Agriculture) (1) 8,777 35,046 351 356 44,530
Pj (Derived products) (2) 825 1,869 0 0 2,694

P2 (Consumption) (3) 0 2,202 98 186 2,486
P3 (Services) (4) 0 0 0 925 925
P4 (Cattle) (5) 0 863 0 5,324 6,187
P5 (Other) (6) 0 0 5 360 365
P (Livestock) (7) 825 4,934 103 6,795 12,657
Zx (Manuf.) (8) 0 531 5 209 745
Z2 (Trade) (9) 0 0 0 1,843 1,843
Z3 (Other) (10) 0 0 5 84 89
Z (Z-goods) (11) 0 531 10 2,136 2,677
.4 + P + Z (12) 9,602 40,511 0 464 9,287 59,864
JSl (Imports in kind) (13) 188 63 213 464
M (Imports in money) (14) 1,038 17,619 910 19,567
N+M+A+P+Z (15) 10,828 58,193 1,123
Wages - N (16) 60 0 0
Wages - M (17) 1,162 0 8,458
Rents - N (18) 60
Rents - M (19) 17
Total 59,864 464 17,745
Net transfer (21) 1,822
Huando
A (Agriculture) (1) 7,821 32,210 0 134 10,050 50,215

Px (Derived products) (2) 2,000 3,000 0 0 293 5,293
Table 4.1 (cont.)
Huando
A +P+Z C I N M Total
(1) (2) (3) (4) (5) (6)
P2 (Consumption) (3) 0 2,710 0 73 2,898 5,681

P 3 (Services) (4) 0 0 0 0 352 352
P 4 (Cattle) (5) 0 1,000 0 0 3,642 4,642
P 5 (Other) (6) 0 0 0 0 0 0
P (Livestock) (7) 2,000 6,710 0 73 7,185 15,968
Zx (Manuf.) (8) 0 0 0 0 0 0
Z 2 (Trade) (9) 0 0 0 21 15,646 15,667
Z 3 (Other) (10) 0 0 0 0 9,083 9,083
Z (Z-goods) (11) 0 0 0 21 24,729 24,750
A +P+Z (12) 9,821 38,920 0 228 41,964 90,933
N (Imports in kind) (13) 0 201 27 228
M (Imports in money) (14) 4,114 46,421 2,536 — — 53,071
N+M+A+P+Z (15) 13,935 85,542 2,563
Wages - N (16) 177 0
Wages - M (17) 8,506 4,520
Rents - N (18) 1,122
Rents - M (19) 371
Total 90,933 228 46,484
Net transfer (21) 0 6,587
Acobamba
A (Agriculture) (1) 6,222 24,888 0 362 4,557 36,029

Pt (Derived products) (2) 2,000 6,000 0 0 0 8,000
P 2 (Consumption) (3) 0 2,240 0 161 1,048 3,449
P 3 (Services) (4) 0 0 0 0 0 0
P 4 (Cattle) (5) 0 1,000 0 0 1,538 2,538
P 5 (Other) (6) 0 0 0 0 92 92
P (Livestock) (7) 2,000 9,240 0 161 2,678 14,079
Zx (Manuf.) (8) 0 117 0 0 5 122
Z 2 (Trade) (9) 0 0 0 0 15,282 15,282
Z 3 (Other) (10) 0 0 0 0 410 410
Z (Z-goods) (11) 0 117 0 0 15,697 15,814
A + P+Z (12) 8,222 34,245 0 523 22,932 65,922
N (Imports in kind) (13) 400 105 18 523
M (Imports in money) (14) 1,138 24,868 1,331 27,337
N+M+A+P+Z (15) 9,760 59,218 1,349
Wages - N (16) 715 0 0
Wages - M (17) 10,335 0 3,841
Rents - N (18) 425
Rents - M (19) 0
Total 65,922 523 26,773
Net transfer (21) 0 564
is in kind. This explains the equality between the value at the bottom
of column 4 and at the end of row 13. For the monetary exchange the
difference of total value of exports and total value of imports is due to
net transfers that families in the community receive from relatives
living in the city. Therefore, the last two figures at the bottom of
column 5 must be equal to the figure at the end of row 14.
The entries in Table 4.1 are all monetary values: soles per year, per
family. Therefore, the values for all communities are comparable
since differences in population have been eliminated. To be sure, the
total annual values for a community can be found by multiplying the
entries in Table 4.1 by the number of families in the community
shown in Table 2.3, and hence Table 4.1 shows the production and
exchange structure of the communities.
Intermediate goods
There are three types of intermediate good (defined according to
source) in the communities: domestic goods, derived from each
family's own production; goods imported from other communities;
and goods imported from the rest of the economy. Table 4.1 shows
that between 45% and 90% of intermediate goods are produced by
the families themselves, and almost all the remainder are imported
from the urban economy. Within the category of domestically
produced intermediate goods, agricultural goods figure most pro-
minently, and among them the principal item is animal feed. The main
intermediate product derived from animals is sheep's wool, while the
Z sector produces very few intermediate goods. The production of
animal feed by the agricultural sector, and of wool for textiles by the
animal sector constitute the main technological relationships between
the three sectors A, P and Z.
Intermediate goods imported from other communities are few.
The table provides estimates for a monetary value for wool, salt,
wood, straw and pasture. Wood and straw are used mostly for the
construction of houses, and there are a few cases of families renting
pasture or stubble for animal feed in neighboring communities. Salt is
also fed to animals.
Tables 4.1 and 4.2 show that intermediate goods imported from
urban economy consist almost entirely of agricultural inputs:
fertilizer and pesticides.1 Table 4.2 also shows that there are wide
variations in the communities' use of modern inputs. Leaving aside
Production and exchange 33
Table 4.2 Average annual expenditure on modern inputs (soles peryear per
family)
Animal
Community Fertilizer Pesticides Seed vaccine Total
JAC 930 622 546 57 2,155

(75)° (64) (42) (81)
CUL 601 505 839 90 2,035
(51) (58) (53) (90)
NIN 404 832 358 40 1,634
(23) (93) (13)
ANC 0 30 119 40 189
(0) (2) (15) (7)
TTI 184 0 124 6 314
(14) (0) (10)
SIH 15 81 179 9 284
(2) (34) (15)
HUA 2,440 946 161 73 3,620
(54) (28) (17) (15)
ACO 364 154 475 8 1,001
(8) (5) (23) (15)
a
Figures in parentheses = percentage of families purchasing modern inputs.
seeds (which are hardly ever improved varieties) and vaccine for
animals (which are an insignificant item in terms of cost), we see that
communities such as Ancobamba, Ttiomayo and Sihuina hardly use
any fertilizer or pesticides at all. Even in those communities where
expenditure on these is relatively high, such as Jacantaya, Culta and
Huando - which, as we shall see, are the least poor of all the
communities sampled — this is accounted for by a small number of
families within them. Of course, the expenditure data do not take into
account deficiencies in the method of application of these inputs,
which is carried out without any technical assistance. These quanti-
tative and qualitative indications, when combined with the data on
capital stocks presented earlier, illustrate the overwhelmingly
traditional character of the technology employed in all the
communities.
Final demand
The production of the A, P and Z sectors goes to consumption
('own-consumption') and export, since accumulation is almost non-
existent (Table 4.1, column 3). In no case does investment exceed 8%
offinaldemand, and it goes as low as 2%, with an overall total of 4%.
Clearly, these communities constitute stagnant economies. They are
unable to generate any investment from their own resources, and
what little investment they do carry out is done with imported goods
(column 3, lines 13 and 14). The main capital goods imported are
cattle, sheep, chickens and tools. The peasant economy does not
appear to have sufficient productive capacity to generate its own
surplus in livestock, and although the interviews contained no
direct question concerning changes in the family's stock of animals, it
was generally agreed that the community's stock was not increasing.
On the contrary, it appeared to be declining, due to disease, accidents
and thefts. Purchases of livestock, therefore, appear to be merely a
replacement of stock rather than its increase. Tools are imported
largely because there is no domestic industry for the production of
these capital goods. In some cases one finds small workshops in local
towns which both carry out repairs and 'manufacture' tools with
discarded iron from the cities.
Exports are mainly monetary transactions, and only a small
fraction is accounted for by transactions with other communities
(columns 4 and 5). The distribution of production between own-
consumption and export is almost half-and-half, except in
Ancobamba, which is the furthest from a road and where own-
consumption is four times as large as exports. Goods exported are
exchanged for intermediate goods, for consumer goods and for
capital goods (lines 13 and 14).
Own-consumption (column 2, line 12) derives mostly from
agricultural goods, followed by animal and their derivatives, with
only a small proportion provided by Z-goods. The table also shows
that a higher proportion of agricultural production goes to own-
consumption than from animal or Z-good production. Agricultural
production is, indeed, mainly for own-consumption; the same can be
said of animal derivatives (milk and eggs), sheep and pigs, domestic
animals such as guinea-pigs and chickens, and, among Z-goods,
woolens and other clothes.
The two activities which are clearly undertaken for the purpose of
export are cattle-rearing and trade. Cattle are a luxury good at these
income levels, and peasant producers prefer to sell their cattle and use
the money to buy other consumer goods, rather than consume them
directly. The rare occasions when they are consumed arise through
commitments imposed at the time of community fiestas.
Trade (line 9) is mainly carried out between the community and the
outside, which is why trade is mainly connected with imports and
exports. In fact, it is best interpreted as an export of services, selling
produce such as cattle outside and bringing products from the cities
or from nearby communities. This last is carried out mainly by people
attending the weekly markets in a community, including both traders
who take community production to the cities and peasants from
neighboring communities. Of the eight communities studied, only
Sihuina, Huando and Acobamba have weekly markets of their own.
The external sector: monetized exchange

The analysis of monetized imports and exports can draw on data for
all eight communities, since data on this subject were collected in the
communities excluded from Table 4.1. Table 4.3 shows the structure
of exports, with trade presented separately from Z-goods, so that
exports of goods are distinguished from exports of services;
Table 4.3 Structure of monetary exports (%)
Agri-
cultural Live-
Community goods stock Z-goods Trade Labor Total
JAC 14.3 16.1 27.1 8.5 34.0 100.0

CUL 1.1 62.5 4.3 8.4 23.7 100.0
NIN 60.9 30.1 2.1 6.1 0.8 100.0
ANC 2.0 38.3 1.6 10.4 47.7 100.0
TTI 33.0 27.4 0.4 0 39.2 100.0
SIH 11.4 27.7 21.0 32.3 7.6 100.0
HUA 21.6 15.5 19.5 33.7 9.7 100.0
ACO 17.0 10.0 0.1 58.6 14.3 100.0
Sample
total 13.5 29.2 37.4 19.9 100.0
Southern
sierra 14.5 25.2 41.1 19.2 100.0
moreover, income from temporary migrations is added, as exports of
labor.
Exports
The first point to emerge from Table 4.3 is the diversified character of
the communities' exports: in no community does one item account
for more than 63% of total export income, and in four of them one
item accounts for 50% or more: cattle in Culta (63%), agricultural
products in Ninamarca (61%), trade in Acobamba (59%) and
temporary migration in Ancobamba (48%). In three of the com-
munities, 80% of export income is accounted for by two items, in
four of them by three items, and in one of them by four items. Given
that each item itself constitutes a range of products, the diversified
character of their exports becomes evident.
Exports vary widely both within and between communities. If one
looks at those sources which account for 33% or more of each
community's exports, one finds that agriculture is important in
Ninamarca and Ttiomayo, livestock in Culta and Ancobamba, and
temporary migration in Jacantaya, Ancobamba and Ttiomayo. Only
Z-goods do not reach this proportion in any community, although
they account for 27% in Jacantaya, 21% in Sihuina and 20% in
Huando.
The specific features of exports vary from community to com-
munity. Jacantaya exports almost all the goods and services which are
included in the study. In agriculture, due to a favorable micro-
climate, it exports mostly green vegetables which are sold at market in
a nearby town (Huancane) and are also exchanged directly with
neighboring^/^ communities. Income from animal exports consists
mainly of cattle. Z-goods exported are mostly fish (mainly the hispi)
from Lake Titicaca sold in Huancane. Trading income is also derived
from dealing in green vegetables.
In Culta, where agriculture is mostly for own-consumption, the
main exports are animals, above all cattle. Apart from raising their
own, many families buy cattle for fattening and sell them after a few
months, financing this activity with their own funds and with loans
from the Banco Agrario, which usually amount to about 20,000 soles.
The most common pattern is for people to buy bulls in September and
October and to sell them in April or May. In this way, they take
advantage of the time when pasture is most abundant and also use the
bulls for ploughing, since at the same time the land is being prepared
for sowing. The bank loans are generally for a year. The use and resale
activity generates a substantial revenue; this enables the peasant to
repay the loan and earn a considerable income. Trading income in this
community is also derived from cattle, and the income from fattening
was divided equally between trade (Z ) and income from animals
(p 4 ).
Ninamarca exports a range of products: potatoes, barley for
consumption, barley for beer, and oats. Sales of barley to the beer
factory in Cuzco, or to intermediaries who then sell to the factory, are
slightly more important than other exported products. Animal sales
include sheep as well as cattle. Ancobamba exports cattle above all.
The community possesses quite a lot of pasture land and cattle are
their main animal product, reared for export. In Ttiomayo, maize
accounts for almost three-quarters of agricultural exports, while pigs
and cattle account for most animal exports.
Sihuina, in contrast, exports the entire range of its agricultural
products, while cattle predominate among its animal exports. It also
exports Z-goods in the form of food, principally bread sold at its
weekly market. The community earns income from trade by taking
valley products — maize, wheat, coca — to the puna where they are
exchanged for sheep's meat and wool, which in their turn are sold in
community markets both in Sihuina and elsewhere.
Huando also exports all its agricultural products in equal pro-
portions, as well as both cattle and sheep. Its mills and trucks provide
Z-exports and cater to the surrounding area and, to a lesser extent, to
Huando itself. Trading income is derived from the community's own
weekly market.
Finally, in Acobamba, 80% of agricultural exports consist of
beans, and cattle and sheep are also sold. The community is located at
the end of a road, and is therefore a commercial center; its market is a
source of trading income for the members of the community.
This list illustrates the diversity of export products both within and
between communities. The one exception to this variation is cattle,
which clearly constitutes a cash-product produced specifically for
export in all the communities, but even then its relative importance
varies, thus further underlining the differences between them which
were emphasized above.
Imports
In contrast, the structure of imports does not exhibit much variation
between communities. Table 4.4 shows that food and drink account
for between 37% and 49% of imports. Expenditure on coca, alcohol
Table 4.4 Structure of monetary imports (%
Communities
T *• 1
Items JAC CUL NIN ANC TTI SIH HUA ACO

i oral ooutncrn
sample sierra
Food (coca, alcoholic
drinks, cigarettes) 41.9 38.1 42.3 36.9 40.0 45.8 46.1 49.0 43.6 44.8
(2.9) (4.3) (14.5) (11.6) (17.1) (13.1) (2.7) (10.2)
Fuel 3.9 1.9 1.9 2.8 3.2 3.7 4.6 2.4
Durable goods 0.5 1.1 1.2 0.8 1.5 2.2 0.7 0.7
Non-durable goods 5.8 4.3 11.2 4.1 9.6 11.0 5.8 7.6
Clothes and footwear 13.9 16.9 14.1 28.7 24.7 15.4 14.2 11.0 15.6 15.1
Education 6.7 5.7 0.5 3.4 1.6 3.7 3.8 2.7
Fares 7.9 8.1 5.4 3.7 4.1 4.1 3.0 2.7
Fiestas 0.7 3.0 7.6 8.4 2.2 4.6 0.5 8.9
Medicines 2.5 2.5 0.3 1.2 2.4 1.5 6.6 0.7
Other 1.1 0.1 0.1 0.1 0.1 0.1 2.3 5.3
Sub-total 84.8 81.7 84.6 90.0 89.3 92.0 87.7 91.0 86.8 87.7
Inputs 9.2 6.5 8.4 5.3 4.8 3.5 7.6 4.1
Investment 6.0 11.8 7.0 4.7 5.9 4.5 4.7 4.9
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
and cigarettes, which is included under food, is associated with
productive activity, since they are used as wage payments in kind, and
these are generally considered to be the community's most important
expenditures. Table 4.4 shows that these products account for
between 3% and 17% of total imports.
A long way behind food and drink, the next most important
imports are clothes and footwear, varying between 11% and 29% of
the total. Consumer items as a whole account for between 82% and
92% of total imports. Modern inputs to production are of little
significance, varying between 4% and 9% of imports. Capital goods
are somewhat more important, varying between 5% and 12%, but
remain a small item.
This homogeneity of imports compared to the heterogeneity of
exports in the communities is explained by the greater homogeneity
of their needs, which are restricted to certain types of food, fuel and
clothes, and also by their low and relatively uniform levels of income,
to which we turn in Chapter 5.
Exchange with other communities

Although there is some monetized exchange between communities,
exchange is mostly carried out in kind, and likewise most exchange in
kind is accounted for by exchanges with other communities. The
estimates of non-monetary exchange can therefore be seen as a
measure, albeit an approximate one, of exchange between com-
munities. According to Table 4.1, these exchanges in kind account
for a very small proportion of both total production (5% at most, as
in the case of Jacantaya) and total exports (7% at most, again in
Jacantaya). However, the two communities which are not repre-
sented in that table show higher proportions: in Sihuina, 11 % of
external exchange is accounted for by transactions with other
communities, and 23% in Ttiomayo, as shown in Table 4.5. Thus one
might conclude that exchange with other communities varies
substantially, having an average value of 6% of total external
exchange.
Our method of accounting for trading activities may have led,
however, to an underestimate of this figure. In communities such as
Sihuina, Huando and Acobamba, which have their own weekly
markets, much of the trading income which has been classified as
'exports of services' in fact derives from exchange with people from
Table 4.5 Katio of non-monetary exchange in peasant communities (soles per

year per family and %)
Value of exports
T 1 * 1
in-Kind
Community Monetary In kind Total share (%)
Jacantaya 37,584 2,861 40,445 7.1
Culta 39,771 312 40,083 0.8
Ninamarca 20,813 1,785 22,598 7.9
Ancobamba 17,745 464 18,209 2.5
Ttiomayo 12,530 3,762 16,292 23.1
Sihuina 19,023 2,593 22,616 11.5
Huando 46,484 228 46,712 0.5
Acobamba 26,773 523 27,296 1.9
Sample total 5.6

Southern sierra 5.7
other communities; if one were to include such transactions the

proportion of production exchanged with other communities would
rise, but not significantly. Moreover, this modification would apply
only to communities which have their own markets.
5
The level and structure of peasant income
The information presented in the previous chapter concerning

intersectoral relations in peasant communities now permits us to
arrive at estimates of the level and structure of income in those
communities. In this way we estimate total peasant income based on
previous understanding of the peasant economy.
The level of total peasant income

One way of calculating peasant income is to consider the value (at
market prices) of consumption and investment. As net peasant
production is dedicated in part to home consumption and in part to
trade, income may be defined as the sum of total consumption and
investment, from which the net balance of external exchange must be
subtracted. As the value of exports (including transfers) is equal to the
value of imports, a second alternative for measuring income is to add
to home consumption the value of exports (which would give the net
product of A + P + Z) and then the value of income from temporary
migrations. From this sum we would have to deduct the value of
imports of intermediate products which were used to obtain the net
product A + P + Z. Of course, here we are using the terms imports
and exports to refer to goods purchased or sold outside the
community.
The third method of calculating income is by way of value added.
The estimations of total product A + P + Z and of inputs (domestic
and imported) permit us to make this calculation, since the value
added is the difference between these two figures. However, income
from wages and rent has been estimated directly in the surveys, and
thus income from self-employment in A + P + Z was estimated by
subtracting out these other components. To the income A + P + Z
obtained in this way (value added), we must add the income from
41
temporary migrations in order to arrive at an estimate of totalpeasant
income. This income is equal to that calculated by means of final
demand. The estimates for the six communities are presented in
Tables 5.1-5.3.
To summarize, in the calculation of income we have considered the
activities A, P and Z in the community, plus the income from
temporary migrations. The allocation of resources to produce goods
has been taken account of in its entirety. Production of income-
generating (monetary and non-monetary) services is also estimated.
The only component of income which is not included is the own-
consumption of household services, such as good preparation,
gathering of fuel and housing repairs.
The average family income levels estimated vary by community
from 47,000 to 82,000 soles per year. On a per capita basis this income
varies between 12,000 and 20,000 soles per year. By comparison,
during the period of the survey the legal minimum wage in Lima was
between 6,900 and 9,000 soles per month, that is, around 100,000
soles annually. The average wage in the manufacturing industry was
at that time around 500,000 soles annually, and the salary of
professionals easily exceeded one million soles per year. To put the
peasant wages under discussion in a global context, we can use the
exchange rate of 200 soles per US dollar which was in effect in late
1978 and early 1979. We thus see that the annual family income is
approximately 250-400 US dollars, while the annual per capita
income is 60-90 dollars. We thus have an estimation of the extreme
absolute and relative poverty in which these families live.
Structure of community income
Own-consumption and exchange

Table 5.1 shows that relative proportions of income derived from
own-consumption (consumption from own production) and ex-
change vary among communities, own-consumption representing
between 45% and 70% of income. The average value for the six
communities is 51% and for the southern sierra 53%. Thus, it may be
concluded that own-consumption is approximately half of total
income.
This result outlines the importance which the market has for the
peasant economy. With this empirical result it is no longer possible to
The level and structure of peasant income 43
Table 5.1 Structure of income of peasant communities: own-consumption
and exchange (soles per annum per family)
Tempo-
Own- rary Impor-
Total consump- Exchange migrated
Community income tion AtP,Z tion inputs
JAC 62,441 27,831 27,674 12,772 -5,836

(44.6)° (55.4)
CUL 68,507 31,175 30,680 9,403 -2,751
(45.5) (55.5)
NIN 47,302 26,584 22,456 142 -1,880
(56.2) (43.8)
ANC 57,494 40,511 9,751 8,458 -1,226
(70.5) (29.5)
HUA 81,518 38,920 42,192 4,520 -4,114
(47.7) (52.3)
ACO 60,003 34,245 23,455 3,841 -1,538
(57.1) (42.9)
Sample
total 67,700 34,406 33,294
(50.8) (49.2)
Southern
sierra 69,071 36,597 32,474
(53.0) (47.0)
"Figures in parentheses = percentages.
refer to the peasant economy as 'self-sufficient' or 'dual'. On the

contrary, it is an economy well integrated into the rest of the Peruvian
economy. Normally when one speaks of countries one refers to an
'open economy' as one in which exports comprise 20-25% of GNP.
On the other hand, the peasant economy has always been seen as
'closed', 'self-sufficient', 'outside the market', when in fact it exports
almost half of its net product.
How can the varying degree of integration with the market be
explained? If the community of Ancobamba is not considered, the
range of the proportion of production for home consumption is
between 45% and 58%. Ancobamba represents an extreme case, and
this may be due to the fact that of the six communities it is the most
isolated from highways, being five to six hours' journey from the
nearest one. (This is also the case with Sihuina, which may have a
structure similar to Ancobamba.) The development of a highway
infrastructure is clearly a determining factor in a community's
integration with the market. What is surprising is that a third of
Ancobamba's income comes from trade with the outside, in spite of
its isolation.
Apart from the case of Ancobamba, in the five remaining
communities there exists an inverse relation between the level of
income and the proportion of it comprised by production for own-
consumption. The three communities with a proportion of own-
consumption representing between 45% and 48% of income
(Jacantaya, Culta and Huando) are the communities with the highest
level of peasant income. The two remaining communities have a
larger proportion (57-58%) and a lower income level. Nevertheless,
these correlations are not very strong, owing to the fact that there are
other variables, such as the diversity of resources and the seasonality
of agricultural processes, which also influence the degree of economic
exchange.
According to productive structure

The data from Table 4.1 allow us to make still another estimate of
peasant income. The communities taken as a whole derive income
from each of the A, P and Z-activities and the temporary migration of
labor. The structure of peasant income broken down into these
components is summarized in Table 5.2, which constitutes, in
addition, an approach for determining the productive structure of the
communities.
Agricultural and livestock activity together represent the largest
source of income in the communities, but their relative importance
varies considerably: between 43% and 90% of total income.1 In
Jacantaya, where the participation of A + P is the lowest, Z-activities
and temporary migration are important components in the income
structure. Within the Z-activities the important components are the
fishing carried out in Lake Titicaca, together with commerce,
particularly that of vegetables. At the other extreme, Ninamarca is a
community whose resources are devoted almost exclusively to
agriculture and livestock husbandry. Only 10% of total product
comes from Z-activities, while temporary migrations do not have
The level and structure ofpeasant income 45
Table 5.2 Productive structure of the communities (%)
Temporary
Community A P Z migration Total
JAC 28.6 14.3 36.6 20.5 100.0

CUL 26.5 47.1 13.7 13.7 100.0
NIN 44.9 44.6 10.2 0.3 100.0
ANC 59.0 21.0 5.3 14.7 100.0
HUA 46.3 17.1 31.1 5.5 100.0
ACO 46.7 20.0 26.9 6.4 100.0
Sample total 43.5 24.5 20.8 11.2 100.0
Southern sierra 46.3 21.9 22.3 9.5 100.0
great importance either. The Z-activities constitute between 5% and

37% of the income in the sample, and are thus in general terms an
important source of income. Note that the Z-activities and the
temporary migrations together make up close to one-third of peasant
income. An implication of this result is that there is an important
underestimation in the calculation of rural income when, as in the case
of the National Accounts, it is taken only as that emanating from crop
and husbandry activities.
By salaries, rents and self-employment

The principal source of income in all the communities is the income
from self-employment, as Table 5.3 shows. The proportion varies
from 72% to 98% between communities; in other words, normally
more than three-quarters of peasant income is from self-employment.
The peasant family of today depends to a great deal on the application
of its labor to its own resources to generate total income.
Property rents (rents from the use of land, animals and tools) are
not very significant.2 There are no families that make their living
primarily in this fashion. Thus, it is wage income which takes second
place to self-employment in the composition of total income. There
are two job markets to consider: local and external. The income from
both markets represents between 4% and 28% of peasant income. If
we do not take into account Ninamarca, a community in which
almost all income is from self-employment, the proportions of
income from wages represent between 17% and 28% of total income
in the remaining five communities. In three communities the external
market is the most important source of jobs, while in two com-
Table 5.3 Structure of income of peasant communities: wages, rents and self-
employment (soles per annum per family)
Total Local wage Self- Temporary

Community income employment Rents employment migrations
JAC 62,441 4,685 72 44,912 12,772

(100.0)fl (7.5) (0.1) (71.9) (20.5)
CUL 68,507 3,052 317 55,735 9,403
(100.0) (4.5) (0.5) (81.3) (13.7)
NIN 47,302 1,045 0 46,115 142
(100.0) (2.2) (0.0) (97.5) (0.3)
ANC 57,494 1,222 77 47,737 8,458
(100.0) (2.1) (0.1) (83.0) (14.7)
HUA 81,518 8,683 1,493 66,822 4,520
(100.0) (10.7) (1.8) (82.0) (5.5)
ACO 60,003 11,050 425 44,687 3,841
(100.0) (18.4) (0.7) (74.5) (6.4)
Total
sample 100.0 7.8 0.8 80.2 11.2
Southern
sierra 100.0 8.8 1.0 80.7 9.5
a
Figures in parentheses = percentages.
munities the local market is of more significance. Employers in the

local market are primarily medium-sized landowners who live side by
side with peasant families in the community.
Monetary transfers
The final grouping in Table 4.1 refers to monetary transfers. In the
communities there exist, on one hand, internal and external transfers;
and on the other hand, monetary and in-kind transfers. In all there are
thus four forms which transfers among families may take. Direct
transfers from the State or from organizations such as the Church
have not been estimated. However, neither is important at the level of
the community.
In-kind transfers between families within the community include
A, P and Z-goods, productive services (especially labor), and
monetary transfers which occur on occasions such as marriages and
other celebrations. These internal transfers have not been estimated.
External transfers have been estimated, but only the monetary ones.
Estimation of non-monetary external transfers was not possible, due
to the difficulties of specifying their quantities and qualities.
Nevertheless, it is known that these transfers are very frequent in
both directions. The movement of cargo in and out is impressive,
above all in communities with access to roads.
The estimates of monetary transfers show the communities as net
receivers. The absolute magnitudes vary between communities, as
can be seen in Table 5.4. The transfers received come primarily from
children who live in urban areas. Cases were also found in which the
head of the household lived in another place for occupational reasons
and sent money to his family in the community. The transfers out of
the community are generally to sons and daughters who study in a
city.
Table 5.4 also shows the relative importance of transfers in the
peasant economy. As a proportion of total peasant income monetary
transfers represent up to 8% or 10%, as in the case of Jacantaya and
Huando. For the rest of the communities the figures are quite low. If
they are compared with total income from exports (including
temporary migrations), monetary transfers increase the purchasing
power of exports up to 17% (the case of Jacantaya). For Huando and
Table 5.4 Money transfers (soles per annum per family)
% of total % of export
Community Ingress Egress Net income income
JAC 7,345 871 6,474 10.4 17.2

CUL 3,715 2,100 1,615 2.4 4.1
NIN 0 0 0 0 0
ANC 2,239 417 1,822 3.2 10.3
TTI 267 0 267 n.a. 2.1
SIH 755 0 755 n.a. 4.0
HUA 7,037 450 6,587 8.1 14.2
ACO 564 0 564 1.0 2.1
Sample
total 8.0
Southern
sierra 7.6
Ancobamba these proportions are 14% and 10% respectively. For

the five remaining communities the numbers are not very significant.
For the southern sierra, the expansion of these sample results gives a
figure of around 8% for monetary transfers as a proportion of total
exports.
The monetary budget of the peasant family

The structure of total income shown in the previous section refers to
the peasant community, and not to the individual family. For the case
of the peasant family, its sources of total income are: A, P and Z-
goods (products of self-employment), income obtained from outside
employment, such as salaries in the local labor market (IF) and in
markets outside the community (W2); income from renting land or
the means of production (R); and finally, income transfers (T). As is
explained in Appendix II, the method of calculation of total peasant
income for the community, shown in Table 4.1, was based partly on
family income data and partly on global estimates for the community.
Thus it was not possible to obtain estimates of to/#/peasant income for
each family, and in this way arrive at an approximation of the
structure of total income of the typical peasant family. Although the
income structure of the community does not have to be equal to that
of the typical family, we will assume here that the differences are not
great.3 Then the result shown in the previous section (in particular,
that on average half of total peasant income is monetary) is applicable
also to the typical peasant family.
Information on the monetary budget was obtained from all the
families and communities. The estimates of monetary income by
source, as well as its destination among the categories of monetary
expenditure, refer to gross income, since allowances have not been
made for the family's input purchases for its production.
Structure of gross monetary income

The first conclusion to emerge from Table 5.5 is that the sources of
money income are quite diversified. The peasant family has contact
with several markets at the same time in order to obtain monetary
income. In the second place, the structure of income varies
significantly between communities. The third conclusion is a con-
sequence of the earlier ones: all sources of income are important for
The level and structure of peasant income 49
Table 5.5 Structure of monetary income of peasant families (%)
Communities
Sample Southern
JAC CUL NIN ANC TTI SIH HUA ACO total sierra
A 11.8 0.6 58.8 1.5 21.2 13.9 12.8 10.9 10.9 11.1
P 11.7 36.7 23.7 29.2 15.1 21.9 10.7 5.5 2Q.7 18.7
Z 17.8 15.2 2.7 7.2 0.1 34.2 21.0 15.9 18.5 18.9
wx 9.0 6.0 5.4 10.3 40.1 10.6 19.3 44.3 15.2 18.3
21.5 19.7 0.3 22.8 15.1 8.6 9.3 11.6 14.5 13.7
R 0.1 0.3 0.0 0.3 5.0 0.1 0.6 0.0 0.6 0.6
T 17.7 10.0 1.5 16.8 0.7 5.3 10.9 3.4 9.6 8.9
0 10.4 11.5 7.6 11.9 2.7 5.4 15.4 8.4 10.0 9.8
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Note on »ew symbols: R = rents; T = transfers; 0 = 'others' (loans, dissavings).
the typical peasant family. This may be seen in the values that appear
in the last two columns of Table 5.5.
If we consider only income obtained from economic activity, that
is, if we exclude transfers and the category 'others' (loans, dissavings)
from Table 5.5, the structure of monetary income of the typical family
of the southern sierra is: 14% from the sale of crops, 23% from the
sale of livestock, 23% from Z-activities (including commerce), 23%
from salaries obtained locally, and 17% from salaries obtained
outside the community. Hence, almost 40% of family money income
is obtained from the sale of labor. In this sense, and to this extent, the
peasant family of today is also a proletarian family.
Structure of gross monetary expenditures

How does the peasant family spend the income which it receives from
different sources ? Three categories or components of expenditures
have been distinguished: intermediate goods and productive services
(salaries and rents) for production, capital goods for accumulation,
and consumption goods. The consumption goods have been sepa-
rated intofivecategories: food and drink, non-durables, clothing and
shoes, educational materials and 'others'.
Table 5.6 presents the results of the estimates made from the
surveys. As might be expected in this type of estimate concerning
Table 5.6 Structure of monetary gross expenditure of the peasant family
/o/ )
( /o)
Communities
Total Southern
Items JAC CUL NIN ANC TTI SIH HUA ACO sample sierra
Inputs 7.7 5.9 7.5 2.8 3.7 3.6 5.8 3.6 5.0 4.7
Services 2.4 1.8 0.2 1.5 3.5 2.3 2.6 1.5 2.1 2.1
Invest-
ment 6.3 10.5 5.8 4.1 2.4 3.5 3.9 2.7 5.3 4.6
Food 41.9 37.1 44.6 43.4 40.7 45.1 47.9 55.9 44.2 45.4
Non-
durables 10.0 6.4 13.4 8.2 12.6 19.5 12.2 14.0 12.1 12.8
Clothing 10.9 12.4 13.4 22.8 25.4 13.9 12.0 10.0 14.2 14.4
Education 5.7 4.6 0.6 3.0 1.1 4.0 4.6 2.7 3.9 3.8
Others 15.1 21.3 14.5 14.2 10.6 8.1 11.0 9.6 13.2 12.2
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
family expenses, one conclusion is that the structure of expenses is

well diversified, as it includes many different products and services.
However, it may be seen from the aggregated data in Table 5.6 that
the category of food is the dominant one in the structure of
expenditures. The typical peasant family spends 37-56% of its
budget on food, depending on the community. The categories that
follow in importance are non-durables, and clothing and shoes. The
remaining categories, including the money spent on inputs and
productive services, are less significant. Nor is a large part of the
monetary budget spent on investments. The second conclusion is
that the structure of monetary expenditures is more homogeneous for
the different communities than that of income. The order of
importance of the different expenditure categories is quite similar in
all of the communities.
The typical peasant family of the southern sierra allocates around
5% of its expenditures to production inputs, and somewhat less than
2% to productive services (salaries and rents). In other words, less
than 7% of expenditures goes toward production. A smaller
proportion, around 4%, is spent on capital goods. This means that
approximately 90% of expenditures goes toward consumer goods
(as seen in the last column of Table 5.6). Thus, the peasant family
allocates the greater part of its monetary income to the acquisition of
consumer goods. The family's accumulation is very much reduced, as
only a small part of its monetary budget is destined for investment.
Net investment is even smaller, almost nil, as much of investment
expenditure (in animals and tools) is replacement. The low expendi-
ture in hiring labor is the result of our definition of peasant family: a
family that does not hire labor. The proportion spent on modern
inputs is also relatively small and clearly indicates the limited
incorporation of modern inputs into the production process.
Almost half of the peasant family's expenditures on consumer
goods is for food. This category includes both processed and un-
processed products, of urban and rural origin respectively. The
monetary expenditures on this latter group are fairly small. There are
of course families who purchase potatoes, maize and barley when
their own harvest is insufficient, or who buy meat, eggs and milk.
There are also families who spend money on foods processed in the
community (Z-goods such as chicha and chuno). However, all of these
products together do not represent a significant part of monetary
expenditure. The principal food purchases are of six processed
products of urban origin (salt, sugar, cooking oil, ice, noodles and
flour) and two products of rural origin but which are brought in from
outside the sierra (coca and cane liquor). The expenditure on these
two products, together with cigarettes, is associated principally with
work. They constitute part of the payment in kind for wage labor or
for reciprocal work within the community.
There are two types of non-durable goods of great importance:
combustibles (kerosene and candles) and cleaning articles (soap and
laundry detergent). The category 'clothing' includes clothing for the
entire family, such as pants, shirts, shoes (normally of synthetic
materials), hats and skirts (the traditional pollera).
Peasant differentiation and forms of entrance into the

market system
Examining the monetary budget, we have shown in the previous
section the distinct ways in which a typical peasant family enters into
the market economy. In each community there are differences
between families as to the form that their incorporation into the
market takes. Not all families participate in the same markets and
derive their income or carry out their expenditures in the same way.
There are income differences in the communities and the question is
whether the poorest peasant families enter into the market in a way
distinct from those who are less poor. In other words, are there
significant differences in the structure of income between families
belonging to different economic strata within a community ? In this
section we present some empirical results related to this issue.
Inequality in monetary income

The most trustworthy result from the empirical point of view relates
to the monetary income and expenditures of the families. As indicated
above, the income from production for own-consumption has been
estimated for several products taking the community as a whole, but
not for the individual families. For this reason it is not possible to
analyze inequalities in total income between families.
One possibility is to measure the inequality in terms of monetary
income only. This measure is quite useful as a quantitative approxi-
mation of the degree of inequality. We would expect monetary
income to increase with total family income, and at an increasing rate.
Thus, income inequality is overestimated when measured only in
terms of monetary income. Table 5.7 presents various indicators of
inequality. As can be seen, inequality is fairly pronounced, even
taking into account the bias indicated above. For example, the ratio of
the percentage of income received by the top quartile to that received
by the bottom quartile has a magnitude of between 4 and 13,
depending on the community. Thus, in addition to a low level of
income in the communities, there is a considerable variation in level
from one family to another. Clearly the peasant communities do not
constitute a homogeneous mass of poverty. Given the income
inequality, the question of how families from different economic
strata enter the market takes on special importance.
Patterns of monetary income in the peasant communities

It is important to study the family income structure by economic
stratum. In this way we can learn which are the most important
sources of income for the poorest peasant families, and for those that
are relatively better-off. Nevertheless, it would make little sense to
speak of the most important sources of income for poor (or rich)
families in general, when we have shown in previous sections that the
Table 5.7 Inequality in family monetary incomes
Coefficient of Income share of
Lower Upper
quartile quartile
Community Variation Gini Skewness (%) (%)
JAC 64.3 0.33 1.3 10 51

CUL 64.1 0.34 1.2 9 48
NIN 86.8 0.36 2.9 8 51
ANC 132.9 0.50 3.7 5 61
TTI 57.2 0.29 1.7 11 45
SIH 80.4 0.36 2.8 8 48
HUA 130.3 0.53 2.8 5 64
ACO 113.7 0.47 3.5 7 59
sources of income vary greatly from one community to another. For

example, in the community of Culta agricultural monetary income is
insignificant, and thus is not of importance for any stratum. In
Ninamarca, on the other hand, agricultural monetary income is very
important, and may have a varying relative importance between
economic strata.
For the reasons discussed above, instead of showing income
structure by stratum in each community, we have opted to examine
whether or not there exist generalized income patterns by stratum in
all the communities. Is agricultural monetary income, for example,
greater in the upper strata, i.e. does it increase with the family's total
income level independent of the community ? The level of agricultural
monetary income will depend on the community with which we are
dealing, due to the above-mentioned differences in sources of income
between communities. However, the slope of the line relating
agricultural monetary income to total income may be similar (if not in
magnitude, at least in sign) between communities. The usefulness of
this procedure is that it permits us to identify the types of monetary
income which vary negatively with total income, and which are thus
the most important for the poorer families. It may also be of interest
to know which types of monetary income vary positively with total
income, and whether these have the same or lesser importance for the
poorer families as for the more prosperous ones. We already know
that all income sources except rent are important for the typical
family. Now we can determine the relative importance of different
income sources by stratum.
In order to carry out the proposed analysis we have designed a
regression model of the following form:
where Y- is the monetary income of type j9 Y is the total monetary

income of the family, and the X^s are dummy variables for each of the
eight communities. With this specification we maintain the difference
in income levels by type between communities, calculated by taking
the differences in the intercepts. We have considered six types of
income, each representing different markets and mechanisms of
market entry. The six types are income from agriculture, livestock,
commerce, unskilled wage labor in the community, skilled wage
labor in the community and wage labor outside the community.
Income from Z-good sale was not analyzed due to its specific
character for each community. The model was applied to all the
families in the sample.
Three equations were specified in addition to the one presented
above: one utilizing the original data in log—linear form, and two
others using deflated values of the variables in conjunction with the
two functional forms. As the survey was conducted during a period
of marked inflation in Peru (around 60% annually), correcting the
nominal values of the variables by deflating with a price index seemed
a proper step before comparing the communities. The time difference
between the first and last community surveys was eight months.
The results of the regression analysis are shown in Table 5.8. In
view of the fact that the results using deflated values for the variables
were similar to those using the original ones, only these latter
equations are shown. In general the equations show the existence of a
significant relation between a specific type of income and total
income. All of the regression coefficients except one are significant at
the 0.05 level. Nevertheless, the R2 values are not very high. This is
the case in particular for income from temporary migration. The
results suggest that there exists a positive relationship between total
monetary income and all sources of income except that from unskilled
labor in the local market. The positive relationship with cropping and
livestock husbandry monetary income can be explained by the fact
that the richer peasant families have a much larger marketable surplus
than the poor families, because they have more land and animals. This
Table 5.8 Summary of the regression analysis
Regression coefficient Global
Variable Value F-value R 2 (adjusted) F-value
Linear
Agriculture (A) 0.113 108 0.370 22
Livestock (P) 0.081 35 0.243 12
Commerce 0.308 243 0.494 36
Wages - local (WJ -0.015 9 0.369 22
Wages - local (skilled) 0.101 98 0.270 14
Wages — external 0.028 5a 0.108 5
Logs
Agriculture {A) 0.865 12 0.400 26
Livestock (P) 1.851 36 0.159 8
Commerce 2.004 57 0.271 15
Wages - local (B^) -1.229 21 0.245 13
Wages - local (skilled) 0.295 1* 0.012 \b
Wages - external 0.859 7 0.109 6
" Non-significant at the 1 % level.

''Non-significant at the 5% level.
would be true even if the proportion of production for own-

consumption were equal for rich and poor families (which may not be
the case). It is to be expected that production for own-consumption
(consisting primarily of food crops) diminishes as a proportion of
total income in the upper economic strata.
The fact that monetary income from commerce increases with the
level of total monetary income indicates that the relatively better-of
peasant families are heavily involved in this activity. Personal
contacts, knowledge of Spanish, available working capital: all make
the wealthier families much more suited for commerce. Poor families,
on the other hand, carry out small-scale commerce, thus obtaining a
lower income from these activities than is the case with the rich
peasant families. In some communities the latter are even able to make
selling trips to the large cities carrying their wares in trucks.
The other income category which is positively related to total
income is the income from skilled labor in the community. Artisans,
particularly bricklayers, obtain relatively high salaries in the labor
market, and the wealthier families derive more income from these
activities.
With respect to wage income from unskilled labor in the local
market, the results show that this type of income is less in absolute
terms among the rich families than among the poor ones. The
wealthier families do not work for wages in the community; on the
contrary, given their resources, many of them are demanders of casual
labor whereas the more impoverished families are the suppliers of
labor. The logical question is, do these poorer families also obtain
more income from temporary migrations than do their better-off
counterparts? Although the statistical result is weaker, there are
indications that income from seasonal migrations varies positively
with total monetary income. This source is relatively more important
for the middle and upper strata. One would expect that the rich
families would have sufficient resources and would not need to
migrate, while the poor families would. Nevertheless, these results
show the opposite. One possible explanation is that it is skilled labor
(which the poor families do not have) which is exported. This 'skill'
refers not only to a particular occupation, such as carpenter,
brickmason, weaver or musician, but also to 'cultural skills'. A
knowledge of the Spanish language, of how to read and write, of how
to conduct oneself in cities and in the valleys where modern farming
methods are practiced - all of these are required, and are found
mostly among families in the upper economic strata.
There is an additional factor which is required to migrate
seasonally: working capital. In order to absent himself from the farm
for two or three months, the head of the household must have
enough funds to cover the expenses of his family during that period,
as well as the costs of traveling, lodging and looking for work. Poorer
families obviously have less of this working capital than richer ones.
It is for this reason that the debt peonage system (enganche in Spanish,
literally 'hooking') of intermediaries in rural labor markets is so
important. The contractor {enganchador) provides the peasant with
capital, in the form of a cash advance. A relatively prosperous peasant
family would certainly not wait for advances before migrating
temporarily.
All of the above leads us to the conclusion that the richer families in
the peasant communities derive more income from all sources than
the poorer ones, with the exception of wage labor within the
community. This is due to the fact that richer families have access to a
large quantity of all the resources which exist in each community.
From the regression analysis we may also make inferences
regarding the proportions of total income consisting of given types of
income, and how they vary with total income. In the case of the
logarithmic regressions it is sufficient to look at the value of the
regression coefficient, since this is equivalent to the elasticity
coefficient. From Table 5.8 we infer that the proportion of total
income made up of agricultural monetary income diminishes as the
level of total income increases. The same thing occurs with wage
income from the communal labor market. In both these cases the
elasticities are less than one. For income from livestock and commerce
the elasticities are greater than one, indicating that this income makes
up a larger share of total income as the latter increases.
In conclusion, there exist clear income patterns among peasant
families, in spite of the differences in the sources of income between
families. Wage income from work in the community has a greater
absolute and relative importance among the poorer families, regard-
less of the community concerned. The same may be said of
agricultural monetary income, except that the importance is only in
relative terms. On the other hand, monetary income from livestock
and commerce has a greater absolute and relative importance for the
families of the upper strata, independently of the community. These
two activities generate income primarily for the relatively wealthier
peasants. Wage income from external markets is relatively more
important for middle and upper strata.
6
The economic behavior of the peasant family
Even though peasant families are organized in communities, the

economic decisions that they make concerning the use of their
resources and labor are fundamentally family decisions. The results
presented up to now show that the peasant family is at the same time a
production and a consumption unit. It is of interest to know the
essential elements that determine the economic behavior of the
peasant family. What are the principles guiding its economic actions ?
What is its economic rationality? From the answers to these
questions, we can derive propositions about the mechanisms that
govern the allocation of peasant family labor to alternative uses and
the reactions to market prices. These are the themes that we shall
attempt to develop in this chapter.
Portfolio of activities and risk-aversion

Any economic theory which attempts to explain the behavior of the
peasant family of the sierra must include as an essential element the
problem of risk. There are two reasons for this. First, practically all
the economic activities of the sierra peasants are subject to risk and
uncertainty. Agriculture in the sierra is certainly a tremendously risky
activity. Since there is normally no irrigation, dependence on rainfall
is complete. The absence of rain, its appearance at the wrong time, or
its over-abundance (creating floods and avalanches) affect production
significantly. At certain altitudes, frost and hail are climatic factors
which also affect production negatively and in an unpredictable
manner. To these problems of nature we must add others, such as
diseases. Livestock production also confronts significant risks, such
as falls (due to the rugged nature of the terrain) and strayings. The
production of Z-goods is in itself perhaps less risky, but since in most
cases it depends on crop and livestock inputs, or on trading activities,
58
The economic behavior of the peasant family 59
it does not escape having a strong chance component. Finally, the
incursion into the external labor market is at least as risky as the other
activities. The possibility of not finding employment, of losing one's
pay, of falling ill, are all clear examples of this problem. The peasants
who do not go out looking for jobs often say of those who do: 'They
only go looking for illness and come back sick.'
In the second place, the peasant families of the sierra are poor. It is
to be expected that the poor family will have an aversion to risk,
avoiding as far as possible endangering its income. A large loss of
income would mean economic disaster. The poor family confronted
with alternatives open to risk will thus prefer a small loss in its income
to a small chance of suffering a large loss.
Given the risk involved in the production and exchange activities
carried out by a peasant family in the sierra, a hypothesis concerning
its economic behavior could be stated in the following way: the
peasant family tries to assure, with the least risk, the attainment of a
bundle of commodities which is culturally determined.
In the case that production and exchange processes show a trade-
off between average income and security for the peasant family, the
way to minimize risk is by means of a combination of activities; that
is, by means of a diversified portfolio of resources and activities. This
diversification, as shown in the present study, is precisely one of the
characteristics of the peasant families and, thus, is entirely consistent
with the hypothesis put forward.
The preference for diversification of activities is clearly revealed in
an affirmation made by a Jacantaya peasant: 'Our agricultural
production is not guaranteed, due to problems of weather against
which we are defenseless. In order not to run this risk we could
complement our activities with other jobs in artisanry or some
industry. We would thus like to have a factory or workshop in order
to work. Perhaps in this way we could improve our situation.'
Portfolio diversification does not refer only to different activities
(A-P-Z-L), however. It refers also to diversification within an
activity itself. In the case of agriculture we observed that the peasant
family sows many different crops, and has numerous small parcels of
land. Why does the family not specialize in a few crops ? Why does it
farm such tiny, scattered parcels? A crop portfolio implies for the
peasant family the acquisition of parcels in different ecological zones,
in order to have access to more resources and a larger set of
production possibilities. As the negative effects of climate and
plagues are not the same in all the ecological zones, the probability of
all crops turning out badly becomes significantly reduced.
Even within the same ecological zone, the peasants take advantage
of the existence of special micro-climates to minimize risk by planting
on several parcels. For example, frosts do not affect an entire area
equally: flat and open fields are affected more than sheltered areas at
the foot of hills or mountains. A good illustration of this fact comes
from the statement of a peasant from the community of Culta: 'Since
the frost falls mostly in the openfieldsbut not on the slopes, if we sow
only in the flat places the frost can wipe out everything and we are
then left without anything to eat. So it is better to sow in different
spots. That way, you win some, and you lose some.' To the number of
parcels which the peasant uses in his crop portfolio we must add
parcels which are lying fallow and will be used for crop rotation. The
number of parcels which the farmer possesses is thus larger than the
number utilized, which is in turn explained by the diversification in
the crop portfolio.1
It is very probable that the uncertainty associated with agriculture
and livestock limits the diffusion of modern inputs. In the sierra it is
not a question of applying fertilizers and automatically obtaining a
good crop. Frost, hail and drought can cause even fertilized crops to
fail. A peasant from Huando sums up this problem in a very clear
way: 'Several peasants have obtained loans from the Agrarian Bank
to buy fertilizers and pesticides and have suffered crop failures due to
bad weather. They are scared now to ask for loans again.'2
Among the economic activities open to the peasant family, not all
of them have the same degree of uncertainty. If this is so, the
alternatives of average net income and its degree of uncertainty open
to peasant families will vary for each mix of A-P-Z-L activities.
A hypothesis on this issue would be to say that the way in which
agriculture and livestock production is organized in peasant com-
munities suggests that these activities (A-P), taken jointly, have a
lower degree of uncertainty (or variance). The mix of resources,
crops, parcels and types of animals indicates a very complex portfolio
of activities in agriculture and livestock production so that the risk of
a complete failure in production is greatly reduced. Also the risk in
exchange is lower because part of the A-P output is for own-
consumption.
In order to illustrate the previous argument, suppose a peasant
family specializes in the production of a single crop. If this crop fails
the family will face an economic disaster; but even if it does not fail,
this crop must be exchanged for other consumption goods. The
attainment of the basket of consumption goods is now subject to the
results of market exchange. Producing, in part, for own-consumption
is a way to reduce the risk in exchange. Hence, agriculture and
livestock production taken together reduce the risk, in production and
exchange, to assure the family's given basket of consumption.
A consequence of this risk-aversion behavior is that agriculture
and livestock production are given a priority in the allocation of
family resources, particularly of family labor. These resources are not
allocated to the different activities so as to maximize average income.
The allocation is biased toward more resources put into livestock
production as compared to the allocation that maximizes average
income; but risk will be reduced. It is in this sense that we can talk
about a hierarchy in the use of resources as another characteristic of the
economic behavior of the peasant family.
This theoretical result is consistent with an answer frequently
obtained from peasants when they are asked why they do not migrate
temporarily to work in other places where wages are comparatively
high: 'I do not go elsewhere to work because I do not have time.'
This is meant to indicate that crop and livestock tasks do not leave
enough time for migration to external labor markets. In many cases
the peasants do not even know what wages are paid in nearby cities or
valleys. Their holding of land offers them an income which is lower,
but more secure.
Agricultural seasonality and the allocation

of family labor
Agricultural activity is seasonal, but in the particular case of the sierra
this seasonality is quite marked, due to the fact that most of the land is
dependent on rainfall for its water. The profile of agricultural
seasonality which is shown in Figure 6.1 corresponds to what has
generally been observed in the communities under study. Turning
over the soil on lands which have been lying fallow takes place in
March and April, months in which the rainy season is ending. This is
particularly true of lands under collective rotation (the laymi).
Planting begins in August or September, with the arrival of the first
rains, and lasts until November of December. Once the planting is
carried out, the calendar of cultural practices is practically de-
Labor
DEC Months
Figure 6.1 Seasonality in peasant communities of the sierra
termined; these are carried out principally between September and

January. The harvest calendar is also determined by the planting date,
the peak months being May and June. The end result is that
July-September and January-April are periods of little agricultural
activity.
The agricultural calendar mentioned is certainly a simplification of
what is observed in each community. There are important differences
between communities in the timing of agricultural activities. This
may be due to the mixture of crops, the particular ecological zones,
the micro-climates or the availability of water for irrigation. Irrigated
lands give two crops per year in the quechua zn&yunga regions; in the
high regions this is less common, because of the strong frosts between
May and July. For this reason, the profile shown in Figure 6.1 is
indicative only of a 'typical' agriculture cycle in the sierra. In this
figure agricultural seasonality includes the labor requirements of
livestock production (equal to OB). To simplify matters, we have
assumed that there is no significant seasonality in livestock pro-
duction (although some activities such as shearing have a fixed
calendar).
The profile shown in the diagram refers to a typical family,
although it could also represent an entire community. If a community
has OE workers, it can supply itself with the labor it needs even in
periods of peak requirement. There will, however, be seasonal
unemployment in agricultural activities. If there are more than OE
units of labor the community is overpopulated, and the excess
population can move elsewhere without causing a decline in crop and
livestock production. If there are fewer than OE units of labor, the
community has a deficit. In the first case we would observe temporary
migrations outside the community even in peak periods of agricul-
tural activity. The second case would be consistent with peasant
immigration from other communities, particularly in periods such as
the harvest. If Figure 6.1 is taken to refer to a peasant family, the
vertical axis measures the size of the family labor supply and the curve
then represents labor requirements for the quantity of resources
which the family economic unit has at its disposal.
How is family labor assigned to A, P, L and Z-activities ?
Consistent with the risk-averse behavior discussed earlier, agricul-
tural seasonality seems to be the labor allocation mechanism. It is only
once the crop and livestock activities are concluded that the peasant
family takes decisions concerning other uses of the family labor force,
whether it is to employ itself in the production of Z-goods or to enter
the job market. Z-goods thus constitute complementary, not com-
petitive, activities with respect to agriculture and livestock in the
countryside. In order to be complementary, these activities should
not be seasonal (i.e. one should be able to start and finish them at any
time); or, if they are seasonal, they should have a cycle opposite to
that of the crops. Most Z-activities comply with one of these
requirements. Some examples of the first category are textiles and
commerce; of the second, construction which is carried on outside
the rainy season (between harvest and the next sowing), as well as the
processing of food which takes place after planting and before
harvest.
In the peasant communities almost everything has its 'season'.
There is a specific form of work organization during the year, which
imposes a certain rhythm and sequence on family labor. The most
important factor in defining this rhythm and sequence is the
agricultural calendar. There are two reasons for this. First, when the
schedules for L and Z-activities are flexible, these can be postponed.
The stages of the agricultural process, on the other hand, cannot be
postponed without risking a loss in production. Secondly, when a
number of activities coincide in their calendar, the A and P ones have
priority for the reasons explained earlier.
The hypothesis of risk-aversion and the existence of a marked
agricultural seasonality together imply that the production of Z-
goods is not affected in a significant way by crop and livestock
production, but rather by the salaries that prevail in the job market. In
the peasant family the allocation of labor to Z-activities and to wage
labor refers basically to the use of off-season labor. This means that
the peasant can dedicate himself only partially to crop and livestock
activity and still consider it to be the most important activity in his
effort to assure a subsistence income for himself and his family. Even
though he may be only a part-time peasant, that is his principal
identity. The plot of land gives him a measure of economic security.
The central hypotheses concerning the economic behavior of the
peasant family have clear implications for its entrance into the labor
market. In the following two sections we present the empirical
evidence.
The labor market in the community

The peasant family does not carry out its productive activity in
isolation from the rest of the families of the community. As we
indicated earlier, we observe three systems of exchange of labor
services in the community: (a) reciprocity; (b) payment in kind; and
(c) monetary wages.
The system of reciprocity, called ayni, has an important require-
ment for its functioning: the net balance for each family should be
zero. If a family has contracted five peasants for its planting, it has the
obligation of working five days for the others. If all families had the
same resources, they would all have to work five days in exchange,
and in this way the entire harvest would be finished with a perfect
communal labor balance. However, the existence of significant
peasant differentiation makes this communal balance imperfect.
There are families with a labor deficit or surplus. In addition, the
richest peasants will always prefer to contract laborers rather than
practice ayni, because they consider their opportunity cost to be very
high. There will also be some households (those consisting of
widows, for example) that cannot pay for services received with their
own labor. Finally, in every community there is a group of families
owning medium-sized farms, whose greater economic resources and
cultural differences with respect to the peasants set them apart. In the
language of the peasants (and of the Peruvian novelist Arguedas),
these are the vecinos ('neighbors'). These 'neighbors' also require
workers. From all this a labor market emerges in the community.
Payments for work done in the communities {total wages) have two
components. One is the wage proper, and the other is a payment in
goods defined according to the socio-cultural matrix of the com-
munity, and considered a 'right'. This 'right' usually includes food,
coca, aguardiente (cane liquor) and cigarettes. A typical wage contract
offered by an employer would include as a 'right' two meals, a fistful
(around an ounce) of coca, a quarter-bottle of aguardiente and one or
two cigarettes to each peasant for a day's work. Even if wages are paid
in kind (for example, work at harvest is paid with part of the crop, and
at planting with seed), the other non-monetary payment ('the right')
is still made. In Huando, where the monetary component of total
wages is more important than in the other communities, the non-
monetary component was estimated equal in value to the monetary
one when evaluated at market prices.
In the case of ayni only the non-monetary part of the wage is paid.
In addition, where the peasants concerned are very poor, they
reduce the quantity of the goods in 'rights' or they eliminate certain
items (food, for example, agreeing that each worker will provide his
own). On the other hand, in the minka system, the total wage consists
primarily of the 'right' to the non-monetary payment. Sometimes it
includes payment in kind, but this is not common. As the employer
does not return the labor received, as in the ayni system, the non-
monetary payment is much greater in quantity and quality than when
there is also payment of wage proper either in cash or in kind. Usually
the worker's family also receives food and care. If any balance remains
in the peasant's favor, the employer will compensate for it with
services at a later date. This latter point implies that minka is typically
a relationship between families from the upper strata (including
'neighbors') and poor peasants, while ayni is carried out mainly
among peasants from the lower strata.
The local labor market exists principally for adult men, as the
opportunities for women and children are much fewer. Only in
Huando, where there exist various families of'neighbors', is there an
important market for domestic help, i.e. women and children. The
most important source of employment for women and children is
helping with the harvest, where wage is paid in kind. The principal
role of these family members is not to obtain wage income directly for
the family (although as we shall see, many children do migrate
seasonally). The main responsibility of the women and children is
to take care of the productive unit (the family's crop and livestock
enterprises) so that the adult man is reasonably free to earn wage
income elsewhere.
Daily money wages vary significantly within the same community
and during a given month. This variance can be explained in part by
differences in the quality and quantity of the 'rights' (the non-
monetary wages). Thus, for example, the monetary wage with food is
less than the monetary wage without food. But in Huando, for
example, the monetary wage with food varied between 80 and 120
soles daily. Part of this difference is due to the slowness of the labor
market's adjustment to inflation. In any case, the money wage
differences also reflect to a great extent the personal relations that
prevail in the community, and the long-term 'balances' between
individuals. In addition, the fluidity in the adjustment of the labor
market to new money wage levels appears to depend also on the role
that personal relations play in the market. Personal relations make
market adjustments less fluid.
Employers in the community frequently complain that there is a
shortage of labor in the community, particularly during periods of
peak agricultural activity. One of the 'neighbors' indicated that when
he needs 10 laborers he contacts 15 'just to be safe'. The peasant
family who has extra labor after self-employment in the plot has
various alternatives, such as self-employment in Z-activities or
migration outside the community. The opportunity cost is not zero
during the off-season for the peasant family's labor force. The
fundamental mechanism attracting these peasants to the local labor
market must be increases in wages. Nevertheless, seasonal variations
in monetary wages are not observed in the community. The
adjustment clearly comes from the increase in payment in kind or in
rights to goods. At harvest time, for example, a common practice is
payment in kind of an amount of the product which has a greater
market value than the average monetary wage. The different
components of the total wage offer considerable flexibility for
adjustments.
One reason not to increase monetary wages in periods of labor
scarcity is the employer's awareness of the downward inflexibility (or
'stickiness') of wages. Huando provides a clear example. Here the
employers prefer to hire workers from neighboring communities
during the harvest season. They are paid an amount which exceeds
the going money wage in Huando or the other communities, because
they are in the area only temporarily. The workers from Huando are
not paid more, because 'one mustn't spoil them', as employers say.
Daily money wages also vary from one community to another. The
average wage for adult men during the period of the survey was 30
soles in Sihuina, 60 soles in Ninamarca, 80 soles in Ancobamba and 100
soles in Huando, Acobamba, Jacantaya and Culta.
The labor supply obviously comes from the family units with a net
labor surplus. This supply will be determined by opportunity costs :
products that can be obtained with self-employment principally in the
production of Z-goods and the net wages which can be obtained by
temporary migration outside the community. In this analysis we must
note that, because of the hypothesis of risk-aversion (and the
hierarchy in labor allocation), crop and livestock production do not
enter fully into the calculation of the opportunity cost. On the other
hand, since ayni is a way of organizing mutually aided production by
means of self-employment, it has no implications for the communal
labor market. Quantities of labor demanded and supplied are always
equal under the ayni system.
The external labor market

The temporary migrations which peasants make to labor markets
outside the community have the characteristic of being seasonal. The
data presented in Table 6.1 are a clear indication of this fact. When we
compare the number of person-days of temporary migration by
month with the average for the year, we see that the months when the
former is greater than the latter are always grouped together. For
instance, the greatest migrations in Jacantaya are in the months of
November-February and May-June. The second characteristic of
the temporary migrations is that the periods when this activity is most
intense coincide with the periods of least intensity in agricultural
activity. In all the communities the periods of greatest migration are
in the summer (December—May) and winter (June—August) months.
The few cases in which the greatest migration occurs in other months
(though always adjacent to the ones mentioned) are due to the specific
agricultural calendar of the respective community. There are com-
Table 6.1 Temporary migrations (number of days per family)
Communities
Months JAC CUL NIN ANC TTI SIH HUA ACO
Jan. 8.0* 4.0* 0.5 4.5* 11.0* 2.2* 4.8* 1.1

Feb. 7.2* 4.4* 0.0 4.5* 4.7* 2.2* 4.3* 3.3*
March 4.3 4.8* 0.0 4.5* 3.3* 2.2* 2.9* 4.0*
April 2.6 0.9 0.5 3.8* 0.7 0.7 2.3 2.5*
May 9.3* 0.9 0.5 1.5 0.0 0.0 0.8 1.5
June 10.6* 5.0* 0.0 0.8 0.4 0.0 0.7 0.9
July 4.7 5.9* 0.0 3.4* 0.7 0.7 0.7 0.9
Aug. 2.0 4.0* 0.0 8.8* 0.2 0.7 5.0* 2.5*
Sept. 2.6 0.9 1.0 1.5 0.6 0.0 3.0* 2.5*
Oct. 4.3 2.0 0.0 1.9 2.1 0.8 0.7 2.8*
Nov. 10.1* 1.2 0.0 1.1 2.0 0.8 1.4 1.7
Dec. 8.4* 2.1 0.0 1.6 3.9* 0.7 2.9* 1.1
Total 75 43 2.5 39 30 14 29 24
Monthly
average 6.2 3.6 0.2 3.2 2.5 1.1 2.5 2.0
Note: Values with asterisk indicate a higher value than the monthly average.
munities that have an agricultural calendar slightly ahead of or behind

the rest.
The evidence presented in Table 6.1 appears to be sufficiently solid
to conclude that temporary migrations are, indeed, seasonal migrations.
They are dependent on agricultural seasonality. This evidence is
consistent with the hypothesis presented above of hierarchy in the
selection of activities, where crop and livestock activities have
priority.
Table 6.1 also shows the average number of person-days of
migration from each community. The lowest average is in Nina-
marca, with 2.5 days; the highest ones are Jacantaya (75 days), Culta
(43) and Ancobamba (39). For the sierra as a whole, our estimate
is that the typical peasant family utilizes 34 person-days of its labor
supply for temporary migration. Since the common pattern is for the
head of the household to migrate, we may conclude that the head of
the household is typically absent for more than one month per year. If
we suppose that the total number of effective workdays is 270-80,
this indicates that around 12% of the available labor time of the head
of the household (i.e. of the adult male) is allocated to obtaining
3
income by means of entrance into external job markets.
What type of markets are these to which peasants migrate ? The
places of migration have been classified in two ways. The first set of
criteria concerned whether the market was regional or extra-
regional. 'Regional' is defined as meaning that the place of
employment was located in the same province or 'zone' as the
community. The idea was to distinguish between better-known,
visible markets (familiar to the majority of the peasants), and markets
which were farther away and less well known. Another characteristic
of the regional market is that it has a lower cost of entry than the
extra-regional one. The empirical results show that most of the
migrants go to the extra-regional markets. As Table 6.2 indicates,
27% of the heads of households migrated to extra-regional labor
markets, and 11% to regional ones.
Secondly, the destination of the migrants was classified according
to whether it was a rural or urban zone. The idea was to measure in
this way the importance of rural and urban labor markets for the
peasant economy (independent of being regional or extra-regional).
Table 6.2 shows that the rural markets have the greater importance,
Table 6.2 Destination of temporary migrants to external labor markets
Destination House-
hrJrl
I1U1Q
Extra- with no
Community Regional regional Rural Urban Various" migrants
JAC 5.6 47.2 27.8 19.5 2.8 41.7

CUL 25.0 27.5 17.5 30.0 5.0 40.0
NIN 9.7 3.2 12.9 0 0 87.1
ANC 40.0 10.0 15.0 32.5 2.5 45.0
TTI 2.8 33.3 36.1 2.8 2.8 58.3
SIH 2.4 24.4 24.4 2.4 0 73.2
HUA 0 30.9 7.1 19.0 0 69.0
ACO 7.7 28.2 28.2 5.1 0 64.1
Total
sample 11.7 27.5 19.5 16.8 57.7
Southern
sierra 10.6 27.0 20.0 15.0 60.1
3
Cases of migrants going to more than one destination.
although the difference is not as pronounced as in the regional—extra-
regional distinction. Although there are some differences between
communities in terms of the more relevant external job markets, for
the peasant families as a whole the conclusions are clear: the largest
group enters extra-regional and rural labor markets.
Finally, Table 6.2 shows the proportion of families in which no
member migrated temporarily in the year preceding the survey. This
proportion varies between communities, ranging from 40—42% for
Culta and Jacantaya to 87% in Ninamarca. These proportions
coincide with the ordering of the number of days spent as migrants.
From this we can conclude that there are some communities
which migrate more than others, and that there are no
families which specialize in migrating, independently of the com-
munity in which they live. If this were the case, we would observe
little relation between communities as far as the proportion of days
spent as migrants and the proportion of families who migrate. For the
southern sierra as a whole, 40% of the families have at least one
member who has temporarily left the community in search of work.
The principal activity which the migrating peasants enter is
agriculture. Of all who left during the year, 54% worked in this
sector. Of these, three-quarters obtained employment working with
crops destined for the national markets, such as rice, vegetables and
fruit (on the coast), and coca and lumber (in the selva, or eastern
jungle). The remaining 25% worked in export agriculture: cotton on
the coast and coffee in the selva, as shown in Table 6.3. The other
important occupation for the migrating peasants is construction,
where 33% of them found employment. The urban service sector
accounted for 19% — unloading trucks, working as street vendors,
domestic servants or waiters and waitresses. Two additional activities
with less importance are mining and handicrafts (including
musicians).
In sum, the labor markets through which most peasants pass are
those related to agriculture and construction. This pattern of
migration well reflects the skills of the peasants. On the other hand,
the employment they obtain is principally on small and medium-sized
farms, or with small firms. One reason they do not work in large
enterprises (the 'modern sector') is that these usually ask for work
papers and identification documents, which the peasants usually do
not have. ('I can only work on small construction jobs — in the big
ones they ask for papers'; 'I always go to the same place - since they
Table 6.3 Employment sectors of temporary migrants (number ofpersons)
Agriculture
IT 1'
x~ianQi-
Community Domestic Exports tion ces crafts Mining Total0
JAC 13 2 2 5 0 0 21
CUL 6 0 9 5 0 0 19
ANC 5 1 11 1 6 1 20
TTI 15 0 1 1 0 0 15
SIH 4 2 1 0 0 0 7
HUA 2 3 4 4 0 3 14
ACO 1 8 0 3 1 0 12
Total 46 16 28 19 7 4 108
Total
sample 217 76 179 105 30 26 542
(%) (40.0) (14.0) (33.0) (19.4) (5.5) (4.8) (100.0)
Note: No data were available for Ninamarca.

a
'Total' does not add up because the same migrants can have employment in more
than one sector.
already know me they give me work, while others ask to see my

documents', are common peasant statements.) One consequence of
this is that even to enter labor markets the peasant becomes
dependent on personal relations. In effect, the majority of the
peasants look for employment through relatives, persons from the
same community, or acquaintances. In this sense they rarely actually
'go out to look for work'. Some peasants revealed to us that in order
to work with large enterprises, they had to pay the person in charge of
hiring. This cost, added to the cost of obtaining work papers, imposes
additional barriers on the entrance of peasants into the 'formal' labor
market.
Another cost that must be faced by the peasant who migrates is the
financing necessary for leaving the community. A three-month
absence involves leaving the family enough cash for everyday
expenses, and having enough for carrying out the trip. The costs of
transportation and lodging, clothing or tools necessary for the job,
the days of waiting before getting the job and before receiving the
first pay all imply the necessity of having a 'working capital' in order
to obtain work. The amount of this 'working capital' has been
estimated in the survey. In the majority of the cases, the answer varied
between 500 and 3,000 soles depending on the place and the
conditions of the migration. These amounts may seem too small to
take into account, but we must remember that the income level in the
communities is quite low, and in that context the necessary amount of
capital becomes an important factor in migrating. In fact, some
peasants indicated that they did not migrate during the previous year
because they had no money.
How do the peasants finance the capital to migrate ? The employer
gave an advance to 25% of the migrants; the rest financed themselves
or borrowed money in the community. The mechanism of cash
advances is an essential part of a specific form of operating the labor
market - the enganche system, already discussed. The employer
usually arrives in the community and gets a commitment from several
peasants to work for his enterprise, usually agricultural, in a given
time period (for example, in the coffee harvest months), and to
formalize the commitment he hands out an advance in cash. We also
found cases in which the employer gives money to a peasant returning
to his community. The latter then finds more workers and pays them
an advance. The enganche system applies mainly to agriculture, where
labor requirements at specific times of the year are crucial, and the
cash advance is the mechanism which assures the provision of that
labor.
Although the enganche system contributes to the functioning of the
labor market in its present stage of development, some peasants feel
that it is a system that places them at a disadvantage. One peasant
from Acobamba illustrates the problem well when he says: 'The
worker who is enganchado is jumping into the frying pan. He is paid
less than the other workers, and treated badly. The ones who use the
system are those who do not have enough money to get to the place of
work, and who are not familiar with it. Those in the know take care of
their own arrangements, and find good work.' As the labor market in
the coastal valleys and eastern selva become better developed, the
enganche system is declining in importance.4
The income from temporary migration that has been estimated for
this study refers to /^/income, i.e. after deducting from the peasant's
monetary wages (including the advance) all the costs incurred in
obtaining this income. (As indicated above, these costs consist of
transportation, lodging, food, and the costs of obtaining a working
capital in order to be able to migrate.) In very few cases was this net
income negative. However, many peasants mentioned to us that they
did not leave the area to look for work because the cost of
transportation, food and lodging has increased considerably in recent
years (this being due to the inflationary process that Peru has
experienced recently). On the other hand, the wage rates that the
peasants received were, in several cases, inferior to those of
permanent employees at the same workplace.
Another question included in the survey concerned how the
income earned from temporary migrations was utilized. The great
majority answered that it was for current expenses, i.e. to increase the
amount of monetary income available to buy consumption goods and
inputs for agriculture. In a few cases the income was destined for a
specific expense, such asfiestas,paying a debt, constructing a house or
buying land. While the incursion into the labor market is sometimes
brought about by the necessity of some particular expense, it is more
than just that. Seasonal migration is fundamentally a regular, stable
activity carried out to cover the regular expenses of the family. The
peasant's migration is temporary, but the income obtained is part of
the family's permanent income. The relationship with the labor
market is, in sum, a social relationship.
It is principally the adult male who participates in migration. Adult
women do migrate, but in very small numbers. The children, on the
other hand, migrate in significant proportions. In Ancobamba, 53%
of the students surveyed left the community to work during the three
months of school vacations (January-March); in Sihuina, this
proportion was 33%, and in Huando, 38%. In all these cases the
proportion of males who migrated was larger than that of females.
The children who migrate are usually employed in agriculture and
services. The children of Huando travel to the Chanchamayo Valley
in the selva region and to the city of Huancayo; those of Sihuina go to
the Quillabamba Valley in the selva^ and to Cusco. In the cities the
children find work as waiters in restaurants, as domestic servants, or
as assistants in small shops. The income obtained is usually destined
to buy school uniforms and supplies. In this way the children finance
part of their own educational costs.
Markets for goods and other markets in the community

A brief review of the functioning of the markets in which the peasants
exchange goods will permit us to complete our picture of peasant
social relations. The communities of Jacantaya and Culta sell their A,
P and Z products principally in the regional fairs of Huancane and
Have, respectively. Peasants from surrounding communities attend
these fairs, which include livestock markets (q'atu). There also the
peasants make most of their purchases of goods of both urban and
rural origin. In both communities shops selling general merchandise
are scarce (one in Culta and two in Jacantaya) and small in size.
The communities of Sihuina, Huando and Acobamba are the sites of
regional fairs. For this reason, most of these communities' purchases
are made in local markets. There are around a dozen general stores in
each place, and several are quite large. In none of these communities is
there a livestock market, as in the case of Puno's altiplano fairs.
Animals are sold to merchants who visit the community periodically
or to local dealers.
In the three remaining communities (Ttiomayo, Ninamarca and
Ancobamba), the goods market is more dispersed. The peasants sell
their products to local or outside merchants (called rescatistas, or
'ransomers'); they also take things to the regional fair, like those from
Ninamarca who go on Sundays to Paucartambo, and those from
Ttiomayo who go to Andahuaylillas and Urcos. In addition, there is
trade in provincial or departmental capitals: Ninamarca with Cusco
and Ancobamba with Abancay, for example. The communities'
purchases of goods are also made at different fairs, including the
general stores of the community. Only in Ninamarca does no store
exist. The purchase of livestock is also done through merchants.
To summarize, there are internal and external markets for the
exchange of goods. In the internal market the peasants can sell their
A, P and Z products in the local fair, or to merchants who visit the
community regularly (and of course also to local 'ransomers'), and
they can buy products from the local fair or the shops that exist in the
community. In the external market, they can exchange their products
in the regional fair in a provincial or departmental capital. Finally, in
the livestock market there is a difference between the altiplano
communities, where the exchange is carried out at fairs, and the other
communities, where it is done through local merchants.
Not only goods are exchanged in the peasant communities. As was
mentioned earlier, services are exchanged between families in an
intensive manner. Apart from labor, there are also markets for land,
land rental, and credit. A brief account of these markets is in order.
The land market in the communities is quite limited. The purchase
and sale of land is carried out very sporadically, and usually only in
cases of emergency. The survey showed very few transactions in the
previous three years. With permanent migration of peasant families
to cities one would expect significant sales and purchases of land.
What we observe, however, is that the migrating families do not sell
their lands. Those who leave the community try to leave their lands
with relatives and still manage them indirectly. The idea appears to be
to hold on to the property in anticipation of any problems that may
arise in the city. This economic behavior will be discussed in more
detail in the next section. Land rental is another activity which is not
very common within the peasant economy. Elderly families tend to
rent out their lands when they can no longer work them and there are
no children around to take over. In other cases, families who migrate
definitively leave their properties under rental. Finally, in some
communities there exist plots belonging to the church or munici-
pality which are leased out.
The credit market appears to be more active than the land rental
market, although the relevant information was difficult to obtain.
Lending money is commonplace. These loans are requested for
emergencies, such as illnesses or burials, to pay for a community
fiesta, or in order to migrate or carry out a business transaction.
Interest is normally not charged directly in cash, but mostly in kind or
services. For instance, the borrower may become obligated to help
the lender in his plot with two days' work. In many cases credit is
obtained in the form of an advance for future work. In this way, the
credit market and the labor market become closely related.
Credit for agricultural 'working capital' is less common. This may
seem paradoxical if we take into account the credit needs implied by
the agricultural process. Nevertheless, there are two reasons for this
situation. In the first place, the agricultural process is self-replacing to
a great extent, since the farmer himself furnishes inputs such as seeds
and labor. In the second place, the necessary 'working capital' is
obtained from the gross monetary income which the peasant obtains
from different sources, such as the case of temporary migrations
which provide income for 'on-farm expenses'.
The logic of exchange in peasant communities:

market and personal relations
To summarize, in the peasant economies there is a considerable
exchange of productive services between families. There is also an
exchange of A,P and Z-goods. For this reason, there exist many
forms of exchange (reciprocity, barter, cash) and many goods and
services that are exchanged. The majority of goods and services have
market prices which are related to the prices in effect in nearby cities
or larger external markets.
Nevertheless, there is a characteristic of the transactions which
prevents us from speaking of market relationships in an unrestricted
way: prices are not always uniform. In many cases the prices vary
depending on the persons who trade; in other words, the exchange
relationships are not always impersonal. The communities are micro-
societies; they are too small for market relationships to eliminate
personal relationships entirely and to reach equilibrium market
prices.
Market relationships, being impersonal, begin and end with the
exchange of commodities in question (including money). There is an
immediate, short-term balance. When the relationships are personal,
they neither begin nor end with an exchange of a pair of commodities.
Thus, there may be an imbalance in the accounts in the short run
which will be resolved in the long run. One peasant may rent out his
oxen to another peasant at less than the 'market' price now, but later
he may request that the latter help him with his harvest when
everyone is busy and labor is scarce in the marketplace.
In short, in peasant communities market relationships generally
predominate, but they are limited to a certain extent by relationships
of a personal nature. This explains the dispersion in the prices (or
exchange rates) of the same good in a community. Although the
average prices in the community follow the changes that occur in the
markets which are external to the community, the adjustment process
is very slow in the community. This is also due to the existence of
personal relations.
Permanent emigrations and returns

In light of the fact that permanent rural-urban migration has been
pinpointed as one of the most important social phenomena in Peru
during the last decades, we decided to obtain information on this
topic in our survey. In particular, we wanted to clear up two issues:
first, what proportion of the children emigrate from the community ?
And secondly, how do they enter into the productive process of the
place to which they migrate?
Regarding the first issue, we were able to estimate the total number
Table 6.4 Residence of the children who left parents' homes (number of
persons per family)
Residence (A)
(A) + (B) ~100
Community Emigrated •(A) Community (B)
JAC 1.73 0.32 84.4

CUL 1.25 0.53 70.2
ANC 1.50 0.45 76.9
TTI 0.46 0.20 69.7
SIH 1.17 0.34 77.5
HUA 1.26 0.29 81.3
ACO 0.65 0.18 78.3
Note: No data were available for Ninamarca.
of children (both sexes) who have left their parents' houses and then
the proportion remaining in the community and the proportion
emigrating. Table 6.4 shows the results. The average number of
children migrating varies between 0.5 and 1.7 per family, depending
on the community. In the southern sierra as a whole there is an
average of one child (son or daughter) per family who has emigrated.
The majority of the children who have left their parents' homes
have also left the community. Table 6.4 shows that between 70% and
84% of the children no longer living at home have left the
community. For the southern sierra the average is close to 75%, i.e.
three out of four children emigrated. The figure of 25% (one of every
four children remaining in the community) can be taken as a measure
of the retention capacity of the community as regards the new labor
force. From these results we can infer that the problem of scarcity of
land (and other resources) in the communities is being resolved
through emigration. The low retention coefficient is a clear indication
that emigration is the peasant economy's answer to the problem of
relative overpopulation.5
The children who emigrate have one common characteristic in all
the communities: they leave at an early age, usually between 15 and
20. A principal motive for migrating so early is to continue studies.
This hypothesis is consistent with the fact that in communities with
secondary schools, the average age at emigration is higher. Thus
Huando has two schools which offer complete primary and secondary
levels and an average migrating age of 20. Ancobamba, on the other
hand, offers only primary education, and has an average migrating
age of 15.
Another fact which supports the hypothesis relating emigration to
supply of education is that the educational level of the migrants
increases after they leave the community. The average additional
schooling was found to be between one and two years. There are of
course individuals whose educational level remains the same after
emigration, even cases of illiterates who do not improve their
situation. However, at the other extreme there are persons who left
their community with barely any primary education, and who are
now studying at the university level.
The current residence of the emigrants is, in the majority of the
cases, in Lima or the capital city of a department (Huancayo,
Arequipa, Puno, Cusco). There are few cases of emigrants who live in
very small cities (such as mining centers), and even fewer migrations to
other rural areas. The general pattern is that the children of the
peasant family emigrate to large cities. That is the goal. This is shown
in Table 6.5.
The occupations of the migrant children are diverse, but with
certain identifiable patterns. An important fraction, around a third of
Table 6.5 Residence and occupation of permanent migrantsa (number of

persons in the sample)
Residence
Large Small
Occupation Lima city* cityc Rural Total
Wage-earner 26 26 12 8 72
Domestic servant 17 19 2 0 38
Self-employed 10 16 4 0 30
Housewife 11 19 8 0 38
Student 2 7 2 0 11
Other 10 9 2 1 22
Total 76 96 30 9 211
a
Includes the cases of Jacantaya, Culta, Ancobamba, Huando, Acobamba.
b
Includes Huancayo, Cusco, Arequipa, Puno.
c
Includes mining towns.
those who emigrate, are not part of the labor force. These correspond
to housewives, students and military personnel. Here we are speaking
of principal activities, since probably there are many cases of students
and housewives who partially participate in income-generating
activities.
Of those who are part of the labor force, the majority are wage-
earners, as presented in Table 6.5. Only around 20% of the emigrants
who work are self-employed (merchants, artisans, street vendors).
The majority of the working emigrants, then, enter urban job
markets rather than self-employment, as is often claimed. The
employment obtained is usually in small enterprises such as repair
shops, tailoring, crafts and construction. An important source of
employment, especially in the case of women, is as domestic servants.
Very few cases were found of emigrants who work in the factories of
the modern sector.
The survey included questions concerning returns of any member
of the family that had lived for more than a year outside the
community (this would give a person 'permanent migrant' status
according to our definition). The results appear in Table 6.6. The
families who do not have any member who has returned represent
various proportions of the total number of families in the com-
munity: in Huando only 17% and in Ancobamba half. The sample
average suggests that the majority of the families of the sierra have at
least one member who has returned after migrating 'permanently'.
This result indicates that the linkage of the peasant family with the
Table 6.6 Returns of members in peasant families (% of families in the

sample)
Members who returned

Standard
Community 0 1 2 3 Average deviation
JAC 26.5 29.4 32.4 11.8 1.29 1.00

CUL 37.5 40.0 15.0 7.5 0.93 0.92
ANC 50.0 35.0 15.0 0 0.65 0.74
HUA 16.7 33.3 47.6 2.4 1.36 0.79
ACO 42.5 32.5 25.0 0 0.83 0.81
Note: No data were available for Ninamarca, Ttiomayo or Sihuina.

rest of the Peruvian economy is quite diverse; it also indicates that the
peasant family's experience with the outside world goes beyond the
relations of exchange of goods and labor during the year. In the same
way, its degree of integration is greater than that which results from
measuring only annual monetary income.
The reasons for returning to the community are various, but some
patterns are recognizable. Of 110 cases of return for which we have
information, we observe in Table 6.7 that half were due to problems
related to the parents of the emigrant, whether they died, fell ill,
became invalids or were very old. This is the case when there are no
more children in the community who can help the parents or take
charge of the land and other resources. The other important cause of
the migrants' return is work. Of the 110 who went back to the
community, 22 did so because they lost their job, had a low income,
or could not adjust to the work (principally the case of those working
in mining). There was also a significant number of returns (14) due to
illness. Finally, there were 14 cases of return due to problems with the
landholdings which the emigrants had left in the community. The
problems include those arising from litigation, squatters and the
agrarian reform.
Although the data in Table 6.7 are not sufficiently statistically
significant to study the problem of returns and arrive at more solid
conclusions at the regional level, they do give an indication of certain
behavior patterns in the peasant family. The cases of return due to
Table 6.7 Reasons for returning (number of persons in the sample)
Communities
Reasons JAC CUL ANC HUA ACO Total
1. Losing the job 3 2 0 1 3 9

2. Low income 2 0 2 0 0 4
3. Maladaptation 0 0 3 4 2 9
4. Sickness 3 0 0 10 1 14
5. Problems with land 2 1 6 4 1 14
6. Problems with
parents 22 12 7 11 2 54
7. Others 4 2 0 0 0 6
Total 36 17 18 30 9 110
work problems or illness (cases 1—4) require that the migrant have a
place to which to return, i.e. that he has his land reserved in some way.
In effect, many declarations of the peasants, as well as personal
observations in the communities, indicate with some certainty that
the migrant does not divorce himself from his plot of land. He does
not 'burn his bridges' upon leaving. As was indicated in the previous
section, there is no land market which emerges from emigration. The
parcel of land thus comprises an income safeguard for the family who
emigrates. One can always return to the farm. Thus, during the
period of the survey, some individuals were returning to the
community because of the current economic crisis in Peru, which is
more severe in the cities.
Reasons 5 and 6 reflect behavior complementary to that of
maintaining one's lands in reserve. Here the issue is not losing title to
the parcel. The death or incapacity of the parents endangers the
possession of the land on the part of the family or heirs. There are
many legal actions in the communities stemming from land, and one
way of avoiding these problems is for a son or a daughter to be
present. In several cases we found persons who had returned as a
result of pressure from other children of the family who could not go
back to the farm so easily. Someone had to return to take care of the
land, animals and buildings; if not, someone else might take it all
away. The reasons for a migrant returning to resolve problems
relating to his own land are even more obvious.
In the reasons for returning we may, in consequence, recognize a
certain strategy in the economic behavior of the peasant. He maintains
links with the community, where he has a 'reserve' consisting of his or
his parents' lands, and to which he can return if he meets failure in the
outside world. The strategy of the family is thus to avoid the risk of
losing the land.6 The phenomenon of emigrations is usually seen as a
flow of people in one direction, from the countryside to the city. The
evidence presented here shows that there is an important feedback in
terms of individuals returning from the city to the country.
Emigrations and returns are movements guided by the economic
strategy of the peasant family.
7
Stagnation in the peasant economy and
the role of demand
The previous chapters have shown production and exchange struc-

tures in today's peasant communities in Peru. The peasant economy
was observed at a given point in time (1978-9). This chapter will
attempt to depict the dynamic of the peasant economy in Peru in the
recent decades. If this is an economy very much integrated to the
market system, has economic growth experienced in the Peruvian
economy reached the peasantry ? Have rapid population growth and
urbanization induced growth in peasants' income through higher
levels of demand for food ?
Growth and distribution in Peru: 1948-75

The income distribution problem presents peculiar characteristics in
the underdeveloped countries. It not only refers to the problem of
inequality, but also includes the problem of absolute poverty. For this
reason, in order to evaluate the performance of an underdeveloped
economy in terms of the distributive problem, it is necessary to
determine the changes which have occurred both in the degree of
inequality as well as in the proportion of the population still
living in conditions of absolute poverty.
In the last three decades, Peru has experienced various patterns of
economic growth, initially based in export of primary products,
followed by import substitution industrialization. Also, it has
undergone periods of both rapid growth and recession. On the other
hand, Peru passed through diverse socio-political contexts. From the
initiation of the government of General Odria in 1948 until 1968,
economic policy was basically liberal. Certainly one can draw
distinctions concerning the degree of liberalism manifested: it was
much more marked in the Odria and Prado regimes; there was a
greater degree of State intervention in the first Belaunde government.
82
Economic stagnation and the role of demand 83
However, overall, the economy in this period (1948-68) can be
characterized as 'free enterprise' or laissez-faire•, wherein the role of
the State in the productive process was very limited.
The military regime of Velasco (1968-75) was characterized, in the
economic domain, by three traits: a reform program treating land and
capital property; a growth program based even more strongly in
import substitution industrialization; and a greater State role in
economic activity, particularly via the formation of public enter-
prises. The Morales period (1975-80), the 'second phase' of the military
government, constituted a period substantially different from that of
Velasco. There were neither reform programs nor growth strategies.
The economic crisis became the central economic problem and
governmental action centered itself around stabilization policies. The
second Belaunde government, inaugurated in July 1980, denotes a
return to liberalism on the part of the State.
The period of economic liberalism: 1948-68

From the immediate post-war era until the final years of the 1960s,
Peru experienced rapid economic growth with price stability. This is
depicted in Table 7.1. The real per capita income increased at an
average annual rate of 2.5% and the average annual inflation rate was
8%. The growth pattern during the Odria and Prado regimes was
based in primary product export, fundamentally of agricultural,
mining and fishing products. Economic policy was directed prin-
cipally toward providing incentives for these exports. Economic
growth was led by the export sector.x The industrial growth observed
in those years was more the result of its technological link to
processing and support of exports (Thorp and Bertram, 1978).
Nevertheless, after the first Belaunde government there began a
change in the growth pattern, toward one based in import sub-
stitution industrialization. This change can be seen in Table 7.1,
comparing the relative stagnation of exports and the relative growth
of the manufacturing sector in the first Belaunde regime.
In order to analyze the distribution problem in Peru, it is necessary
to distinguish three sectors in the productive structure. The large,
modern firms (whether urban or rural) constitute the so-called
'modern sector', and in them capitalist relations of production are
predominant. The 'urban traditional sector' and the 'rural traditional
sector' are composed of the rest of the workers in each area. In these
latter two sectors the small firm and the family economic unit are the
Table 7.1 Peru: growth, inflation, distribution and pattern of growth,
1950-80 (annual rates and %)
Odria Prado Belaunde Velasco Morales

1950-6 1957-62 1963-8 1969-75 1976-80
Growth
Real GNP per capita 3.7 2.9 1.1 2.9 -1.5
Inflation
CPI - Lima 6.6 8.0 12.8 10.7 51.3
Growth pattern
Exports/GNP 15.6 21.2 21.0 16.0 14.8
Manufactures/ GNP 14.0 16.5 18.8 20.5 20.6
Distribution
Profits/Nat, income 15.5 13.3 16.1 20.1 28.4
Growth in average real
income of workers
Modern sector 3.4 3.0 -5.8
Urban traditional 1.7 1.2 -2.2
Rural traditional 0.9 1.0 -2.0
Source: See Appendix IV.
predominant forms of organization of production. About 1961, the

'modern sector', empirically defined as firms with more than five
workers, employed 20% of the workforce; in the 'urban traditional
sector' were 30%; in the 'rural traditional sector' were 50% (Webb,
1977). By far the largest social group in the 'traditional rural sector'
was constituted by peasant families living in the sierra. Given this
economic structure, four income categories appear appropriate for
the study of the income distribution process in Peru: profits and
wages in the 'modern sector' and incomes in the urban and rural
'traditional' sectors, which are composed mainly of wages and
income from self-employment.
Webb (1977) demonstrated that toward the beginning of the 1960s
the composition of the income pyramid was the following: the top
1% was composed of owners of the means of production and of
professionals; the next stratum comprised high-level employees in
the 'modern sector'. These groups constitute what may be called the
Peruvian 'middle class', which is approximately 5% of the popu-
lation.2 The rest of the highest quartile consisted of workers and
employees of the 'modern sector'. The relatively privileged position
of these wage and salary earners is a consequence of the fact that the
owners of the means of production, together with the 'middle class',
constitute a very small group in Peru and this does not signify that
they have a high level of living. The workers of the 'urban traditional
sector' are located in the following stratum in the income pyramid.
Together the above groups constitute the upper half of the income
pyramid. In the lower half we mainly encounter peasant families. The
lowest 30% was constituted almost exclusively by peasants who live
in the sierra.
An estimate of income trends for these groups during the period
1950-66 was also made by Webb. His results indicate that the
average real income of the 'modern sector' increased by 4% annually;
that of the 'urban traditional sector' increased by 2% annually and
that of the 'rural traditional sector' increased by only 1% annually;
while the peasants of the sierra experienced practically no gain in real
income during this period (Webb, 1977: Table 3.7). On the other
hand, the average profit income of owner-families increased in a
greater proportion than did the average income in the nation.3
Therefore, during the liberal period the process of economic growth
mainly benefitted the upper quartile of the income distribution. The
group which benefitted least consisted of the neediest families.
Congruent with the liberal economic context, the most important
redistributive policy during the period was conducted via the govern-
ment budget. In fact, the budget as a proportion of the GNP grew
from 12-13% in 1950 to 17-18% in 1968 (Banco Central de
Reserva, Cuentas Nacionaks). Nevertheless, as Webb also demon-
strated, the net result of the transfers via the government budget was
moderately progressive. Despite the fact that greater tax receipts
came from the 'modern sector', the greater part of government
expenditure also went to families of this sector. Income redistribution
did not change the results of the market.
The period of reforms: 1968-75

The Revolutionary Government of the Armed Forces initiated in
1968 a series of reforms, the intent of which was to alter the structures
of Peruvian society. One of the objectives of the military regime was
to carry through a radical policy of income redistribution. Given the
degree of inequality which exists in Peru, income redistribution is
clearly a policy which benefits those at the base of the income
pyramid. The empirical basis for this assertion arises from Webb's
study: a selective transfer of 5% of the national income, taken from
the upper 1 % and given to the lower tercile of the population, would
reduce the absolute income of the former group by only 16%, but it
would double the incomes of the latter group. In contrast, a very
successful economic growth effort consisting of a sustained growth
rate of 3% annually would require more than 20 years to achieve the
same improvement in the income of the lowest tercile (as the
hypothetical redistribution discussed above). Therefore, and con-
trary to the belief held by many, to redistribute income in a selective
manner in Peru does not imply a redistribution of poverty.
The most important reforms having a direct impact on the
redistribution of income were the sectoral reforms. In the mining,
fishing and industrial sectors these reforms consisted of giving
workers the right to participate in profits, ownership and manage-
ment of the enterprises. The reforms created 'Labor Communities'
{Comunidades Laborales) formed by workers from each firm, with the
right to participate in the profits of the firms in the following
proportions: 25% for industry, 20% for fishing and 10% for mining.
Approximately one-third of the participation was distributed in the
form of income and the balance was provided as an acquisition of the
firm's patrimony in the name of the respective 'Labor Community'.
The enterprises affected by the reform were those of the 'modern
sector' and its beneficiaries, the workers of this sector. The latter
group, as was indicated above, belongs to the richest quartile.
The Agrarian Reform Law, promulgated in 1969, sought to
redistribute land ownership, eliminating indirect forms of agrarian
exploitation so that the 'land belongs to those who work it'. The
affected lands were to benefit: (1) workers organized in cooperatives;
(2) peasant communities; (3) individuals; (4) a special economic
entity, called SAIS, which is a mixture of cooperative and peasant
communities. The initial goals of the reform, both in terms of number
of persons as well as of quantity of land affected, were practically
achieved by the end of 1979. The reform transferred approximately
40% of total land area to 360,000 families (Caballero and Alvarez,
1980: Tables 1 and 6). These beneficiaries represent around 25% of
all the rural families in the country.
Who were the beneficiaries of the Agrarian Reform? In terms of
families, 27% were ex-wage-earners, the majority of whom were
located in the modern farms on the coast; 25% were ex-workers of
the hacienda system, the majority of whom were located in the sierra;
38% were peasant families who lived in peasant communities; the
balance (10%) were 'individual' families (Caballero and Alvarez,
1980: Table 6). But these percentages do not reflect the real
reassignment of lands between families due to the difference in the
size of the lands assigned to each group. The study by Caballero and
Alvarez demonstrates that in the case of peasant families the average
land size adjudicated per family is one-fifth of the total's average
adjudication (ibid.). Thus, the income transfer to peasant families was
very small: the reform reached only 14% of all peasant families and
even for this 14% the quantity transferred per family was much lower
than the average per family transfer given to other groups. Actually,
those who most benefitted were the ex-wage-earners of the large
farms on the coast which belonged to the 'modern sector' and, for this
reason, to the richest quartile in the income pyramid.
The redistributive effect of all the sectoral reforms was the transfer
of around 2% or 3% of national income from the upper percentile to
approximately 18% of the labor force (Figueroa, 1975). This income
transfer took place almost completely within the richest quartile. The
reason is simple: the reforms were concentrated basically in the
'modern sector' of each of the sectors involved. The redistribution of
income within the 'modern sector' — characterized by extreme
heterogeneity in productivity levels among firms - was biased
against the 'traditional sector'. The great majority of workers in the
latter sector, that is, the poorest groups, were left outside the design
of the reforms. The formation of public enterprises, on the other
hand, constituted a restructuring of ownership in the 'modern
sector', without significantly affecting the capitalist rules of pro-
duction and distribution in the economy (Baer and Figueroa, 1981).
The overall Peruvian economic performance during the reform
period was similar to the liberal period as much in terms of economic
growth as in price stability. The average growth rate of per capita real
income was 2.9% annually and the inflation rate was 11 % annually, as
shown in Table 7.1. As for the growth pattern, in this period
significant emphasis was placed on industrial growth via import
substitution. The industrial policy was excessively protectionist, not
only because of the high tariffs but also because of the almost massive
prohibitions on industrial good imports which competed with those
produced domestically.
As a result of the economic growth, of the change in the pattern of
growth and the redistribution of income via the sectoral reforms, the
nature of income inequality showed at the same time change and
continuity with respect to the liberal period. The element of
continuity was that the growth of average worker income in the three
sectors was differentiated: in the 'modern sector' it grew at a greater
rate than the 'traditional sectors', as shown in Table 7.1. In this sense,
inequality between workers increased. The element of change was
that the participation in firms' profits increased in the national
income: from 15-16%, it passed to 20%. Paradoxically, in the period
of reforms and greater state interventionism, the share of profits in
national income increased. Surely, one reason for this increase can be
found in the development pattern based on excessive industrial
protectionism.
Capitalist development and rural incomes:

nature of the problem
The result that peasants' real incomes have remained almost stagnant
in the period 1950-75 challenges the common view that economic
growth in the capitalist modern sector can induce an overall increase
in real incomes in the non-capitalist system of the economy. The
explanation for this failure of the capitalist development in Peru
cannot rely on the hypothesis of 'dualism'. Stagnation in the real
income of peasants is not due to the lack of integration into the
capitalist economy. This study has shown that the peasant economy is
very well integrated into the market system. The outcome must be
explained in terms of the working of the market system and the effects
of economic policies.
On the other hand, the outcome of the land reform program
indicates that exploitation of peasants through non-capitalist re-
lations, that is, through non-wage relations, is not of great impor-
tance. Land reform was directed to the elimination of feudal-type
relations on the sierra and only 10% of the peasantry was part of the
program. Peasants are connected to the rest of the economy basically
through market relations.
Why is it that capitalist development has not generated overall
development — in particular, improvements in the standard of living
of the peasantry ?
One of the main features of economic development in Peru and in
Latin America is the continuous decline in agricultural output to total
output. The share of agriculture in the Gross National Product of the
region has declined by almost a half in 35 years: from 25% in 1940 to
12% in 1975 (UN-CEPAL, 1978). In the case of Peru these ratios
declined from 23% in 1950 to 12% in 1980 (Peru, Banco Central,
1968, 1981). The decline of the share of agriculture in GNP is not an
economic problem per se. Actually this change in the production
structure has also taken place in the developed countries. However,
what is specific to the present Latin American historical stage is the
fact that the relative decline in agriculture is associated with the
prevalence of rural poverty. In Latin America poverty is concentrated
basically in the rural areas.4 Thus, a change in the share of agriculture
in national income has direct implications for income distribution
and, even more importantly, for absolute poverty.
The usual explanation for the decline in the share of agriculture is
that income elasticities for food consumption are very low. This is the
well-known Engel's law of food expenditure. In the case of Peru and
Latin America, however, a rapid increase in population is taking place
which must have a bearing on the demand for food. Given the same
average income, expenditure for food must increase at the same rate
as population growth. To this 'population-effect', we should add the
'income-effect' due to the increase in average income brought about
by the economic development process. Engel's law will apply to the
second effect, whereas the 'population-effect' will be shifting the
Engel curve upward through time. The figures presented for Peru in
the period 1950-74 indicate a rate of growth of approximately 5%
per year in food expenditure in real terms. This is a result of using
2.8% for population growth and 2.5% for per capita real income as
annual growth rates, and 0.7 as the income elasticity.5 However, the
actual average growth rate of agriculture has been only 2.8% for the
same period.
The other feature in the process of development in Peru (and in
Latin America) is the rapid urbanization process. Although the total
demand for food may not be greatly affected by changes in the
urban—rural distribution of population, these changes will increase
the links between the city and the countryside stemming from food
consumption. Moreover, with rapid urbanization, per capita income
in the rural sector should be increasing rapidly as more people will be
spending on food and relatively fewer people will be producing it. If
we consider again the figure for Peru of 5% as the 'theoretical'
agriculture output growth rate and take 1% as the growth rate in
rural population, per capita real income should be growing at a rate of
4% in rural areas. Clearly this is not happening in Peru, at least as a
long-run trend.
The two characteristics of development in Peru - population
growth and urbanization — lead us to expect a much more dynamic
agriculture than is suggested by the income elasticity explanation.
Yet, we observe a rather slowly growing agriculture. This result must
come either from a change in consumption patterns — that is, shifts in
aggregate Engel curves for food expenditure — or from a change in the
production structure of the economy due to changes in the
international division of labor, because in open economies agricul-
tural output depends not only on domestic consumption patterns
but on patterns of specialization in international trade as well.
Alternatively, the rate of growth may be due to a combination of both
effects. In order to analyze these effects a theoretical model is
developed in the next section.
A model of derived demand for rural factors

Given the rapid increase in urban population and also in average
urban income in Latin America, it is clear that the demand for food
comes mainly from the cities. In this section, a simple model is
constructed in order to determine the relationship between urban and
rural incomes derived from the urban consumption of food. The
income generated in the rural sector due to food expenditure in the
cities will be called the 'derived rural income'. The economy will be
divided into three productive sectors: two food-producing sectors,
agriculture and the food-processing industry, and all remaining
sectors which will be called 'rest'. There are three types of food: the
two corresponding to the domestic sectors mentioned above and
imported food. The productive sectors are inter-related, as shown in
Table 7.2, which can be seen as an input-output table of the Leontief
type.
The total rural income (or value added) generated, directly and
indirectly, by a unit offinaldemand in each sector will require to take
into account all the inter-relations of Table 7.2. Solving for all the
'rounds' involved in the process of production, we get:
F ^ 1 = <,01[^11(C*+D1) + ^ 1 2 ( q + D , ) + ^ 1 8 D,] (1)
where aQ1 is the direct coefficient of value added per unit of gross sale
Table 7.2 Input-output table for food production-consumption analysis
Final demand
Agri- Food Food

culture process 'Rest' consump. Others Total
Agriculture I
Food processing
x c\
'Rest'
Imports: food
l 0
ci 0
Imports: others xml 0 Dn
Value added
Total gross value

v\ X
3
in agriculture; the A ±js are the coefficients of direct and indirect

requirements of commodity 1 to produce one unit offinalproduct of
sector/, which come from the solution of the Leon tief system. From
equation (1) we derive total coefficients of value added as:
where BQj = aQ1Axj. Thus, B measures the amount of increase in

agriculture income due to an increase in one unit of food con-
sumption of type 1, that is agricultural food; B measures the same
effect for the case of processed food. Coefficient B does not
concern us here because it is connected to non-food consumption. We
shall assume here that imports and exports are independent, so there
will be no rural income generated from consumption of imported
food (C*).
The consumption expenditures in the three types of food can be
written as follows:
where Cx stands for the expenditure on agricultural food, C2 for

processed food, and Cm for imported food. Since this is the value of
family expenditure in the city, not all this expenditure goes to the
rural sector. In Table 7.2, values are at producers' prices (Cf) but in
equation (3) they are at consumers' prices (Cf). The difference
between the price the producer gets and the price the consumer pays,
usually called 'trade margins', will be assumed to be a fixed amount
for each type of food: tvt^ Therefore, the amount of expenditure
going to the rural sector would be:
C*=(1-/ 1 )C 1 + (1-/ 2 )C 2 (4)
where Cf corresponds to the consumption column of Table 7.2.

From equation (2) we know that for each unit of sales of
agricultural goods, rural income increases by B , and by BQ2 in the
case of processed food. Then, total rural income (Iur) derived from an
urban family expenditure on food will be:
(5)
If we represent equation (5) as a ratio of the family's income (JJ, we
will get the 'derived rural income' ratio:
(6)
This ratio is a measure of the income generated in the rural sector due
to the urban family's food expenditure. It is the rural income content
of the urban family's expenditure. The higher the value of this ratio,
the stronger the connection between urban—rural incomes.
Figure 7.1 summarizes the relations established in the model. The
vertical axis measures expenditure and the horizontal axis income.
The Engel curve for food expenditure is labeled E. From this curve
we derive the proportion of the expenditure which becomes rural
income. Curve R shows this 'derived rural income', which also varies
with the level of family income. The 'derived rural income' ratio is
just the proportion of income that is spent on rural factors of
production, e.g. AN/OA, when family income is OA, in Figure 7.1.
Let us call a the ratio of 'derived rural income' with respect to
average family income in the city (JJ. Thus:
'Total derived rural income' represented by Yur as a proportion of

total urban income (Yu) will also be equal to a. The other ratio of
Expenditure 45°
O A A' Income
Figure 7.1 Engel curve and derived rural income curve
interest is the 'total derived rural income' per head in the rural sector.
Let us call Nu and Nr the populations of the urban and rural sectors,
respectively. Then:
Equation (8) says that the per capita income generated in the rural
sector by the urban expenditure on food depends on three factors: the
'derived rural income ratio', the average urban income, and the ratio
of urban population to rural population. These last two factors are
increasing rapidly in Peru and therefore rural per capita income
should be increasing fast. However, a reduces the effect of those
factors because it decreases as average income goes up. We need to
know more about the determinants of a and its tendencies in order to
assess the combined effect of the three variables on the growth of
rural incomes.
As shown by equation (6), the 'derived rural income' ratio depends
on four sets of coefficients: the average propensity to spend on food,
trade margins, production structure, and the mix of types of food
consumed. In the next section, an estimate of these coefficients will be
made for the case of Peru. This will provide us with an application of
the model as we will estimate empirically curves E and R shown in
Figure 7.1.
An empirical estimate for 1970
The availability of an input-output table for Peru and a family budget
study for its largest city (Lima) for the same period (1969-70) will
enable us to make an empirical estimate of equation (6). The
input-output table has been rearranged according to the sectors
defined in the model. The relevant segment of the rearranged table is
presented in Table 7.3. The direct coefficients computed from this
table are:
aQ =0.7170
a11 = 0.1896 * 2 1 = 0.0522
a12 = 0.1626 a22=0A246
The corresponding B^ coefficients of equation (2) are:
F . 4 1 = 0.8940C*+0.1635C* (9)
Thus, for each sol of sale of agricultural food, 0.89 becomes rural
income; whereas for processed food this becomes only 0.16.
The family budget data for Lima comes from a study carried out
within the ECIEL Program. The structure of family spending by
income quartiles is presented in Table 7.4. Families in the lowest
quartile spend 54% of their budget on food, whereas in the richest
quartile this proportion declines to 31%. The average expenditure
ratio in Lima is 43%.
In spite of the recurrent discussion on the issue of trade margins,
there are no systematic studies in Peru showing the magnitudes
involved. Here we have made some gross estimates based on
Table 7.3 Peru: production structure for food, 1969 (producer's price,
billions of soles)
Agriculture Processing Consumption
Agriculture 6.9 4.7 21.8

Food processing 1.9 3.5 19.2
Rest 1.0 5.6 0
Imported inputs 0.5 4.2 1.5
Value added 26.1 10.9
Total gross value 36.4 28.9
Source: Instituto Nacional de Planificacion (1973).

Table 7.4 Lima: structure offamily spending, 1969 (%)
Quartiles
Item I II III IV Total
Food 54 47 41 31 43
Housing 16 17 21 26 20
Durables 8 10 9 12 9
Clothing 7 7 9 8 8
Transportation 3 6 4 6 5
Others 12 13 16 17 15
Total 100 100 100 100 100

Average expenditure 49 61 100 191 100
Derived rural income ratio0 19 17 15 11 15
a
See text for the methodology used in these estimates.
Source: Figueroa (1974: Table 9).
fragmentary information on particular goods and also on the data

contained in the original input-output table for Peru. For agricul-
tural goods, the coefficient of trade margins has been estimated at
33% and for processed food at 20%. 6
Regarding the mix in which the three types of food are consumed,
we will assume that all families consume them in the same proportion.
There are no data available on these proportions. This assumption
leads to an overestimation of processed food and imported food as a
proportion of total food expenditure for low-income families, which
should not be exaggerated, however. An important proportion of
'popular food' in Peru is imported and/or processed, such as noodles,
bread, oil and fats, milk and sugar. The proportions to be used as
estimates will accordingly be those appearing in the last column of
Table 7.3: agricultural food 51 %, processed food 45% and imported
food 4%.
To summarize, the estimates that have been made so far are:
tx = 0.33 B = 0.89 CJC = 0.51
/2 = 0.20 £ 0 2 = 0.16 CJC= 0.45
As these coefficients are independent of the income bracket of the
family, the 'derived rural income' for families in different quartiles
will vary according to the average propensity to spend on food. Thus
for the lowest quartile of Lima we have:
^ - = 0.54(0.66 x 0.89 x 0.51 + 0.80 x 0.16 x 0.45}

= 0.19
Hence, the poorest quartile of Lima spends 19% of their family
budget on rural factors of production. This proportion for each
income quartile is shown at the bottom of Table 7.4. The average
coefficient for Lima turns out to be 0.15, which means that only 15%
of Lima's total income goes to the countryside as rural income as a
consequence of food consumption. In sum, the first row of this table
shows the empirical Engel curve and the last row the 'derived rural
income' ratio. These are the empirical counterparts of curves E and R
presented in Figure 7.1 in terms of ratios.
This empirical result shows, first, that the 'derived rural income'
ratio is relatively low. The average proportion of family budget spent on food
is 43% in Lima, jet only 15% becomes rural income. If OA were average
income in Figure 7.1, NA would be a third of MA in the case of
Lima. Secondly, the 'derived rural income' ratio declines as family
income rises. Our estimation procedures have led us to a constant
vertical distance between the Engel curve and the 'derived rural
income curve', which need not be the case if the consumption mix of
different types of food varies with income.7
'Derived rural income' trends and income concentration
The main purpose of this and the next section is to present some
hypotheses to explain the slow growth of rural income in Latin
America. One coordinate in the analysis is the intensity in the derived
demand for rural factors of production. For the case of domestic
demand for food we have developed a simple model to relate urban
incomes and expenditures to rural incomes (export demand will be
discussed in the next section). The model was summarized in
equation (8). From that equation, we can advance some hypotheses
on the possible trends of the relevant variables involved so as to
estimate changes in the 'derived rural income'.
It is clear from equation (8) that, other things being equal, a process
of urbanization increases the intensity of demand for rural factors.
More people of equal average income in the cities will be generating a
higher 'derived rural income' which in per capita terms will increase
even more because of the rural population's relative stagnation. If for
each family in the countryside there is a family in the city, this family
will be generating income in the amount alu for his fellow campesino;
however, if there is now a process of urbanization by which for each
family in the countryside there are two families in the city, income
accruing to those rural families will double.
To this 'urbanization effect' we must add the 'pure income effect'
due to the increase in the average income (lu) in Latin American cities,
which is brought about by the economic growth process in the
economy. That increase will augment income for the rural people but
not proportionally, due to Engel's law; that is, the a ratio will decline.
The question is, however, whether a will decline due only to Engel's
law, or whether other variables affected by the particular process of
economic growth could also change a.
Let the curves E and R in Figure 7.1 represent the aggregation of
all individual curves, so that the axes now measure average income
and expenditures. For the E-curve to represent a given pattern of
consumption income distribution must be held constant along the
curve, only in this case the elasticity of the aggregate Engel curve will
have the same value as the individual Engel curve. The 'pure' Engel's
law will then apply to a given pattern of consumption, that is to an E-
curve where income distribution is held constant. Therefore, changes
in income distribution will cause a shift in the aggregate Engel curve.
The economic growth process in Peru has, as shown above, been
accompanied by a higher concentration of income. The effect of the
more unequal income distribution is a downward shift in the E-
curve, which in turn will shift downwards the R-curve. This result
comes from differences in average and marginal propensities to spend
on food between income groups.
The other effect of higher income concentration upon the 'derived
rural income' comes from differences in the mix of types of food
between income groups. It is expected that marginal propensities to
spend on imported food are higher for the rich. The same can be said
about processed food.8 Given the lower content of rural factors in
these latter types of food, as illustrated above, the effect is to shift the
R-curve downward. One implication of the higher concentration of income in
Latin America is, therefore, a further decrease in the 'derived rural income*.
The effect of change in patterns of international trade
In open economies, agricultural output depends not only on domestic
consumption patterns but also on patterns of specialization in
international trade. Therefore, if domestic patterns of food con-
sumption have not created enough stimulus for higher rural incomes,
there exists the alternative that foreign demand might have. After all,
the international division of labor has usually been presented as that
of developed countries producing manufactures and less developed
countries producing agricultural products. The fact is, however, that
foreign demand for agricultural commodities produced in Peru has
not been dynamic. The share of agriculture in total exports from Peru
has declined from 55% in 1950 to 20% in 1974 (Banco Central, 1968
and 1976). On the imports side, Table 7.5 shows an increase in the
proportion of food imports to total production for the basic
foodstuffs between 1943 and 1975.
The changes in the structure of exports and imports in Peru are
certainly a reflection of changes that have been taking place in the
international division of labor, particularly after World War II. First,
in this period world exports have grown faster than GNP in both
developed and underdeveloped countries, which means that the
world as a whole has seen its degree of integration increased.
However, export growth is faster in developed countries (Donges,
1979). Secondly, in terms of exports of agricultural commodities, the
developed countries' share has increased from 49% in 1955 to 61% in
1975, whereas the share of the developing countries has decreased
from 40% to 26%. Currently, the United States supplies 60% of
Table 7.5 Peru: importsjdomestic output ratios for basic imported

food (%)
1943 1960 1965 1970 1975
Wheat 49 70 76 84 86
Corn n.a. n.a. 0 1 52
Oilfats 4 35 34 80 80
Milk 3 22 22 35 41
Beef 2 4 12 28 11
Rice 10 10 32 0 12
Barley 2 6 10 10 32
Source:Lajo (1979).
grains sold in the international market. The less developed countries,
as Donges points out, 'have lost ground in fields where their resource
endowments should have given them a comparative advantage: in
food products, agricultural raw materials...' (p. 11). Thirdly, ex-
ports of manufactures from the underdeveloped countries are
increasing their share in world trade. These changes in patterns of
trade can be seen in Table 7.6.
An explanation for the change in the comparative advantage in
agricultural products lies in the tremendous increase in agricultural
productivity in developed countries. Agricultural economists have
argued that this increased productivity is due to a change from
resource-based agriculture to science-based agriculture. Hayami and
Ruttan (1971), for instance, have reported the fact that for several
products agricultural productivity differences between the developed
and less developed countries have widened. They conclude: 'The
basis for comparative advantage shifted from natural resource
endowments to the endowments of scientific and industrial capacity.
The shift in comparative advantage in agricultural production from
the less developed to the developed countries was accelerated after
World War IF (p. 242). As a result of this technological change, land
has increased in terms of efficiency-units in developed countries
which have become relatively land-abundant.
In addition to this shift in natural comparative advantage,
economic policies have pushed relative prices in the same direction:
developed countries protect agriculture whereas less developed
countries protect manufacturing. Agricultural prices supports, tariffs
and quotas on imports are clear cases of the former. In less developed
countries policies for import substitution industrialization have led to
a decrease in agricultural relative prices.
All these shifts in comparative advantage (natural and induced by
economic policies) have resulted in a change in international trade
patterns. The significant decline in the share of agriculture in total
exports, together with the fact that the share of imported food has
increased with respect to total agricultural output, indicate that Peru
has, in fact, lost comparative advantage in agriculture. Mostly,
agriculture in Peru is still based on natural resource endowments.
A consequence of this shift in the international trade pattern for
agricultural development in Peru is that the production structure of
the national economy shifts away from agriculture. The level of
agricultural output decreases relatively. This is the result of the slow
Table 7.6 Changes in world export structure (%)
Developed countries Developing countries0
Total Excluding OPEC
SITC 1955 1970 1975 1955 1970 1975 1955 1970 1975
Food and related products 0+1 + 22 + 4 48.7 59.0 63.3 42.6 31.8 28.7 41.0 29.3 27.5
Agricultural raw materials 2 excluding 22,
27, 28 49.4 58.4 61.3 40.4 30.3 26.2 38.5 25.6 24.3
Crude fertilizers and minerals 27 + 28 52.6 58.0 54.3 33.0 31.4 32.8 31.5 28.0 29.7
Mineral fuels 3 31.7 26.5 17.4 57.5 63.1 73.9 9.0 10.1 11.0
Chemical products 5 88.1 88.9 87.3 5.1 3.9 5.4 5.0 3.7 4.6
Machinery and transport
equipment 7 86.6 87.6 87.1 0.7 1.6 2.8 0.6 1.5 2.7
Iron and steel 67 86.6 82.5 86.5 0.9 3.3 2.7 0.9 3.1 2.5
Non-ferrous metals and products 68 59.2 63.6 67.9 33.9 29.0 22.0 33.5 28.6 21.2
Other manufactures 6 + 8 excluding
67, 68 82.6 79.9 78.1 8.8 11.2 13.4 8.7 10.9 13.0
Total export 0-9 64.7 71.9 66.2 25.4 17.6 24.1 22.5 11.8 11.4
"Excluding centrally planned economies.

Source: UNCTAD, Handbook of International Trade and Development statistics, 1976; U N , Monthly Bulletin of Statistics, May 1977. Taken
from Donges (1979), p. 12.
growth in exports and, on the domestic demand side, the increase in
the ratio of imports to total production. An increase in this ratio shifts
the R-curve further down because a shift in this curve also comes
from a process of saving rural factors of production, that is by
reducing the BOj coefficients of equation (6). It is an import
substitution process in reverse.
Another effect to be examined is on the structure of the agricultural
production. The question is, what has happened to the growth of
output in the peasant economy? A study made by Hopkins (1981)
shows that in the period 1944-69 agricultural output in Peru grew at
the average rate of 2.8% per year, but output produced basically in
the peasant economy of the sierra grew only by 0.5%. The situation in
the first half of the 1970s must have been very similar.9 Total
agricultural output grew at similar rates and no change of significance
was observed in the peasant economy to expect a different growth
rate in its output.
In conclusion, in the period 1950—75 there was a change in the
level and in the structure of the agricultural production in Peru. The
relative stagnation was more severe in the peasant economy's output.
This change in production structure implies that the peasant
economy plays a less important role as supplier of food for domestic
consumption over time. Food supply for urban markets comes
increasingly from the medium- and large-sized farms and, above all,
from imports.
8
Economic crisis and the peasant
economy, 1975-1980
The second half of the 1970s has been a period of economic crisis in
Peru. The real GNP per capita of 1980 is almost 10% less than that of
1974. The annual rate of inflation has varied between 24% and 70%.
Neither the decrease in real average income, nor the rates of inflation,
nor the duration of the crisis has any parallel in the recent economic
history of Peru. Since 1950 two short periods of crisis are re-
corded (1957-9 and 1967-9) but they were very mild compared to
the present period. This chapter attempts to investigate the effect of
the crisis on the peasant economy. To what extent has this economy
been affected?
The extent of the crisis

In the post-World War II period until the mid-1970s Peru experien-
ced sustained economic growth together with price stability. The real
income per capita in 1974 was twice as high as in the 1950s. Annual
rates of inflation never exceeded 18%. Starting around 1974—5 the
economic situation changed dramatically: the economy stagnated
until 1976, and there was an absolute decline in average real income.
Real GNP per capita dropped by 8% in four years and the level of real
income in 1980 was barely above that already reached back in 1971.
The average annual rates of inflation went from 24% in 1975 to 68%
in 1979. (See Table 8.1.) Politically, a new military government came
to power in mid-1975, headed by General Francisco Morales
Bermudez.
The origin of the crisis is commonly associated with excess demand
in the economy, especially as a result of the government's over-
spending. Table 8.1 shows that government deficit was around
3.1-3.9% of GNP before 1975 and jumped to 5.5-7.5% in the
period 1975-8. In association with this overspending and unfavor-
102
Economic crisis and the peasantryy 1975-1980 103
Table 8.1 Peru: indicators of economic crisis, 1970—81
Real GNP Av. annual Exchange Government Trade

per capita rate of rate US $ deficit/GNP bal./exp.
Year (1970=100) inflation (%) (1970=100) (%) (%)
1970 100 5.0 100 n.a. 32.3

1971 104 6.8 100 3.1 17.7
1972 107 7.2 100 3.7 14.1
1973 110 9.5 100 3.9 7.1
1974 114 16.9 100 3.1 -26.8
1975 114 23.6 104 5.5 -85.2
1976 114 33.5 160 6.3 -54.5
1977 110 38.0 303 7.5 -25.4
1978 104 57.8 453 5.1 17.5
1979 105 67.7 578 1.4 39.8
1980 106 59.7 791 3.4 19.7
1981 107 73.5 1,148 7.8 -19.6
Source: Banco Central de Reserva, Memoria (1974-9), and Instituto Nacional de

Estadistica, various publications. For 1981, Banco Central, Kesena Economical March
1982.
able conditions in international markets, Peru also faced an external

crisis: the balance of trade was secularly negative from 1974 until
1977.1
Economic policies to stabilize the economy were initiated in 1973,
with a selective policy of subsidies on certain products considered
basic (such as foodstuffs) and price-wage controls. These policies
continued throughout the period of crisis. The objective was always
to reduce aggregate demand, including demand for imports.
Government deficits were reduced significantly only by 1979—80 and
the balance of trade turned positive only in 1978. The rate of
exchange has been depreciated significantly since 1976. The period
1975—80 was dominated by economic policies leading to economic
stabilization.
The average income of workers in the modern sector declined
substantially. The average wage income of a typical blue-collar
worker in 1980 allows him to buy only 60% of his purchasing power
of 1974, as shown in Table 8.2. The fall in white-collar incomes is
even more dramatic. Workers at the lower wage levels suffered
relatively less. Table 8.2 shows the case of a hypothetical worker who
entered a firm earning minimum wages and received wage-increases
Table 8.2 Peru: indicators of income distribution in the modern sector
1970-80 (1970= 100)
Average real wage income Profits
White- Blue- Compensated Real Share in nat.

Year collar collar low-paid0 values income (%)
1970 100 100 100 20

1971 107 109 — 95 18
1972 115 118 — 95 18
1973 117 130 — 137 22
1974 112 128 109 157 24
1975 108 115 106 151 23
1976 93 118 102 163 24
1977 81 100 111 165 25
1978 70 90 100 180 28
1979 64 90 106 233 33
1980 68 95 111 234 32
a
Includes 'exceptional compensations due to cost of living increases' decreed by the
government and applied to a hypothetical worker who entered a private firm in
January 1974 earning minimum wages.
Source: Ministerio de Trabajo for data on wages; Banco Central de Reserva for data
on profits {Memorias) and Analisis Laboral for 'compensated low-paid'.
decreed by government 'to compensate for cost of living increases'.

This worker has almost recovered his purchasing power in the last
years. The policy to decree wage-increases in absolute amounts has had
the effect of changing the wage structure and has protected relatively
more the real income of the low-paid worker in the modern sector.
On the other hand, profits have increased continuously in real
terms and in their share of national income during the crisis. The gain
in real terms is shown in Table 8.2. One also notes that the share of
profits in national income went from 23% in 1975 to 33% in 1979. In
the middle of the crisis close to one-third of national income went to
profits.2
Changes in incomes of urban workers outside the modern sector
are much more difficult to estimate. This is a very heterogeneous
social group with a variety of sources of income. The wage income of
this group must have decreased. There are two reasons for this. First,
the minimum wage rate has fallen in real terms (shown in Table 8.3)
and this is the relevant wage rate for this social group, which mostly
Economic crisis and the peasantry, 1975-1980 105
Table 8.3 "Lima: minimum wage, unemployment and underemployment
rates (%
Real minimum Rate of Rate of

Year wage rate* unemployment underemployment2'
1974 100 n.a. n.a.

1975 95 7.5 13.4
1976 91 6.9 15.5
1977 29 8.4 16.1
1978 64 8.0 30.0
1979 83 6.5 29.6
1980 88 7.1 24.2
"Base: 1974=100.
b
Refers to people with incomes below minimum wages.
Source: Ministerio de Trabajo, Situation Ocupational, Cuarto Trimestre 1980 (Lima:
February 1981).
gets temporary jobs. Secondly, expansion of employment has been

very small during this period of crisis. All this is consistent with the
estimates on 'underemployment' in Lima. Table 8.3 shows that the
proportion of Lima's labor force with incomes below the
minimum wage increased from 13% in 1975 to 30% in 1978—9.
The rate of open unemployment has not changed significantly during
this period, however.
The other component, income from self-employment (particularly
that associated with trading and commerce), must have permitted
some degree of adjustment to prevent a further fall in total income of
this social group. Hence, we may conclude that the fall in real income
of the urban poor has been significant, but it was proportionately
lower than that of the workers of the modern sector.
Changes in incomes of peasant families

Peasant families living in the highlands of Peru represent close to one-
third of the country's population. This is the social group which lies
at the bottom of the income pyramid in Peru. It is usually argued that
peasants are outside the market system and that the economic crisis
has no effect on them. However, no evidence has ever been presented
in support of the 'outside-the-market' hypothesis. This study has
shown estimates of the degree of integration to the market of peasant
families in the southern sierra region of Peru. The result is that, on
average, the peasant communities exchange 50% of their income and
therefore dedicate the other 50% to self-consumption. With this
result, which applies to the most 'traditional' peasant families in the
most 'backward' region, the hypothesis of self-subsistence, outside
the market, must be rejected. Peasant families are well integrated into
the market system.
Two types of activities have been considered in the peasant
economy: agriculture (including livestock) and all non-agriculture
(manufacturing, construction, trading). A peasant family allocates its
labor to produce those commodities with the help of resources that
the family possesses. As a consequence of this self-employment the
peasant family produces quantities of X and X 2 during a year, in each
activity. Another use of the family's labor force is to enter into labor
markets and derive wages from temporary employment. There are
two types of labor markets: local (within the community) and
external (outside the community). Total wages earned in local
markets will be equal to the wage rate (i^1) and total number of days
worked (L 1 ). The same concepts will apply to external markets: w
for wage rates and L2 for number of days employed.
Therefore, totalpeasant income will be equal to the monetary value of
agricultural and non-agricultural production (valued at market
prices) and the monetary income from wages. Since part of the goods
produced is consumed by the family (Xl5 X 2 ), and part is sold in the
market (E , J52), the total peasant income can be written as:
In this equation, self-consumption income is equal to the value of

the first parenthesis. As shown in Chapter 4 this is around 50% of the
total income for the typical peasant family in the southern sierra
region in Peru. Monetary peasant income is allocated partly to buy
modern inputs, X 3 (fertilizers, pesticides), and partly to buy con-
sumption goods, X 9 and capital goods, Xfi. Hence:
(0.37 + 0.24 + 0.22 + 0.17 = 0.05 + 0.90 + 0.05)

The numbers shown below the equation are the shares of each
component in the income and expenditure side, as estimated in this
study. With the help of the previous equations we can make some
Economic crisis and the peasantryy 1975-19 SO 107
assessments on the possible effects of the present economic crisis on
the income of peasants.
The decline in GNP has not been a result of a uniform decline in
production in all sectors. Some sectors of the economy have fallen
more drastically than others. The determination of these sectors will
indicate where the employment effect was more severe. These sectors
were construction, manufacturing and agriculture, in that order. In
the cases of construction and agriculture, it is clear that the effect was
negative on peasants' income: wage is earned mostly in these sectors,
as shown in Chapter 5.
Changes in relative prices and wages included in the budget
equation of peasant families will affect their real income in a period of
inflation. To evaluate the effect of inflation on the peasants' income
will require an estimate of the relative change of the five prices and
two wage rates included in the budget equation. We do not have
statistical evidence on these changes. However, based on some
scattered pieces of information, an estimate of the possible effects will
be given here.
Prices received at farm gates for agricultural goods (px) are not
registered in any price index. The price index for food in Lima cannot
be taken as a measure for this, for two reasons: first, the variations
over time in the ratio of consumer's and producer's price; and
secondly, the fact that a significant proportion of food consumed in
Lima is not supplied by highland peasants but is imported — a large
proportion of food consumed in Peru is imported. Domestic prices
of food have been controlled throughout the period of crisis. On the
other hand, subsidies to imported food maintained the prices of the
import competing products depressed. The over-valuation of the
exchange rate had the same effect. As Table 8.1 shows, the rate of
inflation has been higher than the rate of devaluation, compared to
1970 parity and even to 1974 parity. In any event we can assume that
p1 increased at the same rate as the average rate of inflation, due to the
relaxation in price controls, subsidies and import policies that took
place in 1978-9.
With respect to p2, the basic effect is related to the trading activity.
The most important single non-agricultural source of money income
is trading. The cost of trading, the price of commodities imported
into the peasant communities and the cost of transportation all
increased with inflation and have, possibly, created a reduction in the
net income from trading.
The wage rates which are more relevant for peasants entering labor
markets are the legal minimum wage. This is true for local as well as
for external markets. Wage policies implemented in Peru revised
periodically the level of minimum wages by regions. The result was a
fourfold increase in money terms during 1975-9. Inflation in the
same period implied, however, a sixfold increase in price levels. In
addition, the cost of entering into external labor markets increased
substantially (cost of transportation and food). All this clearly led to a
relative reduction in the net money income from wages. In sum, on
the income side of the budget equation, prices and wage rates
increased by less than the general inflation rate (measured by the
consumer's price index of Lima).
We should now review the expenditure side. Here, the relevant
price is pe as 90% of the expenditure is allocated to the imports of
consumer goods. The basic products in the imported consumption
basket of the peasant family are: salt, sugar, cooking oil, rice, noodles,
flour, coca, alcohol, kerosene, soaps/detergents, clothing. A price
index of these commodities was constructed for the period 1973-9. It
shows an index of 950 in 1979 compared to 100 in 1973 (Ruiz, 1980).
This increase is much higher than the rise of the consumer's price
index in Lima. The divergence occurred mainly in 1978-9 when price
controls and subsidies were reduced substantially. The point is that
some commodities under control were part of the basket of
consumption of both urban and peasant families. When prices were
freed the relative increase was higher for peasant families, compared
to Lima's families, indicating that those commodities had a higher
weight in their consumption basket of goods bought in the market.
The evidence on permanent migration indicates that on average
every peasant family has one member (son or daughter) who lives in a
city. This sole fact suggests that private transfers of income must be
an important source of income for peasant families. It was shown in
Chapter 5 that this income is equivalent to almost 10% of their total
money income. To the question of whether transfers from their sons
or daughters were increasing in the last years, the answers we got
from the peasant families in the study are as follows: 11% said yes;
47% said that no significant change has occurred; and 42% said that
it has decreased. They were told that the cost of living in Lima was
increasing substantially and for this reason their relatives were not
able to send the same income as before.
The evidence we have presented indicates that the prices of
Economic crisis and the peasantryy 1975-1980 109
commodities and labor sold by peasants increased by a smaller
proportion than the inflation rate; on the other hand, the price of
consumption goods that peasant families buy from the market
increased more rapidly than the inflation rate. The fact that ft grew
faster than the inflation rate indicates that real income, measured in
terms of purchasing power of imported consumption goods (X ) of
peasant families, clearly decreased during the period of economic
crisis.
We have shown that relative prices moved against peasants during
the inflation period. We now need to estimate the absolute amount by
which real income fell. To experience a decline of 40% in real income,
as in the case of workers of the modern sector, the monetary income
of peasants should have declined by 80% in real terms, since half of
total income is in monetary income. For this to happen relative prices
must have fallen by 80%, on average, during this period. This could
not have happened. An estimate of the relative change in wx\p from
1974 to 1979 shows that this ratio fell by 33%. Taking this ratio as
the average proportion in the decline of relative prices, which is still
an overestimation since some prices (such as p^ increased more
rapidly than wages, would lead to an estimate of 16% in the reduction
of real income during 1970-5. Including the reduction in transfers,
this percentage might fall to 13-15%. This is a much smaller loss in
real income compared to workers of the modern sector and to the
urban poor.
The general conclusion that one may draw from these pieces of
evidence is that inequality and absolute poverty in Peru increased
during the crisis. The major change in inequality is between profits
and incomes from labor. On the other hand, the poverty line for 1980
shows that more people are under it compared to 1974; moreover,
the new families incorporated to this condition are basically urban. In
sum, the shape of the income pyramid has changed in two ways: a
stretch at the top and a squeeze at the bottom.
Economic policies in the early 1980s

The economic policy of the second Belaunde administration (in-
augurated in July 1980) contains new elements to overcome the
crisis: to liberalize the economy from price controls, tariffs and
subsidies. The idea is to attain a higher level of economic efficiency in
the economy. Another policy objective is to reduce inflation, mainly
through cuts in government spending. The effect of gains in
economic efficiency on income distribution is not easy to determine.
However, Arthur Okun (1975) advances the proposition that a trade-
off exists between efficiency and equality. The more equality is
pursued in a market economy the less efficient becomes the economic
system. Hence, the implicit choice of the present policy markets is
biased against equality.
The most important policy of income redistribution now refers to
food stamps. This program attempts to take income from the
government budget and transfer it to the poorest families ('the
extreme poverty') through food stamps. This policy will substitute
the subsidies on food which were eliminated. The success of this
program will depend on several things. First, this program must be
massive to be effective. The experience with food stamps in countries
like the United States cannot be taken as a model. In that country food
stamps are aimed at a very small proportion of the population,
because the poor constitute a pocket in a sea of richness. In Peru,
however, we have the opposite situation: a pocket of richness in a sea
of poverty. Therefore, any program designed to help the poor must
be a massive one. Secondly, as indicated before, with the economic
crisis changes in the income pyramid occurred. Absolute poverty
increased. Approximately 50% of the population in Peru live under
conditions of extreme poverty now. Hence, the redistribution
program must be enormous. Thirdly, the amount of income to be
transferred must be substantial. The food stamp program, however,
amounts to only 2% of the total government budget for 1981. If
we discount the losses that part of these funds will suffer, due
to bureaucratic mismanagement and possible corruption (that is,
through Okun's 'leaky bucket'), the net transfer will be very small.
In fact, the value of the coupons has been set at 3,000 soles per
month, which is equivalent to 10% of minimum wage in Lima.3
In terms of massive transfers, the policy of subsidies on food had a
better performance. Table 8.4 shows the incidence of the food
subsidies in 1980. This incidence is clearly regressive if the rural poor
are excluded. However, the view taken by the government is that the
program is progressive as a whole. The statistical base of these
estimates is very weak, as it relies on a 1972 survey on food
consumption. Particularly, the rural case seems underestimated.
Chapter 5 showed that, on average, 50% of monetary income of the
rural poor is spent on food, which implies 25% of total income, and
Economic crisis and the peasantry, 1975-1980 111
Table 8.4 Food subsidy incidence by families, 1980 (soles per month)
Population Average Average Incidence

Strata (%) income subsidy (%)
High 7.0 180,000 4,200 2.3

Medium 26.5 66,500 3,000 4.5
Urban - low 20.2 38,000 4,100 10.8
Rural - low 46.3 16,000 800 5.0
100.0
Source: Banco Central de Reserva. Data released in a symposium on 'Nutrition and

Food Subsidies', January 1981 (Webb, 1981).
most of those foodstuffs were subsidized, such as sugar, cooking oil,

rice, noodles and wheat-flour.
But even if those statistical estimates are taken as good approxi-
mations, Table 8.4 shows that poor families, particularly those living
in urban areas, got a higher transfer (4,100 soles) than the present food
stamp program, which will transfer 3,000 soles to fewer families. The
target is that at the beginning of the food stamp program 10% of total
families of Peru will be reached. Hence, by substituting the food
subsidy program for food stamps, the cost of redistribution is
reduced from 10% of the government budget to 2%. Finally, even if
the upper 50% of the income pyramid benefits more from the food
subsidy program than the bottom 50%, which is actually the official
argument to arrive at the conclusion that the system as a whole is
progressive, it does not follow that the subsidy program should be
eliminated. This progressiveness can be corrected with other policies,
such as a more progressive taxation, so as to get a net transfer to the
poor through the government budget.
The other argument against food subsidies is the negative effect on
agricultural income. The subsidy is given to imported food, most of
which competes with domestic production; thus subsidies introduce
a price discrimination against domestic farmers. However, the only
alternative to solve this problem is not to eliminate the subsidies. One
possibility was to extend the subsidy also to the production of
domestic farmers. In this manner the price discrimination that they
have suffered for years could have been eliminated, and incentives to
increase agricultural production could have been created. On the
other hand, it was shown in Chapter 7 that the effect of increasing
food prices at the farm gate on urban real income is relatively small:
doubling prices would reduce that income by only 15%.
Structural stagnation in peasants9 incomes

From the post-World War II period until the mid-1970s Peru
experienced a variety of political regimes (liberal, interventionist,
reformist); and different strategies of economic growth (export-led
and import substitution). During this period substantial economic
growth was achieved. Hence, Peru's economic performance has
relevance for the problem of growth and equity, under different
schemes, in less developed countries.
In terms of inequality Peru's performance is disappointing. The
overall income inequality increased over time. Even more unfor-
tunate is the fact that the poorest social group, the peasantry, did not
benefit from the process of growth. Peasant incomes were almost
stagnant during that period. Not only did distributive policies fail to
reach them, but even the programs of social reforms left them almost
untouched. Neither liberal nor reformist governments paid attention
to this social group. Economic growth under different strategies did
not have any significant effect on raising their standards of living.
Capitalist development and the market mechanism also failed.
In the last five years Peru experienced another phenomenon which
is also very typical of less developed countries: recession together
with inflation. Real income of peasant families has decreased during
the economic crisis. A 'best guess' of this decline would be 13—15%
for the period 1975-80. This is a much smaller loss in real income
compared to workers of the modern sector (firms with more than 10
workers), where that percentage ranged between 38% and 45%.
Given this decline and the increase in real profits; and given the
structure of income in that sector, the total decline in the modern
sector could be estimated around 4-5%. Since almost two-thirds of
national income are generated in the modern sector, the rest (the
'traditional sector') must have decreased by 17%. Only then would
the weighted average of both sectors give 8% as a result. Thus, it
seems that peasant families, medium-sized farmers and urban workers
outside the modern sector all experienced a decrease in income of the
order of 17%.
The lowest percentage in the reduction of real income in the
traditional sector seems therefore to apply to peasant families. The
Economic crisis and the peasantry, 1975- 1980 113
fact that, recently, many people have returned to the peasant
communities because of the economic difficulties they faced in the
cities is consistent with this conclusion. Or it may be that these
returns show that the conditions of urban workers in the traditional
sector are much more precarious in the city than in the countryside.
So, even a similar percentage decrease in real income causes a more
serious problem to the urban poor than to the rural poor. The rural
poor have, on one hand, a more diversified economy which reduces
the risk of any major change in their economic situation; on the other
hand, the family can return to the 'natural economy' if it becomes
difficult to buy commodities. There is certainly a much lower floor of
subsistence income in the countryside.4
The economic policies initiated in the second half of 1980 give
higher priority to efficiency, as opposed to equality, and to the
controlling of inflation, particularly in regard to the reduction of the
government deficit. However, in view of the increase in absolute
poverty during this economic crisis, redistributive policies have a
substantial role to play in the present economic policy of Peru. This
does not imply a disequilibrium in the government budget. It does
imply afiscalpolicy oriented toward redistribution, through a change
in the structure of spending accordingly, and/or through a tax reform
to finance redistributive programs. The conspicuous consumption of
the rich in Peru, even in the middle of the economic crisis, is a clear
indication that there is enough room for significant income
redistribution.
9
Conclusions: reality, theory and policy
This study has been concerned with the functioning of the peasant
economy in today's Peru. This particular reality has been neglected in
the economic studies on Peru, in spite of its great significance in the
social system. The peasant economy has remained a reality without a
theory. In this study the basic traits of the peasant economy have been
investigated and then utilized to lay down the principal analytical
coordinates of its prevailing economic rationality. The peasant
economy has also been placed in the context of capitalist development
in order to analyze its changing role in the economic system. In this
final chapter a summary of our empirical findings and theoretical
interpretations are presented. This will be done in terms of the
hypotheses advanced in Chapter 2. Because a theoretical basis has
been reached, economic policies are also discussed.
Poor but efficient

The economy of the peasant communities in the Peruvian sierra is
organized in family units, which are the units of production and
consumption. In the case of the southern sierra, the most economi-
cally depressed area, the average size of the peasant family varies
between four and five members; the average total labor force is four
people and the adult labor force is two. The peasant family depends
upon resources of cultivable land, and usually has no more than five
hectares in area in addition to access to natural pastures, communally
held. The cultivable land is composed of parcels, the average number
of which varies between ten and 80, depending upon the community.
Regarding livestock, a peasant family typically has seven sheep, two
heads of cattle and one horse. Finally, they generally own three
shovels, picks, and other minor tools.
The empirical data cited demonstrates the tiny size of the peasant
114
Conclusions: reality, theory and policy 115
production unit. The peasant family of the Peruvian sierra is a
minifundista family. Its resources are assigned to the production of
agricultural (A) and livestock (P) goods, to a series of non-
agricultural activities which we have called Z-activities. Family labor
is allocated partly to self-employment in the economic unit and partly
to wage labor through temporary employment in labor markets (L).
Besides being small, the peasant economic unit is very diversified in
its productive activities.
Not only is the quantity of resources at the command of a peasant
family small, their quality is very low. The level of their education is
certainly low: approximately 75% of the heads of peasant families
surveyed have not finished primary school. The cultivable lands are
of low quality; the greater part is without water and located on steep
slopes. The communal pastures are also of the poorest quality. Seen in
the perspective of the sierra economy as a whole, and that of Peru, the
peasant economy truly operates on marginal lands. The quality of the
livestock is also low if compared to other productive units in Peru.
Moreover, the equipment used to produce Z-goods still relies on
traditional technology.
In addition, in the technology utilized, there is an absence of
modern inputs. The use of fertilizers and pesticides is the most
widespread, but the proportion of families involved is still small;
furthermore, they are applied without any technical assistance. The
use of new varieties of seeds, of improved species of livestock, of
improved cultivated pastures is absent. On the whole, it is evident
that the peasant economy depends upon poor resources and tradi-
tional technology.
There is no evidence that the peasant economy is inefficient. With
the resources and technical knowledge they currently possess, it is
difficult to imagine ways of significantly increasing production or
their income. Regarding technical efficiency, two aspects of the
peasant economy are commonly cited as examples of the possibility of
increasing production, even with existing resources and technology.
The first concerns the fragmentation of cultivable land. Considering
the risk and uncertainty involved in sierra agriculture, however,
fragmentation is a very rational response. The peasant's economic
behavior — of aversion to risk — is not a personal characteristic but
rather a social characteristic. It is a rational response to the social,
economic and geographic conditions in which the peasantry operates.
It is possible that the fragmentation might be excessive in some cases,
116 Capitalist development and peasant economy in "Peru
in relation to the risk to be covered, but there is no clear evidence that
by joining the parcels one would be able to increase total output
significantly.
The second case of alleged inefficiency refers to the wastefulness of
some agricultural activities, which take the form of fiestas. One may
observe that the corn harvest, for example, is brought in by more
hands than is technologically necessary and consumes more goods
than is necessary for the maintenance of the labor force, but this does
not necessarily imply inefficiency. Inefficiency is a relevant concept
when the process of production and consumption are socially
separate. It is a characteristic of the peasant economy that these
processes are integrated; consumption and production are, in many
cases, intertwined activities. This is certainly a substantive difference
from the capitalist economy, in which these activities are dis-
associated. The 'circular flow' between families and enterprises
employed by economics texts to represent the functioning of the
capitalist economy is not, therefore, applicable to the peasant
economy.
Concerning economic efficiency - meaning that prices and costs
should be incorporated into the decision-making process of the
economic unit in order to obtain the maximum net income from every
allocation of resources - it is commonly thought that the peasant
family follows a 'traditional' economic behavior, in the sense that it
does not evaluate new market situations and does not respond to
price incentives. 'They always do the same', it is said. Besides the fact
that, with changes in prices, one cannot continue 'doing the same'
because the given bundle of commodities is no longer attainable or
there will be income in excess, which demands readjustments in the
quantities produced and consumed, there is empirical evidence which
shows variability in the annual crop mix, in the periods in which
products might be sold, and in the migration periods among peasant
families. The simple fact that the families are very different in their
sources of income is an indication that there is no 'traditional'
behavior in the communities. On the contrary, everything is subject
to continuous economic evaluation.
On the other hand, it is also true that there is a certain inelasticity in
the production of a product, should its price rise. But this inelasticity
is not the result of a 'traditional' behavior; it is, rather, a rational
response in a context of decisions made in the face of risk and
uncertainty. Should the price of potatoes rise in one period, all the
peasants are not going to plant more potatoes, neglecting a well-
established balance in their portfolio of crops. They obviously can
have no certainty that those new prices will prevail in the next period.
With clear indications of the direction of relative prices the peasant
economy responds in a 'viscous' way, as it must revise an entire
portfolio, carefully established and tested. Conceptually, viscosity has
been confused for traditional behavior.
Thus, to attribute the poverty of the peasant economy to
inefficiency in the productive process is due to a conceptional
confusion. The low physical productivity is explained principally by
the quantity and quality of the resources and technology available in
the peasant economy. It is clear that with different resources and
different technology, production would be greater, but this effect
does not correspond to the concept of efficiency. If to this fact one adds
the economic behavior characterized by aversion to risk, which is a
response to the geographic and social context in which the peasant
economy operates, much of what seems to be economic inefficiency
reduces almost to nothing. The peasant families of the sierra are, in
conclusion, poor but efficient. Thus they conform to the thesis of
Professor Schultz (1964).
Relative overpopulation
The hypothesis that there is a permanent excess of labor in the peasant
communities does not seem to be correct either. The empirical results
shown here indicate that the labor of the peasant family is dedicated to
diverse activities. If the marginal productivity of labor is zero, as the
surplus labour hypothesis implies, this means it is zero in all the
activities (A, P, Z and L). It is possible that the marginal productivity
of labor might be zero if the entire labor force were dedicated to a
single activity. This is precisely the conceptual error incurred in all the
studies which attempt to measure excess peasant labor with reference
to the requirements of labor only in agriculture and livestock
production. But if one takes into account #//activities, the marginal
productivity rises. The hypothesis of an order in the use of labor, first
in A — P and then in Z — L, implies that the marginal productivity
could be lower inA—P, as this productivity's variance is less than in
the case of Z — L, but not to the point of being zero. In addition, the
fact that the wages in local and external labor markets constitute
opportunity cost demonstrates that the marginal productivities in A,
P and Z cannot be zero.
Another proof that no excess of labor exists, at least in significant
amounts, is that the wage rates in the peasant communities tend to rise
during periods of peak agricultural activity. As shown in this study,
during harvest time it is common to pay wages in products, whose
market value is superior to the monetary wage ruling in the other
months. Moreover, the employers (proprietors of medium-sized
farms) resort to a series of mechanisms in their personal relations to
assure themselves of labor in 'times of scarcity'.
Finally, the evidence found that the temporary migration to labor
markets external to the community is seasonal again points to the
refutation of the overpopulation hypothesis. If it had been observed
that temporary migration took place at any time of year, this would
have indicated the existence of a structural surplus of labor.
Nonetheless, the scarcity of labor in periods of peak agricultural
activity and the temporary migration at other times offer clear
indications that there is no such structural surplus. That which seems
to be an excess of labor, at least to people unfamiliar with the rural
economy, is the fact that the peasants are not occupied in agricultural
activities during the entire year. Seasonal agricultural unemployment
is thus confused for structural unemployment.
It is true, on the other hand, that areas of minifundio show a greater
population density than areas of modern economic units. But this
obeys the different rationality of the peasant economy, compared to the
capitalist enterprise. The peasant economy is sustained by the control
of the land, and the size of the labor force tends, therefore, to coincide
with the greatest demand for labor in the annual agricultural cycle,
assigning the seasonally unemployed labor to Z-activities and
temporary migrations. In a capitalist context, the same quality of land
and the same cyclical requirements for labor would involve employ-
ing a permanent labor force at the minimum of the cyclical require-
ments, covering the greater seasonal requirements by contracting
temporary labor. This is the way the enterprise minimizes costs.1
If the productivity of labor does not qualify the peasant com-
munities as an overpopulated economy, the level of their productivity
is in most cases below subsistence income. The most plausible
hypothesis is that absolute overpopulation does not exist, but relative
overpopulation does. The fact that wage income is an important
source in the total peasant income implies that the peasant family in
general cannot derive even a subsistence income by the exploitation
of their own resources. Their plot would not give them enough to
subsist upon, even if they dedicated all their time to self-employment.
Therefore, it is entry into the labor market which allows them,
through a division of family labor, to ensure a subsistence income.
This is the role played by labor markets in the functioning of the
peasant economy. The economic viability of the peasant economy is,
thus, assured by this mechanism.
Non-capitalist relations
The idea that most peasant families are integrated into the hacienda
system by means of pre-capitalist relations has been very widespread
in Peru. This concept has given rise to the idea that the poverty of the
peasantry is a consequence of the servile mechanisms of the haciendas.
It is indubitably true that these servile forms still exist but their
significance is minor for a comprehension of the peasant economy in
its totality.
One evidence of the reduced link between the peasant economy
and the sierra hacienda in the productive process is seen in the extent
of the agrarian reform program. This program affected almost all the
medium-sized and large haciendas of the sierra, the beneficiaries
being close to 120,000 families which live in peasant communities.
But this figure greatly exaggerates the number of beneficiaries as it
includes the SAIS system, a type of cooperative, in which various
communities surrounding the hacienda were incorporated as mem-
bers of the new enterprise. A more appropriate figure for measuring
the nexus between peasant families and haciendas in non-capitalist
relations would be approximately 80,000, which is roughly 10% of
the population of peasant families in the sierra. The essential aspect of
today's peasant economy is its relations of exchange through the
market. This is true even for the period before the agrarian reform.
Integration into the market system

Another widespread opinion on the peasant economy is that it is
isolated from the rest of the national economy. 'Self-sufficient
economies' and 'economies outside the market' are two of the phrases
that sum up this thesis. This also gives rise to the idea that Peru is a
'dual economy', that is to say, separated, its parts unconnected. One
of those parts is certainly the peasant economy of the sierra. This is the
common view of the Peruvian economy and SQciety, as of the other
Andean countries. The statistical results of the present study,
however, give a very different picture.
The peasant family produces A, P and Z-goods with its own
resources and labor. Another source of income is that obtained
through employment as wage-earners in the local and external labor
markets. Part of its production of A, P and Z is sold in the market and
part is consumed by the family itself. The monetary value of
everything produced as goods A, P and Z plus the monetary income
from wage employment constitutes the total peasant income. If one
divides this income between the part that the family consumes and the
part that it exchanges with the market (as much of goods A, P and Z
as of labor) the resulting proportion for a typical family of the
southern sierra is 50:50. In as much as half of the peasant income in
the most traditional region of Peru is the result of exchange with the
market, there is no basis to the thesis of 'self-sufficiency' or 'duality'.
One speaks of countries as 'open economies' when they export
15-20% of their Gross National Product. Nonetheless, the peasant
economy has been seen as 'closed economy' despite the fact that it
exports around 50% of its total product. It seems that cultural and
social duality, which clearly exists in Peru, has been confused with
economic duality.
The structure of the monetary income of the typical peasant family of
the southern sierra is 37% from sales of A and P, 24% from Z (where
an important proportion is income from commercial activities), 22%
from wage employment in local markets and 17% from external
markets, reached by seasonal migrations. Thus approximately 40%
of the peasants' monetary income is from wage employment. The
peasant family today is also proletarian. These sources of income
demonstrate, then, the principal characteristics of the peasant
economy in its present historical form in Peru.
The real income of the peasant family therefore depends in part on
the price system of the market. Changes in the prices of agricultural
goods that they produce, or in the prices of the products that they
buy, and changes in the wage rates in rural and urban labor markets,
cause significant changes in the peasant income. As market prices are
modified as a consequence of, among other things, changes in macro-
economic policies, price controls and international prices, it is clear
that the peasant economy is integrated into the national and
international economy. The integration or articulation is achieved
through the products markets and the labor markets. The poverty of
the peasants must then be explained within the income distribution
mechanism that operates in the Peruvian economy.
Dynamics of the peasant economy

In the peasant economy land and other resources have remained
almost fixed while population has increased. This implies a dimi-
nution in the minifundio size per family. If one also considers the fact
that the resources have been deteriorating in quality (as seen in the
erosion of land, the overgrazing of pasture lands, the invasion of
weeds in a large part of those lands), the aforementioned conclusion is
reinforced. The recent introduction of certain modern inputs, such as
fertilizers and pesticides, has at the most somewhat counteracted
these effects. Productivity increases have clearly been practically nil.
The conclusion from above is, hence, inescapable: over time each
peasant family produces less output than before in its own unit of
production. It also follows that over time, given the relative market
prices, the capacity of the minifundio to reproduce the peasant family is
more limited. The peasant family has therefore only two alternatives:
to emigrate, or — if staying in the rural area — to find additional
income as wage labor in temporary jobs. In this way, by a growing
process of proletarianization the peasants' supply of labor to
temporary labor markets is increased.
The growing process of proletarianization of the peasantry is not
the consequence of its population growth only. The pattern of
capitalist development also plays a significant role here. Two
arguments can be presented to support this view. First, the degree of
the peasant economy's integration into the market economy has
increased over time. The proportion of monetary income in the
peasants' total income has risen in the past decades; the peasant
economy interacts with the market much more now than in the
1940s, for example.
To the degree that market exchange expands in the national
economy, the structure of the peasant economy will tend to be
modified as well. A higher level of exchange was not the consequence
of higher levels of productive capacity in the peasant economy. It was
the result of increased opportunities to exchange through markets,
mainly due to improvements in transportation and communication
systems. It was the result of a 'commercial revolution' and not of an
'economic revolution'. Hence more market exchange for the peasant
family implied either a reduction in the level of own-consumption or
a reallocation of labor. It seems that the second effect has been
dominant. In fact, several anthropological and sociological studies
have shown a significant change in the rural life in the sierra since the
1940s, particularly in terms of growing penetration of manufactured
goods (Quijano, 1968). Thus a clear consequence of the expansion of
capitalism is the destruction of rural industry, the Z-activities.2
Therefore, to pay for the increasing imports peasants have reallocated
their family labor from Z-goods to wage employment.
Secondly, peasant agricultural and livestock production (potatoes,
corn, barley, wheat, milk, meat, wool) competes in the urban markets
with capitalist production. As shown in this study, the development
of productive forces is very rapid in the agriculture of developed
countries and food is increasingly imported into Peru. Also capitalist
domestic agriculture has shown some improvements in productivity.
But the peasant economy is almost stagnant. As a consequence there
has been a decline in the proportion of agricultural output produced
in the peasant economy, from 49% in 1948-52 to 31% in 1967-71
(Hopkins, 1981: p. 69). The peasant economy has lost significance as
a food supplier to urban markets. Capital accumulation and tech-
nological change in Peru have been concentrated in the capitalist
sector, leaving aside the peasant economy.
The proletarianization process of the peasantry also indicates a
change in its role in the economic system. Previously, its role was to
contribute to the supply of cheap foodstuffs for the cities, so as to
keep urban wages down. Now, it is becoming more a source of cheap
labor directly. If it were not for the minifundio, labor could not be hired
at minimum daily wage for only a certain period and then laid off.
Because the worker has an economic unit to return to once his labor
contract is terminated, the labor market for casual employment can
exist. Moreover, this is a very cheap system for the capitalist firms
because they pay as wages only part of the annual cost needed to
reproduce that labor force. The other part comes from self-
employment in the minifundio. Hence the peasantry helps to keep
wages down directly. In this sense, the peasant economy performs the
role of the 'reserve army' in the present historical stage of Peru's
capitalist development.
In sum, the processes of population growth and capitalist
development generate an increasing degree of proletarianization of
the peasantry, an increasing need of wage income and wage
employment. But, where will the demand for labor come from ? The
capitalist development has a limited capacity to generate the needed
wage income. First, wage employment in Peru has remained stagnant
Conclusions: reality y theory and policy \2?>
in relative terms; the 'modern sector' employed 21% of the labor

force in 1961 and 1970 (Webb, 1977: Table 6.1). Secondly, peasants
compete for wage income in rural and urban labor markets and these
markets are becoming saturated due to population growth and
migration. Thirdly, the historical experience shown in this study
indicates that although the capitalist sector grew at rapid rates the
peasant economy stagnated. The capitalist road to development has
not induced overall growth, as many people expected; in particular it
has not induced growth for the rural poor.
Some implications for economic policies

Given the limited capacity of the capitalist development to induce
economic growth for the peasantry, why not develop the peasant
economy directly} By increasing productivity and income from self-
employment in the minifundio less will be needed of wage employ-
ment. Economic policy could be directed toward reversing the
current economic process: make the peasant family more farmer and
less proletarian. This peasant road to overcome rural poverty will also
have a positive effect on wage rates: labor supply will decrease and
real wages can rise in urban and rural labor markets.
The results of the present study have many implications for the
design of economic policies for improving the economic condition of
the peasant economy. In what follows, brief mention will be made of
the most notable implications. First, and the most obvious, is to
increase the amount of land to peasant families. However, this would
require an extremely careful policy designed to get land from units
with 'excess' land and to transfer that land to families with the greatest
need. The issue is difficult because this policy implies a step beyond
the agrarian reform recently completed, but the effects can be
significant for the peasantry (Figueroa, 1982).
Secondly, the sources of the total peasant income are very diverse.
Economic policies (designed to improve their income) which are
based on one product, such as a corn program, or based on one
activity, such as a program fostering artisanry, would have only a
small effect in proportion to its importance in the structure of income.
Economic policies must cover a broader range than one price or
product if they are to have real effect upon the peasants' income.
From this perspective, there are two policies which would have a
significant effect. The first concerns the pricing of agricultural and
livestock goods. If one were to raise these prices, relative to the
entirety of goods, there would be two effects on the peasants' income:
one, direct, through their sale of those goods, and the other indirect,
by means of their wage income. The wage income obtained by the
peasants comes principally from two activities: agriculture and
construction, as much in local as external labor markets. Greater
profitability would then foster agricultural growth and greater
employment which would increase the salary income of the peasant
family. And the supply of rural labor lies in the minifundios.
The other policy concerns credit. The peasant families need credit
for their diverse activities: purchase of agricultural inputs, of tools, of
livestock, and of raw materials for artisanry, working-capital for
commerce and working-capital for migration to external labor
markets. The specific necessities of each family are different, and
therefore credit supplied to one sector alone, for example, agricul-
tural credit, would not give the peasant family flexibility in its use.
One would have to think in terms of a policy of peasant credit in
response to the needs of the peasant economy.
Thirdly, as the technology is still traditional there is room for
improvement of the physical productivity in peasant communities.
This implies modernization through the introduction of new and
modern inputs in the production of A, P and Z-goods.
Modernization, however, does not simply, or necessarily, mean the
introduction of fertilizers and pesticides. Modernization has more
meaning when there is a blend of modern inputs and improved
traditional inputs. Improved seeds, improved livestock and im-
proved pastures are some examples of what could be introduced in
the peasant communities. The technological change with the greatest
impact (and greatest diffusion) in the Andes would have to be
directed principally toward the genetic improvement of Andean
crops and products, that is to say those products which are already
adapted to the peasant economy, ecologically and socially, through a
historical process.
Technical assistance is also lacking in the peasant communities.
Every community studied has its irrigation or mineral exploitation
project which needs only technical direction to bring it to fruition.
Disease control for plants and animals is practically non-existent
because there are no technical assistance or extension programs. On
the other hand, neither technical development nor programs of
technical diffusion are designed in terms of the peasant economy. For
Conclusions: realityy theory and policy 125
example, research on new varieties concentrate mainly in increasing
its average yield, but not to reduce its variance (risk).
Modernization is an obvious way to raise the peasants' income.
The results of the present study in this sense corroborate Professor
Schultz's thesis. But they also point out two important differences: on
the one hand, modernization implies also a qualitative improvement
in traditional inputs or in their supply (the supply of water, for
example); on the other hand, given that the peasants' income depends
as much on the productivity of self-employed labor as on the price
system, modernization cannot be judged solely as a technical
problem. Gains in productivity could simply be transferred to the
cities by the price mechanism if they are, at the same time,
accompanied by a policy of anti-rural prices.
Appendix I: The sample
The empirical study of peasant communities was undertaken in the southern sierra
region of Peru. Eight communities were studied. These communities were picked
from a sample universe which constituted the most important community areas in the
southern sierra. Preliminary studies were conducted in seven of these communities
before September 1978, at which time was begun the fieldwork for the present study.
The communities studied are located in four of five departments comprising the
southern sierra; Ayacucho is the department not represented. This does not,
however, present any particular complication to the problems of sample repre-
sentativity in as much as the differences between departments are of minor
importance.
The empirical data were obtained via a peasant family survey, administered as a
uniform questionnaire in all of the peasant communities. Interviews were also
conducted with persons involved in the life of the community (mainly authorities and
ex-authorities) from which were obtained a physical and social perspective which was
of great importance in sharpening the contents of the interview to be administered to
the families. This procedure also permitted the control of the quality of the
interviews by the range of possibilities for response which was given by notable
persons. In each community the survey itself (apart from previous visits) lasted
approximately two weeks and was conducted, on average, by four persons, three of
whom (including the principal researcher) constituted a permanent team, the fourth
interviewer usually being a person from the community. The interviews took place in
Quechua or Aymara when the family surveyed did not have sufficient command of
Spanish.
After concluding the fieldwork in each community, reports were written
describing all that was observed during the stay in the community. These reports
constitute valuable material for the analysis of the quantitative data.
The total number of families surveyed was 306, distributed among eight peasant
communities as detailed in Table 2.3, p. 12. Furthermore, two more qualitative
studies treating specific themes were undertaken in two peasant communities: rural
artisanry in San Pedro, in the department of Cuzco; and migrations in Accomayo, in
the department of Ayacucho.
In each peasant community the sample was obtained after estimating the total
population. The predetermined sample size was around 40 families. This size permits
statistical analysis for each community and, at the same time, permits information
collection in a more detailed and profound manner from each family. The sampling,
including the selection of reemplacements, was random. The first three rows of
126
The sample 127
Table A.I Expansion of the sample (number of families)
Ecolo-
Ecological level Ecological level gical
I II level III
Communities JAC CUL NIN ANC TTI SIH HUA ACO Total
Sample size (//) 36 40 31 40 36 41 42 40 306

Population (IV) 180 290 42 180 92 322 304 100 1,510
5.0 7.25 1.35 4.5 2.55 7.85 7.24 2.5 4.93
Population (N) 512 898 100 1,510

Southern
sierra popu-
lation (AQ 110,000 320,000 70,000 500,000
215 356 700 331
(xR = M/n 1,075 1,559 290 1,602 908 2,795 2,577 1,750 1,632
Southern
sierra popu-
lation (M) 39 62 9 64 33 115 108 70 500
Table A.I show the sample size in community («), the total population of families
(JV), and the respective coefficient of expansion (a) for the eight communities studied.
The expansion of the sample to the southern sierra region required various steps.
In the first place, the total population of the region was estimated based on the
population and agricultural censuses, both from 1972. There are around one million
mintfundista peasants (with fewer than five hectares of land) in Peru; 82% live in the
sierra (Figueroa, 1976).
On the other hand, from the same population census it can be inferred that a little
less than 50% of the rural sierra population is located in the southern sierra region.
Nevertheless, without more precise information it was assumed that the proportion
of sierra peasant families which is located in the southern sierra is much greater than
50%, due to the predominately minifundista character of the region. An estimate of
60% appears more appropriate. This implies an estimate of 500,000 peasant families
living in the southern sierra of Peru.
Secondly, the expansion requires the taking into account of ecological 'levels'
{pisos). The productive structure (and the income structure) of the peasant economy
depends on the community's access to different kinds of land, on control of different
ecological levels. A community which has no access to the quechua zone, for example,
would not produce corn, as was indicated in Chapter 2. In that the intent of this study
is the estimation of peasant family income structure, the communities' access to
different ecological levels becomes an important variable in the expansion of the
sample. The communities Jacantaya, Culta and Ninamarca are typical of the sum
128 Appendix I
zone; Acobamba is typical of thtjunga zone and the rest of the communities are
typical of the quechua zone.
The communities under study control various ecological levels and, as such, the
families surveyed represent distinct proportions of the population. The Peruvian
rural population distribution by ecological levels is presented in Chapter 2. Based on
this information, 500 families were distributed in three ecological levels: sum, quechua
zndyunga. Rows 4 and 5 of Table A. 1 present the sample and population distribution
and by ecological levels, while row 6 depicts the respective expansion coefficients.
Therefore, the sample expansion to the southern sierra population is obtained by
multiplying a and B. The number of families (Af) that each sample family represents
appears in the final row of Table A.I.
Appendix II: Methodological notes concerning
the calculation of peasant income
1. Agricultural production measures the quantity harvested (in dry and in grain).
Moreover it refers to the 'overall' harvest and not the minor harvest, included in
green. There is certainly an underestimation of production.
2. The monetary income from wages derived from migration involves expendi-
tures. It is thus necessary to distinguish between gross and net wages. Here the data
are net. This produces an underestimation of the degree of monetization of the
peasant economy. In particular some expenditure categories were underestimated,
such as travel, lodging, food and beverages. The 'number of workdays' refers to days
of absence from the community. During migration, the peasant faces days without
work as well as days of intense work, be it in two jobs (night-watchman and day-
ditchdigger) or doing overtime. Therefore, it is not permissible to obtain wage rates
dividing total net income by number of days in migration.
3. The monetary and non-monetary incomes derived via the local labor market
include only wages. Goods such as food, coca, cigarettes and aguardiente which the
wage-earner receives from his/her employer are not included.
4. A, P and Z monetary incomes refer to 'value of sales'. Neither monetary costs
incurred via input expenditures (fertilizers, pesticides and others) nor marketing
costs (such as transport) were deducted. In the case of transport, it was considered as
a consumption expenditure (to make purchases or to go on vacation) because there
always exists the possibility of selling the goods in the local market.
5. Monetary income from commerce is net. If a family markets its own product we
consider as Ay P or Z income, according to the case, the opportunity cost of its
production and only the surplus as income from commerce.
6. The only income which is not net refers to A, P and Z-goods. As such, the
estimate presented in the text is called 'gross' monetary income.
7. Trade via barter is evaluated at the average price in force in the local market or
some relevant zonal market. Barter for urban goods (sugar, noodles, oil) is
considered as a monetary transaction. These cases are rare and, on the other hand, the
prices implicit in the trade do not vary much from the market price.
8. To evaluate own-consumption, we used the average prices of monetary
transactions and barter.
9. The estimation of the matrix which appears in Table 4.1 is based on the survey
data, with the exception of the following: eggs, milk and wool were estimated based
on family animal stocks and typical yields of the community; the assignment of
agricultural production to direct consumption and to feed was estimated under the
129
130 Appendix II
assumption of a relation of 4:1 for all the communities. The latter is based on
information collected at various stages of the fieldwork.
10. The manner in which monetary expenditures and incomes are estimated
produces an overestimation of expenditures in relation to income. In that the period
of study was one year (over which time were distributed expenditures and income),
income was measured at different times during the year, while expenditures (in their
most important rubrics, such as food), were measured for the week or month
preceding the survey. Based on these reference periods (week or month), we
extrapolated to the entire year. This method has the defect of exaggerating annual
expenditure when there is significant inflation. Expenditures are evaluated at the
prices of the most recent week (or month) but incomes are evaluated at prices of
different periods. The fact that in the majority of cases the average monetary
expenditure in the communities was greater than the respective income appears to be
consistent with the above-mentioned bias.
11. In the face of discrepancies between the total value of exports and imports, it
was assumed that export values were more trustworthy. The level of imports (which
was always greater than that of exports) was then reduced to the level of exports,
without modifying its structure.
Appendix III: Andean food plants
1. Potatoes: Solatium spp., the majority of native potatoes grown in Peru; Solatium
tuberosum subspecies indigena.
2. Oca: Oxalis tuberosa.
3. Olluco: Ullucus tuberosus.
4. Mashua: Tropaeolum tuberosum.
5. Achira: Canna edulis.
6. Quinua: Chenopodium quinoa.
7. Canihua: Chenopodium pallidicaule.
8. Achita: achis = quihuicha. Amaranthus caudatus.
9. Maize: Zea mays.
10. Tarhui: Lupinus mutabilis.
131
Appendix IV: Methodology used in the calculation
of Table 7.1
1. Growth
The series of real GNP per capita estimates for the period 1950-74 was obtained
from the Cuentas Nacionales published by the Banco Central de Reserva del Peru in
various volumes (1950-65,1960-7,1960-9,1960-74). For the period 1975-80, we
used the series published by the Banco Central in its annual Memorias.
2. Inflation
The index is the Lima CPI. Published data obtained from the Instituto Nacional de
Estadistica were used for the entire period. (The Institute's name changed at various
points during the period.)
3. Growth pattern
The series of manufactures and exports as percentages of GNP were obtained from
the same sources that were used for the GNP series. In the case of exports there was
no significant discrepancy between the Cuentas Nacionales and the Memorias during
the period of superposition in the calculations (1972-4). Thus both series were used
as if there were only one. In the case of manufactures there were discrepancies
between the sources during 1972-4. The figures which appear in the Cuentas
Nacionales are, on average, 83% of the figures in the Memorias. In order to have a
long-term series which did not display important methodological changes in the
calculations, it was considered appropriate to correct the figures in the Memorias by
that coefficient (83%) for 1975-80.
4. Income distribution
National income and functional distribution

The series of national income and firms' profits were obtained from the Cuentas
Nacionales and the Memorias of the Banco Central. Both estimates coincide for the
period 1970-4. Thus it was assumed that for the period 1975-80 we could expand
the series with the data of the Memorias. The profits data refer to total profits, without
deducting taxes nor revaluation adjustments for inventories.
The profits figure for 1980 indicates insignificant growth in real terms. If it is
132
Methodology used in the calculation of Table 7.1 133
related to the data on profit taxes ('tax on juridical persons' published by the Banco
Central) we find that nearly 19% of profits went to tax payments. Nevertheless, this
percentage never exceeded 11% during 1974-9. This discrepancy suggests that the
profits data for 1980 were underestimated. No tax reform of significance took place
which could have caused the percentage to change so drastically. Even so, the profits
figure for 1980 was maintained; this gave greater solidity to the conclusion that the
change in profits during the crisis was substantial.
Incomes of workers in the (modern sector'

This series refers to incomes from wages for workers and salaries for employees in
firms with more than 10 workers. Two sources were used for the period 1950-66:
the study of Richard Webb (1977), which indicates an annual growth rate of 4.1% in
real terms for this period; and the series published by the Ministerio de Trabajo,
Direccion General de Empleo, Encuestas de Establecimientos, which refers to
Metropolitan Lima. These latter publications present a series starting in 1957, for
which it was possible to obtain estimates for real wages and salaries for 1957-66; this
implied as a result a smaller growth rate, nearly 3.2%. For the period 1967-80, we
used the series from the Ministerio de Trabajo.
Incomes in the 'urban traditional sector'

For the period 1950-66, we used the estimate made by Webb (1977) which showed
growth in average real income totalling 2.1 % per year. For 1967-70 we assumed that
there occurred a similar growth in national per capita income, for lack of additional
information. Starting in 1970 we used the evolution of the legal minimum wage for
Metropolitan Lima as an indicator of the average income in urban employment
which was not in firms employing more than 10 workers. This series was obtained
from the review Analisis Laboral, January 1981, p. 8. To convert this series to real
terms we used the Metropolitan Lima CPI.
Incomes in the 'rural traditional sector'

Webb's estimates (1977) for the period 1950-66 were also used in this income
rubric; he estimated an average annual growth rate in real income of 1.3%. For the
period 1971-4, we used the growth rate of 1% per year, based on the author's
evaluation of the impact of agricultural price policies on rural incomes (Figueroa,
1981). Finally, the arguments presented in the text (Chapter 8) imply an estimate of
— 2.7% as the change per year in real income during 1975-80.
Thus, the estimates of workers' incomes are based on various assumptions and
derived from several sources of varying quality and trustworthiness. However, they
were performed independently for the three sectors. A consistency analysis was
undertaken in order to verify whether the real income series per sector, weighted by
worker population per sector and then summed, produce values approaching total
labor income estimated via the functional distribution. The latter analysis, performed
for various years, displays errors of varying magnitudes (between —2% and 15%);
however, the estimates appear acceptable as first approximations.
Notes
3 The economic unit and economic organization

1 The lay mi land always starts production with potatoes, followed by cereals. After
3—4 years of cultivation the land is again left fallow for 5—10 years, depending on
the quality of the land and its scarcity in the community.
2 There are communal cultivable lands in some communities which are distributed
annually among peasant families, but these communities are very few.
4 Production and exchange

1 An exception is Jacantaya, where inputs are imported for boat construction and
repairs. This community, located on the shore of Lake Titicaca, uses boats for
transportation and for fishing.
5 The level and structure of peasant income

1 Because imported inputs are allocated basically to agriculture, the total value of
these imports was deducted from the gross sales of agricultural products in order
to arrive at net peasant income.
2 Data on rents are very weak. There are too many ways of paying rents and too
many things rented to get good estimates. However, incomes from rent are not an
important source due, in part, to the fact that most rent is paid with other
productive services.
3 The difference between the income structure of the community and that of the
family is due, technically speaking, to the fact that 'the ratio of means' is not the
same as 'means of ratios'. If, for example, wage income is more important for poor
families than for rich ones, this type of income will have a lower share in the
income structure of the community compared to the family. The first case is
weighted by income, the second by number of families. A comparison of the two
measures therefore indicates how the different types of income are distributed
among income strata. A family budget study in Lima showed little difference
between the two structures - 54% and 57% for the case of wage income, for
example (Figueroa, 1974). Thus, the assumption of considering both structures
very similar in the case of peasant incomes, where levels and sources of income are
less heterogeneous than in the case of Lima, seems reasonable.
134
Notes to pages 60-97 135
6 The economic behavior of the peasant family
1 The mix of A, P, L and Z-activities and the variety of products implied in each
activity show a highly diversified portfolio. Yet diversification is more than that.
Even in one single product diversification is present. For example, potatoes are
cultivated using a mix of varieties with different yields and resistance power to
frost.
2 This behavior is consistent with the use of credit. Peasants who obtain credit from
the agrarian development bank for the purchase of agricultural inputs usually
allocate part of the credit to other activities, although it is illegal to do so.
3 No estimates were made about the number of days worked in the local labor
market. A best guess would be to assume at least an equal number of temporary
migration for all communities. Hence, above 25% of the labor time of the
household head is probably allocated to work outside the parcel.
4 On the evolution of the labor markets in rural Peru see Cotlear (1979).
5 Another consequence of the empirical result of Table 6.4 is that the rural-urban
link takes on a new form: income from transfers. We showed in the previous
chapter the importance which transfers have for the income of peasant families.
6 On the peasant family strategy regarding emigration and returns see Vega (1979).
7 Stagnation in the peasant economy

and the role of demand
1 Several studies support these conclusions. See, for example, Hunt (1975), Roemer
(1970), Thorp and Bertram (1978).
2 If ownership of an automobile is taken as the criterion for belonging to the 'middle
class', the result is exactly 5% of all families, as much in 1961 as in 1980.
3 This result was obtained from the fact that the share of profits in national income
rose slightly during the period (see Table 7.1) and from the assumption, which
appears reasonable, that the population of 'capitalist families' grew less than the
total Peruvian population.
4 For an empirical support of this statement see Weisskoff and Figueroa (1977).
5 Estimates of income elasticities for food and other consumer goods for 10 Latin
American cities can be seen in Musgrove (1978).
6 Given the tremendous dispersion of producing units and the significant real costs
involved in moving commodities due to poor conditions in the systems of
transport and also given imperfect market structures, these coefficients are
underestimates of true values. However, this will give a bias in favor of a smaller
difference between curves R and E in Figure 7.1, which we are trying to challenge.
7 An implication of this result is that the city-countryside conflict is not as acute as it
is usually said to be. The prices for the food sold by rural people could be doubled
and their income doubled, yet urban real income would decline by only 15%.
There is much room for income redistribution through price policies for food. A
further development of these issues can be seen in Figueroa (1981).
8 The substitution of agricultural food by processed food seems also related to
another much deeper trend in the Latin American economies: the irrevocable
expansion of the market system. This expansion requires an increasing amount of
136 Notes to pages 101-22
processed goods. A more recent, and related, phenomenon is the presence of
multinational corporations in the food processing industry.
9 A study made by Alvarez (1980) shows, in effect, that stagnation in peasant output
was even more dramatic during 1967-77.
8 Economic crisis and the peasant

economy, 1975-1980
1 Given that Peru is a net importer of capital, its balance of services always shows
deficit. Hence, the surplus in the balance of trade is crucial to maintain long-run
external equilibrium in the economy.
2 Clearly the burden of the crisis did not fall on capitalists; on the contrary, it seems
that the crisis was very profitable for them. For the case of commercial banks, for
instance, the average rate of profit jumped from 15.9% in 1975 to 22.5% in 1979
{Latin American Regional Report. The Andean Group, 23 January 1981, p. 5).
3 As anticipated by these arguments, the results obtained with this policy were not
satisfactory and the government cut the program by the end of 1981.
4 At the present time the consumption of urban goods is decreasing in favor of a
return to goods produced in the community, as in the case of substitution of animal
fat for cooking oil, of treebark (of the taxana, for example) for soaps and
detergents, roof-tiles in place of corrugated iron and transport by foot rather than
in trucks. In reference to these substitutions, one effect of the market exchange has
rarely been noted: in this process the traditional technology is lost. In the present
period of crisis the peasants desire to return to their traditional technology, but in
many cases it has already been lost.
9 Conclusions: reality, theory and policy

1 In this discussion one should include cooperative enterprises as well as capitalist
enterprises to be more precise with regard to the organizational form of modern
enterprises in Peru in the Post-Agrarian Reform period. Nonetheless, the
argument is equally valid in the case of the cooperative enterprise, as its economic
behavior with respect to permanent and seasonal employment is not very different
from that of the capitalist enterprise. For a theoretical analysis, see Caballero
(1978).
2 Foods processed in the communities have given way to urban products, as in the
case of chicha replaced by beer. In textiles and manufacture, the peasants' clothing
now shows a greater urban influence - synthetic fabrics in place of their own
production of wool (woolen manufactures). The difference between generations is
remarkable among the peasants; only the older people wear clothing made mainly
of wool. Products made of plastic have invaded the countryside, displacing wool,
as in the case of sacks and ropes, and native ceramics — pots, containers and
cooking utensils.
Bibliography
Alvarez, Elena (1980). Politico Agrariaj Estancamiento de la Agricultura, 1969-1977.

Lima: Instituto de Estudios Peruanos.
Baer, Werner and Figueroa, Adolfo (1981). 'Equity and State Enterprises: Some
Reflections Based on the Cases of Brazil and Peru', in T. Bruneau and P. Faucher
(eds), Authoritarian Capitalism: Brazil's Contemporary Economic and Political
Development. Boulder: Westview Press.
Caballero, Jose M. (1978). 'Los Eventuales en las Cooperativas Costenas Peruanas:
Un Modelo Analitico'. Economia (Lima), Vol. I.
Caballero, Jose M. and Alvarez, Elena (1980). Aspectos Cuantitativos de la Reforma
Agraria, 1969-1979. Lima: Instituto de Estudios Peruanos.
Cotlear, Daniel (1979). El Sistema de Enganche a Principios del Siglo XX: Una Version
Diferente. Bachelor's thesis in Economics. Lima: Universidad Catolica.
de Janvry, Alain (1974). 'The Political Economy of Rural Development in Latin
America: an Interpretation'. American Journal of Agricultural Economics, Vol. 57.
Donges, Juergen (1979). 'The Economic Model Order at the Crossroad: Past
Development Trends, the Commodities Problem, and North-South Trade Rules'.
Published in Spanish version in J. Donges et al., America Latinaj la Economia
Mundial. Buenos Aires: Instituto Torcuato Di Telia.
Figueroa, Adolfo (1974). Estructura del Consumo j Distribucion de Ingresos en Lima
Metropolitana, 1968-1969. Lima: Universidad Catolica.
Figueroa, Adolfo (1975). 'The Impact of Current Reforms on Income Distribution
in Peru', in A. Foxley (ed.), Income Distribution in Latin America. Cambridge, Eng.:
Cambridge University Press.
Figueroa, Adolfo (1976). El Empleo Rural en el Peru. Geneva: ILO (mimeo).
Figueroa, Adolfo (1978). 'La Economia de las Comunidades Campesinas: El Caso de
la Sierra Sur del Peru', in E. Valencia et al., Campesinado e Indigenismo en America
Latina. Lima: Ediciones CELTAS.
Figueroa, Adolfo (1981). 'Agricultural Price Policy and Rural Incomes in Peru'.
Quarterly Review of Economics and Business, Vol. 21, No. 3.
Figueroa, Adolfo (1982). Reestructuracion Agraria en la Sierra Peruana. Lima:
Confederacion Campesina del Peru, Escuelas Campesinas, No. 1.
Hayami, Yujiro and Ruttan, Vernon (1971). Agricultural Development: An Inter-
national Perspective. Baltimore: The Johns Hopkins Press.
Hopkins, Raul (1981). Desarrollo Desigual j Crisis en la Agricultura Peruana,
1944-1969. Lima: Instituto de Estudios Peruanos.
137
138 Bibliography
Hunt, Shane (1975). 'Direct Foreign Investment in Peru: New Rules for an Old
Game', in A. Lowenthal (ed.), The Peruvian Experiment: Continuity and Change under
Military Rule. Princeton: Princeton University Press.
Hymer, Stephen and Resnick, Stephen (1969). 'A Model of an Agrarian Economy
with Non Agricultural Activities'. American Economic Review, Vol. LIX,
September.
Lajo, Manuel (1979). 'Industria Agroalimentaria y Transnacionales: El Caso
Peruano'. Publicaciones CISEPA, No. 43. Lima: Universidad Catolica.
Musgrove, Philip (1978). Consumer Behavior in Latin America, Income and Spending of
Families in Ten Andean Cities. Washington: The Brookings Institution.
Okun, Arthur (1975). Equality and Efficiency. The Big Trade-Off. Washington: The
Brookings Institution.
Pulgar Vidal, Javier (1972). Geografia del Peru. Lima: Editorial Universe
Quijano, Anibal (1968). 'Tendencies in Peruvian Development and in the Class
Structure', in J. Petras and M. Zeitlin (eds), Latin America: Reform or Revolution.
New York: Fawcett.
Roemer, Michael (1970). Fishing for Growth: Export-led Development in Peru,
195Q-1967. Cambridge, Mass.: Harvard University Press.
Ruiz, Liliana (1980). Terminos de Intercambio Campo-Ciudad. Bachelor's thesis in
Economics. Lima: Universidad Catolica.
Schultz, Theodore (1964). Transforming Traditional Agriculture. New Haven: Yale
University Press.
Thorp, Rosemary and Bertram, Geoffrey (1978). Peru 1890-1975: Growth and Policy
in an Open Economy. London: Macmillan Press.
UN-CEPAL (1978). Series Historicas del Crecimiento de America Latina. Santiago.
Vega, Gabriela (1979). Migracibn de Retorno a las Comunidades Campesinas. Bachelor's
thesis in Sociology. Lima: Universidad Catolica.
Webb, Richard (1977). Government Policy and the Distribution of Income in Peru,
1963—1973. Cambridge, Mass.: Harvard University Press.
Webb, Richard (1981). 'Aspectos Economicos de los Subsidios al Consumo', speech
of the President of the Central Bank of Peru in the Symposium on Nutrition and
Food Subsidies. Lima: January (mimeo).
Wiesskoff, Richard and Figueroa, Adolfo (1977). 'Viewing Income Pyramids in
Latin America'. Latin America Research Review, Vol. XI.
Statistical sources
Instituto de Planificacion (1973). Relaciones Interindustriales de la Economia Peruana.
Lima.
Oficina Nacional de Estadisticas y Censos (1970). Anuario Estadistico del Peru. Lima.
Oficina Nacional de Estadisticas y Censos (1975). Censo de Poblacibn, 1972.
Banco Central de Reserva. Memorias (1974-9); Resena Econbmica (1981) and Cuentas
Nacionales 1950-1974.
Ministerio de Trabajo (1981). Situacion Ocupacional Cuarto Trimestre 1980. Lima:
February.
Index
agricultural development in Peru 88-90 exchange relations in peasant communities

agricultural goods barter 21-2, 76
consumption-exchange 34 market (monetary) 21, 73-5
production 5 personal-impersonal relations 23
technology 7 reciprocity 21-2
agricultural seasonality expenditures in peasant communities
role in peasant-labour allocation 63 consumption 34-5, 108
in the sierra region 61-2 inter-regional 35-9
and wage determination 66 intra-regional 39-40
level 25-31
campesino see peasant family structure 24, 32-3
capital goods
accumulation 34 income in peasant communities
types 17-19 inequality 52
capitalist development in Peru monetary and non-monetary 42-4
features 90 stagnation 112
interrelation with agriculture 1 structure by sources 44-6
policies 82-3,103,109 total 41-2
role of peasantry 88 transfers 41, 46-8
credit markets trends 44, 52-7, 105
in peasant communities 75 income of peasant families
policies 124 level 43, 46, 47
structure 48-51
derived demand for rural factors transfers 49
concept 90-3 integration to the market system
measurement 94-6 degree 4, 36-9, 42-4
trends 96-7, in Peru and Latin market channels 35, 38
America 98-101
dualism labor markets
economic 4 costs of entry 56, 71
social and cultural 49 local-external 67-73
rural-urban 69
economic rationality of peasantry temporary see migrations
dynamics 121 wage determination 65-7
hierarchy in activities 61 land markets
risk-aversion 17, 58-9, 115 services (rent) 45, 75
'traditional' behavior 62 stocks 74-5
uncertainty 59 land reform in Peru 86, 88
engancbe system (debt peonage) 56, 72 land use in peasant communities
Engel's law 89 communal land 15
139
140 Index
land use in peasant communities {cont.) labor force 13-14
fragmentation 16 resource endowments 17-20
laymi system 15, 23 size 13-14
rotation systems 15-16 Peruvian economy
tenure 15 economic crisis 83, 102
Latin America growth and distribution 82
agricultural growth 98 inflation 83
comparative advantage 99 liberalism 83-5
economic development 1, 2 reforms 85-8
income concentration 97 state intervention 82
international trade 98 portfolio diversification
Lima and ecological levels 60
structure of family spending 94 of activities 58-61
wages and employment 105 of income 59
livestock of resources 17, 60
consumption-exchange-accumulation 34
production 5 risk-aversion behavior see economic
technology 7 rationality of peasantry
migrations
sierra region (highlands)
permanent 13, 75, 78 ecology and microclimates 8-10, 21
returns 76, 79 population 9
temporary 13, 42, 44, 56 southern sierra, as the most traditional
see also labor markets region 10-12
peasant communities
dynamics 44, 58, 82 technology
economic organization 5-8 and ecological levels 19
inefficiency 3 intersectoral relations 5, 7, 24
integration to the market 4, 42-4, 119 policies 125
non-capitalist relations 45, 119 traditional-modern 114-17
overpopulation 3, 77, 117
production structure 3, 5, 25-31 Z-goods
peasant family consumption-exchange 34-5
as economic unit 13, 23, 41 definition 5
education 78 production 6
technology 7
CAMBRIDGE LATIN AMERICAN STUDIES
1 Simon Collier. Ideas and Politics of Chilean Independence, 1808- 1833

3 Peter Calvert. The Mexican Revolution 1910-1914: The Diplomacy of Anglo-
American Conflict
4 Richard Graham. Britain and the Onset of Modernisation in Brazil, 1850-1914
6 Leslie Bethell. The Abolition of the Brazilian Slave Trade: Britain, Brazil and the
Slave Trade Question, 1807-1869
.7 David Barkin and Timothy King. Regional Economic Development: The River Basin
Approach in Mexico
8 Celso Furtado. Economic Development of Latin America: Historical Background and
Contemporary Problems (second edition)
10 D. A. Brading. Miners and Merchants in Bourbon Mexico, 1763-1810
11 Jan Bazant. Alienation of Church Wealth in Mexico: Social and Economic Aspects of
the Liberal Revolution, 1856-1875
12 Brian R. Hamnett. Politics and Trade in Southern Mexico, 1750-1821
14 Peter Gerhard. A Guide to the Historical Geography of New Spain
15 P. J. Bakewell. Silver Mining and Society in Colonial Mexico, Zacatecas 1564-1700
16 Kenneth R. Maxwell. Conflicts and Conspiracies: Brazil and Portugal, 1750-1808
17 Verena Martinez-Alier. Marriage, Class and Colour in Nineteenth- Century Cuba: A
Study of Racial Attitudes and Sexual Values in a Slave Society
18 Tulio Halperin-Donghi. Politics, Economics and Society in Argentina in the
Revolutionary Period
19 David Rock. Politics in Argentina 1890-1930: the Rise and Fall of Radicalism
20 Mario Gongora. Studies in the Colonial History of Spanish America
22 James Lockhart and Enrique Otte. Letters and People of the Spanish Indies: The
Sixteenth Century
23 Leslie B. Rout, Jr. The African Experience in Spanish America: 1502 to the Present
Day
24 Jean A. Meyer. The Cristero Rebellion: The Mexican People between Church and State
1926-1929
25 Stefan de Vylder. Allendes Chile: The Political Economy of the Rise and Fall of the
Unidad Popular
29 Anthony Hall. Drought and Irrigation in North-east Brazil
30 S. M. Socolow. The Merchants of Buenos Aires 1778—1810: Family and Commerce
31 Charles F. Nunn. Foreign Immigrants in Early Bourbon Mexico, 1700-1760
32 D. A. Brading. Haciendas and Ranchos in the Mexican Bajio
Cambridge Latin American Studies
33 Billie R. DeWalt. Modernisation in a Mexican Ejido: A Study in Economic
Adaptation
34 David Nicholls. From Dessalines to Duvalier: Race, Colour and National
Independence in Haiti
35 Jonathan C. Brown. A Socioeconomic History of Argentina, 1776-1860
36 Marco Palacios. Coffee in Colombia 1850-1970: An Economic, Social and Political
History
37 David Murray. Odious Commerce: Britain, Spain and the Abolition of the Cuban Slave
Trade
38 D. A. Brading (ed.). Caudillo and Peasant in the Mexican Revolution
39 Joe Foweraker. The Struggle for Land: A Political Economy of the Pioneer Frontier
in Brazil from 1930 to the Present Day
40 George Philip. Oil and Politics in Latin America: Nationalist Movements and State
Companies
41 Noble David Cook. Demographic Collapse: Indian Peru, 1520-1620
42 Gilbert Joseph. Revolution from Without: Yucatan, Mexico and the United States,
1880-1924
43 B. S. McBeth. Juan Vicente Gbme^ and the Oil Companies in Venezuela,
1908-1935
AA J. A. Offner. Law and Politics in A^tec Texcoco
45 Thomas J. Trebat. Brazil's State-owned Enterprises. A Case Study of the State as
Entrepreneur
46 James Lockhart and Stuart B. Schwartz. Early Latin America: A History of
Colonial Spanish America and Brazil

Adolfo Figueroa - Capitalist Development and The Peasant Economy in Perú (1984) PDF

Uploaded by

Copyright:

Available Formats

Adolfo Figueroa - Capitalist Development and The Peasant Economy in Perú (1984) PDF

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

Adolfo Figueroa - Capitalist Development and The Peasant Economy in Perú (1984) PDF

Uploaded by

Copyright:

Available Formats

CAMBRIDGE LATIN AMERICAN STUDIES

CAPITALIST DEVELOPMENT AND THE

The right of the

CAMBRIDGE UNIVERSITY PRESS

Cambridge University Press

© Cambridge University Press 1984

This publication is in copyright. Subject to statutory exception

First published 1984

Library of Congress Catalogue Card Number: 83—18861

ISBN 978-0-521-25397-0 hardback

List of tables and figures Page viii

2.1 Intersectoral relations in the economy of a peasant

This book summarizes much of my research work of the last five

Peasant families represent approximately 25% of the population in

Hypotheses on peasant economies

The production process

The process of exchange

The analytical framework

Table 2.1 Inter sectoral relations in the economy of a peasant community

Ecological levels and micro-climates

nation's population, 25% live in Lima. The next most important

Table 2.3 The sample communities

Accha-Sihuina (SIH) Sept. 1978 41 Cuzco

In order to study the functioning of the peasant economy the basic

Size of family and labor force

Size of family Total Adult

Jacantaya 4.6 (2.6) 3.9 (2.2) 2.6 (1.3)

Note: X= average; S = standard deviation.

Land: uses, tenure and fragmentation

Community Type of tenure Cultivated Fallow Total

Jacantaya Private 19.5 15.1 34.6

Culta Private 11.3 0.6 11.9

Ninamarca Private 2.3 1.4 3.7

Ancobamba Private 6.5 0.9 7.4

Ttiomayo Private 2.9 0 2.9

Huando Private 8.5 1.0 9.5

Acobamba Private 2.5 0.3 2.8

Note: Data for Sihuina were not available.

of crops JAC CUL NIN ANC TTI HUA ACO

Total (%) 100.0 100.0 100.0 100.0 100.0 100.0 100.0

Note: Data for Sihuina were not available.

in the production process. One type is used for consumption of its

Table 3.4 Animal stocks: number of animals per family

The economic unit and economic organization in

Table 3.5 Stocks of implements per family

A (Agriculture) (1) 3,700 14,896 0 2,854 5,401 26,851

Z (Manuf.) (8) 0 3,825 0 0 482 4,307

N+M+A + P + Z (15) 10,526 66,330 2,584

A (Agriculture) (1) 4,966 19,866 0 287 429 25,548

A (Agriculture) (1) 2,345 9,380 0 837 12,685 25,247

Z (Z-goods) (11) 0 2,900 0 116 1,712 4,728

A + P+ Z (12) 4,325 26,584 0 1,785 20,671 53,365

N+M+A+P+Z (15) 6,205 44,782 2,520

Net transfer (21) 0

A (Agriculture) (1) 8,777 35,046 351 356 44,530

Pj (Derived products) (2) 825 1,869 0 0 2,694

A (Agriculture) (1) 7,821 32,210 0 134 10,050 50,215

P2 (Consumption) (3) 0 2,710 0 73 2,898 5,681

A (Agriculture) (1) 6,222 24,888 0 362 4,557 36,029