Front Cover
 Title Page
 Table of Contents
 List of Illustrations
 Background to the case studies
 The case studies
 Back Cover

Group Title: Kumarian Press case studies series
Title: The impact of male out-migration on women in farming
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00086608/00001
 Material Information
Title: The impact of male out-migration on women in farming
Series Title: Kumarian Press case studies series
Physical Description: xviii, 78 p. : ; 28 cm.
Language: English
Creator: Palmer, Ingrid
Population Council
Publisher: Kumarian Press
Place of Publication: West Hartford Conn
Publication Date: c1985
Subject: Women in agriculture -- Developing countries   ( lcsh )
Migrant labor -- Developing countries   ( lcsh )
Rural women -- Developing countries   ( lcsh )
Migrant agricultural laborers' spouses -- Developing countries   ( lcsh )
Femmes en agriculture -- Pays en voie de développement   ( rvm )
Travailleurs migrants -- Pays en voie de développement   ( rvm )
Femmes en milieu rural -- Pays en voie de développement   ( rvm )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Spatial Coverage: Lesotho
Bibliography: Bibliography: p. 76-78.
Statement of Responsibility: Ingrid Palmer ; prepared under the auspices of the Population Council.
 Record Information
Bibliographic ID: UF00086608
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 11812039
lccn - 85005240
isbn - 093181622X (pbk.)

Table of Contents
    Front Cover
        Front Cover
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
        Page vii
    List of Illustrations
        Page viii
        Page ix
        Page x
        Page xi
        Page xii
        Page xiii
        Page xiv
        Page xv
        Page xvi
        Page xvii
        Page xviii
    Background to the case studies
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
    The case studies
        Page 15
        Page 16
        Southern Africa
            Page 17
            Page 18
            Page 19
            Page 20
            Page 21
            Page 22
            Page 23
            Page 24
            Page 25
            Page 26
            Page 27
            Page 28
            Page 29
            Page 30
            Page 31
            Page 32
            Page 33
            Page 34
            Page 35
            Page 36
            Page 37
            Page 38
            Page 39
            Page 40
        The Near East
            Page 41
            Page 42
            Page 43
            Page 44
            Page 45
            Page 46
            Page 47
            Page 48
            Page 49
            Page 50
            Page 51
            Page 52
            Page 53
            Page 54
            Page 55
            Page 56
            Page 57
            Page 58
            Page 59
            Page 60
            Page 61
            Page 62
            Page 63
            Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
    Back Cover
        Page 79
Full Text
r omenI' oles & Genderf
Differences in Development

SheThe II





The Impact

Women's Roles & Gender
Differences In Development.

The Impact
Ingrid Palmer

Prepared Under the Auspices of The Population Council

West Hartford

Copyright 1986 Kumarian Press
29 Bishop Road, West Hartford, Connecticut 06119
All rights reserved under International and Pan-American Copyright
Conventions. No part of this book may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopy, recording or
any information storage andc retrieval system, without prior written permission
of the publishers.

Printed the United States of America

Cover design by

Timothy J. Others

Library of Congress Cataloging in Publication Data
Palmer, Ingrid.
The impact of male out-migration on women in farming
(Women's roles and gender differences in develop-
ment; 7)
Bibliography: p. 76
1. Women in agriculture-Developing countries.
12 Migrant labor-Developing countries. 3. Rural
women-Developing countries. I. Title. II. Series:
'Women's roles and gender differences in development,
cases for planners ; 7.
HD6073.A292D4472 1986 331.4'83'091724 85-5240
ISBN: 0-931816-22-X


PREFACE ..................................................................... ix

SUM MARY ................................................................. xi

BACKGROUND TO THE CASE STUDIES ........................................ 1

INTRODUCTION ........................................................ 1
TYPES OF MIGRATION ............... ............................... 2

Theories of Pure Gain ............................................. 2
Theories of Private Gain and Social Loss: The General Challenge ..... 3

THE ISSUES FOR WOMEN LEFT BEHIND .............................. 4

Lack of Male Family Labor and Solutions .......................... 4
Decision-making Authority of Women ............................. 6
Alternative Deployment of Land and Women's Labor ................. 8
Rural Income Distribution and Its Effect on Women ................. 9
Family Health and Welfare ........... .............. .............. 10
Demographic Implications ......................................... 12
Economic and Social Changes During Long Absences of the Migrant .. 12
Return Migration and Its Aftermath ............................... 14

THE CASE STUDIES .......................................................... 15

SELECTION OF SOURCE MATERIAL .................................. 16

SOUTHERN AFRICA ........................................................ 17

LESOTH O ............................................................... 18
The Nature of Migration and Remittances of Earnings ............... 18
Utilization of Remittances ........................................ 19
The Impact on Agriculture ......................................... 20
Gains and Losses to Women Due to Male Migration .................. 22
Rural Income Distribution ........................................24
Changes in Some Determinants of Fertility ............. ........ ..... 2

BOTSWANA ........................... ........... .............. ...... ..... 26
The Nature of Migration and Remittances of Earnings ............... 26
Farm Assets and Total Incomes ..................................... 27
The Impact on Agriculture .............................. ...... 31
Alternative Labor Deployment ..................................... 33
Rural Income Distribution ........................................34
Changes in Some Determinants of Fertility .................... ... 34

SWAZILAND ...... ......................... ........... 3

The Nature of Migration and Remittances of Earnings ............... 3
Utilization of Remittances and Savings ............................. 36
Labor Power and Tractor Use in Farming .. ......................... 36
Women's Options and Decision Making ..................... ........ 39

THE NEAR EAST .............. ......... ...... ....... .. .. .. .. ... 41

YEMEN ARAB REPUBLIC .................... ......................... 41

The Nature of Migration and Remittances of Earnings ............... 41
The Impact on Agriculture ............... ...................... 42
Nonproductive Expenditure and Its Impact on Women .................. 44
Rural Income Distribution ....................................... 46

PAKISTAN .............. ................. ......................... 46

The Nature of Migration and Remittances of Earnings ............... 46
Utilization of Remittances ......................................... 48
The Impact on Agriculture ....... ............. ..... .............. 60
Nonproductive Expenditure and Its Impact on Women .............. 52
Rural Income Distribution .......... ... .... ............. ..... 53
Changes in Some Determinants of Fertility ......................... 64

EGYPT ... ............. .................... ......................... 64

The Nature of Migration and Remittances of Earnings ............... 66
The Utilization of Remittances ............................... ..... 66
The Impact on Agriculture ........................................ 87
New Responsibilities and Status for Migrants' Wives ................. 89
Rural Income Distribution ................ ....................... 61
Changes in Some Determinants of Fertility .......................... 61

TURKEY .............. ....... ....... .......................62
The Nature of Migration and Remittances of Earnings ............... 62
The Utilization of Remittances ............ ....................... 6
The Impact on Agriculture .... ... ......... ............. 65
Women's Growing Independence and Authority ..................... 63
Rural Income Distribution ........................................ 64
Changes in Some Determinants of Fertility .......................64

CONCLUSION ................ .................. ......................... 66

CONCLUSIONS FROM THE CASE STUDIES ..... ......................... 6

Difficulties and Incentives in Maintaining Farm Output .............. 66
Lack of Investment in Higher Productivity Farming .............. .. 66
Women's Limited Farming Resources and Authority, and
Their Other Options ........................................... 67
Family Welfare ..................................................... 68
Nonmigrant Households and Wage Laboring Income ................. 69
Changes in Some Determinants of Fertility ......................... 69

IMPLICATIONS FOR POLICIES ........................................ 70

Maintenance of Former Agriculture in the Absence of the
Male Head of Household .......................................... 71
Inducements to Productive Investment Before and After the
Permanent Return of Migrants .................................. 73
Maintenance of Real Income in Poorer Nonmigrant Households ...... 74

BIBLIOGRAPHY ........................................... .... ................ 76


Table 1. Percentage Distribution of Replies to "Which Are the Two
Greatest Problems?" ............................................

Table 2. Replies to the Question "Is It Better to Educate a Son or a
Daughter?" By Age and Sex of Respondent ........................... 2

Table 3. Composition of Male and Female Headed Households .................. 28

Table 4. Income and Value of Livestock and Farm Equipment, By Age of
Male and Female Heads' of Households ...............................28

Table 6. Income Levels and Percentage Composition of Income
By Household Category ...............................................31

Table 6. Sources of Power for Plowing by Type of Household,
1976 78 (Percent) .................................................32

Table 7. Person Primarily Responsible for Agricultural and
Livestock Care Tasks (Percent) ......................................37

Table 8. Source of Labor Assistance to Women Who Are Primarily
Responsible for Tasks (Percent) .....................................38


Why should development planners and scholars of
development be concerned about women's roles and gender

No project that expresses its goals in terms of
production gains or increased benefits can afford to ignore
the economic potential and needs of one-half of the
population. Guidelines for the design and evaluation of
development projects sensitive to women's roles have often
been applied only to a narrow range of "women's projects."
Our view at the Population Council is that all development
efforts could be improved if the differential impact on both
class and gender groupings were considered.

The series of case studies on Women's Roles and Gender
Differences in Development was developed to demonstrate that
such analyses are not only essential, but also feasible
within existing structures.

These case studies make clear how inattention to
women's roles and gender differences is played out as
projects are implemented. Excluding gender as a variable,
or limiting women's roles to the welfare sector, results in
unintended effects, sometimes positive, but more frequently
negative. Many of the stated objectives of the development
schemes under study were not attained because project
designs were predicated on an incomplete picture of the
society to be served and drawn into participation.

The case studies draw largely from material that
existed originally in other forms (such as exceptional Ph.D
dissertations). From these materials has been extracted the
"case:" (1) salient aspects of the culture and society in
which the development project was placed, (2) the project
dynamics themselves, and finally, (3) an assessment of gains
and losses in different goal areas. To complement
individual case studies, this series for planners includes
monographs on broader development phenomena whose effects
are seen outside the confines of specific development
schemes. As of this writing, the series includes two
monographs, one dealing with the effects of male out-
migration on rural women's roles and a second on the impact
of different styles of agrarian reform on women's roles and

These materials are intended to be used by students of
development and professionals in the field, including those
at the highest planning levels. By providing examples of
how individual development schemes have operated vis-a-vis
gender, we hope they stimulate in the reader an interest in
exploring what these effects might be in development
projects being designed, implemented, or evaluated. For
some time now, an understanding of class dynamics has been

seen as essential in designing projects for successful
outcomes. We have the same conviction regarding the
importance of understanding gender differentials. We hope
that this study series positively advances that notion and
provides its readers with new skills and insights by raising
questions and suggesting alternatives.

We wish to thank each of our individual authors for the
exhaustive work they have put into forming their material
into case studies. We commend Marilyn Kohn for her fine
editorial work in finalizing the material. Finally, we
acknowledge with appreciation the role of the Ford
Foundation in providing support for developing three of the
five cases and both monographs in this series.

Judith Bruce
The Population

Ingrid Palmer
Editor of the


There are two opposing views on the impact of male out-
migration on the rural household and the community. The
first claims that the impact is pure gain for both. Migrant
earnings are adequate for investment in improved agriculture
and therefore for income. If labor is not hired to
substitute for the migrant, kin or communal labor groups
step in to assist the wife on her own. Where labor is hired
this leads to better rural income distribution by raising
the real wage. The second view assumes private gain to the
migrant and his household, but a social loss.

Evidence points to declining agricultural output, at
least of subsistence crops. Opportunities for investing in
agriculture are poor. Migrant earnings are spent on
consumption items.

Neither of these views offers insights on how women
left behind see their options and make accommodations. The
issue of how migrant savings are allocated, and by whom, is


Lack of Male Family Labor and Solutions

Absence of male family labor affects phasing and land
preparation in particular. Harvesting is commonly done by
men and women, with women immediately involved with
processing crops--a task that cannot be delayed. Are
remittances for farm purposes adequate for hiring labor
necessary to maintain output? Is the technology and
frequency of phasing (and therefore yields and acreage
planted) affected by male out-migration? Do women have
problems in hiring and supervising labor? Do kinship and
communal networks assist women with land preparation and

Decision-making Authority of Women Over the Deployment of Resources

Some writers suggest that migrants' wives enjoy greater
managerial authority. But the real issue is how many more
decisions women make, how freely, which ones, and with what
resources? Can they decide on the crop mix and acreage
planted? Is land rented out? Can they choose to undertake
more off-farm work? Are they able to dispense remittances
according to priorities as they see them? How important are
considerations of risk in women-managed farming?

Economic and Social Changes During Long Absence of the Migrant

It cannot be assumed that the authority of migrants'
wives, their problems of accommodation, and farming
capability remain constant over many years. Much depends on
the size of remittances and the cultural background. Are
early remittances morel generous and regular than later ones?
Do women's lives become more or less difficult and their
responsibilities more or less onerous as time passes? If
remittances are large, do wives gradually withdraw from
agricultural' work, perhaps taking up some other economic

Family Health and Welfare, and Demographic Implications

The combination of maintenance of farm output and the
use of remittances determines the pattern of health and
welfare gains. This, raises questions on the ordering of
expenditure on nutrition, education, electrification, and
housing. The process of arriving at that ordering also
merits investigation. If male migration leads to the
nucleation of households, what are the gains for families?
Does male migration have an effect on fertility, apart from
the obvious one of partial separation of the couple?

Rural Income Distribution and Its Effect on Women

If farm output levels remain constant, male wages
should rise. But one of the accommodations that migrants'
wives may make is to farm less land. If this is accompanied
by local inflation through high-consumption. use of
remittances, it is uncertain that real male wages will rise.
Large farms usually rely on large households in which the
loss of one, or even two men may not have much impact on the
rural male labor market, except seasonally. The outcome for
real female wages is even more questionable. If migrants'
wives withdraw from field work the poorest women should
gain. But in some circumstances their numbers will be
swelled by migrants' wives who opt for more off-own farm


The Nature of Migration and Remittances of Earnings, and Utilization of Remittances

Men migrate to work in South Africa, spending one and a
half to two years away on each contract. In between
contracts they spend several months at home. Over a

lifetime, a man might be absent for a total of fifteen years
or so. In Swaziland there has been a shift to internal
migration, which allows for frequent home visits.

Earnings in South Africa are large compared with farm
income, but small by international migrant standings.
Remittances are extremely small, migrants preferring to
repatriate savings when they return between contracts.

The best information on utilization of remittances is
from Lesotho. The small regular remittances appear to be at
the disposal of wives for living and farming expenses.
There is general agreement between migrants and wives that
the first claim in the large "repatriated" sum is education,
followed by livestock acquisition, improved housing and
household effects, and clothing. The area of conflict of
interests is the migrant's allocation of his savings between
remitted and repatriated sums. In Swaziland internal
migrants (mainly being husbands) tended more to remit money
for plowing costs while external migrants (mainly being sons
in extended households) tended more to save for cattle

The Impact on Agriculture

Hiring labor is costly and, except in some cases in
Swaziland, remittances appear to be inadequate to cover
these costs. The general evidence is that male kin offer
negligible assistance. Arranging for the hire of male labor
and supervising it present problems for women. Difficulties
faced by women farming on their own were reflected in the
decline in farm assets in female headed households.

There is general agreement in the literature that
decisions affecting long-term investment are made by
migrants, and the Botswana information suggests that this
investment, if made at all, is made upon final return of the
migrants. But there is little doubt that day-to-day
management of the farm is undertaken by women. Women
interviewed in Lesotho claimed that they decided what crops
to plant and when and where to plant them. A study in
Botswana showed that female-headed households plant a
smaller percentage of their holdings, and plow the land less
frequently than male-headed households.

Migrants' wives demonstrate their ability to decide on
the deployment of their own labor resources. Their marginal
opportunity cost of own-farming is clearly not zero. In
high-risk agriculture, with difficulties in marketing grain
surpluses, and with great need of cash income, there are
strong incentives to work on other farms, to brew beer, etc.
That they shift some of their time from own-farming to other
economic activities is illustrated by the fact that in
Botswana women in female-headed households spend a smaller

proportion of Itheir time on their crops than do women in
male headed households.

Gains and Losses to Women Due to Male Out-migration

It is difficult ito see how many theories of pure gain
can be applied to migrants' wives in1this region, apart from
children's i education, land the exercise of managerial
functions. Even the latter must be qualified; it is a case
of managing meager resources. On the other side of the
coin, a survey in Lesotho revealed wives' anxieties about
lack of food and money and problems of obtaining medical
care. The greater effort by women did not lead to welfare
gains, and strain increased with duration of a husband's

Rural Income Distribution

This is not an area of total landlessness. Moreover,
the small remittances would not have done much for a rural
labor market. The literature points to income distribution
being an issue of differentials between women-headed and
men-headed farming households.

Changes in Some Determinants f Fertility

Fertility rates remain high. Of the commonly held
determinants of a decline in fertility, only one--
educational improvements--appears to be activated by male
out-migration. There is evidence from Lesotho that mothers
favor daughters' education in order to give them
independence. But this positive development has to be offset
against women's iAsecurity, partly through conjugal
instability, and their need of children's assistance in


The Nature of Migration and Remittances of Earnings, and Utilization of Remittances

Labor migration from this region is to oil-producing
countries of the Middle East, and from Turkey to Europe.
First remittances are delayed and modest when debts have to
be repaid. Among the poorest migrant families, as in
Pakistan, net remittances may not be large enough to relieve
wives of !additional burdens forl a couple of years.
Established remittances from the Middle East can be
extremely large; fori instance, $200 to $450 a month was

recorded in Yemen, while in Pakistan regular remittances
could be up to five times and irregular remittances as much
as 160 times the local (monthly) male wages.

Utilization of remittances depends on who receives
them, whether the migrant's wife uses them to establish
separate residence and economic activities from the extended
household, social attitudes to women, and profitability of
investment in land and agriculture. In Yemen, for instance,
wives are left in the custody of male kin, and because low-
yielding food production is unable to compete with cheap
food imports remittances are diverted into high consumer
spending. In Pakistan women in extended families purchase
more food, small livestock for household-consumed protein
foods, and sometimes a buffalo, while women in nuclear
households, after an early struggle during which land might
be rented out, hire labor. Pakistani wives gradually
withdraw from most field tasks as remittances allow. But
poor opportunities for improving agriculture lead to much
larger savings being spent on house construction and
consumer durables when the migrant finally returns.

A small study in an Egyptian village with a
comparatively good village and market infrastructure showed
that women in extended households, most receiving
remittances directly, spent the money in disengaging from
the household's collective consumption and production, and
then moved to separate new residential units. This seems to
be a dominant motivation of migrant and wife with the wife
acquiring major new financial, productive, and supervisory
responsibilities. Some invested in farming, but this
appears frequently to have been combined with the wife
investing in a nonagricultural occupation. Opportunities for
doing so would seem to overcome an initial subservient and
non-decision making status of women.

The Impact on Agriculture

In this region remittances, once established, are large
enough to cover hired plowing services. After the first
years of problems there should be no financial reason why
farm output levels cannot be maintained. Nuclear farming
households established before the departure of husbands
appear to achieve this. But women who disengage from
extended households face problems in establishing a new farm
on their own, and may settle for a nonagricultural
occupation, perhaps combined with some farming. Although
there was no information on the impact on the farming of the
household they left, there is no reason to suppose it was

It may be too soon to draw conclusions on any long-term
investment in farming after the migrants' return. But what
evidence there is suggests that former migrants see their

future in nonagricultural, usually urban-based, self-
employment. I A weak market in land, or inauspicious
opportunities for raising agricultural productivity, are
amongst the reasons.

Nonproductive Expenditure and Its Impact on Women

There is little doubt that the greater expenditure on
food and other consumer items improved the welfare of women.
Where potable water and electricity; were available wives
could experience only a rise in living and working
conditions. Being able to use grain mills further reduced
their work burden. But any improvement in the social status
of women that followed could be defined by high consumption
(as in Yemen) or by new independent managerial roles (as in
the small Egyptian study). Women who remained in extended
households shared physical benefits with other members, but
their status remained changed. Nonproductive expenditure
does not in itself appear to have improved the status of
women in the community except in terms of venturing more
into shops.

Rural Income Distribution

Male wages have clearly benefited, though by how much
depends on the amount of local inflation generated by
remittance expenditure. The impact on female wages is
likely to be even less uniform. In IPakistan some migrants'
wives gradually withdrew from (own) field work, while
others, together with wives of small farmers who had to give
up farming because of higher male wages, entered the wage-
labor market. In Yemen some poor women found employment in
migrant households, but nonmigrant small- scale farming
households faced lower returns to marketed produce because
of higher male wages aid cheap food imports.

Changes in Some Determinants of Fertility

Predictions on fertility changes have to be speculative
apart from the influence of absent husbands. Big increases
in income, expenditure on better nutrition, children's
education and marriage expenses should, all other things
being constant, allow for the critical expectations of a
decline in fertility. In the long term it is likely that
children's labor in farming would decline. Migrants are also
likely to return more knowledgeable about family planning
and with new social ideas. But against this is the absence
of any change in women's social status--except in the cases
of migrants' wives in Egypt who achieved management of a
viable livelihood.


Maintenance of Farm Output

Where plowing costs affect wives' incentives to
maintain former levels of output, subsidized tractor
stations might bear a positive social benefit:cost ratio,
and even encourage migrants to remit larger sums fo:
farming. Concentrating on other aspects of female-managed
farming could bring alternative gains. Which food crops do
women favor to produce, process, and sell on their own? Can
women's interest in small livestock be capitalized on? In
general there is a need to raise women's authority in the
eyes of hired laborers and servicing and marketing
organizations, and accord them access to enabling inputs and
services. Greater effort will have to be made where the
free-market infrastructure is weak.

For wives who will receive large remittances after
initial years of struggle, a collective rotating credit fund
would assist continuity of farming.

Inducements to Positive Investment Before and After the Permanent Return of Migrants

Measures to assist wives to maintain farm output can be
seen as laying the basis for wives to manage new
investments. Utilizing extension services to discuss such
possibilities with migrant and wife before the former's
departure, and sending extension officers to concentrations
of nationals abroad make economic sense if they lead to a
more productive utilization of remittances.

The studies tended to confirm that it is high-risk
agricultural areas that supply large numbers of migrants.
Private investment in agriculture remains unattractive
unless collective or social investment in the environmental
base is made. It might be desirable to combine migrants'
contributions with government contributions to upgrade the
local environment.

Maintenance of Real Income in Poorer Nonmigrant Households

Improving the infrastructure of food supply from other
areas should help to moderate local inflation of food prices
where this is a problem. Professionalizing small-scale
agriculture while husbands are absent should help to
maintain or increase the demand for female labor. A program
to upgrade the environment would raise farm labor by
increasing the production-bearing capacity of the land. A
concentration on small-scale female-managed farms (nuclear


households) also restrains any tendency toward labor-
displacing machanization.

In these ways t e social costs of male out-migration
can be reduced by empowering women to manage better on their
own and by recognizing that this can be a precondition to
diverting migrants' savings from conspicuous consumption to
productive investment.



Male migration from rural areas is a growing world
phenomenon that is now recognized as having far-reaching
consequences for the areas of migrants' origin. In the early
1980s the number of migrant workers was estimated at twenty
to twenty-one million worldwide (Bohning: 1984 p. 24).

Migration was initially seen in terms of labor
responding to market forces, and therefore allocating itself
more efficiently. Governments have also seen the potential
of remitted earnings, and have initiated policies directed
toward facilitating migration and, in some cases,
encouraging migrants' savings into national bonds and
similar deposits. Today, while there is widespread
agreement that migration results in a clean net benefit to
the migrant himself, the assumption of net benefits to his
family, the farming community he leaves behind, and
ultimately his country has been challenged. Yet policy
makers have given little attention to the migrant's family
and its economic base.

This study looks more closely at the dilemma posed by
male out-migration from the perspective of the women left
behind in rural areas. The principal purpose is to throw
light on the options faced by them and the accommodations
they make, as well as their reasons for making them, and on
long-term investment practices of migrant families. While
such an investigation cannot be expected to answer
questions about all the consequences of male out-migration,
it is firmly believed that a study that focuses on gender
issues can illuminate many dark corners of the general
controversy over migration, and thereby enable policy makers
to design interventions appropriate to national priorities.

It also hoped that the case studies will together
provide policy makers with a framework in which to pursue a
more comprehensive information base on the impact of male
out-migration in their respective countries as an aid to

The variety of rural backgrounds of migrants and the
influences of migration on them is, worldwide, enormous. In
order to retain depth in presentation of the case studies,
and to point out some sharp comparisons, only two regions of
the world from which migration originates are included:
Southern Africa and the Near East.

It is often said that the poorest of the poor, or the
absolutely poor, do not migrate since they do not have the
resources to finance the travel and maintenance costs before
income is received. Therefore a range of households--from
deficit farming to large-surplus farming households--should
be the principal suppliers of migrants. The larger the farm
the more likely is it to be operated by a large household
(three generations or married siblings working and living
together). In households of this type the male family labor
supply will be larger than average so that more than one
male member (married or unmarried sons) might seek migratory
employment. In contrast, the small farm is normally
cultivated by a small or nuclear family. The migrant is
usually the husband, often the only adult male in the

Internal migration is defined as a movement to another
part of the migrant's own country. External migration means
leaving the country i of de jure residence. In general,
internal migration brings less return, but it is easier and
involves less initial capital cost. Also, in the majority
of cases it allows more frequent home visits. The frequency
of home visits, howev er, varies enormously by region and by
distance to the migratory employment. Thus a simple
distinction between internal and external migration is not
meaningful when assessing impact. The term oscillatoryy
migration" has been used to describe situations with home
visits of nonseasonal regularity. In some instances there
is a pronounced seasonal element to visits, associated with
the agricultural timetable. When we look at information on
the total number of years spent by a migrant away from his
home during a working life, the term semipermanentt" also
comes to mind. Because of the mix of long-term absence
broken by short-term visits the pertinence of the terms
oscillatory and semipermanent should be considered in
relation to the nature of the felt impact.


Theories of Pure Gain

The neoclassical approach: This is a largely abstract
approach based on the theory that labor is deployed such
that it maximizes its returns. Neoclassical theory states
that the marginal product value of labor (which is assumed
to determine the rural wage) is less than the wage that can
be earned from migration. The migrant is able to remit to
his family! a sum adequate to hire labor at the going wage
rate to substitute for himself if this is required. A bonus
to his family is that his absence means one less resident
adult unit of consumption, so that there is a rise in the

household's per-capita food consumption. Furthermore, the
migrant's net gains allow investment in new farm resources
thereby overcoming past constraints on breaking out of low-
productivity agriculture. Farm output is not only
maintained, but should actually increase. Rural income
distribution improves. The decline in the supply of labor
leads to a rise in wages. While the poorest section of the
community, which cannot afford to migrate gains from this,
the largest farmers who employ most labor must lose.

The traditional kinship approach: This approach
recognizes that there is a specific seasonal need for the
migrant's labor at home, and that it cannot be assumed that
hired labor is available or affordable. Kinship relations
(with extended family substituting for the migrant) and
exchange labor provide adequate support. According to this
theory, with many slack periods during the annual
agricultural timetable the absence of a migrant's labor is
felt only at specific times of the year when the total
household labor force is fully employed. In large extended
households this can be covered by new work arrangements.
Where a nuclear household experiences difficulties because
its principal or sole adult male is away, the husband's kin
step in to replace the missing labor. This traditionalist
theory emphasizes the most important seasonal task of the
male migrant, land preparation. Where kin support falls
short of the measure there can sometimes still be a
fortunate complementarity between seasonal male migration
and maintenance of output. Net migrant earnings are pure
profit for the household. This theory, which does not
include an active wage-labor market, implies no change in
rural income distribution.

Theories of Private Gain and Social Loss: The General Challenge

Much of the recent empirical data on rural areas of
male out-migration point to declining agricultural output,
at least of subsistence crops. There is also evidence of
worsening income distribution. Loss of national production,
and greater inequality of income distribution are described
as "social losses." All this has led to more qualified, but
still largely speculative, comments on the benefits of
migration. While it is still assumed that the migrant
enjoys a private net gain, concern over these social costs
of migration has resulted in questioning the use to which
private gain is put, and the strategies of adapting
agriculture to a reduced household labor:land ratio if
remittances are not earmarked for hiring substitute labor.
It is not questioned that remittances are made, but the
prioritizing of their use has become a matter for concern.

The empirical evidence for the use of remittances to
maintain output is patchy and inconsistent. Family


consumption expenses (especially school fees) appear more
significant than support of farm output levels. It is clear
that in the absence of a new government agricultural policy
there is no noticeable investment in higher-productivity
agriculture. It has been argued that frequently there is no
obvious potential for such investment. Had there been, the
cost of travel to migratory employment might already have
been so invested. Available evidence on the proportion of
remittances used for nonproductive purposes also suggests
that few see an opportunity for improved agriculture. Where
there are no obvious opportunities for upward mobility
through agricultural improvement upward mobility has to be
sought through consumer expenditure, which provides outward
signs of social arrival.

But such a conclusion points merely to possible long-
term strategies on the part of migrants, and leaves a void
in the matter of what is happening back on the farm where
roles must be changing. There is still the nagging thought,
"Why not at least maintain levels of farm output, especially
since so much of remittances goes to family support?"
Finally, it is worth remembering that the challenge to the
ideas of pure gain has focused solely on the dualism of
private gain to the migrant himself and social loss. Aside
from the migrant and society, there is the migrant's wife
and family. How do they fare? It is at this stage that
gender relations of production, exchange, and decision
making must be introduced.


Lack of Male Family Labor and Solutions

Male out-migration presents the farm household with a
diminution of the supply of labor for male-typed and shared
tasks. The sexual division of labor and the seasonality of
labor requirements render the notion of "marginal product of
labor" unhelpful. If the plowing cannot be done at all, or
cannot be done every year, this lowers the marginal (and
average) product of planting, weeding, and harvesting labor.
In fact, the technology is different. Plowing and other
land preparation--commonly men's tasks--have to be done.
Where there is relatively little landless labor available
for this intensive job, which has to be done within a short
time period, the cost of hiring labor will bear no relation
to any abstract notion of marginal product value. What this
means is that the cost of hiring male labor for crucial
tasks is probably much greater than theoreticians have
supposed. If draft animals are owned and used their year-
round care may be a new imposition on remaining working
members ofl the household. The direct cost plus the
opportunity cost of Caring for these animals may be greater
or less than the cost of hiring animals for the plowing.

But an additional cost somewhere in the household farm
system will be for draft power with a diminution of year-
round family labor. An alternative to draft power is to
hire a tractor. This may or may not be more expensive than
draft power, depending on supply and demand, respectively,
for draft animals and tractors.

The labor market facing women employers can include
"imperfections." First, women usually have less ease of
entry into information networks of the male labor market;
likewise the hired tractor and draft animal markets. Thus
women have greater difficulty shopping around for the best
bargain. Second, much of the labor will have recurrent
seasonal agreements with certain farmers (most likely from
the larger farms and households where there is a resident
male head) who have first call on available labor for hire.
The outcome could be that women are able to obtain
assistance only at the end of the plowing season, well into
the planting season, and the job may be rushed. A lower-
quality product for the common going price then ensues.
Third, the ability of women to supervise hired male labor
may be less than that of men. The shift to hired labor
"reduces the incentives to heavy and efficient labour-input,
especially if many of the supervisors are women, who may not
easily command the respect of labourers who are used to
taking instructions from men" (Lipton 1976: 30). Again the
outcome is lower-quality product for the price. Or the
hired labor may not even turn up for the agreed job. All
these factors are dependent on gender relations in the

The plowing problem is only the starkest example of
difficulties women may face. Men and women usually harvest
and carry the crop together. It is difficult to see how
women, even with the assistance of their children, can work
more intensively and longer to offset the absent husband in
these seasonally rushed jobs. Because of the seasonally
high demand for labor, hiring of extra hands will be at a
premium, and once again bear no relation to abstract ideas
of marginal products. The alternative is additional
seasonal work stress for women.

The solution offered by kin and other support to the
women left to farm on their own may not always be rooted in
reality. Even with a husband present the traditional ideal
is not always followed. Where male out-migration is
massive, labor assistance from relatives declines because so
many families are short of labor (Gordon 1981: 62; Kossoudji
and Mueller 1981: 3). Moreover, where relatives and friends
believe a woman is receiving large remittances they may
assume that she is not in need of traditional support

All these considerations are much more relevant to
small or nuclear households in which male migration halves

or eliminates the adult male family labor force. In
extended or joint households the remaining men are more
likely to cope with men's tasks or arrange for hired labor
to be used effectively.

The questions that arise for this study are:

*Are remittances for farm purposes adequate for hiring
labor necessary to maintain output?

*Is the technology of plowing affected by male migration?

*Do women have problems in hiring and supervising labor?

*Is plowing undertaken less frequently or on less land?

'Do kinship and communal networks !assist women with land
preparation and harvesting?

Decision-making Authority of Women

In a general study on the impact of male out-migration
Lipton (1976: 19) comments that with the supply of working
men depleted, women gain by being more important in the
workforce, and through the formation of women-headed
households. The implication is that an emancipating
authority follows. Colison (1970) suggests that women's new
independent managerial role could mean appropriating not
only the decision-making role of husbands but also the
functions of the extended family.

However, when it comes down to meaningful detail there
is a fog surrounding the subject of wives' and absent
husbands' spheres of decision making. The question at issue
is not whether women on their own are making more decisions,
but how many more decisions, how freely, and which ones.
Women may be able to' make day-to-day decisions on family
maintenance while absent husbands may have the final say on
major farm decisions. i Some observers imply that such major
farm decisions are capital investments and credit raising.
Within these constraints are a host of other decisions to be
made. If an absent husband stipulates that a certain acreage
must be cultivated 'or a certain amount of some crop be
produced, farm resources can be shuffled while still meeting
these targets.

A woman' s authority over the flow of remittances may
or may not be comparable to her authority over existing
productive resources. li One migrant husband might leave his
wife with decisions on land use but retain control over the
destination of his remittances; another might decree land
use and crop mix but be generous with remittance use. The
consequences for cash crop-production, food production,
maintenance of farm assets, and general welfare will be

quite different in each case.

Of course the preferences of migrant and wife on total
resource deployment might coincide. But there are three
reasons they could differ. First, long absence and
infrequent communication could lead to the migrant being
unaware or unappreciative of difficulties and opportunities
arising at his home base. Second, his long-term strategy of
saving as much as possible for a transformation of his
livelihood on his return could clash with his wife's
shorter-term interest in more immediate compensation for his
absence. Third, it might be culturally regarded as the
wife's responsibility to maintain the children from the land
left in her charge, and the husband's right to retain the
bulk of, or to strictly dispense, his earnings.

A woman's decision-making authority should not be seen
relative only to her husband's. Migrant earnings sometimes
offer the potential of expanding productive assets, and a
greater asset portfolio requires new areas of decision
making per se. If there is collaboration between migrant and
wife both their decision-making roles in farming will be
expanded, jointly or separately. This would be particularly
true if migration had the effect of separating extended-
household farming into nuclear-household farming in which
more assets were under the control of couples. A woman's
decision-making authority can also be examined in the
context of the wider community.

The above discussion has frequently focused on women in
nuclear households. It is easy to assume that the weak
authority of junior women concerning farm decisions will
remain weak, especially if remittances are sent to a father
or brother who then retains ultimate management. However,
even in extended households, particularly those of married
siblings, the migrant might own a separate piece of land
with separate draft animals. The produce might be seen as
his wife's, contribution to the entire household's

A number of questions arise for the purposes of this

What kinds of decisions affecting farm output can a wife

Is cash-crop production foregone because of problems of
access to credit, extension services, and markets in the
absence of an adult male household member?

How large and how frequent are remittances?

Are remittances earmarked by the migrant for farming and
consumption purposes?

* How great a capital expenditure can a wife make without
consulting her husband?

SIf a migrant's wife is observed to enjoy more resources
does this Iaffed her standing with cooperatives,
extension services and market suppliers? Does this
extend her decision making into realms where she
didn't previously contribute to decisions?

And for women remaining in extended households:

* Who receives the remittances, and what powers of control
over them accrue to the recipient?

What changes are affected in the decision-making
authority of migrants' wives in farming matters?

If a migrant has individual ownership of a piece of land
usually farmed by the whole household, does his wife
assume managerial and produce appropriation rights over

What kind of a collaboration in production or investment
decisions passes between migrant and wife? Is this
collaboration direct, or is it mediated and censored
by a resident male?

Does partial privatizing of surplus accumulation occur
in a large extended household? And is the role of the
migrant's wife relevant to the process of transforming
remittances into private productive asset accumulation?

Alternative Deployment of Land and Women's Labor

If cash remittances from migrants do not allow for
production and consumption "as usual,'" their wives are bound
to make changes in production and income-gaining activities.
How extensive these 'changes are, and to what degree they
affect agricultural production and rural income distribution
depend on women's authority to redeploy resources and on
incentives accruing to them personally. Here we consider
possibilities and reasons they would have for that

Plowing and other tasks that men and women
traditionally share may present serious constraints to the
continuation of farming patterns after male out-migration. A
different use of the family holding is an obvious solution.
Working the entire holding with the usual crop mix would
lead to lower yields and net returns. Altering the crop mix
by shifting to less labor intensive crops or to a crop mix
that moderates seasonal labor demand peaks is another
alternative. Its common observation suggests that it
results in a| smaller fall in net returns than attempting to

continue as usual. However the criterion of net returns
might well be less important than somehow getting around the
labor bottleneck.

A reduction in acreage cultivated--almost inevitably
accompanied by an alteration in the crop mix--is another
solution. The outcome for aggregate agricultural production
and rural income distribution depends on whether women rent
out the uncultivated land, and to whom.

Women might choose to completely redeploy their labor
on the family's holding by a different use of that land.
They might elect to undertake wage employment on the other
larger farms for the first time or to increase it if done
previously. Introducing this possibility challenges the
accepted wisdom that the opportunity cost of farming for
women remaining in rural areas is virtually zero since they
can earn so little off their own farms. However, incentives
to take wage employment can arise not only from calculations
of marginal returns to the separate parts of a mixed work
portfolio. The net returns to hiring labor and purchasing
farm inputs are uncertain. With new, untested work
arrangements, and kin support in the absence of husbands,
women farming on their own are bound to consider new
strategies of risk avoidance.

Therefore, the questions we need to ask include:

Do wives of migrants reduce the acreage they cultivate,
and if so which crops are most affected?

What are the determinants of this strategy as seen by the
women themselves?

Is land rented out, and if so is there one overriding

How important are risk and uncertainty in farming in
decisions to take wage employment on other farms?

Rural Income Distribution and Its Effect on Women

Male out-migration will lead to some increase in the
demand for male hired labor. All other things constant,
male wages should rise. Do we assume that the female wage-
labor market remains unchanged? If wives of migrants choose
to take up or increase their wage employment because of
difficulties in farming on their own, the supply of female
wage labor increases and, all other things constant, the
female wage will decline. Furthermore, if family land
remains uncultivated as a result of migrants' wives turning
to wage employment, this is the equivalent of numbers of
women entering the area from outside seeking work. The
supply of female labor rises. On the other hand, if the

land is rented out, demand for female labor will rise; more
so if land is rented to large landowners than to small
farmers. Even so it is unlikely that this will prevent a
depression of female wages with a swelling of the numbers of
women seeking wage employment. The result should be a
widening of! the sex differential inl wages. For landless
couples gains could outweigh losses. But for (de facto or de
jure) women heads of households or main breadwinners there
would be a clear loss.

Owners iof larger farms will lose from higher male
wages, and gain from lower female wages. But in many cases
they have large household labor forces from the combination
of several nuclear families. Farm income losses from lack of
crucial male labor are unlikely to occur to them. If they
can retain remittances to the whole household their surplus
accumulation should be rapid. It could be that this ability
to accumulate surplus rather than paying a larger wage bill
will be the main contribution to, future rural income
distribution. Therefore, the ability of a migrant's wife
initially left in an extended household to control
remittances and establish a separate household may be a
factor determining the outcome of male out-migration for
rural income distribution.

Farming households in which a male member has not gone
to migratory employment are particularly affected by a rise
in wages of hired labor. Moreover, these households
together with wage-laboring households face the inflationary
impact of remittances. | It is far from clear that male wages
always rise as fast as the local cost of living, at least
after the first few years of a large exodus; and it is
scarcely credible that .female wages should do so.

The questions to be asked concerning rural income
distribution are:

'What has determined any increase in demand for male
and female wage labor?

Has an increase in demand come mainly from nuclear
migrant households?i

*Has there been an increase in the supply of female wage

Have wages risen faster than local inflation?

Family Health and Welfare

The large proportion of remittances that are alleged to
be used for family consumption suggests that nutrition is
improved. Remittances might be readily earmarked for school
fees if education of! children or younger siblings is one
I !

clear goal of migration. Medical expenses, however, are
often unforeseen, and could result in a situation that
requires women to find additional money on short notice.
Improvements in the living environment, particularly housing
are believed to be a popular use of migrant earnings. The
effect on women's day-to-day workload is uncertain. But
electrification and a better water supply would be clear
gains in terms of reducing women's domestic burden.
Therefore, the nature of remittance use for family
consumption, and the chronological ordering of different
consumer durable acquisitions, need to be investigated.

Work and social stresses on women with absent husbands
need not always be compensated by purchases. It is
pertinent to explore whether overall stress increases or
decreases over time, and if so what the sources of stress
may be. Speculation about farming conditions, type of
household, cultural attitudes toward the rights of men to
retain their earnings and toward women's family maintenance
responsibilities, and the effect of long separation does not
lead to obvious hypotheses about the worst years of
husbands' migration for women's health and welfare.

Wives initially left in extended households may or may
not remain there. If they do remain they may still be able
to use remittances to purchase food for themselves and their
children, and to improve their children's education. If
they plan to move out new house construction should mean
better living conditions.

Some relevant questions to ask are:

Does household nutrition benefit from remittances?

Are remittances spent on improved education?

SWhat kinds of consumer durables are purchased, and how do
they affect women's domestic workloads?

Is housing improved by new construction or the installa-
tion of electricity?

Does stress for wives increase or decrease over time?

What are the main sources of stress?

Does the social status of a woman on her own increase or
decrease over time?

And for women left in extended households:

*Does the whole household gain equally from remittances
spent on daily consumption items, or does a migrant's
wife and her children obtain preferential nutrition and


Do a wife and her children gain more in health and
welfare if they move to a separate residential

Demographic Implications

The impact on short-term fertility will depend on the
frequency of the migrants' home visits. In the long term
fertility could be influenced by modern ideas on family
planning and suitable family size that migrants bring back.
But whether these ideas take root will also depend on
whether the size of migrant earnings and savings provide the
hope of a future livelihood not dependent on a large family
labor force, on improved expectations of old age security,
and on whether educational expenditure leads to a
perception of higher costs of raising children. Because
information ;on fertility changes in the case studies that
follow is insignificant, only speculative comments can be
made in the second section on the direction taken of some of
the assumed determinants of fertility change, such as the
use of children's labor, investment in education, and better
standards of living.

Economic and Social Changes During Long Absence of the Migrant
It is reasonable to surmise that changes at home
accumulate over the many years of a migrant's absence.
Every change need not move in the same direction as the
years pass. A wife, grows older and children grow up.
Production and consumption patterns change.

The most important single determinant of the direction
of change must be the flow of remittances and the way their
utilization alters. Regardless of whether the size of
remittances increases| or decreases, their distribution
between production and consumption purposes is likely to
alter, as will their distribution between productive
investment and consumer durables on the one hand, and
expenditure on variable production costs and immediate
consumption on the other.

These questions and issues will have great influence on
the nature of the outcome for women of long-term male
migration. The size of migrant earnings will clearly be an
important factor, and it is tempting to suppose that the
major difference will be between cases of distant external
migration on the one hand, and internal or nearby external
migration on the other. However, there are also important
underlying factors. High-risk agriculture cannot always
profitably absorb investment. Poor general infrastructure,
no electrification, and a weakly developed consumer market
are likely to influence consumption patterns.

If Lipton's suggestion that women gain through their
headship of households over a long time carries some truth,
in what way does this come about? And is it a real gain or
just greater responsibility to make ends meet and things

An important social change that could have lasting and
widespread effects is a new basis to the marital bond. A
long separation does not necessarily mean a weaker alliance
between husband and wife. A useful approach is to explore
the dependence of the migrant on his wife for maintaining or
improving a rural livelihood to which he intends to return.
If this dependence is great, the couple are likely to
consult each other in earnest. The migrant must rely on the
achievements of his wife, and what she does achieve should
gain his respect.

One particular situation could bring them much closer.
If the migrant departs from a multiple generation or joint
household he may have it in mind to work toward a nuclear-
household economic base so that surplus accumulation is not
dissipated but remains under his control. In this case he
is relying on his wife not only to develop managerial skills
but to undergo something of a social transformation during
the years of his absence. It is difficult to see how this
can be realized without a stronger forging of the marital
bond because it requires a move from extended social
relations to focusing on the working alliance between the
two partners. For a woman to carry out this change without
the support of her husband's presence involves a high degree
of real personal emancipation.

Some relevant questions to ask are:

How rational is the distribution of purchases between
producer and consumer items in terms of women's capabili-
ties and needs?

Are early remittances most generous while later ones
taper off as migrant and family get used to
separate worlds? If so, do farm assets and consumption
reach a peak in early years and then decline?

If early remittances are very small, how severe are the
problems that women face in the early years?

If remittances increase, do wives hire more labor,
including female labor, and gradually withdraw from
agricultural work, perhaps taking up some other economic
activity on their own account?

SHow rational is the distribution of purchases between
producer and consumer items in terms of women's
capabilities and needs? How far can women influence pur-
chases? What kind of consumer durables are purchased, and


* how do they affect women's domestic workload?

* Does a migrant'si dependence o his wife to turn
remittances into productive or other assets create a new
relationship between them?

Return Migration and Its Aftermath

The immediate effect of the migrant's permanent return
is an increase in the supply of male household labor and a
resolution of the particular labor bottleneck imposed by his
absence. On his return the farm is likely to be in a
different condition then when he left. It may be run down
and need to be built up. Low (1982) poses a model of farm
changes during and after male out-migration for Lesotho in
which low yields, land left uncultivated and a depletion of
farm assets are tolerated so that' savings from migrant
earnings are large enough to invest in a transformation to
higher productivity agriculture on his return. In other
situations, where remittances have been used to hire labor,
improve land resources, and introduce mechanization, the
migrant returns to a better endowed farm. The difference
for women is that in the latter situation they will have
acquired new managerial skills and might also have withdrawn
from some fieldwork.

But an overriding consideration could be whether low-
productivity or high-risk agriculture has a potential for
improvement that competes with opportunities to invest
savings in a nonagricultural livelihood. If it does not,
the migrant will probably take up another occupation. But
it is unclear what his wife will do. Attachment to land may
lead to retaining the low yielding (but gilt edged security)
farm that is then worked by the wife together with hired
labor. Or if earnings and savings are very large,
agriculture may be abandoned and the land sold, with the
wife withdrawing to the home, especially if there is a move
to an urban area.

If it is true that large savings are spent on better
housing, consumer durables, and bigger marriage expenses,
all indicating upward social mobility, what is the effect on
women? Does this encourage withdrawal of women to the home
and an increase in dowries? Do women trade off a smaller
total work burden for more social restriction? Or are they
able to capitalize on the skills and status they won when
they had to cope alone?


It is apparent that governments' hopes for benefits of
migrant earnings have not been realized. Yet there has been
almost no policy intervention. The principal purpose of
this monograph is to throw some light on the accommodations
that migrants' wives make, their reasons for making them,
and on long-term investment practices of migrant families to
enable policy makers to design interventions appropriate to
national priorities.

It is also hoped that the case studies will provide
policy makers with a framework in which to pursue a more
comprehensive information base on the impact of male out-
migration in their respective countries as an aid to


With such a profusion of information over a broad
literature on different aspects of the consequences of male
out-migration, it was not easy to select which sources to
use. One option was to glean the literature on one aspect
at a time. This was rejected on the grounds that a common
issue can spring from different sets of causes just as its
consequences can be great or small depending on the actual
overall situation. To extract detail from noncomparable
contexts runs the risk of error, in both diagnosis and
treatment, Ifrom reductionism. Instead it was decided to
present case studies separately. This left the problem of
selecting the studies.

Settling on regions or subregions of the world has one
substantial advantage:i regional overviews are provided. In
addition, inasmuch as ithe nature of male migration is likely
to be similar in an area, looking at several countries
within it should build up a more comprehensive overview of
issues from a number of studies that address different
aspects. Where differences in issues between countries of a
region come to light,i these are of more analytical value
than differences noted between regions of the world.

The two regions of Southern Africa and the Near East
were selected for the contrasts they present in farming
systems, traditional role of women, and the size of
remittances. In addition, Southern Africa illustrates well
the problem of other parts of Africa where male out-
migration is increasing, namely the difficulty of
identifying and treating the obstacles to maintaining
agricultural! production, especially food production, on
female-managed farms. i This problem, together with that of
making agricultural livelihoods more attractive so as to
stem the rising dependence on food imports, is becoming more
apparent in the Near Eastern countries where a large exodus
of rural men has been more recent.

The issues and questions raised in the first section
are wide-ranging and detailed. The case studies were not
undertaken with the purpose of answering all of them, and
most of the studies provide information on only a few.
Moreover, when examining the impact on women they do not
include a common core of issues. Because the studies focus
on the impact on women of male out-migration, especially on
women's new work obligations and status, the information on
actual farming changes is the most deficient. The Southern
African studies offer much more information on farm
management problems. This can be put down to the longer
listing of large-scale male out-migration and researchers'
greater awareness of the consequences for agricultural
production. But even within this region the several case
studies tend to focus on different aspects of farm
management. : If we can assume that differences in overall

situations of female-managed farms are not very great among
these countries, the case studies together build up a
substantial profile. The studies from Near East countries
are much more deficient in farming information. This may be
due to the very large remittances inclining researchers to
examine changes in the status of women resulting from
greater purchasing power. Nevertheless, the case studies
provide pointers to policy makers who wish to design
effective agricultural programs.

Because of the variation in issues addressed, it was
decided to treat each case study with respect to its own
merits, and to convey as full a picture of the lives of
migrants' wives as space allowed. But some consequences
inevitably follow. While the ordering of the issues laid
out in the first section provides the general framework for
deriving analysis of the case studies, not all presentations
share the same structure, and a few have very different
structures. Comments in the nature of drawing comparative
conclusions on the issues are kept to a minimum here in this
section, they are raised in the concluding section.


Peasant agriculture in Southern Africa is reputed to
have always been low yielding and, for decades, inadequate
for family maintenance. With men contributing labor
primarily at the preplanting stage it allegedly made sense
for men to leave the homestead to earn wages elsewhere.
Internal male migration varies by country, and has changed
over time with the pattern of development. But the
proximity of the South African economy and the attraction of
relatively high wages in its mines, industry or construction
has encouraged a long-established and extensive movement of
men from Lesotho, Botswana, and Swaziland.

Semipermanent, oscillatory, and seasonal male migration
are all in evidence. The first represents absences of
several years at a stretch, and much longer absences in
total, perhaps with permanent return only at the end of an
active work life. The term oscillatory migration has been
applied not only to internal migration (Vletter 1978: 20),
but also to migration to South Africa for the legal limit of
two years for a single work contract followed by leave of
some months before the next departure (Gordon 1978: 3,5).
However, when internal migration includes weekly or monthly
visits, as in the Swazi (Vletter 1978: 33, 51), it is bound
to incorporate a seasonal element. Indeed, national
employers have complained of regular absenteeism at plowing

This region's agriculture is characterized by a poor
environmental base, and high risk due to drought,
hailstorms, pests, and crop diseases. Maize and sorghum are


the staples. Cattle figure prominently as a source of
traction power and a savings reserve to be liquidated in
emergencies or for marriage payments. Cattle are more
important in Swaziland than in Lesotho and Botswana, and
more important in the mountains than in the lowlands.
Farming is therefore a very lowly yielding, gilt-edged
security that barely', or inadequately, underwrites the
reproduction of the population. Class divisions are
evidenced by differences in assets and by the hiring in or
out of household labor.


The Nature of Migration and Remittances of Earnings

Since '1963 Basotho women have not been allowed to work
in South Africa so that all more recent external migrants
can be considered male. An estimate of over 110,000 male
Basotho working in South African mines in 1973 would
represent almost half the adult male labor force. This is
believed to have risen to 121,000 in 1976, with another 30-
80,000 working in agriculture, construction, and industry
(Vletter 1978: 132). Since the 1976 Census recorded a total
of 234,160 married women, between 40 and 60 percent of
married women in Lesotho are left to manage on their own for
most or all of the year.

It has been estimated that external migrants spend on
average fifteen years (to thirty-five percent of a working
life) in the mines.' This period is punctuated by home
leaves. Average duration of a single contract in a gold
mine was 14.4 months in 1974 and 10.6 months in 1976. More
than two-thirds of migrants took up another contract within
six months of returning home (Murray 1980: 23).

Households with larger herds of cattle and more land
have markedly more members (and are therefore likely to be
extended households) than households of low-asset status.
Therefore, the better-endowed households tend to send one or
more sons away to wage employment while the poorest tend to
send husbands who ma'y be the only adult male in the unit.
Male migration has the effect of leading to female-managed
farms among the least well endowed households. Also, larger
remittances accrue to those households with greater farm

Earnings in South Africa are high compared with farm
income. Even the lowest-paid job there can return in a
month more than a year's effort in farming in Lesotho
(Sibisi 1980: 7). Mueller (1977: 68-69) quotes a 1966
estimate of 18.7 percent of migrant earnings being remitted.
In 1973 it was estimated that 66 percent was repatriated
(including remittances, transfers of deferred wages, and

goods and cash brought by hand, the last amounting to 42
percent of total repatriated money). This 66 percent would
amount to 160 Rands (or 13.3 Rands a month) out of total
earnings of 239 Rands.

But in her study in 1974 Mueller found that remittances
averaged 7 Rands a month. They appear to be made monthly,
after a delay of four to seven months after starting

Who receives and dispenses remittances is influenced by
household type. Young migrants' wives living with parents-
in-law will not normally handle their husbands' remittances.
The head of household usually largely determines who
migrates and controls resources. A young wife is
subservient to her mother-in-law and follows directions on
work. She will have no fields of her own to cultivate or
from which she has produce to dispose of on her own.
Without home-based resources under her control, or separate
maintenance responsibilities, sending remittances to her
might be seen as inappropriate by the migrant who will still
be under the influence of his father. While this appears to
be the structured majority situation, individual instances
show variation (Ibid.: 176-77). The determining factors are
likely to be the age of the migrant, how long he has been
married and how many children he has, and the approach of a
disengagement from the larger household. At some stage a
migrant will build his wife a separate house. While the
wife will have an obvious interest in moving to a separate
establishment, her parents-in-law have an obvious stake in
keeping her with them.

The wife of a migrant in a nuclear household will
receive remittances directly; she will also handle her own
sources of cash income. But she may have less freedom over
the use of remittances than migrants' fathers enjoy.
Moreover, the calculations of remittances mentioned above
are averages and their regularity was assumed. Showers
(1980: 9) mentions that "meager and often erratic sums of
money" are sent because migrants see their income as their

Utilization of Remittances

Showers also points out that remittances are frequently
earmarked by the migrant for certain purposes such as
livestock, household furniture, and clothing. (Other items
of expenditure from remittances include school fees, soap,
candles, tea, and sugar). Murray (1980) confirms this,
adding that men's and women's priority expenditures do not
always coincide. Other items of expenditure from
remittances include school fees, soap, candles, tea and
sugar. What does appear clear from the literature is that
the migrant expects food requirements to be met from


cultivation of the holding and from women's small cash
income. Perhaps as a consequence of this, investment in
farming (excluding livestock) does not figure at all.
i I
Mueller offers a more structured examination of the way
remittances (or, rather, repatriated savings) are earmarked.
Migrants prefer to send only small sums in regular
remittances, and to bring the bulk of their savings back
with them in Icash and goods. The effect of this is a
separation of the use of migrants' savings between day-to-
day living and farming expenses (from regular remittances);
and choices made between education expenditure, livestock
acquisition,! improved" housing and household effects, and
clothing (from savings brought back).

If there are quarrels between husband and wife over the
issue of utilization of his savings they are over the very
small amounts allocated to every day expenses and the
husband's lack of appreciation of his wife's daily struggle
to take care of the family. Because the consequences of
this are seen mainly1 in wives' decisions on farming and
other productive work, they are discussed later.

Mueller (1977: 180-86) found that men and women agreed
that education had the priority claim on the larger sums of
money brought back. In spite of men being interested in
building up herds of cattle, her survey showed that
household effects were the next priority for both sexes
(although women showed a stronger affirmation than men for
this investment), followed by livestock investment. It is
possible that men's answers were influenced by the fact that
care of larger herds of cattle was not practical while they
were still in migratory employment. Expenditure on clothes
came last with both men and women. The conclusion is that
there are no fundamental conflicts over the use of the bulk
of savings.i But this accord is reached on the basis that
decision making is confined to within, not between, the two
types of expenditures1. Once the migrant has determined the
sizes of remittances and savings, the main options for their
wives are closed.

The Impact on Agriculture

With women doing all or most of field agriculture
except land preparation and plowing it could be hypothesized
that if male labor could be hired or male kin relied for
planting and plowing, I the maintenance of farm output would
present no problem.

Male kin cannot be relied on to substitute for absent
male family labor. Fathers-in-law may be old, and brothers-
in-law in migratory employment. Also, relationships with a
husband's kin weaken! when the husband is away for a long

All the available references leave no doubt that the
absence of male family labor is the cause of a decline in
small-farm output. Hiring labor is costly. Remittances are
too small to cover this additional cost when there are
competing claims made on them. Farm input costs peak in
certain months, and a woman's cash flow cannot cope with
this. As we have seen, the larger savings brought back by
migrants are not considered relevant to farm output

Care of cattle, including the time-consuming task of
collecting fodder, is usually undertaken by men and boys.
While women can and do care for cattle when necessary, it is
less compatible than crop cultivation with domestic work and
child care. The time constraint also means that care of
cattle must compete with agriculture for women's time. The
effect of this is that over the years of a husband's absence
livestock numbers decline because there is inadequate
incentive to replace animals lost due to death or forced
sales during drought years. When no draft animals are owned
cash costs are incurred in hiring them.

How much freedom do wives enjoy in redeploying land and
their own labor to take account of new agricultural cost
structures or lower net returns to their own farm effort?
There is some debate over the demarcation of migrants' and
wives' farm decision making. Showers states that men have
ultimate authority, and that a woman with an absent husband
must still follow his directions, risking serious punishment
if she acts independently. But she adds that whereas the
day-to-day decisions are theirs, "some of the long term
decisions must [also] be made by women" (1980: 10).
Conflicts can arise when men have no clear idea of the
options and problems of the farm and the household. Gordon
(1978) revealed the difficulty of disentangling decision
making. While 80 percent of her sample of women replied
that absent husbands made important decisions, 74 percent
claimed that agriculture and livestock were their own
responsibilities. This does not take us very far. Being
responsible for working the fields and caring for the
animals might allow the right to determine how the work is
done, but choosing the crop mix, the intensity of labor
application, and the leaving of some land uncultivated are
also important. All these decisions may well have to be
made by women. Certainly women's minuscule amounts of
working capital, and their lack of direct access to credit
limit their options regardless of husbands' directives.
While Gordon believes that husbands still make long-term
decisions concerning agriculture and livestock (presumably
concerning capital investment), she suggests that there is
an area of joint decision making in the shorter term.
Mueller is clearer. She states that women claimed during
interviews that they usually made the decisions on what
crops to plant, when and where to plant them, and whether to
use exchange labor practices (1977: 204). When the husband

is present these decisions are made jointly. Extended male
absence allows women considerable latitude and independence
in decision making.

There are, of course, decisions other than those
exclusively concerned with the management of own-farm
agriculture and livestock care, but that bear on how much
time is devoted to this primary production. All the
references cited here give evidence of alternative sources
of gainful work for women. Should a woman do more laboring
on other farms and less beer brewing, or vice versa? Or more
of both at the cost of own-farm agriculture? That women on
their own do have the authority to decide on the deployment
of their own labor is revealed by the range of their (cash
and imputed) income sources.

The range of off-own-farm sources of gainful employment
allows women to reshuffle their work portfolio to maximize
income and avoid risk. By all accounts self-provisioning
agriculture is not strengthened through greater
femininization of own agriculture, and it is quite wrong to
assume that the opportunity cost of women's own agriculture
is anywhere near zero.i!

Income can be raised from the sale of small livestock,
weeding and harvesting on neighbors' fields (for cash or
kind), brewing beer for sale, or working on food-for-work
projects. Weeding and harvesting for larger farmers is
usually the most regular off-own-farm work. These larger
farmers are usually the most efficient farmers, they can
better bear risks, and even when rainfall is poor they do
better than small farmers. The key ingredient of beer is
sorghum, which may be obtained from winnowing the grain from
others' farms. Beeri brewing is the most important single
source of cash income under the control of women, and
households with no migrant remittances might obtain all
their cash income from beer sales (Gay 1980: 51). But
acquiring cash income' from these sources entails foregoing
some own-farm work, which invariably means leaving part of
the holding uncultivated rather than cultivating the entire
holding by less satisfactory methods.

Gains and Losses to Women Due to Male Migration

Where migrants' earnings facilitate or accelerate
departure from an extended household,! there are a number of
possible benefits for women: the end of a subservient role,
greater ability to rationalize their labor deployment, more
control over the returns to their agricultural work, and the
opportunity to handle some cash income. When migrants' wives
give up some own farming directed at self-provisioning to
obtain cash income they exercise further independence in
decision making.

Expenditure on children's education is no doubt a
source of satisfaction to them, and investment in the house
should improve part of their standard of living.

But against this, Gordon paints a picture of an
intensifying struggle. Her study of 521 migrants' wives
revealed that strain is felt more strongly by women over
twenty-five years, and it rises with age (1978: 25-28).
Particular strain was shown to be associated with husbands
being away more than ten years, the presence of four or more
children, and residence in the mountains (where livestock
herds are more important). Why should poverty weigh more
heavily on women heads of households as the years pass?

Gordon asked her sample what were the two greatest
problems they faced. Their responses are shown in Table 1.


Greatest Problem Greatest problem
n =351 n = 193

Agriculture 47 16
Livestock 8 12
Illness, getting medical care
children 6 17
self 4 3
others 3 8
Hunger, lack of food 8 8
Poverty, lack of money 5 3
Attacks, robberies 4 3
Family affairs 4 5
Miss husband 1 4
Other 9 20

Source: Gordon 1978: 41.

The responsibility for what is almost wholly self-
provisioning agriculture is by far the most frequently
mentioned problem. The fact that the problems of managing
production from the holding looms so large does not suggest
that significant assistance is forthcoming from husband's
kin. The majority of the women interviewed did not think
that male relatives substituted for husbands. "Any
assistance that kinsmen may provide therefore appears to be
not extensive or effective enough to significantly influence
the way in which the wives function or the burden that they
bear" (Gordon 1981: 71). It is reasonable to suppose that
kin support, if any, declines over the years.

There are other sources of strain on the women. Table 1


showed that difficulties encountered in coping with illness
and looking after livestock are comparable. Murray (1980:
37) also mentions illness as a problem felt by women: for
children this meant gastroenteritis, pneumonia, and
malnutrition; for adults malnutrition, sexually transmitted
diseases, and respiratory diseases. Medical expenses then
compete with food deficits for scarce cash income.
Allocating inadequate resources carefully on a day-to-day
basis requires constant decision making. This kind of
independence' cannot be counted as a gain, and Gordon
concludes that "Independent decision making, thus, cannot be
seen to be an activity enjoyed by wives, but rather one
associated with unpleasant stress" (1978: 68). Yet three-
quarters of her sample claimed that their husbands carried
all the responsibilities for their children. Those who
believed their husbands did not concern themselves with
their families tended to be older women whose husbands had
been migrants for more' than ten years. It is likely, then,
that material and nonmaterial support from migrant husbands
weakens over time.

What could be interpreted as inconsistencies in the
responses led Gordon to conclude that women were resigned to
expecting husbands' minimal participation in family affairs
and planning. "Theyj appear to expect little from their
absent husbands and should some small' indication of support
be forthcoming on his part, he is seen as fulfilling his
role (1978: 69). While women may adjust to male migration
in the sense that they'and their children survive, Gordon's
data show that adjustment, or ability to cope, does not
improve over the life cycle. On the contrary, strain
increases with age.

Finally, something should be said of the social effects
of male migration. I I spite of the upheaval caused by the
migration, cultural demarcation of the sexes' roles, land
inheritance, patrilocal marriage, and agnatic structures
have remained intact. Yet Murray (1980: 31-35) shows that
alongside the stability in these areas there is much
turnover of household members, conjugal instability,
illegitimacy, desertion, and family breakup. Children may
move to other households, and women may return to relatives.
Mueller (1977: 177-78) gives a slightly different view,
referring to "not infrequent separations and unions," and
also pointing outithat divorce rates are low because with
men controlling; resources women have almost no viable
alternative to marriage.

Rural Income Distribution

Migrants' savings should raise the nominal amount of
total cash \and imputed incomes accruing to the sum of all
household members. Large households will in all likelihood
do much better than smaller ones, especially if they rent

out draft animals. Men who hire out their labor will also
enjoy increased wages. But it is not clear that income
resources at the disposal of women have increased. The
issue of the impact of male migration on rural income
distribution rests mainly on distribution between women and
men, and on disposable income for family maintenance between
male- and female-headed households.

Changes in Some Determinants of Fertility

The priority given to children's education in the
allocation of the bulk of migrants' savings signifies a
movement from the economic value of children's labor to
incurring greater costs of children. What is also
significant are signs that daughters' education is receiving
more approval, especially from mothers. This might be
related to concern over old- age security. Mueller reports
women claiming that daughters are more loyal than sons to
their natal families. She quotes one response: "When a man
marries he forgets his parents. A daughter never forgets"
(Ibid.: 139). Table 2 shows her information on men's and
women's preferences relative to daughters' and sons'


Son Daughter Both Son
and Daughter

Age of Respondent

18-29: men 3 1 2
women 2 17 3

30-39: men 6 1 8
women 8 15 16

40-49: men 3 4 10
women 11 7 22

50-59: men 4 1 16
women 6 6 16

60+: men 4 0 14
women 8 7 27

Source: Mueller

1977: 142.

Respondents' explanations for these replies were
strongly influenced by their opinions on which child would
later support them. l But not only the dividends to parents
are at stake. Respondents stated that education gave girls
some compensation for erratic financial support from their
future husbands, and even that educated daughters would have
an alternative to marriage. There was also evidence that
the majority of women (95.5 percent between 18 and 29 years)
preferred to see their daughters work than marry. The
majority of men under fifty years showed the same preference
(in the eighteen-td-twenty-nine-year age group, 83.3
percent). Mueller sees this as a new ideology recognizing
that "for |a woman work yields a better life than does
marriage." i Many mothers and young unmarried women called
marriage a trap for fools.

Brideprice is not as regular as formerly, and is
difficult to enforce. Mueller comments that parents are
beginning to consider a daughter'is income as a more
effective means of social security than brideprice cattle.

What comes through this discussion is that determinants
of fertility rates are in transition, but that the need for
later social security is probably inducing a "wait and see"
strategy on the parts of.mothers at least with respect to
the earnings and loyalty of educated grown daughters.


The information for this country is of a different
nature than that for Lesotho. Whereas information on Lesotho
emphasized the disposal of migrant savings and the nature of
difficulties in carrying on alone as expressed by the women
themselves, the studies on Botswana provide quantitative
information on farm-management issues of households with and
without adult male labor. To the extent that the issues for
women and farming are similar in the two countries, the two
sets of information ican be seen i as complementary and
mutually illuminating.! In particular, this case study helps
to explain why migrants' wives find life more difficult with
the passage of time.

The Nature of Migration and Remittances of Earnings

The proximity of South Africa again accounts for male
migration. !The 1971 census indicated that male:female
residential ratios were 54:100 in the twenty-to-twenty-four-
year age group, 50:100 in the twenty-five-to-twenty-nine-
year age group, and 67:100 in the thirty-to-thirty-four-
year age group (Kossoudji and Mueller 1981: 27). The
ratios in rural areas were even less because of internal
migration of rural-based men. According to the Rural Income
Distribution Survey (RIDS) 43 percent of rural households

were headed by women.

The same survey included an item of "transfer payments"
in estimates of household income. But 40 percent of these
payments pass between households. It is possible that they
are indirectly influenced by migrants' remittance--although
there is no information on this-but they do not represent
direct remittances. In extended households remittances are
sent to fathers rather than to wives. The survey showed
that 95 percent of female-headed households and 88 percent
of male-headed households received transfer payments. The
former also received larger sums. Female-headed households
with no adult male present received transfer payments of 80
Rands bringing their total income to 434 Rands, while male-
headed households with adult males present received 46
Rands, bringing their income to 903 Rands (Ibid.: 40).

Farm Assets and Total Incomes

Before discussing problems of farming some data on
resource differences between types of household are given as
an aid in understanding initial and changing farm cost
differences. But first, because of the importance of some
adult male family labor and of dependency ratios (or
consumption demands on family labor power), data on member
composition of types of household are offered.

The term "female-headed household," which is used in
the quantitative data, arises from the RIDS asking
respondents to identify the household head (providing that
he or she is a regular member).

The definitions of four categories of household used in
the survey are given in Table 3. In the third category the
designation of a female head (with adult male present) is
either due to the woman's age and widow status or because of
her effective control over assets. The large size but low
dependency ratio of these households reflects the presence
of adult children. Although women on their own without any
adult male labor (FH-NMP) have the smallest households, they
have the highest dependency ratios.

Table 4 includes only the categories MH-MP and FH-NMP;
it shows farm assets of the two extreme types of households.
However, the data need to be treated with caution. They do
not follow the same households through their life cycles.
When migrants return permanently their households move to
the male-headed category. Also, a woman head over fifty
years with no adult male labor is likely to be a widow or


I (0-14 yrs)
Percent of Average
Total House- Number of Per Adult Per Adult
holds in Members Male and Female
Sample Female


1. Male head:
male aged 20-64
present (MH-MP) 49 7.3 1.3 2.5

2. Male head:
no male aged
20-64 present
(MH-NMP) 8 6.4 2.2 2.2

3. Female head:
male aged 20-64
present (FH-MP) 13 7.6 1.3 2.4

4. Female head:
no male aged 20-64
present (FH-NMP) 28 5.1 2.5 2.5

5. Head unidentified 2

Source: Kossoudji and Mueller 1981: 8,39.


Age of Mean Cattle Mean Small Mean Equivalent
Head Mean Income Value Livestock Value Value


less than
30 627 616 561 610 58 69 118 131
30-39 1359 493 1506 218 148 58 306 57
40-49 1027 481 1343 450 186 76 360 121
50-59 1072 379 1792 309 151 63 332 102
60 up 1057 374 1554 325 166 60 542 133
i~ j

Source: Kossoudji and Mueller 1981: 41.

As the age of the male head rises there is rapid cattle
accumulation. But in the case of a woman head with no male
labor, cattle holdings fluctuate around a long-term decline.
How much of this decline is due to forced sales in bad years
or voluntary sales because cattle are too troublesome for
women who have alternative income opportunities is not
known. While these are only averages for each category,
other data suggest even wider discrepancies. The 1979 Water
Points Survey revealed that 57 percent of female-headed
households owned no cattle, against only 19 percent of male-
headed households. Carol Bond's 1974 survey indicated that
over 50 percent of female-headed households owned no cattle,
and that those that did averaged fewer animals than male-
headed households (Fortmann 1980: 3,10-11). Therefore it
seems fair to conclude that probably at least half of
female-headed households have no draft animals.

In the case of small livestock and farm equipment,
while there is no clear change in female-headed households'
assets over time, there is a considerable increase in these
assets in male-headed households.

Table 4 shows that income of male-headed households is
closely correlated with value of cattle assets, less closely
with the value of small livestock and equipment. This
demonstrates the very close relationship between adult male
presence and cattle on the one hand, and household income on
the other. The relationship between cattle value and mean
income is much weaker in the case of female-headed
households over the life cycle. Cattle ownership and care
are men's province. But minimal numbers of animals are
useful as draft power. However, also in the case of female-
headed households there is no recognizable relationship
between small livestock or equipment (which are roughly
maintained) on the one hand, and income (which declines) on
the other. The data do suggest growing leakages of income
from some other source as the woman ages. One explanation
might be declining remittances, or terminated remittances
with the death or desertion of the husband, which could
affect averages; other explanations include accumulating
farm management and cost problems, or weakened ability to
undertake off own-farm work.

Like Lesotho, Botswana has high-risk agriculture.
Assets acquired from migrant savings can be lost by the
poorest households through forced sales in a drought. The
male head category must include cumulations of households
with permanently returned migrants as average age of male
head rises, and therefore must include numbers of recent
female-headed households that lost assets during droughts.
The sharp drop in cattle and farm equipment of female heads
between the ages of thirty and thirty-nine years is the
clearest evidence of the difficulties faced by women whose
husbands are long-term migrants. Higher age groups of women
heads would include larger numbers of widows.

Disparities in land assets between household types are
not quite as great as in cattle ownership. The average
holding is fourteen acres for MH-MP and MH-NMP households,
twelve acres for FH-MP households, but only six acres for
FH-NMP households (Kossoudji and Mueller 1981: 44). All
female-headed households averaged one-third less land than
all male-headed households. Women usually gain access to
land through their husbands or the fathers of their
children. They can also, like men, obtain land from the Land
Board on their own application, on condition that they are
able to cultivate it. Bettles (1980: 5-6) suggests that
women heads may start out with less land than male heads or
have land confiscated because they are unable to work it.
She provides information to show that although 71 percent of
male-headed households and 58 percent of female-headed
households have more than seven acres, the former plant 61
percent of their holdings but the latter only 51 percent.
Because women on their own are either unable or unwilling to
cultivate all the land at their disposal (see below for
reasons), they are unlikely to take the opportunity to
acquire the use of more land.

Overall, female-headed households average less than
half the earned income of male-headed households. Transfer
payments narrow the differential only slightly. When we
take account of size' and age composition, female-headed
households are 25 percent worse off than male-headed
households in terms of total income per adult equivalent
(Kossoudji and Mueller 1981: 11).

Mean-income figures in Table 4 indicate the net effect
of women heads' struggles over time. Theirs appears to be a
losing battle, while male-headed households gain, then lose,
and then enjoy a moderate improvement. The latter is very
likely due to migrants returning to a rundown farm to become
male heads again, later to build up the farm.

What is of particular interest is that in proportionate
terms male-headed households obtain a higher percentage of
their total income from agriculture than do female-headed
households, even though they have larger total incomes (see
Table 5). Female-headed households rely relatively more on
wage employment as well as on manufacturing (including beer
brewing) and gathering' of wild foods. These figures shatter
the assumption that women have no better alternative to only



Average income, excluding transfer
income (Rands) 857 749 645 354
Average income, including transfer
income (Rands) 903 802 749 434
Average total income per adult
equivalent (Rands) 178 188 142 136
Sources of income (% of total):
Crops 13.1 8.6 6.7 8.7
Animal husbandry 46.7 69.1 39.6 32.2
Wage labor 20.0 3.2 23.4 22.8
Manufacturing 3.0 3.8 3.8 5.5
Trading 1.2 0.1 2.9 1.0
Services & construction 3.1 1.4 1.9 1.2
Hunting & fishing 1.8 1.1 0.6 0.1
Gathering 4.6 5.4 5.6 7.8
Transfers 6.4 7.2 15.5 20.6

Source: Kossoudji and Mueller 1981: 40, 42.

The Impact on Agriculture

There is a fairly clear demarcation of men's and
women's agricultural tasks. Carol Bond's referential survey
is commonly quoted (Bettles 1980; Fortmann 1980; and
Kossoudji and Mueller 1981). Men customarily do about 80
percent of all land preparation. Women assume dominance from
then onward, except that men make a substantial contribution
to transporting crops from the field. Women can and do take
over all these male-typed jobs out of necessity, but this
requires giving up something else they are doing. If they
choose not to forego these other economic activities and not
to maintain draft animals (assuming they have not sold
initial stocks of cattle to buy food), they must find the
working capital and incur cash costs to get somebody else to
do the plowing, at least. If male kin or some exchange
group system is still available, the effective hiring cost
should be less. However, without a resident male head, male
kin support gravitates away, and male exchange labor is
impossible. Fortmann points out that women hire plowing
services at a higher cost than the imputed cost for
households who exchange male labor (1980: 8).

Another particular problem of plowing that emerges
strongly from the literature is timeliness and reliability
of quality work. If draft animals or tractors have to be
hired, women may find themselves at the end of the queue,
the resulting late plowing affecting yields. Hired plowmen

may not care how well they do the job on someone else's

One consequence if these difficulties is that female-
headed households may not plow every year. Surveys
conducted in 1976, 1977, and 1978 indicated that the
percentage of male-headed households that did not plow in
those years was 16, 19, and 38, respectively, compared with
34, 34, and 54, respectively, for female-headed households
(Fortmann 1980: 12). Table 6 shows the source of power used
by those who did plow.
1976 78 (PERCENT)

Male-headed households Female-headed Households
1976 1977 1978 1976 1977 1978

Used Own Oxen 78 77 79 51 47 61
Used Hired/borrowed
Oxen 10 12 7 36 34 22
Used a Tractor 115 15 13 12 14 12

Source: Fortmann 1980: 13.

aThese figures relate only to those households which
did plow. Figures totaling more than 100 percent are
due to farmers using more than one means. There is no
explanation for totals for female-headed households
not reaching 100 percent.

The differences between the two types of households
must be interpreted through the additional findings that (1)
cattle owners were significantly more likely than non-cattle
owners to plow; (2) male- and female-headed households that
owned cattle were equally likely to plow; and (3) male- and
female-headed households that didi not own cattle were
equally likely to plow. Of those which did plow, there was
no significant difference in the incidence of use of
tractors, but a very great difference in the use of hired or
borrowed oxen. From this we can conclude that the danger of
less timely and reliable plowing is greater for female-
headed households, as they are far less likely to have

The costs of acquiring and retaining cattle (against
pressures to sell cattle to raise money to buy food and to
retain other time-consuming economic activities) are high.
They may be higher than the cost of hiring draft animals or
a tractor. Either way women are at a cost or income-losing

With no sexual division of labor we can easily see the
sense of reducing planted acreage in proportion to the loss
of adult household labor if hiring labor is uneconomic in
terms of alternative work. But with a strong sexual
division of labor and no household adult male labor, the
incentive to reduce acreage planted is stronger. Kin
support and exchange labor for crucial tasks are rarely
available. Plowing costs involve paying the highest
seasonal rural wages. Borrowing money for working capital
raises the costs of production further.

According to the RIDS, female-headed households
averaged net profits of 7.91 Rands per acre compared with
11.57 Rands per acre for male-headed households (Bettles
1980: 6). If female-headed households planted six acres,
then their additional farming costs would amount to 21.96
Rands. This sum is just over one-quarter of their total
transfer payments received, and amounts to two-thirds of the
difference between male- and female-headed households'
transfer payments (80-46 Rands). Were female-headed
households to attempt to cultivate as much as nine acres
(still much less than the fourteen-acre average of male-
headed households), the whole of the difference between male
and female heads' transfer payments would be swallowed up by
higher farming costs. These comparisons are given only to
indicate the order of magnitude of these higher costs. A
woman's decision on how much land to cultivate will depend
on how she views returns to alternative gainful work.

Data on yields suggest that a reduction in acreage
planted is preferred to attempting to cultivate the entire
holding using less labor-intensive methods (which makes
sense in terms of plowing costs per unit of output).
Fortmann (1980: 9) quotes a survey undertaken in 1972 in
Manyana that shows that female-headed households averaged
1.2 bags of sorghum per acre, while male-headed households
averaged 1.9 bags. On the other hand, the RIDS revealed
equal levels of productivity. The survey also comments that
the extra burden on women heads of households results in
lower proportions of the holding being cultivated.

The clearest evidence of lower net economic returns to
own agriculture is that women in female headed households
spend a smaller proportion of their time on crops than do
women in male-headed households (Kossoudji and Mueller 1981:

Alternative Labor Deployment

Table 5 shows that female-headed households obtain
higher proportions of income from wage laboring and
manufacturing, and smaller proportions from own agriculture
than do male-headed households. Wage laboring on other
farms must largely compare with own-farm work. Therefore,


working for other farmers would appear to be an income-
maximizing strategy. But there is also the consideration of
risk avoidence, especially if money is spent on plowing.
Inasmuch as the women work for the largest farmers, in
periods of poor rainfall they stand a better chance of
maintaining part of their usual income. Finally, working
for others for part payment in sorghum also assures a supply
of raw material for b'er brewing. Beer brewing, cooking,
and selling snack foods are undertaken in slack periods in
the agricultural timetable, and so do not directly compete
with own cultivation. The cash income they provide has many
consumption claims made on it.

That cropping is maintained at all may be due to
requirements for retaining the land, having some food supply
under sure control, limited wage-laboring opportunities, or
simply because' of the weight of tradition. Low (1982)
makes the :point that! the big difference between farm gate
and market prices !of staple foods encourages self-
provisioning farming. Migrants' wives are unlikely to find
enough cash-earning employment to purchase all their family
food requirements. But that they cut off self-provisioning
agriculture at a point where they still appear to suffer a
food deficit, and then seek cash income to purchase food,
indicates some very fine marginal calculations on the most
profitable way of allocating their labor.

Rural Income Distribution
The RIDS provides substantial evidence on widening
income and asset differentials between male- and female-
headed households as age of household head rises. The
economic status of households with older women heads
suggests that where women are left to manage farms on their
own for a long time, male migration has an adverse impact in
terms of disposable income for women and children. There is
no evidence that migrants' savings are used to support, let
alone improve, farming during their absence.

Changes in Some Determinants of Fertility

Despite male migration women in Botswana average one of
the highest fertility rates in the world. Frequent home
visits by husbands facilitate this,! but the motivation to
continue having large families is not apparent. Kossoudji
and Mueller'suggest that enlarging the network of people who
can assist economically may be a risk-reducing strategy.
Botswana has the same conjugal instability and variety of
conjugal formations seen in Lesotho.

With !no statistically significant difference in
fertility rates between women living with or without a
resident male partner, "migration rather than lowering

fertility seems to maintain it at high levels; and it
appears to do so by undermining the economic security of
women" (Kossoudji and Mueller 1981: 29).

The motivation to enlarge the network of supportive
people in Botswana, and the opinions on the importance of
educating daughters in Lesotho, need not be seen as opposing
determinants of fertility. The nature of the information
offered is different for the two countries. Investing in
both daughters' education and a supportive network of people
could be common to both countries. In Lesotho women could
still be following a strategy of "wait and see" for the
impact of daughters' education. The fiscal outcome must
depend on daughters being able to realize good incomes in


Information from this third Southern African country is
of a different nature then that of Lesotho and Botswana, as
it includes details on sources of labor. While the
countries share many similar background characteristics,
there are three relevant differences: cattle assume a
greater role in Swaziland's economy; there has been an
expansion of national wage-employment opportunities
resulting in a shift from external to internal migration of
men; and an improved maize cultivation program has
successfully raised yields. The last provides an
opportunity to focus on whether migrants' net earnings are
invested in improved agriculture, and to what extent.

The Nature of Migration and Remittances of Earnings

According to the 1976 Swazi Census external migration
(almost entirely to South Africa) amounted to about 10
percent of the adult male population. This decline from
about 18 percent in the mid 1960s is mainly due to the rise
in internal migration to sugar and citrus plantations and
cattle ranches. Internal migration is now much larger than
external migration, although it has itself been declining in
the last few years. A 1976 survey reported that 35 percent
of all households did not have a resident male over sixteen
years, and that 75 percent had at least one absent adult

Whereas both male- and female-headed households have
equal tendencies to have a migrant member, the majority of
de jure male heads who migrate remain within the country's
borders, while sons who migrate tend to be equally divided
between internal and external migration. Therefore,
husbands in nuclear households who leave wives in charge of
the farm mostly choose internal migration, while larger
households send sons ito both types of migratory employment.


Socioeconomic divisi ns among farming households largely
emerge from differences in size of cattle herds, the care of
which is men's occupation. It can therefore be supposed
that larger male-headed households are of higher (cattle)
asset status.

Internal migrants often return for plowing or to hire
tractors, and may visit their homes at weekends and
holidays. They are generally able to retain closer contact
with their families and be more aware of farming operations
than external migrants. They tend also, of course, to be
husbands rather than sons. The frequency of their home
visits allows them to participate in own agriculture at
certain times of the year. Mineworkers in South Africa
generally follow a form of oscillatory migration, returning
home for several months every few years, marking the period
between contracts. However, Vletter (1978: 33, 51) mentions
that weekly or monthly visits by migrants in South Africa
are common, and some onger visits can be timed to coincide
with a plowing period.

Replies from migrants' wives indicate that about 58
percent of husbands are absent between six and twelve
months, but that of fall absent husbands about 71 percent
make weekly, fortnightly, or monthly return visits
(Government of Swaziland Survey on "Roles, Tasks, Needs and
Skills of Rural Womeni" 1978-79).

External 'migrants remit money erratically and
accumulate savings that they bring back with them. Internal
migrants, on the other hand, remit more to their homes on a
regular basis.

Utilization of Remittances and Savings

Given differences in migratory earnings and closeness
of contact with homes of the two types of migration, it is
not surprising that external migrants accumulate their
savings for investment in cattle. In contrast, the
remittances of internal migrants tend to go more to
subsistence requirements or farm-equipment purchases
(Vletter 1978: 28, 43-i 46).

Labor Power and Tractor Use in arming

The greater frequency of husbands' returns in Swaziland
means they are able to be actively involved in home-based
farming at certain crucial points in the agricultural
timetable. Their presence helps to explain an adoption rate
for hybrid maize in demarcated Rural Development Areas
(RDAs) that exceeded all expectations. Statistics show that
national maize yields per acre more than doubled. But the
mobility of migrant husbands does n6t explain why, in the
process, acreage planted to maize was halved. For that we

need to examine sources of power year round and women's

Because of the frequency of husbands' visits the
Government Survey asked the sample of thirty women (with and
without migrant husbands) two questions concerning the
division of labor: "Who is primarily responsible for this
task?"; and, "When you are primarily responsible for this
task, from whom do you obtain assistance?" In the light of
their frequent home visits, husbands could figure in both of
the answers. The replies, shown in Tables 7 and 8, provide
illuminating information on sources of labor. The survey
does not reveal the amount of assistance, only its

The highest percentage of men being primarily
responsible for a task is in plowing (61.9 percent). This
figure has to include some husbands who are not usually
resident because the proportion of female-headed households
in the country is higher than 38 percent. "Some migrant
husbands either take leave or send part of their wages home
for hired labor during the ploughing season" (Government of
Swaziland Survey, 1978-79: 12). "Other relatives" are
negligible as persons primarily responsible for plowing.
For the women who are primarily responsible for plowing
(one-quarter of the total), only 26.3 percent receive
assistance from "relatives and friends." Another 17.5
percent hire labor. Family labor (including children) and
hired labor overwhelm kin in land preparation and plowing
when husbands appear not to be present. What is also
interesting is that the date in the two tables imply that of


Self Other
(Woman) Husband Children Relatives

Preparing land 34.7 54.6 9.2 1.5
Fertilization 39.7 47.2 10.6 2.5
Plowing 24.4 61.9 12.7 1.0
Planting 52.7 35.8 7.8 3.7
Hoeing 88.5 1.9 3.8 5.8
Weeding 91.0 0 3.8 5.2
Harvesting 92.4 1.3 0.4 5.9
Sorting & storing 88.7 6.0 1.0 4.3
Preservation of food 96.4 0 0 3.6
Sheep and goat care 47.3 21.8 27.3 0
Cattle care 46.7 20.6 32.7 0
Going to the cattle dip 34.6 30.7 33.9 0.8

Source: Government of Swaziland, 1978-79: 10.


Relatives Hired
Husband Children & Friends Labor

Preparing land 19.6 37.3 35.4 9.8
Fertilization 25.3 47.4 22.1 5.3
Plowing I 21.1 35.1 26.3 17.5
Planting 130.4 37.8 25.9 3.9
Hoeing 22.6 40.9 29.6 6.9
Weeding 12.6 40.8 41.7 4.3
Harvesting 23.4 48.1 24.7 3.9
Sorting & storing 42.9 23.5 30.3 3.4
Preservation of food 3.2 35.4 61.3 0

Source: Ibid:12.

aThese percentages do not represent share of labor
input, but' frequency of any assistance coming from
different people.

the relatives and friends who are primarily responsible for
a task or who give assistance, women are more important than
men because their support is most clearly seen for the
female-typed tasks of hoeing, weeding, harvesting, sorting
and storing, and food preservation. This support comes
mainly through women'silexchange-labor groups. Hired labor
for hoeing, weeding, and harvesting seems to be a last

Children's assistance is significant. The data in
Table 7 imply that teenage sons play an important
"responsible" role in land preparation, plowing, and
livestock care. Children generally are the most frequent
source of assistance in land preparation and plowing when
women are primarily responsible for this work. But their
support is very important in all agricultural activities,
except sorting and storing. The Survey notes that plowing,
planting, and hoeing take place in the school summer break,
and harvesting, sorting, and storing in the winter break.
Children are active in weeding as well, but the increase in
frequency of support from relatives and neighbors may
reflect a smaller amount of work actually done by children.

The use of tractors in land preparation (mainly
plowing) has increased sharply in the last decade or so.
Only one-quarter of rural homesteads used tractors in the
early 1970s. By 1980 about 40 percent were doing so, and in
some RDAs the figure was up to 85 percent (Vletter 1982:
23). As internal wage-employment] opportunities became
scarcer and more competitive in the late 1970s migrants did

not leave their jobs for the plowing period for fear of
losing them. Instead they started remitting money for
hiring tractors. Mineworkers in South Africa had until
recently been able to plan their extended home visits more
easily because of the different contract options. But now
they too are facing job competition and less variable
contracts, and have started remitting money for tractor use.
However, the use of tractors in migrant households was still
comparatively low at the time of the Survey. Only 18
percent of women who considered themselves primarily
responsible for plowing used hired tractors, a figure much
lower than the national average. The reason given for not
hiring a tractor was lack of money. For them hiring a
tractor meant hiring an operator. Some women in the sample
suggested that women should be trained to operate tractors.

The figures in Tables 7 and 8 strongly suggest women
being under a time constraint. It is well known that high-
yielding varieties of cereals require more labor application
in female-typed field tasks, and it is easy to speculate
that when facing a time constraint women choose to limit the
area they plant under hybrid maize. For female-managed
farms less land having to be plowed also means smaller cash
costs for the same amount of output if they adopt the
higher-yielding maize.

Women's Options and Decision Making

There are other plausible explanations for the large
reduction in maize acreage when hybrid maize is used, which
center on women's decision-making roles.

Letter points out that the considerable differential
between farm gate and purchase prices of maize encourages
farming households to produce as much of their own food
requirements as possible (1982: 22, 23). Yet household food
deficits have not been eliminated by the new maize. Vletter
points out that very few homesteads are inclined to sell
maize and therefore produce a surplus. But a "no surplus
maize production" strategy cannot explain the food deficits.
Moreover, other food crops are sold. Some further
explanation is required.

Sibisi writes that a marriage gives a woman her own
fields, and she "can decide for herself what to plant and
how the crops are to be used" (1980: 13). On the other
hand, the Government Survey made clear that the majority of
women encountered problems in making decisions during their
husbands' absence because "decisions on major changes in
cropping were in men's hands." Women cannot on their own
acquire more land from the chief. They also cannot raise
credit without the consent of their husbands who hold the
collateral. The Survey pointed out that women sometimes
travel long distances to obtain their husbands' signatures

on credit application papers.

Seasonal returns of migrants, the hiring of tractors,
and the widespread adoption of hybrid maize indicate
cooperation between migrants and wives over maize
cultivation. However, it is not clear from the literature
that had there been a surplus of maize for sale the women
would have been allowed to dispose of the cash income
freely. "It is by no means certain that the woman and her
children will receive a share in any increase in income
which is commensurate with the time and energy she has
devoted to increasing yields" (Ibid.: 19). If wives cannot
control the maize cash income, but are able to decide how
much of their own labor is spent on maize cultivation, then
we can see both incentive and power of women to limit the
acreage under hybrid maize. The Government Survey showed
that women seek other :sources of cash income over which they
have undisputed control. Many grew other crops for sale,
the ostensible reason being that marketing these crops was
"easier," which may be partly a euphemism for saying that
they cannot sell maize as their own property. Beer brewing
for sale was !also popular for the same reason. These
activities do not involve men's "major decisions." Because
women needed an assured supply of cash under their own
control, continuing with these activities--and even
expanding them--is likely to be a safer bet than increasing
maize output. Were both men and women in the household to
earn and consume in a more fully corporate manner, then
women's desire to follow own-account economic activities
should not be so strong, and maize acreage might not have
fallen so much as yields rose. The problem of the time
constraint for female-typed seasonal tasks in hybrid-maize
cultivation could then be isolated.

This case study offers another dimension to the
analysis of Southern African countries. Migrants keep in
closer touch with their wives; they frequently supply their
own labor at seasonal times or remit money for plowing
costs. The new technology of maize production has been used
to maintain output levels, but no more. With the
encouragement of a government agricultural program some of
the tenets of a neo-classical approach to migration were
practiced as far as maintaining production. But further
improvements in agriculture from migrant' earnings are
likely to be obstructed by economic relations within the


This migration across national borders is to the oil-
producing countries of the Gulf. Earnings are very large in
comparison with migrants' home income. There is little
scope for seasonal return visits, and absences can last for
many years interrupted by annual or biennial visits.
Yemen Arab Republic, Pakistan, Egypt, and Turkey are
reviewed here. Despite the commonality of earnings of
migrants the consequences in the four countries have been
very different. Findings from one case study cannot be
generalized to other areas in the country because there are
different agroeconomic and infrastructural conditions within
the countries.


The Nature of Migration and Remittances of Earnings

Rural areas in this country have long been a supply of
migratory labor, especially after the British established a
port colony at Aden in 1839. But part of this migration was
seasonal with men returning to the villages when their labor
was most required there. Even so, the gradual increase in
male migration from rural areas led to women taking a
prominent position in food agriculture with old men active
in land preparation.

The expansion of Middle East oil earnings gave
migration a new dimension in terms of numbers, duration of
absence, and earnings potential. The 1975 Census provides a
figure of 1.23 million people abroad out of a de jure
population of 6.47 million. Both these figures have been
considered too high, and another estimate, based on analysis
of the Census returns, gives the figure of 300,000 to
348,000 men abroad out of a population of about 5 million
(Birks et al. 1978: 2, 20, 29). By 1977 the estimate of
migrants had risen to 541,000, or approximately 30 to 36
percent of the male labor force between ten and sixty years.
Almost all migrants are men, mainly young and from rural
areas. Between one-half and two-thirds are married (Hammam
1981: 9).

The incidence of male-outmigration is greater from
poorer subsistence-farming areas than wealthier cash-crop
areas. This is well illustrated by three villages studied
by Swanson: (1977) in Village A (in a fertile valley at
7,000 feet, with a regular rainfall for grains, but with
hard frosts limiting the production of qat* between 75 and

* Qat is a leaf that is chewed. It is extremely popular,
and qat parties are held. It is a profitable cash crop
yielding five to eight times the return of grain production.

80 percent of the adult males had gone abroad; in Village B
(at 6,500 feet, with less regular rainfall, and only the
best and irrigated fields reliably producing grains, and qat
output limited by cold weather) 75 !percent of adult males
had left; and 'in Village C (in a broad valley at slightly
lower elevation, with some irrigation from wells and with
large qat production) only 48 percent of adult males had

The duration of absence is believed to have fallen
sharply in recent years. Estimates for migrants in Saudi
Arabia show that around 1970 it fluctuated between 5.6 and 8
years, but by ,1974 it was only 2.4 years. This has led
Birks and Sinclair to 'comment that increases in earnings and
reduced length of absence support the "hypothesis that
Yemeni migrants go abroad with a savings target and return
to Yemen when their target sum is reached" (1980: 58,146).
In 1977 Myntti recorded in the village she studied
remittances of $200 to $450 a month.

In Yemen migrants leave their wives in the protection
of a male kin ,or neighbor. Remittances are made almost
entirely for household maintenance and therefore are finally
received by wives through other men. Larger sums are
brought back by migrants on their visits or final return. No
major investment decisions are made in their absence. "The
major realignment of power, with regard to expenditure of
household income, has not been women gaining from men but
rather younger men (the migrants) gaining compared to the
men of the senior generation" (Myntti 1983: 16).

The Impact on Agriculture

The effect of the massive exodus of rural men in the
1970s on the role of women in agriculture is in some
dispute. Birks et al., for instance, suggest that secondary
labor reserves have emerged through fuller utilization of
male labor and more extensive participation by women,
although they caution that the existence of labor reserves
is no guarantee that they will be utilized. Myntti, on the
other hand, provides evidence thatlwomen do not move into
male economic roles to fill the vacuum created by absent
men. The reality is that the impact has varied by locality
according to environmental conditions of farming and to the
crop mix.

In predominantly; rainfed (sorghum and maize) grain
production,' women and children are active in several tasks
of soil preparation; they dominate sowing, thinning,
weeding, and harvesting; they assist men in threshing; and
they are (or were prior to the establishment of grain mills)
responsible for manual grinding. Men have usually, though
not entirely, cultivated qat. This conforms with the
different incidence of male out-migration from subsistence

and cash-crop areas.

Male labor in grain production is most important in
land preparation (maintaining terracing and plowing) and
threshing. If an adequate amount of male labor is not
retained for land preparation (a vacuum that women do not
fill) then the year round work of women in other production
tasks declines by default. Systems of exchange labor within
sex specific tasks do not resolve the problem. But if the
land is such that only a few men are needed to maintain
opportunities for women's contribution, then grain output
can be maintained. This was the situation Myntti found in a
poor highland community with infertile soil and variable
rainfall, where only drought-resistant sorghum, millet, and
cowpeas were grown (1983).

But the sharp increase in migration in the 1970s has
also drawn male labor from areas dominated by more
profitable cash crops. Qat is not everywhere cultivated
solely with male labor and, although Myntti (1978) does not
say so explicitly when referring to its increasing
popularity as a crop, it is possible that wherever qat is
substituted for some millet and sorghum, women's labor is
more involved in its cultivation than formerly, even when
migration is moderate.

Agricultural change has also been influenced by
investment in tractors, wells, and threshing and grinding
mills. This has further promoted better-endowed areas,
which is well illustrated by the varying pattern of these
investments Swanson found in his three villages. In Village
A it was not feasible for the new private wells to be used
for irrigation. In Village B broader terraces allowed for
tractor plowing, and two tractors were in operation at the
time of his study. There had also been some investment in
irrigation pumps, although this was limited by the amount of
water downstream that could be pumped up. But in Village C
the climate permitted extensive qat production. Ground
water is raised from many wells in which pumps have recently
been installed. There has been a shift from grain production
to qat production. As Swanson concludes: "It appears that
willingness to invest in agricultural mechanization is more
a function of the applicability of available technology to
specific environmental conditions than to increased wages
deriving from emigration (1977: 16-17). To the extent that
women work on qat production, their agricultural
participation might have increased as a result of this
investment in new technology.

Mechanical mills for threshing and grinding grains have
been widely introduced. Since men shared in threshing, this
is a form of substituting for male labor. But it has

derived benefits |for women, particularly in that manual
grinding of grains on stone mills is rapidly disappearing
(Dorsky 1981: 36).

How far has hired labor substituted for the absence of
male family labor? i Given the extent of male migration and
the critical need for male labor for land preparation, it is
not surprising that there has been a big increase in wages
for male agricultural ;labor since the mid 1970s. Both
Myntti (1983) and Swanson conclude that higher wages have
been a constraint on maintaining food production,
particularly in the case of drought-resistant grains.
Certainly aggregate food production has fallen. But it
could hardly be the case that a migrant member's remittances
were inadequate to cover hired labor costs. If labor has
priced itself out of the market in some grain-producing
areas, this must be because of lowi profits there. Huge
injections of purchasing power through remittances have
encouraged large [imports of food, which has made food
production using hired labor uneconomic in less well endowed
areas. For instance, between 1975 and 1977 agricultural
wages rose about 300 percent while grain prices rose only
about 30 percent (Swanson 1977: 19). The result of the new
wages: grain price ratio has been that some lower yielding
land has been left uncultivated, and higher yielding land
has been put under the more profitable crop of qat. For
this reason women in some areas can be found to be doing
more fieldwork, in other areas less.

Nonproductive Expenditure and Its Impact on Women,

Because they are left in the protection of other men,
and because there has been very little investment in food
production through which they might have emerged as
innovating farmers, rural women have not moved into new
managerial roles in farming. Therefore, any improvement in
their standing has not come from new economic roles.
Instead, those with access to remittances are acquiring
status through consumption purchases. Women are now able to
buy imported wheat, rice, tea, sugar, and tinned fruit in
local shops.! Beautiful dresses are sought, and gold jewelry
is acquired.! New ]house construction also confirms the new
high standard of living.

It is too early to draw conclusions on changes in men's
attitudes to women. Yemeni men have rights superior to
those of women in matters of inheritance, employment,
mobility, marriage choice, divorce, and child custody. As
already noted, women have not gained in standing relative to
men. Any tensions that are emerging between men and women
are probably imaskedL-or sedated-Tby some women's much
greater purchasing power. They are able to display their
higher standard of living to others through some traditional
customs. For instance, although financial resources permit

withdrawal of women from productive work and greater
expression of the religious practice of women's seclusion,
there is no reason to believe that women's regular visiting
and wide-ranging contacts with other women are being
undermined by stricter seclusion. For those women who have
benefited materially from migrants' earnings, there is
likely to have been a trade-off between less arduous work
and a greater degree of certain forms of seclusion and the
acquisition of jewelry as security. Traditional structures
determining women's place in society do not appear to have

Rural Income Distribution

The magnitude of the increase quoted above in wages (in
grain purchasing terms) for rural male labor over a two-year
period in the mid 1970s suggests that the poor who have been
able to sell their labor have improved their situation. But
this wage labor is often supplementary to some own farming
which, if self-provisioning, would merely maintain its
former real imputed value, or which, if formerly partly for
sale, would now have to compete with cheaper food imports.
Moreover, increases in the rural wage vary by local
agricultural conditions. In Swanson's Village A, for
instance (much male out-migration, tractors not possible,
and only subsistence agriculture), the wage rate rose by 100
percent between 1972 and 1975; but in Village B (much male
out-migration but tractors introduced) and Village C (less
male out-migration because of qat production increases), the
wage rate rose by only 66 percent.

The poorest women also work for wages in agriculture.
Where women in migrant households have withdrawn partly or
wholly from fieldwork, demand for female wage-labor must
have risen unless the withdrawal of female family labor was
matched by land going out of production. The shortage of
male labor has also moderated the sexual division of hired
labor; women can be seen working alongside men on irrigation
construction. Households that have gained much from male
migration are also employing women from nonmigrant
households for water collection and housework. Whether this
kind of employment arises from the preference of poor women
for nonagricultural work or from their displacement in
agriculture due to declining production, and whether it
therefore indicates a rise in female wages, must vary by
area. Once again, then, environmental conditions are the
main influence on the outcome.

A rise in the wage rate need not bring a proportional
rise in annual income because of seasonal slack periods and
a mixed work portfolio. Furthermore, the heavy food
imports, which can so effectively compete with national
grain production, hardly make grain prices an adequate cost
of living index. Birks et al. (1978: 28) cautioned that

around the mid-1970s! rises in the cost of living in the
capital, Sana, could! well have wiped out the value of
increases in per-capita GNP. Even if this claim has to be
moderated, it suggests that while many have gained, some
might have lost.

What is fairly clear from the literature is that
further socioeconomic groups are emerging. Myntti (1983:
11-12, 14-16) has summarized the impact on women by
describing their roles in three delineated new classes. In
the poorest class, which cannot rely on remittances, women
who are able to work labor in irrigation and house
construction, carry water and clean animal quarters for
better-off women, and hire themselves out as day laborers
for seasonal agricultural tasks. In the village she studied
these women--together with women too old or ill, or unable
to cultivate their own land because of a lack of labor and
working capital--were found in 15 percent of households. A
second class consists of those who have steady and sizeable
remittances and work as before in field agriculture. They
enjoy a reduced workload in threshing and grinding grains,
and cook on butane gas so that firewood collection is ended.
These women were in 75 percent of households. In the
remaining 10 percent of households the third and most
privileged class of women no longer worked in visible
production. i If land is still owned it is rented out or
worked by hired labor; while the family is urban based but
has a holiday home in the village.

For this country only one study has been used. Shaheed
(1981) investigated a sample of extended and nuclear migrant
households in two villages in the Punjab and one village in
Rawalpindi District. The latter survey included a high
proportion of households with an "internal" migrant. The
main value of this study lies in the contrast between the
isolated Punjabi villages, which suffered serious water-
logging problems in agriculture on the one hand, and the
Rawalpindi village wiih its more developed infrastructure
and market connections. The Rawalpindi village had a longer
established pattern of significant out-migration also. The
highlights of Shaheed's investigation include the way
remittances are managed, the particular farming problems of
migrants' wives in nuclear households, and the impact on
wage laborers.


The Nature of Migration and Remittances of Earnings

Since independence there has been considerable internal
migration between and within provinces. Most migrants have
gravitated to urban areas. The growth of the administration
and the military sector has offered men the incentive of

wages higher than those they could earn in farming. In the
late 1970s external migration increased sharply from both
rural and urban areas due to employment opportunities in the
Middle East. Of about 120,000 migrants leaving the country
each year almost three-quarters go to the Middle East.
Although nearly half find jobs in unskilled work, they are
not from the poorest households because the cost of travel
requires an ability to raise a sizeable sum from assets or

Both internal and external male migrants from rural
areas are mostly from large extended households; this too
suggests that they are not from the poorest households since
there is usually a degree of covariation between size of
household and assets. It also hints at migration being
influenced by how much household labor can be spared without
a fall in farming activity.

This is supported by a report of a comparative field
investigation. A study of two isolated villages in
Bhawalnagar District in South Punjab, from which there was
both internal and external migration, found that former
tenants and smallholders with water-logging problems
supplied most of the migrants. The low profitability of
farming constituted a strong push factor. But in a village
in Rawalpindi District, which was close to a city and from
which migration was mostly internal, men from larger scale
farming households tended to be the migrants. Men from
small-scale farms in both districts had more crucial
responsibilities in viable agriculture and could not be
spared. Joint households were able to supply proportionally
larger numbers of migrants than nuclear households. The
very poorest households, those with little or no land, were
usually unable to raise travel costs.

There is an enormous difference between remittances of
internal and external migrants. Shaheed (1981: 11) reports
that regular remittances from external migrants vary between
Rs. 500 and Rs. 2,000 a month. Less frequent remittances
could be any sum up to Rs. 65,000. On the other hand, men
in the Pakistani army might be remitting, at most, Rs. 400 a
month. The value of these remittances is indicated by local
landless laborers' earnings of Rs. 300 to Rs. 400 a month,
and the price of a draft animal between Rs. 3,000 and Rs.

However, there is a minimum period of three to four
months after obtaining employment before an external migrant
is able to send money home, and first remittances might be
spent on repaying loans incurred by travel costs. The
burden of this delay is heavier for those households that
find themselves short of necessary male household labor, and
the impact can be felt long after net remittances are

The great majority of rural women in Pakistan are not
accustomed to handling money or entering market situations.
They are also accustomed to dealing only with their own
household or men familiar to them in their own villages.
Yet sufficiently largellsums are being remitted on a monthly
basis by external migrants to provide opportunities for
investment and consumption expenditure on an unprecedented

Shaheed argues that since women have never been allowed
to control any property it is inconceivable to most
villagers that they should now have authority over
remittances. Her study revealed that certain exigencies and
proximity to developed market exchange (including financial
institutions) exerted a moderating influence on tradition,
at least in terms of small sums of money.

Only five women out of a sample sixty-four households
draw from the two isolated Punjabi villages had bank
accounts in their own name. What is notable is that all five
came from nuclear households. It has to be borne in mind
that opening a bank account entails journeys outside the
village--andi few men in these villages have bank accounts.
It is much easier for a male relative to make the journey to
the nearest town.

In contrast, in Vatni in Rawalpindi District, with a
bank in the village, fifteen migrants' wives had bank
accounts. Four of these women were from nuclear households.
Although the interviewed sample size was twenty-four Shaheed
states that these figures are for wives with bank accounts
out of the whole village population. But we are told that
the four represented a higher proportion of all nuclear
households than the remaining nine represented of all
nonnuclear households (Ibid.: 12). Shaheed points out that
it is not only travel and administrative obstacles that
inhibit women from opening accounts of their own, since
otherwise even more of the migrant wives in Vatni would do

It can be concluded that wives in nuclear households
are more likely than other wives to have their own bank
accounts, and therefore to be accountable exclusively or
very largely to their migrant husbands and not to male

Utilization of Remittances

Given the tradition of women not handling money, it
might be expected that without her own bank account a
migrant's wife must obtain the agreement of a male relative
to withdraw and use part of the remittances, and in the
process need to reveal her purpose and gain approval for her
proposed expenditure. Shaheed confirmed this. The male

relative was either a father-in-law or a brother-in-law.
One respondent who was obliged to obtain the agreement of
her brother-in-law stated: "Often he gives it to me if it is
a small amount, but sometimes he says he must first write
and ask my husband. How do I know what he writes since I
cannot read, and how can my husband know what I need since
he is so far away?" (Ibid.: 13). The women generally
mentioned that they were expected to keep accounts and had
to report these to a male relative. We are not told whether
women in nuclear households also had to ask a male relative
of their husband for access to remitted money. Shaheed
simply comments that this kind of dependence on a male
relative was especially true of joint households, and that
some wives of migrants living in joint households complained
that they had more say in deciding expenditure, even if
indirectly, when their husbands were present.

With bank accounts in their own name the five women
from nuclear households in the two Punjabi villages were
found to be still under constraints. First, they were not
necessarily actually operating their accounts themselves.
Second, there is also either self-censorship or external
censorship on using remittances to meet their own needs.
"Very few women have the self-assurance needed to spend a
large sum of money; and so far none have invested in
agriculture, whether in terms of land or machinery" (Ibid.:
13). The largest item of expenditure that involved some
measure of decision making by migrants' wives concerned the
purchase of draft animals or small livestock (goats and
chickens). In the case of the purchase of buffaloes and
goats this was never done without prior consultation with
male relatives, even though livestock care is women's
responsibility. Of the thirteen women who bought buffaloes,
eight lived in nuclear households, and of these five bought
buffaloes at the suggestion of their husbands, and three
said the idea was their own but that they intimated their
intention to their husbands. But all the wives living in
joint or extended households claimed the idea of purchasing
buffaloes came from a husband or male relative. The
purchase of chickens, on the other hand, was decided by
women in all types of households.

In the village of Vatni, although a much higher
proportion of migrants' wives had their own bank accounts,
the handling of remittances was not significantly different.
With the availability of goods in local shops Vatni women
did spend small sums of money on themselves. But although
the women here were more likely to take the initiative on a
desired sizeable purchase, the opinion of a male relative
was also important. Therefore, farm decisions requiring
substantial investment sums could not be taken by the women
alone however strongly they held their opinions. In
extended or joint households plowing would be arranged by
other men in the household. In nuclear households there
would appear to be close collaboration between wife and


migrant on the purchase of draft animals.

The most important use of remittances for women was to
ameliorate the effects of poverty and to provide some area
of maneuverability, including easing their greater workload.
Status of household jlis relevant for not only would the
poorest have incurred the largest debts in sending a member
to Saudi Arabia, but in nuclear households the absence of
an adult male would be felt most by migrants' wives. Where
intensity of poverty and nucleation coincide the need for
financial relief is greatest but slowest in coming.

The Impact on Agriculture

Women's customary contribution to agriculture consists
of their participation in planting, sole responsibility for
weeding, assisting as much as possible in harvesting
(including cooking extra meals and taking them to the men in
the fields), and processing crops. But much of their time
can be spent in necessary but very labor-intensive domestic
and expenditure-displacing work, such as making fuel cakes
from cow dung, and sorting, spinning, and weaving cotton.

How much field agriculture women do depends on the
socioeconomic status of their households and, of course, on
the availability of migrants' remittances. As ability to
hire landless female laborers for female-typed tasks
increases, women withdraw gradually from fieldwork. The
last tasks they give up appear to be cotton picking and
delivering food to the men. In the Punjabi villages only 13
percent of! the sample of women stated that their workload
had been reduced since remittances had been received. For
the rest it had increased or remained constant. But it
should not be assumed that male migration has the effect of
progressively easing women's agricultural workload from the
date of departure or even of the first remittance.

The absence of some male family labor always has an
immediate effect. Remaining male members might be able to
manage the plowing if they work longer hours. There is a
cost incentive to utilize household labor more intensively.
Male wages in the area rose from Rs.i 150 to Rs. 400 a month
over a period of three years in the late 1970s. At planting
and harvesting time female members will contribute more
labor to these tasks.

In nuclear households plowing presents an acute
problem. Since these Ihouseholds are among the poorer small-
scale farming households, their ability to pay for hired
labor is less than average. Whilelremittances, when they
arrive (net of debt repayment), can cover plowing costs, the
problems caused by the waiting period need to be solved. One
option that is sometimes taken is to rent out land to a
sharecropper, but this means a halving of farm income.

There is no alternative source of gainful employment for
these women unless they are prepared to join the ranks of
the very poorly paid landless women laborers. If they
manage to continue some farming until remittances can pay
for plowing assistance, they still have the incentive to
apply their labor more intensively in tasks normally shared
by men and women. Their only other recourse is to utilize
more of their children's labor. But the strong emphasis on
female virtue has the effect of keeping growing daughters
more confined to the house when fathers are absent. It is
just in the poorer nuclear households, where any process of
surplus accumulation is slowest, that a woman might have to
wait three to four years before remittances are large enough
to permit the hiring of sufficient labor to reduce her
workload in farming to the level it was before her husband's

Wives in nuclear households face the additional
prospect of having to supervise the farm. This entails
learning how to obtain seeds and fertilizers, and how to
sell the produce. Shaheed's study does explain the impact
of problems encountered on farm productivity, but it is
reasonable to assume that total production declines. For
this reason also land may be rented out to a sharecropper.

Between the better-off large household with a large
farm, and the poor, nuclear household with a small farm or
tenancy, there is a spectrum including poorer large
households and better-off nuclear households. It is clear
that a process rather than a sudden transformation is
underway. But in the early years of that process some
farming households are vulnerable to failure, notably those
in which the women were already working the hardest.

In the village of Vatni the women were found to have
stabilized their position after long-established male
migration to the army or other internal wage employment.
Remittances are small and, except for the well off, women
continue to work intensively in the fields. The better
institutional infrastructure within which women can move
with relative ease allows them to utilize remittances for
maintaining farm productivity more effectively than in the
isolated villages.

So far only problems of maintaining levels of farm
output have been discussed. There is very little investment
of remittances in land or higher agricultural productivity.
In the Punjabi villages this is largely due to the scarcity
of land for purchase and the serious water-logging problem
that requires large, community-wide investments. These
factors, together with the isolation and continued
dependence of women on male relatives, precludes the chance
of women building up the farm's viability or starting a
separate farm on the basis of remittances. The
deteriorating environmental base and the poor infrastructure

provide no springboard for women to develop their managerial
and entrepreneurial abilities. At the same time there is
deep reluctance on the part of families to sell land;
renting out is preferred. The only agricultural investments
have been in tractors and mechanical threshers, and then by
the better off. These investments may raise productivity
slightly on large farms.

Nonproductive Expenditure and Its Impact on Women

For the wives of migrants the main hope for the future
must rest in remittances taking the edge off poverty and
sharing in conspicuous consumption of the returned external

There seems little doubt that those families receiving
regular remittances from abroad enjoy a higher standard of
living than before. Shaheed (Ibid.: 18) singles out
nutrition as the most important change. More food is
complemented by better food, especially more milk and milk
products and eggs from the newly acquired livestock. This
protein production is not sold. In extended households
women's accountability to a male relative over utilization
of remittances can be expected to affect the distribution of
food in the whole household. Yet migrant wives are able to
favor their children. However, Shaheed states that while
women are eating better, men get preferential treatment in
food consumption. Women in nuclear households face the
difficulty of venturing out to make food purchases formerly
undertaken by their husbands. Similarly, they must cope
with the acquisition of medicines or medical advice from men
who are strangers.

Expenditure on clothes and children's education has
risen, with migrants' families sending daughters to school
more frequently. Improved cooking utensils are a clear
benefit to women, while the acquisition of sewing machines
points to higher labor productivity in an old expenditure-
displacing task.

But the poor opportunities for investing profitably in
agriculture have led to the large savings passing into
conspicuous consumption. Even the poorest-paid external
migrants will bring back radios, cassette players, and other
luxury household effects on visits or their final return.
The priority extensive outlay is on marriage expenses,
including dowries. Lavish marriages are a favored means of
signaling improved family status, and may be used for
attempting further upward social mobility. After marriages
house improvements or new residential construction absorb
accumulated savings. One would hope that this eases women's
domestic chores. Yet Shaheed writes that the women were
less enthusiastic than men about investing in new houses, in
spite of the fewer repairs necessary to more solid

constructions. Her explanation is that women's living
quarters are not improved, and the accumulation of furniture
does not make cleaning any easier. She does not mention
kitchens, but many observers in other countries have noted
that the kitchen is the last room in the home to receive
improvements. "On the whole, therefore, the manner in which
the bulk of the remittances is used benefits the women
mainly in terms of the increased social status she shares
with the family and only in a limited sense in tangible
terms, since her workload remains untouched while she uses
the pucca room rarely" (Ibid.: 17).

It is not made clear whether migrants' earnings are
used to withdraw female family labor from field agriculture
before or after money has been spent on marriage and
housing. An ingredient of high family social status is
usually the withdrawal of women from visible production, and
stricter seclusion. Certainly the most successful migrants
will achieve all these outlays, and to that extent women's
total workload should be reduced. But this has been
achieved only in a small minority of households. It is
understandable that women are not averse to the seclusion
involved in higher status. And status itself can provide a
compensatory sense of well-being.

There is one last point to mention, which may belong to
the future. Shaheed states that migrants from the Punjabi
villages envisage the possibility of ultimately setting up a
business in a town (Ibid.: 21). If this should happen and
the village home and land retained, the end product of male
migration could be more women and children left in the
villages as remittance female-headed households while their
men are elsewhere. Most of the women interviewed said they
did not want to move to a town. New remittances might
provide a standard of living higher than their mother and
grandmothers enjoyed, but other strains are likely to
develop from this situation.

Rural Income Distribution

Shaheed studied the inflationary effect of remittances
in Vatni where a great deal of past internal migration had
been recently augmented by migration to the Middle East.
Men in the army might be remitting at best Rs. 400 a month.
Their families were suffering "at least a relative
impoverishment" compared with those who had men in Saudi
Arabia (Ibid.: 24). Those who had men neither in the army
nor in Saudi Arabia were even more severely affected, and it
can be assumed that they suffered a clear fall in real
purchasing power. But increases in both prices of
consumption goods and wages have also influenced land tenure
by forcing some out of farming. Small-scale farmers with
little room for maneuver have been most affected by rises in
wages, and some have rented out their land. The smallest

farmers and sharecroppers have felt the cost squeeze most
sharply of all, and many men from these households have left
to find work'in the towns. Thus the wave of male migration
generates another. Iotis not difficult to imagine the new
predicament of the wives left behind.! Women do not continue
to sharecrop on their own as leases are made between men

The large wage increases apply to male labor hired for
the cropping season or for a year. Women are hired on a
casual basis. Furthermore, the numbers of women seeking
paid work have swelled (from natural increase and from small
farmers and sharecroppers going out of farming) since women,
unlike more mobile men, do not have alternative sources of
employment. At the same time the introduction of mechanized
wheat threshing has reduced their employment opportunities
in postharvest activities. As a result, their wages have
increased only marginally, far below the rate of local

Unfortunately, there are no data to gauge which groups
of households have gained or lost in absolute terms. But if
some are moving toward conspicuous consumption while others
are entering the laboring class or having to send their
menfolk to the towns, some differentials may be widening.
To suggest that women join the wage-labor force (in
preference to own farming) merely as a response to new
labor-market conditions is to ignore the value placed on
women's virtue and social status. In this cultural context
it is a measure of desperation.

Changes in Some Determinants of Fertility

It is not known what the impact on fertility rates or
active family planning has been. Long absence of husbands
presumably lowers the fertility rate, although frequent
visits by internal migrants could lead to no great
difference. The better off who are able to invest in
children's education, marriage expenses, and new housing
will no doubt become more conscious of the costs of raising
a large family. But for others whose farming livelihoods
are threatened by rising wages, the benefits of children may
still outweigh the costs. A new contribution to family
planning from an improved standing of women looks extremely


A single study has been used here. Khattab and El
Daeif's 1982 survey cannot be seen as representative of the
whole country because the village they selected was densely
populated with a well-developed market infrastructure and
rural labor market. Also, their sample was very small, only

twenty households. The value of the study is that it
demonstrates the advances that migrants' wives can make, and
their ability to establish an independent economic and
social position when their husbands remit to them
substantial sums of money. A few of the women interviewed
had emerged from very subservient positions in extended

The Nature of Migration and Remittances of Earnings

The lifting of restrictions on the emigration of labor
in the early 1970s led to very large numbers of unskilled
men seeking employment in the major oil-producing countries
of the Middle East. In 1976 there were around 400,000 to
600,000 such migrants (Birks and Sinclair 1980: 44). It can
be assumed that Egyptians' earnings and savings were
comparable to those of Yemeni and Pakistani migrants working
in these countries.

The case study described here comes from the village of
Babel Wa Kafr Hamam in Tala County in Menoufia Governate.
Large-scale emigration to Saudi Arabia began in the latter
half of the 1970s. The study consisted of twenty
households, of which fifteen had migrant husbands. The
remaining five were used as a control group. Of the
migrants' households eight were at the time nuclear
(sometimes including a widowed mother and/or unmarried
sibling), four were extended, and three were joint (married
siblings' families).

During the first year or so of absence, migrants, who
generally have no experience of banks, send remittances home
via friends who are making home visits. But later they send
them through banks located in the nearest town.

Of the fifteen migrants' wives, eleven received
remittances directly, although in four cases bank
transactions were made by a male in-law (Khattab and El
Daeif 1981: 47, 48). It is not explicitly stated, but
possibly all but one of the wives in nuclear households were
included in these eleven (one husband did not share any
decisions with his wife, and allowed his brother to manage
remittances). If this is true, then four of the seven wives
in extended/joint households were directly receiving
remittances. That the personal benefit of handling money
was anticipated by the women is strongly indicated by the
fact that eleven of the fifteen wives played a major role in
their husbands' decision to migrate, most of them initiating
the idea and contributing to the financing of the travel by
disposing of their personal property (gold, livestock, or
copper utensils) or by borrowing from their relatives
(Ibid.: 59). This display of will and involvement at the
outset plausibly contributed to the later trust shown by
husbands in their wives' management of savings.


The Utilization of Remittances

Only two of the fifteen migrants' wives in the sample
were deprived of any management of their husbands'
remittances--and property. The rest were actively involved,
and some were the ultimate decision makers. There is no
doubt that improved i nutrition was seen as the immediate
benefit. The majority of women stated that the priority
utilization 'of remittances, after purchasing better food for
their children, was the purchase of land on which to build
an independent home. For those already with such a home,
improving it with firmer materials and installing
electricity and potable water was the next priority.

The initiative 'demonstrated by the small sample of
wives in moving out of extended/joint households to
establish separate enterprises and residences is so unlike
the behavior of Yemeni and Pakistani wives that some
cautionary remarks need to be made. The village, with a
population of about 6,500 has a council, three primary
schools, shops, and a health center. Electricity and
potable water sources were introduced in 1957. There is
canal water,i but it is not stated whether this is used much
for irrigation. That it is in a densely populated locality
suggests that irrigation is applied and that agriculture is
of relatively high productivity, in turn implying a land
market and a sizeable wage-labor force. The institutional
infrastructure and local resources therefore hint at a vent
for large cash injections in productive investment. This
may not be typical of other areas of the country. Hammam,
for instance, writes that very little of migrants' earnings
is invested in agriculture in Egypt (1981: 10).

Probably the most important power of remittances was to
establish separate nuclear households. It is easy to
understand why wives would wish to move out of larger
households with members farming together. In an extended
household the young wife is likely to be a junior female
member whose labor is under the direction of the senior
female. Until she produces a son her position resembles
that of a servant, and she is given the more fatiguing tasks
in and out of the house. In a nuclear household not only
can she allocate her labor more rationally and decide on her
mode of work using hired labor (in consultation with her
husband), but her husbands' remittances can more assuredly
be utilized to benefit only her and her children. The
incentive of a migrant husband to.see his wife move out of
an extended household to form one of her own is not
altogether clear. With the desire to protect women's
virtue, and past negligible experience of women in dealing
with agricultural institutions and agents, it might be
supposed that absent men would feel happier if their wives
remained within their own (the men's) families. At the same
time the opportunity afforded by large savings from migrant
earnings to buy land, or to buy out of joint ownership of

family property and to set up as independent farmers, is
very tempting because it means that the migrant does not
have to contribute to a wider family business. At any rate,
Khattab and El Daeif found in the village they studied a
clear trend toward the emergence of migrant nuclear families
from extended/joint families.

The dissolution appears to develop in stages (Ibid.:
24-27). The wife's financial status improves if she
receives the remittances directly. Giving her children
better food marks the first stage of a separation of
financial arrangements. Later she moves out with her
children and takes up a new separate residence. Land is
bought and a house built.

The three migrant joint households of the sample were
all observed in the process of change. They had already
dissolved joint financial arrangements. The actual move out
may be awaiting the result of negotiations over selling the
migrant's share in the property or buying out the other
brother, or a separate house may still be under

How much land is purchased is not made clear. In some
instances it seemed that a house plot was acquired. Further
land acquisition depends on whether the wife will devote
time to active farming or use her skills in alternative
gainful employment. Where the couple decide that farming
should start immediately, labor is hired.

Other earmarking of remittances might include a tractor
or truck, livestock, electrical appliances for the home, and
trousseaus for marriageable daughters. The information
indicates that apart from acquiring a house there is a
marked preference for investing first in producer labor-
displacing goods, later in consumer goods. The
concentration on producer goods demonstrates a husband's
wish to increase the economically productive resources of
his wife.

The Impact on Agriculture

It was probably too soon at the time the case study was
undertaken to have evaluated the impact on farm output.
Certainly no information on changes in yields and output is

Apart from women performing domestic chores, they
undertake economic production including livestock care,
assistance in fieldwork (such as planting, weeding,
harvesting, picking cotton, and transporting produce to the
house) during periods of heavy labor requirements, and
bringing lunch to those in the field. The overall sexual
division of labor and the workload varies over the life

cycle of women and according to household composition.
Junior women in a large household and women in nuclear
households work hardest.

In most extended/joint households other male members
were able to replace the migrant's labor input. However,
this created problems of distributionlof the proceeds. In
some instances the migrant's wife and children continued
with their usual agricultural work, but were also compelled
to increase their labor input to substitute for the migrant.
Understandably, migrants' wives may wish to see an end to
shared farming. Some were behind a division of the shared
property. Remittances mean that agriculture is no longer
the primary source of income for family maintenance, and
this provides wives with the opportunity to enter some other
gainful employment, which gives them independence from their
husbands' families. One wife persuaded her husband to sell
his share of family land to his uncles to induce the final
economic break (Ibid.: 41).

Investing further in a new separate or larger farm
enterprise while husbands are away must be influenced by the
decision of migrants to take up farming again on their final
return. The choice of gainful employment made by some wives
suggests that this decision is not taken in the early years
of a migrant's absence. During Khattab and El Daeif's field
investigation of the !fifteen migrants' wives, five were
working as seamstresses, three as agricultural laborers,
five were in the family farming business supervising land
and livestock, and two confined themselves to domestic work
(Ibid.: 16). Unfortunately, we are not told the type of
household they came from. All the seamstresses had acquired
the skill before the departure of their husbands, but most
started this self-employment only after their husbands had
left. The 'attractions of this work include the small
capital outlay and the ability to work at home. Even when
they are supervising their own farms and hiring labor, wives
may undertake sewing as a sideline. It is not possible to
judge whether the returns to sewing are greater than to
working as much as possible in own agriculture to save on
the wages bill. Moreover, such a calculation may not be the
primary motive for the actual deployment of labor. What
pervades the case study are stories of women endeavoring to
switch from self-provisioning agriculture (but not
necessarily livestock) raising) to monetizing their labor
power, and of recognition by both migrant and wife that she
is using her newj access to resources to establish an
economically viable separate household. Even the
acquisition of a tractor was not for purposes of overcoming
plowing problems, but for renting out. This wife used the
income from the tractor to pay off loans incurred in
building a new home.

There is no indication that labor input per unit of
land declined or that some land went out of production.

Extra family effort or hired labor filled any gaps left by
departing migrants. Separate or joint farming was continued
under altered work organization or a share in a combined
property was sold to other family members.

New Responsibilities and Status for Migrants' Wives

Receipt of remittances does not mean powers of managing
them. Neither does nonreceipt signify no powers. Eleven
wives claimed they received remittances directly. Of the
fifteen wives in the sample, ten admitted to making
decisions on their management themselves. Presumably this
refers to regular remittances for maintenance and running
costs. Two stated that they made joint decisions with their
husbands, two left the decisions to their husbands (one of
whom was in a nuclear household but had a dominating
husband), and in one case the father-in-law made decisions
(Ibid.: 48). Bearing in mind that these remittances must
have been substantial compared with local income, this
figure constitutes an impressive record of confidence in
wives unused to handling much money.

Migrants' wives in nuclear households face quite new
conditions and experiences of work. They appear to take an
active part in the building of the house as well as
supervising construction. They assume leadership of a
household and manage any land and livestock. This does not
necessarily mean any lessening of their workload; rather the
contrary, to which the burden of responsibility must be
added. The women in this Egyptian study have more resources
under their control, their household economies are evolving
and expanding, and the future holds the promise of surplus
accumulation. Life is still hard, but the signs are
promising and encourage response.

Roles and status changed for migrants' wives in other
households even before making the break from extended or
joint households. However, there can be serious strain
during the process of disengagement. It does not take much
imagination to appreciate the jealousy and resentment of
others who see great benefits from the relatively huge
remittances coming the way of a junior female member. It is
a difficult time for her, and a test of her determination
and newly felt power. Her husband's farm labor contribution
will be missing, and there will be (mostly successful)
attempts to reduce her share of the farm's proceeds for this
reason and because of her new income. The most difficult
relationship is usually with her mother-in-law because
financial disengagement carries with it the potential for
disengagement from a work role subordinate to the mother-in-
law. From dependence on the in-laws for subsistence, a
migrant's wife moves to an egalitarian position, sometimes
even ending with payment of an allowance to her mother-in-
law. Or the mother-in-law might appear submissive, pleading

poverty in order to obtain financial support from her
daughter-in-law. One! wife said that her mother-in-law had
refused to lend her son money to travel, claiming that they
were two separate families, but later showed interest in her
son's affairs on the grounds that they were all one family.
While residence with husband's kin provides shelter and some
economic base, family relationships are also an early
training ground for the wife's assertiveness and sense of

Raising children on their own islnot seen by the wives
as an extra burden jin terms of Ieconomic maintenance.
Instead, standing in for fathers means the extra
responsibility of disciplining children and supervising
homework, organizing tuition, and buying books. But
domestic chores can be alleviated by remittances. Some
wives had used remittances to install electricity and were
applying for potable water. Greater access to cash income
has meant that others have chosen to purchase ready-made
foods such as bread, cheese, and butter.

The acquisition of items such as washing machines and
refrigerators, purchased locally during the migrant's first
return visit, indicates something about the nature of the
new partnership being forged between migrant husband and
newly independent wife. A touching story was told by a
vacationing husband who said that before becoming a migrant
he had never appreciated the effortsjof his wife in washing
clothes, and that he was going to buy her a washing machine.
Another husband acquired a cassette recorder with his first
wages and sent itltolhis wife to facilitate communication
between them. The'village midwife told the story of a wife
who asked her family to record the birth of her baby on
cassette so! that her husband could hear his child's first

That the wives themselves understand the personal value
of the changes brought about by their husbands' migration
was made very clear in their comments. Most of them
mentioned that their self-improvement was the basic reason
for wanting their husbands to seek work in another country.
They are aware of their new self-esteem. Having to deal
with people and institutions beyond the family has also
given them a social personality. When the government
deprived migrant families of food'ration cards on the
grounds that they enjoyed higher incomes, the women had no
difficulty in obtaining alternative supplies of food, as
suppliers respected their new purchasing power and satisfied
their needs promptly.

Changes in husbands' attitudes are probably what most
benefit women because they determine how far the wives have
managerial control of productive assets. In all that has
been described above, one significant factor emerges:

Migrant husbands are dependent on their wives' services and
drive. It is premature to draw conclusions on the
permanency of this after the husbands' final return. But
one vacationing husband explained his new appreciation of
his wife thus: "Not only did she work to earn money to
provide for her family, but she knew how to manage the
irregular remittances sent home" (Ibid.: 28). All expenses
during his absence had been recorded (as by the other wives
too), and he was very proud that his wife had paid off all
their debts and had started saving. Some wives enjoyed the
new experience of hearing their husbands speak highly of
them in front of their families.

The contrast of these gains with the stationary
situation of nonmigrant wives or migrants' wives who have
not changed their roles is all too apparent. The latter
continue with their usual chores, and enjoy no improvement
in authority.

Rural Income Distribution

The case study did not mention changes in rural income
distribution. But it is plausible to infer that male
agricultural wages rose as migrants' wives hired more labor
for agriculture and house construction. The developed
market infrastructure might well have moderated inflationary
consequences, but it is difficult to see that any increase
in demand for female wage labor was sufficient for this wage
rate to keep up with inflation. Therefore, landless
households probably made only modest gains. What is clear
is that an income gap must have developed between migrant
and nonmigrant small-scale farming households.

Changes in Some Determinants of Fertility

Fertility can be expected to fall with long separation
of young couples. But much of the above intimates changes
in some of the commonly regarded determinants of fertility.
If migrants' wives are following a strategy of monetizing
their productive work, or combining managing the farm with
sewing, this withdrawal of their labor from family farming
is most likely to have been preceded by the withdrawal of
their children's labor. Khattab and El Daeif point out that
during their visits to the village they did not hear of a
single case of migrant families withdrawing children from
school for wage employment (Ibid.: 44). The desire for
children's (including daughters') education was confirmed by
the enrollment of children in schools, and by the care
mothers took to ensure that homework was done. The cost of
education can also be hypothesized to influence the total
number of children.

Even were the migrant to take up own farming again on
his final return, it is doubtful whether he would use his
children's labor except during school vacations or at
harvest time since that would defeat one of the main
purposes of his migration. If his wife continues with her
separate gainful employment, future farming would in all
likelihood be undertaken with the assistance of hired labor.


The study of this country concentrates on the growing
authority over farming matters of migrants' wives in nuclear
households, land the emergence of all wives with much new
purchasing power in the market. The emphasis is therefore
mainly on the social effects of male out-migration on women.
But some speculative comments can be made on what women's
new status could mean for their role in farming. Abadan-
Unat uses information from a 1975 survey in Bogazliyan,
Yozgat Turkey, covering a very large sample.

The Nature of Migration and Remittances of Earnings

As in many other! countries, in rural Turkey extended
households are i usually those with larger farm holdings,
while nuclear households are disproportionately represented
among the landless. A study in the villages around
Bogazliyan in Yozgat Turkey showed that 32 percent of a
sample of returned migrants had undertaken migratory
employment because they owned no land (Abadan-Unat 1977:
47). If they had been tenant farmers previously they could
not be considered among the absolute poor. But the figure
does imply that a substantial proportion of migrants came
from households that were nuclear at the time of their
departure. iAt the time of the survey 56 percent of the
members of migrants' families remaining in the villages were
in nuclear households,' suggesting that nucleation initially
extended far into small-scale farm ownership, and/or that
nucleation spread as a consequence of men's migration.

Of a sample of 113 returned migrants who had stayed
abroad at least seven years, 84 had made at least five home
visits. This i indicates close involvement with their

Remittances are sent on a regular basis to the
effective remaining head of household, and are obtainable at
banks or post offices. These sums are larger than what is
required for daily essentials since their cumulations are
drawn on for the purchase of consumer durables and new house

The Utilization of Remittances

The enhanced authority of women and the size of regular
remittances imply that daily needs are met more easily and
the ingredients of a higher standard of living are obtained.
Conspicuous consumption in the form of new furniture and
electrical goods (even before electricity is available) is
also undertaken since this signals upward social mobility.
Greater purchasing power is used to reduce time-consuming
economic activities, such as looking after sheep.

The Impact on Agriculture

There is very little information on how women on their
own maintain or invest in farming activity. Nuclear
households are likely to have small farms, if they own any
land at all. According to Abadan-Unat, "Whenever there is
some land left, it is the women who decide what work is
going to be done, when and by whom" (Ibid.: 49). It is
possible to speculate further. If a woman chooses not to
continue caring for sheep personally, she is demonstrating
an ability to sell the sheep or hire labor for the task.
Either way she clearly has some authority over farming
matters. The degree of her control over remittances
suggests she has financial capacity at least to maintain
farm output in her husband's absence. What significant
final decisions are made on maintaining or improving farming
are therefore likely to be made through migrant and wife
collaboration after they have pooled their profit
calculations and consumption/work preferences. This
involves further information on the basic profitability of
farming and on personal values placed on leisure and status.
But it can reasonably be concluded that women's new
authority and resources are adequate at least to maintain
output, if this is desired.

Women's Growing Independence and Authority

From a traditional position in which their husbands had
final authority in family matters, a large proportion of
migrants' wives have assumed new decision-making roles. In
nuclear households women receive remittances and manage
regular expenditure. As a result, they are dealing for the
first time with financial institutions and retailers, and
visiting government agencies in the towns. Eighty-six
percent of wives in nuclear households do the shopping
against only 14 percent living with husbands' families;
while the proportions of those who make financial decisions
are 68 percent and 28 percent respectively (Ibid.: 49, 50).
Shopkeepers now readily serve women, where before they might
not allow them in their shops. We are not told how
decisions are made on more expensive items, but with
migrants' frequent return visits we might assume

collaboration between husband and wife. Since these visits
are not lengthy wives must frequently be left to implement
expenditure 'decisions.! Those in nuclear households now
enjoy authority over matters concerning children, and a
stronger desire for daughters' education is in evidence.

Male migration is having the effect of increasing the
proportion of nuclear households. First, wives (in original
nuclear households) insist that they do not join with other
family members when their husbands leave. Second, migrant
husbands are establishing separate households for their
wives. A wife's motivation to be free of her in-laws is
obvious. One could speculate that migrants are motivated by
a removal from extended family patriarchy and the
possibility of rapid accumulation of private assets.

Rural Income Distribution

This also has to be speculative. Demand for female (as
well as male) wage labor should rise if migrants' wives wish
to reduce their own labor input. Inflation of food prices
is probable even if more food is supplied from new sources
because more food is bought by migrants' families. Any new
income differentials ;should be dominated by the division
between migrants' and nonmigrants' families because many
landless families have sent men to migratory employment.
Abadan-Unat Ipoints out that in villages and small towns the
new social Idivision can be seen by the larger houses of
families with a member employed abroad.

Changes in Some Determinants of Fertility

The Bogazliyan study revealed that the birth rate has
risen. A midwife reported that migrants' wives were not
interested in birth control, the main reason being that
there was enough money to look after children properly.
Migrants' visits from Europe in July lead to a peak of
births nine months later (Ibid.: 48). But there are signs
among younger women, more recently married, that an interest
in birth control is emerging. It is likely that young
migrants in Europe are picking up ideas on birth control
faster than; oldermigrants. In the longer term the new
interest in: daughters' education may have an impact on



Difficulties and Incentives in Maintaining Farm Output

The case studies show that women on their own in
nuclear households face serious problems in maintaining farm
output. Migrants' wives in extended or joint households do
not share these problems as they are integrated in a larger
farming enterprise, although, as in Egypt, they may be asked
to contribute more labor and have to negotiate shares of the
produce. But planners should be aware that despite
difficulties of own farming, women have good reasons to
disengage from larger households and that male migration
encourages the nucleation of farming households.

In southern Africa the very small regular remittances
are either not available or are inadequate to cover the
extra costs (see Botswana) of farming of female-headed
households. Assistance in plowing from male kin is not in
much evidence; data from Swaziland suggest that it is
female relatives and friends who are more important. Females
also help with female-typed or shared tasks when children's
assistance reaches its school vacation limits. There are
problems of timely hiring and supervision of male wage
labor. Female heads of households plow less often than male
heads of households (see Botswana). Leaving part of the
holding uncultivated but maintaining comparable yields per
acre with male-headed households fits a strategy of
minimizing costs of hired plowing power. How much land is
left uncultivated depends on net marginal returns to farming
compared with the value placed on alternative sources of
income. The case studies show that the (marginal)
opportunity of farming for migrants' wives is not zero.
Also, the fact that women in female-headed households spend
a smaller proportion of their time on own farming than do
women in male-headed households (see Botswana) means that
the suspected feminization of agriculture because of male
migration needs to be more closely defined.

Declining farm assets and income over the years (see
Botswana) are signs of accumulating struggle with farm
problems. Complaints from migrants' wives that stress
increases in the long term and that managing the farm is the
biggest problem (see Lesotho) corroborate this.

Women certainly have incentives to provision the family
as much as possible from own farming, but the need for cash
income, and a risk-avoidance strategy and the consequential
competing demands on their time, can lead them to stop own
farming short of food self-sufficiency. Doubt over their
control of income from sales of surplus maize would

certainly inhibit going beyond self-sufficiency even were
this to generate more income per se than alternative work.
That women rationalize their position by leaving some land
uncultivated, choosing crop mixes and alternative work is
evidence that 'they 'are able to make some important

Migrants' wives in the Near East have very different
experiences which are determined by the degree to which
women are protected, agroeconomib conditions, infra-
structural development, and alternative sources of
employment. In Egypt, at least, there is evidence that
women desire to move out of extended or joint households,
but this might be influenced by the local opportunities to
"go it alone." Theimagnitude of regular remittances is
quite sufficient for hiring draft power and other labor
assistance, but there is a long delay in the arrival of such
remittances net of debt repayments (see Pakistan). This
delay can be several years for the poorest, usually nuclear,
households. There ,are problems of negotiating and
supervising male labor in this region also. The result of
these difficulties sometimes leads to land being rented out,
and wives may enter wage employment if only temporarily. In
the Egyptian case study the wives had, or developed, a great
deal of self-confidence. Problems were experienced with the
farmers' cooperative but women do not appear to have been
intimidated by hired laborers. Not all of the small sample
of wives in nuclear households undertook own farming (most
likely to do so, although it was not stated, were those who
had disengaged from larger households). Establishment of a
new separate farm must depend on the migrant's future
intentions. In the meantime his wife finds other economic
activities, which in some cases are combined with
supervising farming.

Problems of a shortage of male household labor in
Yemeni food agriculture are compounded by a sharp rise in
male wages and cheap food imports, which together lower the
profitability of food production. The very large
remittances,1 especially in poor grain-producing areas from
which migration is greatest, also obviate any struggle with
low-yielding and risky agriculture. The two regions
illustrate combinations of factors that refute simple neo-
classical assumptions that male migration assists the
maintenance of farm output.

Lack of Investment in Higher Productivity Farning

It has to be recognized that many areas of male out-
migration are not amenable to (present) higher-yielding
seeds and their accompanying cash inputs. Furthermore, as
long as migrants' wives are disregarded by farmers'
cooperatives and extension services, any promotion of
higher-productivity agriculture while men are absent runs

into service access problems.

The one example we have of an agricultural project that
was successful in terms of raising output per unit of land
cultivated is hybrid maize in Swaziland. The frequent home
visits of internal (and some external) migrants, and later,
when job competition inhibited these visits, the fact that
migrants remitted funds to hire tractors, suggest that when
migrants keep in close contact with their families a program
to raise productivity can be successful. But in Southern
Africa, at least, women's time constraint and their lack of
clear rights to dispose of income from surplus staple
production limit acreage planted and, therefore, total
output. There is a need to undertake further research
separating the issues of physical constraints and service
access problems on the one hand, and women's incentives (and
effective powers of veto) to implement the goals of an
agricultural project on the other.

Small livestock appear to be a part of farming favored
by women. They are normally women's responsibility and they
have the advantage of not being dependent on rainfall or
good land. In Botswana holdings of small livestock were
maintained over the years while cattle holdings and total
income declined. In Pakistan women purchased small
livestock freely for self-provisioning protein food
production. Women's interest in this part of farming has
been a neglected area of agricultural planning.

Nearly all of the areas from which these case studies
came were of low-yielding, high-risk agriculture. Only the
real possibility of making agriculture a high-yielding,
high-status profession can counter the tendency toward
spending savings on conspicuous consumption and
nonagricultural livelihoods. In the Near East savings are
large enough to overcome some of these obstacles,
particularly if resources are pooled for environmental
improvement. The migrants' reluctance to invest further in
farming, and the predeliction for expenditures on houses and
consumer durables attest to farmers' views of the likelihood
of this happening.

Women's Limited Farming Resources and Authority, and Their Other Options

The weakness of remittance support for farming (Lesotho
and Botswana), the decline in farm assets and income
(Botswana), the lower proportion of time spent on own
farming by women in female-headed households and their
smaller percentage of income derived from own farming lead
to the conclusion that migrants' wives are partially
alienated from agriculture. Both lack of resources and
authority (with their husbands, hired labor, and farmers'
services) contribute to this. But women's strategy of

reducing own agriculture and developing a wider income
portfolio carries the social cost of reduced total farm
output. Because migrants' savings are not corporately
utilized, that is, not made available to meet the farming
needs (at least) of their wives, migrants' wives are left
underresourced for effective farming!. The private gain to
migrants has social costs via the accommodations made by
their wives. I

Women in the Egyptian case study appeared to opt partly
or wholly for nonagricultural occupations for different
reasons. Where they did farm they mustered the authority to
surmount any difficulties in hiring and supervising labor.
But taking up sewing or buying a tractor to rent out could
be seen as easier options. If planners see social costs in
this private strategy the solution lies in offering the
couple greater incentives to farm, and adequate servicing
support to the wife.

Where women do take up off-own-farm occupations the
costs in production terms may be less than where their
status and autonomy in the household and community are so
low that they play out their role by sharing in unproductive
conspicuous |consumption supported by remittances. This
could be seen as one of the differences between the Egyptian
and Yemeni case studies. But culture is not independent of
the economic environment. For example, had local economic
activities been available to women in Yemen, conspicuous
consumption might not have been so heavily resorted to and
women not confined to the house.

The different accommodations migrant wives have made
lead to corresponding economic and social transformations of
their roles. Those who have changed their work and income
portfolios while their husbands are away are offering clear
signals to planners as to openings for women in the market
and social structures.

Family Welfare

The evidence ofj the case studies from the Near East
points to improved children's nutrition being a priority
concern of wives. In the Egyptian study the separate
feeding of one's children in a large household, with all the
likely intrahousehold strains that would ensue, is a clear
illustration of this. iInvestment in protein-producing small
livestock in the Punjab is another form of expression. In
Southern Africa working for less risky cash income in the
absence of adequate regular remittances can be seen as a
demonstration of the importance of food security.

The importance placed on investment in children's
education suggests that financing better education is one
of the main!purposes of male migration. In Lesotho it was

overwhelmingly the priority claim of both migrant and wife
on the more substantial savings he brought back. Illiterate
Egyptian women did what they could to supervise their
children's homework. One significant aspect is the gain in
daughters' schooling. There were no data on this, but
stated preferences and comments suggest a shifting view to
valuing the education of girls as well as boys.

Other aspects of family welfare can depend on the
overall health and contentment of mothers. Stresses--
including greater workloads--on women can increase in the
long term (as in Lesotho) or be greater in the early years
(as in Pakistan and Egypt). There are reports of conjugal
instability from Southern Africa.

Very large remittances in the Near East allow for
better housing, household electrical appliances, and further
consumption such as splendid marriages, which are all
practical or pleasing gains to women. Such gains sometimes
accompany stricter forms of female seclusion.

Nonmigrant Households and Wage Laboring Income

Nonmigrant households are generally found among the
poorest. There is also some evidence that where there is
both internal and external migration, some husbands in
nuclear households in the poorer small-scale farm category
will choose internal migration to keep in closer contact
with the families (see Swaziland) or because they cannot
afford large travel costs (see Pakistan). In Pakistan
internal migrant households, like nonmigrant households,
face local inflation arising from the demands for goods
generated by the large remittances of external migrants.

Male wages certainly rise, and probably in real terms.
But female wages are reduced by certain forms of
mechanization, such as mechanical mills, and by some
migrants' wives taking off-own-farm wage employment either
to maximize their income or reduce risk (Southern Africa) or
because they have rented out their land (Pakistan). Changes
in demand and supply of female wage labor will determine how
far the female wage rate is able to keep up with inflation,
but it is doubtful whether real female wages will rise.

Therefore, the impact of male out-migration on the
income of the poorest households and rural income
distribution cannot be viewed only in terms of the greater
demand for male wage labor.

Changes in Some Determinants of Fertility

There are no direct data on changes in fertility, but
we can surmise from the impact on certain supposed

determinants of fertility that in the long term fertility
will decline!in migrant families.

More education, especially of daughters, and less
dependence on children's labor are likely to be the most
important determinants of a fertility decline. Upward
social mobility and greater marriage expenditure should also
inhibit large families. But against this is resistance by
men to altering social;relations between the sexes. Female
virtue is still very important in the Near East, and the
circumscribing of women's control of remittances in a long-
established outmigration village such as Vatni, in Pakistan,
suggests that other means by which women can control their
lives might be resisted by men. What daughters are allowed
to do with their education must be a crucial factor in the
long term. On the positive side, a revolt by young migrants
against extended family patriarchy could have derived
effects on fertility through new concepts of the family.
Nucleation of households should promote surplus
accumulation, the hiring of farm labor, and new vistas for

In southern Africa women clearly depend on their
children's labor, and this is likely to continue in the
foreseeable future. The use of migrants' savings for
children's education, and women's preferences for investing
in daughters' education (see Lesotho)Isignify a hope, if not
a real expectation, that both sons and daughters will enjoy
better employment opportunities than their parents.
However, migrants' wives may be following a wait-and-see
strategy, particularly concerning their daughters' futures.
In the meantime migrants' wives have good reasons to have
children to widen support networks. There is no evidence of
a fertility decline so far.


This study indicates that there is no reason to believe
that agricultural output of extended or joint households is
threatened by male out-migration. While migrants' wives in
extended households sometimes bear the private costs of
greater workload and subservience to family elders,
amelioration of these difficulties is not a realistic area
of policy intervention Therefore, this discussion is
confined to policies directed at ameliorating the conditions
of nuclear farming l! households and poorer nonmigrant

Maintenance of Former Agriculture in the Absence of the Male Head of Household

There are three main interrelated aspects to be
considered: (1) the size of regular remittances in relation
to extra costs of farming; (2) authority of women
farmers vis-a-vis farmers' cooperatives, extension services,
input suppliers, and marketing outlets; and (3) farm
management and net total income factors that encourage women
to hold a different cropping, livestock raising, and off-
own-farm portfolio.

Whenever regular remittances (net of the most immediate
claims made on them) are not adequate to cover extra costs
of farming, the third aspect--farm management and net total
income--assumes greater importance. Assuming that policy
makers are not in the business of persuading migrants to
remit more of their savings to their wives, they must seek
to understand why women farmers choose the livelihood
portfolio they do, and determine whether government's
interest in maintaining certain crop output levels is strong
enough to offer women inducements to alter that portfolio.
Sometimes a government's goals will not be specific to
certain crops, but will focus on concern about land left
unutilized or plowed less frequently.

If grain production, which uses land more extensively
than production of other food crops, is uppermost in policy
makers' minds because of increasing foodimports, the nettle
of plowing costs has to be grasped. Subsidized tractor
service stations, with disciplined and respectful male labor
or trained womanpower, might bear a positive social
benefit:cost ratio if they shift women farmers' incentives
back to more grain production. Substitute tractoring might
also have the effect of interesting migrants in remitting
more of their earnings. The interest of Swazi migrants in
hybrid maize has a parallel here.

Aside from tractors or draft power, there are other
cultivation tasks in extensive grain production, usually
shared by men and women, which place a greater workload on
women when men are absent. Women's exchange-labor networks
could be augmented by drawing on women from land-poor
households if food-for-work programs incorporated them.
Instead of working for large-scale (usually extended
household) farms women would be directed to small-scale
female-managed farms. Usually food-for-work schemes are
used to increase the production-bearing capacity of the
rural economy. This proposal aims at applying more labor to
existing capacity that might otherwise be underutilized.
One advantage is that there would be less crowding of women
on larger farms in search of money or food. Another
advantage is that a fall in the supply of female wage labor
on larger farms should raise female wages.

However, when the calculations of social costs and
gains of possible intervention on grain production have been
made it might be found to be more realistic to concentrate
on other aspects of female-managed farming. What are the
other food crops thatjiwomen favor for sale or making into
cooked foods? Can women's interest in small livestock be
capitalized 'on? Why not work with their chosen crop and
livestock mix and make it more profitable, at least for the
duration of their husbands' absence?

All these proposals assume that policy will seek to
improve the authority of women farmers in the eyes of
servicing and marketing organizations. There have been many
calls to award women equal access with men to credit,
extension assistance, 'and marketing advice. We underscore
the need here. Implementation of this access for migrants'
wives becomes an imperative if social costs are not to

In the Near East remittances, after a long delay, are
quite sufficient to cover extra farm costs. The challenges
to policy in this region are different from those in
Southern Africa. They are basically threefold. First,
assistance is needed! in the first 'few years of migrant
absence to avoid a severe workload for women and to dissuade
them from renting out land. The early years of debt
repayments and small net remittances could be ameliorated by
a collective credit fund, rotating over successive groups of
migrant families. ; The later very large remittances of
successful migrants could bear both credit and interest
repayments (which could include the writing off of bad debts
of unsuccessful migrants). Second, remittances can
eventually be so large as to make farming unnecessary.
Women's bad experiences in the first years after husbands'
migration could encourage the abandonment of farming. On the
other hand the experience of receiving the attentions of a
credit fund and of managing borrowed working capital could
have the effect! of instilling lin women a new and
professional1 interest in farm maintenance. It could also
encourage the spread of bank accounts in the names of
migrants' wives, which could lead to wives gradually being
less accountable to 'male kin while still seeking their

Third, many women in the Near East are not accustomed
to the minimum amount of social intercourse required to
manage a farm, and ;farmers' cooperatives or extension
services are not yet directed to offer them assistance. In
this case it is not only a matter of extending specific
services to women, but of encouraging women to move in a
sphere beyond the home. It cannot be assumed that rural and
"protected" women will not respond to new opportunities.
Women in the densely populated villages in Egypt, presented
with a developed infrastructure, saw green lights beckoning
when large remittances flowed to them even while they were

still in a subservient role in an extended household. They
had opportunities to express their entrepreneurial talents,
and apart from building houses, conspicuous consumption did
not last long. In the other Near East countries reviewed
here women had no rent for their talents, and conspicuous
consumption continued, or the security of gold jewelry was
sought. In thinly populated and isolated areas in
particular, cooperative and parastatal officials have to
substitute for free-market agents. An example of this might
be the hiring and supervising of male labor. The presence
of a cooperative official or extension worker on the first
occasion could prove invaluable. Where there are no
nonagricultural employment opportunities, if women are to be
agents of deflecting migrants' savings from consumption,
then agriculture has to be made more professional before the
men finally return.

Inducements to Productive Investment Before and After the Permanent Return
of Migrants

Maintenance of agriculture is a precondition to
investment in improved agriculture. For this reason the
preceding section can be seen as laying the foundations for
further productive investment. It has often been said that
migrants' earnings are not invested in higher-productivity
agriculture in the absence of an agricultural promotion
program. The evidence supports this. But the adoption of
hybrid maize in Swaziland shows what can be done while
husbands are still in migratory employment. When husbands
are farther afield then they were in this country, and
communication of couples is weaker, it is desirable to make
it possible for migrants and wives to discuss farm matters
together with extension officers before the migrant leaves.
In areas of mass male migration adding this service to what
exists makes economic sense. A husband will know before he
leaves what input and extension assistance will be available
to his wife and what her opinion of it is. He will also be
made aware of the seriousness of government approaches to
female-managed farms. All this should influence the
migrants' strategy on earmarking remittances. Another role
of extension officers could be to help the couple open a
bank account in the wife's name.

A followup might also be considered that takes the
extension service to migrants' work locations. This would
also reach migrants who missed a predeparture briefing.
What is proposed is that extension officers should travel,
perhaps once a year, to areas of migrant concentration to
explain existing and revised facilities offered to their
wives, and to offer the opinions of wives in general. This
is not intended to take the place of any private
communication between migrants and wives; on the contrary,
it should enhance it. What is envisaged is a faster and
more accurate learning process for migrants so that they do

not have to wait for a later home visit to appreciate what
their wives have achieved or are contemplating. Under this
scheme a visiting migrant would arrive home with a long list
of family subjects to discuss and questions to ask, in all
probability to be met by his wife with her own list.

Most of the case studies reflected the widespread
belief that it' is high risk agricultural areas in which
individual private investment cannot hope to yield a
reasonable Ireturn that supply large numbers of male
migrants. If agriculture is ever to become attractive as a
private investment proposition it will first require social
investment--organized above the household level.
Consolidation of fragmented holdings, small-scale irrigation
construction, land drainage, and reforestation can be
undertaken only on a social scale. Some countries, such as
Egypt, are attempting to divert migrants' savings from
conspicuous consumption by offering special issues of bonds
(Birks and Sinclair 1980: 104). But the money raised is
intended for national development unrelated to the area from
which the migrants came, and perhaps not even to agriculture
in general. A more pertinent policy intervention is to
devise schemes whereby migrants from the same area
collectively invest in the local environment to upgrade the
quality of all their holdings. This would of course benefit
nonmigrant households that could not afford to contribute as
well as noncontributing migrant households. Antagonism to
the idea of benefiting noncontributors could be overcome by
an arrangement whereby the government contributes X dollars
for every Y dollars the migrants make available. More
detailed incentives could be offered to induce agreement.
For instance, government subsidies for on-farm terracing or
ditch digging could be offered to contributors. In this way
collectivized contributions and private gains would not
appear as unrelated.

An environmental upgrading program of this kind would
generate demand for labor from nonmigrant households that
could benefit further if they undertook their own on-farm
improvements. An alternative arrangement might be that if
they gave some free labor to the general area improvement
they could obtain subsidies for improvement on their own

Maintenance of Real Income in Poorer Nonmigrant Households

As we have seen,' we need to be concerned with wage
levels of both men and women. The departure of so many men
almost certainly increases male wages. The overriding
threat to the real value of these increases comes from local
inflation. Food purchases are a major item of expenditure
in wage-laboring households. Encouraging output on female-
managed farms should contribute to moderating price

increases of both grains and higher nutritional value foods.
But in isolated areas the price elasticity of increased
demand from migrant households for certain foods can still
be high. Improving the infrastructure of supply from other
areas should help.

It is doubtful that female wages rise anywhere near as
fast as the price of essentials except in instances of a
massive withdrawal of the labor of migrants' wives from
field agriculture. Professionalizing the agriculture of
small-scale farming while husbands are still away should
help to maintain or increase the demand for female wage
labor. It would prevent land from going out of production,
raise the value of farm output, and incline women farmers to
shift away from some fieldwork to roles of managing and
supervising the farm. An emphasis on making the small farms
managed by migrants' wives more viable also restrains the
tendency toward the labor-displacing mechanization that is
seen first on larger farms. Were a program of environmental
improvement as described above undertaken, stability of
demand for both male and female wage labor could be
maintained for many years. In the long term such a program
would increase the greater production-bearing capacity of
the land, and that should provide more opportunities for
hired labor in agriculture.

This monograph began by suggesting that studying the
accommodations that migrants' wives make can illuminate
areas of conflicting theories on the social and private
gains and costs of male out-migration. It ends by pointing
to ways in which the costs might be reduced by empowering
women to manage better on their own, and by recognizing that
this can be a precondition to diverting migrants' savings
from conspicuous consumption to productive investment.


Abadan-Unat, Nermin. "Implications of Migration on Emanci-
pation and Pseudo-emancipation of Turkish Women."
International Migration Review 11, 1, (1977).

Bettles, Fiona. Women's Access to Agricultural Extension
Services in Botswana. Nairobi, Kenya: Ford Foundation
Workshop on "Women in Agriculture in Eastern and
Southern Africa, 1980.

Birks, J. S., and C. A. Sinclair. International Migration
and DevelopmentI in the Arab Region. Geneva:
International Labour Office, 1980.

___ __ __ and J. A. Secknat. Internation-
al Migration Project Country Case Study: Yemen Arab
Republic. Department of Economics, University of
Durham, 1978.

Bohning, W.R. Studies in International Labor Migration. New
York: St. Martin's Press, 1984.

Colson, Elizabeth. "Family Changes in Contemporary Africa."
In Black Africa, edited by J. Middleton. London: Mac-
millan, 1970.

Dorsky, Susan Joan. Women's Lives in a North Yemeni High-
lands Town. Ph.D. thesis. Cleveland: Case Western
Reserve University, 1981.

Fortmann, Louise. Women's Involvement in High Risk Arable
Agriculture. The Botswana Case. Ford Foundation Work-
shop on "Women in Agriculture in Eastern and Southern
Africa." Nairobi, Kenya: 1980

Gay, Judy. "Wage Employment of Rural Basotho Women: A Case
Study." South African Labour Bulletin 6, 4, (1980).

Gordon, Elizabeth. The Women Left Behind: A Study of the
Wives of the Migrant Workers of Lesotho. World
Employment Programme Working Paper. Migration for
Employment, WEP 2-26/WP 35. Geneva: International
Labour Office, 1978.

"An Analysis of the Impact of Labour
Migration on the Lives of Women in Lesotho." In African
Women in the Development Process, edited by Nici
Nelson. London: Frank Cass, 1981.

Government of Swaziland'. The Survey of Roles, Tasks, Needs
and Skills of Rural Women in Swaziland, 1978-79.
Ministry of Education and UNICEF, 1980.

Hammam, Mona. "Labor Migration and the Sexual Division of
Labor." MERIP Reports, March/April (1981).

Khattab, H. A. S. and S. G. El Daeif. Impact of Male Labor
Migration on the Structure of the Family and the Roles
of Women. New York: The Population Council, 1982.

Kossoudji, Sherrie, and Eva Mueller. The Economic and
Demographic Status of Female Headed Households in Rural
Botswana. Research report No. 81-10. Ann Arbor:
Population Studies Center, University of Michigan,

Lipton, Michael. Migration from Rural Areas of Poor Coun-
tries: The Impact on Rural Productivity and Income
Distribution. Washington: Employment and Rural Develop-
ment Division, Development Economics Department, IBRD,

Low, Allen R. C. "From Farm-Homestead Theory to Rural
Development Policy in Swaziland and Lesotho." Labour
Migration and Agricultural Development in Southern
Africa. Rome: Food and Agricultural Organization, 1982.

Mueller, M. B. Women and Men in Rural Lesotho: the Periphe-
ry of the Periphery. Waltham, MA: Brandeis University,

Murray, Colin. "The Effects of Migrant Labour: a review of
the evidence from Lesotho." South African Labour
Bulletin 6, 4 (1980).

Myntti, Cynthia. Women in Rural Yemen. London: Department
of Anthropology, London School of Economics, 1978.

Male Migration and the Position of Women: A
Case Study from the Yemen Arab Republic. Mimeograph
1983. Revised and reprinted as follows:

"Yemeni Workers Abroad: The Impact on
Women." MERIP Reports 124, 14 (June 1984).

Shaheed, Farida. Migration and Its Effects on Women in the
Villages of Provenance, Pakistan. WEP 10-4-04, 24-1-
158-8. Mahabaleshwar, India: Tripartite Asian Regional
Seminar, Rural Development and Women, 1981.

Showers, Kate. "A Note on Women, Conflict and Migrant
Labour." South African Labour Bulletin 6, 4 (1980).

Sibisi, H. N. Feminine Dilemmas in Rural Swaziland. Nairobi,
Kenya: Ford Foundation Workshop on "Women in
Agriculture in Eastern and Southern Africa," 1980.

Swamy, G. International Migrant Workers' Remittances: Issues
and Prospects. World Bank Staff Working Paper No. 481.
Washington, D.C.: World Bank, 1981.

Swanson, Jon C.i Some!Consequences of Emigration for Rural
Economic Development in Yemen. \ Detroit: Wayne State
University, 1977.

Vletter, Fion de. Migrant Labour in Swaziland:
Characteristics, Attitudes and Policy Implications.
World Employment Programme Working Paper, Migration
for Employment, WEP 2-26/WP 22. Geneva: International
Labour Office, 1978.

"Labour Migration and Rural Development
in Swaziland." Labour Migration and Rural Development
in Southern Africa. Rome: Food and Agriculture Organi-
zation, 1982.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs