Professional Documents
Culture Documents
Gender and Social Policy in A Global Context: Shahra Razavi and Shireen Hassim
Gender and Social Policy in A Global Context: Shahra Razavi and Shireen Hassim
in a Global Context
Uncovering the Gendered Structure
of ‘the Social’
Edited by
Shahra Razavi and Shireen Hassim
Gender and Social Policy in a Global Context
Social Policy in a Development Context
General Editors: Thandika Mkandawire and Huck-Ju Kwon, both of UNRISD
Social Policy in a Development Context is a new series which places social policy at the centre
of research while maintaining the United Nations Research Institute for Social Development
(UNRISD)’s unified approach to social development. The series provides a new and exciting
contribution to the literature in economic development and social policy. In economic devel-
opment, social policy has been recognized as an integral part of development, but the litera-
ture often falls short of elaborating social policy for a unified approach to economic and social
development. In social policy, analysis has concentrated mainly on European and North
American countries, and studies on developing countries often lack comparative rigour.
The bridge between economic development and social policy will not only contribute to the
academic research but also inform the policy debate at the international and national levels.
Titles include:
Olli Kangas and Joakim Palme (editors)
SOCIAL POLICY AND ECONOMIC DEVELOPMENT IN THE NORDIC COUNTRIES
Massoud Karshenas and Valentine M. Moghadam (editors)
SOCIAL POLICY IN THE MIDDLE EAST
Economic, Political and Gender Dynamics
Huck-Ju Kwon (editor)
TRANSFORMING THE DEVELOPMENTAL WELFARE STATE IN EAST ASIA
Maureen Mackintosh and Meri Koivusalo (editors)
COMMERCIALIZATION OF HEALTH CARE
Global and Local Dynamics and Policy Responses
Thandika Mikandawire (editor)
SOCIAL POLICY IN A DEVELOPMENT CONTEXT
Shahra Razavi and Shireen Hassim (editors)
GENDER AND SOCIAL POLICY IN A GLOBAL CONTEXT
Uncovering the Gendered Structure of ‘the Social’
You can receive future titles in this series as they are published by placing a standing order. Please contact
your bookseller or, in case of difficulty, write to us at the address below with your name and address, the
title of the series and the ISBN quoted above.
Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21
6XS, England
Gender and Social Policy
in a Global Context
Uncovering the Gendered Structure
of ‘the Social’
Edited by
List of Figures ix
Notes on Contributors xi
Foreword xvii
6 Social Policy Reforms and Gender in Japan and South Korea 130
Ito Peng
v
vi Contents
12 Health Sector Reform in China: Gender Equality and Social Justice 258
Jufen Wang
Index 340
Tables
1.1 Time spent in unpaid and paid work by men and women in
two adult families with a child under 5 years old (average
hours per day). 11
1.2 Women’s hourly earnings as a percentage of men’s
hourly earnings 13
3.1 Entitlements and financing of family benefits, Poland, 2003 73
3.2 State social support benefits in the Czech Republic as
of 1995/96 74
3.3 Three family benefits, total spending and spending as a
percentage of GDP, the Czech Republic, 1996–2000 75
5.1 Government budget: size and distribution, South Africa 123
6.1 Ageing population (% population over 65) in Japan and
Korea, 1980–2030 134
6.2 Total average fertility rate in Japan and Korea, 1970–2030 134
6.3 Family structures in Japan and Korea 1980–2000 135
6.4 Divorce rate (crude divorce rate per thousand population),
1980–2002 135
6.5 Percentage of single parents in Japan 135
6.6 Percentage of single parents in Korea 136
6.7 Changes in the population coverage of the four major
social insurances in Korea 139
9.1 Employment pattern of husbands and wives in urban,
poor households, 2002 Bolivia and Ecuador Sample Survey,
by sex 203
9.2 Access to benefits by workers, 2002 Bolivia and Ecuador
Household Sample Survey, by type of benefit and by sex 204
9.3 Monthly earnings and earnings variability of
self-employed workers, 2002 Ecuador and Bolivia
Household Sample, by sex 205
9.4 Patterns of household decision-making in couple
households, 2002 Bolivia and Ecuador Household
Sample, by sex 207
9.5 Mean debt service ratio of borrowers, 2002 Ecuador Household
Sample, by sex and type of household headship 209
9.6 Loan characteristics by gender and credit source, 2002
Ecuador Household Sample 210
10.1 Gendered risks and vulnerabilities associated with different
stages of the life cycle 220
vii
viii List of Tables
ix
Acronyms and Abbreviations
x
Notes on the Contributors
Lourdes Benería is Professor of City and Regional Planning and Women’s Studies,
and Director of the International Studies in Planning Program at Cornell
University. Her work has focused on gender and development issues, paid/
unpaid work, globalization, structural adjustment and Latin American development,
among others. She is the author of many articles and books on these topics. Recent
publications include Gender, Development and Globalization; Economics as if all
People Mattered (2003), with its Spanish translation published in 2005, and Global
Tensions; Opportunities and Challenges in the World Economy, edited with Savitri
Bisnath (2003).
xi
xii Notes on the Contributors
Jyotsna Jha was awarded a PhD in Economics of Education from Jawaharlal Nehru
University, New Delhi, in 1995. She has worked for both government and non-
governmental organizations, and for national and international agencies, for the
past several years in India. She has conducted research and evaluation studies
relating to various aspects of education policy, programming and financing. She
has also worked closely with teachers and administrators on matters of equity
in educational planning, research techniques, development of social learning
curriculum and educational management issues. She is the co-author of the recently
published book, Elementary Education for the Poorest and Other Deprived Groups (2005).
Jane Lewis was Barnett Professor of Social Policy at the University of Oxford and
is now Professor of Social Policy at the London School of Economics and Political
Science. Her recent publications include: Should We Worry about Family Change?
(The 2001 Joanne Goodman Lectures) (2003); The End of Marriage? Individualism
and Intimate Relations (2001); with Paul Bridgen, Elderly People and the Boundary
between Health and Social Care 1946–1991: Whose Responsibility? (2000); and with
Kath Kiernan and Hilary Land, Lone Mothers in Twentieth Century Britain (1998).
Maxine Molyneux is Professor of Sociology at the Institute for the Study of the
Americas, University of London. Among her recent books are Women’s Movements
in International Perspective (2000), M. Molyneux and Razavi, S. (eds), Gender Justice,
Development and Rights (2002), M. Molyneux and S. Lazar, Doing the Rights Thing
(2003), M. Molyneux and N. Craske (eds), Gender and the Politics of Rights and
Democracy in Latin America (2002), and E. Dore and M. Molyneux (eds), The Hidden
Histories of Gender and the State in Latin America (2000).
Relations and Outcomes: A Framework and Case Studies for Gender Aware-Planning
(1999).
Jufen Wang is Professor at the School of Social Development and Public Policy,
Fudan University, Shanghai, China, where she teaches courses on demography
and gender studies. She has published over 30 papers in various journals. Her main
publications include The Resolution of Pre-marital Pregnancy (2002) and Legal
Protection of Women’s Rights (2004). Her research interests are reproductive health,
women’s employment and social security issues.
Preface and Acknowledgements
This book is part of a series coming out of UNRISD’s timely and innovative
research programme on Social Policy in a Development Context. The programme
has launched a wide-ranging inquiry into models of social policy that are inclu-
sive, developmental and democratically grounded – especially for middle- and low-
income countries. Inclusive social policy, the programme maintains, is not a
luxury that only affluent countries can afford, but a necessity of social and eco-
nomic development. The programme has thus set itself the task of understanding
the social, political and economic conditions for the edification of a state-society
nexus that is developmental, inclusive and democratic.
As editors of this volume we would like to thank all the contributing authors
for their patience and persistence in responding to our numerous requests for
revisions. At UNRISD, Thandika Mkandawire was a major source of intellectual and
moral support; we also benefited from many useful discussions with Huck-Ju Kwon.
Our greatest debt goes to Alexandra Efthymiades for providing excellent research
assistance and for seeing the book through, with great patience and good humour.
We would also like to thank those who contributed to our work by agreeing to
review manuscripts, act as discussants, and provide advice at different points in
the life cycle of this project, especially Armando Barrientos, Ulla Björnberg,
Claudia Bonan Janotti, Sarah Cook, Monica Erwer, Éva Fodor, Tekaligne Godana,
Beth Goldblatt, James Heintz, Ping-Chun Hsiung, Elizabeth Jelin, Helen Schneider,
Stephanie Seguino, Ylva Sörman Nath, Ida Susser, Dzodzi Tsikata and Ann
Zammit.
xv
xvi Preface and Acknowledgements
We would also like to express our appreciation to the Centre for Global Gender
Studies at Göteborg University, especially Ann Schlyter, Monica Erwer and Monica
Lindberg-Falk, for organizing a memorable conference on the beautiful Marstrand
Island in Göteborg (Sweden) in May 2005.
The UNRISD research programme on social policy consists of both regional and
thematic components. This volume presents the comparative research carried out
by a network of gender specialists, largely though not exclusively from developing
countries, exploring the gender dimensions of social policy. Although it was
always the intention of the social policy programme to address gender issues
across all regional and thematic components, we were also convinced that this
effort would have to be complemented by a focused analysis of social policy
through a gender ‘lens’. Without the latter, we feared, ‘mainstreaming’ gender
would risk diluting it.
From their diverse regional perspectives the contributions to this volume show
how social and economic rights are shaped by economic structures and reforms,
institutional designs and capacities, and political processes which are all deeply
gendered. It underlines the importance of thinking beyond states and markets in
social provisioning, including in the analysis the interactions between these and
other social institutions, especially the family and community that take on much
of the burden, particularly in developing countries where formal social provision-
ing is thin. Although there have been changes in the balance of work and care in
many societies, this book shows that in many contexts these changes have rein-
scribed rather than eroded gender inequalities. The volume illustrates why both
academic research and policy thinking need to factor-in gender hierarchies and
structures if they are to address some of the key challenges of contemporary
societies: the widespread informality and insecurity of paid work, and the crisis
of care.
UNRISD would like to take this opportunity to thank the Swedish International
Development Cooperation Agency (Sida) and the United Kingdom Department
for International Development (DFID) for their financial support of the project.
for its financial support of the project. As is the case with all other UNRISD pro-
jects, work on this project would not have been possible without the core funding
of the governments of Denmark, Finland, Mexico, Norway, Sweden, Switzerland
and the United Kingdom.
Thandika Mkandawire
Director, UNRISD
xvii
1
Gender and Social Policy in a Global
Context: Uncovering the Gendered
Structure of ‘the Social’
Shireen Hassim and Shahra Razavi
The past decade has witnessed a renewed interest in social policies, and some
governments have increased social spending to soften the impacts of economic
reform. These changes have come in the wake of widespread realization of the
failure of the neoliberal economic model to generate economic growth and
dynamism, and to reduce poverty. At the same time, processes of political liberaliza-
tion have opened spaces for social movements in many parts of the developing
world to articulate demands for more effective social policies that mitigate the
effects of market failures and reduce inequalities.
These contestations have coincided with a rediscovery of ‘the social’ in the
policy oriented literature, widely understood to embrace the cluster of social and
political institutions, norms, and relationships that define the boundaries of
market exchange, reduce transaction costs and enhance social and political stability.
Polanyi’s (1957) seminal work that showed the market to be a political and social
construct is widely cited today to explain the failure of the structural adjustment
packages that narrowly focused on ‘getting the prices right’, and to redirect atten-
tion to the institutional underpinnings necessary for successful market capitalism
(Ruggie 2003).1 However despite the movement away from the standard neoliberal
approach of the 1980s, and the increasing recognition given to institutions and
the state, there is little agreement on a number of critical issues. These include the
scope of social policy and the appropriate interface between social policy and
macroeconomic policy (Elson 2004; Mkandawire 2004; Tendler 2004); the role of
the state, not just as ‘regulator’ but also as a provider of social welfare; and the values
underpinning public policy, in particular core values of equality and redistribution
which seem to have been displaced by the discourse on poverty (Phillips 2001).
A gender perspective on social policies in the South, as in the North until quite
recently, has remained on the margins of these debates. This volume is an attempt
to move the gender analytical framework closer to the centre of social policy
thinking. From their different regional perspectives, the chapters in the volume
map out the complex ways in which social policies are filtered through social
institutions – families and communities; markets; care arrangements; health and
1
2 Shireen Hassim and Shahra Razavi
education systems; the public sector – that are ‘bearers of gender’. Moreover attention
to gender reveals the extent to which inequalities (of class, gender and region) are
being intensified as a consequence of shifts in the global economy, and processes
of privatization and commercialization taking place within countries. Women’s
unpaid care work continues to form the bedrock on which social protection is sub-
sidized, with erosions in state provisioning impacting most strongly on women.
Despite women’s increasing participation in paid work, labour markets continue
to reproduce gender-based segmentations and inequalities in wages/income,
work-related social benefits, and social security. As this book demonstrates, social
institutions are by no means homogeneous: economic, institutional and cultural
variations across countries and regions shape the nature of both risks faced and
forms of social protection available for women.
In this chapter, we root a theorization of gender and social policy in three key,
interrelated arenas: the nature of labour markets, the institutional basis for social
policy formulation (families, communities, markets and states) and the nature of
political contestation around social policy. In the first section, we lay out the
gendered nature of economic transformations in the late twentieth century,
drawing out the implications for gender equality of shifts in the nature of labour
markets and the relationships between paid and unpaid work. In the following
section, we link these changes in the structure of labour markets to a discussion
of the impacts of social sector restructuring. Here we examine the gender impli-
cations of commercialization and privatization of social services and income
supports as well as the policy turn to targeting and social insurance as a response
to the exclusionary effects of markets. The third section explores the institutional
basis for social policy formulation, examining more closely the assumptions
about gender roles and entitlements, especially in the key institutions of family
and community, and how these interface with the state. The relationship between
political democratization and the development of gender equitable social policy
is then examined. We are particularly concerned with exploring women’s
agency in relation to advocating for social policy change in ways that meet their
various needs.
To facilitate our discussion of the chapters, Figures 1.1 to 1.7 (pages 3 to 6)
capture some of the standard indicators of relative (or absolute) female status for
the countries included in this volume, ranked by per capita GDP (USD PPP
adjusted). These include indicators of educational status (gross enrolment rates at
the primary, secondary and tertiary levels, and adult literacy rates), health status
(maternal mortality rates), political status (percentage of parliamentary seats held
by women), and economic status (economic activity rates). While enrolment rates
in primary education (Figure 1.1) provide a generally more egalitarian picture
(India being an exception), inequality sets in at the secondary level (Figure 1.2)
especially for some low-income countries. Gross enrolment rates at the
tertiary level (Figure 1.3) provide a more complex picture with gender gaps being
in favour of males in some countries (such as Japan, South Korea, Bolivia, Tanzania),
and in favour of females in others (such as Sweden, Argentina, Jamaica), cutting across
3
Figure 1.1 Ratio of female to male gross primary school enrolment rates
Note: Data are from 2001–02.
Source: World Development Indicators 2005, World Bank.
Figure 1.2 Ratio of female to male gross secondary school enrolment rates
Note: Data are from 1999–2002.
Source: World Development Indicators 2005, World Bank.
4 Shireen Hassim and Shahra Razavi
Country (per capita GDP, $US, PPP adjusted)
income groups. Figure 1.5 captures maternal mortality, which seems to be more
sensitive to income levels, although again there is great diversity between coun-
tries reflecting differences in health systems and the accessibility of health care
(with Argentina and South Africa doing particularly badly for their income level).
5
1,600 1,500
1,400
1,200
1,000
800
600 540
420
400
230
200 130 87
56 76 44 43 83 31 13
82
9 20 10
2
0
In ($2 )
ua ($ 0)
m ($ 0)
($ 0)
($ 0)
($ 0)
($ 0)
ex ($ 0)
ut Ch ($8 0)
ric ($9 0)
Ar land 10, )
ze tina 10, )
So h R ($1 60)
re 15 )
ed $1 80)
)
26 0)
)
80
Ko . ($ 80
Ja n ($ 950
40
Ec dia ,46
Ja dor ,67
ca ,58
Th na ,98
nd ,58
os an ,01
69
o 84
7
Po a ( ,82
07
($ 5
liv ($5
ile ,9
Sw a ( ,7
n 6,0
,9
6,
ic 8,
ut ep 0,
6,
2
pa 2
Bo nia
n ($
ia
M ica
a
nz
ai
hi
la
Ir
e
R
C
ai
Ta
Af
ta
ge
h
c
h
C
So
C
Country (per capita GDP, $US, PPP adjusted)
Figure 1.5 Maternal mortality rates, 2000 (maternal deaths per 100,000 live births)
Source: Maternal Mortality in 2000: Estimates Developed by WHO, UNICEF, UNFPA, Geneva: WHO, 2004.
50.0% 45.3%
45.0%
40.0% 35.1%
35.0% 30.7%
29.8%
30.0%
25.0% 21.4% 22.6%
20.2% 20.2%
20.0% 18.5% 16.0% 17.0%
15.0% 11.7% 12.5%
8.8% 9.2%
10.0% 5.9% 7.1%
4.1%
5.0%
0.0%
In ($2 )
ua ($2 0)
m ($ 70)
C a ($ 80)
Ir $4 )
C aila $6, )
ta d ( 90)
8, )
ut Chi ($8 )
9, )
Ar lan 10 )
ze tina 10,5 )
)
0)
en 6, )
pa 26 )
26 )
)
0
( 80
M a ($ 10
o 840
Af e ($ 970
0
0
ed ($1 80
Ja ($ 950
($ 50
40
58
Th an ,58
Po a ( 82
ge d ( ,07
8
4
c 3,5
na 3,9
,0
ut ep 0,8
Sw rea 5,7
,0
,9
($
,
R $7
1
ia
$
$
Ko ($
an
(
ia
Ec ia
Ja or
os n
ic
ic
n
liv
d
nz
ai
hi
ric
ex
Bo
R
Ta
n
ch
h
h
So
So
C
Figure 1.7 Difference between male and female activity rates (economically active population
as a percentage of total population, 15⫹)
Note: Age ranges vary for some countries: Jamaica: 14⫹, South Africa: 15–65, Sweden: 16–64. For India and
Iran, estimates are projections based on EAPEP (Economically Active Population Estimates and
Projections) data, ILO.
Sources: LABORSTA Database, ILO; EAPEP Database, ILO; Labour Force Survey, September 2004, Pretoria:
Statistics South Africa, March 2005.
Figure 1.6 captures women’s relative access to parliamentary seats, which seem to
bear little relation to income levels, with Sweden and South Africa being the two
front-runners. Gender gaps in economic activity rates are shown in Figure 1.7, with
Sweden and Tanzania at the two ends of the income spectrum displaying the
narrowest gender gaps, while the widest gaps appear in Iran, India and Mexico
with income levels that are much higher than Tanzania’s.
We make two central arguments in this section. First, we note that historically a
major preoccupation of social policy has been to respond to labour market risks
(such as unemployment, old age, maternity) through transfer payments condi-
tional on previous employment.2 However, women’s access to paid work does not
easily translate into their enjoyment of social protection mechanisms that flow
from paid employment. Gender-blind analysis of social policy has tended to
underestimate the extent to which labour markets are themselves gendered
(for example, sex-segregation of jobs, differential pay, gendered definitions of
‘skill’), features which in turn shape and constrain women’s access to such
Gender and Social Policy in a Global Context 7
towards a more universalistic, even if highly stratified, model (Filgueira & Filgueira
2002).
In short, in deeply hierarchical and segmented societies, where some crucial
forms of social protection were linked to formal employment and where the latter
never included more than a small fraction of the population, vast sections of the
population, including the majority of women, were frequently excluded from
coverage. Yet it would be wrong to assume that women were absent from state
social provisioning and protection altogether. Not only did women make up a
significant proportion of social security beneficiaries as wives and daughters
(‘dependents’ in the language of social insurance) of male breadwinners, they were
also direct beneficiaries of some public services (health, education) as well as being
targets of so-called maternalist programmes aimed at mothers and their children
(Molyneux this volume). Women have also traditionally constituted a significant
share of state employees in the social sectors, as teachers, nurses and carers.
with women’s activity rates nearing men’s in some countries. The forces propelling
women into the work force are complex and diverse. In the OECD countries, the
trend of going to work is most noticeable among mothers with young children –
the group that was most likely to drop out of employment in the post-second
world war period (Orloff 2002). However, workforce participation rates are crude
indicators of working women’s situations. Women in these countries, for example,
tend to work on a part-time basis far more frequently than men (ibid). While
taking on part-time work may involve an element of real choice in some cases, for
others it may be ‘involuntary’, because of either the lack of full-time employment
options or the need to accommodate care in the absence of other viable sources for
its provision, or both.
Feminists have been particularly interested in Nordic social democracies that
appear to have achieved relatively high levels of gender equality through female
labour-force participation and the redistributive mechanisms of social policy.
Sweden, for example, has been able to forge what Hobson (this volume) calls ‘par-
ticipation parity’ in the labour market with nearly equal numbers of women and
men in the workforce (Figure 1.7). Yet this achievement masks the manner in which
gender inequalities have been reinscribed in new ways. For example, a significant
portion of women – over a third – are working part time4 and more than 90 per cent
of all part-time workers in Sweden are women. As Hobson argues, the Swedish model
is thus in practice a ‘one and three-quarters’ model: men work full time and invest in
their careers, while women work part time in the public sector, where it is easier to
combine employment with having a family. Even though parental leave schemes
and ‘daddy leave’ quotas were designed to be gender-neutral, the outcomes are
not. Throughout the 1990s, between 10 and 12 per cent of parental leave days
were taken by fathers. In 2005 men’s share rose to 17 per cent – still a far cry from
the equal participation of men in unpaid work. As Lewis and Giullari (this volume)
remind us, although the balance of contributions women and men make to house-
holds in the form of cash and care is changing it remains gendered. In most coun-
tries, women have added paid work to their existing responsibilities for care, while
men have decreased the amount of paid work they do and increased their care-
work only slightly (see Table 1.1 for a snapshot of selected OECD countries).
Recent research finds that despite some improvements in the 1990s, levels of
gender segmentation in the labour market remain high throughout the world
(Anker et al. 2003). Women tend to congregate in relatively low-paid and low-
status work at the bottom of the occupational hierarchy, and also to have little job
security. Gender segmentations in labour markets are more difficult to capture in
developing countries with pervasive agrarian and informal sectors, which are not
sufficiently covered in most large-scale statistical surveys.
For many developing countries increasing poverty and the commodification of
the economy are changing the coping strategies of households and communities
in a multitude of ways, causing upheavals in gender and generational patterns of
work and responsibility. Research from sub-Saharan Africa, for example, shows
that in the context of economic crisis and reform, it is becoming increasingly
necessary for all household members – whether female or male, young or old – to
Gender and Social Policy in a Global Context 11
Table 1.1 Time spent in unpaid and paid work by men and women in two adult families
with a child under 5 years old (average hours per day).
Source: OECD Employment Outlook, Paris: OECD, 2001, Table 4.5, p. 140.
take on paid work (Bryceson 1999). Much of this tends to be badly remunerated,
with women being overwhelmingly clustered in low-entry, low-return type
activities (Whitehead 2004).
According to the ILO (2002) informal employment5 comprises between one-half
and three-quarters of non-agricultural employment in developing countries. But
there are also important stratifications within the informal economy, based on
productivity and income as well as gender. The informal economy tends to be a
larger source of employment for women than for men in most countries (ILO
2002); Figure 1.8, based on a recent six-country study, shows strongly gendered
patterns of employment in the formal and informal economy (with the exception
of El Salvador where male and female rates are comparable). Women informal
workers tend to be over-represented in the more precarious and less remunerative
segments of informal work: they are more likely to work as own-account workers,
domestic workers, and unpaid contributing workers in family enterprises than are
men, while men are more likely to work as employers and wage workers (ILO
2002). This is confirmed by Beneria and Floro’s evidence (this volume) from
Bolivia and Ecuador where women tend to have relatively more precarious jobs
than men. Furthermore, as time allocation studies show, while informal work,
especially home-based work, may enable women to combine paid and unpaid
work (such as sewing garments on a piece-rate basis while looking over children
and doing housework), it entails an intensification of work which can seriously
affect women’s well-being (see Beneria and Floro’s chapter).
It is very difficult to find gender-disaggregated data on income and poverty.
The reliance on poverty lines and household expenditure data has profound impli-
cations for how gender issues are analysed. Measuring poverty on the basis of
12 Shireen Hassim and Shahra Razavi
Formal
Non-agricultural
Employers 56.6 n.a. 78.9 n.a. n.a.
Own-account 62.1 n.a. 45.1 54.6 n.a.
Private wage 84.6 151.9 87.5 n.a. 89.5
Public wage 87.7 107.6 116.2 84.2 95.6
Agricultural
Private wage 85.1 n.a. 105.4 n.a. 66.3
Informal
Non-agricultural
Employers 97.6 n.a. 83.8 n.a. 83.6
Own-account 50.6 n.a. 65.1 80.2 59.6
All wage n.a. 263.3 n.a. 69.8 n.a.
Private wage 79.5 317.1 75.4 n.a. 107.0
Public wage n.a. n.a. 135.2 88.0 99.2
Domestic 57.4 n.a. 56.2 n.a. 100.0
Agricultural
Own-account 53.3 n.a. 56.9 65.0 n.a.
Private wage n.a. n.a. 86.4 n.a. 98.5
Public wage n.a. n.a. 177.6 n.a. n.a.
Note: n.a. indicates that data were not available or that there were insufficient observations to derive
statistically significant estimates.
Source: Heintz, James. ‘Summary of Country Case-Studies and Tabulations for UNIFEM’s Progress of the
World’s Women 2005’. Background paper prepared for Progress of the World’s Women 2005: Women, Work,
and Poverty. New York: UNIFEM.
120%
100% 96%
80%
80% 72% 71%
60%
40%
20%
0%
Costa Rica (2003) El Salvador (2003) Ghana (1998/9) South Africa (2003)
There have been tremendous changes in the development policy mindset with
regard to the role of the social sectors in the last two decades. In this section we
address the implications of these shifts for the development of gender-equitable
social policies.
By the late 1980s it became increasingly evident that the poverty and social
disruptions caused by stabilization and adjustment were not the ‘transitional
phenomena’ or ‘frictional difficulties’ the international financial institutions (IFIs)
had initially assumed; they were pervasive, long-term and systemic. Studies of the
impacts of adjustment, and popular opposition to key adjustment-related
measures combined to bring into question dominant policy prescriptions. Global
policy pronouncements became less assertive about the imperative of cutting
social spending, more apologetic about the imposition of ‘user fees’, and began to
acknowledge that social policy could have a vital role to play in the development
process.
In the early 1990s the World Bank grudgingly accepted that adjustment
packages had paid too little attention to social privations and that in view of
empirical studies documenting the economic and social pay-offs from investments
in health and education it would be wise to prevent the ‘depreciation of human
capital’ during the adjustment process. Despite this shift, the dominant view was
that ‘fiscal restraint’ had to be strictly observed. The dilemma of how to respond
to social needs while remaining within the constraints of macroeconomic stabi-
lization was resolved by attempting to ‘target’ social expenditures to populations
most in need (Vivian 1995). Certain expenditures were thereby reallocated, for
example from secondary to primary education; and supplementary programmes,
‘safety nets’ and ‘emergency funds’ were developed for the poor.
A consensus seemed to be emerging on the importance of social provision and
social protection, especially in a globalized economy where vulnerabilities were
accentuated due to exposure to external risks. That this was not merely a rhetorical
shift becomes apparent from the increasing share of Official Development Assistance
(ODA) going to the social sectors (UNRISD 2005: Figure 8.1). Moreover, in several
countries, levels of public social spending were restored in response to popular
pressure and discontent (UNRISD 2005: Table 3.3). At the same time, the state’s
role in the development process as well as in ensuring effective governance was
increasingly recognized (World Bank 1997). But what precisely was the scope for
state action in the realm of social development and welfare policy? And how was
it to triangulate with the institutions of the market, family and civil society?
The 1997 Asian financial crisis prompted the G7 to request the World Bank to
formulate ‘social principles’ and ‘good practice of social policy’ as a guide to policy
makers worldwide. The new emphasis was reflected in the World Bank’s
2000/2001 World Development Report (Attacking Poverty), which identified ‘social
risk management’ (SRM) as the most sustainable basis for coping with risk and
reducing the vulnerability of the poor. In this framework, the state was expected
to provide ‘risk management instruments where the private sector fails’ in
Gender and Social Policy in a Global Context 15
addition to ‘social safety nets’ for risk-coping for the most vulnerable (Holzmann
and Jorgensen 2000: 18) – displaying continuities with the earlier generation of
minimal safety nets. As we will see later, the family and the community were seen
as central institutions in SRM, working in tandem with the market.
The welfare pluralist approach was adopted in a context in which the provision of
social services and social protection was being rapidly commercialized. Selectivity in
social policy thus went hand-in-hand with a trend towards multi-tierism in modes
of provision in several important areas – pensions in particular, but also healthcare
and education. While selectivity means narrowing the targets for support, multi-
tierism means reducing the state component and partially privatizing social protec-
tion. Market-based, individualized entitlements (such as private pensions and
health insurance) are thus seen as appropriate for those who can afford them, while
scarce public resources are to be channeled or ‘targeted’ to the poor in the form of
elusive safety nets and ‘basic’ public health and education services.
Hence, behind the apparent consensus about the importance of social policy
lurks the ‘traditional great divide in social policy debates’ (GASSP 2005) between
the so-called universal redistributive model with an emphasis on universal access
to social services and a focus on equity and redistribution, and the residual
approach which ‘sees social policies as a residual measure to address the plight of
the poorest and most vulnerable’ (p.3). The ascendance of the latter approach was
at least partly related to the particular policy emphasis on poverty alleviation in
the 1990s, now in full bloom with the Millennium Development Goals (MDG),
which sought to ensure that public resources were geared towards only the poor-
est (GASSP 2005). By contrast with developed countries, the elaboration of social
assistance and social security in developing countries has thus been rooted in the
new global discourse of poverty, which has in turn fed into and reinforced a
bifurcated view of social policy (markets for those who can afford them and ‘safety
nets’ for the poorest). Below we consider the key building blocks of the currently
dominant model of social policy – commercialization, targeting, and the new
discourse of welfare developmentalism – from a gender perspective.
referred to as Health Sector Reform (HSR), has entrenched the shift toward greater
commodification of public services. In sub-Saharan Africa, as Mackintosh and
Tibandebage (this volume) show, HSR has been promoted in a context of severe
poverty and in the wake of economic crisis. In Tanzania, for example, user fees
were introduced in the early 1990s at a time when cuts in government spending
and removal of subsidies on basic goods were disproportionately affecting the
poor. In a context where a significant proportion of the population cannot afford
its basic needs, the imposition of fees for healthcare has been impoverishing.
The currently dominant model of health service provision has particularly prob-
lematic implications for women – as users of health services, as health service
workers, and as providers of informal care. As Mackintosh and Tibandebage show,
HSR in Africa seems to be based on unrealistic assumptions about poor women’s
ability to muster the economic resources needed to access services for themselves
and their dependents, about their unlimited time and capacity to provide unpaid
care when formal care remains out of reach, as well as problematic assumptions
about the interests of healthcare workers and clients (for example, seeing them as
antagonistic).
A common policy response to the exclusionary effects of ‘user fees’ has been the
promotion of mutual health insurance schemes as well as social insurance
schemes. Unlike social insurance schemes which are employment-based, mutual
health insurance (MHI) schemes are voluntary schemes to promote the inclusion
of the poor and vulnerable by pooling their risks and providing exemptions for
those unable to pay. However most community-based MHI schemes face the
problem of low participation rates and lack of financial sustainability; in Tanzania,
for example, many rural Community Health Fund (CHF) schemes have not been
able to extend their participation rates beyond 10 per cent of eligible households
(Tibandebage 2004). Inability to pay constitutes one of the main reasons for non-
enrolment, which is likely to affect women more severely given that they are more
cash-constrained than men, and are likely to have more health needs. It is also not
clear how such schemes can provide exemptions for the poor and ensure financial
sustainability in the absence of significant subsidies from the state – given the
difficulties of having cross-subsidies from the better off in small-scale voluntary
schemes in poor communities.
Enrolment in social insurance programmes is very often employment-based,
with mandatory affiliation. In the case of developing countries, coverage has
tended to be limited due the large size of the informal economy, and the high rate
of evasion of contributions, even by employers and employees in the formal sector
(Huber 2000). For these reasons even though social insurance schemes facilitate
resource mobilization via contributions, they may not be the most effective
vehicle for extending coverage to the majority of the population, particularly
women who, as we showed in the previous section, tend to be informally
employed. But in some countries efforts are being made to extend social protec-
tion mechanisms to formerly excluded categories of workers (Lund, this volume).
The full implementation of these schemes, however, remains a challenge given the
administrative difficulties of reaching dispersed workers with erratic incomes,
Gender and Social Policy in a Global Context 17
monitoring the implementation of their new rights, and making their membership
financially sustainable.
The health insurance scheme built by SEWA (Self Employed Women’s
Association) in India, discussed in Lund’s chapter, is an example of an employment-
based scheme that is successful in reaching women informal workers; this is done
through an integrated insurance scheme which also offers life and asset insurance.
The reasons for its success include exceptionally strong leadership, top-flight
expertise from insurance experts, a willingness to respond to members’ needs and
set realistic rates, as well as the fact that it has been built on existing patterns of
solidarity of SEWA as a trade union and as a set of cooperatives for women with
strong organizational capacities. Replicating such conditions will not be easy.
Health sector reform in China, discussed in Jufen’s chapter in this volume, is
taking place alongside a fundamental restructuring of the labour force marked by
massive unemployment in the state industrial sector and large-scale migration of
the rural workforce into export-oriented factories. This stands in marked contrast to
the ‘full-employment’ scenario of pre-reform China when industry was exclusively in
state hands and welfare was provided through the enterprise. To be sure, gender
segmentations in the workforce often translated into stratified rights to welfare in
the pre-reform era: men greatly outnumbered women as permanent state workers
and also predominated in senior grades, while collectives constituted a feminized
sector (Lee 2005). But the proposed schemes for health insurance do little to
reduce gender segmentations, and are in fact likely to exacerbate gender-based and
other forms of exclusion.
The ‘basic health insurance scheme for urban workers’, which is the principal
component of China’s health insurance scheme for the urban population, covers
those who are in the formal workforce and have permanent residence permits,
thereby leading to the exclusion of informal workers, migrant labourers and those
who are not part of the workforce. These exclusions are exacerbated by the
scheme’s highly individualistic design which does not even provide coverage for
the ‘dependents’ of the insured. Furthermore, as Jufen’s contribution to this
volume clearly shows, a social insurance model with gender-neutral design and
individualized accounts is likely to produce very unequal outcomes for men and
women in terms of access to benefits (relative to need) when it is filtered through
structural inequalities, especially inequalities in wages/income and years of
employment. If coverage in social insurance programmes remains employment-
based and individualized with little subsidy from the state, then women’s labour
market disadvantages are likely to feed into their weaker claims on healthcare.
In the education sector the logic of ‘targeting’ promulgated at the international
level has prioritized primary education, where the ‘rate of return’ is presumed to be
higher and where public expenditure is considered to be pro-poor. In countries such
as India, discussed in Jha and Subrahmanian’s chapter, this logic seems to have led
to a significant reallocation of public social expenditure from higher education to
primary education. While this redistribution and the accompanying reforms in
Indian education have been vital for rapidly boosting India’s abysmal education
enrolment rates (Figures 1.1–1.4), it has also led to the unfortunate neglect of
18 Shireen Hassim and Shahra Razavi
The current jargon – ‘fiscal restraint’ and ‘efficient allocation of resources subject
to budget constraints’ – seems to suggest that the problems of targeting are
technical. By contrast, Mkandawire (2005) underlines the ideological and political
imperatives which determine the choice of instruments used to address poverty,
inequality and insecurity. As the literature on social assistance programmes pro-
viding unreciprocated aid to the ‘deserving’ poor has repeatedly shown, there is a
strong element of control that underpins targeted programmes which goes against
current notions of citizenship and empowerment. In measures such as community
targeting much discretionary power can be vested in local unaccountable admin-
istrators who then wield enormous powers over ‘matters of life and death’
(Mkandawire 2005).
It has long been argued in the context of US welfare policies that government
programmes were divided into two, gendered streams: those with the most
legitimacy guaranteed secure entitlements to some citizens in return for their
contributions, while other programmes provided unreciprocated aid to the
‘deserving’ poor (Fraser and Gordon 1994). The gender-coded contract-versus-
charity dichotomy persists today in many countries in the opposition between
social insurance and public assistance programmes. Welfare benefits targeted to
women as mothers, for example, tend to identify women and their children as
needy and poor, while shading out their contributions as workers/carers (whether
paid or unpaid) and their rights as citizens. Social insurance programmes, on the
other hand, confer entitlements on workers in return for their labour/financial
contributions, even if in fact benefits depart from actuarial principles and do not
reflect financial contributions (Fraser and Gordon 1994).
Current anti-poverty initiatives also invoke and depend on gendered assumptions
about identities, interests and responsibilities of citizens. In recent years women
have been heavily present in targeted social assistance programmes. This can, per-
haps, be seen as a response to feminist policy advocacy in many national and
global forums where women have been identified as ‘the poorest of the poor’ and
governments were requested to take appropriate action. One response has been in
the form of micro-credit initiatives, which have been directed at women on the
basis of their assumed thriftiness, efficiency and entrepreneurial skills.
Elsewhere social assistance is provided on the condition that clients conform to
certain normative understandings of ‘good motherhood’ or appropriate female
behaviour. This is most evident in the Mexican programme Oportunidades,
explored by Molyneux in this volume. This is a programme that aims to enhance
human development by focusing on children’s health, education and nutrition. It
is premised on strongly normative – but implicit – assumptions about the ‘natural’
role of mothers as full-time carers and nurturers of children. The programme
seems to be driven by the expectation that low-income mothers would accept ben-
efits for their children even if these are made available on demanding terms (the
unpaid work expected of them) as well as intrusive levels of control exercised by
programme managers. This may very well be the case, given the lack of any better
option as well as the side benefits that women may derive from involvement in
such programmes, such as greater self-confidence, an ability to access public
Gender and Social Policy in a Global Context 21
spaces and connect with other women beyond kinship and marriage networks.
But such objectives are not always built into the programme (through training and
group activities, for example), nor is it clear whether participation in such
programmes enhances women’s economic autonomy and security.
Debates about the relative balance of responsibilities between state and non-state
institutions in the provision of social services and social support are underpinned
by normative assumptions about the role of the state as well as about gender roles.
The residualist approach to social policy is not just an innocent statement describ-
ing the limited role played by the state in social provisioning, but one prescribing
how states ought to behave. As feminist critics have repeatedly argued, the shed-
ding of state responsibilities shifts even further the burden of social provisioning
to the unpaid providers of care within families, households, and communities. On
the other hand, defining the state ‘as the central locus of social policy’, as
Molyneux (forthcoming) notes, also carries normative assumptions, alluding to a
vision of a strong, capable and socially responsive state. While much of the debate
between neoliberals and their critics tends to be cast in terms of markets versus
states, in many countries, low-income populations have to creatively combine
24 Shireen Hassim and Shahra Razavi
social supports from a mix of formal and informal social institutions for their secu-
rity. As Molyneux goes on to argue, ‘social reproduction is in such cases secured by
a variety of social practices and institutions that exist independently or work in
conjunction with “state action”; yet this interface, if it is acknowledged at all, is
rarely analysed in the social policy literature’ (Molyneux forthcoming).
In this section we address the interface between the state and other social
institutions. Firstly, we discuss the implications for gender equality strategies of
the assumptions made by states about the role of the family in social provisioning,
and the ways in which state policies attempt to reform the family. Secondly, we
address the implications for developing gender-responsive social policies of
differing levels of capacity, reach and legitimacy of state institutions. Finally, we
consider the ability of women’s organizations to extract benefits for women out of
processes of political liberalization and democratization.
reproduce and entrench gender inequalities. Who in the community does the car-
ing and on what terms? Who does the voluntary work? Who benefits from social
provisioning provided by religious-based voluntary institutions and on what
terms? Who in the household does the bulk of social provisioning and caring?
These questions are not new. They lie at the heart of debates to measure the extent
of social rights in welfare states and are central to contemporary processes that
have been loosely termed ‘reprivatization’. The care burden imposed by the AIDS
epidemic, as Mackintosh and Tibandebage note, has cruelly exposed the inadequacy
of the assumptions about women’s coping capacity and unlimited labour supply.
This is forcing onto the policy agenda questions about care and its provision, and
the costs to the carer of unpaid and voluntary work.7
Steinhilber’s chapter shows that despite the divergent reform paths taken in
Poland and the Czech Republic, in both countries the family was addressed as ‘the
most natural’ social institution to ensure care for its members; as increasing
responsibilities were assigned to families in the reform process, supporting families
in coping with the social fallouts from the transition was considered eminently
important. Gender relations in families, however, and the impact of family
benefits on them did not become an issue for public debate. Indeed, reform
discourses have invoked and aimed to recreate traditional gender arrangements
with motherhood as a full-time dedication. These idealized notions of the family
and of women’s roles are often in conflict with the reality of women’s day-to-day
lives and their continued attachment to the labour market (more so in the Czech
Republic than in Poland). Indeed, women face greater difficulties in the post-
reform era in combining employment and family responsibilities than they did
prior to the reforms. Ironically, despite the pro-family rhetoric, in both countries
aggregate expenditure for family benefits declined over the course of the 1990s. In
Poland, the share of family benefits in total social expenditure declined while total
social expenditure increased.
A more radical engineering of the family is evident in the Islamic Republic of Iran
where a revolutionary Islamist state has sought to restructure the family, and society
more broadly, along traditionalist lines by attempting to segregate all public
spaces, and to fully domesticate women.8 Moghadam argues in chapter 4 that state
‘developmentalism’ in the Middle East was undermined by its ‘neopatriarchal’
approach to women, gender, and the family, codified in family laws/personal status
codes that define women largely in terms of their filial, marital, and maternal roles,
place them under male guardianship, and deny them equality in access to family
wealth. Although state social expenditures did allow for social mobility and access
by some women to education and employment, on the whole, oil wealth and fam-
ily laws served to prevent the ‘commodification’ of women’s labour. On these
grounds she argues that processes of individuation and commodification need to
be supported as without these developments women are unlikely to be considered
as economic agents and right-bearing citizens. Hence, some disembedding of social
provisioning may be desirable as far as gender justice is concerned.
Individualist approaches to development have, however, been criticized for
assuming that poverty is the result of individual (or even broader cultural) deficits,
Gender and Social Policy in a Global Context 27
extent to which collective struggles for greater protection are successful. The next
section examines the interplay between these factors more closely.
1992). Gordon White argues that the effective linking of democratic and
developmental goals depends on the construction of ‘an effective developmental
state that is regulatory, competent and redistributive, and has the political authority
to manage social and political conflicts’ (quoted in Minogue 2002: 127).
The ‘good governance’ approach of the World Bank, which promoted democra-
tization through institutional development, recognizes the importance of ‘getting
the state right’. However, the emphasis is on strengthening the institutions that
are considered to be essential for capitalist development such as those responsible
for financial management, private property rights and rule of law. The aim of con-
tributing to more efficient government has not necessarily been linked to the
value of promoting government accountability downwards to citizens. Women
clearly have an interest in a responsive and accountable state, but one that is
responsive to their particular needs. There are gender-specific capacity failures in
all public institutions targeted for reform. Public expenditure management
systems fail to acknowledge women’s needs or distribute budgetary resources
equally. The civil service or judiciary may be dominated by men antipathetic to gen-
der equality. Women public-sector workers clustered at the bottom of state bureau-
cracies may be the first to be fired when cost-cutting efficiencies are introduced
(UNRISD 2005: Chapter 11).
As the Latin American and East Asian chapters suggest, competent public bureau-
cracies that are at least internally accountable can be made responsive to the needs
of women. In these regions states appear to be more able to act on political agree-
ments struck between political parties and other actors. Latin American states
achieved consolidation relatively early, often in association with demands for mass
incorporation into political institutions (Rueschemeyer et al. 1992). Although
there are variations in the social composition and democratic credentials of
states, on the whole Latin American states have been pervasive, relatively well-
institutionalized and with a strong history of interventionism. Grindle (1986: 13)
argues that ‘the Latin American state has played a considerable role in the mediation
of social conflict through co-optation, manipulation and coercion and this helps
account for the complexity and conflict that exists within the bureaucratic apparatus
of individual countries’.
The institutional context of democratization was also different in East Asian
states compared to sub-Saharan Africa. East Asian states inherited highly evolved
bureaucracies from a combination of Japanese models as well as from strong
networks between state and political and economic dynasties. This left a strong
foundation for the developmental states (indeed the authoritarian legacy allowed
developmental states to intervene in markets in productive ways). In these
contexts, the development of gender-equitable social policies was dependent far
more on winning political support and social consensus over the direction of
social policy than on state capacity to absorb women’s demands.
Weak state capacity also impacts on the kinds of strategies feminists can employ
in making social policy more gender-equitable. One of the most innovative of
these strategies, gender-responsive budgeting (GRB), has suffered from the com-
bined problems of political will and state capacity. This strategy is often adopted
30 Shireen Hassim and Shahra Razavi
because it is seen as the key tool for exposing and thereby re-directing government
policies and spending patterns to more gender equitable ends. As a ‘technical’ tool,
it is often regarded as a strategy that can bypass political and cultural obstacles to
equality. However, as Budlender argues in chapter 15 of this volume, GRBs can
often become a panacea for the more politically difficult work of redirecting
political will and institutional capacity. Budlender very clearly demonstrates the
differential outcomes of particular political contexts as well as the effects of the
lack of receptivity of politicians and state bureaucracies to change. Policy inertia
seems to dominate while opportunities for change (brought about by a combina-
tion of strong politicians, often female, with effective alliances in the face of
government receptivity to change) seem to be very rare moments.
allies in political parties allowed the adoption of a residualist welfare model that
seriously undermined women’s social rights.
A different dynamic was at play in East Asian developmental states, where the
process of democratization was more successfully pushed by local actors, and with
clear consequences for the expansion of social protection. Peng shows how male
bias in both the Korean and Japanese systems only softened as a consequence of
demographic shifts – declining birth rates and an ageing population – that in turn
became the touchstone for political competition. These demographic changes,
together with broader social changes, facilitated the erosion of traditional living
arrangements, increasing divorce rates and increasing numbers of single mothers,
accompanied by the increasing employment rates of women (including married
women). These social trends created a tension between caring needs on the one
hand (of children and the elderly) and the availability of women to provide
unpaid care on the other. Together these factors resulted in shifts in social policies
with increasing welfare expenditures. Care of the elderly shifted from being
means-tested to being rights based, while public supply of childcare was increased
and parental leave extended. All of this was facilitated by political regime shifts,
the extent to which social policy became an electoral issue and increasing
numbers of women in political office.
On the other hand, as Hassim’s chapter shows, similar processes of democratiza-
tion in South Africa have not had the same effects on the social welfare system.
There, the potential redistributive effects of regime shift and expansion of women’s
access to political office were mitigated by a dominant party system in which social
policy did not become part of electoral contestation, and by a labour market
characterized by high levels of unemployment rather than labour shortages.
Women’s movements in developing countries have an uneven record of
organizing for better and more appropriate forms of social protection. Many
women’s movements in developing countries have eschewed state-centred poli-
tics, questioning whether women’s citizenship could be expanded by states that
were fundamentally undemocratic in character. States are clearly not neutral in
designing and implementing social policies, but they are not self-evidently
patriarchal either. Gender relations are shaped by, and themselves shape, the
nature of the state and its relationships to other social institutions; in this sense
institutions are interpretive systems that give meaning to particular notions of
social positioning and citizenship. Yet unlike institutions such as the family and
community, states may be more permeable to women’s interests in contexts of
strong social conservatism. This may offer openings for women’s movements to
extend the reach of social programmes in ways that address women’s gendered
vulnerabilities. For example, in Latin America collectivization among poor women
did lead to successful pressures being brought to bear on ruling parties that were
seeking to extend their legitimacy among poor citizens. Populist governments in
the region sought to expand the basis of their legitimacy by expanding the welfare
net in ways that benefited both men and women.
It can be argued, therefore, that women have successfully made claims on the
state, very often by harnessing their maternal roles to political claims-making and
32 Shireen Hassim and Shahra Razavi
advocacy for better conditions and social support for women. Although maternal-
ist politics has had contradictory and different outcomes in different countries,
this form of claims-making shares an implicit acceptance that the rights women
were claiming should come in return for certain pre-given responsibilities tied to
traditionally-ascribed gender roles. This acceptance of traditional gender roles ren-
dered maternalist movements and demands controversial. The protean character
of maternalism also leant itself to subtle shifts from ‘a vision of motherhood in the
service of women to one serving the needs of paternalists’ (Koven and Mitchel
1993: 5). As such maternalism may reinforce the patriarchal gender order and
entrench women’s economic dependence on men – it can become ‘a cloak for
paternalism’ (ibid). For example, Islamist women activists in Iran have tended to
use maternalist arguments for greater social protection focusing on rewarding
women’s unpaid labour without addressing the underlying power relations of
gender. Consequently, Moghadam argues for a secular politics that focuses on
women’s access to the public sphere of paid labour and political participation.
From this platform, she argues, women are more likely to succeed in making family
relations more equitable.
spaces for feminist interventions. Rather closer to the experiences of the Middle
East, post-socialist countries ‘retraditionalized’ the family, upholding the distinc-
tion between the public and the private spheres (Haney and Pollard 2003). The
family was seen as ‘a site of refuge amidst chaos and unpredictability … and served
as a model and metaphor for transition’ (p.7). In Poland, even the several brisk
changes from left-to right-wing governments during the 1990s did not result in
progressive policies for women, because of the dominant social conservatism
(Steinhilber, this volume). Women were poorly represented in government and
the bureaucracy and there were few openings for feminists to make policy inter-
ventions. However, in response to the scaling back of benefits women’s organiza-
tions are beginning to make social policy reform part of their political activism.
Similarly in countries such as Iran the family is idealized as the antithesis to
Western individualism, making feminist struggles for more egalitarian social
policies that recognize women’s rights extremely difficult to pursue.
Concluding remarks
We have noted that historically state social provisioning and protection was
premised on a normative male-breadwinner/female-carer model, even if in reality
many women were in the labour-force. That model has eroded over the past two
decades, with the global increase in women’s labour-force participation. However,
despite the convergence in men’s and women’s crude economic activity rates,
gender segmentation is not disappearing from the world of work – not even
in Sweden where men continue to work full time and invest in their careers,
while women work ‘part time’ in the public sector, where it is easier to combine
employment with having a family. The link between paid work and entitlements
to social benefits has been further weakened by processes of labour informalization
or casualization, which have been part and parcel of liberalization policies over the
past two decades. A greater proportion of women’s paid work (compared to men’s)
tends to be of the informal kind, and within the informal economy women tend
to dominate the more casual and less remunerative niches.
Labour market segmentation and casualization mean that increasing numbers
of people, particularly women, are likely to be excluded from access to social
services and income supports if these are provided on a commercial basis or on the
basis of labour ‘contributions’ as in the social insurance model – ironically the two
paths that have dominated social sector restructuring in the 1980s and 1990s. In
theory it is possible to extend the coverage of social insurance programmes to
include informal workers – as some countries in East Asia and Latin America have
attempted – but only if their contributions are heavily subsidised (by the state). An
effective means for reducing gender-based poverty and inequality would be public
provision of accessible and accountable social services (especially health and
education) as well as citizenship-based entitlements to basic income support
(pensions and family/child allowances).
Despite significant differences played by families and households in social
protection and provisioning (these being the ultimate safety nets in many poorer
Gender and Social Policy in a Global Context 35
Acknowledgements
We thank Alexandra Efthymiades for excellent research assistance, James Heintz for the
preparation of tables, and Francie Lund and Silke Steinhilber for extensive comments on an
earlier draft of this chapter.
Notes
1 One response has been the ‘good governance’ agenda promoted by the World Bank,
which rehabilitates the state by emphasizing the role of ‘lean’ state bureaucracies and
judiciaries in creating the conditions for market competition through the enforcement of
private property rights and contracts, and by ‘regulating’ private industry and commer-
cial social services (World Bank 1997, 2003; Fukuyama 2004).
2 The welfare state broadly defined not only includes transfer payments but also the public
funding and delivery of goods and services, such as education and health. However, the
quantitative literature on the development of the welfare state contains a one-sided focus
on transfer payments (and neglect of social services), which is related to the other major
lacunae in this research – the role of gender (Huber and Stephens 2000a).
3 The other categories were ‘strong male breadwinner states’ where Ireland and Britain were
placed, and ‘modified male breadwinner countries’ which included France (Lewis 1992).
4 Though long part-time, 20 to 30 hours per week.
5 ‘Informal sector’ covers informal enterprises. Informal employment is a broader category
that includes all workers who work without secure contracts, worker benefits or social
protection; so it can include those who work as informal wage workers for formal
enterprises or households (e.g. casual labourer, domestic worker) (ILO 2002).
Gender and Social Policy in a Global Context 37
6 ILO, UNRISD, UN/DESA, UNESCO, UNDP (and even some parts of the World Bank) are
among the organizations often mentioned as those promoting this global discursive shift
(see GASPP 2005).
7 There are some similar gender concerns about the burden and division of care work in the
more affluent countries in the context of ageing (Stark 2005).
8 What the Islamist state has attempted to do however should not be mistaken for social
reality. Processes of social change, in which women have been major change agents, have
subverted many of the Islamization measures promulgated by the state.
9 For developing countries, however, we need more systematic analysis of unpaid work
than is currently available through the existing time use surveys.
References
Anker, R., H. Melkas and A. Korten (2003) Gender-Based Occupational Segregation in the 1990s.
ILO Working Paper No. 16. Focus Programme on Promoting the Declaration on
Fundamental Principles and Rights at Work. Geneva: ILO.
Bates, R. (1994) ‘The Impulse to Reform in Africa’. In J. A. Widner (ed.) Economic Change and
Political Liberalization in Sub-saharan Africa. Baltimore: Johns Hopkins University Press.
Blondet, C. (2002) ‘ “The Devil’s Deal”: Women’s Political Participation and
Authoritarianism in Peru’. In M. Molyneux and S. Razavi (eds) Gender Justice, Development
and Rights. Oxford: Oxford University Press.
Bryceson, D. F. (1999) African Rural Labour, Income Diversification and Livelihood Approaches:
A Long-Term Development Perspective. ASC Working Paper No. 35. Leiden: African Studies
Centre.
Cho, Hyoung, Ann Zammit, Jinjoo Chung and In-Soon Kang (2004) ‘Korea’s Miracle and
Crisis: What Was in It for Women?’ In Shahra Razavi, Ruth Pearson and Caroline Danloy
(eds) Globalization, Export-Oriented Employment, and Social Policy: Gendered Connections.
Basingstoke: Palgrave.
Devereux, Stephen and R. Sabates-Wheeler (2004) Transformative Social Protection. Working
Paper No. 232, Brighton: IDS.
Elson, Diane (2004) ‘Social Policy and Macroeconomic Performance: Integrating “the
Economic” and “the Social” ’. In T. Mkandawire (ed.) Social Policy in a Development Context.
Basingstoke: Palgrave.
Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism. Princeton: NJ Princeton
University Press.
Filgueira, F. (forthcoming) ‘The Latin American Social States: Critical Junctures and Critical
Choices’. In Y. Bangura (ed.) Democracy and Social Policy. Basingstoke: Palgrave.
Filgueira, F. and Filgueira, C. (2002) ‘Models of Welfare and Models of Capitalism: The Limits
of Transferability’. In E. Huber (ed.) Models of Capitalism: Lessons for Latin America.
University Park PA: Pennsylvania University Press.
Fraser, N. and L. Gordon (1994) ‘Civil Citizenship Against Social Citizenship? On the
Ideology of Contract-Versus-Charity’. In van Steenbergen (ed.) The Condition of Citizenship.
London: Sage.
Fukuyama, F. (2004) ‘The Imperative of State-Building’. Journal of Democracy, Vol. 15, No. 2,
pp.17–31.
GASPP (Globalism and Social Policy Programme) (2005) Copenhagen Social Summit Ten Years
On: The Need for Effective Social Policies Nationally, Regionally and Globally. Policy Brief No 6,
Helsinki: GASPP/STAKES.
Grindle, M. (1986) State and Countryside. Baltimore: Johns Hopkins Press.
Haggard, S. and R. R. Kaufman (1995) The Political Economy of Democratic Transitions.
Princeton: NJ Princeton University Press.
Haney, L. (2003) ‘Welfare Reform with a Familial Face’. In L. Haney and L. Pollard (eds)
Families of a New World: Gender, Politics, and State Development in a Global Context. New
York: Routledge.
38 Shireen Hassim and Shahra Razavi
Haney, L. and L. Pollard (2003) ‘In a Family Way: Theorizing State and Familial Relations’. In
L. Haney and L. Pollard (eds) Families of a New World: Gender, Politics, and State Development
in a Global Context. New York: Routledge.
Hassim, S. (2006) ‘Women in African Parliaments: The Virtuous Circle of Representation’.
In G. Bauer and H. Britton (eds) Women in African Parliaments. Boulder: Lynne Rienner.
Heintz, J. (2005) The Growing Informality of Labour Markets. Paper presented at the
UNRISD/CGGS Conference on Gender and Social Policy, Marstrand, Sweden, 28–29 May.
Hobson, B. (1990) ‘No Exit No Voice – Women’s Economic Dependency and the Welfare
State’. Acta Sociologia 33, 235–250.
Holzmann, R. and S. Jorgensen (2000) Social Risk Management: A New Conceptual Framework
for Social Protection and Beyond. Social Protection Discussion Paper Series, No. 6,
Washington DC: World Bank.
Huber, E. and J. D. Stephens (2000a) ‘Partisan Governance, Women’s Employment, and the
Social Democratic Service State’. American Sociological Review, Vol. 65, No. 3, 323–42.
Huber, E. and J. D. Stephens (2000b) The Political Economy of Pension Reform in Latin America.
OPG No. 7, Geneva: UNRISD.
Huber, E. (2000) Social Policy in a Development Context: Notes on Social Security and Pensions
Systems. Presentation at the UNRISD Conference on Social Policy in a Development
Context, Stockholm, 23–24 September.
International Labour Organization (ILO) (2002) Women and Men in the Informal Economy:
A Statistical Picture. Geneva: ILO.
Korpi, W. and J. Palme (1998) ‘The Paradox of Redistribution and Strategies of Equality:
Welfare State Institutions, Inequality, and Poverty in the Western Countries’. American
Sociological Review, Vol. 63, 661–87.
Koven, S. and S. Michel (1993) ‘Introduction: Motherworlds’. In S. Koven and S. Michel (eds)
Mothers of a New World: Maternalist Politics and the Origins of Welfare States. New York:
Routledge.
Lee, C.K. (2005) Livelihood Struggles and Market Reform: (Un)making Chinese Labour After State
Socialism. OP No. 2, Geneva: UNRISD.
Lewis, Jane (1992) ‘Gender and the Development of Welfare Regimes’. Journal of European
Social Policy, Vol. 2, No. 3, 159–73.
Mackintosh, M. and M. Koivusalo (2005). ‘Health Systems and Commercialisation: In Search
of Good Sense’. In M. Mackintosh and M. Koivusalo (eds) Commercialization of Health Care:
Global and Local Dynamics and Policy Responses. Basingstoke, Palgrave.
Minogue, Martin (2002) ‘Power to the People? Good Governance and the Reshaping of the
State’. In U. Kothari, and M. Minogue (eds) Development Theory and Practice: Critical
Perspectives. Basingstoke: Palgrave.
Mhone, G. (2004) ‘Historical Trajectories of Social Policy in Post-Colonial Africa: The Case of
Zambia’. In T. Mkandawire (ed.) Social Policy in a Development Context. Basingstoke: Palgrave.
Midgley, J. (1995) Social Development: The Developmental Perspective in Social Welfare. London:
Sage.
Mkandawire, T. (2005) Targeting and Universalism in Poverty Reduction. Programme Paper
No. 23, Geneva: UNRISD.
Mkandawire, T. (2004) ‘Social Policy in a Development Context: Introduction’. In
T. Mkandawire (ed.) Social Policy in a Development Context. Basingstoke: Palgrave.
Molyneux, M. (forthcoming) Poverty Relief and the New Social Policy in Latin America: Mothers
at the Service of the State? Programme Paper, Geneva: UNRISD.
Orloff, A. S. (1993) ‘Gender and the Social Rights of Citizenship: The Comparative Analysis
of Gender Relations and Welfare States’. American Sociological Review, No. 58, 303–28.
Orloff, A. S. (2002) Women’s Employment and Welfare Regimes: Globalization, Export Orientation
and Social Policy in Europe and North America. SPD No.12, Geneva: UNRISD.
Phillips, A. (2001) ‘Feminism and Liberalism Revisited: Has Martha Nussbaum Got It Right?’
Constellations 8, No. 2, 249–66.
Gender and Social Policy in a Global Context 39
In Latin America as elsewhere in the world, gender bias and masculine prerogative
have prevailed in social policy as in social life more broadly, with entitlements
resting on culturally sanctioned and deeply rooted notions of gender difference
and patriarchal authority. These have generally accorded with idealized assump-
tions about the asymmetric social positions occupied by the sexes with male
breadwinners and female mother-dependents receiving benefits according to
these normative social roles. Such assumptions have proved remarkably universal
and enduring even where, as in Latin America, gender divisions have been
modified by women’s mass entry into the labour force and by equal rights
legislation.
This chapter considers the changes and continuities in social policy provision in
Latin America through a focus on the ways that women, in particular mothers, are
positioned within the new anti-poverty programmes that have followed structural
reform. It examines a flagship anti-poverty programme known as Oportunidades
(Opportunities) established in Mexico at the end of the last decade.2 Seen by some
commentators as a quintessentially neoliberal programme, and embodying many of
the main ideas of the ‘New Poverty Agenda’,3 Oportunidades represents a novel com-
bination of earlier social policy approaches with the contractarian, co-responsibility
models associated with new approaches to social welfare and poverty relief. This
chapter is in two main parts; the first provides the background context for the
emergence of the new approaches to poverty; the second describes and critically
examines the Mexican programme’s selective construction of social need.
43
44 Maxine Molyneux
widely different degrees, some forms of social provision began to evolve. These
were principally concentrated on the education and health sectors and, where
Bismarckian models were influential, as in Mexico and Chile, state pension
schemes, along with other forms of social insurance for privileged (predominantly
masculine) sectors of the labour and armed forces, accompanied the process of
state formation.
From the first decade of the twentieth century social rights increased as a result
of successful demands by organized labour and socialist parties for social reform,
with an incremental assumption of social responsibility by governments. In the
1920s and 1930s ‘improving the “race” ’ in order to secure the conditions for
development and head off threats of disorder became the leitmotif of the social
reform and eugenics movements. Many women were among the promoters of
‘social hygiene’ and its derivative, the science of puericultura (child development).
They energetically supported policy and legal changes which were maternalist in
orientation, demanding benefits and services for mothers and children. Mothers
were among the first to be recognized as social policy claimants whether as mar-
ried women or as ‘unfortunates’, that is, impoverished single mothers. However, it
was often stated in the discussion of these provisions that it was primarily in the
interest of their children that women might receive benefits of a financial, educa-
tional or medical kind. In other words it was in the construction of children’s
needs that their mothers received entitlements, that is, in order to better fulfil
their maternal responsibilities.
The era of nationalist state-centred development under corporatist populism
was inaugurated by the crisis of 1929 but was more securely established in the post-
war period. It brought some expansion in entitlements, notably for organized
labour, the natural constituency of corporatist regimes and a relatively privileged
sector for long afterwards. Social rights correspondingly expanded in Mexico,
Chile, Argentina and Brazil among others, and even as populist corporatism
waned, the technocratic developmentalism that replaced it continued to expand
the social sector.
By the end of the 1960s all but the poorest states had established the main
planks of social welfare, if at times in skeletal form. Health and education were
publicly funded, and social insurance systems covered some categories of formal-
sector workers. Regional policies were now influenced by the ECLAC (United
Nations Economic Commission for Latin America and the Caribbean) or CEPAL-
ISTA guidelines which drew on human capital theory to anchor social policy more
firmly in a discourse of development priorities, as Latin American states presided
over a rapid expansion of literacy programmes and primary education. At this
time too, the ‘basic needs’ approach was gaining support, leading to some greater
attention to ‘subsistence rights’6 through the provision of food to the poor, sani-
tary works, potable water and affordable housing. Positive growth rates, rapid
urbanization and social mobilization, all caused Latin American states, irrespective
of political inclination, to embark on programmes to meet rising social demands
and expectations. These decades saw Latin America leading the developing coun-
tries in terms of social expenditure and social coverage. There was a corresponding
Mothers at the Service of the New Poverty Agenda 45
entitlements remained closely bound to the family, and were accessed by virtue of
their status as wives and mothers. If, in the public realm of work, women employees
gained recognition as mothers and were granted comparatively generous mater-
nity leaves, in the private realm, on marriage, they lost certain citizens’ rights.
Instead, like children, they gained a measure of protection, as those of dependent
status. Widows of soldiers, professionals and some formal-sector workers were able
on this basis to claim their deceased husbands’ pensions. Yet most working women
were located in low-paid jobs, in unorganized sectors of employment, and in work
that was considered supplementary to the male wage and lacked social protection.
In Mexico, the cataclysmic upheavals of the revolution did not greatly alter this
general picture. The constitution of 1917 and the civil code of 1928 accorded
women legal equality and gave them some new rights, such as the right to
divorce, but denied them full civil rights and political rights; universal female fran-
chise was not granted until 1953, later than in most other Latin American states.
Over time the law placed limits on men’s authority over their wives, but it was
only in 1974 that the code making women responsible for the domestic sphere,
and abandoning the home a specifically female offence, was repealed (Varley
2000).
As elsewhere in Latin America, Mexican women workers and women’s organiza-
tions pressed for the regulation of their working hours, in order to protect them
from over-exploitation. They gained popular support for these demands in a
context where there were widely expressed concerns that this was necessary to
safeguard their ‘maternal functions’.8 Paternalist sentiments were aroused by such
claims, with women and children positioned in this discourse as requiring protection;
many women protested that this was an excuse to deny them the right to equal
work and well-paid jobs. Mexico’s ‘perfect dictatorship’, a blend of one-party rule
and corporatism, combined revolutionary rhetoric, radical reform measures and
social conservatism. Catholicism retained a hold over the population and, despite
the secular disposition of the early revolutionaries, exerted a continuing influence
over social policy and women’s rights. In other ways too, women lost out in the
process of post-revolutionary state formation. Corporatism established a political
bond between the worker and the ruling party ensuring the loyalty of the more
powerful sectors of organized labour. Male-dominated trade unions were the prin-
cipal beneficiaries of corporatist social contracts that enrolled men in the service
of the state as workers and patriots, their compliance secured through negotiated
pacts over wages, working conditions and social security (Rosemblatt 2000).
Where women had acquired a significant presence in the workforce, they laboured
in poorly paid, less-organized sectors. Not only were they marginal to the contrac-
tual negotiations of the corporatist state, but they also occupied an ambiguous
place in wage negotiations since their very presence in the workforce could com-
promise the historic demand of organized labour for a ‘family wage’, one premised
on female dependency and the presence in the family of the full-time housewife
and mother.
Corporatist bargaining did however secure the passage of some welfare
measures, but what evolved was a selective, predominantly urban system of
Mothers at the Service of the New Poverty Agenda 47
welfare, with rural and indigenous women largely excluded from these and other
entitlements. Despite the agrarian reform’s revolutionary momentum, women
were more often than not denied rights to property in land, which through law or
custom remained a male prerogative, thus revealing the gender bias as well as the
limits of the ruling Party’s inclusive pretensions. If women gained entitlements as
workers, these too were not only restricted to a small section of the female
population, but were often unclaimable in practice.9 What little social protection
was available to women was more likely to be accessed through marriage and
family law which specified that it was a husband’s duty to provide for his wife and
children, and afforded women conjugal property rights.
While there occurred some limited individuation of women’s rights from the family
as a result of reforms spearheaded by feminist movements, these general features of
women’s social rights endured in Latin America. The restricted reach and scope of
social policy, the poor quality and difficulty of accessing many of the services and
benefits, meant that most low-income women could not and did not look to the state
for much in the way of support on their own behalf. They might be fortunate enough
to attain some minimum provision in education and health, and some support for
their children, but individual entitlements such as income support and pensions
were distant dreams for the majority. Security, such as it was, came from paid work
where it could be found, from marriage, kin and community, and from the church.
The fragility and inefficiency of the social security systems prevailing across much
of Latin America were features that were sharply accentuated by the broader socio-
economic trends that set in from the mid-1970s. The oil shocks, the debt crisis and
subsequent recession of the 1980s combined with demographic pressures – and in
much of the region with political conflict – to erode the social sector at precisely
the moment when its expansion was most needed. This was a period which saw
more women entering the labour force, while households sought to cut consump-
tion, substituting market-purchased goods and services with reproductive labour,
largely provided by women (González de la Rocha 1994).
The human and social costs of the first phase of the structural reforms helped to
revive longstanding debates over social sector reform which now took place both
within the region and in international development policy arenas. Cornia et al.’s
UNICEF study, Adjustment with a Human Face (1987) is widely acknowledged as a
‘wake up call’ to international agencies to pay attention to the social costs of
adjustment, but it was also important for the policy recommendations that it
made. While presaging future policies, these recommendations also confirmed
trends that were already under way; these included more targeting as a means to
enhance the redistributive role of the state, employment-creating labour-intensive
public work schemes, subsidies of certain items, and, crucially, for the scope of
social policy to be widened during times of adjustment.
These ideas were already achieving wide acceptance in development institutions,
but would take time, political will, and resources to implement across the region.
48 Maxine Molyneux
As Brock et al. (2001) argue, the evolution of development policy is not usefully
seen as following a uni-linear dynamic; the social policies and development
Mothers at the Service of the New Poverty Agenda 49
for reform are essential in understanding the ways in which social policy was
refashioned in the changed circumstances of the 1990s.
The Latin American experience of post-adjustment policy-making was marked
by an ad hoc accommodation to the precepts of the New Economic Model,
combining new and old policy elements, while at the same time generating some
original trends and initiatives that entered the global debate on poverty relief. In
practice policies range along a spectrum from social liberal variants with univer-
salist inclusionary principles to those based on targeted provision with a greater
role for privatized services. Brazil for example introduced a universal public
pension to which everyone over a certain age is entitled. Indeed the Brazilian
Constitution of 1988 was the first to combine proposals for an alliance between
state and civil society, with a commitment to universal social polices and raising
total public expenditure, along with new institutional structures, such as manage-
ment councils and public hearings, where state and civil society could ‘work
together to ensure that priority-setting matched the public and private interests
and secure accountability in the definition and delivery of social policies’ (Coelho
et al. 2002). In all of these variants, the Latin American region has seen the
establishment of a wide range of new agencies and institutional structures for
providing access to social services. Government-sponsored poverty programmes
have been established to complement the continuing work of Social Investment
Funds, and consultative processes and institutions have been put in place at local
and national levels. As elsewhere, the official policy discourses and forms of enti-
tlement that are being created in Latin America tend to place more emphasis on
individual responsibility, while social security is defined in official statements as
no longer residing solely with the state. It now involves the ‘co-management of
risk’: that is, the individual has to make responsible provision against risks
(through education and employment); the family too, must play its part (through
better care); while the market (through private interests) and the community
(through decentralization, ‘co-responsibility’ and the voluntary sector) are all
involved in the decentring of expectations of welfare from the state.
The specificity of the Latin American region not only stamped its mark on how
these ideas would materialize in policy but also how they would be received by
citizens. In a context of widespread distrust of the state and weak social protection,
the refiguring of state–society relations offered by the NSP received a mixed
response, not by any means all negative. The core ideas at least seemed to offer
some potential for advancing much needed reforms, if social and political condi-
tions allowed. Decentralization, ‘good governance’, accountability, participation
and urgent attention to poverty, resonated with the reform agendas of democratic
parties, movements, and civil society organizations that were working to democ-
ratize politics and society following years of military rule.12 From the 1980s calls to
‘deepen’ democracy, and to address the ‘social deficit’ of the adjustment years
converged with some of the Good Governance and state reform agendas. The
human rights movement in the 1990s was enjoying a particularly prominent
international role, and this impacted on Latin America at a time of considerable
receptivity to the new inclusions of women’s and children’s rights and indigenous
52 Maxine Molyneux
claims for recognition and justice (Molyneux and Lazar 2003). Women’s organiza-
tions of various kinds were particularly active in promoting women’s rights,
working simultaneously within communities and at state level to advance reforms
in the areas of violence against women, legal and political representation and
reproductive rights. They also helped to establish and sustain popular health
movements, leadership and legal literacy training for women throughout the
1980s and 1990s.13
Where was the state in this scenario? Analyses of neoliberal restructuring have
documented the scaling down of entitlements and of governments’ commitments
to universal provision showing the trend towards a greater reliance on the private
and third sectors for welfare delivery. These trends have been seen as evidence of
the shrinking of the state, its ‘hollowing out’, ‘evacuation’, even ‘disappearance’.
Does this accurately describe what has occurred in Latin America? The evidence
suggests a more complex picture, a less ‘zero sum’ situation. There has undoubtedly
occurred a decisive shift from the state-centric principles that previously governed
social policy, but this has not gone along with the ‘evacuation of the state’ from
social provision. This widely held view of post-1980 reforms does not allow us to
capture what is different about current social policy figurations of the state, or the
substantial changes that the state itself has undergone during the decades of
‘neoliberal hegemony’. In Latin America social reproduction in the domains of
health and educational provision remains by and large the responsibility of the
state across a range of countries, despite decades of creeping privatization and
underfunding. After the critical watershed years of the debt crisis when social
expenditure per capita fell to unprecedented levels,14 by 1991 it recovered the
levels registered at the beginning of the 1980s, and in recent years social expendi-
ture has generally risen across the region. While these figures must be treated with
caution, the role of the state in public welfare remains significant.
The state itself however has undergone reform in this process, led by efforts to
advance good-governance agendas designed to make state institutions more effi-
cient and accountable and by democratic reform parties and movements. This has
gone along with support for decentralization and deconcentration, with Latin
America taking the lead in the 1990s as the region that advanced furthest down
this path. Re-democratization involved a wave of constitutional reform across the
region, and decentralization was one of the democratic principles incorporated in
the new frameworks to redress a historic legacy of over-centralization. Across the
Latin American region from the 1980s, states have been engaged in a process
aimed at strengthening and reforming local government, while devolving a
greater share of the budget to locally administered state agencies. This process of
‘municipalization’ has brought the state back into the domain of welfare provi-
sion, albeit in a new guise. As we will see in the case of Mexico, poverty relief has
engaged states, both central and local, in a wide range of programmes involving
millions of dollars of public and international funding. None of this is to suggest
that the decentralization process in Latin America has overcome distributive prob-
lems or secured adequate citizen representation. Devolved resources remain sparse
and without plans to tackle regional economic regeneration, decentralization has
Mothers at the Service of the New Poverty Agenda 53
PROGRESA/Oportunidades
of the economically active population depends upon the informal sector for its
income, and has access to few benefits. Moreover the size of the informal sector
means that Mexico collects only 11 per cent of GDP in tax, well below the average
for Latin America (which is 18 per cent) and below that of relatively low-tax
countries such as the United States.16
The election of Partido Acción Nactional (PAN) leader Vicente Fox in 2000 ended
71 years of one-party rule by the Pártido Revolucionario Institucional (PRI), and
was accompanied by efforts to reform existing institutions along more democratic
and accountable lines. Fox pledged to make social justice a priority of his govern-
ment, recognizing that poverty was a ‘multidimensional phenomenon’, and
raising social expenditure by an average of almost 10 per cent per annum. The
main poverty relief programme directed at those in extreme poverty, PROGRESA
was modified and relaunched in 2002 under the name of Human Development
Opportunities (Desarrollo Humano Oportunidades) known today as Oportunidades.
The programme’s coverage, formerly restricted to the rural poor, was extended to
include urban and semi urban areas,17 and the number of those inscribed in the
programme was expanded from 2.6 million families (in 1999), the equivalent of
40 per cent of all rural families (Rocha Menocal 2001: 520), to 4.2 million families
in 2002 (of whom 2.9 were rural) (González de la Rocha 2003: 14). By 2005 it
covered five million households with an estimated 25 million beneficiaries.
Oportunidades is one of the most extensive programme of its kind in Latin America.
It is also considered to be the most successfully developed example of the region’s
NSP-inspired anti-poverty programmes.18 It has been judged to be particularly
effective in meeting its goals, and this is attributed to an unusually high degree of
presidential support and inter-ministerial collaboration along with an annual budget
(in 2004) of 25 billion pesos and a recent loan of $1 billion from the IDB. An
undoubted strength of the programme is that it is subject to regular evaluations,
including some by outside bodies, and has been responsive to suggestions for
improvements and modifications. Oportunidades is a targeted cash transfer pro-
gramme that attempts to combine short- and long-term objectives of sustainable
poverty reduction, as advanced by the social risk-management approach. As noted
earlier this approach aims to tackle poverty though helping the poor to ‘cope, mit-
igate or reduce’ their risk of falling into or being trapped in poverty. Oportunidades
aims to improve human development by focusing on children’s education, nutrition
and health. It is based on the assumption that poor households do not invest enough
in their human capital, and are thus caught in a vicious circle of intergenerational
transmission of poverty with children dropping out of school and destined to
suffer the long-term effects of deprivation.
Families selected for the programme are therefore helped, through cash transfers,
with the financial and opportunity costs of having children in school. The transfers
are primarily in the form of ‘scholarships’ for children to attend school,19 supple-
mented by additional cash to improve nutrition where required. The practical func-
tioning of the programme centres on mothers as the key to securing improvements
in the life chances of their children, born and unborn. It seeks to strengthen,
through workshops and monitoring, the mothers’ responsibilities for children’s
Mothers at the Service of the New Poverty Agenda 55
health and education and to improve the nutritional status of their children
(and of themselves, if they are pregnant or breastfeeding). Secondary outcomes
such as building the mothers’ capacities, empowerment, citizen participation and
strengthening community ties are part of the programme’s goals, but how these
are interpreted has varied over time and the quality of what is on offer under these
headings depends upon local authorities and cooperating professionals.
Oportunidades’ guiding principles are designed to differentiate it from assisten-
tialist programmes through an emphasis on the participants’ ‘active management’
of their risk through ‘co-responsibility’ (cogestion) or co-management. Responsibility
for health and education is to be recognised as not only the government’s but the
whole of society’s, and therefore should be assumed by the entire community.
However, the responsibility of the ‘entire community’ is perhaps better described
as being devolved to mothers who are those designated as being primarily respon-
sible for securing the programme’s outcomes. Co-responsibility is formalized
through a quasi-contractual understanding that in return for the entitlements
proffered by the programme, certain obligations are to be discharged by the two
parties, that is, the programme and the participating mother. This conditional
form of entitlement, although well established in other regions and emanating
from the United States, is of more recent origin in Latin America but it is being
widely adopted. In this case the responsible participants (mothers) receive their
stipend on condition that they fulfil the duties laid out by the programme
managers: this involves taking children for regular health checks, meeting targets
for ensuring their children’s attendance at school, attending workshops on health
and programme coordinators’ meetings, and contributing a set number of hours of
work to the community, typically for cleaning buildings or clearing rubbish.
Failure to comply with the requirements can lead to being struck off the
programme.
On the available evidence collected through regular evaluations, the programme
has been largely successful in its own terms.20 Its stipends have reduced household
poverty, and have improved school attendance of children, as well as the health
and nutritional levels of all those inscribed in the programmes.21 These are impor-
tant gains, and are extensively discussed elsewhere. Here we will focus on some of
the more contentious aspects of the programme as highlighted by the women
participants in interviews and evaluations before considering the question of the
programme’s gender impacts.
As far as the design of the programme is concerned there appear to be two main
criticisms made by participants which raise some general issues of principle. These
are the issue of targeting and that of co-responsibility. With regard to the first, one
large-scale multi-site assessment carried out by the World Bank and others22 found
that while none of the women who participated in the programmes doubted that
Oportunidades had helped them in their struggle against poverty, there were some
criticisms of the way targeting was applied. Despite the rigour of the selection
mechanisms, and despite the claim that the programme is intended to be seen as
a way to ‘[access] a social right in a situation of social inequality’ (Rivero 2002: 6),
the targeting process attracted the strongest criticism from participants in some
56 Maxine Molyneux
evaluation exercises. Along with a general sense that more information should be
made available on the programme and on the means-testing mechanism itself,
dissatisfaction was expressed over the selection process, which was felt to be
arbitrary, excluding people whose needs were considered just as pressing as those
included in the programme. Means-testing was felt to ‘generate a lack of trust,
social divisions and feelings of envy and exclusion’ among those not selected.
(González de la Rocha 2003: 17). These are common problems faced by targeted
social-protection programmes in contexts where poverty is extensive and deep,
although Oportunidades’ coverage is more extensive than most programmes, a
factor which has caused it to be dubbed by those in charge of it, ‘a near universal
programme’.
A second general complaint voiced by participants was that they felt ‘discrimi-
nated against’ by its demands on their time. They believed that they were ‘treated
badly or […] were asked to do things in ways that offended their dignity’ (Rivero
2002: 4). As they expressed it, because they were ‘paid by the government’ they
were expected to perform community work, such as cleaning schools and health
centres, while others in the community did not.23 In another evaluation women
complained of being made to do ‘absurd’ tasks just for the sake of keeping them
occupied. The requirement to do community work had been incorporated into the
earlier programme and was continued into the new post-PROGRESA design by the
Fox administration, but following recommendations by evaluators the amount of
work time contributed was reduced, and it is still an issue under consideration.
In light of such findings it is not surprising that there was, among some
communities, resistance to accepting the notion of ‘co-responsibility’. Rather, the
requirements of the programme were seen in terms of ‘obligations’ and partici-
pants felt that genuine co-responsibility would also oblige teachers to accept their
‘responsibility not to miss classes so much’. This ‘inequality of responsibility’
made some participants resentful of the way they were expected to meet targets set
for monitoring the health and education of their children and could be ejected from
the programme for failing to do so. Why, they asked, should a teacher’s salary
not be reduced if they fail to turn up to teach, since mothers were fined for not
meeting their targets? This latter point reflected a general criticism that there
were few reliable mechanisms of accountability where complaints regarding the
behaviour of officials or professionals could be processed. Nor were the partici-
pants given an active role in the design, management and evaluation of the pro-
gramme (González de la Rocha 2003). It is hard to square these findings with
the view that the programme was intended to function as a way of exercising
civil, political and social rights and as a means to achieve full citizenship
(Rivero 2002).24
not only incorporated into, but is central to, the management and design of the
Oportunidades programme. There are four main aspects to this gender sensitivity:
first, the programme was one of the earliest in Latin America to give the financial
transfers (and principal responsibilities associated with them) to the female head
of participating households; second, the transfers associated with children’s school
attendance involved an element of affirmative action: stipends were 10 per cent
higher for girls than for boys at the onset of secondary school which is when the
risk of female drop-out is highest;25 and third, the programme’s healthcare bene-
fits for children were supplemented by a scheme which monitors the health of, and
provides support for, pregnant and breastfeeding mothers, and children under two
years of age. The fourth aspect of the project design which displays gender
sensitivity is the goal to promote the leadership and citizenship of the women
subscribed. These goals represent a combination of equality measures (for the girls)
and maternalist measures for their mothers, but with what outcomes and gender
impacts?
There is a paucity of detailed evidence on this question, and far from sufficient
to make an accurate estimate.26 Most information that is available comes from a
large-scale survey by Adato et al. (2000) and qualitative research by Escobar and
González de la Rocha (2004). These allow certain general points to be established.
In the first place, as is well known, improving the educational opportunities of girls
has strong potential to enhance their self-esteem and life chances, while at the
same time sending a message to households and to communities that girls are
‘worth investing in’. Secondly, stipends paid directly to mothers are widely
accepted to benefit their households through more equitable redistribution, but in
giving women direct control over cash resources, their standing in their commu-
nities as well as their leverage within the household can be enhanced. The evidence
from evaluations of the Mexican programme confirm these trends, although, as
one evaluation noted, while the mothers enjoyed some increased autonomy, this
did not necessarily translate into empowerment, since the latter depended on
more factors than control over a small money income (Escobar and Gonzalez de la
Rocha 2004). Women did however appear to feel that their self-esteem was
enhanced as a result of the stipends.27 They also appreciated the programme’s edu-
cation and training projects (including health and community leadership) where
these were well organized, but they wanted more access to education and training
(Adato et al. 2000).
More research into the gender impacts of the programme is needed to establish
if it is producing a redistribution of power and status within households and if so
to explain what effects this status reordering has on household livelihoods and
well-being. Transfers paid directly to women have the potential to generate con-
flict if men feel that they are entitled to control money resources and resent any
undermining of their authority. However, existing data indicate that no strong
relationship has been found linking the programme and the incidence of violence
in the home.28
These positive developments however might need to be qualified in the context
of more critical appraisals. While those available refer to the earlier years of the
58 Maxine Molyneux
programme they indicate issues that arose, some of which are ongoing, but also
the reflexive response of the programme to these problems. One evaluation of
Oportunidades by the Network of Rural promotoras and assessors (Red de
Promotoras y Asesoras Rurales) concluded that there was no significant improve-
ment in women’s position in their families,29 the stipend was insufficient to
overcome poverty, and the programme did not generate employment opportuni-
ties for school-leavers which would enable the cycle of poverty to be overcome.30
Escobar and Gonzalez de la Rocha (2004) further noted that the programme did
not take sufficient account of women’s income-generating and other activities
such as collective community work ( faenas) and that as a consequence women
could be overloaded with competing demands on their time.
In sum, Oportunidades has several positive features as a new human development
programme beyond its successes with regard to improving children’s health and
life chances. It has, over time, expanded its coverage and has sought to respond to
some of the gaps in its provision taking account of evaluations in the modifica-
tions of its programme. It has also made some headway in detaching poverty relief
from political patronage although much still remains to be achieved. It has also
helped low-income households to cope financially with the demands of school-
age children. It has remained, however, in essence a maternalist programme in
that it aims to fortify and normalize the responsibilities of motherhood as a way to
improve the life chances of children. Its human development rationale accurately
describes the programme’s goals as far as children are concerned, but its combined
focus on mothering and reproductive health has made it less likely to develop a
more differentiated set of capacities for the mothers. In effect Oportunidades creates
a dependency on a subsidy which confirms mothering as women’s primary social
role, one which may enhance their social status and self respect, but nonetheless
puts them at risk of remaining in poverty for the rest of their lives.
Oportunidades exemplifies the maternalism at the heart of many of the new anti-
poverty programmes being established in Latin America, and its organizational
principles raise some important questions for gender analysis. Feminist theory
and practice has suggested that if gender equality is to be tackled in development
and welfare programmes these must have some potential to empower women and
enhance their capabilities in ways that enable them to challenge relations of
inequality and subordination and at the same time provide some scope for female
economic autonomy. The new anti-poverty programmes may successfully identify
some unmet needs within poor households and communities but attending to the
needs of the women (the mothers) who are central to the functioning of these
programmes is not their explicit aim, any more than is gender equality a key
objective. The social construction of need in these programmes is child-centred, as
is their overall organization. Women are incorporated into programme design
(i.e. are visible) but in a way that depends upon the gender divide for its success. Thus,
even as women might be empowered within these structures (through managing
Mothers at the Service of the New Poverty Agenda 59
the subsidy), such programmes in effect reinforce the social divisions through
which gender asymmetries are reproduced.
In the first place they depend upon women fulfilling their ‘traditional’ social
roles and responsibilities. Oportunidades does so by basing its programme on nor-
matively ascribed maternal responsibilities, in effect making transfers conditional
on ‘good motherhood’. Men are not incorporated in any serious way, and no effort
is made to promote the principle that men and women might share responsibility
for meeting project goals. These programmes unambiguously rest on normative
assumptions concerning ‘women’s roles’ so that the work women undertake in
ensuring that children’s needs are met is taken for granted as something that moth-
ers ‘do’. The social relations of reproduction therefore remain unproblematized,
and the work performed easily naturalized.
Latin American cultural constructions of femininity are strongly identified
with motherhood, and serving the needs of children and household is generally
considered a primary maternal responsibility. Motherhood is often offered as the
explanation for political or civic activism, and allied with moral virtue, altruism
and self-sacrifice. It is likewise assumed by programme managers and participants
alike that any actions that improve the well-being of children are not, as
Bradshaw and Quirós Víquez express it, a ‘burden’ for women, and any ‘costs’
they bear are ‘simply part of the mothering role’ (2003). If femininity is closely
bound up with an affective investment in a self-sacrificing or altruistic motherhood,
the ideological site for contesting the demands of maternalist programmes is not
one that is easily occupied. Beneficiaries who miss a clinic appointment or a
workshop because they were working, lay themselves open to the charge of
being ‘bad mothers’ who do not care for their children (Bradshaw and Quirós
Víquez 2003).
The Mexican programme seeks even greater commitment from mothers
through the regulation of their domestic responsibilities, situating them as the
principal managers of their families’ needs. This does however involve some status
reordering in the family in favour of mothers but within the traditional division of
labour. While much is said about the ‘individuation of the social’ in regard to neo-
liberal policies, this does not apply to the women in these programmes who are
bound ever more securely to the family. If there is a new element beyond the
purely technical administration of the project, it is that which is introduced by
some (albeit limited) sensitivity on the part of the designers to issues of gender
equality at least in the case of the girl children, if far less so in the case of their
mothers. One must conclude that gender equity considerations had some influ-
ence in the design of these programmes in recent years, sometimes as a result of
feminist advocacy through NGOs, sometimes as a result of the shift in public and
professional attitudes occasioned by the spread of feminist ideas.
It remains the case that the women in such programmes are primarily
positioned as a means to secure programme objectives; they are a conduit of policy,
in the sense that resources channelled through them are expected to translate into
greater improvements in the well-being of children and the family as a whole.31
Such benefits as are derived by the mothers themselves as a result of participation
60 Maxine Molyneux
in the programme are a by-product of servicing the needs of others. This is com-
pounded by the fact that there is little in the design of the programmes that
advances women’s economic autonomy or security. Training for the job market is
limited or non-existent, and there is scant if any, childcare provision for those
women who want or need it because they work, train or study. Poor women are
often involved in income-generating activities which, while precarious, may leave
them without much disposable time or flexibility.32 Indeed, while rarely acknowl-
edged in the case of women, participation in anti-poverty programmes can have
negative consequences in incurring opportunity costs by preventing or restricting
women’s freedom to engage in paid work (Bradshaw and Linneker 2003). In the
Oportunidades programme the assumption that mothers were available to carry out
‘their’ stipulated duties in respect of their children takes no account of the fact
that there was according to one evaluation, an ‘increasing dependency’ on
women’s earnings, even though these were often meagre.33
This is not to deny that many women might choose not to work and might not
perceive the programme’s demands as anything other than helpful in relieving
some of the pressure on them to obtain paid work – especially if little is available
to them. However given the importance of women’s lifelong economic precari-
ousness and their need to secure cash incomes, the relative lack of attention to this
issue is striking.34
Conclusions
tackling social need, but in their narrow vision of how to overcome poverty;
stipends are no substitute for economic regeneration, and without attention to the
household livelihoods and long-term prospects of the poor, including women, such
programmes despite their good intentions, remain fundamentally trapped in
‘assistentialist’ paradigms.
Notes
1 This chapter is extracted from a comparative research paper entitled Poverty Relief and the
New Social Policy in Latin America: Mothers at the Service of the State? (Molyneux
forthcoming). I would like to thank Edurne Larracoechea for her invaluable research
assistance on this project. Thanks also to my colleagues at the Institute for the Study of
the Americas – Helga Baitenmann, Fiona Macaulay and Rachel Sieder – as well as to
Sarah Bradshaw, Jasmine Gideon, and María de la Paz López for their encouragement
and generous help with materials. Thanks, too to Elizabeth Jelin for her comments and
to the Oportunidades programme directors for their cooperation.
2 The analysis draws on conversations with Oportunidades Director Rogelio Gómez
Hermosillo and Concepión Stepa, Director of Planning and Evaluation, as well as on field-
visits to Huachinango, San Miguel de Tenango and Zacatlán de las Manzanas in July 2005.
3 This term was first used by the World Bank but it has acquired a wider currency since.
4 Colombia is a possible sixth according to some analysts. For overviews and analyis of
Latin American social policy see inter alia Abel and Lewis 1993 and 2002, Abel 1996,
Mesa-Lago 1994, Haagh and Helgo 2002, and Tulchin and Garland 2000.
5 It is significant that this group includes a socialist (Cuba), a market (Chile) and a mixed
economy (Costa Rica) model of welfare. See Mesa-Lago 2000 for elaboration of these
comparative observations.
6 See Eckstein and Wickham-Crowley (2003: 19) on subsistence rights.
7 Filgueira and Filgueira (2002) differentiate between countries characterized by ‘stratified
universalism’ (Uruguay, Chile, Argentina); ‘dual regimes’ (Brazil, Mexico) and ‘exclusionary
regimes’ (DR, Central America, except Costa Rica), Bolivia and Ecuador.
8 As early as 1906 in Uruguay bills were proposed to give rights to maternity leave, and leg-
islation to restrict women’s working hours were first introduced in Argentina in 1905.
9 Barriers to claim-making included administrative obstruction, low female educational
attainment, ignorance of rights; and for indigenous women, the lack of an identity card
would be sufficient to bar them from their entitlements.
10 By 2005 the talk was of the ‘post-post-Washington non-consensus’ reflecting the greater
pluralization of approaches to development in Latin America and in policy circles.
11 But one that was singularly gender blind. For a critical analysis see Molyneux 2002.
12 See Jelin and Hershberg 1996 for further discussion.
13 See Craske and Molyneux 2002 for case studies of women’s movement activism in the
1990s.
14 In real terms social spending per capita declined by 10 per cent between 1982 and 1986.
Even as it grew afterwards it remained 6 per cent below 1980 at the end of that decade,
and only recovered slowly in the 1990s (IDB 1996).
15 ECLAC and the Mexican government’s estimates broadly agree that 45 per cent of the
population live under conditions of poverty while other estimates put the figure as high
as 61 per cent in poverty and 26.5 per cent (25 million) in extreme poverty (Urquidi in
Middlebrook and Zepeda 2003). The top 20 per cent accounted for 59.3 per cent of
income in 1989; the bottom 20 per cent for 3.9 (Grindle 2000: 20).
16 In 1997 social expenditure accounted for 7.8 per cent of GDP. It has risen slightly since.
17 PROGRESA was preceded by PRONASOL also known as Solidarity, Mexico’s first large-
scale anti-poverty programme. Established in 1988 its conception of poverty relief was
Mothers at the Service of the New Poverty Agenda 63
quite different from PROGRESA and had party political objectives. It was designed by the
Salinas’ administration to offset the political consequences of the adjustment years and
revive the flagging political support of the PRI. According to Molinar and Weldon,
PRONASOL’s regional priorities were developed with three aims in mind: to reward PRI
loyalists, to reconvert PRD supporters and to punish PAN supporters (in Rocha Menocal
2001: 524). Such manoeuvres delivered the expected returns to the ruling party, but
the programme was discredited. When Zedillo assumed the presidency in 1994, he
replaced PRONASOL with PROGRESA claiming that his new anti-poverty programme
did not have a political agenda (Rocha Menocal 2001: 513). Although some political bias
continued, it was much reduced, the PRI lost the 2000 elections to the opposition
PAN party. Oportunidades has since sought to distance itself from this record of political
clientelism, with a public campaign message stating that social protection is a right and
allegiance is not due to any political party.
18 Some observers have preferred to ignore the Mexican contribution to the development
of the programme and have seen it as originating with the World Bank as it appears to
be in sympathy with its broader recommendations. PROGRESA was not imposed by the
World Bank and was intended to run only on federal funds with no direct funding from
the World Bank. Oportunidades is government-funded with loan support from the IDB as
above.
19 Monetary and educational grants are provided for each child under 22 years of age
who is enrolled in school between the third grade of primary and third grade of high
school.
20 The International Food Policy Research Institute website contains a number of evaluation
reports on different aspects of PROGRESA/Oportunidades. See in particular Skoufias and
McClafferty 2000, which covered the three years up until 1999, as a result of which the
programme was extended to rural areas. The results of a qualitative evaluation carried
out in six communities by Escobar and González de la Rocha were published in 2004 and
will be referred to here. Another large-scale evaluation is currently under way.
21 Escobar and González de la Rocha’s qualitative research confirmed Skoufias and
McCafferty’s findings that the largest positive impacts were on children in secondary school.
The latter’s survey finds a 10 per cent increase in enrolment for boys and 20 per cent
for girls along with an overall narrowing of the gender gap in education, particularly in
primary school.
22 Cited in Rivero 2002.
23 In the PROGRESA programme such work involved on average 29 hours per month.
24 Since the programme strives to separate social entitlements from political clientelism,
this is another sense in which citizenship is understood.
25 Grants rise with the age of the child and the sex difference in the grant starts with
secondary school, which is normally when girls drop out. In the third year of secondary
school monthly grants are about US$58 for boys and US$66 for girls.
26 An audit of the gender effects of the programme is currently under discussion.
27 Escobar and González de la Rocha 2004.
28 This is clearly contentious and requires further in-depth research. According to Adato
et al. 2000 and my own interviews with specialists in gender and poverty, there has been
considerable evidence of violence against women over control of the stipend in some
regions (Author’s interview, Oaxaca, July 2005).
29 Based on responses from 309 beneficiaries and 60 professionals in eight communities.
30 A new component has been added to the programme since these findings, the ‘Jovenes
con Oportunidades’ which provides youth training and work experience. However on a
field visit young people interviewed saw themselves as having no future in their
localities and dreamt of migrating. Without attention to rural livelihoods Oportunidades
risks educating the young for the US labour market.
31 BID data cited in Bradshaw and Quirós Viquez 2003.
64 Maxine Molyneux
References
Abel, C. (1996). Health Hygiene and Sanitation in Latin America c. 1870–1950. Institute of Latin
American Studies, University of London.
Abel, C. and Colin Lewis (eds) (1993). Welfare, Poverty and Development in Latin America.
Basingstoke: Macmillan.
Abel, C. and C. Lewis (eds) (2002). Exclusion and Engagement: Social Policy in Latin America 2,
Institute of Latin American Studies, University of London.
Adato, Michelle, B. de la Briere, D. Mindek and A. Quisumbing (2000). The Impact of Progresa
on Women’s Status and Intra-household Relations. Final Report. Washington, DC: International
Food Policy Research Institute, Washington D.C.
Angell, A. and C. Graham (1995). ‘Can Social Sector Reform Make Adjustment Sustainable
and Equitable?’ Journal of Latin American Studies, 27.
Arriagada, Irma (1998). ‘The Urban Female Labour Market in Latin America: The Myth and
The Reality’. Mujer y Desarrollo Series, No. 21, Santiago, Chile, ECLAC.
Barrientos, A. (2004). ‘Latin America: A Liberal-Informal Welfare Regime?’ In I. Gough et al.
(eds), Insecurity and Welfare Regimes in Asia, Africa and Latin America. Cambridge:
Cambridge University Press.
Bradshaw, S. (2002). Gendered Poverties and Power Relations: Looking inside Communities and
Households. Fundación Puntos de Encuentro, Managua.
Bradshaw, Sarah and Brian Linneker (2003). Challenging Women’s Poverty. London: Catholic
Institute of International Relations.
Bradshaw, Sarah and Ana Quirós Víquez (2003). ‘The Social Protection Network in Nicaragua:
A Gendered Analysis’. Unpublished draft.
Brock, K., A. Cornwall and J. Gaventa (2001). Power Knowledge and Political Spaces in the
Framing of Poverty Policy. Working Paper 143. Institute of Development Studies, Sussex.
Cardelle, A. (1997). ‘Health Care in the Time of Reform: Emerging Policies for Private–Public
Sector Collaboration on Health’. Miami, North South Centre, Vol. VI, No. 1.
Castameda, T. (2002). Tendencias de Largo Plazo en Tamamo, eficiencia y Focalización del gasto
Social en America Latina y el Caribe. Report prepared for the Social Protection Division of the
Banco Internacional de Desarrollo.
Chant, S. (1997). Woman-Headed Households; Diversity and Dynamics in the Developing World.
Basingstoke: Macmillan.
Coelho, S. et al. (2002). ‘Deliberative Fora and the Democratisation of Social Policies in
Brazil’. IDS Bulletin, Vol. 33, No. 2.
Cornelius, Wayne, A. Craig and J. Fox (eds) (1994). Transforming State-Society Relations in
Mexico: The National Solidarity Strategy. San Diego: La Jolla, California Centre for Latin
American Studies.
Cornia, Giovanni, R. Jolly and F. Stewart (eds) (1987). Adjustment with a Human Face. Oxford:
Clarendon Press.
Craske, N. and M. Molyneux (eds) (2002). Gender and the Politics of Rights and Democracy in
Latin America. Houndmills: Palgrave.
De Vylder, S. (2003). Gender Equality and Poverty Reduction Strategies. Discussion Paper
(unpublished).
Eckstein, S. and T. Wickham-Crowley (eds) (2003). Struggles for Social Rights in Latin America.
New York and London: Routledge.
Economic Commission for Latin America and the Caribbean (ECLAC) (1999). Social
Panorama of Latin America 1998. Santiago: ECLAC.
Mothers at the Service of the New Poverty Agenda 65
Economic Commission for Latin America and the Caribbean (ECLAC) (2000). The Challenge
of Gender Equity and Human Rights on the Threshold of the Twenty-First Century. Report to the
eighth session of the Regional Conference on Women in Latin America and the Caribbean.
Lima, Peru: ECLAC.
Enrique Rosas, Rocío (2002). ‘Women and Survival Strategies in Poor Urban Contexts: A Case
Study from Guadalajara, Mexico’. Journal of Development Studies, 18 (2–3): 81–108.
Escobar, A. (2002). ‘The PROGRESA Programme and social change in Rural Mexico’. In Louise
Haagh and C. Helgo (eds.), Social Policy Reform and Market Governance in Latin America.
Basingstoke: Palgrave.
Escobar Latapi, A. (2003). ‘The PROGRESA Programme and Social Change in Rural Mexico’.
In A. Escobar and M. González de la Rocha (eds), Evaluación Cualitativa del Programa de
desarollo humano Oportunidades. Mexico City: Serie, Documentos de investigacion 3,
SEDESOL.
Escobar Latapí, A. and M. González de la Rocha (2004). Evaluación Cualitative del Programe
Oportunidades. Mexico City: Centro de Investigación y Estudios Superiores en Antropología
Social.
Filgueira, Carlos and F. Filgueira (2002). ‘Models of Welfare and Models of Capitalism: The
Limits of Transferability’. In Evelyne Huber (ed.), Models of Capitalism. University Park PA:
Pennsylvania State University Press.
Gideon, Jasmine (2002). ‘Decentralization, Participation and Inclusion: Reassessing Primary
Healthcare Delivery in Chile’. In L. Haagh and C. Helgo (eds.), Social Policy Reform and
Market Governance in Latin America. Basingstroke: Palgrave Macmillan.
Goetz, A. M. and D. O’Brien (1995). ‘Governing for the Common Wealth: The World Bank’s
Approach to Poverty and Governance’. IDS Bulletin 26, 2.
González de la Rocha, M. (1994). The Resources of Poverty: Women and Survival in a Mexican
City. Oxford: Basil Blackwell.
González de la Rocha, M. (2001). ‘From the Resources of Poverty to the Poverty of Resources:
The Erosion of a Survival Model’. Latin American Perspectives 28: 72–100.
González de la Rocha, M. (2003) ‘Oportunidades y Capital Social’. Paper presented at the
Comisión Económica Para America Latina, Workshop on Social Capital and Anti Poverty
Programmes, Chile.
González Bombal, I. and R. Villar (eds) (2003). Organizaciones de la Sociedad Civil e Incidencia
en las Politicas Publicas. Buenos Aires: Libros de Zorzal.
Graham, C. (1991). From Emergency Employment to Social Investment: Alleviating Poverty in
Chile. Washington DC: Brookings Institute Press.
Graham, C. (1992). ‘The Politics of Protecting the Poor During Adjustment: Bolivia’s
Emergency Social Fund’. World Development, Vol. 20, No. 9.
Graham, C. (1994). Safety Nets, Politics and the Poor: Transitions to Market Economies.
Washington, DC: Brookings Institution Press.
Grindle, Merilee (2000). ‘The Social Agenda and the Politics of Reform in Latin America’. In
Tulchin and Garland (q.v.): 17–52.
Haagh, Louise and Camilla Helgo (eds) (2002) Social Policy Reform and Market Governance in
Latin America. Basingstoke: Palgrave Macmillan.
Hirst, P. and G. Thompson (2000). Globalization in Question. 2nd edition, Cambridge: Polity.
Inter-American Development Bank (IDB) (1996) Economic and Social at Progress in Latin
America. Report, Making Social Services Work, special edition, Washington DC: Johns
Hopkins University Press.
IFRI at http://www.ifpri.org
Jackson, C. (1998). ‘Rescuing Gender from the Poverty Trap’. In C. Jackson and R. Pearson
(eds), Feminist Visions of Development: Gender Analysis and Policy. London: Routledge.
Jelin, E. and E. Hershberg (eds) (1996) Constructing Democracy: Human Rights, Citizenhip, and
Society in Latin America. Boulder: Westview.
Kabeer, N. (1997). ‘Editorial. Tactics and Trade-Offs: Revisiting the Links Between Gender and
Poverty’. IDS Bulletin 28, 3: 1–25.
66 Maxine Molyneux
Kabeer, N. (2003). Gender Mainstreaming in Poverty Eradication and the MDGs. London:
Commonwealth Secretariat.
Kaufman, R. and G. Trejo (1997). ‘Regionalism, Regime Transformation and Pronasol: The
Policies of the National Solidarity Programme in Four Mexican States’. Journal of Latin
American Studies, Vol. 29, No.3: 117–45.
Laurell, A. C. (2003). ‘The Transformation of Social Policy in Mexico’. In K. J. Middlebrook
and E. Zepeda (eds), Confronting Development: Assessing Mexico’s Economic and Social Policy
Challenges. Stanford CA: Stanford University Press.
Linneker, Brian (2003). ‘Gender Comparisons of Capital Influences on the Well-Being of
Women and Households Experiencing Poverty in Nicaragua’. Unpublished report, mimeo.
Lipton M. and S. Maxwell (1992). The New Poverty Agenda: An Overview, IDS Discussion Paper
No. 306.
Middlebrook, Kelvin and Eduardo Zepeda (eds) (2003) Confronting Development: Assessing
Mexico’s Economic and Social Policy Challenges. Stanford CA: Stanford University Press.
Mesa-Lago, C. (1994). Changing Social Security in Latin America: Toward Alleviating the Social
Costs of Economic Reform. Boulder CO: Lynne Reinner.
Mesa Lago, Carmelo with Alberto Arevas de Mesa, Ivan Brenes, Verónica Montecinos and
Mark Samara (2000). Market, Socialist and Mixed Economies. Comparative Policy and
Performance: Chile, Cuba and Costa Rica. Baltimore: Johns Hopkins Press.
Mkandawire, T. (2004). ‘Disempowering New Democracies and the Persistence of Poverty’. In
Max Spoor (ed.), Globalisation, Poverty and Conflict. Dordsecht, London and Boston:
Kluwer.
Molinar Horcasitas, Juan and J. A. Weldon (1994). ‘Electoral Determinants and Consequences
of National Solidarity’. In Wayne A. Cornelius, Ann L. Craig and Jonathan Fox (eds),
Transforming State-Society Relations in Mexico: The National Solidarity Strategy. San Diego:
Center for US–Mexico Studies.
Molyneux, M. (2002). ‘Gender and the Silences of Social Capital: Lessons from Latin
America’. Development and Change, Vol. 33, No. 2: 167–88.
Molyneux, M. (forthcoming). Poverty Relief and the New Social Policy in Latin America: Mothers
at the Service of the State? Programme Paper, Geneva: UNRISD.
Molyneux, M. and S. Lazar (2003). Doing the Rights Thing: Rights-Based Development and Latin
American NGOs. London: Intermediate Technology Publishing Group.
Moser, C. (1996). Confronting Crisis: A Comparative Study of Household Responses to Poverty and
Vulnerability in Four Urban Communities. World Bank, Washington DC.
O’Donnell, G. (1993). ‘On the State, Democratization and Some Conceptual Problems: A
Latin American View with Glances at Some Postcommunist Countries’, World Development,
Vol. 21, No. 8: 1355–69.
OECD DAC (2001). Guidelines for Poverty Reduction, at http://www.oecd.org/
dataoecd/47/14/2672735. pdf.
Pastor, M. and C. Wise (1997). ‘State Policies, Distribution and Neoliberal Reform in Mexico’.
Journal of Latin American Studies Vol. 29: 419–56.
Piester, K. (1997). ‘Targeting the Poor: The Politics of Social Policy Reform in Mexico’.
In D. Chalmers, C. Vilas, K. Hite, S. Martin, K. Piester and M. Segarra (eds), The New Politics
of Inequality in Latin America. Oxford: Oxford University Press.
Red de Promotoras y Asesoras Rurales (2000). ‘Dinero del Diablo’. Unpublished report,
mimeo, May.
Rivero, Margot Aguilar (2002). ‘Oportunidades y Derechos Sociales: Un Proceso de construcción
social de Ciudadanía’. Unpublished report, mimeo.
Rocha Menocal, Alina (2001). ‘Do Old Habits Die Hard? A Statistical Exploration of the
Politicisation of Progresa’. Journal of Latin American Studies, Vol. 33: 513–38.
Rose, Nikolas and Peter Miller (1992). ‘Political Power Beyond the State: Problematics of
Government’. British Journal of Sociology Vol. 43 (2): 173–205.
Rosemblatt, K. (2000). ‘Domesticating Men: State Building and Class Compromise in Popular
Front Chile’. In E. Dore and M. Molyneux (eds), Hidden Histories of Gender and the State in
Latin America. Durham NC: Duke University Press.
Mothers at the Service of the New Poverty Agenda 67
Schell, P. (1999). ‘An Honourable Vocation for Ladies: The Work of the Mexico City Unión de
Damas Católicas Mexicanas, 1912–1926’. Journal of Women’s History, Vol. 10, No. 4: 78–103.
Schild, V. (2002). ‘Engendering the New Social Citizenship in Chile’. In M. Molyneux and
S. Razavi (eds), Gender Justice, Development and Rights. Oxford: Oxford University Press.
Skoufias, E., B. Davis and S. de la Vega (2001). ‘Targeting the Poor in Mexico: An Evaluation
of the Selection of Households into PROGRESA’, World Development, Vol. 29, Issue 10,
October: 1769–84.
Skoufias, Emmanuel and B. McClafferty (2000). ‘ “Is Progresa Working?” Summary of the
Results of an Evaluation by International Food Policy Research Institute (IFPRI) and Food
Consumption and Nutrition Division (FNCD)’. Discussion papers No. 118, Washington
DC, at http.//www.ifpri.org.
Stallings, B. and W. Peres (2000). Growth, Employment, and Equity: The Impact of the Economic
Reforms in Latin America and the Caribbean. ECLAC: Brookings Institute Press.
Tendler, J. (1997). Good Government in the Tropics. Baltimore: Johns Hopkins University Press.
Ticknell, A. and J. Peck (2003). ‘Making Global Rules: Globalisation and or Neoliberalisation?’
In J. Peck and H. Yeung (eds), Remaking the Global Economy: Economic Geographic
Perspectives. London: Sage.
Tulchin, J.S. and A. Garland (eds) (2000). Social Development in Latin America: The Politics of
Reform. Boulder CO: Lynne Reinner.
United Nations Development Programme (UNDP) (1995). Human Development Report: Gender
and Human Development. Oxford and New York: Oxford University Press.
Urquidi, Victor. (2003). ‘Mexico’s Development Challenges’. In Kevin Middlebrook and
E. Zepada (eds), Confronting Development: Assessing Mexico’s Economic and Social Policy
Challenges. Stanford CA: Stanford University Press.
Varley, A. (2000). ‘Women and the Home in Mexican Family Law’. In E. Dore and
M. Molyneux (eds), Hidden Histories of Gender and State in Latin America. Durham NC and
London: Duke University Press.
Weyland, K. (1995). ‘Social Movements and the State: The Politics of Health Reform in
Brazil’. World Development, Vol. 23, No.10: 1699–712.
Whitehead, Ann (2003). Failing Women, Sustaining Poverty: Gender in Poverty Reduction Strategy
Papers. Gender and Development Network Report, Oxford: OXFAM and Christian Aid.
Whitehead, L. (1993). ‘On Reforms of the State and “Regulation of the Market” ’. World
Development, Vol. 21, No. 8: 1371–93.
Wood, C. (2003). ‘Adjustment with a Woman’s Face: Gender and Macroeconomic Policy at
the World Bank’. In S. Eckstein and T. Wickham-Crowley (eds), Struggles for Social Rights in
Latin America. New York and London: Routledge.
World Bank (1997). World Development Report: The State in a Changing World. Oxford and
New York: Oxford University Press.
World Bank (2000). World Development Report 2000/2001: Attacking Poverty. Washington DC:
World Bank.
World Bank (2002). The Gender Dimension of Bank Assistance: An Evaluation of Results
Washington DC: World Bank.
Yaschine, I. (1999). ‘The Changing Anti-Poverty Agenda: What Can the Mexican Case Tell
us?’ IDS Bulletin, Vol. 30, No. 2: 47–60.
3
Gender and Post-socialist Welfare
States in Central Eastern Europe:
Family Policy Reforms in Poland and
the Czech Republic Compared
Silke Steinhilber
Introduction
Since 1989, post-socialist1 Central and Eastern European countries have under-
taken significant reforms of their social policies and social security systems. One of
the key challenges in so doing was balancing the tension between a tradition of –
and in some countries continued commitment to – extensive welfare provisioning
and income redistribution on the one hand, and the residualist social policy set-up
advocated by the currently dominant global neoliberal economic framework on
the other. Among the multiple challenges faced during the reforms, abolishing
gender inequality was given little, if any, priority.
Through an analysis of family benefits after 1989, including maternity and
childcare benefits, as well as transfers to families, this chapter illustrates that
reforms have nevertheless had important gender implications. I trace how and
why ‘costly’ benefits and services supporting women’s dual role as worker and
mother – a feature of the socialist past often considered ‘women-friendly’ –
changed dramatically. Thereby, I explore how the broader dynamics of post-socialist
social policy reforms have transformed the institutional and social environment
in which gender relations are reenacted and renegotiated.2
Both Poland and the Czech Republic faced the ‘competing demands of building
capitalism and democracy’ at the beginning of the 1990s (Orenstein 2001: 6). Yet
the two countries responded somewhat differently: Poland chose a rapid and rad-
ical economic transformation strategy – ‘shock therapy’. In the Czech Republic, on
the other hand, neoliberal and social democratic elements were combined into a
‘social liberal strategy of reform’ (Orenstein 2001: 7).
Both countries also followed different reform paths in the field of social policy.
During the 1990s Poland moved further than other post-socialist countries toward
a residualist and familial model of a welfare state, scaling down state involvement
68
Gender and Post-socialist Welfare States in Central Eastern Europe 69
and benefits had been introduced as early as 1924. The benefit for caring for a sick
child was introduced in 1954.
Parental leave schemes and benefits were of a more recent origin. They were
introduced in response to declining birth rates, yet were subject to political con-
flict throughout. In Czechoslovakia, family benefits were debated from the second
half of the 1950s. A flat-rate birth grant was introduced in 1956, and an extended
two-year maternity leave in 1970, initially unpaid. At the same time, a maternity
allowance was introduced for mothers of two or more children who stayed at
home. In Poland, child-raising leave was first introduced in 1968, initially consist-
ing of a one-year leave without pay. In the 1970s, with a low demand for labor
because of economic difficulties, and with declining birth rates, the state encour-
aged women to stay home for longer time periods (Lohmann and Seibert 2003: 78).
Unpaid leave was extended to three years in 1972 (Wóycicka et al. 2003: 194). Paid
leave for two years was introduced only in 1980 in response to popular demand,
and as a concession of the Communist government to the opposition Solidarity
movement which had initiated a campaign on the issue. A great number of Polish
women took advantage of the paid parental leave because the benefit level
was decent, and discontent with the workplace environment widespread.
Moreover, the supply and quality of institutional childcare was unsatisfactory
(Cichocinska 1993).
Since childcare was typically a mother’s responsibility in both countries,
childcare facilities were crucial to women’s ability to combine employment and
family responsibilities. In comparison with many Western European countries,
childcare was relatively widely accessible, generous and comprehensive in
Czechoslovakia. However, in Poland demand for childcare places greatly exceeded
supply throughout the entire postwar period (Lohmann and Seibert 2003: 70).
Also, regional differences in the availability of childcare services were substantial,
especially between urban and rural settings. Criticism about the services was
widespread: they were often overcrowded, and the ratio between children and
employees was high.
In both countries, nursery care enrolment, i.e. for children below age three, was
always lower than kindergarten enrolment. Places were rare, the quality of nursery
services was low, and state incentives for home care for young children were
strong. Nursery care enrolment in Poland declined even further during the second
half of the 1980s. The number of factory-run nurseries decreased by a quarter, the
number of places by a third, and the number of children in nurseries by almost half,
from a peak in the mid-decade. Kindergarten attendance also declined noticeably
over the 1980s (Cichocinska 1993).
Poland
Poland went through profound economic, political and social changes after 1989,
experiencing a triple challenge of transition to a market economy, integration into
a globalized world economy, and transformation for accession to the European
72 Silke Steinhilber
Union. Popular discontent with the economic reforms caused several swings from
left- to right-wing governments. Despite their divergent political orientations,
however, the different governments have on the whole followed a surprisingly
similar macroeconomic and social policy reform recipe. Macroeconomic austerity
measures dominated all reforms in the highly indebted country. In its broad
effects, gender equality policy – or the absence thereof – has also been surprisingly
similar under different governments, as the discussion below illustrates.
Following the postulates of the ‘Washington Consensus’, highly indebted
Poland initiated a rapid economic reform process which focused on macroeco-
nomic stabilization, price and trade liberalization, and privatization (Lavigne
1999; Orenstein 2001). Among the severe consequences was an initial steep
decline of GDP by about 20 per cent, a significant decline in real incomes, and a
drop in economic activity rates, as well as high unemployment and growing
poverty. During 1992 more than 25 per cent of all Polish families could not afford
to purchase the basic necessities of food and clothing (Seibert 2001: 42). The share
of the population below the poverty line hovered between 12 and 17 per cent in
the period 1995–2000 (Lohmann and Seibert 2003: 16).
Poland today is regarded a successful economic reformer in CEE. Relative
economic stabilization with fast economic growth ensued during the mid-1990s,
followed by a slowdown of the economy at the end of the decade and renewed
faster growth after 2003. Yet growth has had only a limited employment impact:
both women’s and men’s employment rates have been decreasing massively since
1990, and are low by international standards. Women’s employment rates are
significantly lower than men’s. Unemployment has remained at high levels of
between 15 and 20 per cent since the early 1990s, with women’s unemployment
consistently higher than men’s. Gender differences in wage levels are in the range
of 15–30 per cent, the highest gap being found in the highest wage group
(Chlon-Dominczak 2004). A joint EU/ Polish government analysis concluded that
‘there remain significant inequalities of opportunities between women and men
in the labor market’ (European Commission/Government of Poland 2001).
Three major social policy reforms were enacted in 1998/99: reform of the
education system, the national health insurance system, and the national pension
scheme, which was partially privatized. Family benefit reforms during the 1990s –
while no less profound in their impact – were designed and implemented in mul-
tiple steps: among the most significant was the decoupling of the entitlement to a
number of family benefits from employment. They were now paid from the state
budget – in some cases the central, in others the local. A majority of family benefits
became income-tested (with the exception of maternity benefits and short-term
childcare benefits) (see Table 3.1). Some benefits were cut, and the real value of
some further declined through inadequate adjustment for inflation (Wóycicka
et al. 2003).
Means-tested/
Basis for Financing not means
Benefit entitlement source tested?
1998, unemployment had surpassed 6 per cent for the first time since 1989, with
wide regional differences. It was as high as 15.6 per cent in the hardest-hit locali-
ties. Almost 60 per cent of those unemployed were women and about a third
young people. Women’s unemployment has been consistently higher than men’s
throughout the decade of the 1990s, with particularly high rates of long-term
unemployment.
The fiscal deficit widened in 2001 to 3.2 per cent of GDP as compared to 1.6 per cent
in 1999. An austerity programme therefore foresaw cutbacks in state spending. In
2000 and 2001, the Czech economy recovered to above 3 per cent growth on aver-
age. The unemployment rate, however, has been in the 8.5 to 9 per cent range
since 1999 and above 9 per cent in 2002.
Czechoslovakia, as Poland, entered the 1990s with a comprehensive social
security system. Yet since the old system was inadequate to address some of the
new needs emerging from the transition, such as unemployment and poverty,
social security reforms were supported across the political landscape. Early
reforms included the introduction of a household subsistence minimum, which
serves as a poverty benchmark, and the formulation of cost-of-living increases
of benefits (Kucharová et al. 2003). Two new family benefits were introduced in
1990, the state compensatory allowance – intended to buffer the negative effect
of consumer price liberalization – and a parental allowance for at-home care of
children.14 Other early reform steps included the revision of the maternity
benefit formula, and the decoupling of the child allowance from employment.
Instead, it was made dependent on the number of children in a family, and
their age.
A major reorganization of social security happened in 1995–96. Benefits were
grouped into three subsystems: social insurance, social assistance, and state social
support. Most family benefits are provided as state social support, with parental
allowance, child allowance and social allowance forming 85 per cent of total state
social support spending (Kucharová et al. 2003). State social support benefits are
financed from general revenues and paid out through the district authorities.
Some of the benefits are income-tested, but income-testing is not applied in a very
restrictive manner so far (see Table 3.2).
Several broad effects of the family benefit reforms became clear at the end of the
1990s. First, aggregate expenditure for family benefits in both countries declined
over the course of the 1990s:17 In Poland during the period 1990–98, from 1.7 to
1.06 per cent of GDP; and in the Czech Republic during the period 1996–2002,
from 1.6 to 1.2 per cent of GDP (Fultz and Steinhilber 2004). See Figure 3.1 and
Table 3.3.
It is noteworthy that in Poland, during most of the 1990s, the share of family
benefits in social expenditure declined, while total social expenditure increased.
This suggests that family benefits were more affected by cuts than other social
security benefits, for example pensions (Wóycicka et al. 2003: 207). Further
savings in family benefits were achieved by reducing benefit adjustments for
inflation. Consequently, the share of family benefits in household income
declined significantly (Wóycicka et al. 2003: 195).
A second effect was a declining number of beneficiaries and concentration of
benefits on low-income households. In the Czech Republic, the proportion of
families receiving the three most important family benefits fell by half. At the
same time, those still eligible for all three benefits experienced increases between
2.5 2.3
2.3 2.1
2.0
1.7 1.7
1.5
1.3 1.1 1.07
1.0 1.06
0.5
0
1990 1991 1992 1993 1994 1995 1996 1997 1998
Table 3.3 Three family benefits, total spending and spending as a percentage of GDP, the
Czech Republic, 1996–2000 (CZK, thousands)
Source: Government of the Czech Republic, quoted in Fultz and Steinhilber 2004: 260.
76 Silke Steinhilber
3 and 10 per cent of net family income (Fultz and Steinhilber 2004; Kucharová et al.
2003: 125).
In Poland, by 1998, family benefits were significantly more important for low-
income households, as compared to higher-income households. Family benefits
represented 12 per cent of household income for families in the lowest income
decile, but just 0.15 per cent for those in the top decile. The numbers of beneficia-
ries of employment-related benefits and of recipients of the family allowance
declined drastically, between 30 and 60 per cent, depending on the benefit scheme.
The decline is explained by high unemployment (in the case of employment-related
benefits), stricter benefit entitlements and the sharp decline in birth rates.
While on the whole fewer families receive benefits in Poland, the one benefit
which most directly serves women and their children, the alimony benefit, was also
the one with dramatically increased numbers of beneficiaries throughout the
1990s. Initially, the income-test for the benefit had been waived in 1989, prompt-
ing increased numbers of beneficiaries (Wóycicka et al. 2003). Yet, even after the
reintroduction of income-testing, the numbers of beneficiaries increased, as did
overall spending. Between 2001 and 2002 the fund saw a 6.1 per cent increase of
beneficiaries, and a 16 per cent spending increase (Ministry of Economy, Labour
and Social Policy 2003: 44). While this is an indication of the precarious situation
of single mothers in the country, the alimony fund was abolished in 2004,
sparking a wave of protest among women all over the country.18
Third, evidence suggests a gap between de jure entitlements to family benefits,
and their practical use and take-up (Heinen and Portet 2002; Steinhilber 2003).
There are many accounts of employers discriminating against women workers
with family responsibilities, and there is evidence that women in both countries
do not fully use their rights for fear of reprisals by employers, such as job-loss or
unfavourable reassignment (Fultz and Steinhilber 2004: 259).19
While Poland and the Czech Republic undertook profound social policy reforms
since 1989, gender equality was not a high-priority topic in either country. For the
Czech Republic, Eermáková et al. conclude that at least until the end of 1997 –
when the Department for Equality of Men and Women was established – ‘women’s
issues were the least of governmental and state administration concerns’
(Eermáková et al. 2000: 31). Similarly, Wóycicka et al. argue that the family
policies implemented over the course of the 1990s range from ‘passive to harmful’
in their effect for gender equality in the reconciliation of parental and work
responsibilities (Wóycicka et al. 2003: 221).
Social policy reform debates in both countries did, however, focus widely on the
family and highlight the importance of combating family poverty. The family was
addressed as ‘the most natural social group which should (…) ensure the care for
its members and form their material and intellectual needs from childhood until
old age’ (Prusa 1994, quoted in Eermáková et al. 2000: 42). Since families were
regarded so important and were assigned increased responsibilities in the reform
Gender and Post-socialist Welfare States in Central Eastern Europe 77
process, supporting them to cope was considered eminently important. The gen-
dered nature of family relations, however, and the impact of family benefits on
them did not become an issue for public debate.
The proclaimed family focus of reforms matched popular demand. Before 1989,
support for families had been a constitutive part of state responsibilities, and was
regarded by citizens as at least partial compensation for the inadequacies of the
planned economy. The importance of family networks for day-to-day life in an
economy of scarcity has been highlighted (Veeerník 1995; Veeerník and Matßju
1999; Marksová-Tominová 2003; Seibert 2001; Cichocinska 1993; Giza-Poleszczuk
1992). In continuation, many citizens called for state-support to families as a matter
of social justice after 1989 (Steinhilber 2003; UNDP / Charles University Prague 2003).
In this climate, women’s needs were addressed indirectly and conceptualized in
stereotypical gender terms. Women benefited as members of other social groups,
such as members of low-income households or the unemployed. Echoing the
strong norm of motherhood and the importance of women’s role as caregivers in
both countries in the period after 1945, women today continue to be addressed as
mothers, possibly also as heads of households, but not as individuals or economic
actors (Kucharová et al. 2003: 134).
Changes in entitlement criteria have shifted the status of beneficiaries from
‘holders of personal rights to petitioners of the state’, required to prove their need
(Fultz and Steinhilber 2004: 260). Because of the pervasive practice of means-testing
of benefits and the contraction of the state in social security provision, this effect
is more pronounced in Poland than in the Czech Republic. Yet a renewed debate
about stricter means-testing, for example of the child allowance, in the Czech
Republic in 2004 (Kucharová 2004), illustrates that means-tested benefits are more
vulnerable to changing political conjunctures.
During the reforms, a debate about ‘good motherhood’ re-emerged in both
countries. Notably, good mothers were not referred to as mothers who care for
their children while also being employed. Instead, motherhood was proclaimed as
full-time dedication. In the Czech Republic, income restrictions associated with
the parental allowance exemplify this. While receiving the parental allowance, a
mother should not maintain her links to the labour market. A proposal of the gov-
erning Social Democrats to lift the income restrictions was countered by conserv-
atives who were concerned about Czech women becoming ‘bad mothers’ if the
restrictions were lifted. Women would then focus on making money instead of
caring for children (Marksová-Tominová 2003: 56).
In Poland, the strength of Catholic family values and the traditional image of
the ‘Polish Mother’ have also shaped the debate on ‘motherhood’ and family ben-
efits (Plakwicz 1992; Seibert 2001; Heinen and Portet 2002). Here, religious values,
the church and faith-based organizations play an important role in private and
political life (Inglehart and Norris 2003). Since 1989, abortion and women’s repro-
ductive rights have remained on the political agenda, and religious values have
influenced politics widely across party lines. The Church exerts particularly strong
influence on some areas of social policy, such as abortion, and remains a powerful
advocate for traditional family values and gender roles (Einhorn 1993; Heinen and
78 Silke Steinhilber
Concluding remarks
In the face of similar challenges to construct a market economy and provide social
protection to the citizens, post-socialist Czech Republic and Poland have opted for
different social policy reform paths. These differences have in turn had important
implications for gender equality, a topic which has received only scant, if any,
attention in both countries. During the 1990s Poland moved further than other
post-socialist countries in the direction of a residualist and familial model of a wel-
fare state by cutting, individualizing and familializing benefits. The central state’s
role in social provisioning was reduced massively by shifting responsibilities to the
local administration and individual citizens and their families. The Czech
Republic, in turn, where reforms often were justified in similar ways as in Poland,
has maintained a greater role for the state in the provision of social welfare, and
has upheld important elements of the inherited redistributive welfare structures.
The individualizing and familializing effect of Czech reforms has thus been more
moderate than in Poland.
In the international economic reform climate of the 1990s, characterized by a
clear focus on macroeconomic stability and fiscal austerity, budgetary concerns
have had a strong presence in social policy debates. The Czech Republic has seen
itself under weaker constraints in this respect than the highly indebted Poland. Yet
fiscal pressures as such do not explain where budgets are cut: the analysis of fam-
ily benefit reforms illustrates that economic reform ideologies and political
dynamics and power struggles interlink with family and gender ideologies of
reform actors. In the case of Poland, for example, family benefits suffered greater
losses than other social security benefits.
Despite the differences in reform approaches and their gendered results, family
benefit reforms in both countries show a number of important commonalities.
Most importantly, family benefits in both countries continue to embody a conser-
vative family model. Rarely during the post-1989 reforms were family benefits
referred to by policy-makers as instruments to advance gender equality. Moreover,
the discourse on family benefits has not included ‘non-standard’ families as legit-
imate family forms. Also, no incentives have been developed so far to stimulate
change in men’s gender roles, or to support the de-gendering of care work.
Policy-makers in both countries have upheld a longstanding tradition of seeing
family benefits as instruments to redress declining population growth – a practice
already well in use during socialist times, with no empirically clearly proven effect.
Pronatalism through family benefits, accompanied by strict anti-abortion policies, is
more pronounced in Poland, but is a topic on the rise in the Czech Republic as well.
While the analysis has illustrated the continuities with state socialist family policy,
it also shows a profound shift in the nature of family benefits and in their justifi-
cation since 1989. In the past, family benefits served the dual role of supporting
women in combining paid work and family responsibilities, while also forming
part of the state-sponsored system of income support and redistribution. Despite
women’s continued attachment to the labour market, even more so in the Czech
82 Silke Steinhilber
Republic than in Poland, women today no longer benefit from such a well-
furnished benefit structure. Therefore, today women face greater difficulties in
combining employment and family responsibilities and suffer from a widening
gap between entitlements in law and their practical value.
To different degrees, post-socialist governments have withdrawn from the
tradition of state-organized income-distribution and welfare provision, now
prioritizing the reduction of poverty through means-tested benefits while limiting
state intervention and emphasizing subsidiarity, i.e. the provision of support at
the lowest possible level. In fact, this often means reliance on women’s unpaid
family work. While similar in its direction, this change has been more brisk in
Poland than in the Czech Republic.
As a consequence of reforms, the avenue to access social rights for women has
become more narrow, and women’s need for state support is interpreted differently
today. A few benefits are still linked to employment – yet finding and keeping
employment has become more difficult under conditions of a market-economy.
Entitlement to other key family benefits is based on fulfilling a traditional gender
role as mother and caretaker under precarious economic circumstances. Particularly
those women who do not live in low-income households have therefore lost their
entitlement to benefits. This policy is questionable from the point of view of
gender equality: to enhance gender equality, support for women as a matter of
individual right is as important as support for combining the dual roles in the family
and the world of work. Both need to be accompanied by targeted interventions to
promote a shift in gender roles through policies specifically addressed at men.
Reforms in Poland and the Czech Republic have not moved toward this goal so far.
The legacy of past benefit systems, economic and political shifts since 1989, and
power dynamics between key reform actors and their gendered values and
ideologies constitute a multifaceted web in which gender is being negotiated in
the daily lives of women and men. In the name of ‘women’s interests’ and ‘needs’ –
but without their adequate political representation and participation in the
decision-making process – family benefit reforms have impacted women’s and
men’s employment and life choices. Reforms have created new challenges to
women’s economic independence, public presence and reproductive options,
while reform discourses have invoked a traditional gender structure, often in
contradiction to women’s day-to-day reality of full employment plus unpaid care
responsibilities. The durable conservatism with respect to gender roles and values
in both countries, together with the political marginalization of women, have
come to bear very forcefully on the reform processes since 1989. Despite the many
differences, it is here that both countries show many similarities – and are not very
different from their neighbouring countries, and some (fellow) EU member states.
Acknowledgement
The author is currently supported by the Berlin Programme for Equal Opportunities for
Women in Academia (Berlin Programm zur Förderung der Chancengleichheit für Frauen in
Forschung und Lehre).
Gender and Post-socialist Welfare States in Central Eastern Europe 83
Notes
1 The terms ‘real-existing socialism, ‘real socialism’ and ‘state socialism’ are used here
interchangeably, to distinguish the Central and Eastern European regimes from the
ideals of socialism. See True 2003: 181.
2 For earlier gender analyses of post-socialist social reforms, see, for example, Einhorn
1993, Funk and Mueller 1993, Aslanbeigui et al. 1994, Kligman 1994, Lobodzinska 1995.
3 The ironies of a discourse on greater individual responsibility which in fact hands
women the responsibility of picking up the slack have repeatedly been pointed out by
feminists.
4 See also Chapter 1 of this volume.
5 For a more extensive discussion, see Fodor 2004.
6 Poland ratified ILO Convention 100 on equal pay for equal work and work of equal
value (1953) in 1954, the Czech Republic in 1957.
7 Unless noted otherwise, data before 1993 used here relate to Czechoslovakia, thereafter
to the Czech Republic.
8 On the problems and contradictions within state-socialist social policy, see Castle-
Kanerová 1992, Millard 1992, Deacon 1992.
9 Before 1989, maternity and childcare leave together were regularly termed ‘maternity
leave’ or ‘extended maternity leave’. Here, ‘maternity leave and benefits’ is used in the
sense of ILO Convention 183 as related to the protection of mother and child in relation
to the child’s birth. ‘Childcare leave’/ ‘parental leave’ is granted for a temporary absence
from employment to raise a child. Childcare benefits may or may not be linked to leave.
Conversely, leave may or may not come with a cash benefit.
10 Maternity benefits to insured women existed from 1928, yet the law was not
consistently implemented until after World War II (Pavlik 1985).
11 Short-term sick-child leave.
12 Means-testing of alimony benefits was waïved in 1989 and reinstated in 1999.
13 The guaranteed periodic benefit is payable to a single parent who has exhausted her/his
right to unemployment benefits.
14 The maternity benefit is available to mothers for a maximum of 28 weeks (37 weeks in
case of multiple births). It is insurance-based, paid from sickness insurance, with a
replacement rate of 90 per cent of the net income, up to a ceiling (1990). While
maternity benefits are provided to fathers only under exceptional circumstances (death
or disability of the mother), fathers can receive the parental benefit for full-time care
directly following the birth of a child (maternity leave is not mandatory for women), but
the parental benefit is lower than the maternity benefit.
15 Allowance for the families of soldiers.
16 The following section is based on Fultz and Steinhilber 2004.
17 Note that different levels of aggregation of family income data by household size
complicate cross-country comparisons.
18 Further reforms foresaw some additional support for single mothers, boosting a unified
social assistance benefit.
19 New data from the Czech Republic indicates an increased take-up of parental leave by
women: 60 per cent now return to the labour market only when their child is three.
Worse labour market conditions, an increased interest in full-time care for children at
home, a small increase in fertility (second children) and the shortage of childcare facili-
ties may account for this (V. Kucharová, private communication, June 2005).
20 A much-quoted example is the strict anti-abortion legislation which was not relaxed
under the left-wing government either.
21 Restrictions about childcare use have been maintained. Only from 2006 children can
attend kindergarten for four hours per day without any impact on the parental allowance.
22 Periods of right-wing governments also contributed to the consolidation of conservative
and Catholic women’s organizations (M. Fuszara, personal conversation June 2005).
84 Silke Steinhilber
Bibliography
Aslanbeigui, Nahid, Steven Pressman and Gale Summerfield (1994) Women in the Age of
Economic Transformation. Gender Impact of Reforms in Post-socialist and Developing Countries.
New York and London: Routledge.
Castle-Kanerová, Mita (1992) ‘Social Policy in Czechoslovakia’. In B. Deacon (ed.) The New
Eastern Europe: Social Policy Past, Present and Future. London: Sage.
Eermáková, Maria, Hana Hasková, Alena Krizková, Marcela Linková, Hana Mariková and
Martina Musilová (2000) Relations and Changes of Gender Differences in the Czech Society in
the 90s. Prague: Institute of Sociology of Academy of Science of the Czech Republic.
Chlon-Dominczak, Agnieszka (2004) ‘Gender Dimensions of Social Security Reforms in
Poland’. Presentation at the seminar, The Gender Dimensions of Social Security Reform in
Central and Eastern Europe: Case Studies of Poland, Hungary and the Czech Republic.
organized by the International Labour Office and Central European University, Budapest,
23 February 2004.
Cichocinska, M. (1993) ‘Gender Aspects of Dismantling the Command Economy in Eastern
Europe: The Case of Poland’. In V. Moghadam (ed.) Democratic Reform and the Position of
Women in Transitional Economies. Oxford: Clarendon Press.
Deacon, Bob (1992) ‘Eastern European Welfare: Past, Present and Future in Comparative
Context’. In B. Deacon et al. (eds) The New Eastern Europe. Social Policy Past, Present and
Future. London: Sage.
Einhorn, Barbara (1993) Cinderella Goes to Market. New York: Verso.
European Commission / Government of Poland. (2001) Joint Assessment of Employment
Priorities in Poland. http://europa.eu.int/comm/employment_social/employment_
analysis/japs/poland_en.pdf, accessed on 24 April 2005.
Fajth, Gaspar (1994) Family Support Policies in Transitional Economies: Challenges and
Constraints. Innocenti Occasional Papers, Economic Policy Series, No. 43. Florence: UNICEF.
Fodor, Eva (2004) ‘Women at Work: The Status of Women in the Labor Markets of the Czech
Republic, Hungary, and Poland’. Occasional Paper written for the preparation of the report
Gender Equality: Striving for Justice in an Unequal World. Geneva: UNRISD.
Fultz, Elaine and Silke Steinhilber (2003) ‘The Gender Dimensions of Social Security Reforms
in Central and Eastern Europe’. In E. Fultz, M. Ruck and S. Steinhilber (eds) The Gender
Dimensions of Social Security Reforms in Central and Eastern Europe. Case Studies of the Czech
Republic, Hungary and Poland. Budapest: ILO, Subregional Office for Central and Eastern
Europe.
Fultz, Elaine and Silke Steinhilber (2004) ‘Social Security Reform and Gender Equality: Recent
Experience in Central Europe’. International Labour Review, Vol. 143, No. 3: 249–73.
Funk, Nanette and Magda Mueller (eds) (1993) Gender Politics and Post-Communism.
Reflections from Eastern Europe and the Former Soviet Union. New York and London:
Routledge.
Gal, Susan and Gail Kligman (2000) The Politics of Gender after Socialism. Princeton NJ:
Princteon University Press.
Giza-Poleszczuk, A. (1992) ‘Familism and System Transformation’. Polish Sociological Bulletin,
Vol. 3, No. 4: 315–27.
Heinen, Jacqueline and A. Matuchniak-Krasuska (1995) ‘Abortion in Poland: A Vicious Circle
or a Good Use of Rhetoric’. Women’s Studies International Forum, Vol. 18, No. 1: 27–33.
Gender and Post-socialist Welfare States in Central Eastern Europe 85
Heinen, Jacqueline and Stephane Portet (2002) ‘Political and Social Citizenship: An
Examination of the Case of Poland’. In M. Molyneux and S. Razavi (eds) Gender Justice,
Development and Rights. Oxford: Oxford University Press.
Ingelhart, Ronald and Pipa Norris (2003) Rising Tide: Gender Equality and Cultural Change
around the World. Cambridge: Cambridge University Press.
International Labour Office (ILO) (1984) ‘Family Benefit’. In Introduction to Social Security,
Chapter 11. Geneva, Switzerland, 1984.
International Labour Office (ILO) (1999) Information communication by the ILO national
correspondent in Poland on the Bill on Equal Status of women and men and general pre-
sentation of the situation of women. Internal Communication. Warsaw, 22 April 1999.
Kligman, Gail (1994) ‘The Social Legacy of Communism: Women, Children, and the
Feminization of Poverty’. In J. Millar and S. Wolchik (eds) The Social Legacy of Communism.
Washington DC and New York: Woodrow Wilson Center and Cambridge University Press.
Kucharová, Vera (2004) ‘Gender Dimensions of Social Security Reforms in the Czech
Republic’. Presentation at the seminar, The Gender Dimensions of Social Security Reform
in Central and Eastern Europe: Case Studies of Poland, Hungary and the Czech Republic.
Organized by International Labor Office and Central European University. Budapest,
23 February 2004.
Kucharová, Vera, M. Kotynková and Ladislav Prusa (2003) ‘The Gender Dimensions of Social
Security Reform in the Czech Republic’. In E. Fultz, M. Ruck and S. Steinhilber (eds) The
Gender Dimensions of Social Security Reforms in Central and Eastern Europe. Case Studies of the
Czech Republic, Hungary and Poland. Budapest: ILO, Subregional Office for Central and
Eastern Europe.
Lavigne, Marie (1999) The Economics of Transition – From Socialist Economy to Market Economy.
New York: St. Martin Press.
Lewis, Jane (ed.) (1993) Women and Social Policies in Europe. Work, Family and the State.
Aldershot Edward Elgar.
Lewis, Jane (1997) ‘Gender and Welfare Regimes: Further Thoughts’. Social Politics:
International Studies in Gender, State and Society, Vol. 4, No. 2: 160–77.
Lobodzinska, Barbara (1995) Family, Women and Employment in Central-Eastern Europe.
Westport, CN: Greenwood Press.
Lohmann, Kinga and Anita Seibert (eds) (2003) Gender Assessment of the Impact of EU
Accession on the Status of Women in the Labour Market in CEE. National Study: Poland.
Warsaw: Karat Coalition.
Marksová-Tominová, Michaela (ed.) (2003) Gender Assessment of the Impact of EU Accession on
the Status of Women in the Labour Market in CEE. National Study: Czech Republic. Karat
Coalition. Prague: Gender Studies Center.
Marody, M. and A. Giza-Poleszczuk (2000) ‘Changing Images of Identity in Poland: From the
Self-Sacrificing to the Self-Investing Woman?’ In S. Gal and G. Kligman (eds) Reproducing
Gender: Politics, Publics, and Everyday Life after Socialism. Princeton NJ: Princeton University
Press.
Marx Ferree, Myra (1993) ‘The Rise and Fall of ‘Mommy Politics’: Feminism and Unification
in (East) Germany’. Feminist Studies, Vol. 19, No. 1: 89–115.
Millard, F. (1992) ‘Social Policy in Poland’. In B. Deacon (ed.), The New Eastern Europe: Social
Policy Past, Present and Future. London: Sage.
Ministry of Economy, Labour and Social Policy (2003) Poland 2003: Report on Labour Market
and Social Security. Warsaw: Ministry of Economy, Labour and Social Policy.
Offe, Claus (1993) ‘The Politics of Social Policy in East European Transitions. Antecedents,
Agents, and Agenda of Reform’. ZeS, Arbeitspapier No. 2/93, Bremen.
Open Society Institute (OSI) Network Women’s Program (2005a) Your Sex Matters in Poland.
Equal Opportunities for Women and Men. Monitoring Law and Practice in New Member
States and Accession Countries of the European Union. Fact Sheet. http://www.
soros.org/initiatives/women/articles_publications/publications/equal_20050502/
poland.pdf, accessed on 1 July 2005.
86 Silke Steinhilber
Open Society Institute (OSI) Network Women’s Program (2005b) Your Sex Matters in the Czech
Republic. Equal Opportunities for Women and Men. Monitoring Law and Practice in New
Member States and Accession Countries of the European Union. Fact Sheet.
http://www.soros.org/initiatives/women/articles_publications/publications/equal_200505
02/czechrep.pdf, accessed on 1 July 2005.
Orenstein, Mitchell (2001) Out of the Red: Building Capitalism and Democracy in Postcommunist
Europe. Ann Arbor: University of Michigan Press.
Pavlik, Z. (1985) ‘Czechoslovakia’. In V. Bodrova and R. Anker (eds) Working Women in
Socialist Countries: The Fertility Connection, Geneva: ILO.
Plakwicz, J. (1992) ‘Between the Church and the State: Polish Women’s Experience’. In
C. Corrin, Super Women and the Double Burden. London: Scarlet Press.
Regulska, Joanna (1998) ‘The ‘Political’ and its Meaning for Women: Transition Politics in
Poland’. In J. Pickles and A. Smith (eds) Theorizing Transition: The Political Economy of Post-
Communist Transformation. New York: Routledge.
Research Institute for Labour and Social Affairs (RILSA)/ United Nations Development
Programme (UNDP) (1999) Human Development Report: Czech Republic 1999. Prague: RILSA.
Sainsbury, Diane (ed.) (1999) Gender and Welfare State Regimes. Oxford: Oxford University
Press.
Seibert, Anita (2001) ‘From Matka Polka to New Polish Woman: Women and Restructuring in
Poland’. Unpublished PhD thesis, Monash University, Clayton, Victoria, Australia.
Siemienska, Renata (1983) ‘Women and the Family in Poland’. In E. Lupri (ed.) The
Challenging Position of Women in Family and Society. Leiden: E. J. Brill.
Steinhilber, Silke (2003) ‘Women’s Views on Social Security Reform: Qualitative Survey’. In
E. Fultz, M. Ruck and S. Steinhilber (eds) The Gender Dimensions of Social Security Reforms in
Central and Eastern Europe: Case Studies of the Czech Republic, Hungary and Poland. Budapest:
ILO. Subregional Office for Central and Eastern Europe.
True, Jacqui (2003) Gender, Globalization and Postsocialism: The Czech Republic after
Communism. New York: Columbia University Press.
United Nations Children’s Fund/Monitoring Eastern Europe Project (UNICEF/MONEE)
(1999) Women in Transition. Florence: UNICEF/ MONEE.
United Nations Development Programme (UNDP)/ Charles University Prague (2003) Human
Development Report. Where Do We Come From, What Are We, and Where Are We Going?
Prague: UNDP and Charles University.
Veeerník, Jiri (1995) ‘Household Incomes and Social Policies: The Czech Republic in the
Period 1989–1995’. SOCO Project Paper No. 25. Vienna.
Veeerník, Jiri and Matfju, Petr (1999) Ten Years of Rebuilding Capitalism: Czech Society After
1989. Prague: Academia.
Wolchik, S. and A. Meyer (eds) (1985) Women, State and Party in Eastern Europe. Durham NC:
Duke University Press.
Wóycicka, Irena, Bozena Balcerzak-Paradowska, Agnieszka Chlon-Dominczak, Irena
Kotowska, A. Olejniczuk-Merta and Irena Topinska (2003) ‘The Gender Dimensions of
Social Security Reform in Poland’. In E. Fultz, M. Ruck and S. Steinhilber (eds) The Gender
Dimensions of Social Security Reforms in Central and Eastern Europe: Case Studies of the Czech
Republic, Hungary and Poland. Budapest: ILO, Subregional Office for Central and Eastern
Europe.
Zielinska, E. (2000) ‘Between Ideology, Politics, and Common Sense: The Discourse of
Reproductive Rights in Poland’. In S. Gal and G. Kligman (eds) Reproducing Gender: Politics,
Publics, and Everyday Life after Socialism. Princeton NJ: Princeton University Press.
4
Maternalist Policies versus
Women’s Economic Citizenship?
Gendered Social Policy in Iran
Valentine M. Moghadam
That social policy is gendered is now a truism in the feminist literature. But what
are the different patterns of (gendered) social policy formation and implementa-
tion across regions in the world economy? In the Middle East, for example, what
are the state policies, practices and institutions that directly influence the welfare
and security of various groups within a particular society, especially as far as
women are concerned? And if social policy is partly about protecting against risks
and contingencies while also providing social equity, do all social policies serve to
expand women’s citizenship? In this chapter I examine the gendered nature of
social policy in Iran, with a focus on the problems associated with women’s
economic citizenship that have resulted from Iran’s economic structure and its
cultural/ideological institutions. An oil-dependent economy and gender relations
codified in Muslim family law have had implications for women’s access to
employment and economic resources, as well as overall citizenship.
The literature on social policy is overwhelmingly concentrated on the experi-
ences and models of the welfare state in advanced capitalist countries, and in par-
ticular Europe and North America. The feminist component of this literature
consists largely of historical investigations of maternalist politics and policies in
the United States,1 or comparative studies of gender and welfare regimes in the
core countries,2 or studies of the distribution of care and the problem of employed
women’s double burden.3 Gosta Esping-Andersen’s typology of the three worlds of
welfare capitalism and T. H. Marshall’s theory of the evolution of civil, political,
and social rights of citizenship often conceptually frame these investigations.4 The
emerging literature on social policy in a development context suggests that East
Asia, among other regions, might represent a different experience and model
(Kwon 2005). The present chapter expands the empirical understanding of
gendered social policy in a development context, while also proposing a distinc-
tive experience and model based on the characteristics and consequences of an
oil-dependent economy and Muslim family law in one Middle Eastern country.
Elsewhere I have argued that in contrast to the rights-based model of the Nordic
welfare state or the developmentalist East Asian experience, social welfare
87
88 Valentine M. Moghadam
provisioning in the post-colonial Middle East and North Africa (MENA) was tied to
the imperatives of state-building and legitimation, was financed by the regional
oil economy, and reflected and reinforced a ‘patriarchal gender contract’ premised
on women’s family roles.5 The MENA state’s ‘developmentalism’ was undermined
by its ‘neopatriarchal’ approach to women, gender, and the family, codified in
family laws/personal status codes that define women largely in terms of their fil-
ial, marital and maternal roles, place them under male guardianship, and deny
them equality in access to family wealth. Although state expansion and social
expenditures did allow for social mobility and access by some women to education
and employment, on the whole oil wealth and family laws served to prevent
women’s ‘labour commodification’ in the MENA region, in contrast to other
regions in the world economy. Through most of the period between 1950 and
1990, the MENA model allowed for generous if non-inclusive social provisioning
and high wages, and provided the (rather small) salariat with generous benefits. But
in the subsequent period, the family wage was undermined by economic crisis and
then by the neoliberal prescription, while the reinforcement of Muslim family law
became the focus of Islamist groups (Moghadam 1998; 2003, ch. 2).
These are early days in the study of social policy in the Middle East. The grow-
ing literature on poverty, the limited work on pensions and social security reform
in Arab countries, and the more extensive literature on socio-economic develop-
ment lack attention to the gender dynamics of social policies.6 The focus of the
emerging literature on women’s citizenship in MENA is on the problems of
women’s civil and political rights of citizenship, and deals only minimally with
social rights associated with employment, or with economic citizenship.7 On the
other hand, family law is a social policy, and a highly gendered one that has
implications for women’s access to employment and income, although the large
and important literature on family law has not explored the links between family
law, women’s economic activity, and social development.8
In the sections that follow, I begin by elaborating on this chapter’s theoretical
framework, including the concepts of economic citizenship, maternalist politics,
the rentier and neopatriarchal state, Muslim family law, and the patriarchal gender
contract. I then turn to the gender/social policy regime in Iran, starting with a
brief historical overview and drawing attention to continuities and discontinuities
before and after the 1979 Islamic revolution. I argue that the state remained
rentier and neopatriarchal after the revolution, but a key difference was the
strengthening of Muslim family law and a state-driven maternalist discourse and
politics. The paper ends with reflections on the future of Iran’s gender/social
policy regime.
Conceptual overview
The central premise of a feminist approach to social policy is that the sexual division
of labour – which is influenced by the organization of economic production –
shapes policy decisions and results in different outcomes for women and
men. Marxist-feminist research has shown how women’s and men’s differential
roles in production and reproduction, and especially women’s greater responsibilities
Maternalist Policies versus Women’s Economic Citizenship? 89
the Islamic Republic of Iran in the 1980s and 1990s, where women sought
to improve their legal status and social positions within the acceptable discursive
frameworks of revolutionary motherhood and religious duty (see Gheytanchi
2001). Maternalist policies refer to mother and child welfare policies that emerged
in the late nineteenth and early twentieth centuries, especially in the United States
and Britain, and which channelled social benefits directly to women as mothers.
There is some consensus that American middle-class women reformers – cut off as
they were from institutionalized politics – used the language of motherhood to
legitimize their political activism and to advocate better conditions and policies
for poor, working-class, or immigrant women. There is less consensus as to
whether these policies were ‘feminist’ or if they served to reinforce the patriarchal
social order. Some studies show that mother–child policies were palliative, or
temporary, or intended to strengthen rather than undermine the patriarchal family
(Lewis 1980; Ladd-Taylor 1994). Skocpol (1992) has argued that maternalist social
policies – which she defines as policies by women for women – characterized the
early welfare state in the United States.9 Koven and Michel (1993) distinguish
between radical and conservative maternalists, and demonstrate that maternalism
is a political discourse that is sometimes ‘a cloak for paternalism’ and sometimes
deployed by women for women. They write:
We apply the term to ideologies that exalted women’s capacity to mother and
applied to society as a whole the values they attached to that role: care, nurtu-
rance, and morality. Maternalism was the central and defining core of some
women’s vision of themselves and of politics.
Koven and Michel 1993: 4
They show, too, that at the heart of maternalism lies the paradox of entering the
public political arena by reinforcing the traditional female sphere of children,
family, nurturance and care.10
These feminist insights – along with a political economy approach – frame the
present study of gender and social policy in Iran, as well as the Middle East as a
whole. The regional oil economy combined with Muslim family law produced a
patriarchal gender contract that rendered the majority of MENA women economic
dependents. The dominant gender ideology has held that men are the providers of
women and children, and women are primarily mothers and wives. Until rela-
tively recently – that is, until the post-oil-boom downturn – the family wage and
the family law reinforced this situation. The beneficiaries of social security systems
thus have been largely men and their dependants; women’s welfare has been
presumed to be attained within the family or through the provisions of Muslim
family law; and the vast majority of Middle Eastern women have relied on family
support systems for protection vis-à-vis risk, contingencies and disasters.
This has been the case also in Iran, where, especially after the Islamic revolution
of 1979, women’s family roles were emphasized and female labour-force participa-
tion declined. In the late 1980s, Islamic women bureaucrats began to campaign for
improvements in women’s legal status and social conditions, and they did so
within a conservative religious/maternalist discursive frame. Accordingly, the
92 Valentine M. Moghadam
Islamic Republic of Iran adopted ‘maternalist’ policies for women that were
predicated on the patriarchal gender contract rather than on feminist concepts of
individual and collective women’s rights.
In the more recent neoliberal era, non-governmental organizations of all types,
including women’s organizations, have taken on an increasingly important role in
Iran, as elsewhere in the world economy. NGOs appear to have a three-fold role:
(a) they fulfil certain neoliberal requirements in the context of the withdrawal of
the state in welfare provisioning;
(b) they are part of the democratization process, or political liberalization, and
help to constitute the emerging civil societies across the region; and
(c) they play a key role in the women’s movement, with cultural, political, and
economic/employment implications for women and for gender relations.
features of family law are similar across Muslim countries, as they commonly
reflect, and reinforce, a patrilineal logic.
The Islamic marriage contract requires the consent of the bride, and in some
countries women may insert stipulations into the contract, such as the condition
that she be the only wife (An-Naim 2002). However, marriage gives the husband
the right of access to his wife’s body, and marital rape is not recognized
(Welchman 2001). Children acquire citizenship and religious status through their
fathers, not their mothers. Muslim women are not permitted to marry non-
Muslim men, and non-Muslim widows cannot inherit from Muslim husbands.
A key aspect of Muslim family law is that women are required to obtain the per-
mission of father, husband, or other male guardian to marry, seek employment,
start a business, or travel. This is the source of a common legal inconsistency in the
Middle East, because placing women under the guardianship of a male contradicts
constitutional guarantees of equality of all citizens and labour laws which ensure
rights for men and women workers. As Pateman (1988) argued in relation to the
‘sexual contract’ in early modern Europe, Muslim family law denies daughters and
wives their full civil rights.
The unequal inheritance aspect of Muslim family law and civil codes compro-
mises women’s economic independence but is a sensitive issue; it is seen on one
level as a divine imperative revealed in the Koran and on another level as an impor-
tant part of the patriarchal gender contract whereby women are provided for by
their fathers, husbands or brothers. Islamic law gives women the right to own and
dispose of property, but they inherit less property than men do. Sons inherit twice
as much as daughters, but they are also expected to look after their parents in old
age. In a polygamous context, a deceased man’s inheritance and his pension are
divided among his widows, children and other relatives that he may have been sup-
porting; this could leave the widows with insignificant pensions. Thus, even though
Islamic norms require that fathers and husbands financially support their daughters
and wives, it is also the case that divorced, widowed or abandoned women without
access to jobs or a steady source of income, especially among the low-income social
groups, are often left in a state of impoverishment. Indeed, Loewe (2000) found that
among Egypt’s poorest are divorced, abandoned and widowed women.
In Iran (and elsewhere), a husband has the legal right to forbid his wife or
unmarried daughter to seek employment or continue in a job. As such, Muslim
family law contravenes the ICESCR, which is the main framework of economic
citizenship. The ICESCR prescribes the right of people to a freely chosen job;
equitable and equal wages for work of equal value; dignified working conditions
for workers and their families; professional training; equal opportunities for
promotion; protection for families, especially for children; maternity protection;
protection of boys, girls and teenagers against economic exploitation. Inasmuch as
Muslim family law and norms in Iran and elsewhere prevent women from apply-
ing for or staying in a job without the permission of father or husband, and in
some countries certain occupations and professions are off-limits to women (such
as the profession of judge in Iran), this denies women the right to enjoy the
ICESCR’s provision for ‘a freely chosen job’.
94 Valentine M. Moghadam
The mahr – or dower – is a sum of money promised by the groom to the wife
before marriage. Typically, the wife receives a portion of it at the time of marriage,
and the balance is deferred. But whereas this practice was initially meant to pro-
vide a sort of social insurance for the wife in the case of divorce or widowhood, in
the modern era it symbolizes women’s economic dependence.12 Moreover, legal
statutes, based on Islamic law, assert that men owe their wives material support or
maintenance and cannot command them to contribute to the family economy.
Indeed, one of the grounds for which women can legitimately petition for divorce
is non-support by the husband. This legal requirement, too, may have been salu-
tary in early Islamic history, but in the modern era it has functioned to perpetuate
the patriarchal gender contract.
In some ways, Muslim family law may be seen not only as a pre-modern or pre-
feminist code for the regulation of family relations, but also as a way of retaining
family support systems in the place of a fully functioning welfare state predicated
on concepts of citizen contributions and entitlements. The welfare of wives and
children remains the responsibility of the father/husband. When a woman seeks a
divorce or is divorced, her maintenance comes not in the form of transfers from
the state but in the form of the mahr that is owed to her by her husband, or (in the
Islamic Republic of Iran) the ojrat-ol-mesl, which is the monetary value of the
domestic work she has performed over the years. I discuss this new element of
Iran’s gender/social policy regime in the section below, but let us first examine the
historical context.
Women have a long history of economic activity in Iran, particularly in the carpet-
weaving and textiles sectors. In the late Qajar period (the turn of the twentieth
century), women workers were largely concentrated in home-based and small
workshop enterprises. Iran joined the International Labour Organization in 1914,
and in 1920–21 the ILO informed the government of its concern over working
conditions in the carpet industry in Kerman. This led to a series of decrees
designed to improve working conditions with respect to length of the work-day,
minimum age of employment for girls and boys, safety, sanitation and health.
According to Floor (1985: 88), two articles of the decree of 17 December 1923 read
in part: ‘boys and girls shall work in separate workshops; mixed workshops are
absolutely forbidden’, and ‘foremen (those who dictate the pattern to the workers)
shall not enter the girls’ workshop, where forewomen shall be employed.’ Iran’s
Civil Pension Fund was established in 1923, although there are no data to deter-
mine how many women, if any, were enrolled at that time. In the 1930s, under
Reza Shah, women’s social participation expanded as more schooling was made
available and the growing state sector absorbed some of the educated female
labour force (Moghadam 2000). In other areas, however, women remained
disadvantaged, as they could not vote and were subordinated within the family.
Moreover, they were all but excluded from Iran’s rapidly expanding oil industry,
which at the time was not only a labour-intensive industry but also completely
male-dominated.
Maternalist Policies versus Women’s Economic Citizenship? 95
After Reza Shah’s forced abdication in 1941, the coalition government of Ahmad
Qavam sought to implement a wide-ranging social programme. In 1946 it
promised distribution of state lands, women’s suffrage, provincial assemblies,
elimination of unemployment, construction of rural clinics, schools and irrigation
projects. Qavam established a Supreme Economic Council that was charged with
implementing a Five Year Plan, setting of minimum wage, protecting national
industry, drafting a plan to distribute crown lands, and supporting the peasantry.
He recognized trade unions and appointed a Supreme Labour Council to introduce
unemployment insurance, wage scales, and negotiating committees between man-
agement and labour. His government, however, did not survive to pursue these
radical policies (Messkoub 2005). Thus by the mid-1960s illiteracy rates remained
high in Iran: just half of the urban population was literate compared with only
15 per cent of rural population. The female literacy rate of 17 per cent was less
than half that of men (40 per cent). The situation improved somewhat by the early
1970s – after nearly a decade of a rural literacy campaign spearheaded by the
Literacy Corps.
During this period, Iran enjoyed unprecedented economic growth, largely as a
result of domestic investments fuelled by rising oil prices. Oil revenues accrued to
the state, and its ever-growing budget paid for development projects, prestige
projects (such as expensive monuments and royal celebrations), military procure-
ments, infrastructural development and social programs. Skilled workers and
high-level manpower were imported from South Korea and the United States,
among other countries, but Iran had its own large workforce to man the oil indus-
try, manufacturing and the booming construction sector. Oil revenues allowed for
high wages and low taxes. Karshenas has studied manufacturing wage levels
comparatively and has established that wages were higher in the Middle East,
including Iran, than in southeast Asia and Mexico. At the same time, income and
other taxes were among the lowest in the world (Karshenas 1997, 2001). High wages
and generous social policies for government employees – financed not by income
taxes or employee contributions but by oil revenues and other rents – also were
meant to strengthen and expand the state, establish national integration, and build
a social base of support for the state.
As the Shah of Iran sought to establish national unity and mobilize support for
his rule, the government established an array of social policies and welfare benefits.
Government employees received pensions, subsidized staples, and interest-free
loans for housing; a workers’ shareholding program applied to those in large
establishments; the urban poor could use free schools, free health clinics, and
cheap food; and peasants benefited from the land reform, a literacy campaign and
a healthcare campaign. For women, Shah Mohammad Reza Pahlavi extended the
vote to women and modernized the family law – largely due to advocacy by a group
of elite women in and around the state or with connections to the royal family.
subsidies (utilities, petrol, bread), and paid maternity leaves for women
professionals. Various state enterprises and ministries had their own policies and
benefits packages. The most generous were offered by the armed forces and the
National Iranian Oil Company, while the Ministry of Roads had a hospital and
seaside resort for its employees.13 Social security benefits (sickness, disability,
death, pensions) were available to all regular employees, including women. State-
owned factories employing a certain number of women (usually 50 or 100) were
required to establish nurseries, and working mothers were also entitled to nursing
breaks. Social welfare services for the poor, including poor women, were chan-
nelled in part through institutions such as the Pahlavi Foundation and the
Women’s Organization of Iran.
During the 1960s and 1970s, the female labour force grew, and by 1976 women
comprised 12–14 per cent of the total labour force. Women were found in the civil
service (largely in teaching, healthcare and social services, with some participation
also in public administration), in large-scale factory employment and of course in
traditional small-scale activities. Though the oil and petrochemicals industries
were male-dominated, a small proportion of women worked in clerical positions
in the National Iranian Oil Company. Those in the modern sector were beneficia-
ries of labour legislation and social insurance. In the 1970s women retired on
average two to three years before men but with similar annual pensions; women
comprised around 15 per cent of all pension recipients (Messkoub 2005).
Class, urban, and gender biases were inscribed in the social policies, and
especially in the social security policies and provisions of labour law. For the most
part, the principal beneficiaries were men, members of the urban middle class or
the ‘labour aristocracy’. Social security and labour legislation did not extend to
domestic workers/servants, irregular or casual workers, unpaid family workers, or
peasants, rural workers and small farmers. Given the size of the rural population in
Iran, this meant that the majority of citizens relied on family support systems – a
situation that helped to reproduce the patriarchal family through large, extended
families with a preference for sons, and the unpaid labour of female members.
However, this also was a time when elite women in Iran began to organize and
mobilize for improvements in women’s social conditions and their legal status.
Much of that work was carried out by the Women’s Organization of Iran (WOI). By
the late 1970s, the WOI had over 400 branches and 120 women’s centres through-
out the country that offered daycare centres, legal literacy, counselling, family
planning, health, nutrition, and vocational training (Afkhami 2003; Afary 2005).
The Family Protection Law of 1967 and 1975 provided unprecedented rights for
Iranian women in the areas of divorce and child custody; for example, a wife’s
permission was required should her husband seek a second wife. However, a man
still had the right to forbid his wife from working outside the home, if he found
her job ‘dishonourable’; and she could not travel outside the country without her
husband’s written permission.
Given the huge increases in the national income in the 1970s, more invest-
ments should have been made in creating a literate, educated and skilled labour
force, both male and female. Public services did expand to a large extent, but Iran
Maternalist Policies versus Women’s Economic Citizenship? 97
did not achieve full literacy or eradicate child and maternal mortality. The 1979
revolution also set back socioeconomic gains by at least a decade. Thus by 1990,
Iran’s educational attainment rates were far below those of other countries at com-
parable levels of development or income. In particular, illiteracy was high among
adult women, and gender gaps were very wide. The mean years of schooling of
adult men and women (aged 25 and over) was 4.6 and 3.1, respectively. But in far
poorer Vietnam, the corresponding figures were 5.8 and 3.4 years. The Philippines
did much better, at 7.8 years for men and 7.0 for women.14
(a) poor economic conditions, inadequate domestic and foreign investments, and
subsequently low levels of job creation;
(b) a preference for investments in capital-intensive, male-intensive sectors such
as oil, gas and nuclear energy;
(c) a constitutional clause extolling the virtues of motherhood; and
(d) the ubiquitous Islamic criteria, including the fact that under the Islamic
Republic’s Sharia-based provisions on women and the family, women cannot
seek jobs without the approval of their fathers or husbands.
of all manufacturing jobs (formal and home-based alike) was 14 per cent and their
overall share of paid employment was 12 per cent (Moghadam 2003, chs. 2, 7).
After the revolution, a massive ideological campaign was launched to tie women
to their family roles, and the new Islamic family law restored men’s rights to
polygamy, unilateral divorce, custody of children after divorce, and control over
their wives’ mobility. In matters of inheritance, women became severely disad-
vantaged. For example, a man could inherit all of his wife’s wealth, but she was
entitled only to one-fourth (if he had no child) or one-eighth (if he had children)
of his movable property and of the value of his estate. This pertains to cases of
permanent marriage; partners in a temporary marriage receive no inheritance.15 In
polygamous marriages, wives must divide among themselves the allotted one-
fourth or one-eighth. Article 1117 of the Civil Code stipulates that a man has the
right to prevent his wife from taking employment ‘if he deems such employment
would be at variance with their family interests and values’.
Although women’s educational attainment rates have risen steadily, the result of
both Iran’s political economy and its cultural/ideological institutions has been low
levels of female salaried employment, high rates of inactivity, disproportionately
high female unemployment, and a huge gender-based income gap. Iran’s labour
force has been predominantly masculine. The burgeoning private sector remains a
largely male domain, with work conditions that are not appealing to most women.
Gender bias has been rampant in employment and in the country’s social insurance
programs. Even though occupational restrictions were lifted, the profession of
judge remains exclusively male. Male and female workers alike have paid 7 per cent
of their wages toward their retirement insurance, but in the event of a woman’s
death, her family could not collect retirement benefits. All working women are
required to appear in hejab, which is at minimum a large scarf covering all the hair
and neck, and a long-sleeved and loose-fitting smock or manteau that covers the
body’s contours. In some government agencies, more conservative attire (chador)
is required of the women employees. Of course, professional women in the civil
service have been beneficiaries of social policies, and the time of writing they rep-
resented about 36 per cent of Iran’s total civil service employees. However, they
constitute only a small proportion of the total female economically active popula-
tion, and though they constitute only 12 per cent of the total salaried labour
force, their unemployment rate in 1996 was 12.5 per cent (IRI 1997; 1999: 99).16
Clearly, Iranian women’s economic citizenship is seriously circumscribed.
This may be one reason for the growth of women’s NGOs in Iran. When job
opportunities in the formal sector are foreclosed, women form their own businesses
or associations. This is an area that remains under-researched in Iran, although
Povey (2004) has undertaken some preliminary work on women’s NGOs. Women’s
NGOs include publishing houses, environmental protection groups, research
centres rural women’s centres, charitable agencies and professional associations.
The women’s press in Iran has been proliferating, and consists of magazines and
women’s studies journals such as Zanan, Hoghough-e Zanan and Farzaneh, and a
publishing house, Roshangaran Press. The Women’s Cultural Centre has produced
calendars, compendiums and a feminist journal, and maintains a website.
Maternalist Policies versus Women’s Economic Citizenship? 99
The WID-oriented NGOs engage in advocacy for enhanced women’s rights, sometimes
by using the Islamic Republic’s own maternalist discourse against it (see discussion
below on ojrat-ol-mesl). Forming women’s NGOs may be seen, therefore, as a strat-
egy to counter women’s economic dependence and high unemployment rates, as
a way of enhancing women’s social participation, and as a form of advocacy for
women.
Lack of job opportunities also may be behind another phenomenon, similarly
under-researched and largely anecdotal at this time – the expansion of the practice
mehrieh, which had been in decline in the final years of the Pahlavi era, especially
among the educated, westernized elite. By all accounts, mehrieh has become more
rather than less entrenched in Iranian society. The Islamic state’s ideological pressure
is surely one reason, but what I am suggesting here is that young Iranian women
and their parents may very well be insisting on mehrieh due to economic and
employment uncertainties.
commentator has explained, with approval and pride: the Iranian legal system
does not obligate women to do any household activity. Even the wife’s nursing
of her own child ‘is not one of her unwavering responsibilities’. He states, citing
Article 1176 of the Civil Code: ‘The mother is not obliged to suckle her own
baby unless the baby cannot be fed by any other means’ (Mehrpour 1995: 60).
Here the commentator appears unaware of the larger implication of such a
‘right’ on the part of the wife and mother – which is to exempt her from any
labour whatsoever other than sexual services to her husband and childbearing
and therefore to deny her any economic citizenship. In the event of divorce, he
continues,
when the wife is free from any blame or fault, [the husband] is obliged to pay a
lump sum to his wife, which payment is in addition to his obligations on mar-
riage portion [mehrieh], recompense of the wife’s past services in married life
[ojrat-ol-mesl, of which more, below], and alimony payment during the speci-
fied time after divorce when she is religiously prohibited from a new marriage
Mehrpour 1995: 61; my emphasis
The Islamic Republic may have exalted mothers in its official discourse, but the
Civil Code stipulates a fractional share of a deceased husband’s wealth for the wife.
The stated justifications for women’s lesser shares are ‘the marriage portion and
provisions of women’s living expenses by the husband’ (Mehrpour 1995: 58).
The Civil Code also cites the Qur’anic verse ‘God has ordained that amongst the
children, the son’s share is double the daughter’s’ to establish unequal inheritance
for sons and daughters. Nonetheless, advocates of the Islamic Republic’s legal and
social policy frameworks insist that
men and women are entitled to equal rights in possessing and executing the
privileges of their legal capacity, and as such there is no distinction between men
and women. All the means and tools by which men can obtain economic
resources are legally at the disposal of women as well. … The only limitations on
the employment of women concern jobs which defy family values and interests.
Mehrpour 1995: 61
Many feminists, however, have noted that despite its ‘motherist’ discourse, the
Islamic Republic of Iran has in practice devalued motherhood through its policies
and laws. Lawyer Mehrangiz Kaar has highlighted the contradiction between the
Islamic state’s formal praise for mothers and Iranian mothers’ lack of legal protection,
such as the right to custody of their children, the right to open a bank account for
their children, and control over their own bodies (Kaar 1996). Her articles in the
women’s magazine Zanan – which came to exemplify the Islamic feminist per-
spective in the 1990s – concentrated on women’s real status under Iranian civil law
and family law and called on women to engage in ijtihad (independent reasoning
and interpretation) to claim their rights on behalf of their status as mothers
(Gheytanchi 2001: 566). Ojrat ol-mesl was one such example of women’s ijtihad to
Maternalist Policies versus Women’s Economic Citizenship? 101
further their social rights. Another was to argue that in cases of divorce, the courts
should calculate the wife’s deferred dower according to an index updated for
inflation (Poya 1999: 101–2).
Ojrat-ol-mesl, or the law mandating wages for housework, was passed by the
Majles (parliament) in late 1992. The regime’s Islamist women supporters played
an important role in the adoption of this policy – largely because of their dismay
over the ease by which men could divorce their otherwise good Muslim wives. As
believing women, their protests and recommendations were delivered in a
religious idiom. As Hoodfar explains:
Islamist women activists argued that women, like all other Muslims, are
entitled to the fruit of their labour on the grounds that Islam is against exploita-
tion; that in Islamic tradition wives have no duties to their husbands beyond
being faithful and are not required to work in their husbands’ homes, to the
extent that women are not even obliged to breast-feed their children without
payment from their husbands. Therefore, because all women do, in fact, work
in their husbands’ homes, they are entitled to the fruit of their labour!’
Hoodfar 2000: 311
Such women may have been advocating a maternalist policy in Skocpol’s sense.
However, not all Iranian feminists supported the new policy, and many felt
ambivalent about the use of the discourses of republican motherhood and the
rights of mothers under Islam. Although the law clearly gave women additional
benefits and increased the ‘cost’ of divorce to men, ojrat-ol-mesl was functional for
the Islamic Republic in that it carried no financial implications for the govern-
ment, it reinforced women’s maternal roles, and there was no onus on the
government to provide jobs or social assistance to divorced women.18 Moreover, as
we have seen above, these entitlements are due only to wives who are deemed to
be not at fault in the case of divorce. The legal scholar An-Naim (2002: 110)
emphasizes the point: ‘A woman is entitled to half of her husband’s assets if the
court finds the divorce was initiated by the husband and was not caused by any
failure on the wife’s part.’ This would appear to make ojrat-ol-mesl a conservative
maternalist policy – in Koven and Michel’s sense – that reinforces the patriarchal
gender contract rather than expands women’s economic citizenship. Moreover,
ojrat-ol-mesl ‘is difficult to apply in practice, partly because of the difficulty in
assessing wages for housework’ (An-Naim 2002: 110). Anecdotal evidence suggests
that it is far from an equal sharing of marital assets. Unless the husband has put
property or bank accounts in his wife’s name, he is entitled to keep all ‘his’ assets
except for that portion which is calculated as the ojrat-ol-mesl – or the wife’s
entitlement for the years of household and childcare.19
And what of other maternalist policies in the Islamic Republic of Iran, especially
for working women? In the early 1990s, and in response to its supporters’ concerns
about the status of women, the government of Hashemi Rafsanjani established the
Bureau of Women’s Affairs, which worked with a number of Iranian WID experts
and helped to establish WID or women’s units in various ministries. It also had
102 Valentine M. Moghadam
At least two of these measures are similar to the family-friendly policies found in
continental Europe, but the overall thrust of the measures is to reinforce women’s
family roles. Given women’s overall lower incomes, an average retired woman’s
pension can be rather low, and certainly insufficient to support a family. In fact,
the policies have been designed on the assumption of a salaried male household
head as the real breadwinner.
The Islamic Republic’s 1990 labour law also includes maternalist measures that
are similar to the maternal protection policies of many countries. The labour law
bans women and all workers under 18 from working at night. For pregnant
women, the law proscribes the assignment of heavy work; requires employers to
transfer pregnant women to lighter work at the same pay when their regular work
is deemed heavy or otherwise harmful to their condition; and prohibits women
from carrying heavy loads during pregnancy and for 10 weeks after giving birth.
The law entitles women to 90 days of maternity leave, at least half of which must
be taken after childbirth, for the first three children. New mothers are given time
off to nurse their babies in the workplace crèche (IRI 1990; Centre for Women’s
Participation 1999). The crèches in which new mothers may put their babies and
nurse them in the course of the work-day are usually found in factories.20
However, because factory work is generally deemed inappropriate to women,
relatively few women are involved in the manufacturing sector and there have
been no campaigns to encourage working-class women’s employment. The leaves,
nursing breaks, crèches and childcare centres, therefore, are enjoyed by only a
fraction of the economically active female population.
Healthcare services have expanded in the Islamic Republic, and special
programmes for pregnant women are in place, such as immunization against
diphtheria and tetanus. Maternal mortality rates have fallen, and couples are
required to attend a seminar in family planning prior to receiving their marriage
certificate. This has been accomplished under the public healthcare system through
the network of Rural Health Centres and Urban Health Centres, and since 1990 the
Ministry of Health and Medical Education has employed the services of women
Community Health Volunteers.21 According to a 1999 government publication,
about 80 percent of urban mothers and 73 percent of rural mothers have been
attended to more than twice during their pregnancy. Trained professionals
Maternalist Policies versus Women’s Economic Citizenship? 103
These policies and outcomes may reflect social welfare goals, concerns for the
health of mother and child, and the advancement of women. But conservative
legislators and jurists have blocked bills that would raise the age of marriage to 16,
give women the same inheritance rights as men, or award a temporary stipend to
widows disadvantaged by inheritance laws from their late husbands’ estates
(Shahidian 2003: 222). A bill that would ban sexual integration at hospitals and
public health facilities was cancelled only due to a public outcry. In frustration,
Mehrangiz Kaar wrote in Zanan (cited in Gheytanchi 2001: 573):
Nobel laureate Shirin Ebadi similarly has observed a discrepancy between the
stated objectives of some of the laws and policies of the Islamic Republic of Iran,
and their real objectives and implications:
Conclusions
This chapter has demonstrated how Muslim family law, as a social policy, impedes
women’s economic citizenship, and why it is that Iranian women activists (like
activists in other MENA countries) have prioritized reform of family law as a
necessary condition for the expansion of women’s rights and welfare. Whereas the
reform of family law is not in the interests of the rentier neopatriarchal state, it is
necessary for the strategic gender interests of Iranian women. I have argued that
Muslim family law today is an anachronistic social policy that reinforces the
patriarchal gender contract, undermines women’s economic citizenship, and does
little to provide for women’s welfare.
For these reasons, Iranian women have embarked on a quest for economic
citizenship. Although economic and cultural barriers continue to prevent large-
scale labour incorporation of women, middle-class women who are in the labour
force have been pushing for reform of those barriers – such as Muslim family law –
while also engaging in other strategies, such as forming or joining NGOs, in order
to enhance their social participation, employment opportunities and access to
economic resources.
This study also has confirmed the contradictory nature of maternalist discourses
and policies, especially as they have been practised in the Islamic Republic of Iran.
Following Koven and Michel, we note that maternalism in Iran has been a ‘cloak
for paternalism’ more than a radical discourse and vision emanating from the
women’s movement. Whereas Islamic women advocates have deployed a religious-
motherist discourse and were responsible for the adoption of ojrat-ol-mesl, secular
feminists have stressed the need to make Iran’s legal frameworks, including its
family law, more consistent with international standards and norms. Similarly,
because of the official discourse of women’s domesticity and the ideal maternal
role, secular feminists have steered clear of a maternalist discourse that seeks
rewards for women’s unpaid labour or housework. Rather, they emphasize the
need for more employment opportunities and equality in social insurance. Unlike
Europe, where ‘the highly unequal distribution of unpaid work and especially car-
ing labour underlies much of the inequality found in employment and reproduced
in social security’ (ILR 2000: 115), the issue in Middle Eastern countries like Iran is
the highly unequal distribution of paid work, along with explicitly discriminatory
family laws, that underlies the gender inequalities found in the region.
Indeed, in almost all Middle Eastern countries, maternalist politics and
discourses are associated with the neopatriarchal state and with Islamist move-
ments, including their female members. The masculine nature of the labour force
and the predominance of the discourse of motherhood is surely why MENA
feminists tend not to engage in a maternalist politics or advocate for policies
pertaining to women’s care work. Their priorities, instead, are reform of family law
in order that family relations are made more equitable and that women’s economic
and political participation is enhanced (Moghadam 2002a). Many non-Islamist
women have distanced themselves from religious-maternalist discourses and
campaigns, or have used them only instrumentally.23
Maternalist Policies versus Women’s Economic Citizenship? 105
There is no doubt that many Iranian families ignore Muslim family law and
provide daughters with the same inheritance as sons. A husband also may put
property in his wife’s name. (This was common in pre-revolutionary Iran, especially
among the elite.) But the fact is that the laws disadvantage women and apparently
so does much social practice. This, after all, is why Islamic women eventually cam-
paigned against the most egregiously discriminatory aspects of Muslim family law;
and why they lobbied for a mechanism to provide divorced women with some
share of the husband’s assets. One may also note that, having faced the limits of
the religious-maternalist discourse, Islamic women in and around the Centre for
Women’s Participation have deemed it necessary to lobby for Iran’s accession to
the Women’s Convention. How the state will respond remains to be seen.
Notes
1 This includes research by Linda Gordon, Theda Skocpol, Sonya Michel and Seth Koven,
Sonya Michel and Robin Rosen, and Molly Ladd-Taylor – who tend to have different
approaches to and definitions of ‘maternalist’ in the social policy or welfare state
context.
2 Representative of this body of research are studies by Gisela Bock, Helga Hernes, Barbara
Hobson, Ann Shola Orloff and Diane Sainsbury.
3 See, for example, studies by Nancy Folbre (1994), Arlie Hochschild (1989), Jane Jenson
(1986) and Jane Lewis (1980).
4 Esping-Anderson (1990) and Marshall (1964). Marshall’s theory remains powerful and
appealing, inasmuch as it integrates development and social policy frameworks and
offers a historical perspective.
5 The argument is made in a larger study of gender and social policy in the Middle East,
with a three-country focus on Iran, Jordan and Tunisia (Karshenas and Moghadam
2005). The present chapter draws on some of the conceptual discussion there.
6 See, for example, Richards and Waterbury 1996; Tzannatos and Kaur 2002.
7 Exceptions are Moghadam (1998) and Amira Sonbol (2003). See also the essay by Barbara
Swirski (in Joseph 2000), on differential access to social rights and the state’s welfare
benefits by Jewish and Palestinian women citizens in Israel.
8 On family law see, for example, Esposito and DeLong-Bas 2001; An-Naim 2002; Joseph 2000.
9 Among other things, Skocpol was keen to dispute the notion of the United States as a
relative ‘laggard’ among the core countries in terms of social welfare provisioning, and
to make a case for the revival of generous social policies in her country. What was dis-
tinctive about the early American policies, she argues, was the extent to which women
reformers were involved in their design and implementation. This has been challenged
by other scholars, who take issue with Skocpol’s definition of maternalist policies as
woman-centered, largely progressive, feminist, and cross-class. See the debate between
Skocpol and the historian Linda Gordon in the journal (now defunct) Contention, Vol. 2,
No. 3, Spring 1993: 139–89.
10 It is worth noting that maternalist discourses and politics have a long history in the
women’s movement, and have been especially evident in women’s peace and human
rights movements.
11 There are four Sunni schools (Hanafi, Hanbali, Maliki, Shafii) and one Shia school
(Jaafari) of Islamic jurisprudence. For details on Islamic family law across countries and
figh schools, see An-Naim (2002).
12 Mahr also commercializes the marriage contract. See Fatemeh Moghadam (1994).
13 Information from members of the author’s family, retired civil servants from the ministries
of agriculture, roads and transportation, and culture. Of course, the vast majority of
employees of the armed forces, the Oil Company, and the Ministry of Roads were men.
106 Valentine M. Moghadam
14 The figures are from UNESCO (1994), and the full table may be found in Moghadam
(1998), ch. 2.
15 Temporary marriage, known as sigheh, is an Iranian Shia phenomenon.
16 According to UNDP estimates, women’s earned income in 2002 was $2,835, compared
with men’s earned income of $9,946. See UNDP (2004), Table 24.
17 Family planning was revived and aggressively implemented in the wake of the economic
and social difficulties created by the eight-year war with Iraq, growing poverty due to
declining oil revenues and the state budget, and an exceedingly high birth rate during
the 1980s. In 1993 the parliament passed the Family Planning and Population Control
Act. This was the beginning of the ‘rationalization’ of social policy – within limits – in
the Islamic Republic. By the new century, Iranian (and Tunisian) women had, at 2.2
births per woman, the lowest fertility rates in the region, followed by Lebanon and
Turkey.
18 Yesim Arat (2000: 278–79) reports that a similar debate took place in Turkey, when
women’s groups proposed that an amendment to the civil code allow women to select
the option of sharing assets acquired during the marriage. Some feminists opposed this
measure, fearing that women would continue to see housewifery as an attractive
prospect, decline to work outside the home, and continue their economic dependence.
19 In early 2005, the sister of a friend of the author was seeking, without success, a larger
share of her husband’s assets after 22 years of marriage during which time the husband
had been frequently abusive. The amount of ojrat-ol-mesl that was due her was consider-
ably less than the average price of a home/apartment in Tehran.
20 In 1994, the author visited a crèche and childcare centre in a large textiles and garments
factory outside Tehran, and found them to be of high quality. See details in Moghadam
(1998), ch. 7.
21 In the early 1990s, the Iranian government mobilized around 20,000 women volunteers
to help implement its new family planning policy, but later refused them permission to
set up an independent organization. See Hoodfar (1998).
22 Shirin Ebadi, Women’s Rights in the Laws of the Islamic Republic of Iran (2002), reproduced
in the Iranian on-line feminist journal Badjens at http://www.badjens.com/ebadi.html
(accessed 20 December 2004).
23 Personal communication, Latifa Jbabdi of Morocco’s Union de l’Action Feminine,
Helsinki, 10 September 2004. Jbabdi was describing the feminist strategy of mobilizing
support for the reform of Morocco’s very conservative family code, the Mudawanna. She
told the author: ‘We had to break the taboo [of tampering with the family law, seen as
divine], and so we studied the Qur’an. We found that many aspects of the family law
contradicted the Qur’an. We are not Islamist feminists, but our country is a Muslim
country, and so we used religious language.’ The Iranian feminist lawyer Mehrangiz
Kaar, a staunch secularist, nonetheless urged ijtihad as a way of motivating believing
women to advocate change in the Islamic Republic.
References
Afary, Janet (2005) ‘Mahnaz Afkhami: A Memoir’ (book review essay). Journal of Middle East
Women’s Studies, Vol. 1, No. 1 (Winter): 147–57.
Afkhami, Gholam Reza (ed.) (2003) Jame-eh, Dowlat va Jonbesh-e Zanan-e Iran, 1342–1357:
Mosahebeh ba Mahnaz Afkhami (Society State and Women’s Movement in Iran: 1963–1978:
Interview with Mahnaz Afkhami). Bethesda: Foundation for Iranian Studies.
An-Naim, Abdullahi (2002) Islamic Family Law in a Changing World. London: Zed Books.
Arat, Yesim (2000) ‘Gender and Citizenship in Turkey’. In Suad Joseph (ed.) Gender and
Citizenship in the Middle East. Syracuse University Press.
Center for Women’s Participation (1999) Special Laws and Regulations for Women in the Islamic
Republic of Iran. [in Persian]. Tehran: Office of the President, Centre for Women’s Participation.
Maternalist Policies versus Women’s Economic Citizenship? 107
Esping-Anderson, Gøsta (1990) Three Worlds of Welfare Capitalism. Princeton NJ: Princeton
University Press.
Esposito, John, with Natana J. DeLong-Bas (2001) Women in Muslim Family Law. Syracuse NY:
Syracuse University Press.
Floor, Willem (1985) Labour Unions, Law and Conditions in Iran (1900–1941). University of
Durham Center for Middle Eastern and Islamic Studies, Occasional Papers Series No. 26.
Folbre, Nancy (1994) Who Pays for the Kids? Gender and the Structures of Constraint. London
and New York: Routledge.
Gheytanchi, Elham (2001) ‘Civil Society in Iran: Politics of Motherhood and the Public
Sphere’. International Sociology, Vol. 16 (4): 557–76.
Hernes, Helga (1987) Welfare State and Woman Power: Essays in State Feminism. Oslo:
University of Oslo Press.
Hobson, Barbara (1998) ‘Women’s Collective Agency, Power Resources, and Citizenship
Rights’. In Michael Hanagan and Charles Tilly (eds) Recasting Citizenship. Totawa NJ:
Rowman & Littlefield.
Hochschild, Arlie (1989) The Second Shift: Working Parents and the Revolution at Home. New
York: Viking.
Hoodfar, Homa (1998) ‘Volunteer Health Workers in Iran: Social Activists?’ Montpelier:
WLUML Occasional Paper no. 10.
Hoodfar, Homa (2000) ‘Iranian Women at the Intersection of Citizenship and the Family
Code: The Perils of Islamic Criteria’. In Suad Joseph (ed.) Gender and Citizenship in the
Middle East. Syracuse University Press.
ILO (1990) Yearbook of Labour Statistics: Retrospective Edition on Population Censuses
1945–1989. Geneva: ILO.
ILR (2000) ‘Introduction: Social Policy and Social Protection’. International Labour Review,
Vol. 139, No. 2: 113–117.
Islamic Republic of Iran [IRI] (1999) Human Development Report of the Islamic Republic of Iran,
1999. Tehran: Plan and Budget Organization and United Nations.
Islamic Republic of Iran [IRI] (1990) [1369]. Labour Law and Social Security of the Islamic
Republic of Iran. Tehran: Korshid Publications (in Persian).
Islamic Republic of Iran [IRI] (1997) Statistical Yearbook 1375 [1996]. Tehran: Central
Statistical Office.
Jenson, Jane (1986) ‘Gender and Reproduction: Or, Babies and the State’. Studies in Political
Economy, No. 20 (Summer): 9–46.
Joseph, Suad (ed.) (2000) Gender and Citizenship in the Middle East. Syracuse University Press.
Kaar, Mehrangiz (1996) ‘Women and Personal Status Law in Iran: An Interview with
Mehrangiz Kaar’. Middle East Report, No. 198 (January–February): 36–8.
Karshenas, Massoud (1997) ‘Macroeconomic Policy, Structural Change, and Employment in
the Middle East and North Africa’. In Azizur Rahman Khan and M. Muqtada (eds)
Overcoming Unemployment. London: Macmillan.
Karshenas, Massoud and Valentine M. Moghadam (eds) (2005) Social Policy in the Middle East:
Economic, Political, and Gender Dynamics. Basingstoke: Palgrave Macmillan.
Kerber, Linda (1980) Women of the Republic: Intellect and Ideology in Revolutionary America.
Chapel Hill: University of North Carolina Press (1st edn).
Kessler-Harris, Alice (2001) In Pursuit of Equity: Women, Men, and the Quest for Economic
Citizenship in 20th Century America. New York: Oxford University Press.
Koven, Seth and Sonya Michel (1993) ‘Introduction’. In Seth Koven and Sonya Michel (eds)
Mothers of a New World: Maternalist Politics and the Origins of Welfare States. New York: Routledge.
Kwon, Huck-ju (ed.) (2005) The Developmental Welfare State and Policy Reforms in East Asia.
Basingstoke: Palgrave Macmillan.
Ladd-Taylor, Molly (1994) Mother-Work. Women, Child Welfare, and the States, 1890–1930.
Chicago and Urbana: University of Illinois Press.
Lewis, Jane (1980) The Politics of Motherhood: Child and Maternal Welfare in England,
1900–1936. London: Croom Helm.
108 Valentine M. Moghadam
Loewe, Marjus (2000) ‘Social Security in Egypt: An Analysis and Agenda for Policy Reform’.
Background Paper for the 7th Annual Conference of the ERF, Amman (26–29 October).
Marshall, T. H. (1964) Citizenship and Social Class. Cambridge: Cambridge University Press.
Mehrpour, Hosein (1995) ‘A Brief Review of the Woman’s Economic Rights in Iranian Legal
Systems’. Farzaneh, Vol. 2, No. 7 (Fall/Winter): 49–64.
Messkoub, Mahmood (2005) ‘Constitutionalism, Modernization, and Islamization:
A Political Economy of Social Policy in Iran’. In M. Karshenas and V. Moghadam (eds)
Social Policy in the Middle East. Basingstoke: Palgrave Macmillan.
Meyer, Mary K. and Elisabeth Prugl (eds) (1999) Gender Politics in Global Governance. Lanham
MD: Rowman & Littlefield.
Michel, Sonya and Rianne Mahon (eds) (2002) Child Care Policy at the Crossroads: Gender and
Welfare State Restructuring. New York and London: Routledge.
Moghadam, Fatemeh E. 1994 ‘Commoditization of Sexuality and Female Labor Participation
in Islam: Implications for Iran 1960-90’. In Mahnaz Afkhami and Erika Friedi (eds) In the Eye
of the Storm: Women in Post-Revolutionary Iran. Syracuse, NY: Syracuse University Press.
Moghadam, Valentine M. (1988) ‘Women, Work and Ideology in the Islamic Republic’.
International Journal of Middle East Studies, Vol. 20 (2) (May): 221–43.
Moghadam, Valentine M. (1998) Women, Work and Economic Reform in the Middle East and
North Africa. Boulder CO: Lynne Rienner Publishers.
Moghadam, Valentine M. (2000) ‘Hidden from History? Women Workers in Modern Iran’.
Iranian Studies, Vol. 33, No. 3–4 (Summer/Fall): 377–401.
Moghadam, Valentine M. (2003) Modernizing Women: Gender and Social Change in the Middle
East. Boulder CO: Lynne Rienner Publishers (2nd edn).
Orloff, Ann Shola (1993) ‘Gender and the Social Rights of Citizenship: The Comparative
Analysis of Gender Relations and Welfare States’. American Sociological Review, Vol. 58
(June): 303–28.
Orloff, Ann Shola (1996) ‘Gender in the Welfare State’. Annual Review of Sociology, Vol. 22:
51–78.
Pateman, Carole (1988) ‘The Patriarchal Welfare State’. In Amy Gutmann (ed.) Democracy and
the State. Princeton NJ: Princeton University Press.
Povey, Elaheh Rostami (2004) ‘Trade Unions and Women’s NGOs: Diverse Civil Society
Organizations in Iran’. Development in Practice, Vol. 14, Nos. 1/2 (February): 254–66.
Poya, Maryam (1999). Women, Work, and Islam. London: Zed.
Richards, Alan and John Waterbury (1996) A Political Economy of the Middle East. Boulder CO:
Westview (2nd edition).
Sainsbury, Diana (1996) Gender, Equality and Welfare States. Cambridge: Cambridge
University Press.
Shahidian, Hammed (2003) Women in Iran: Gender Politics in the Islamic Republic. Westport
CT: Greenwood Press.
Sharabi, Hisham (1988) Neopatriarchy: A Theory of Distorted Change in Arab Society. New York:
Oxford University Press.
Skocpol, Theda (1992) Protecting Soldiers and Mothers: The Political Origins of Social Policy in the
United States. Cambridge MA: Harvard University Press.
Sonbol, Amira al-Azhary (2003) Women of Jordan: Islam, Labour, and the Law. Syracuse NY:
Syracuse University Press.
Tzannatos, Zafiris and Iqbal Kaur (2002) ‘Welfare State Policies in the Middle East and North
Africa’. In Ethan P. Kapstein and Branko Milanovic (eds) When Markets Fail: Social Policy
and Economic Reform. New York: Russell Sage Foundation.
UNESCO (1994) Education for All: Status and Trends 1994. Paris: UNESCO.
UNDP (2004) Human Development Report 2004. New York: OUP.
Welchman, Lynn (2001) ‘Capacity, Consent and Under-Age Marriage in Muslim Family Law’.
The International Survey of Family Law, 2001 Edition. Cambridge: Cambridge: University
Press.
5
Gender Equality and Developmental
Social Welfare in South Africa
Shireen Hassim
Introduction
109
110 Shireen Hassim
The democratic government inherited a state that from a social policy perspective
was unique in Africa. Social assistance was already inscribed as state responsibility,
albeit in a racially discriminatory form, prior to the creation of apartheid. State
pensions, the major area of government spending on welfare, have always been a
non-contributory form of social assistance and although means tested were
awarded regardless of work status and with no reciprocal obligations on recipients.
Pensions were supplemented by a range of other non-contributory, means-tested
social assistance mechanisms. In general, as Lund et al. (1996: 97) have characterized
it, the South African system is a ‘strange combination of mainly British welfare tra-
dition and apartheid policy’ where the distinction between the ‘deserving’ and the
‘undeserving’ poor is marked by race (Van der Berg 1997). Race was regarded as
shaping lifestyle patterns as well as entitlements. Under apartheid, African exclu-
sion from welfare was justified on the grounds that ‘people accustomed to modern
lifestyles and consumption patterns had greater need of social protection than those
in rural subsistence agriculture, who were not proletarianized and were thus
presumed to be better placed to meet traditional subsistence needs’ (Van der Berg
1997: 486).
Gender Equality and Developmental Social Welfare in South Africa 111
The provision of pensions and other grants (primarily child maintenance and
disability) expanded in the period between the two world wars partly in response
to unionized white working-class struggles and as a mechanism for addressing the
‘poor white’ problem. Inclusion of poor whites (mainly Afrikaners) was a corner-
stone of the Afrikaner nationalist movement and the creation of safety nets
targeted at this group legitimated ideological processes of building a racially
exclusive state system. Africans – even those who were workers in the urban
formal economy – were excluded from welfare provisions on the grounds that the
burdens of social reproduction would be carried by extended families. Indians
were regarded as temporary residents in South Africa; a repatriation policy was still
adhered to formally. On the basis of recommendations of the Carnegie
Commission of Enquiry, which sat in 1929, a state welfare department was
established for the first time in 1937.
African urbanization and the rise of urban slums created enormous anxieties for
the state during this period, although the response was not the comprehensive
welfare system that developed to support poor whites. Over time some benefits,
such as pensions and grants for the blind and disabled, were extended to African
and Indian people (men and women) although benefits were pegged at different
levels on the basis of racist assumptions about the basic needs of different groups.1
For example, African pensions were set at one-tenth the amount for whites and
the means test employed for Africans was far more stringent (Van der Berg 1997:
487). Nevertheless, the acceptance of the idea that all groups were entitled to some
benefits is startling given the denial of political and civil rights to all but whites.
During the apartheid period, the racial spread of welfare benefits was retained but
the gap in the levels of benefits between whites and Africans widened considerably.
Apartheid also began to be inscribed not just in the ideological premises of policy
but also in the ways in which services were delivered to citizens. Administratively,
there was a shift in the 1950s from a single Welfare Department in government for
the whole population to separate departments for different race groups, although
all departments were governed by the same welfare legislation (Lund 1988: 24).
Lund (1988) points out that not only did the apartheid government force private
voluntary agencies to stop offering services on a racially inclusive basis, it also
insisted that voluntary welfare bodies not have racially mixed management com-
mittees. State subsidies to private welfare agencies stipulated racially discriminatory
salaries for social workers.2
The 1950s were also a period of heightened political mobilization against the
unfolding of the apartheid state, with women being among the most vociferous
critics of attempts to control the movement of African people and to limit African
women’s access to the labour market through the imposition of pass controls. An
independent, non-racial national women’s movement, the Federation of South
African Women (FSAW) emerged that aligned itself with the ANC and linked
women’s struggles for emancipation with those of the national liberation
movement. The FSAW played a central role in redefining women’s political roles
away from being ‘the tea ladies of the struggle’ to articulating concrete political
demands, within a radical motherist frame (Wells 1992). The ideology of
112 Shireen Hassim
between social and economic development is not mere rhetoric. A key plank in the
developmental social welfare platform is the creation of employment; this is a
responsibility of macroeconomic policy in general but also specifically of the
Department of Social Development (previously Welfare and Population
Development) through the Community Based Public Works Programme, funded
out of the social security and welfare budget. Public works programmes are seen to
have the twin benefits of addressing the infrastructural needs of the country as
well as reducing poverty and long-term dependence on state assistance. They are
therefore seen as developmental in their impact, and considerable resources have
been directed into these programmes.
The White Paper identifies a wide and impressive set of guiding principles, 17 in
all, including democracy, human rights, justice, transparency and accountability.
An examination of the list of principles suggests the policy is based on wide-ranging
and expansive conceptions of citizenship. In the particular formulations of the
guiding principles, it is compatible with the capabilities approach developed by
Amartya Sen, which focuses less on the elaboration of the entitlements of citizen-
ship and much more on whether all members of society are capable of achieving
an enhanced quality of life. For example, non-discrimination is specified in
addition to equity; where equity refers to the distribution of material resources,
non-discrimination focuses on ‘tolerance, mutual respect, diversity, and the inclu-
sion of all groups in society’. The policy is based on the principle that the quality
of life of all people should be raised through a redistribution of resources and ser-
vices. In yoking the cultural concept of ubuntu (humanity) the White Paper signals
the importance of cultural norms and values, particularly the principle of caring
and mutual interdependence, to the project of development. Elsewhere, in the
‘Agenda for Action’, the policy emphasizes the need for government programmes
to ensure the realization of citizens’ ‘dignity, safety and creativity’.
This expansive notion of development initially found resonance and widespread
support in civil society. The White Paper was developed under the new ‘rules of
the game’ with regard to public decision-making introduced by the ANC govern-
ment, in which policies would be developed through extensive consultations
between government and civil society, and where public policies in a range of
areas – the economy, health and welfare, trade and so on – were assumed to be
synergistic. The White Paper was developed in a context of general optimism
about the ability of the state to lead a process of transformation and a faith in the
democratic process.
As a policy document produced out of compromise between different interests,
the White Paper inevitably embodies numerous tensions that are left to particular
programmes to deal with. From a feminist perspective, the most noteworthy tension
is between the capabilities approach and neoliberal caveats. Social assistance pro-
grammes were to be based on a principle of affordability and sustainability – that is,
they were to be ‘financially viable, cost efficient and effective’. These neutral terms
obscure the extent to which welfare budgets are vulnerable to the imperatives of
‘fiscal responsibility’. The plan of action for the White Paper recognized the fiscal,
economic and infrastructural constraints on government’s ability to implement
116 Shireen Hassim
the principles of developmental social welfare. Social security provisions were thus
to be ‘phased in’ on the principle of the progressive realization of benefits (this
wording conforms to the constitutional provisions on social rights) as well as
sustainability. The issue of how rapidly this would be done and what the targets
for inclusion would be are not specified in the White Paper and, like the question
of affordability, would be intensely contested in the welfare debates that ensued.
Despite the emphasis on participation, poor women were relatively poorly orga-
nized and had very little voice in national-level policy debates, making it unlikely
from the outset that they would have much power over budgetary choices.
Although developmental welfare seeks to provide a third way between inequality-
enhancing residualist systems and expensive social democratic systems,
commitments to social justice are relatively weak. There is an inbuilt normative
choice in the emphasis on public works programmes as opposed to expansion of
the scale of welfare benefits that sets up a two-tier system of benefits, with people
on work-related programmes treated as ‘deserving’ poor and those on welfare (and
particularly mothers drawing the child support grant), as either passive and depen-
dent subjects or cunning exploiters of the system. Reflecting on the notion of
developmental social welfare in 2003 in the context of a debate on the Basic
Income Grant, Ravi Naidoo of the trade-union research agency NALEDI argues
that the ‘trendy concept of developmental social welfare has failed miserably’ and
that it has embedded notions of ‘an undeserving poor’ and ‘promoted ways for the
“able-bodied” to pull themselves up by their own bootstraps’ (Naidoo 2003).
The White Paper names the importance of the non-state welfare sector (religious
organizations and non-governmental organizations) as well as informal special
support systems, including community care, in meeting social service needs.
Again, this is posed in the White Paper in the spirit of ‘national collective respon-
sibility’, but it leaves open the question of the precise balance between the different
sectors of welfare provisioning. In this respect the emphasis on the cultural value
of caring might perhaps be seen as loading the dice against women, who bear the
practical burdens of care-work within families and communities. As Lund et al.
(1996: 115) note, ‘a double equation is at work which assumes that community
care is equal to care by families which is equal to unpaid care by mostly women’.
Women’s caring burdens have dramatically increased as the HIV/AIDS infection
rates have assumed pandemic proportions. Indeed, early evidence from the
pandemic is showing that it is not only women who are carrying an even greater
burden of caring but that children are increasingly having to take on these roles.
The shift away from the language of rights and entitlements in the White Paper
would seem to dilute the particular (and greater) responsibility of the state in
meeting social security needs through the redistribution of public resources.
A final area of tension in the White Paper relates to the ways in which ‘the
family’ is invoked in the document. Under the subheading ‘The Family’,
the White Paper simply states: ‘The family is the basic unit of society. Family life
will be strengthened and promoted through family-oriented policies and pro-
grammes’. It is not difficult to understand why ‘strengthening family life’ is a
desirable goal for many in South Africa, given the ways in which the migrant
Gender Equality and Developmental Social Welfare in South Africa 117
labour policies of the colonial and apartheid state, and the dominance of
residentially based domestic work denied basic human comfort and intimacy to so
many Africans. The White Paper leaves open, perhaps intentionally, the definition
of what constitutes family (while stipulating that social security provisions should
include homosexuals). These specificities are left to particular policies to articulate
and while in some instances (such as the Lund Committee on the implementation
of a child support grant) opportunities were taken to shift away from nuclear,
male-headed family forms as the norm, this approach is not guaranteed in the
overarching policy framework, leaving perhaps too much to the political will and
ideological perspectives of particular policy-makers.
The increasing centralization of macroeconomic decision-making from 1996
undermined the assumptions of consultative, participatory decision-making
assumed by the White Paper. For women’s organizations, the ability to leverage
the symbolic power and legislative representation of women into policy outcomes
was severely undermined. Government’s assertion of fiscal restraint introduced a
new discourse into policy-making: the debate because less concerned with what
was desirable and more concerned with what was possible. Affordability was often
assessed in narrow fiscal terms and by prioritizing gross inequalities rather than a
concern with the long-term costs of failing to address pervasive systemic inequal-
ities. The formal provisions of the constitution proved inappropriate in dealing
with the ways in which government prioritizes spending. Although the right to
social security is entrenched in the socioeconomic-rights clause in the constitu-
tion (section 27) the implementation of this right is by no means automatic, nor
does it guarantee that the extent of social security provided will be adequate to
ensure a decent standard of living. The important proviso to the right to social
security is a qualifying clause in section 27 of the Bill of Rights, which defines the
state’s obligations as limited to ‘available resources’ (Government of South Africa
1996, section 27).
The immediate consequences of the emphasis on affordability were seen in the
process of overhauling the system of child and family benefits, which the
Department of Welfare instituted in 1996 and which coincided with the public
debate on the White Paper on Social Welfare. The changes to the welfare provisions
for children began in a transitional context of translating broad policy formula-
tions into concrete programmes, and the deliberations of the Lund Committee,
which spearheaded the changes, reveal the tensions between pursuing equity (by
ensuring that poor African families would benefit from social assistance) and
affordability.
By July 1996, poor mothers were paid a total monthly grant of R565 (made up
of a parent allowance of R430 and a child allowance of R135) for a maximum of
two children up to the age of 18 (that is, to a maximum of R700). For those fami-
lies receiving it, the grant played a major poverty-alleviation role in raising overall
household income above the household subsistence level. However, as noted
above, the grant reached very few African households, with Indian and Coloured
families being the primary beneficiaries. Attempts to entrench the racial equity of
the grant produced the perverse consequence that while more poor families
118 Shireen Hassim
received the grant the monetary value of the grant was drastically reduced. The
Committee had to work within the government’s decision not to increase overall
spending. At the time, government had already cut defence spending significantly
(virtually halved between the 1990/91 and 1996/97 fiscal years) and increased
social security spending by 120 per cent over the same period (Gelb 2004). There
was a strong perception in the Committee that no further increases were likely and
that the demands on the existing budget were likely to increase under the impact
of the HIV/AIDS epidemic. The Committee’s recommendations were therefore
based on a fiscally constrained scenario in which the already inadequate welfare
budget faced further cuts from the central government.
A number of concerns have surfaced with regard to the effective implementa-
tion of the child support grant. Firstly, women’s organizations have criticized the
two-tier means test to establish eligibility for the grant. The means test requires
that the primary caregiver prove that he or she is a member of a household with a
combined income below R9,600 (approximately $1,300) per annum for urban
households and R13,200 (approximately $1,800) per annum for rural dwellings or
those in informal areas. Other elements of the means test include a requirement
that the primary caregiver shows that she/he is actively seeking employment. The
Lund Committee was itself not convinced of the benefits of means testing, favouring
instead a universal grant. The Committee noted that the means test imposed
administrative costs on the system as it required extensive assessment and
monitoring and imposed costs on parents, which could act as a negative incentive,
particularly for foster parents. The Committee also warned that means testing
could be a biased process. ‘Communities, especially in poorer areas, are so greatly
under the domain of traditional leaders with extensive powers of patronage that
caution should be exercised in this approach. The track record of civic associations
in impartial decision-making is likewise uneven’ (Lund Committee 1996: 29).
However, Department of Welfare officials insisted on the retention of this mecha-
nism, concerned that there might otherwise be overwhelming numbers of people
applying for the grant and that many of them might be ‘unworthy’, at least in
economic terms. The Committee stressed that should a means test be adopted, it
should be a simple economic test and should not ‘in any way depend on a defini-
tion of a family’ (Lund Committee 1996: 92). However, the old means test appears
to still be in place. Recent research shows that many welfare officials informally
add on new ‘tests’ of eligibility, such as proof of tax being paid, and letters certified
by police officers (Goldblatt 2005). Anecdotal evidence suggests that welfare offi-
cials turn mothers who are overtly lesbian away. The South African NGO Coalition
(SANGOCO) has called for the existing means test to be replaced with one that is
based on the income of the primary caregiver, arguing that this will be easier to
administer. As Liebenberg points out, ‘this is particularly important in view of
the fact that the child support grant requires a doubling of the capacity of the
welfare system to process grants. The present system is already over-burdened with
huge backlogs in poverty-stricken areas’ (Liebenberg 1998: 11).
These problems are exacerbated by poor management and delivery systems,
and in some cases corrupt practices at the provincial level, which initially led to
Gender Equality and Developmental Social Welfare in South Africa 119
under-spending of welfare budgets for three consecutive years, with the child
support grant showing the slowest take-up rate (EPRI 2002). In an assessment of
the effectiveness of the grant commissioned by the Department of Welfare and
conducted in 2000, the Community Agency for Social Enquiry (CASE) found that
the capacity and competence of the Department of Welfare to administer the grant
was limited by a range of problems: insufficient information, outdated application
forms, lack of coordination between the Departments of Welfare, Health and
Home Affairs, poor departmental cooperation with NGOs and community organi-
zations and by ‘indifferent and even hostile attitudes on the part of Welfare staff’
(CASE 2000: 120). The government has sought to deal with some of the adminis-
trative problems related to delivering grants by privatizing grant payments and
creating a common agency for grants payments that would take back control over
grants payments from the provincial to the national level of government. This has
by no means made the grants more accessible; indeed many pensioners have com-
plained that queues remain as long as ever and that the private companies are
using the payment process as an opportunity to sell insurance policies and even
short-term loans to pensioners. The provinces with the best resources (Gauteng
and Western Cape) have been most successful in reaching delivery targets, while
those with the greatest need (Eastern Cape and Limpopo) also have the least effective
delivery systems and have been the least successful in reaching their targets
(Guthrie 2002).
Most street children and children in child-headed households cannot access the
grant because they do not have the necessary identity documents to apply, or they
do not know the procedures for application. The Alliance for Children’s
Entitlement to Social Security (ACESS) estimates that 75.8 per cent of South
African children live below a poverty line of R400 per month. Many NGOs have
recommended increases in the level of the grant, but noted that ‘this would
require a political decision involving a trade-off with other grants and budgetary
items’ (Guthrie 2002: 122). The government has been responsive to these recom-
mendations, improving information about the grants and increasing the amount
and age limit of the grant. By 2003, close to 3 million poor children were
receiving the grant (Woolard 2003) and by December 2004 this number had
increased to 5.4 million, despite the administrative difficulties described above
(Taylor 2004).
A major discursive shift introduced by the Lund Committee was the move away
from the ‘familist’, male-worker model of social policy. The Committee chose to
adopt a ‘follow the child’ policy, with emphasis on the child rather than the carer –
in other words, the grant would be paid to the primary caregiver on behalf of the
child.4 Means-testing would therefore no longer comprise any component of
moral assessment of the worthiness of carer (at least in theory); nor would carers
who were not mothers be excluded from the grant. The Committee adopted the
policy in recognition of the fluidity and diversity of household forms. This was a
significant shift away from the normative assumption that the nuclear family was
the desirable model, which was inscribed in apartheid-era social welfare; and,
importantly, it addressed the rights of children born into polygamous families.
120 Shireen Hassim
At least in formal terms, the effect was to de-link childcare from normative
assumptions about family forms or gender responsibilities and particularly to de-
link care work from mothering. As the Committee noted, ‘it resolves the problem
of how to define the family in such a complex and multi-cultured society. It says
that children, however many in a household, of whatever status, are important
and need to be protected’ (Lund Committee 1996: 91).
Of course, these shifts in the core assumptions of the child support grant do not
as yet translate into shifts in the views of welfare officials and the media. In reality,
moral discourses continue to infect social security provision, crowding out rights-
based arguments for social security. In debates about the child support grant, there
was a strong view among many in government that welfare grants reinforced
apartheid privileges, had no developmental potential and should be phased out in
favour of greater attention to programmes such as community-based public works
(Francie Lund, personal communication). Even some ANC women members of
parliament took a conservative view of welfare as reinforcing a ‘culture of entitle-
ment’, with welfare grants seen as handouts that reinforced dependency on the
state. As one MP argued, ‘women should look at developing themselves’
(Parliament of South Africa 2000). Fraser-Moleketi herself accused poor people of
not doing enough: ‘communities had to change the thinking of those who held
out their hands for help but kept their sleeves down, a sign that they were not
willing to work’ (Koopman 1999). The child support grant continues to be blamed
for increasing teenage pregnancy (that is, that young girls are getting pregnant in
order to access cash), that women are spending the money on ‘clothing and
lipstick’ and even an allegation by the Minister of Social Development that mothers
have ‘rented’ out their children to others so that they can claim social assistance.
Newspaper reports castigate ‘runaway mothers’ who ‘claim the child support grant
meant to feed their offspring’ (Rhupiso 2003). Even the Chairperson of the
Commission on Gender Equality had to be gently reminded by feminist activists
not to fall into the trap of stereotyping women receiving the child support grants
as undeserving.
The Lund Committee’s brief was a narrow one, and the debates which accompanied
the release of its recommendations and subsequent problems with the implemen-
tation of the child support grant as well as major inefficiencies in the delivery of
all social grants have reopened larger questions about the overall system of social
assistance. A central concern among many welfare activists was the narrow reach
of social security. However redistributive in effect, by their nature pensions only
reach a limited number of people. Although by 2004 uptake of the child support
grant outstripped government expectations, at the moment poor children over
14 do not receive support. Up to 60 per cent of the poor – mainly those
between the ages of 14 and 60 – are not getting any social security at all. It is
estimated that 11.8 million of the poorest 23.8 million South Africans live in
households that receive no social assistance (Lund 2004). The trickle-down effects
Gender Equality and Developmental Social Welfare in South Africa 121
of the Growth, Employment and Redistribution strategy (GEAR) are not material-
izing; levels of growth remained well below the rate required to address underlying
social needs. Unemployment has continued to rise and employment creation
through the public works programme was disappointing. It has been estimated
that the Community Based Public Works Programme only created between 13,000
and 33,000 jobs per annum between 1996 and 2001 (McCord 2003). By the end of
the 1990s, the Congress of South African Trade Unions (COSATU) as well as a
range of civil society organizations were arguing strongly for a more inclusive sys-
tem of social security that would have poverty-reducing effects.
Poverty is, unsurprisingly, a deeply gendered phenomenon. Statistics from
South Africa’s labour force study in 2003 showed that women on the whole had
lower incomes, higher unemployment rates and less access to assets than men
(Seidman-Makgetla 2004: 2). African women make up 42 per cent of the workforce
but only 30 per cent of the employed population. Those women who are
employed find themselves in the worst-paid sectors of the labour market, notably
in domestic and retail work. In 2003, 96 per cent of domestic workers were black
(that is, African, Indian and Coloured) women and 93 per cent of these workers
earned under R1,000 (approximately US$180) per month (Seidman-Makgetla
2004: 7). African rural women are the poorest category of citizens: in 1997, 65 per
cent of African female-headed households in rural areas were poor compared to
54 per cent of male-headed households (Seidman-Makgetla 2004: 40). At 29.4 per
cent, the mortality rate among African women was more than twice that of white
women in 1994 (11.5 per cent) (Seidman-Makgetla 2004: 41). It has been estimated
that 53 per cent of South Africans, including 60 per cent of the country’s children,
live in households with the lowest per capita consumption (Samson 2002: 71). On
the United Nations’ Gender and Development Index, even though women and
men had comparable school enrolment and adult literacy ratios, men earned twice
as much as women (Budlender 2004: 10). These gender vulnerabilities are com-
pounded by the HIV/AIDS pandemic. African women are most vulnerable to HIV
infection, more women than men are HIV-positive and women are likely to
become more infected at a younger age than men (Albertyn 2003). The pandemic
imposes additional burdens on women in their roles as primary carers of family
members who are HIV positive. These caring tasks, moreover, have to be performed
in the context of poor basic services such as the availability of clean water,
electricity and modern sanitation.
To what extent could social policy address these gendered dimensions of
poverty? There have been few concrete proposals put forward by the women’s
movement. Instead, many women’s organizations have allied themselves to the
proposal for a Basic Income Grant (BIG), put forward by COSATU as a universal
poverty-reduction mechanism. COSATU envisaged that a relatively small universal
grant (R100 per month) would be introduced for all individuals (including
children). Although neither business nor government were enthusiastic about a
BIG,5 COSATU kept up pressure for the grant both within the tripartite alliance
and in the media. In July 2001 the BIG Coalition involving 12 organizations was
formed. By this time, government had appointed a Committee of Inquiry into a
122 Shireen Hassim
Education 18 21 22 20 20 20 20 19
Health 9 10 11 11 11 11 11 11
Housing, other 13 5 3 4 4 5 5 5
social services
Social services 46 46 48 48 49 50 51 51
(total)
Protection 20 17 16 17 17 17 16 16
services
Economic 14 11 9 11 12 13 13 13
services
Interest 12 19 20 17 15 13 13 13
Other 8 7 8 7 6 7 6 6
Total 100 100 100 100 100 100 100 100
Source: Stephen Gelb, Inequality in South Africa: Nature, Causes, Responses ( Johannesburg: The Edge
Institute, 2004).
Policy Research Institute (which has been associated with the BIG Coalition), shows
that rising expenditure has begun to impact positively on poverty, labour-market
participation and productivity. Minister of Social Development Zola Skweyiya
argues that ‘income support is more than a safety net for the poorest and most
vulnerable people in our society. It is also a trampoline that enables many people
in these households to jump over the barriers of economic and social exclusion’
(Ministry of Social Development 2004). The study showed a 66.6 per cent reduction in
poverty, when the destitution poverty line (a measure of relative destitution based
on the household expenditures of households in the lowest 20 per cent of the
income distribution) is used as a benchmark. The study found that the progressive
extension of the child support grant up to the age of 14 would reduce the poverty
gap by 57 per cent. Receipt of the child support grant correlates to increasing
enrolment of children in primary school, with Lund (2004) reporting a remarkable
8 per cent increase in enrolment in grant-receiving households. The EPRI study also
found that the major social grants are significantly and positively associated with
greater household expenditure on food and consequently better nutritional
outcomes. The impact of existing social grants also seems to be addressing other
areas that are of concern to the proponents of a Basic Income Grant. The study
found that ‘social grants provide potential labour market participants with the
resources and security necessary to invest in high risk/high reward job search, [and
that] living in a household receiving social grants is correlated with a higher success
rate in finding employment’ (Ministry of Social Development 2004).
Nor surprisingly, the most prominent debates about the BIG have related to afford-
ability and administrative capacity. The impact of the Basic Income Grant on
124 Shireen Hassim
women has not been directly addressed by the major participants in the BIG
Coalition, but the proposal raises a number of issues for feminist analysts. A range
of women’s organizations have argued that the BIG should be supported by
women on the grounds that women carry the major burden of poverty, and that
‘women are often more responsible than men in using income collectively and
developmentally’ (Goldblatt 2003: 2). The universal nature of the grant may
remove the stigma attached to other social assistance grants, and the absence of a
means test will reduce the degree of moral regulation of women. However, there
are a number of assumptions embedded in arguments for the BIG that bear closer
attention. These relate to assumptions about intra-household behaviour, the
impact of a direct cash grant on women’s autonomy and the consequences of a
universal grant for enhancing women’s citizenship entitlements and shifting from
notions of female dependency. I will deal with these very briefly.
There has been little attention among BIG proponents to the internal power
dynamics within households. It is assumed that members of a household will pool
their grants and that these will then be used for the benefit of the household as a
whole. This assumes also that the interests of the household are unitary and that
there will be consensus as to the spending priorities. There is inadequate research
into internal household dynamics in South Africa and much of the discussion is
thus highly speculative. However, existing research suggests that households are
not monolithic entities in which incomes will be pooled (Posel 2001). Some of
these concerns have been raised in relation to the BIG. As Beth Goldblatt asks, ‘is
it possible or likely that lack of trust will prevent women and men from pooling
the BIG or deciding together how it should be spent?’ (Goldblatt 2003: 2). Francie
Lund (2004) has pointed out that there is a further assumption that people live in
just one household, alluding to the fact that household forms in South Africa are
fluid and diverse. Eva Harman has shown how the distribution of social grants is
‘mediated by social relations, historical dynamics and material conditions’
(Harman 2003: 3). While her research shows that grants are pooled within some
poor households, this does not mean that distribution within the households is
fair and grants may indeed increase dependency of non-grant recipients on recipi-
ents. Merely receiving a grant, such as a pension, does not mean that the recipient
will have the power to decide how the grant will be used. The elderly, for example,
may receive pensions and thereby secure their place and some degree of care from
their families, but may give up the grant to younger (possibly male) and more
powerful members of the household. From this point of view, a direct grant to
each member of the household may reduce such dependencies, but lack of power
in the private sphere may lead to grants being appropriated. Work on the spread of
HIV shows the extent to which women lack power over decision-making within
households – even where the risk of not asserting their voice is extreme illness and
even death (Albertyn 2003). Some have argued that a BIG will reduce the extent of
domestic violence as women will have access to cash and therefore independence.
This seems a far-fetched expectation.
An area in which there is more likely to be a gendered impact lies in the poten-
tial shift that would result in rural households away from remittances to reliance
Gender Equality and Developmental Social Welfare in South Africa 125
on a BIG. Although migrant remittances are crucial to the survival of rural house-
holds, they are also part of a cycle of power. Dorrit Posel’s research on the varied
patterns of distribution of migrant remittances within rural households suggests
that we cannot assume that migrants remit to rural households out of altruism
(Posel 2001). Posel found that remittances are being directed not at households
but at particular individuals in households, and that migrants may be more
responsive to the needs of some household members than to others. Some
claimants within households – such as old people and children – are seen as valid
claimants whilst others (notably young women) are seen as invalid.
A universal grant may therefore have the effect of uncoupling support for rural
households from the vagaries of the remittance system and might offer rural
women greater autonomy and enable them to overcome other barriers to eco-
nomic security. Although women have yet to gain equal rights to ownership and
access to communal land, they are able to purchase other forms of land through
accessing a state subsidy (Walker 2001). The major barrier to this thus far has been
lack of cash (purchasers are required to make a cash deposit before the state will
issue a subsidy). An income grant might make it possible for women in households
where there is more than one woman to pool their income and purchase the land
they are currently working. A basic monthly income might make it easier for
women to access low levels of credit to increase their productivity. Evidence sug-
gests that there is a definite gender pattern in household spending. Lund’s work on
pensions shows that the grant is most generally put to common household use
and that, moreover, pension money that goes to women is spent better (that is, on
food, health and education) than that which goes to men (Lund 2004). Directing
cash grants to women is therefore likely to enhance the quality of life of poor
women and children. However, these assumptions have not yet been tested by
empirical research and cannot be said to decisively provide support for the idea
that a small cash grant will be empowering for women.
The paucity of gendered analysis in social assistance debates is partly attributable
to the decline in feminist activism since the advent of democracy. Gender politics
in the last ten years has centred on increasing women’s representation in various
institutions, while the original link between representation and equality outcomes
appears to have been broken. In the apartheid era, a clear line was drawn between
struggles for formal equality and those for substantive equality. Formal equality –
the achievement of equal rights and opportunities – was regarded as an inadequate
conceptualization of liberation. The achievement of formal political and civil
rights, while an important gain in itself, was understood as a weak form of equal-
ity that would have little impact on the lives of poor women. What was needed
was substantive equality, understood as the transformation of the economic con-
ditions that produce gender equality.6 The Women’s Charter for Effective Equality,
adopted by the Women’s National Coalition in 1994, articulates a notion of equal-
ity that is closer to the vision of substantive equality, with a very clear emphasis on
the structural and systemic underpinnings of women’s subordinate status.
Definitions of equality of course need to be grounded in the particular context
in which claims are being made. I would argue that a strong notion of equality,
126 Shireen Hassim
one that would provide some guidance about appropriate policy choices in South
Africa, would rest on the extent to which formal discrimination in law and policy
is reduced,7 the extent to which overall poverty is reduced, the degree to which
women have autonomy and are able to make choices free of the constraints of care
work within families and communities, as well as free of the pressure to remain in
oppressive and violent relationships (Orloff 1993: 319),8 and the extent to which
women feel safe in society. This notion of equality has specific implications for
social policy, as it would require that resources be directed in such a way that they
serve not only to address the needs of the poorest women, but also become part of
an incremental process of enhancing women’s autonomy and full participation in
political and economic processes.
In South Africa and in other developing-country contexts the conceptualization
of equality primarily in terms of labour-market participation and autonomy from
family responsibilities is flawed. Firstly, the issue of access to labour markets is
different in the context of high and chronic joblessness. Secondly, the caring
needs of middle-class women have not resulted in demands for publicly funded
systems of family support. The availability of cheap childcare and domestic work
for the middle class meant that there was virtually no demand for high quality
state-provided childcare. The economic and political processes that accompanied
demands for greater public responsibility in providing care in Northern welfare
states (Lewis and Giullari, chapter 8 of this volume) do not exist in South Africa.
In South Africa the supply (of private childcare) has outstripped demand and
consequently the bulk of care work is likely to remain within the private sphere.
Conclusions
This chapter has shown that although the transition to democracy has led to the
elaboration of a wide-ranging set of civil, political and social rights, the gendered
patterns of poverty and inequality have not been significantly reduced. The dera-
cialization of social assistance has included some poor women in the social security
safety nets, but the gap in coverage excluded the majority of poor women between
the ages of 14 and 60. The extent to which poor women have been able to access
their citizenship rights has been limited by the faltering political will to address
poverty in a comprehensive manner, by an overarching macroeconomic frame-
work that prioritizes fiscal restraint over redistribution and by an administrative
system and infrastructure that is unable to fulfil basic tasks of service delivery.
Despite some progress in using social grants as part of a developmental strategy
and despite the increasing monetary expenditure on the social sector, the type of
welfare system being built in South Africa retains many of the features of a residual
system, reacting only to the worst effects of market or family failures and provid-
ing assistance to social groups seen as ‘deserving’. Women access support only
through their children until they reach the age of 60. Provision of benefits is not
generous, benefits are not universal but income-tested, and access to benefits is
difficult. Unless the basis of entitlements changes in ways that recognize women’s
ntitlements as citizenship rights, poor women will continue to be excluded from
Gender Equality and Developmental Social Welfare in South Africa 127
Acknowledgement
I would like to thank Beth Goldblatt for many hours of debate about gender and welfare in
South Africa, and for her critical reading of earlier versions of this article. Thanks also to
Francie Lund for her thoughtful comments and corrections of an earlier draft.
Notes
1 Notably, however, family allowances for large low-income families, instituted in 1947,
did not include Africans.
2 Lund (1988) points out that welfare agencies accepted these changes because they were
highly dependent on state subsidies. However, some did make private arrangements to
equalize salaries of their social workers.
3 See Patel 1992, chapter 3, for a useful discussion of these initiatives.
4 However, Goldblatt (2005) argues that the removal of the parental allowance (almost
exclusively accessed by mothers) that was part of the state maintenance grant denies
women access to money that they were in the past able to claim as their own entitlement.
5 Paradoxically, the Democratic Party, which in many respects can be seen as the party of
business, has supported the introduction of BIG.
6 Elsewhere Catherine Albertyn and I have argued that even this conceptualization of
freedom is limited, as it fails to address the social and cultural dimensions of inequality.
See Albertyn and Hassim (2003).
7 Of course, certain forms of gender discrimination are both inevitable and desirable, such
as in relation to maternity and reproductive health benefits.
8 Orloff (1993) takes this argument much further in suggesting that social policies should
aim at decommodification of gender relations by enabling women to form and maintain
autonomous households. I am hesitant to apply this notion to women in the South
African context, given the particular cultural attachments and support systems that
women value within family and communities. It could also be argued that the high num-
ber of women-headed households in South Africa suggests that women are indeed free to
form autonomous households, but this has patently not empowered women to become
full and equal citizens.
Bibliography
Albertyn, C. (2003) ‘Contesting Democracy: HIV/AIDS and the Achievement of Gender
Equality in South Africa’. Feminist Studies, Vol. 29, No. 3: 595–615.
Albertyn, C. and S. Hassim (2003) ‘The Boundaries of Democracy: Gender, HIV/AIDS and
Culture’. Development Update 4(3): 138–64.
Bernstein, Ann (2005) ‘Let the Market Rule in Welfare’. Business Day, 4 May.
Budlender, Debbie (2004) ‘Women and Poverty’, at www.genderstats.org.za/poverty.html,
accessed 30 August 2005.
CASE (2000) A Social Impact Study on the Phasing-in of the Child Support Grant. Johannesburg:
CASE.
Case, A. and A. Deaton (1996) Large Cash Transfers to the Elderly in South Africa. Research
Program in Development Studies, Princeton NJ: Princeton University.
128 Shireen Hassim
EPRI (2002) ‘Social Security Reform and the Basic Income Grant for South Africa’. EPRI Policy
Report, 31.
Gelb, Stephen (2004) Inequality in South Africa: Nature, Causes, Responses. Johannesburg: Edge
Institute.
Goldblatt, B. (2003) ‘Teen Pregnancy and the Abuse of the Child Support Grant: Addressing
the Myths and Stereotypes’. Agenda, No. 79: 59–81.
Goldblatt, B. (2005) ‘Gender and Social Assistance in South Africa’s First Decade of
Democracy: A Case Study of the Child Support Grant’. Politikon, November.
Government of South Africa (1996) Constitution of South Africa.
Government of South Africa (1997) White Paper on Social Welfare.
Guthrie, T. (2002) ‘Family Social Security Benefits in South Africa’. Social Dynamics 28(2):
122–45.
Harman, Eva (2003) ‘Social Grants and their Social Circulations’. Paper presented at the
Second Annual USBIG Conference of the Eastern Economic Association, 21–23 February.
Hassim, S. (2002) ‘A Conspiracy of Women: The Women’s Movement in South Africa’s
Transition to Democracy’. Social Research, 69(3), Fall: 693–732.
Koopman, Andre (1999) ‘Poor Urged to Roll Up Their Sleeves’. Cape Times, 25 May.
Liebenberg, S. (1998) ‘Social Security and Human Rights: Current Issues’. Economic and Social
Rights in South Africa Review, 1(2): 1–2.
Lund Committee (1996) Report of the Lund Committee on Child and Family Support. Pretoria:
Government Printers.
Lund, F. J. (1993) ‘State Social Benefits in South Africa’. International Social Security Review,
46(3): 5–25.
Lund, Francie (1998) ‘State Restructuring of Welfare’. Transformation 6.
Lund, Francie (2002) ‘Crowding in Care, Security and Micro-enterprise Formation: Revisiting
the Role of the State in Poverty Reduction, and in Development’. Journal of International
Development, 14(6): 1–14.
Lund, Francie (2004) ‘Gender and Social Security: What We Know, What We Think We Know
and What We Really Don’t Know’. Paper presented at Centre for Applied Legal Studies
Workshop in Gender and Social Security, December.
Lund, Francie, with Sally Ardington and Mary Harber (1996)’Welfare’. In Debbie Budlender (ed.)
The Women’s Budget. Cape Town: IDASA.
Matisonn, Heidi and J. Seekings (2002) ‘Welfare in Wonderland? The Politics of Basic Income
Grant in South Africa, 1996–2002’. Paper presented to the BIEN 9th International Congress,
Geneva, 12–14 September.
McCord, A. (2003) ‘Public Works as a Component of Social Protection in South Africa’. Paper
presented to the Development Policy Research Unit/ Trade and Industry Policy Secretariat
Forum, The Challenge of Growth and Poverty: The South African Economy Since
Democracy, 8–10 September.
Ministry of Social Development (2004) ‘Press Release on Income Support Findings’,
5 December.
Mkandawire, Thandika (2004) ‘Social Policy in a Development Context: Introduction’. In
T. Mkandawire (ed.) Social Policy in a Development Context. Basingstoke: Palgrave.
Orloff, Ann Shola (1993) ‘Gender and the Social Rights of Citizenship: The Comparative
Analysis of Gender Relations and Welfare States’. American Sociological Review, 58(3):
303–28.
Naidoo, Ravi (2003) ‘The Basic Income Grant: Poverty, Politics and Policy-making’. Harold
Wolpe Memorial Lecture, University of Natal, August.
Parliament of South Africa (2000) Minutes of Portfolio Committee on Welfare, 13 October
2000.
Patel, Leila (1992) Restructuring Social Welfare: Options for South Africa. Johannesburg: Ravan
Press.
Posel, Dorrit (2001) ‘How do Households Work? Migration, the Household and Remittance
Behaviour in South Africa’. Social Dynamics 21(1): 165–89.
Gender Equality and Developmental Social Welfare in South Africa 129
Rhupiso, Victor (2003) ‘Dumped with Daddy’. Sunday Times Metro, 23 February.
Samson, Michael J. (2002) ‘The Social, Economic and Fiscal Impact of Comprehensive Social
Security Reform for South Africa’. Social Dynamics, 28(2): 69–97.
Seidman-Makgetla, Neva (2004) ‘Women and the Economy in South Africa’, at www.
genderstats.org.za/economy.html, acccessed 30 August 2005.
Taylor Committee (2002) Transforming the Present – Protecting the Future. Report of the
Committee of Inquiry into a Comprehensive System of Social Security for South Africa. Pretoria:
Department of Social Development.
Van der Berg, Servaas (1997) ‘South African Social Security under Apartheid and Beyond’.
Development Southern Africa, 14(4): 481–503.
Van der Berg, Servaas (2001) ‘Redistribution Through the Budget: Public Expenditure
Increase in South Africa, 1993–1997’. Social Dynamics, 21(1): 140–64.
Vetten, Lisa (2005) ‘ “Show me the Money”: A Review of Budgets Allocated Towards the
Implementation of the Domestic Violence Act’. Politikon, November.
Walker, Cherryl (1981) Women and Resistance in South Africa. London: Onyx.
Walker, Cherryl (2001) ‘Piety in the Sky? Gender Policy and Land Reform in South Africa’.
Paper presented at the UNRISD Conference on Agrarian Change, Gender and Land Rights,
Geneva, 6–7 November 2001. Geneva: United Nations Research Institute for Social
Development.
Wells, Julia C. (1992) We Now Demand! The History of Women’s Resistance to Pass Laws in South
Africa. Johannesburg: University of Witwatersrand Press.
Woolard, I. (2003) ‘Social Assistance Grants, Poverty and Economic Growth in South Africa’.
Research paper, Development Policy Research Unit, University of Cape Town.
6
Social Policy Reforms and Gender in
Japan and South Korea
Ito Peng
Introduction
Feminists are sharply divided into two camps on the issue of family and gender
policies in Japan and South Korea (henceforth Korea). On the one hand are those
who claim that despite the remarkable developments in the social, economic and
political spheres, very little has changed in terms of women’s status (Moon 2002).
Proponents of this ‘no real change’ perspective point to these countries’ poor
records in the Human Development Report (HDR) indices on gender equality
(Gender Development Index or GDI, and Gender Empowerment Measure or GEM)
as evidence of the persistence of gender inequality. Indeed, both Japan and Korea
are amongst the lowest in the international rankings, and Japan has fallen even
further behind in recent years despite the government’s claimed efforts to advance
gender equality.1 Based on these figures, there is good reason for scepticism about
these developmentalist economies’ abilities to ensure gender equality. As feminist
studies have shown, economic growth in many East Asian countries has occurred
simultaneously with gender inequality in both public and private spheres
(Seguino 1997; Hsuing 1996).
On the other hand, there are those who argue that despite their poor standings
on international measures, noticeable changes have been made in terms of gender
equality in Japan and Korea (Osawa 2000, 2003; Lee and Park 2003). Rather than
focusing solely on outcome measures as the means to evaluate progress towards
gender equality, more attention needs to be paid to the processes of policy change
and to how gender issues are being incorporated into social policies.2 Evidence
suggests that a wide range of policy measures have been introduced in both Japan
and Korea since 1990 to address the issues of family and gender, and attempts have
been made in the direction of gender mainstreaming.
This chapter cautiously adopts the latter approach, examining family and gender
policy reforms in Japan and Korea in the 1990s with an aim to explaining why
these reforms have been made, and what they mean in terms of progress towards
gender equality. I argue that some important changes have been made in social
welfare policies to support family and gender equality, and that the reasons for the
130
Social Policy Reforms and Gender in Japan and South Korea 131
changes are both sociological and political. In both countries changes resulting
from demographic ageing and low fertility, greater participation of women in the
labour market and in politics, and political regime shifts have created pressures on
the government to push for gender equality. However, there remain significant
structural and institutional barriers to women’s full participation in the labour
market and in politics, and a significant gap exists between individual ideas and
policy claims about gender equality on the one hand, and actual practice on
the other. The reasons for policy changes in relation to the family and gender are
post-industrial changes and political regime shifts. The two most important post-
industrial changes are firstly, demographic ageing and declining birth rates and,
secondly, changes in family and gender relations.3 In Japan and Korea the rapid
ageing of the population combined with declining fertility has sparked fears of an
increased dependency rate, labour shortages, greater social security burden, econ-
omic slowdown, and ultimately population decline. To exacerbate the situation,
the increase in women’s labour-market participation rates and widening genera-
tional divide in the ideas about family and gender relations have brought issues of
child and elderly care, and the role and capacity of the family in providing indi-
vidual welfare, to the centre of policy debates. These have undoubtedly created
imperatives for policy change.
Both countries also experienced political regime shifts in the 1990s, which
disrupted the earlier structure of social policy-making, creating opportunities for
new actors to engage in policy-making process and for new policy ideas to be
pushed forward. In Japan political realignments following the collapse of the con-
servative Liberal Democratic Party’s (LDP) political hegemony in 1993 created a
number of opportunities for social policy reforms. In Korea the Asian economic
crisis of 1997 created openings for political regime shift and enabled subsequent
governments to initiate welfare expansion. In both Japan and Korea, important
changes in family and gender policies happened therefore in the context of polit-
ical regime shifts. I argue that while post-industrial pressures may have forced
Japanese and Korean people and governments to re-think social policies, changes
in domestic political dynamics provided important openings for social and politi-
cal actors to carry through the reform agenda.
This view of social policy changes in Japan and Korea stands against the earlier
developmentalist perspective of East Asian welfare states which sees social poli-
cies as subjugate responses to economic development. In highlighting the
importance of the post-industrial and political factors, this chapter challenges
such a static view of East Asian welfare states, and offers an alternative perspective
that explains how structure and politics can facilitate welfare state change.
Following a brief discussion of the limits of the developmentalist view of
East Asian welfare states, I compare domestic post-industrial pressures confronting
Japan and Korea. I will then outline social policy reforms in these countries
since the 1990s, and explain how structural and political changes affected wel-
fare state restructuring in these countries. Finally, in the last section I will reflect
on the implications of family and gender policy reforms for women in these
countries.
132 Ito Peng
The comparative study of East Asian welfare states has come a long way since the
early 1990s, when scholars began to undertake more systematic studies of
countries such as Japan and Korea. Building on the idea of the developmental
state, many first-generation East Asian welfare-state scholars focused on the role of
the state and the bureaucracy in shaping economic and social policies (Goodman
and Peng 1996; Goodman, White and Kwon 1998). Growing out of this literature,
the economic productivist approach sought to explain welfare developments in
East Asia in terms of the state’s prioritization of economic development, and hence
the subordination of social policies to economic and industrial policies. Holliday
(2000), for example, argues that in Japan and Korea while the state underpins the
market and families with some universal programmes such as health insurance
and pensions, only those linked to the ‘productive’ elements of society, typically
the male workers in core industries, are accorded minimal social rights.4
Alternatively, others examined the Confucian underpinnings of the East Asian
welfare regime. Jones (1990, 1993) has pointed out the pervasive influence of
the Confucian tradition underlying the success of East Asian welfare states. The
Confucian principles of filial piety and industry are thought to contribute to a
strong familial sense of welfare obligation and self-help. The notions of gender
role segregation and women’s subordination to men and to the family are also
used by the state as the rationale for its reliance on women’s unpaid work at home
and in the community (Palley 1992; Sung 2003). These factors in turn shape the
normative foundation of the dual labour-market structure that overwhelmingly
privileges male workers in core industries, and leads to strong assumptions about
women’s welfare responsibilities.
These perspectives, however, are being challenged. First, social policy has
become much more politicized in the last couple of decades. In Japan, social policy
and welfare reform have become an important policy issue and political parties
have became more vocal in advocating particular social policy choices. In Korea
while the democratization process helped to mainstream social policy, the Asian
economic crisis of 1997 further highlighted the political salience of social policy in
electoral politics. Simply put, the political currency of social policy in both Japan
and Korea rapidly appreciated in the 1990s.
Second, the productivist welfare-state argument also faces some difficulties in
explaining the evident expansion of social welfare programmes aimed at the ‘non-
productive’ sectors of society in the 1990s. If the productivist welfare-state logic pre-
vails, social policies in these countries should be more focused on the ‘productive’
sectors of the society as the economy weakens, and one would expect preferential
allocation of resources to ‘productive’ sectors and greater reliance on the family’s wel-
fare responsibilities, as was the case for Japan in the 1980s. In the 1990s, however, the
exact opposite dynamics were observed. In Korea significant social welfare expansion
followed after 1997; in Japan, ongoing economic malaise notwithstanding, welfare
expansion continued. What is more, not only was there expansion in social services,
Social Policy Reforms and Gender in Japan and South Korea 133
most of the social policy reforms were targeted to ‘non-productive’ sectors of these
societies such as the elderly, children and the unemployed.5 In Japan, social care for
the elderly and children expanded alongside reforms to strengthen gender equality
and to protect women and children in vulnerable situations. In Korea, universaliza-
tion of healthcare benefits and pensions took place together with radical reforms of
unemployment insurance, social welfare, and family and gender policies. The focus
on ‘non-productive’ sectors of society is further confounded by the new emphases on
gender equality, and work and family harmonization, concepts that are clearly at
odds with Confucian principles of gender role segregation.
Of course, one may argue that these reforms could simply be counter-cyclical mea-
sures, and that state policies remain basically ‘productivist’ in nature. True, job cre-
ation through public investment can be an effective counter-cyclical measure in times
of economic crisis. The promotion of active labour-market policies in both countries
can be seen as essentially ‘productivist’ as well. However, this begs two questions.
First, why social welfare, and why at this time. In the past, the typical counter-cyclical
measure for both the Japanese and Korean governments was to work with employers
to sustain high rates of employment, and if necessary, governments used public
construction and other forms of public-works programmes to kick-start the economy.
Indeed, that was precisely what the Japanese government did in the early 1990s, until
they ran out of steam with little success and a massive government deficit. These mea-
sures made ‘productivist’ sense because they targeted the productive sectors of society
and provided male employment. Implicit in the productivist logic was a highly gen-
dered notion of economic growth: productivism through protection and enhance-
ment of male workers in productive sectors. Promoting gender equality in the labour
market was not part of the productivist counter-cyclical logic. Even if we see the cur-
rent welfare-state expansion as part of the same productivist counter-cyclical logic,
we have to explain why there has been a shift in policy understandings of both the
causes of economic malaise and solutions proposed to counter it. What are the
underlying causes of this shift in policy thinking, and what are its implications?
Second, the idea that East Asian welfare states are somehow uniquely produc-
tivist may be misplaced. As illustrated by the historical experiences of Nordic wel-
fare states, social policies have always been implicitly linked to economic
development. Indeed, as Mkandawire (2004) points out, social policy is basically
developmental. What is important, then, may be not the productivist nature of
social policy reforms, but rather the mechanisms and motivations underlying pol-
icy changes. In other words, we need to move beyond identification and classifi-
cation of East Asian welfare regimes to analysing causal processes of change.
Welfare-state expansions in Japan and Korea therefore require more in-depth
examinations which go beyond description.
Post-industrial pressures
Demographic shifts
Japan and Korea have been experiencing similar demographic changes over the
last few decades. In both countries population is ageing and fertility is declining
134 Ito Peng
Table 6.2 Total average fertility rate in Japan and Korea, 1970–2030
Sources: NIPSSR, Population Statistics, 2004; KNSO (2005) Future Household Projection.
(Tables 6.1 and 6.2). Although Korea is still demographically young compared to
Japan and other OECD countries,6 the pace of ageing is rapid. As shown in
Table 6.1, the elderly population in Korea will constitute about 23 per cent of the
total population by 2030. The main cause of demographic ageing in both
countries is the low fertility rate. Given the current total fertility rate, demographic
ageing in Korea will occur within an extremely compressed time frame. The issue
for Korean policy-makers therefore is not simply that of declining fertility and an
ageing population, but more importantly, the speed at which these demographic
changes are taking place. There is concern that Korea will not be prepared for
changes of this magnitude in terms of its social infrastructure.
1980 2000
Japan
Nuclear 60.3 58.4
Three-generation 19.7 13.6
Single-person 18.1 24.1
Korea
Nuclear 76.8 75.0
Three-generation 17 8.4
Single-person 4.8 15.5
Table 6.4 Divorce rate (crude divorce rate per thousand population), 1980–2002
% of
single
parent *
Total Widowed Divorced Unmarried Others families
* Proportion of single parent families in relation to all families with children under 18 years of age.
Source: MHWL (2005) Report of the Heisei-15th year National Survey of Single Mother and Other Families (Heisei 15-nen do
Zenkoku Boshisetai tou chousa kekka houkoku); NIPSSR, Population Statistics, 2004.
136 Ito Peng
Surveys of public attitudes towards marriage and divorce also suggest that
divorce is becoming more socially acceptable in Japan and Korea (Cabinet Office –
Japan 2005; Hong 2005; Kim 2003b). Moreover, there are also some changes in the
intergenerational contract, the most significant being the weakening of the
parental expectation of receiving support from children in old age, and a corres-
ponding decline in children’s expectation of supporting parents in their old age
(Cabinet Office, Japan 2005; Retherford et al. 2001).
Finally, employment structures in both countries have changed over the last
couple of decades. There has been a steady informalization of work throughout
the 1990s. In Japan, the proportion of people in non-standard forms of work
increased from 20.9 per cent to 30.4 per cent between 1995 and 2003 (MHWL
2004), while in Korea, the proportion of those employed in non-standard work
rose from 45.5 per cent in 1990 to 52.1 per cent in 2000 (KNSO 2005).7 In both
countries the majority of non-standard workers are women. The employment rate
of married women in Japan rose from 48 per cent in 1980 to 56 per cent in 2000,
while that in Korea increased from 40 per cent to 48 per cent over the same
period.8 In both cases, while the informalization of work has pulled more women
into the labour market, it has also widened the gap between child and elderly care
needs and the availability of unpaid care provided by women.
The two governments’ main response to these social and demographic changes
has been to ‘activate’ the labour-market participation of women and the younger
elderly, while encouraging childbirth. This approach has been preferred to develop-
ing policies to encourage immigration, for example. In Korea, the postwar family
planning policy was officially terminated in 1996, and a number of woman-
friendly and gender-positive policies have been introduced to encourage women’s
social, political and labour-market participation. In Japan a series of woman-friendly
policies were introduced in the 1990s to help families harmonize work and family
responsibilities. These include: the Gold Plan in 1989, the Angel Plan in 1994,
Social Policy Reforms and Gender in Japan and South Korea 137
introduction of parental leave and flexible hours for workers with childcare
responsibilities in 1996, and the Long-term Care Insurance in 1997.
Social expenditure as a proportion of GDP in Japan grew from 11 per cent to
17 per cent between 1990 and 2000, and in Korea from approximately 3 per cent
to 7 per cent between 1990 and 1999 (OECD 2004). Although a large portion of
these expenditures is accounted for by pension and health insurance, social wel-
fare expenditures also grew. In Japan, social welfare expenditure doubled from
¥4.8 trillion in 1990 to ¥10.9 trillion in 2000 (NIPSSR 2002), most of which was
accounted for by the expansion in long-term elderly care and childcare. In Korea,
social welfare expansion was even more remarkable: government expenditure on
the Employment Insurance Program increased from 4.7 million Won in 1996 to
306,172 million Won in 1999; the budget for childcare rose from 41,876 million
Won to 436,903 million Won between 1991 and 2002; and elderly welfare
increased from 37,861 million Won in 1990 to 407,767 million Won in 2003
(Lee and Park 2003).
gender equality more systematically. The law included provision for the
institutionalization of gender-equality principles throughout the national and
local governments, and the establishment of a monitoring system to measure the
level of change. The Council also broadened its influence throughout the 1990s.
For example, its policy proposals have been incorporated in the recent social
policy reforms (Cabinet Office, Japan 2005; Osawa 2003),10 including reforms to
divide the employee pension between husbands and wives, allowing divorced
women the right to claim half of the husband’s pension, and the extension of the
employee pension to part-time workers. Similarly, the Council has also introduced
a proposal to abolish the special tax break for married women, which was criti-
cized by feminists for discouraging married women’s full-time employment and
privileging middle-class male breadwinners. It has also been successful in pushing
for family policy reforms through the Low Fertility Measures Plus One Program
(Shoshika Taisaku Plus One), which includes targets of abolishing waiting lists for
public daycare, increasing the parental leave take-up rate to 10 per cent for fathers
and 80 per cent for mothers, a minimum five-day paternity leave, and shorter
working hours for workers with small children.
The policy-making process has also become more open to women’s groups. The
LTCI legislation was one of the first cases of direct engagement by women’s groups
in the social policy-making process (Eto 2001; Peng 2002a). The Council of Gender
Equality’s core group, for example, consists of 25 members, 13 of whom are min-
isters of various government ministries, while 12 others are ‘experts’ made up of
university professors and representatives of business and NGOs, most of whom are
women. In addition, the Council also set up a community advisory committee
called Egalité Network comprised of representatives from 88 different civil society
groups. Women politicians are also taking a more active role in policy reform. The
Domestic Violence Law (1999) and Child Abuse Prevention Law (2004), for
example, both involved a special parliamentary bill (gi’in rippo) put forward by a
cross-party coalition of women MPs (Gelb 2004; Iida 2004). This method of legal
reform involves MPs making a draft bill proposal directly in the parliament, and
bypassing the bureaucracy.
Table 6.7 Changes in the population coverage of the Four Major Social Insurances
in Korea
1997 2000
EI WCI PI HI EI WCI PI HI
Firm size ⫺5 X X X O O O O O
(number of regular 5–9 X X O O O O O O
employees) 10–29 O O O O O O O O
30⫹ X O O O O O O O
Daily workers X X X O 䊐 䊐 O O
Temporary workers X X X O 䊐 䊐 O O
Self-employed X X X O X O O O
Unpaid family workers X X X O X O O O
Coverage for other social insurance schemes also expanded after 1997. The
National Pension Plan (NPP) was extended to all workers, including daily, tempo-
rary, self-employed and family workers (Kim 2003). It is, however, important to
point out that the outcome in relation to women is still small. Although the total
number of people insured by NPP has more than doubled between 1990
(4,651,678) and 2002 (12,248,483), women make up less than a third of the total
(29.5 per cent in 1990, and 30.1 per cent in 2001). The percentage of women who
are covered by NPP increased from 6.4 per cent of the total population in 1990 to
15.6 per cent in 2002, but this is still significantly lower than the coverage given to
men (15.2 per cent and 36.7 per cent respectively) (KWDI 2004). The main reasons
for this are women’s lower employment rates and the newness of the NPP. The
National Health Insurance coverage, on the other hand, has expanded since
the end of the 1970s, and currently nearly the entire population is covered.
However, the level of co-payment (user fees) for NHI is still relatively high, making
it difficult for low-income families to utilize healthcare.
The Livelihood Protection Programme, which provided basic income support to
the very poor, was also reformed in 1998 to become the National Basic Livelihood
Security Programme (NBLS), resulting in a 50 per cent increase in the number of
recipients within a year (from 1 million to 1.5 million) (Sung 2003). A work-fare
component called the ‘self-reliance policy’ was added to the NBLS in 1999 to
engage recipients in public-works programmes, while providing them with medical
assistance, childcare and school fees, and counselling services. In 2001, approxi-
mately 46,000 people were involved in this programme, 70 per cent of whom were
women, and half of those were middle-aged single mothers (Lee and Park 2003:
16). The expenditures for NBLS nearly quadrupled from 900 billion Won in 1997
to 3,400 billion Won in 2002 (Lee 2005).
The Mother-Child Welfare Programme was introduced in 1999 to provide
housing support, job training, and employment and childcare assistance to
140 Ito Peng
The changes described above suggest that the 1990s was indeed a watershed
decade for both Japan and Korea. Faced with similar social and demographic
imperatives, both countries had little choice but to undergo welfare-state restruc-
turing. However, translating these imperatives into policy reforms requires political
will. As illustrated by the experiences of European welfare states, despite similar
post-industrial pressures, social policy responses have diverged, largely because of
different domestic political conditions and policy options.
Similar post-industrial pressures in North America, for example, did not translate
into social policy expansion in either Canada or the US. It would be therefore
premature to draw direct relationships between post-industrial pressures and
welfare-state expansion in Japan and Korea. It is important to trace mechanisms
through which these pressures have been translated into policy changes. For this
we need to turn our attention to the domestic political dynamics. In Japan and
Korea, the structural changes of the 1990s also coincided with political regime
shifts: in Japan with the end of the conservative Liberal Democratic Party’s (LDP)
political hegemony in 1993, and in Korea with the transfer of power from the
conservative party to the opposition government in 1998. In both cases, the shifts
disrupted the political status quo and opened opportunities for change.
Social Policy Reforms and Gender in Japan and South Korea 141
Japan
In Japan the elderly population had already reached 12 per cent by 1990, and the
basic social security system, introduced in the early 1960s, was beginning to
experience structural pressures associated with the maturing of the system. Not
surprisingly, then, a great deal of the increase in social security expenditure in the
1990s did result from rising pension and healthcare costs owing to population age-
ing. The important policy shift in the 1990s was therefore not so much associated
with the increase in health and pension benefits, which would have been
expected, but rather with the unexpected expansion of social care services. The
causes of social care expansion were in fact functional as well as political. While
post-industrial changes did affect people’s thinking about social welfare, not much
happened until the 1990s. Social care reforms in Japan are therefore a good
example of post-industrial imperatives facilitated by political regime shift.
In Japan, the so-called ‘crisis of the family’ had been brewing since the 1980s
when women’s groups began to agitate for more social care. This idea, however,
did not make much impact on social policy debates in the 1980s, until the fertil-
ity rate declined below the historical low of 1.58 in 1989. By the early 1990s the
idea of family crisis was promoted not only by the government but also by the
media and civil society groups. The image of the crisis ranged from the notion of
the family as a ‘caring hell’ (kaigo jigoku) for women (Higuchi 1997; Koreika Shakai
wo Yokusuru Josei no Kai 1998; NFDR 1997, 2001), to young people being depicted
as ‘Parasite Singles’ (Iwakami 2000; Yamada 1999), and the fear of labour shortage
and population decline (MHW 1997).13 Underlying these disparate characteriza-
tions was a shared sense of crisis about the adverse effects of social and structural
changes, especially fertility decline and rapid ageing, on the Japanese family and
society.
Work–family tensions for women were evident. Not only was the married
women’s employment rate increasing, but more women were working outside of
the home. The proportion of women in self-employed work or family businesses
declined from 39.0 to 18.4 per cent between 1975 and 1999, while the proportion
of women working outside the home rose from 59.8 to 81.4 per cent over the same
period (NFDR 2001). The majority of women were also experiencing stress and
psychological burdens as a result of having to care for their elderly relatives while
working (Ministry of Labour 1998). Opinion polls showed a decline in support for
the idea of the male-breadwinner/female-housewife model, and an increase in
support for women’s lifetime employment (Cabinet Office, Japan 2005). Women
in their twenties and thirties were less willing to accept the traditional division of
labour within marriage and were more willing to consider divorce as an option
(Retherford et al. 2001).
This shared sense of the family being in crisis due to social and economic
changes in Japan emerged in the context of a political regime shift. The collapse of
the LDP’s political dominance in 1993 led to a chaotic political scramble as politi-
cians and parties tried to realign themselves in a new and uncertain political situ-
ation. Within a few years, no less than a dozen new political parties appeared and
disappeared, while politicians jumped from one party to another in an effort to
142 Ito Peng
the former military elites (the Hanahoe) and the jailing of the coup leaders under
the military regimes. Like post-1993 Japan, in Korea new political parties emerged
and new political alliances were forged and broken, as politicians vied for political
advantage (Kim 2000). In the meantime, civil society groups also reorganized, and
shifted their focus from the broader democracy movement to specific social policy
advocacy. As Seong (2000) points out, the civil society movement in Korea after
1987 entered the mainstream as new civic organizations, representing diverse
popular concerns such as environmental protection, social welfare, women’s
issues, educational reforms and crime control, began to gain hegemony over
special-interest groups such as employers’ associations and labour unions. They
pushed for welfare-state expansion and the mainstreaming of social policies in the
1990s.
The suddenness of the Asian economic crisis in 1997 caught everyone off guard:
overnight, foreign investments were pulled out, stripping Korea of its assets and
reserves; stocks plunged; the economy halted; and finally, in what President Kim
Dae-Jung called ‘the country’s second humiliation’ (the first being its colonization
by Japan from 1910 to 1945), Korea had to turn to the IMF for an emergency bail-out.
The IMF conditions were stringent: currency devaluation, economic liberalization
and labour-market restructuring. Ironically, it turned out to be a watershed for
political regime shift in Korea. In the 1997 presidential election, public discontent
with the Kim Young Sam government’s ineffectiveness in handling the crisis
turned into a swing vote for the opposition candidate Kim Dae-Jung who cam-
paigned on the platform of social policy reform on the one hand, and modernization
and responsibility on the other.
Kim Dae-Jung’s electoral coalition was far from united. Indeed, from the
beginning significant tensions existed between the pro-welfare and pro-business
interests within his camp. The imperatives of the economic crisis, however, offered
him the opportunity to create a government–employer–labour tripartite council
and to force the ‘grand consensus’, a two-pronged, if not contradictory, strategy of
economic liberalization and welfare-state expansion. The labour movement
agreed to radical market liberalization as directed by the IMF, in exchange for the
expansion of the social safety net. Clearly many of the social policy reforms in
Korea took place against the backdrop of social and economic imperatives follow-
ing the economic crisis in 1997. However, we should not understate the impact of
the economic crisis on public support for change and on civil society groups’
willingness to back a reformist leader like Kim Dae-Jung. Certainly, pro-welfare
civil society groups constituted Kim Dae-Jung’s support base. Once the welfare-state
reforms were introduced they became the basis from which civil society began to
push for further expansion.
Even after the economic crisis the expansionary reforms continued. With the
economic crisis behind, the social policy reforms were increasingly informed by
the idea of changes in demographic patterns and family and gender relations.
Policy debates on elderly care and childcare expansion since 1998, for example,
mirror the Japanese family policy debates. Despite its relatively young population,
demographic ageing and low fertility are now the priority issues in the Korean
Social Policy Reforms and Gender in Japan and South Korea 145
government’s social policy agenda (Park 2005; Lee 1999b; Cho 2000).17 Long-term
Care Insurance pilot projects (based on the Japanese model) will begin in 2007.
Similar attempts are being made in terms of the expansion of public childcare services.
The policy focus has also shifted from the economic crisis to the persistence of
low-income households, particularly those headed by women (Lee and Park 2003),
with much of the recent family and gender policy reforms being informed by
studies of poverty among women. Lee (1999) found that women were particularly
hard hit by the economic crisis, with two-thirds of the poor and four-fifths of poor
elderly people being women. Government studies also show the negative implica-
tions of the ‘dismantled’ families not only as they relate to financial hardships, but
also in relation to problems associated with childrearing, caregiving, and emotional
status. Finally, the combination of defamilialization and increased employment
among married women has also been identified as creating problems for elderly
care (Sung 2003; Lee and Park 2003). In summary, the mainstreaming of social
welfare in Korea since 1987 was propelled by the economic crisis that brought
about a shift from pro-business to a pro-welfare government. Once the crisis was
over, social policies remained informed by the idea of the family crisis caused by
the changes in demographic structure, and family and gender relations. Like
Japan, social policy reforms occurred in the context of structural and political
changes.
Have any of these policy reforms made a difference to the lives of women in Japan
and Korea? The Gender Empowerment Measure and Gender Development Indexes
indicate very little change. However, at the ground level more subtle changes seem
to be taking place. For example, the number of women parliamentarians in Japan
increased from 26 (3.4 per cent) to 79 (10.8 per cent) between 1980 and 2001,
while their proportion at the prefectural and local government levels rose from
1.2 per cent in 1980 to 7.9 per cent in 2003. More evident changes in women’s
political representation are observed at the city level. In large cities such as Tokyo
and Osaka, the proportion of women politicians increased from 8.9 per cent in
1990 to 21.5 per cent in 2003. There has been an increase in the representation of
women in national policy advisory councils as well: in 1980, less than half of these
councils (46.2 per cent) had a woman member, and women represented a mere
4.1 per cent of all the council members; by 2003, however, the proportion had
changed to 98 per cent and 26.8 per cent respectively (Cabinet Office, Japan 2005).
In Korea, similar changes can be seen. In the 1992 parliamentary election there
were 35 women amongst a total of 1,106 candidates from local and national con-
stituencies (3.2 per cent); of this number, three were elected from a total of 296
seats (1.0 per cent). In the 2004 election, the percentages of women candidates
and electees increased to 10.4 and 13.0 respectively. Part of the reason for the
increase in elected women politicians is the introduction of the proportional
representation system through electoral reform. Similarly women’s representa-
tion in government advisory councils also increased from 9 per cent in 1990 to
146 Ito Peng
27.3 per cent in 2002, and the proportion of women officials in the government
bureaucracy also increased from 23.9 per cent in 1990 to 34.0 per cent in 2003
(KWDI 2004). In both Japan and Korea it appears that some effort has been made
to raise women’s representation in public and political office over the last couple
of decades.
In addition to women’s entry into the government and bureaucracy, there has
been a noticeable change in women’s participation in civil society groups as well.
In Japan, the number of registered NGO groups increased from 1,176 to 14,657
between 1998 and 2003. A national survey of NGOs in 2001 found women made
up nearly half (48.5 per cent) of all workers while elderly and retired people made
up about a third (31.4 per cent) (Cabinet Office, Japan 2005). In Korea, women’s
organizations have emerged as the main advocacy groups in the social policy-
making process (Lee 2005). The number of women’s organizations in Korea
increased from 62 in 1994 to 93 in 2001 (KWDI 2004). Studies of civil societies in
both countries show NGOs are becoming increasingly politically active, and
governments too have become more open to civil society groups’ engagement in
public policy-making (Lee and Peng 2005; Shipper 2005). What these trends
suggest is a possible change in the governance structure vis-à-vis public policy-
making. As social policies in both countries become more decentralized and as
civil society becomes more closely engaged in policy-making, public policies may
be further re-shaped.
Finally, we cannot understate the importance of ideational shifts in these coun-
tries. Public opinion polls in both countries suggest noticeable changes in men’s
and women’s attitudes towards gender relations. In Japan, the proportion of those
who claim that a woman should continue to work even if she has a child has risen
from 26 per cent to 38 per cent for women and 19 to 37 per cent for men between
1992 and 2002. Similarly, during the same period, the proportion of respondents
who endorse the male-breadwinner/female-housewife model has declined from
55.6 to 43.3 per cent for women, and 65.7 to 51.3 per cent for men; while the pro-
portion who reject the model has increased from 38.3 to 51.1 per cent for women
and 28.6 to 42.1 per cent for men (Cabinet Office, Japan 2005). In Korea, similar
attitudinal shifts with respect to women’s employment are observed.
Yet, these gains made at political, community and ideational levels are offset by
significant structural resistance, particularly within the labour market. In both
Japan and Korea employers are reluctant to adopt new gender policies, and the
legislation is inadequately enforced. In Japan, despite the public campaign to
promote parental leave, the rate of take-up has increased only marginally from
46 per cent for women in 1996 to 64 per cent in 2002, and in the case of men it
actually declined from 0.8 per cent in 1996 to 0.3 per cent in 2002 (MHWL 2004).
Workers are reluctant to take full parental leave because of negative pressures from
employers and work colleagues, and the fear of being made redundant (Cabinet
Office, Japan 2005). Workers will be reluctant to exercise their rights if they fear
negative consequences and if the legislation does not fully and actively enforce
their rights. In both Japan and Korea, despite the increase in public childcare
spaces since 1990, a significant number of children are still on the waiting list. In
Social Policy Reforms and Gender in Japan and South Korea 147
both countries, but particularly in Korea, the large and progressive informalization of
employment also undermines women’s ability to achieve economic security. Despite
the extension of social insurance coverage to non-standard workers, a significant por-
tion of them remain uncovered, and more importantly, the social security systems in
both countries remain a poor defence against insecure employment and inadequate
income. These suggest that until the labour market and employment practices
change, the journey towards gender equality will indeed be slow and painful.
Acknowledgement
I would like to thank Kwansei Gakuin University for a travel grant to enable me to conduct
interviews in Japan and Korea during the academic year of 2001/2, and Social Sciences and
Humanities Research Council of Canada (SSHRCC) for the research grant (2003–7) which has
enabled me to continue and develop this research.
Notes
1 The UNDP report on the Gender-related Development Index (GDI), ranked Japan 12th
and Korea 29th amongst 174 countries in 2002. The same report also ranked Japan 38th
and Korea 68th in the Gender Empowerment Measures (GEM).
2 Interview with Mari Osawa, Professor of Gender and Social Policy, Institute of Social
Sciences, University of Tokyo, and also a member of the Gender Equality Council and
the Pension Reform Advisory Council, 17 February 2004.
3 Post-industrial changes also include exogenous factors such as economic globalization
and internationalization. Though clearly important, it will be beyond the scope of this
paper to deal with them adequately.
4 It should be pointed out here that in both Japan and Korea the male workers in core
industries, such as those represented by large business conglomerates, only represent
about 20 per cent of the total labour force.
5 In Japan, the economy stayed stagnant throughout the 1990s. Hope of economic recov-
ery in 1996 was, however, dashed by the Asian economic crisis of 1997. Korea was one
of the worst-hit countries in East Asia, and required financial rescue by the IMF.
6 The current OECD average is approximately 13 per cent.
7 Non-standard work normally means salaried and waged work that is part-time, tempo-
rary, contract or day work. In Japan, the majority of non-standard workers are part-
timers, while in Korea, most of them are contract and day workers.
8 These figures are for both standard and non-standard employment. In both cases the
majority of married women are non-standard workers.
9 The Gold Plan set targets for service expansion in elderly care for the next ten years.
Some examples of expansion are: an increase in the numbers of home helpers from
31,400 to 170,000, day services for the elderly from 4,300 to 17,000, beds in short-stay
centres from 1,000 to 60,000, and elderly healthcare institutions from 28,000 to 280,000
between 1989 and 1999.
10 Interviews with: Mari Osawa, 17 February 2004; Jozuka, director of the Gender Equality
Commission, Home Ministry, 20 February 2004; Akaishi, co-director of Femin, a national
women’s organization in Tokyo, 24 February 2004.
11 The percentage of children under six attending childcare increased from 7.4 per cent to
21.5 per cent between 1995 and 2002 (Lee and Park 2003). In Japan, it increased from
27 per cent to 33 per cent between 1995 and 2000.
12 Interviews with: Hyekyung Lee, Professor, Yonsei University, Korea, 1 February 2002;
Dr Wha-Soon Byun, Dr Yeon-Ran Park, and Dr Dayoung Song, Korea Women’s
148 Ito Peng
Bibliography
Cabinet Office, Japan (2005) Danjo Kyodo Sankaku Hakusho, 2004 (2004 White Paper on Gender
Equality), Tokyo: Cabinet Office.
Cho, Namhoon (2000) ‘Social Impact of Korea’s Low Fertility Rate’. Chislin IIbo, 1 October.
Eto, Murase Mikio (2001) ‘Public Involvement in Social Policy Reform: Seen from the
Perspective of Japan’s Elderly-Care Insurance Scheme’. Journal of Social Policy, 30(1): 17–36.
Democratic Party of Japan (2002) Tayona Lifestyle wo Ikiru Jidai no Jiritsu to Anshin no Seisaku
(Policies for Independence and Security in the Age of Different Lifestyles). Tokyo:
Minshuto Seisaku Chosa Kai.
Gelb, Joyce (2004) Gender Policies in Japan. New York: Palgrave.
Goodman, Roger and Ito Peng (1996) ‘East Asian Social Welfare States’. In Gosta Esping-
Andersen (ed.) Welfare States in Transition. London: Sage.
Goodman, Roger, Gordon White and Huck-ju Kwon (1998) The East Asian Welfare Model:
Welfare Orientalism and the State. London: Routledge.
Higuchi, Keiko (1997) ‘Kaigo no Shakai-ka wo Koshite Susumeta’ (How We Promoted
Socialization of Care). In Higuchi, Keiko (ed.) Chiiki de Miru, Minna de Mitoru: Joseiga
Susumeru Kaigo no Shakaika (Everyone in the Community Takes Care of the Elderly: Women
Promoting Socialization of Care). Kyoto: Minerva Shobo.
Holliday, Ian (2000) ‘Productivist Welfare Capitalism: Social Policy in East Asia’. Political
Studies, Vol. 48: 706–23.
Hong, Joo-Yun (2005) ‘Poll Shows Shift in Korea’s Priorities’. Joong Ang Daily, 23 November.
Hsiung, Ping-Chan (1996) Living Rooms as Factories: Class, Gender and the Satellite Factory
System in Taiwan. Philadelphia: Temple University Press.
Iida, Keisuke (2004) ‘Human Rights and Sexual Abuse: The Impact of International Human
Rights Law on Japan’. Human Rights Quarterly, 26(2): 428–53.
Iwakami, Mami (2000) ‘Shoshi-Koreika Shakai no Oyako Kankei: 30dai wo shoten ni shite
kangaeru (Parent-Child Relationship in the Low Fertility and Ageing Society: with a Focus
on People in their 30s)’. Seikatsu Kyodo Kumiai Kenkyu, Vol. 2: 19–26.
Social Policy Reforms and Gender in Japan and South Korea 149
Jones, Catherine Finer (1990) ‘Hong Kong Singapore, South Korea and Taiwan: Oikonomic
Welfare States’. Government and Opposition, 25 (Autumn): 446–62.
Jones, Catherine Finer (1993) ‘The Pacific Challenge: Confucian Welfare States’. In Catherine
Jones (ed.) New Perspectives on the Welfare State in Europe. London: Routledge.
Kim, Byung-Kook (2000) ‘The Politics of Crisis and a Crisis of Politics: The Presidency of Kim
Dae-Jung’. In Kongdan Oh (ed.) Korea Briefing, 1997–1999: Challenges and Change at the
Turn of the Century. Armonk, New York: M.E. Sharpe, 2000.
Kim, Sang Kyun (2003) ‘A Systemic Reform of Korean National Pension Scheme’. Paper
presented at the Canada-Korea Social Policy Research Symposium, Seoul, 22–23 November.
Kim, Young-hoon (2003) ‘Women Delay Marriage, Report Shows’. Joong Ang Daily, 3 July.
Korean National Statistics Office (KNSO) (1996) Report on Population Census. Seoul, KNSO.
Korean National Statistics Office (KNSO) (2001) Report on Population Census. Seoul, KNSO.
Korean National Statistics Office (KNSO) (2003) Future Household Projection. Seoul, KNSO.
Korean National Statistics Office (KNSO) (2005) Report on Population Census. Seoul, KNSO.
Korean Women’s Development Institute (KWDI) (2004) Statistical Chart on Women. Seoul:
KWDI.
Koreika Shakai wo Yokusuru Josei no Kai (ed.) (1998) Tomodaore ka Tomodachi Shakai he
(Together We All Collapse or Together We Support Together). Tokyo: Akashi Shoten.
Kwon, Huck-ju (2001) ‘Globalization, Unemployment and Policy Responses in Korea:
Repositioning the State?’ Global Social Policy, 1(2): 213–34
Lee, Hyekyung (1999a) ‘Globalization and the Emerging Welfare State – the Experience of
South Korea’. International Journal of Social Welfare, 8: 23–37.
Lee, Hyekyung (2005) ‘Civil Society and Welfare Reforms in Post-Crisis South Korea’. Paper
presented at Canada-Korea Social Policy Research Cooperation Symposium, Munk Centre
for International Studies, University of Toronto, Canada, 26–27 January.
Lee, Heykyung and Yeong-Ran Park (2003) ‘Families in Transition and the Family Welfare
Policies in Korea’. Paper presented at the Canada–Korea Social Policy Research Symposium,
Seoul, 22–23 November.
Lee, Hyekyung and Ito Peng (2005) ‘Civil Society and Social Policy Reform in Japan and
Korea’. Paper presented at Association of Asian Studies Annual Congress, 31 March–4 April,
Chicago.
Lee, Hyun-Seung (1999b) ‘Korea’s Aging Population’. Monthly Chosun, September.
Mainichi Shimbun (2000) Mainichi Shimbun Seron Chosa (Mainichi Shimbun National Opinion
Survey). Tokyo: Mainichi Shimbun.
Ministry of Health and Welfare (MHW) (1997) Heisei 9-nendo Kosei Hakusho. (1997 White
Paper on Health and Welfare), Tokyo: MHW.
Ministry of Health, Welfare and Labour (MHWL) (2001) Heisei 13-nendo Kosei Rodo Hakusho.
(2001 White Paper on Health, Welfare, and Labour). Tokyo: MHWL.
Ministry of Health, Welfare and Labour (MHWL) (2004) Heisei 16-nendo Kosei Rodo Hakusho.
(2004 White Paper on Health, Welfare, and Labour). Tokyo: MHWL.
Ministry of Health, Welfare and Labour (MHWL) (2005) Heisei 17-nendo Kosei Rodo Hakusho.
(2005 White Paper on Health, Welfare, and Labour). Tokyo: MHWL.
Ministry of Labour (1998) Heisei-9 Hataraku Josei no Jijo (Situation of Working Women, 1997).
Tokyo: 21-seiki Shokugyo Zaidan.
Mkandawire, Thandika (ed.) (2004) Social Policy in a Development Context. New York: Palgrave
MacMillan.
Moon, Seung Sook (2002) ‘Carving out Space: Civil Society and the Women’s Movement in
South Korea’. Journal of Asian Studies, 61(2): 473–500.
National Institute for Population and Social Security Research (NIPSSR) (2002) Jinko Dotai
Tokei (Population Statistics). Tokyo: NIPSSR.
National Institute for Population and Social Security Research (NIPSSR) (2004) Population
Statistics, 2004. Tokyo: NIPSSR.
150 Ito Peng
Introduction
151
152 Barbara Hobson
maintained that there were dynamic and interactive processes at work in women-
friendly states: policies that have facilitated participatory citizenship through
incentives to enter the labour market produce active recruitment of women into
unions and politics. These processes laid the ground for mobilizing women’s
constituencies to make claims on the state for more power in the shaping of
welfare institutions (Siim 1994).
Alongside this view that highlights the importance of Swedish social policies in
promoting participatory citizenship in the spheres of family, paid work and poli-
tics (Hernes 1987; Bergqvist 1999), one also finds less optimistic and critical voices
toward women’s inclusion into state institutions and policy-making. For instance,
some feminists argue that the Swedish welfare state has produced a system of
segregating men and women into different responsibilities with different power
positions (Hirdman 1989). Equality has meant privileging male norms. In the
same vein, others maintain that Sweden’s women-friendly policies are top-down
policies not framed in women’s interests but rather in the interest of family and
children (Eduards 1991; Ohlander 1992) and that there is feminism without femi-
nists (Gelb 1989). Taking the most pessimistic view of feminism in Sweden, Elman
(1995) claims that there is neither feminism nor feminists as seen in the failure to
address radical feminist issues, such as violence and sexual harassment.
This contested view of Sweden and gender equality reflects ideological rifts over
what feminism is, debates over gender difference and gender equal strategies, over
autonomous feminism versus institutionalized feminism. This is a dead-end
debate. My purpose in this paper is to broaden the conceptual terrain in gender
and social policy away from the ideological debates over autonomous and institu-
tional feminism in Sweden toward more process-oriented approaches that
highlight the mechanisms of power and agency in shaping gender-equality
outcomes. My approach assumes that gender strategies are contextually bound
and connected to social movements. This is to take seriously the notion of
capabilities (Sen 1992; and Lewis and Giullari in chapter 8 of this volume), and to
consider what kinds of policy frameworks and institutions promote individual
choices and participation in different spheres of life – the economic, social and
political. It is also important to consider how enhanced individual capabilities and
agency facilitate collective agency, that is, the ability of organizations and con-
stituencies of women to make claims, to have a voice and influence in creating
gender-sensitive social citizenship rights (Hobson and Lister 2002). Within the
comparative research on gender and welfare states there is an assumption that
societies with high rates of labour-force participation and defamilialization poli-
cies reduce agency inequalities (Korpi 2000); they increase women’s capabilities to
be full citizens. These in turn lead to greater participatory citizenship and collec-
tive political presence, thus creating women publics. The example of the former
socialist states demonstrates that high rates of labour-force participation and weak
economic dependency do not in themselves lead to the creation of women’s polit-
ical presence. Hence we need a more nuanced analysis of participation parity and
agency inequality in order to unravel a complex set of processes: How and on what
basis women are included in policy-making spheres shapes not only outcomes in
The Evolution of the Women-Friendly State 153
citizenship rights, but also political identities and the meanings of emancipation.
What kinds of claims are made and which women actors come to speak for
women, is historically and contextually grounded, and rooted in previous interac-
tions between social actors and institutions.
In this chapter I trace the role of women’s collectivities in shaping the women-
friendly Swedish welfare state. I address three sets of questions. Firstly, how have
feminist political actors (individuals and organizations) and their allies in political
parties and trade unions shaped the contours of Sweden’s ‘women-friendly’ wel-
fare state? Secondly, how have alliances and coalitions with political parties and
unions acted as constraints for women’s claims for equal participation in political,
economic and social spheres? And finally, which groups have been left out or
marginalized in the participatory model of gendered citizenship? Linked to the
question of which groups were not recognized, we need to consider which dis-
courses became recognized as embodying gender equality and which actors
became the privileged spokespersons for feminism.
Embedded in these questions are mechanisms of power and agency. Two that
I have used in my own work (Hobson 2003) are highly relevant to social
movements involving women’s movements and claims-making: brokering and
boundary-making. Brokering is a concept in social-movement theory that I have
defined as the pursuit of allies who gain advantages from their linkages with a
movement or collectivity, who can be sympathetic, lukewarm, resistant or even
antagonistic to the claims a group is making (Hobson 2003). In the case of Swedish
women’s movements, these allies have been leading men in the Social Democratic
Party and Liberal Party as well as trade union representatives. Boundary-making is
a mechanism that reveals the processes by which particular actors (and often their
organizations) become the certified spokespersons for the group, the recognized
actors in public discourse who are interviewed in the media and/or serve on
government commissions (Hobson 2003). These actors have the power to shape
which issues and claims are heard and legitimated, and which are shaded out and
marginalized. Because Swedish feminism early on became institutionalized and
linked to the Social Democracy Party and its ideologies, it is possible to view
the clear path-dependencies in discourse and policy around gender equality,
seen through the privileging of discourses that coupled paid work to women’s
emancipation.
In the following sections I provide a contextualized account of these processes
in the following periods: the first phase of Social Democratic governance in the
1930s, the second phase of welfare-state expansion in 1970s, and the third phase
of global restructuring and the emergence of feminist politics and feminist
political parties in the 1990s.
Foundations
The long duration and hegemony of the Swedish Social Democratic Party is well
known; it has been the only party with the possibility of winning a parliamentary
majority: from 1932 to 1988 the Social Democrats’ share of the vote was between
154 Barbara Hobson
42 and 50 per cent. In the policy realm, Sweden is a society where politics and
markets have been coupled. Up until the mid-1980s, there was a highly institu-
tionalized corporatist bargaining system composed of employer representatives,
unions and members of government. According to one historian of Swedish
politics, the political organization and political culture have inhibited the devel-
opment of social movements based on identities other than class, such as gender,
ethnicity and race (Åmark 1992).
Whereas most studies of Sweden and gender equality begin in the ‘golden age’
of welfare-state expansion in the 1960s and 1970s, in order to understand women’s
agency in the development of women-friendly policies, it is useful to go back to
the first phase of welfare-state formation in the 1930s and the feminist activism
that emerged in the same period. This was a period of political opportunity for
women’s groups as the Social Democratic Party was courting women as voters. It
was also a time in which there was a flowering of women’s organizations and
cross-party alliances.
Political opportunity
After a serious electoral defeat in 1928, the Social Democratic strategy shifted to a
mass-membership party, which downplayed the rhetoric of class struggle and
revolution. This election, called the Cossack Election – the label that conservatives
used in the campaign rhetoric – resulted in a decrease in Social Democratic
representation from 86 to 75 seats. Talk of revolution was replaced by security
(trygghet) (Olsson 1990: 94). Leaders of the party sought to broaden their base and
realized that it was clearly in their interest to get women to become party members
(Karlsson 1990: 323). However, there was no attempt to place women on the
voting lists nor were women’s issues notably present in the party platform.
It is important to keep in mind that the surge in Swedish feminist mobilization
came not during suffrage but in the decade after, the 1930s, a period in which
many Western feminist movements were in retreat. In Sweden large numbers of
women joined organizations for the first time. Sweden’s Social Democratic
Women’s Union increased their membership four-fold.1 Almost twice the number
of women were mobilized compared to those who fought for suffrage. Women’s
organizations that had been divided by class, education, political loyalties and
ideological differences (Lindholm 1991), in the 1930s began cooperating across
party lines. In 1939 a feminist wrote in Morgonbris (the Social Democratic women’s
journal), ‘I have experienced a shift in the women’s movement from a strictly
bourgeois phenomenon, led by middle-class women with great opposition from
working-class women, to a movement that has become everyone’s concern’
(Morgonbris 1939: 5).
This cooperation among women’s groups was facilitated by the shift in rhetoric
within the Social Democratic Party, which opened up new discursive and political
opportunities, as seen in a metaphor that became a master-frame in Social
Democracy, the People’s Home (Folkhemmet). Immortalized in the famous speech
of Prime Minister Per Albon Hansson, the People’s Home was compared to a home
in which there ‘are no privileged or deprived members, no pets and no stepchildren.
The Evolution of the Women-Friendly State 155
Depression era (Hobson 1993). The number of unemployed rose more than five-
fold between 1931 and 1932, and then reached its maximum in 1933 when over
186,000 Swedes were registered as unemployed (Tingsten 1973: 285). Married
women were seen as taking men’s jobs and every party from left to right
demanded restrictions. In the official statistics, only 10 per cent of working
women were married, but neither part-time workers nor married women working
alongside their husbands were considered as being in the labour force (Nyberg
1987). Moreover, it is important to keep in mind that half of the adult middle- and
working-class women lived alone with or without children and had to support
themselves up until the 1930s. Perhaps most importantly, there was the symbolic
import of denying married women the right to work. In the Social Democratic
worker state, in which paid work was the marker of citizenship, to exclude married
women was to create a gendered category of lesser citizens.
In response to the pressure for a marriage bar in employment, a branch of Open
Door International, an organization formed to protect women’s right to employ-
ment, was launched in Sweden. The whole spectrum of women’s organizations,
including the National Housewives Association, defended women’s right to work
on the basis that it was a citizenship right, though their preference was for
mothers to be at home (SOU 1938). An article that appeared in the Social
Democratic women’s journal Morgonbris was reproduced in two feminist journals
with a middle-class base. The link between participatory citizenship and the right
to employment was clearly made: ‘the right to walk through the door of the labour
market, not as a subservient, but as full citizens, equal to men. As long as there is
democracy, equality and justice, the bitter war between the sexes should not exist’
(Morgonbris 1937: 4; Tidevarvet 1937: 2; Hertha 1937: 3).
The Farmers’ Party, the Conservative Party and the right wing of the Liberal
Party all advocated laws restricting married women’s employment and asserting
the rights of male breadwinners. The majority of Social Democratic men (including
the leader of the party, Per Albin Hansson) hoped that women would return to the
home after they were married, but Albin Hansson was against restrictions.
However, he did not envision a law protecting married women’s rights to employ-
ment (Franguer 1998). The trade unions were against the marriage bar, not because
they supported women’s rights, but because they opposed any government inter-
ference in worker–employer contracts (Rothstein 1992). A small band of Social
Democratic men were strong supporters who joined forces with women’s groups
to turn the tide against the laws against married women’s work. They were
instrumental in getting a law passed that in fact prohibited the firing of married
women, pregnant women (married or unmarried) and women with children
(Hobson 1993).
work as the path to women’s emancipation. Two key spokespersons in the debate
were the chair and secretary of the Government Committee on Married Women’s
Work. The chair, Kerstin Hesselgren, was the first woman to be elected to the first
chamber of the Swedish parliament in the Liberal Party. Her experience as the first
woman factory-inspector made her an obvious choice to speak for the women’s sit-
uation in the labour market. The secretary of the Committee, Alva Myrdal, was
chairperson of the Professional Women’s Organization, and a Social Democrat with
an inside track into party politics through her husband, Gunnar Myrdal.2
The committee not only steered the discourse in women’s journals and the
media, but also set the terms for the debate and the framing of gender equality in
the future. Myrdal represented a 180 degree turn from the earlier maternalist femi-
nism of Ellen Key, the most prominent writer and speaker on women’s issues in the
first decades of the twentieth century. Key opposed suffrage and advocated a femi-
nism of difference, in which woman’s power derived from her maternal role and
female consciousness. She maintained that feminists who championed the vote
and careers for women were misguided (Key 1912). In contrast, Myrdal imagined a
gender equality in which women and men would both be active participants in the
labour market, and put forward proposals for collective housing and cooperative
kitchens. These utopian schemes did not stick in Sweden, but what did and would
be revived in the 1960s and 1970s was her vision of women’s participatory citizen-
ship through paid work. By the end of the 1930s the maternalist position had lost
ground. Although issues involving motherhood were part of feminist claim –
making – women’s groups put forward demands for maternal health care, paid
vacations for mothers alongside childcare, maternal leave benefits for employment –
the basis for women’s activism was not bound by a maternalist world view that
women had unique political identities based upon their motherly roles.
In the 1930s women’s groups, through their pressure on Social Democratic
government, achieved rights and protections for women in paid work (Hobson
1993; Hobson and Lindholm 1997). These included:
● Protection for all women against the loss of their jobs after marriage or
pregnancy. It was against the law to fire a woman from her job because of
marriage or pregnancy (the law also protected unmarried mothers).
● Maternal leave for all women in waged work, but those in the public sector were
given a paid leave benefit.
● Income maintenance for unmarried and divorced mothers, in which the state
gave advance payments to divorced or unmarried mothers when fathers failed
to pay their support.
At the end of the 1930s, a decade after the Women’s Committee on Married
Women’s Work published its findings and recommendations, Alva Myrdal pro-
claimed the successes that women had achieved: ‘This is a year of social political
duties. We have a government under our rule. We have a Parliament elected on a
worthy social program. We have women’s increased political awareness and also
family politics in the forefront’ (Myrdal 1937: 21).
158 Barbara Hobson
This optimism was short-lived. During the war and postwar years, women’s
issues were placed at the margins of the ‘real politics’. Rationing produced a
reassertion of traditional female skills of canning and domestic activities.
Furthermore, as the postwar years revealed, the consolidation of the Social
Democratic power base shaded out women’s voices and weakened women’s
brokering power with party leaders. It should not be forgotten that the historic
Saltsjöbaden agreement was signed at the end of the 1930s, which laid the
foundation of corporatist bargaining structures between employers, unions and
the government. These postwar years from 1945 until the 1960s represented a
bracketing of feminist politics and it would take over two decades before a revived
feminist movement could emerge. Still, the contours of Swedish feminism that
emerged in the 1930s in this early phase of welfare-state formation shaped future
feminist politics and claims for gender equality.
Social Democratic Party, but many of the participants in the group were from dif-
ferent political parties (Social Democrats, Liberals and Communists), which is why
they met in secret.5 They were young radicals, men and women, often husband-
and-wife teams, and the most notable attending these meetings were Lisbeth and
Olof Palme.
Generational differences were a key divide in the contests over the male-
breadwinner norm and the state’s role in promoting women’s employment
(Bergqvist 1999). The male-breadwinner norm still had strong supporters in the
Social Democratic Party and even within the Women’s Federation. One of the stal-
wart defenders of the full-time housewife was a long-time activist in the women’s
section of the Social Democratic Party, Nancy Eriksson. She published her book
Just a Housewife to counter the images in the media of housewives as parasites or
prostitutes (Florín 1999). Eriksson launched a crusade against the individualiza-
tion of taxes, which as she rightly understood devalued the carework of mothers
in her generation, and which would ultimately lead to the demise of the male-
breadwinner wage that supported housewives. Realizing that she had sympathizers
among Social Democratic women, the party gave Eriksson a position on the com-
mission to investigate tax reform, but she was outflanked by the younger radical
men and women on the committee.
The winds of change were moving in another direction. In the 1950s, 77 per
cent of women without children were in the labour market, and 55 per cent with
one child were employed; about 28 per cent of those with two children were in
paid work (Florín and Nilsson 1999). A survey of young women in Swedish schools
in the 1950s revealed that young women did not want to follow in the footsteps
of their mothers as housewives ( Johansson 2000). By the time the tax reform was
implemented in the 1970s, nearly half of women were already in the labour market
(Axelsson 1992). Those women who championed the dual-earner model would sit
on the dozens of commissions on marriage, family and childcare throughout the
1970s. Within the Social Democratic Party, they became the privileged speakers for
women’s interests.
As the research from the former Soviet countries has illustrated, activation of
women into the labour force in itself does not necessarily lead to gender equality
in outcomes, or even a discourse on equal participation of men and women in
paid and unpaid work. In Sweden, the dual-earner norm was coupled to a gender-
equality discourse in which men were considered to be part of the emancipatory
project. From the perspective of Eva Moberg, feminist spokesperson for the Liberal
Party, ‘the fulfillment of the goals of feminism requires a radical change of the
habits of living, attitudes and values of the average man …’ (Moberg 1962: 116).
Yet although radical men in Group 222 supported women’s participation in paid
work and even envisioned gender parity in the labour market, they could not
imagine the transformation in gender roles and the organization of paid and
unpaid work that this would entail (Hobson 2003).
The Evolution of the Women-Friendly State 161
the law ending widows’ pensions came into effect, there were 64,000 marriages
(the average had been about 40,000 a year (Hoem 1991)), a reflection of the
gap between gender-neutral ideologies and policies and gendered practices in
everyday life.
Sweden was the first society in the Western world to denaturalize child-leave
policy in 1975. No longer maternal leave, it became parental leave and the bene-
fits were generous and the conditions incredibly flexible: fathers and mothers
could divide the leave up in any way that they chose and use the time up to the
eighth birthday of their child. By 1989, parents could take up to one year and
were compensated with 90 per cent of their salaries up to a very high ceiling6 and
a further three months with a flat benefit of about 60 crowns a day (equivalent to
around 7 euros a day). The policy was gender-neutral; the practices were not. By
the early 1990s, less than a third of men took off any time for parental leave and
the proportion of days fathers took off was still no more than 10 per cent of the
couple’s permitted leave.
The idea that men should share the leave was an idea that had existed since
the 1970s in different policy circles (Bergqvist 1999). For example, when the
Commission on Family and Marriage published their report in 1972, they not only
recommended that the leave be gender neutral, but also suggested that the leave
be shared. This was meant to strengthen women’s labour-market position and
increase men’s responsibility for care (SOU 1972: 41).
Women’s organizations across the political spectrum sought legislation that
would designate a portion of the leave for men. They reasoned that otherwise the
reform would be meaningless and perpetuate lower wages and fewer career
prospects for women. Two parliamentary members of the Centre Party proposed a
12-month parental leave in which no parent would use more than eight months.
Perhaps succumbing to the realities of policy-making, Social Democratic women
insisted that one month be reserved exclusively for the father out of the eight
months allowable (Bergqvist 1998). It took two decades before the ‘daddy month’
(one month of parental leave reserved for the father) was enacted (1995), which in
1999 was extended to two months. Men took the lead in this reform; men in men’s
groups and men in key policy-making roles who had clout in political parties,
most notably the head of the Liberal Party and at that time, the Minister of Social
Affairs, Bengt Westerberg.
Though not equal participation by any stretch of the imagination,7 the ‘daddy
months’ embody an awareness that equal participation in the labour market
implies change in the division of unpaid work in the home. Moreover, this policy,
though couched in gender-neutral terms, which states that each parent shall have
at least two months of the leave, is understood as targeted and gender specific – for
fathers (Bergman and Hobson 2002).
There is a widely held assumption in the scholarly literature on the Swedish welfare
state that policy interventions around gender were a response to labour-market
The Evolution of the Women-Friendly State 163
shortages – that trade unions preferred women to immigrants to fill the expanding
service economy (Elman 1995; Jonung and Thordarson 1980). More recently
scholars have rejected this functionalist account of gender equality and high-
lighted the range of actors, men and women, who were involved in the debates on
sex roles and gender equality that began in the 1960s (Bergqvist 1999; Florín and
Nilsson 1999; Karlsson 1996). Rather than the main determinant of enabling poli-
cies for women to enter labour-market work, according to Florín and Nilsson
(1999), the shortage of labour offered a window of opportunity for brokering with
men in the Social Democratic Party who were lukewarm towards feminism and
with men in the blue-collar trade unions, the LO, who were even less sympathetic.
Unions became crucial allies of feminists in the 1970s in the debates around
publicly supported day care and individual tax reform. Their backing was essen-
tial. The large industrial unions (LO) were the main power bloc in Social
Democracy. Although they supported women’s issues that were workers’ issues –
they had opposed restrictions on married women’s right to work in the 1930s and
they backed equal pay for women workers in the 1950s – still they had little
sympathy for feminism. They tended to cast women’s movements for gender
equality as bourgeois, driven by professional women and often seen at odds with
the working man’s struggle for the male-breadwinner wage to maintain a full-time
housewife. The LO preference for women’s labour versus immigrant labour to
solve the labour shortage in Sweden is well known (Kyle 1979), and they put pres-
sure on the government to restrict immigration. Activist women in the unions and
Social Democratic Party seized upon this window of opportunity. Women in the
industrial and service-sector unions had been pushing the day-care issue since the
1940s, but in the 1960s they found allies among men in LO, and they were able to
push forward the day-care issue onto the LO agenda at a time when the shortage
of labour was on the minds of policy-makers, employers and union officials
(Hirdman 1998). Considering brokering in unions, we should keep in mind that
women had little voice in the shaping of policy within the trade unions. Although
LO was interested in women’s recruitment into unions, they did not recruit
women into the union hierarchies until the 1990s (Mahon 2002; Hirdman 1998).
Constraints
Alliances in policy-making necessitate forms of brokering that not only involve
strategic framing of issues, but also set the parameters for what kinds of claims will
have play in the policy-making forums. From this perspective, every alliance can
be an exclusion (Hobson 2003).
We can see specific examples of this in response to feminist claims around
working times and pay equity. Since the 1970s, women in the Social Democratic
Party continuously have sought a shorter working week as a women-friendly policy.
However, unions have always won out in the battles over shorter working days
versus longer vacations: the mandated vacation is now six weeks. Policies to alter
the discriminatory gendered patterns in pay, job recruitment and promotion have
engendered the most controversy within the unions. The industrial unions pro-
moted a wage-solidarity policy for low earners in their centralized bargaining with
164 Barbara Hobson
The belief that women’s and men’s roles would converge as more women entered
the labour market appears naïve with hindsight. And the data on practices
The Evolution of the Women-Friendly State 165
45
40
35
30
25
20
15
10
0
1990/91 1990/91 2001/02 2001/02
Men Women Men Women
Figure 7.1 Women’s and men’s paid and unpaid work, weekly hours 1990–91, 2000–01
Source: Statistics Sweden: www.SCB-se/statistik7be0503/tidsavanv.asp
70
60
50
40
30
20
10
0
SW95 SW00 UK99 FR94 GE94 GE00 NL94 SP90 IT95 IE96 US00 HU99 AS94
consequences for mothers after divorce. Women not attached to male earners are
poorer in terms of time and money. This has intensified since the 1990s, a period
of economic downturns and global restructuring and the fiscal conservatism that
EU membership requires.
Though Sweden continues to have the lowest poverty rates among solo mothers
compared to other industrialized welfare states (see Figure 7.2), since the 1990s,
there has been a decline in solo-mother earnings as a result of their deteriorating
labour-market position. Solo poverty rates have doubled (Fritzell 2000). The
employment rates for solo mothers are now lower than for partnered mothers,
which represents a reverse of the earlier pattern. Not only higher unemployment,
but also longer periods of unemployment, underemployment and withdrawal
from the labour market characterized the experience of solo mothers in the 1990s
(Björnberg 1997; Statistics Sweden 1998; Sainsbury 2000).
Non-Nordic immigrant groups are also bearing the brunt of a highly competitive
and discriminatory labour market. At the beginning of the 1990s, 70 per cent of
immigrants were employed, but by the end of the 1990s, the level had dropped to
55 per cent. Non-Nordic women immigrants fared the worst, with only 43 per cent
holding onto their jobs (Sainsbury 2000). And immigrant women have not had a
voice in feminist organizations or immigrant organizations (Hobson et al. 2006).
Notes
1 The Social Democratic Women’s Organization increased its membership from 7,302 in
1930 to 26,882 in 1940. In 1937 the Fredrika Bremer society (a Swedish organization
dedicated to advancing feminism) had 37 local organizations with over 6,000 members.
The National Association for Housewives’ mailing list grew from 10,000 newsletters sent
in 1930 to 23,550. The Swedish Women’s Business Organization boasted 4,000 members
at the end of the 1930s (Åkerman 1983: 122).
2 The committee produced a 500-page report that included masses of statistics that under-
scored the extent of gender inequality in the labour market (SOU 1938: 2).
3 See Leira (1992) for a discussion of Norwegian exceptionalism in this regard.
4 One might compare the women-friendly policy initiatives to those developed in the
1990s, a period with dramatic increases in women’s labour-force participation in Europe
(see Bruning and Plantenga 1999).
5 The name of the group is the address of a well-known feminist, Annika Baude, who
became a key person in the debates on women’s work.
6 At that time it was the equivalent of 3,200 euros.
7 Though over three-quarters of fathers take some leave, the proportion of leave remains
at 17 per cent.
8 The first anti-gender discrimination law was passed during a term when the Social
Democrats were out of power in the mid-1980s.
9 Swedish Equality Ombudsman brought the case against the Örebro county council who
employed both workers at the hospital (case a 153/95).
10 The case was taken to the EU Court twice and rejected by the Swedish Labour Court: the
first time arguing that the two jobs were not comparable, and the second time claiming
that it would undermine collective bargaining (Case C-236/98). The case has not been
appealed again due to cost of EU cases (Interview with Jamställdhetsombudsmannen).
11 Women have decreased their hours by about two and a half hours per week; men by
about an hour per week.
12 The Social Democratic Party in the 1994 election made the decision to alternate women
and men on all of their party lists. This meant greater representation in real terms.
Bibliography
Acker, J. (1992) ‘Två Diskurser om Reformer och Kvinnor i den Framtida Välfärdsstaten’ (Two
Discourses on Reforms and Women in the Future Welfare State). In Acker (ed.) Kvinnors och
Mäns Liv och Arbete. Stockholm: SNS Förlag.
Åkerman, B. (1983) Vi Kan, Vi Behövs! – Kvinnorna går samman i egna föreningar ( We Can! We
are needed! – Women Organize themselves in Their Own Moverments). Stockholm: Förlaget
Akademilitteratur.
Åmark, K. (1992) ‘Social Democracy and the Trade Union Movement: Solidarity and the
Politics of Self-Interest’. In K. Misgeld, K. Molin and K. Åmark (eds) Creating Social
The Evolution of the Women-Friendly State 169
Democracy, A Century of the Social Democratic Party in Sweden. University Park PA:
Pennsylvania State University Press.
Axelsson, C. (1992) Hemmafru som forsvann. Övergången till lönearbete bland gifta kvinnor i
Sverige 1968–1991 (The Housewife that Disappeared: On the Shift to Wage Labour amongst
Married Women in Sweden 1968–1991). Stockholm: Swedish Institute for Social Research.
Baude, A. (1992). ‘Inledning’ (Introduction). In A. Baude (ed.) Visionen om jämstalldhet
(Vision of Equality). Stockholm: SNS.
Bergman, H. and B. Hobson (2002) ‘Compulsory Fatherhood: The Coding of Fatherhood in
Sweden’. In B. Hobson (ed.) Making Men into Fathers: Men, Masculinities and the Social
Politics of Fatherhood. Cambridge: Cambridge University Press.
Berqvist, C. (ed.) (1999) Equal Democracies? Gender in the Nordic Countries. Oslo: Scandinavian
University Press.
Björnberg, U. (1997) ‘Single Mothers in Sweden: Supported Workers Who Mother’. In
S. Duncan and R. Edwards (eds) Single Mothers in an International Context. London
University College: London Press.
Bruning, C. and J. Plantenga (1999) ‘Parental Leave and Equal Opportunities: Experiences in
Eight European Countries’. Journal of European Social Policy, Vol. 9, No. 3: 195–209.
Bryson, V. (1999) Feminist Debates: Issues of Theory and Political Practice. London: Routledge.
Cass, B. (1994) ‘Citizenship, Work and Welfare: The Dilemma for Australian Women’. Social
Politics, Vol. 1, No. 1: 106–124.
Dahlström, E. (ed.) (1962) Kvinnors liv och arbete (The Life and Work of Women). Stockholm: SNS.
Eduards, M. (1991) ‘Toward a Third Way: Women’s Politics and Welfare Policies in Sweden’.
Social Research, Vol. 58, No. 3: 677–705.
Eisenstein, H. (1996) Inside Agitators: Australian Femocrats and the State. Philadelphia: Temple
University Press.
Eklund, R. (1996) ‘The Swedish Case – the Promised Land of Sex Equality?’ In T. Hervey and
D. O’Keefe (eds) Sex Equality Law in the European Union. New York: John Wiley.
Elman, A. (1995) ‘The State’s Equality for Women; Sweden’s Equality Ombudsman’. In
D. McBride Stetson and A.G. Mazur (eds) Comparative State Feminism. Thousand Oaks CA: Age.
Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism. Princeton NJ: Princeton
University Press.
Flood, L. and U. Gräsjo (1997) ‘Tid för barn, tid för arbete’ (Time for Children, Time for
Work). In G. Ahrne and I. Persson (eds) Familj, makt och jämställdhet (Family, Power and
Equality). Stockholm: Statens offentliga utredning (SOU), 138: 159–188.
Florín, C. (1999) ‘Skatten som befriar: Hemmafruar mot yrkeskvinnor in 1960-talets
skattningsdebatt’ (The Tax that Liberates: Housewives against working women in the
‘Separate taxation’ debate of the 1960s). In C. Florín, L. Sommerstad and U. Wikander (eds)
Kvinnor mot Kvinnor: Om systerskapests svårigheter (About the Problems of Womanhood).
Stockholm: Nordstedts.
Florín, C. and B. Nilsson. (1999) ‘Something in the Nature of a Bloodless Revolution … How
New Gender Relations became State Policy in Sweden in the 60s and 70s’. In Rolf
Torstendahl (ed.) State Policy and Gender System in the Two German States and Sweden
1945–1989. Stockholm: Opuscula Historica Upsaliensis, No. 22.
Frangeur, R. (1998) Yrkeskvinnor eller makens tjänarinna? Striden om yrkesrätten för gifta kvinnor i
mellankrigstidens Sverige ( Working Woman or the Servant of Power? The Fight about Labour
Rights for Married Women in the Mid-War Period ). Stockholm: Arkiv; PhD thesis, Department
of History, Stockholm University.
Fritzell, J. (2000) ‘Inkomstfördelningens trender under 1990-talet’ (Trends in Wage
Differences during the 1990s). In J. Palme and J. Fritzell (eds) Välfärdens finansering och
fördelning (The Financing and Distribution of Welfare). SOU: 57. Stockholm: Fritzes.
Gelb, J. (1989) Feminism and Politics: A Comparative Perspective. Berkeley CA.: University of
California Press.
Gustafsson, S. and M. Bruyn-Hundt (1991) ‘Incentives for Women to Work: A Comparison
between The Netherlands, Sweden and West Germany’. Journal of Economic Studies, 18.
170 Barbara Hobson
Hernes, H. (1987) Welfare State and Woman Power: Essays in State Feminism. Oslo: University
of Oslo Press.
Hirdman, Y. (1989) Att Lägga livet till rätta. Studier i sventsk folkhems politik (To Adapt Life the
Right Way. Studies of the Swedish ‘People’s Home’ Policy). Stockholm: Norstedts Förlag.
Hirdman, Y. (1998) Med kluven tunga: LO and Genusordningen (Divided Tongue: The Labour
Union and the Role of Gender). Stockholm: Atlas.
Hobson, B. (1990) ‘No exit, No Voice: Women’s Economic Dependency and the Welfare
State’. Acta Sociologica 33, 3: 235–50.
Hobson, B. (1993) ‘Married Women’s Right to Work in the U.S. and Sweden During the
1930s’. In S. Koven and S. Michel (eds) Mothers of a New World. Maternalist Politics and the
Origins of Welfare States. New York: Routledge, Kegan and Paul.
Hobson, B. (1994) ‘Solo Mothers, Policy Regimes, and the Logics of Gender’. In D. Sainsbury
(ed.) Gendering Welfare Regimes. London: Sage.
Hobson, B. (2003) ‘Recognition in Universal and Gender Distinctive Frames’. In B. Hobson
(ed.) Recognition Struggles and Social Movements: Contested Identities, Agency and Power.
Cambridge: Cambridge University Press.
Hobson, B. and M. Lindholm (1997) ‘Collective Identities, Power Resources, and the Making
of Welfare States’. Theory and Society (Fall): 1–34.
Hobson, B. and R. Lister (2002) ‘Citizenship’. In B. Hobson, J. Lewis and B. Siim (eds)
Contested Concepts in Gender and Social Politics. Cheltenham: Edward Elgar.
Hobson, B. and M. Takahashi (1997) ‘The Parent-Worker Model: Lone Mothers in Sweden’.
In Jane Lewis (ed.) Lone Mothers in European Regimes: Shifting Policy Logics. London: Jessica
Kingsley Publishers.
Hobson, B., M. Carson and R. Lawrence (2006) ‘Recognition Struggles in Trans-national
Arenas: Negotiating Identities and Framing Citizenship’, Critical International Reviews of
Social and Political Philosophy, 8(4).
Hoem, J. (1991) ‘To Marry Just in Case … ; The Swedish Widow’s Pension Reform and the
Peak in Marriages in 1989’. Acta Sociologica 34: 127–35.
Huber, E. and J. Stephens (1998) ‘The Internationalization and the Social Democratic Model:
Crisis and Future Prospects’. Comparative Political Studies, Vol. 31, No. 3: 353–97.
Huber, E. and J. Stephens (2000) ‘Partisan Governance, Women’s Employment and the Social
Democratic State’. American Journal of Sociology, Vol. 45: 323–43.
Johansson, U. (2000) Normalitet, klass och kön: Liv och lärande i svenska läroverk 1927–1960
(Normality, Class and Gender: Life and Learning in Swedish Upper Secondary Schools
1927–1960). Stockholm: Arkiv, PhD. dissertation, History Department, Stockholm
University.
Jonung, C. and B. Thordarson (1980) ‘Sweden’. In A. Yohalem (ed.) Women Returning to Work:
Policies and Progress in Five Countries. Montclair NJ: Landmark Studies.
Karlsson, G. (1990). Manssamhället till behag? (Conforming to the Patriarchy?) Tidens.
Karlsson, G. (1996). Från broderskap till systerskap, Det socialdemokratiska kvinnoförbundets
kamp för inflytande och makt i SAP (From brotherhood to sistership, The social democratic
womens unions’ struggle for influence and power in SAP). Lund: Arkiv.
Key, E. (1912) The Women’s Movement. New York: Putnam.
Korpi, W. (1980) The Democratic Class Struggle. London: Routledge.
Korpi, W. (2000) ‘The Faces of Inequality: Gender, Class, and Patterns of Inequalities in
Different Types of Welfare States’. Social Politics, Vol. 7, No. 2: 127–91.
Korpi, W. and J. Palme (1998) ‘The Paradox of Redistribution and Strategies of Equality and
Poverty in Western Countries’. American Sociological Review, Vol. 63: 661–87.
Kyle, G. (1979) Gästerbetskerka in mannsamhället (Female Guest Workers in the Patriarchy).
Stockholm: Liber.
Leira, A. (1992) Welfare States and Working Mothers. Cambridge: Cambridge University Press.
Lewis, J. (1992) ‘Gender and the development of welfare regimes’. Journal of European Social
Policy, Vol. 2, No. 3: 159–73.
The Evolution of the Women-Friendly State 171
Vestbro, D.U. (1992) ‘Från liberal rörelse till socialistisk kamp mot patriarkatet’ (From Liberal
Movement to Socialist Struggle against the Patriarchy). In Annika Baude (ed.) Visionen om
jämstalldhet. Stockholm: SNS.
Västberg, D. (1931) ‘Manlig konservatism och kvinnlig sterilitet’ (Male Conservatism and
Female Sterility). Morgonbris, March: 8–9.
Introduction
173
174 Jane Lewis and Susanna Giullari
There have been a variety of reactions in the academic literature to the shift
towards an adult-worker-model family. Some have argued that women’s ‘moral
rationalities’ are at variance with the new model and that many want to, and do,
put the work of informal care first (Duncan and Edwards 1999; Barlow and
Duncan 2000). This position is in some respects bolstered by Hakim’s (2000) argu-
ment that only a minority of women want to be and choose to be ‘career women’.
Others have argued that because the shift towards an adult-worker-model family is
part of a much wider trend towards emphasizing the importance of responsibilities
rather than rights, and of so-called ‘active’ rather than passive ‘welfare’, it is
impossible to oppose (Orloff 2002). The issue then becomes the terms and condi-
tions on which the shift to the new model is undertaken. Many feminists have,
after all, long argued in favour of women’s financial independence. But the nature
of the choices women and men face and the pursuit of gender equality depend on
the extent to which policies address the issue of care work as well as on the posi-
tion of women in the labour market. Choice is socially ‘embedded’ and ‘genuine
choice’ or ‘real freedom to choose’ in respect of the balance of paid and unpaid
work at the level of the household (and hence gender equality) will involve not
only a rebalancing of paid work between men and women, but a complicated
rebalancing of unpaid work between the market, the state, and men and women.
A mother who wishes to take up a good job offer may well hesitate if she can find
only poor-quality (formal or informal) care for her child.1 Choice has become a
central value in the restructuring of welfare states ( Jenson and Sineau 2001), but
this example signals the central difficulty in ensuring real or genuine choice for
women, which in turn impacts on gender equality.
Something that is missing from the debate is a discussion of the principles on
which a move towards the adult-worker model should be based, especially in regard
to the issue of care work. Yet policy-makers usually operate on (often implicit)
assumptions as to what is ‘good’ and desirable, especially in the field of family policy
(Kaufman 2002). The first part of this chapter explores the very different ideas fram-
ing the shift to an adult-worker-model family at the EU level. These ideas tend to be
dominated by economic considerations, which means that the importance
attached to the choices of the individual are considered in relation to economic
goals. In addition, when gender equality is acknowledged it is often defined in a
particular, partial and instrumental way: in respect of the importance of labour-
market participation, but not care work. This also serves to limit genuine choice.
We review the nature of the policy discourse on employment and care at EU
level and go on to suggest that there are real limits to the pursuit of a full adult-
worker model based on the commodification of care. If this argument is accepted,
and given that care work is necessary for human flourishing (particularly, but not
exclusively, in respect of young, old and disabled people), it means that it must be
made possible to choose to care. This in turn raises issues about the recognition
and valuing of care, and about the way it is shared by men and women, and by the
household and the collectivity. In our search for firmer principles to inform
policies, we examine the possibilities offered by the capabilities approach (CA)
developed by Amartya Sen and Martha Nussbaum (Sen 1985, 1992, 1999;
The Adult-worker-model Family and Gender Equality 175
Nussbaum and Sen 1993; Nussbaum 1999, 2000, 2003), who have explicitly
addressed both the issues of gender equality and care. We suggest that the empha-
sis placed by the CA on the individual’s real freedom to choose something he or she
considers to be worthwhile to do or to be provides a substantial basis for the recog-
nition and valuing of care work, as well as providing a means of countering the
more rank instrumentalist concerns expressed at the level of EU policy-making.
However, gender equality also requires care to be more equally shared between men
and women, which is more difficult to encapsulate with a capabilities approach. In
societies where the position of men and women is unequal, it is possible that the
exercise of men’s freedom to choose between work and care serves to limit women’s
choices. This raises in an acute form the old issues of how to reconcile choice and
equality and what kind of state intervention can be justified.
The trend in Western welfare states has been towards a reworking of the relation-
ship between social provision and employment, with the emergence of what
Goodin (2001: 39) has called ‘a new constellation of work-and-welfare variables’.
Policy-makers have sought to shift the emphasis from rights to responsibilities,
and from so-called ‘passive’ to ‘active’ welfare, such that claimants on the welfare
system are ‘encouraged’ into work and the work is made ‘to pay’ (Lödemel and
Trickey 2001). The relationship between employment and social provision has
always been central to modern welfare states (Supiot 1999); what is striking about
this reworking of the relationship is the way in which it has been couched in
gender-neutral language. It is now assumed that women as well as men will be
‘citizen workers’.
work or short-term, more precarious employment can be seen as a way for women
to enter the labour market, or as a type of employment that women find very hard
to escape.
Furthermore, in the recent economic strategy documents of the Commission,
there has been relatively little reference to the problems raised by the need to
reconcile family and work responsibilities. Indeed, there is increasing evidence
from recent EU documents and from the work of academic policy advisers that the
extension of the ‘citizen worker’ strategy to women is part of the dominant
competitiveness and growth agenda. It is significant that recent EU-level employ-
ment policy documents have put more emphasis on the provision of formal child-
care services, which have been shown to be the most effective way of promoting
female labour-market participation (Bradshaw et al. 1996; Gornick et al. 1997).
In 2002 a benchmark for childcare was set, whereby by 2010 at least 33 per cent of
children under three years of age and at least 90 per cent of children between three
and the mandatory school age should have access to childcare (EC 2003).
The academic arguments put to EU level bodies since 2000 in favour of the
adult-worker-model family have tended to assume the desirability of a ‘de-
familialization strategy’ for care work (for children and adult dependants),
whereby it is commodified and put into the public, formal arena (Esping Andersen
et al. 2001, 2002). The overwhelming emphasis has been on gender equality
defined in terms of increased female employment. In practice, in those countries
where women’s employment has increased the most dramatically over the past
decade, there has been a tendency for women to enter more ‘flexible’ forms of
employment. Thus gender equality defined in terms of individualization and
labour-market participation has not managed to deliver ‘better jobs’ in the sense
of equally secure, equally well-paying jobs, to both men and women. Rather, in
the case of women, ‘better jobs’ have tended to be defined in terms of the extent
to which provision is made for reconciling work and family responsibilities, and
all too often the focus in respect of these policies has been on women alone, rather
than on men and women.
The economic productiveness and competition goal has been accompanied by a
‘work first’ strategy for women and an emphasis above all on the de-familialization
(and commodification) of care work via the provision of childcare services.
Differing amounts of emphasis have been placed on the provision of cash transfers
(e.g. for parental leave) as a means to achieve more female labour-market partici-
pation, but very little attention has been paid to the pursuit of gender equality in
respect of unpaid work at the household level. It is also significant that the princi-
ple of gender equality is deployed in EU-level documents in such a way as to
eclipse the equally difficult issue of ‘choice’. The economic imperative to increase
female employment has resulted in some extraordinarily instrumentalist arguments.
Thus high female labour-market-participation rates are required for the health
of the economies of EU member states and the EU social model, and are
justified in terms of achieving ‘gender equality’ (Esping Andersen et al. 2002). The
de-familialization of care work necessarily follows, and there is no further discussion
of gender inequalities in unpaid work. Nevertheless, very different ideas about the
178 Jane Lewis and Susanna Giullari
The key issue underlying the goal of making further progress towards a full adult-
worker-family model within existing policy approaches is thus, above all, how far
it is possible to defamilialize and hence ‘commodify’ care. This in turn begs
questions about the nature of care work. It also raises issues about gender equality,
for policies to reconcile work and family have historically been focused in the
main on women only.
The commodification of care can take place in the formal and informal arena,
with very different implications for defamilialization. Ungerson (1997) argues that
the different kinds of cash payments for informal caregivers and care receivers
introduced across welfare states, indicate a general trend towards the commodifi-
cation of informal care, but obviously do not amount to a clear-cut defamilializa-
tion of care strategy. Still, payments for care can differ in relation to the distinction
between formal and informal care – some payments may not be used to pay for
relatives – and indeed this distinction is sometimes also differently drawn between
child and adult care, and between countries (McLaughlin and Glendinning 1993;
Land 2002; Millar and Warman 1996; Ungerson 1997). If care is commodified and
put in the formal arena either through the provision of cash subsidies to purchase
care on the market, or by directly providing public child and elderly care services,
the scope for defamilialization is clearly widened. Yet we shall argue that such a
strategy nevertheless results in incomplete defamilialization, because it is not pos-
sible fully to commodify care (Folbre and Nelson 2000). However, it should be
noted that we are not following those who argue that care should not be com-
modified, because the marketization of care jeopardizes the affective dimension of
care (e.g. Noddings 1984).
Care is more than a task; it involves emotional labour and relationship (Finch
and Groves 1983). Care is also active and passive, involving physical and non-
physical presence (Land 2002). Balbo’s (1987) concept of doppia presenza incorpo-
rates these elements of passive care and non-physical presence, which show up in
the uneasy accounts of mothers who report that they are thinking simultaneously
about their jobs and about whether their children are being looked after properly.
The interaction between the passive and active dimension of care suggests that
the shift from ‘passive’ to ‘active’ that has underpinned the reworking of the
work/welfare relationship is particularly disadvantageous for women who have
caring responsibilities, and makes the valuing and rewarding of care even
more difficult.
Folbre and Nelson (2000) show that there has been a rapid expansion of
commodified care services in the US, and Daly (2001) has also argued that care
services, public and private (market), have been an area of expansion in what has
The Adult-worker-model Family and Gender Equality 179
The adoption of the adult-worker-family model at the policy level has been
based on an instrumental approach to gender equality and care. But care cannot
be fully defamilialized or commodified. Thus we need to make a case that first,
enables the recognition of care work as something that is worthwhile and necessary,
which in turn necessarily involves valuing it. Human beings need care – it is a
human activity – which means that we must also consider the way in which care
work is shared in societies, between men and women at the household level as
well as between the individual and the collectivity. The current policy debates
tend to treat care work as something that has to be reconciled with employment
for women, and which can be addressed via commodification, whether through
cash payments of some kind (through the tax/benefit system or directly to carers)
so that carers may purchase care on the market, or through the provision of
public-sector care services. Such policies are poorly developed compared to the
other major social services (health and education) in most countries. But if it is in
any case not possible to commodify all care work, then the issue of how it is shared,
not just between individuals and the collectivity, but also between men and
women at the household level, must also be addressed.
Amartya Sen and Martha Nussbaum’s work on the ‘capabilities approach’ (CA)
(Sen 1985, 1990, 1992, 1999; Nusbaum and Sen 1993; Nussbaum, 1999, 2000,
2003) is promising ground for establishing a basis on which to address the issue of
care, not least because gender equality is of central concern in the work of both
authors. There are important differences in their approaches, the most fundamental
being that Nussbaum argues for a universal list of core capabilities whilst Sen does
not (Robeyns 2003). In addition Nussbaum (2000, 2003) devotes a lot of attention
to care, which hardly appears in Sen’s writings. However, first, it is important to
recognize the way in which the work of both authors has served to shift attention
from inequalities in resources, outcomes and preferences, to inequalities in
capabilities, that is, to the ‘real freedom’ that a person has to be and to do what
she has reason to value, and the significance of this shift for addressing the issue
of care.
This notion of individual freedom is pivotal to the CA, which recognizes that
genuine freedom must be underpinned, while also making important distinctions
between: the resources available to a person (means); what she is and does
(functionings); the personal, social and environmental factors that affect her ability
to transform means into functionings (conversion factors); and the combination of
being and doings that she has the real freedom to achieve (her capability set).
A further crucial distinction is made in the CA between well-being freedom and
agency freedom (Sen 1985). Sen (1985) argues that the outcome measure of well-
being has major limitations because it fails to address the pivotal and distinctive
part that agency plays in a person’s life. He argues that it is important to take
account of agency above and beyond the role that it plays in securing well-being,
because of the importance that is attached to the individual’s ‘real freedom’ to
The Adult-worker-model Family and Gender Equality 181
choose in accordance with her conception of ‘the good’, regardless of whether this
would actually be to her advantage (Sen 1985: 206), for instance, in respect of a
decision to forgo paid employment in order to undertake care work. Advocates of
the capabilities approach would argue that her capabilities set should enable her to
choose to care as well as to choose not to care.
There is another fundamental reason for making a distinction between well-
being freedom and agency freedom, which Sen (1999) stresses in Development as
Freedom, where he argues that the real freedom of individuals is the motor of
development and social change, rather than economic growth per se. The recogni-
tion that Sen accords women’s agency in achieving social change parallels the
increasing interest in the question of individual and collective agency in work on
gendered citizenship and on gender and the development of welfare states more
generally (Lister 1997; Misra and Akins 1998; Siim 2000; Williams et al. 1999).
has long been argued by feminist analysts that the value attached to particular
activities is affected by who does them. The activities of the less powerful are
devalued (e.g. Bradley 1989).
The CA also accounts for human diversity in a more fundamental way, for in
order to guarantee equality of real freedom to choose and achieve functionings,
different individuals will require different kinds and amounts of resources (which
is why equality of resources is in and of itself an insufficient condition for the
achievement of chosen functionings). The notion of personal, social and environ-
mental conversion factors accounts for the fact that people have different capacities
to gain access to the same resource, and different potentials for converting
resources into chosen functionings. To choose employment and care an individual
requires either more money to pay for childcare or affordable childcare services, as
well as a right to reduced working time and paid time in order to care. But real
agency freedom also requires there to be a genuine choice to reject alternative
functions. A carer should be as free as a non-carer to choose to work. Clearly this
requires anti-discrimination legislation, the right to return to the same job, and
the offer of retraining. It also requires rights to income based on caring, at a suffi-
ciently high level to secure the effective freedom to choose care. Gender equality
in capabilities for paid work and for care highlights the need for a combination of
a variety of social entitlements, and brings to the fore the issue of time as well as
money. It thus makes manifest the biases and limitations of the current policy
trend that focuses on wages as the best form of welfare, active labour-market
policies, and care services aimed at the defamilialization of care.
Although Sen and Nussbaum both recognize that human beings differ in regard
to the values they attach to different beings and doings, they differ considerably in
respect of the process of defining and selecting functionings to be included in the
capabilities set. In regard to valuing care, Nussbaum’s approach appears at first
sight to be more useful than that of Sen, who leaves the construction of the capa-
bilities set open, and with it, the possibility of excluding care. But, despite
Nussbaum’s concern with gender justice, care is not explicitly included in her list
of the core capabilities necessary for human flourishing (unless it is taken to be
part of ‘capability for affiliation’, which includes ‘showing concern for other
human beings’), which is symptomatic of the problem of drawing up a core list of
capabilities that are universally applicable.
In contrast, in Sen’s approach the importance of human diversity is embraced
further, as it is argued that real agency freedom requires the inclusion and partici-
pation of diverse voices in the definition and selection of capabilities sets. As
Robeyns (2003) notes, Sen stresses that capabilities sets are context-dependent,
and that their legitimacy is dependent on democratic processes of inclusion and
on who is involved in their definition, valuation and selection (see also Bonvin
and Farvaque 2003). Thus, paradoxically perhaps, the fact that Sen leaves the def-
inition of capabilities open, insisting that in order to achieve equal agency free-
dom for men and women it is necessary to include diverse voices in the process of
defining and selecting the functionings that are to be part of the capabilities set,
is, if it can be achieved, a more gender-sensitive and gender-equal path to the
The Adult-worker-model Family and Gender Equality 183
recognition and valuing of care. After all, policies based on universality, equality
and justice have all too often been modelled on the male norm, and have resulted
in the exclusion and subordination of women’s claims (Lister 1997; Okin 1989;
Young 1989).
In addition, by requiring the capability for voice and by arguing that policies
should be designed, implemented and evaluated on the basis of the real freedom
that a person has to be and to do what she has reason to value, the CA exposes the
democratic deficit of the instrumental approach that has informed the shift to the
adult-worker model.
words, the capabilities sets of individuals are interdependent (Basu and Lopez-
Calva 1999), and such interdependence is particularly acute at the household level
(Iversen 2003). A child’s capabilities are dependent in large measure on her
mother’s choice to care or not to care for her, while her mother’s and her father’s
capability to choose to engage in paid work or political activism are dependent on
each other’s choices to care or not to care for their child. Nussbaum’s conceptual-
ization of care is therefore limited because it does not enable us to recognize fully
that women’s agency freedom is unequally restricted by their caregiving.
In Sen’s (1990, 1999) cooperative–conflict bargaining model the individual
enters into cooperation solely to maximize his or her welfare, which makes it
difficult to acknowledge the kind of relational interdependency that care entails,
which in turn cannot be broken down into individual benefits and costs (Held
1993). As Peter (2003) argues, Sen’s account of women’s low perception of self-
interest ignores that women’s agency is situated in relationships of care, and there-
fore that concern for others needs to be taken seriously as an expression of
autonomy. Agency is unequally restricted along gender lines largely because men
can be concerned about their relationships without being constrained by under-
taking care work.
If we maintain, as both Sen and Nussbaum do, that agency freedom requires
that the real freedom to achieve the same capabilities set must be available to all,
then, given the interdependence that characterizes care relations, it would seem to
us that the key issue is not how to make women more self-interested – which Sen
(1999: 144) argues will come about via their increasing attachment to the labour
market – but rather how to promote conditions that foster responsibility for shar-
ing care between men and women and that enhance women’s agency freedom by
making men more accountable for their responsibility to care for others. But this
is particularly difficult within the CA.
The CA makes agency freedom of the individual pivotal, which means that care
can only be conceptualized as an opportunity, a real individual choice. But the
argument we have put forward about care suggests that it is inherently relational.
This means that the individual’s autonomy and interdependence must be recog-
nized. Ultimately, it is only when all persons are conceived from the start as
autonomous and interdependent – that is as persons who need, give and receive
care (Fraser 1997; Lister 1997; Tronto 1993) – that gender equality in respect of
agency freedom can be embraced. We all need others to care for us in order to
develop agency freedom, and gender justice therefore requires care to be valued
and conceptualized as both a ‘legitimate’ opportunity/choice and as a necessary
central human activity to be shared between men and women.
Choices are made in the context of gendered inequalities in power relations, in
all their economic, political and discursive manifestations, which skew the inter-
dependency of men and women’s individual capabilities sets at the household
level. Thus Phillips (2001) has criticized the CA’s departure from the more tradi-
tional concern with equality of outcomes and resources. It is impossible to ignore
group inequalities of this kind. Inequalities in ‘group functionings’ (for example,
the fact that women persistently do more care work than men) signify gender
The Adult-worker-model Family and Gender Equality 185
If there is to be ‘genuine choice’ between work and care for women and men, then
a wide range of policies – in respect of time to care, cash for care, care services and
the regulation of working hours – is necessary. In practice, existing measures have
usually been invoked to enable individuals (mainly women) to choose to add work
to care (the supported-adult-worker model of the Scandinavian countries and dis-
cernible in some, but not all, EU-level policy documents) and they rely primarily
on sharing care between (women in) households on the one hand, and some form
of paid provision on the other. However, we have argued that care must also be
shared between men and women at the household level, and that it is necessary to
recognize the importance of interdependence and unequal power, as well as the
individual’s agency freedom in order to achieve this. Is there a place for compul-
sion in this respect? When she considers the issue of sexual harassment in the
186 Jane Lewis and Susanna Giullari
● The household level (in terms of the provision of cash and time for men and
women)
● The collective level (in terms of services and cash, and workplace-based changes
especially in respect of measuring the work accomplished, rather than the
‘face-hours’ devoted to it).
The task of devising social policies that promote real or genuine choice for both
men and women in respect of paid and unpaid work poses huge difficulties. Both
employment and care are necessary for human flourishing. Care is not just ‘a good
thing’, it is crucial to the welfare of the person-cared-for; employment is not just
‘a good thing’, it is crucial to the welfare of the adult (and likely to become more
so as governments assume the capacity for self-provision will increase). From the
point of view of human welfare, it is impossible to choose not to care or not to
work. In this sense, to choose between them has the hallmarks of tragedy as much
or more than opportunity, which makes it all the more important to address
seriously the policies that play a part in structuring people’s choices.
Acknowledgement
The authors gratefully acknowledge financial support from the EUROCAP project, funded by
the EC’s FP5 Programme and coordinated by Professor Robert Salais, and Susanna Giullari
The Adult-worker-model Family and Gender Equality 187
Note
1 The reasons why the father of the child may not hesitate, at least to the same extent, are
much debated between those who assume biological essentialism, which underpins
much of the classical literature, e.g. Parsons and Bales (1995), Becker (1981), also Gilligan
(1982), and those who insist on the power of socialization which underpins most of the
feminist literature, e.g. Oakley (1981). New variants of the arguments have come to the
fore in the face of observations that some lone mothers put the obligation to care before
paid employment for reasons that include structural variables (Duncan and Edwards
1999), while Hakim (2000) argues that lifestyle choice alone is the main explanation.
Bibliography
Balbo, L. (1987) ‘Crazy Quilts: Rethinking the Welfare State Debate from a Woman’s Point of
View’. In A. Showstack Sasson (ed.) Women and the State. London: Hutchinson.
Barlow, A. and S. Duncan (2000) ‘New Labour’s Communitarianism: Supporting Families and
the “Rationality Mistake”, Part II’, Journal of Social Welfare and Family Law 22(2): 129–43.
Basu, K. and L. F. Lopez-Calva (1999) ‘Functionings and Capabilities’. In K. Arrow, A. Sen and
K. Suzumura (eds) Handbook of Social Choice and Welfare. Amsterdam: Elsevier
Science/North Holland.
Beck, U. and E. Beck-Gernsheim (2002) Individualization. London: Sage.
Becker, G. (1981) A Treatise on the Family. Cambridge MA: Harvard University Press.
Bojer, H. (2002) The Social Contract: Unpaid Child Care And Women’s Economic Capability. Paper
presented at the VHI conference ‘Promoting Women’s Capabilities: Examining
Nussabum’s Capabilities Approach’, Cambridge, 9–10 September.
Bonvin, J. and N. Farvaque (2003) Towards a Capability-Friendly Social Policy: The Role of Local
Implementation Agencies. Paper prepared for the VHI conference ‘Transforming Unjust
Structures: Capabilities and Justice’, Cambridge, 26–27 June.
Bradley, H. (1989) Men’s Work Women’s Work. Cambridge: Polity Press.
Bradshaw, J., S. Kennedy, M. Kilkey, S. Hutton, A. Corden, T. Eardly, H. Holmes and J. Neale
(1996) The Employment of Lone Parents. London: Family Policy Studies Centre.
Commission of the European Communities (CEC) (1993) Growth, Competitiveness and
Employment – The Challenges and Ways Forward into the 21st Century. Luxembourg: CEC.
Crompton, R. (ed.) (1999) Restructuring Gender Relationships and Employment: The Decline of
the Male Breadwinner. Oxford: Oxford University Press.
Daly, M. (2001) ‘Alternative Ways of Compensating for Care in Western Europe’. In M. Daly
(ed.), Care Work Security. Geneva: ILO.
Duncan, S. and R. Edwards (1999) Lone Mothers, Paid Work and Gendered Moral Rationalities.
Basingstoke: Macmillan Press.
Esping-Andersen, G., D. Gallie, A. Hemerijck and J. Myles (2001) A New Welfare Architecture for
Europe? Report submitted to the Belgian Presidency of the European Union, Final Version.
Esping-Andersen, G., D. Gallie, A. Hemerijck and J. Myles (2002) Why We Need a New Welfare
State. Oxford: Oxford University Press.
European Commission (EC) (1996) ‘Council Directive 96/34/EC of 3 June 1996 on the
framework agreement on parental leave concluded by UNICE, CEEP and the ETUC’.
Official Journal of the European Communities, L 145.
188 Jane Lewis and Susanna Giullari
European Commission (EC) (2000) Social Policy Agenda. COM (2000) 379 final, Brussels.
European Commission (EC) (2002a) Increasing Labour Force Participation and Promoting Active
Ageing. COM (2002) 9 final, Brussels.
European Commission (EC) (2002b) Taking Stock of Five Years of the European Employment
Strategy. COM (2002) 416 final, Brussels.
European Commission (EC) (2003) Scoreboard on Implementing the Social Policy Agenda. COM
(2003) 57 final, Brussels.
European Commission and Council of the European Union (2002) Joint Report on Employment,
at http://www.europa.eu.int/comm/employment_social/employment_strategy/report_2002/
jer2002_final_en.pdf, accessed on 9 June 2005.
Ferrera, M., A. Hemerijck and M. Rhodes (2000) The Future of Social Europe: Recasting Work and
Welfare in the New Economy. Report for the Portuguese presidency of the European Union.
Finch, J. and D. Groves (eds) (1983) A Labour of Love: Women, Work and Caring. London:
Routledge & Kegan Paul.
Finch, J. and J. Mason (1993) Negotiating Family Responsibilities. London: Routledge.
Folbre, N. (1995) ‘ “Holding Hands at Midnight”: The Paradox of Caring Labour’. Feminist
Economics, 1(1): 73–92.
Folbre, N. and J. Nelson (2000) ‘For Love or Money – Or Both?’ Journal of Economic Perspectives
14(4): 123–40.
Fraser, N. (1997) Justice Interruptus: Critical Reflections on the ‘Post-Socialist’ Condition. London:
Routledge.
Gershuny, J. (2000) Changing Times: Work and Leisure in Post-industrial Society. Oxford: Oxford
University Press.
Gilligan, C. (1982) In a Different Voice. Cambridge MA: Harvard University Press.
Glendinning, C. and J. Millar (1987) Women and Poverty. Brighton: Wheatsheaf.
Goodin, R. E. (2001) ‘Work and Welfare: Towards a Post-productivist Welfare Regime’. British
Journal of Political Science, 31(1): 13–39.
Gornick, J. C., M. K. Meyers and K. E. Ross (1997) ‘Supporting the Employment of Mothers:
Policy Variation across Fourteen Welfare States’. Journal of European Social Policy, 7(1):
45–70.
Hakim, C. (2000) Work-Lifestyle Choices in the Twenty-first Century: Preference Theory. Oxford:
Oxford University Press.
Hantrais, L. (ed.) (2000) Gendered Policies in Europe: Reconciling Employment and Family Life.
Basingstoke: Palgrave.
Held, V. (1993) Feminist Morality: Transforming Culture, Society and Politics. Chicago:
University of Chicago Press.
Iversen, V. (2003) ‘Intra-Household Inequality: A Challenge for the Capability Approach?’
Feminist Economics, 9(2–3): 93–115.
Jenson J. and M. Sineau (eds) (2001) Women’s Work, Childcare, and Welfare State Redesign.
Toronto: University of Toronto Press.
Kaufman, F-X. (2002) ‘Politics and Policies Towards the Family in Europe: A Framework and
an Inquiry into their Differences and Convergences’. In F-X Kaufman, A. Kuijsten,
H-J Schulze, K. P. Strohmeier (eds) Family Life and Family Policies in Europe, Vol. 2. Oxford:
Oxford University Press.
Kittay, E. (1999) Love’s Labour: Essays on Women, Equality and Dependency. New York:
Routledge.
Kleinman, M. (2002) A European Welfare State? Basingstoke: Palgrave Macmillan.
Land, H. (2002) ‘Spheres of Care in the UK: Separate and Equal’. Critical Social Policy, 22(1):
13–32.
Lewis, J. (1997) ‘Gender and Welfare Regimes: Further Thoughts’. Social Politics, 4: 160–77.
Lewis, J. (2001a) ‘The Decline of the Male Breadwinner Model: The Implications for Work
and Care’. Social Politics, 8(2): 152–70.
Lewis, J. (2001b) The End of Marriage? Individualism and Intimate Relations. Cheltenham:
Edward Elgar.
The Adult-worker-model Family and Gender Equality 189
Siim, B. (2000) Gender and Citizenship: Politics and Agency in France, Britain and Denmark.
Cambridge: Cambridge University Press.
Silva, E. B. and C. Smart (1999) The New Family? London: Sage.
Smart C. and B. Neale (1999) Family Fragments? Cambridge: Polity Press.
Supiot, A. (ed.) (1999) Au-dela de L’Emploi (Beyond Employment). Paris: Flammarion.
Tronto, J. (1993) Moral Boundaries: A Political Argument for and Ethic of Care. London:
Routledge.
Ungerson, C. (1997) ‘Social Politics and the Commodification of Care’. Social Politics, 4: 362–81.
Ungerson, C. (2000) ‘Thinking About the Production and Consumption of Long-Term Care
in Britain: Does Gender Still Matter?’ Journal of Social Policy, 29(4): 623–43.
Vandenbroucke, F. (2001) ‘European Social Democracy and the Third Way: Convergence,
Divisions, and Shared Questions’. In S. White (ed.) New Labour: The Progressive Future?
Basingstoke: Palgrave.
Williams, F. (2001) ‘In and Beyond New Labour: Towards a New Political Ethics of Care’.
Critical Social Policy, 21(4): 467–93.
Williams, F., J. Popay and A. Oakley (1999) Welfare Research: A Critical Review. London: UCL
Press.
Young, I. M. (1989) ‘Polity and Group Difference: A Critique of the Ideal of Universal
Citizenship’. Ethics, 99: 250–74.
Young, I. M. (1990) Justice and the Politics of Difference. Princeton NJ: Princeton University Press.
Part II
Labour Market Informality and
the Search for Social Security
9
Labour Market Informalization,
Gender and Social Protection:
Reflections on Poor Urban
Households in Bolivia and Ecuador
Lourdes Benería and Maria S. Floro
Introduction
Conceptual framework
193
194 Lourdes Benería and Maria S. Floro
Since its inception, the notion of ‘informal sector’ was subject to debates and
criticisms, particularly in terms of the heterogeneity and artificiality of the
formal/informal division. The initial 1970s formulations by Keith Hart and by
the ILO mission to Kenya in 1972 implied that modernization and economic
growth would facilitate the absorption of the ‘informal sector’ by the modern sec-
tor. By contrast, during the 1980s the literature emphasized the links rather than
the divisions between the two sectors, pointing to the dependency of the modern
on the informal sector as a way of lowering costs, thus implying that there was no
incentive for the former to absorb informal activities (Benería and Roldán 1987;
Portes and Castells 1989). The 1990s saw a deepening of these processes and an
expansion of informality, thus increasing the need to understand further the
factors that have reversed the prediction of the initial formulation. Hence the
continuing debates on the topic and their increasing relevance to understand
the persistence of poverty and the increasing vulnerability in the midst of an
expanding global economy (De Soto 2000; ILO 2002).
This raises questions regarding the differences between earlier and current stages
of informalization. Have conditions changed and, if so in what way? Several obser-
vations can be made in this respect.2 First, since the 1980s, the macroeconomic
context has changed significantly due to the implementation of neoliberal policies
in many countries, including Ecuador and Bolivia. Privatization, market deregula-
tion and economic restructuring have contributed to set up the conditions for an
expansion of informal activities and precarious working conditions. We provide
specific examples related to these countries in subsequent sections.
Second, the expansion and deepening of markets has extended the links, direct
or indirect, between formal and informal activities while the distinction between
the two has become increasingly blurred. Large firms have developed increasing
links with informalized production through outsourcing and subcontracting. This
trend has been examined by several studies in the case of Thailand as well as for
the Latin American region (Boonmathya et al. 1999; Balakrishnan and Huang
2000; Pérez-Sainz 1994, 2005; Buechler 2002). Nevertheless, as argued below,
many other types of informal work remain disconnected from more formal
production particularly those related to survival activities.
Third, informal activities are no longer viewed as the anomaly that eventually will
disappear with modern economic growth. On the contrary, the trend has been in
the opposite direction; the decline in formal employment, particularly in the public
sector, and the dynamism of the informal economy in many cases has enhanced the
relative attractiveness and the predominance of the latter. Likewise, social exclusion
has become more prevalent among unskilled labour and rural-to-urban migrants.
Fourth, although the traditional association of informal work with low skills and
low productivity still holds, the last two decades have introduced many changes in
this respect. With the increase in outsourcing and subcontracting, several
segments of informal activities have their centre of gravity in the core firms that
generate them. This implies that productivity levels in these segments tend to be
high and competitive, particularly because technology transfers and skills training
take place through these linkages.
Labour Market Informalization, Gender and Social Protection 195
Figure 9.1a Areas of economic activity and labour use during economic growth
Formal
Economy
Household
Economy
Combined formal market
and household activities
Figure 9.1b Areas of economic activity and labour use during severe economic crisis
adjust labour input without much resistance from workers, thus providing
maximum flexibility for capital. At the same time, many governments, including
those of Bolivia and Ecuador, have adopted labour laws that tolerate and even
promote labour flexibility without much concern about safety nets and unem-
ployment compensation schemes during periods of high unemployment and
underemployment.
A different question is the extent to which women’s increasing participation in
paid work is transforming gender relations. At the general level, the transforma-
tion of gender roles during the past decades has been profound, affecting many
aspects of women’s and men’s lives. Broadly speaking, and for the purpose of this
paper, these transformations fall into two categories.
Labour Market Informalization, Gender and Social Protection 197
One has to do with the division of labour and the location of women and men
in the world of work, such as the increasing participation of women in the paid
labour force. To be sure, gender remains a major axis of differentiation in terms of
work and labour market adjustments. Women workers tend to earn lower wages
than their male counterparts as a result of discrimination, gender norms and other
factors that permeate economic and social institutions. These factors and norms
vary across countries and cultures, thus influencing the ways in which gender
inequality takes concrete forms (Kabeer 2003; Benería 2003; Salzinger 2003). In
Latin America, and despite the fluctuating economic performance of several
economies since the 1980s, there have been remarkable changes for women in the
labour market (Duryea et al. 2004). At the same time, although women tend to be
placed at the bottom of labour hierarchies, we have also observed increasing
differentiation among women in terms of work and income (McCrate 1995;
Lavinas 1996; Piras 2004). Thus, while educated women have made inroads into
the professional world that used to be predominantly male, the large majority of
women are confined within the more precarious jobs and informalized
working conditions; they continue to be constrained by their low mobility, few
alternatives and by social norms that reduce their bargaining power over their
earnings and working conditions.
The second category of gender transformations operates at a deeper level of
gender constructions and has to do with the way men and women, as well as society
in general, see themselves and each other with regard to gender norms and gender
identities. As women enter the labour force in increasing numbers, an understand-
ing of the relationships between income and capabilities becomes key to evaluate
these changes (Sen 1992; Nussbaum 2003). Whether or not a worker is able to trans-
late his/her earnings into capability-functioning depends upon a variety of factors,
such as the nature of household relations and decision-making processes in addition
to the terms of employment. As our study illustrates, we can observe both change
and continuity in the ways gender roles and gender differences manifest themselves.
Changes in gender constructions at the deeper level of the individual and society,
including the most intangible ways in which gender inequality operates, are
neither straightforward nor easy to measure even though they may be observable.
Yet, they are crucial for our understanding of the march towards gender equality.
Here an interdisciplinary analysis is important if we are to comprehend the
multiple levels through which they manifest themselves. To illustrate this with one
specific example from Latin America, Gutmann’s anthropological study (1996) of
gender transformations in a working-class community of Mexico City provides
very useful insights into how these changes can take place. His analysis of what he
calls ‘degendering’ – or processes through which gender stereotypes weaken –
shows how gender constructions are subject to change, even in the face of appear-
ances to the contrary. In this sense, and as Gutmann argues, ‘the gendered
character of social life is never transparent’ (p. 12). Thus, how the evolving
division of labour and the world of work translate themselves into changes at
the deeper level of identity construction may not be obvious or it may appear in
contradictory ways.
198 Lourdes Benería and Maria S. Floro
example that has inspired further work (Lund, chapter 10 in this volume).
Ultimately, however, the discussion and design of social protection schemes closer
to the notion of universal provisioning remains a challenge that cannot go away.
We will return to this topic in the last section.
The socioeconomic developments in Ecuador and Bolivia since the 1980s display
both similarities and differences, and the expansion of the informal economy
during the period has been a common feature. In Ecuador, this expansion was
facilitated by the pattern of economic growth that resulted in the formal market
economy’s incapacity to generate adequate jobs. Industrialization during the
1970s was capital-intensive and generated relatively little employment growth. By
the early 1980s, serious balance-of-payments problems ushered in an IMF-led
stabilization and structural adjustment programme (Floro 1992).3
The deepening of trade and financial liberalization policies during the 1990s
brought about a pattern of uneven development characterized by greater concen-
tration of wealth and further deterioration of working conditions. The rapid inte-
gration of Ecuador’s economy into the global market led to the increased use of a
flexible labour force by the productive sectors. Additional measures were thus incor-
porated in the Ecuador labour code in the early 1990s such as the use of temporary,
part-time, seasonal and hourly contracts and the replacement of indefinite labour
contracts with fixed-term contracts. At the same time, the reforms increased
restrictions on the right to strike, collective bargaining and the organization of
workers (CELA-PUCE 2002; Weinberg 2002).
Ecuador’s informal economy further grew in the context of persistently high
income inequality and poverty during the 1990s and a severe macroeconomic and
financial crisis in the later part of the decade (CELA-PUCE 2002). The Gini index
of income inequality increased from 52 to 54 between 1995 and 1999 (Parendekar
et al. 2002). By 1999, the extreme poor accounted for 21 per cent of the total pop-
ulation (12.4 million) and the poor accounted for 52 per cent of the population
(UNDP 2003).4 The Open unemployment rate increased during the period
1989–99, with women’s and men’s reaching 14.4 per cent and 19.6 per cent (UNDP
2002; ILO 2002). At the same time, underemployment remained at about 60 per cent
of the economically active population (CELA-PUCE 2002).
The Ecuadorian government response to the economic crisis further sustained
the deterioration of macroeconomic conditions. Its decision to adopt the US
dollar as its currency (called dollarization) resulted in high inflation and further
reduced the purchasing power of workers’ incomes (Solimano 2002). By 2001,
after dollarization, the real minimum wage had been reduced to 20 per cent of its
1980 value (CELA-PUCE 2002). Moreover, Ecuador’s fiscal austerity measures and
regressive taxation system have led to inadequate, ill-funded and poorly designed
social protection programs. Social spending decreased by more than a third, as
public spending on social services and social programme fell from US$78 to US$51
per person between 1995 and 2000. Ecuador’s level of social health services as a
200 Lourdes Benería and Maria S. Floro
result were reduced. The cut in social-services expenditures adversely affected the
Bono Solidario, a cash-transfer programme targeted at poor households, the school
meals programme and integrated early childhood care, and prenatal and neonatal
programmes so that their scope, type and degree of protection became even more
limited (Parandekar et al. 2002). Only a small proportion of the population, mainly
regular workers in medium and large firms as well as the public sector had access to
the social security system. Even then, the beneficiaries suffered huge losses as the
purchasing power of their monthly benefits declined drastically.
In the case of Bolivia, the dramatic expansion of the informal sector has resulted
from deep economic restructuring and the neoliberal reforms which began in the
early 1980s and were intensified with the structural adjustment programme (SAP)
adopted in 1985. Throughout the 1970s, rising foreign debt and the fall of export
commodity prices generated an economic crisis that peaked during the first half of
the 1980s. The crisis manifested itself in negative growth rates, hyperinflation, and
the declining value of the Bolivian peso. By 1985, the inflation rate had reached
levels way above 1,000 per cent,5 with debt service representing 45–50 per cent of
GDP (Morales Anaya 1987). This situation led to the adoption of the IMF-
sponsored SAP in 1985, which resulted in the ‘new economic policy’ (NEP).
The 1985 policies focused on structural reforms that redefined the role of the
government, turned the economy in a free-market direction through liberalization
of both the goods and labour markets, and promoted opening of the economy to
the rest of the world. Typically, it imposed wage freezes, reduction of government
spending, and a shift towards an export-oriented model of development. At the
same time, the 1980s saw the closing of the tin mines that had been a key export
sector and had provided secure employment and a way of life for an important
proportion of the working population. This led not only to social upheaval but
also to a rapid relocation of the mining population in 1986 (McFarren 1989).
The marked tendency towards increasing labour market informalization in
Bolivia was intensified with the 1994 Law of Popular Participation which included
several components emphasizing administrative and fiscal decentralization,
privatization, educational reform, and the restructuring of the country’s social
insurance system, among others (Kohl 1999; Benería-Surkin 2003). The economy
went through a period of recovery during the 1990s and economic growth was pri-
marily based on the growth of capital-intensive sectors, such as electricity, trans-
portation and financial services rather than that of the more labour-intensive
sectors such as manufacturing and agriculture (Republic of Bolivia 2001).
This implied that growth was achieved with insufficient job creation, resulting
in the persistence of unemployment and underemployment. Other chronic
problems also reappeared: the economy has been stagnant since 1999, with an
annual growth rate of 1.6 while the urban population has increased at 3.7 per cent
annually (UDAPE 2003). The current conditions of high unemployment rates,
government deficits, and foreign debt pressures, as well as critical levels of poverty
and significant disparities in regional economic growth have resulted in political
and social tensions. Under these circumstances, the informal economy has
continued to absorb a large part of the growing workforce.
Labour Market Informalization, Gender and Social Protection 201
We first examine the nature of informality that characterizes jobs in the urban
poor neighbourhoods of Bolivia and Ecuador’s largest cities. To highlight the
gender dynamics, we focus on 506 husband-and-wife respondents in a purposive
samples of urban, low-income households in these countries. Using information
collected from these respondents, we attempt to capture the different levels of
202 Lourdes Benería and Maria S. Floro
Table 9.1 Employment pattern of husbands and wives in urban, poor households, 2002
Bolivia and Ecuador Sample Survey, by sex (percentage of total in parentheses)
Bolivia Ecuador
By degree of
informality of Women Men ALL Women Men ALL
main job1 N ⫽ 137 N ⫽ 137 N ⫽ 274 N ⫽ 116 N ⫽ 116 N ⫽ 232
A. No. of employed
1. Formal/low 0 (0) 10 (7) 10 (4) 11 (10) 34 (29) 44 (19)
2. Medium 97 (71) 96 (70) 193 (70) 60 (45) 52 (45) 113 (49)
3. High/severe 22 (16) 20 (15) 42 (15) 22 (19) 29 (25) 51 (22)
All workers 119 (87) 126 (92) 245 (89) 9 (80) 115 (99) 208 (90)
B. Not 18 (13) 11 (8) 29 (11) 23 (20) 1 (1) 24 (10)
employed2
All respondents 137 (100) 137 (100) 274 (100) 116 (100) 116 (100) 232 (100)
Notes:
1
The categories on degree of informality are defined in the text.
2
‘Not employed’ refers to all heads or spouses who did not look for work in the previous month. Rounding
of percentages may not add up to 100.
Source: Authors’ calculations.
producers in traditional crafts, furniture, shoes, leather and garments sectors who
supply most of their goods to middlepersons. Over all, women workers in our sam-
ple tend to have relatively more precarious jobs than men in both countries,
implying greater vulnerability.
Categorization of employment by degree of informality of a worker is made
complicated by the decision of some workers, with the decline in real wages and
job security, to take on multiple jobs. Nearly a sixth of total respondents in the
Bolivia sample hold more than one job. The proportion is less in the case of
women and men respondents in Ecuador – about 10 per cent and 8 per cent
respectively. The multi-job dimension for some workers in Bolivia shows a coping
mechanism that serves as a buttress against job and income insecurity. This how-
ever may have welfare consequences in terms of lengthening the working day or
intensifying the work activities performed that can lead to, among others, stress
and deterioration of the health of the worker.
There is also a general absence of labour or social protection among these
workers, as reflected in Table 9.2. Only a very small proportion of the respon-
dents receive social-security benefits, and they are predominantly male workers
in the case of Ecuador. The lack of job benefits indicates the extent to which
economic insecurity permeates the lives of these workers, particularly women.
This is reinforced by the high degree of household-income variability examined
below.
204 Lourdes Benería and Maria S. Floro
Table 9.2 Access to benefits by workers, 2002 Bolivia and Ecuador Household
Sample Survey, by type of benefit and by sex (percentage of total in parenthesis)
Bolivia Ecuador
By type of
benefit Women Men Women Men
1
These are workers who answered the question(s). Several respondents left the question
of benefits unanswered and are therefore missing in this table.
Volatility in earnings
While the average monthly earnings of workers in the sample are not the lowest in
Bolivia and Ecuador, we have found a high degree of fluctuation, particularly among
workers with moderate and highly informal jobs. For some workers, this variation is
somewhat predictable due to the seasonal character of their activities; thus, the prob-
abilities of expected earnings in ‘low season’ or in ‘high season’ are likely to be known.
But for others, these probabilities are unknown and income variation tends to be idio-
syncratic and unpredictable. Using information based on the estimated ‘high’, ‘aver-
age’ and ‘low’ by the workers and their predicted or ‘subjective estimation’ of length
of ‘high’, ‘average’ and ‘low’ profit periods, we estimate an earnings-variability index
that allows us to compare across workers with different earnings patterns.
Table 9.3 shows average monthly earnings among self-employed workers with
predictable income variability which can dip to as much as a tenth of what they earn
during ‘high’ profit periods. The earnings variability index, called Zvar, is calculated
as the absolute value of the ratio of the weighted difference of the incomes during
high and low earnings periods to the total sum of the weighted incomes between
high and low earnings periods or:
Table 9.3 Monthly earnings and earnings variability of self-employed workers, 2002
Ecuador and Bolivia Household Sample, by sex (earnings in US dollars)
Bolivia Ecuador
Notes:
1
Refer to workers with predicted low- and high-level earnings period only. Reported monthly
earnings is for the previous month.
2
The earnings-variability index is discussed in the text.; a higher index (closer to one) means
higher variability; a lower one (closer to zero) means more stable earnings. Reference period is for
the past 12 months.
It is evident that low money incomes and job precariousness represent only one
facet of economic insecurity experienced by poor and vulnerable groups; intensi-
fication of work is another important facet that tends to shift the burden of
economic insecurity to certain members within the household. The interconnect-
edness of paid and unpaid work, vulnerability and gender roles is illustrated in the
case of home-based workers in two studies based on the 2002 UPHWS Thailand
survey data.
Home-based work is carried out by two types of workers who do remunerative
work within their homes, namely: a) independent own-account producers and
206 Lourdes Benería and Maria S. Floro
studies have examined (Benería and Roldan 1987; Kabeer 2000). If this is the case
and a renegotiation of household division of tasks occurs, women’s increased
participation in paid work may result in having other members take on her
domestic chores. But if gender norms are resilient to any change in money income
contributions, women may be compelled to increase the length of the working
day or intensify their work effort by overlapping activities, say combining
childcare and market work. As pointed out earlier in this chapter, gender roles in
Latin America have been changing rapidly in terms of the more observable and
measurable indicators. However, at the deeper level of gender norms and distribu-
tion of tasks between paid and unpaid work, we observe change as well as conti-
nuity. For example, a subsample of 29 households in Santa Cruz, Bolivia,
interviewed in January 2005, typifies these trends. When asked whether they
thought that gender roles were changing, 82 per cent replied affirmatively.7
However, the answers as to how they saw this change were more ambiguous;
57 per cent pointed out that women had more financial responsibility now
because of their greater involvement in paid work; about 43 per cent agreed with
the statement that it was easier for women to find jobs than for men; and 82 per
cent responded affirmatively to the question of whether they thought that gender
equality had increased.
Table 9.4 lists several proxies or indicators of the relative position of workers in
household decision-making. Based on the 2002 Bolivia and Ecuador household
workers sub-sample, the results show that there are relatively more women respon-
dents than men respondents who say that they require permission to work from
their partners/spouses. Roughly the same proportion of women and men reported
that they decide solely on the use of their own earnings. Another indicator of
Table 9.4 Patterns of household decision-making in couple households, 2002 Bolivia and
Ecuador Household Sample, by sex
Bolivia Ecuador
bargaining power is with respect to loan decisions. About 20 per cent of workers in
the Bolivian sample reported having made sole decision of when to borrow; the
majority reported that this is a joint decision. The proportion of individual deci-
sions to obtain credit is much higher in the case of Ecuador, especially among
men. But a different pattern emerges when it comes to the question of who pays
the loan or shoulders the burden of loan payment. A higher proportion of women
respondents in both Bolivia and Ecuador reported that they are solely responsible
for loan repayment compared to the men respondents. These have repercussions
in terms of a worker’s vulnerability and her future earning capacity as our case
study of Ecuador shows.
The mere existence of employment does not define the economic status and well-
being of a worker or a household. It needs to be seen as acting together with a
number of other factors, the most crucial being access to and control over
resources – mutual support networks, credit, savings and physical assets – in the
face of income shortfalls and consumption expenditure shocks.8 Income variabil-
ity, alongside the persistence of job insecurity, heightens the household’s need to
find ways of smoothing its consumption and designing survival strategies. Coping
strategies among the poor have received much attention in the literature on eco-
nomic and financial crises since the 1980s. As this section will illustrate, there is a
strong propensity for households with irregular and variable sources of income to
become increasingly differentiated. There is also a tendency for the risk burden to
be shouldered unequally among household members.
Table 9.5 Mean debt service ratio of borrowers, 2002 Ecuador Household
Sample, by sex and type of household headship
borrowers (0.378) than that among men (0.088). Given their dual roles as income
earners and primary care providers, Ecuadorian women workers in our sample seem
to have a more compelling need for credit both to meet the needs of their livelihood
activities, and to meet their household consumption expenses. The economic inse-
curity associated with informalized jobs therefore may have increased the demand
for credit to help smoothen consumption but with lessened inability to repay.
There can be important, far-reaching effects on future income from job insecu-
rity if some of the loans that are intended for productive purposes are themselves
used to buffer current consumption from income fluctuations. Home-based workers
who may need their own resources to pay for raw materials, equipment and other
inputs in their job, can find themselves severely capital-constrained and hence
unable to increase their earnings. Using the 2002 Ecuador survey data, the study
by Messier (2005) on poverty traps and credit provides empirical evidence on how
the allocation of credit, especially by self-employed women workers, for con-
sumption use (e.g. to meet food and educational expenses, household repair,
health care, and so forth) can adversely affect the level of investment in their own
enterprises. As a result, their income in the next period is likely to be lower or more
precarious, thus threatening social reproduction and, again, reflecting risk and
vulnerability.
provided by friends, relatives and/or neighbours comprise nearly 32 per cent of the
total number of loans, indicating the presence of socialized income-maintenance
schemes based on either extended kinship ties or locational affinity in urban poor
communities. In the case of one urban poor community in Quito called Itchimbia,
the strong cohesion and active community organizations have facilitated a high
degree of social cooperation and mutual support among its residents.
While annualized interest rates are roughly 24 per cent on formal/semi-formal
loans (predominantly from banks and microcredit organizations), and 73 per cent
and 44 per cent on informal loans (from private moneylenders) to men and
women borrowers respectively, kin loans are interest-free and require no collateral
(Table 9.6). Overall, women more often approach their relatives, friends and
neighbours for loans. Loan provisioning, as our Ecuador case study demonstrates,
is part of the income-maintenance schemes and mutual support systems that have
evolved to cope with the basic requirements and contingencies of life. Women
have played an important role as participants in such networks.
To some extent, community-based formal and informal networks have played
some role in meeting the transport, health care and social-service needs of the
community. This is the case among the urban poor communities in Bolivia and
Ecuador that we have encountered in our field work. Community-based trash,
sewer and medical services as well as some form of recreation facilities are
Table 9.6 Loan characteristics by gender and credit source, 2002 Ecuador Household Sample
(percentages in parentheses)
Formal/Semi- Formal/Semi-
formal Informal Kin formal Informal Kin
N ⫽ 29 N ⫽ 10 N ⫽ 19 N ⫽ 41 N ⫽ 25 N ⫽ 30
Notes:
1 Loan amount in dollars.
2 Nominal interest rates were annualized as follows: Interest rate ⫽ [(Repaid amount⫺Loan)/(Number of
monthly payments]*12. the inflation rate in 2002 was 12.5 per cent, using consumer price index.
3 Instalments refer to single payments made by the borrower to cover loan interest and principal.
4 Average dollar value of the monthly instalment.
5 Number of loans that required some form of collateral to guarantee the loan. Collateral included land,
house, durable goods, etc.
6 Cosigners include spouse, parents, other family members and neighbours.
Source: Floro and Messier (2004).
Labour Market Informalization, Gender and Social Protection 211
Concluding remarks
Third, one of the most striking findings from our study is the extremely low
degree of social protection among the households in the sample. Only a very small
proportion of the respondents received any type of social security benefits – the
exception being 19.1 per cent of male workers in Ecuador. Neoliberal policies have
had an implicit bias against any type of social protection programmes; we have
observed a privatization of security in the sense that, for the most part, it has been
left to kin and family networks to provide some means of protection. Mutual
support mechanisms and systems of reciprocity exist for example to provide loans
and as a way of smoothing consumption. Families also design living strategies
around housing and living spaces, such as relatives moving into a family home at
a time of crisis or loss of employment, or in cases of home-switching among family
members according to the affordability of the highest rent.
Ultimately, our study leads us to emphasize the notion that poverty elimination
programmes are not likely to succeed unless they are accompanied with policies to
create decent jobs for those living under the conditions of informality described in
this paper. Second, the study joins the voices that call for new forms of universal
social provisioning. Both point to a more interventionist state.
Notes
1 The sample survey is referred to here as 2002 Urban Poor Home-based Worker Survey
(UPHWS). The authors would like to specially thank John Messier for the fieldwork in
Ecuador and data processing, UNIFEM-Andean Region for the financial support received
and FLACSO (Facultad Latinoamericana de Ciencias Sociales), REMTE (Mujeres
Transformando la Economia) and FEDESO (Guayaquil) for their institutional support.
Many thanks also to Lourdes Montero and CEDLA (Centro de Estudios sobre el Desarrollo
en America Latina) in La Paz for their invaluable help with the Bolivia survey. We also
want to acknowledge the assistance provided by Lucia Salamea Palacios (UNIFEM-Andean
Region), Gioconda Herrera (FLACSO), Magdalena de Leon (REPEM), Maria Rosa
Anchundia (REMTE and University of Guayaquil), and Elizabeth Estrella (UNIFEM-
Andean Region) for their help with the training workshop in Ecuador and conference in
Quito in March 2004. Our thanks also to Sonali Duggal and Carmen Triana for their
generous assistance.
2 An elaboration of some of these points can be found in Benería 2001.
3 One of the main elements of this package was a set of labour-policy measures. These took
the form of: a) reduction of mandatory employment stability to all activities, b) wage-
setting at market-determined levels, and c) abolition of a firm obligation to pay workers
on strike (World Bank 1988).
4 A household is classified extremely poor if its total consumption expenditure is below the
food poverty line.
5 In fact, prices increased by 20,000 per cent from August 1984 to August 1985, and by
60,000 per cent between May and August 1985 (Sachs 1987). These represent the highest
inflation rates in Latin American history and one of the highest in the world.
6 The diversity of low-income neighbourhoods in the metropolitan cities can take the form
of old versus newly settled areas, proximity to or distance from main city centres, density of
population, and the degree of social cohesion represented by the presence or paucity
of grassroots organizations. It is important to note that the community selection for the
survey took into account the presence of contact such as a local community organization
or a community-based researcher which helped establish trust and initial rapport.
This facilitated the ‘entry’ of the survey team in the area.
214 Lourdes Benería and Maria S. Floro
References
Antonopoulos, R. and M. Floro (forthcoming) ‘Asset Ownership Along Gender Lines: Evidence
from Thailand’. Journal of Income Distribution.
Balakrishnan, R. and M. Huang (eds) (2000) The Hidden Assembly Line: Gender Dynamics of
Subcontracted Work in a Global Economy. Bloomfield CT: Kumarian Press.
Bardhan, P. and C. Udry (1999) Development Microeconomics. Oxford: Oxford University Press.
Benería, L. (2001). ‘Shifting the Risk: New Employment Patterns, Informalization, and
Women’s Work’. International Journal of Politics, Culture and Society, Vol. 15, No. 1,
September: 27–53.
Benería, L. (2003) Gender, Development and Globalization: Economics As If All People Mattered.
London: Routledge.
Benería, L. and M. Roldan (1987) The Crossroads of Class and Gender: Homework, Subcontracting
and Household dynamics in Mexico City. Chicago: University of Chicago Press.
Benería-Surkin, J. (2003) ‘Power, Conservation, and Indigeneous Livelihood: Guarani
Strategies for Conquering Political Space in Izozog, Bolivia’. PhD dissertation, University of
California, Los Angeles.
Boonmathya, R., Y. Praparpun and R. Leechanavanichpan (1999) ‘The Situation of Women
Subcontracted Workers in the Garment Industry in Bangkok, Thailand’. Working Paper,
the Asia Foundation, Bangkok, October.
Buechler, S. (2002). ‘Enacting the Global Economy in Sao Paulo, Brazil: The Impact of Labor
Market Restructuring in Low Income Women’. PhD dissertation, Columbia University.
CELA-PUCE (2002) ‘Evaluacion de Los Impactos Economicos y Sociales de las Politicas de
Ajuste Estructural en el Ecuador 1982–1999’, at http://www.saprin.org/ecuador/research/
ecu_res_ej_fin.pdf, accessed on 8 July 2003.
De Soto, H. (2000) The Mystery of Capital. Bantam Press/Random House.
Duryea, S., A. C. Edwards and M. Ureta. (2004) ‘Women in the LAC Labor Market: The
Remarkable 1990s’, in Claudia Piras (ed.), Women at Work: Challenges for Latin America.
Inter-American Development Bank, Washington DC: 27–56.
Economic Commission for Latin America and the Caribbean (ECLAC) (2004). Panorama
Social para América Latina y el Caribe 2002–2003. Santiago de Chile: CEPAL.
Floro, Maria (1992) ‘Macroeconomic Analysis Background Paper: The Case of Ecuador’.
Urban Poverty and Social Policy in the Context of Adjustment Project, Working Paper,
Washington DC: World Bank.
Floro, M. and J. Messier (2004) ‘Credit, Gender and Debt Servicing: The Case of Urban Poor
Households in Ecuador’. Paper Presented at the Southern Eastern Economics Association
Meetings, 19–20 November 2004, New Orleans.
Floro, M. and R. Antonopoulos (2005). ‘Asset Pawning among Urban, Low Income
Households in Thailand’, Levy Economics Institute Working Paper, forthcoming.
González de la Rocha, M. (2000) Private Adjustments: Household Responses to the Erosion of
Work. UNDP, Social Development and Poverty Elimination Division, Conference Paper
Series No. 6, New York: UNDP.
Gutmann, M. (1996) The Meanings of Macho: Being a Man in Mexico City. Berkeley/Los
Angeles/London: University of California Press.
International Labour Organization (ILO) (2001–02). Key Indicators of the Labour Market
(KILM). Geneva: ILO.
Labour Market Informalization, Gender and Social Protection 215
ILO (2002). Women and Men in the Informal Economy. Geneva: International Labour
Organization.
ILO and HomeNet Thailand (2002). Impact of the Economic Crisis on Homeworkers in Thailand.
Bangkok, International Labour Office.
Kabeer, Naila (2000) The Power to Choose: Bangladesh Women and Labour Market Decisions in
London and Dhaka. London and New York: Verso.
Kabeer, Naila (2003) ‘Labor Standards, Women’s Rights, Basic Needs: Challenges to Collective
Action in a Globalizing World’. In L. Benería and S. Bisnath (eds) Global Tensions:
Challenges and Opportunities in the World Economy. London: Routledge.
Kohl, B. (1999) ‘Economic and Political Restructuring in Bolivia: Tools for a Neoliberal
Agenda’. PhD dissertation, Cornell University.
Lavinas, L. (1996) ‘As mulheres no universo da pobreza: o caso Brasileiro’. Estudos Feministas,
4(2): 454–79.
McCrate, E. (1995) ‘The Growing Divide Among American Women’. Unpublished paper,
University of Vermont, Department of Economics.
McFarren, W. (1989) ‘Economic Crisis and the Politics of Adaptation: Survival Strategies of
Displaced Bolivian Mining Households’. MA thesis, Cornell University.
Messier, John (2005) ‘Dynamics of Poverty Trap and the Role of Credit’. Unpublished
doctoral thesis, Economics Department, American University, Washington DC, August.
Morales Anaya, J.A. (1987) ‘Estabilización y Nueva Politica Económica’. EL Trimestre
Economico, LIV: 179–211.
Nussbaum, M. (2003) ‘Women’s Capabilities and Social Justice’. In L. Benería and S. Bisnath
(eds) Global Tensions: Opportunities and Challenges in the World Economy. New York:
Routledge.
Parandekar, S., R. Vos, and D. Winkler (2002) ‘Ecuador: Crisis, Poverty and Social Protection’.
In P. Beckerman and A. Solimano (eds) Crisis and Dollarization in Ecuador: Stability, Growth
and Social Equity. Washington DC: World Bank: 127–76.
Pérez-Sáinz, J. P. (1994) El Dilema del Nahual. Globalización, exclusión y trabajo en Centramérica.
Costa Rica: FLACSO.
Pérez-Sáinz, J. P. (2005) ‘Labor Exclusion in Latin America: Old and New Tendencies’. In
L. Benería and N. Kudva (eds) Rethinking Labor Market Informalization: Poverty, Precarious
Jobs and Social Protection. Ithaca, NY: Cornell e-Publishing.
Pichetpongsa, Anant (2004) ‘Work Intensity and Well-being: The Case of Women
Homebased Workers in Thailand’. Unpublished doctoral dissertation, American
University, March.
Piras, C. (ed.) (2004) Women at Work: Challenges for Latin America. Washington, DC: Inter-
American Development Bank.
Portes, A. and M. Castells (1989) The Informal Economy. Baltimore: Johns Hopkins University
Press.
Republic of Bolivia (2001) Poverty Reductions Strategy Paper. La Paz.
Sachs, J. (1987) ‘The Bolivian Hyperinflation and Stabilization’. American Economic Review,
May: 279–83.
Salzinger, L. (2003) Genders in Production: Making Workers in Mexico’s Global Factories.
Berkeley: University of California Press.
Sebstad, J. and M. Cohen (2001) Microfinance, Risk Management and Poverty. Washington DC:
Consultative Group to Assist the Poorest.
Sen, A. (1992) Inequality Reexamined. Cambridge MA: Russell Sage Foundation and Harvard
University Press.
Solimano, A. (2002) ‘Crisis and Dollarization: An Overview’. In P. Beckerman and
A. Solimano (eds) Crisis and Dollarization in Ecuador: Stability, Growth and Social Equity.
Washington DC: World Bank: 1–16.
Unidad de Análisis de Politicas Sociales y Económicas (UDAPE) (2003) Crisis, Crecimiento y
Pobreza. La Paz, January.
216 Lourdes Benería and Maria S. Floro
Introduction
217
218 Francie Lund
work falls to private individuals (in the family) and to ‘community care’ or
‘volunteer work’. It is overwhelmingly women who take the larger share of this
unpaid caring work. In the period post-Washington consensus, there is more scope
for the role of the state, and poverty relief is on the social policy agenda (see
Molyneux, chapter 2 in this volume) – but women’s low paid and unpaid work is
not central to the analysis.
Especially for poorer women who work informally, the labour market is itself a
source of risk. Formal social-protection provision is not accessible; private (self-
funded) provision is not affordable, and small project-based provision is often not
sustainable (or affordable by the very poor). At the same time, poorer workers are
excluded from possibilities for skills generation or regeneration. While women’s
responsibility for health and welfare – in their own homes and in their communities –
continues, there seems to be little gendered exchange of productive and reproduc-
tive roles, even where there are high rates of male unemployment. Where chronic
poverty coincides with the AIDS pandemic, the caring obligations of women, and
to a lesser extent of men, increase and this influences access to and exit from the
labour market.
In 1952 the International Labour Organization (ILO), listed its core of contin-
gencies of social security: healthcare, incapacity for work due to illness, disability
through work, unemployment, maternity, child maintenance, invalidity, old age,
and death of the breadwinner. In advanced industrial economies, there was the
expectation that being in formal employment meant that there were contractual
obligations between employers and employees, which included some employer
responsibility for protection against risk, whether ex ante, anticipating risks that
would happen, or ex post, in response to various contingencies after they occurred.
This was frequently done with contributions from the state as well.
While the ILO in the 1950s expressed its hope that the contingencies for social
security could be expanded to all workers worldwide, there has in fact been a
reversal in social protection since then, and this has happened not only in those
developing countries which had some social security, but also in developed coun-
tries. What, actually, are poorer people supposed to do to cover their own risks
when they cannot afford the services provided by the private sector and when
their attempts at creating their own long-term formal and informal savings
founder because of the pressing needs of the present? Sustainable schemes of co-
insurance among poor people are difficult to achieve, as the core insurance principle
of cross-subsidization is absent. Further, expecting poor people to build their own
‘institutions for the poor’ is contradictory, in terms of investments of both time
and money required.
This chapter explores an approach to social protection which places women’s
low paid work as an important element for consideration in social policy. It holds
that patterns of production generate inequalities which disproportionately affect
women, and the growing size of the informal economy means that this will be a
key policy issue for the foreseeable future. Social policies which seek to address
both poverty and gender inequality will need comprehensive interventions from
multiple interest groups – informal workers, organized labour, the private sector,
Working People and Access to Social Protection 219
the state, and organizations in civil society. It is too ambitious to expect that all
workers, everywhere, will be able to get access to comprehensive protection mech-
anisms in the near future. The search for appropriate economic and social policy
responses will be assisted by reframing the standard risks and contingencies with
a focus on informally working women. This should take into account vulnerabil-
ity over the life cycle, and the specific contingencies and risks faced by informal
workers in their different places of work.
Many systems of social protection were based on a ‘family model’ or ‘male bread-
winner model’, with the idea of a male ‘primary breadwinner’ in full-time employ-
ment, and with access to insurance benefits secured through the relationship of
family members to the working person. Accommodating the fact that increasing
numbers of women are in the labour force, some detect a shift in welfare states to
an ‘adult-breadwinner model’ – with the assumption still that being in employ-
ment will bring with it access to measures of protection against risk. Men’s and
women’s risks are dependent on their relationship to other people in the
household; and households themselves, and members within them, go through
life-cycle stages, which affect men, women and children in different ways.
Different types of risks and vulnerabilities are associated with different stages in
the individual’s life cycle as well, and there are deeply gendered dynamics at play.
There are clearly overlaps between the different life stages, and some people may
not go through some of the stages. There are gender similarities and gender differ-
ences. Boy children may be at least as likely as girl children, in some countries, to
be withdrawn from school for household or other labour. Childbearing may begin
in adolescence and continue in later life. Not all women and men marry; single or
lone parenting has become more common, as has divorce. A woman may become
a grandmother at 50, and be responsible for the care of the children of her work-
ing daughter at the same time as being responsible for the care of her own 75-year-
old mother. Older men who are widowed are far more likely to remarry, and not be
alone, than older women.
For people in precarious employment, there is a close link between economic secu-
rity and social security. Loss of income may have different causes, for example sea-
sonal failure of produce, through loss of a contract, through inability to secure a
contract, or through unreliable and late payment for work done. Loss of assets may
arise through theft, fire and natural disaster; poorer informal workers such as street
vendors may lose assets through the destruction or confiscation by local authori-
ties of equipment such as selling-tables, carts, sewing machines, shoe cleaning out-
fits, and other tools. Another common reason for loss of income is ill-health, of
the worker herself, or of someone in her family for whom she has to care. Even
short interruptions of work can be a major disaster for the worker who has no paid
220
Table 10.1 Gendered risks and vulnerabilities associated with different stages of the life
cycle
leave, no paid sick leave, and no healthcare scheme. There is a vicious circle where
poor health lowers income, which in turns contributes to poor health.
Women worldwide take greater responsibility for childcare than men do. As
increasing numbers of women enter the labour market, this has an effect on the
need for childcare. Even poorly paid women hold that an advantage of working
from home is that multiple tasks can be seen to simultaneously, and in particular
that an eye can be kept on children. However, the presence of children, and the
need for children to be attended, requires inputs of both time and energy from the
worker/carer, and this will affect her productivity and ability to earn. Most welfare
states have a number of benefits surrounding the bearing and rearing of children.
A comprehensive package would include free healthcare, the right to extended
paid leave, the right not to be retarded in progressing along the career ladder by
having taken such leave, and in some countries, the rights of fathers to some
paternity leave (‘daddy leave’) as well. Some countries allow time off for breast-
feeding while at work, and work-based childcare facilities are provided. Workers
with no maternity benefits are at risk of loss of income when they take time off
work to have children, and are likely to return to work sooner than is good for
their own health and for the health of the newborn child, in order to earn again.
Women in export processing zones may be required to sign an agreement declaring
that they are not pregnant at the time of hiring, and that they will lose their
employment if they do get pregnant.
Though there are marked country and regional differences, girl children may be
at particular risk of having their education cut short so that they can work in the
family enterprise, or look after siblings while mothers work elsewhere. This may
reduce their chances of acquiring the skills necessary for earning a decent income.
However other factors are at work besides the demand for labour. Caste, custom
and religion can mire girl children into trajectories of little education, early mar-
riage and early pregnancy.
The vulnerability of unprotected elderly people has become a significant social
problem. The elderly may have been economically active all their lives, and
women in addition may have played a significant health and welfare caring role
all their lives, yet there is no certainty of security in their older years. There are crises
in pension systems worldwide, with demographic shifts (more people living far
longer), and cutbacks in social spending. Even those with formal coverage find
that, especially where the levels are not inflation-linked, the pension does not
stretch far enough. Many women’s access to financial security in old age is
through their working husbands. In most societies, women live longer than men.
Widowed men are more likely to remarry and have someone to care for them. A
woman in an advanced welfare state may get widow’s benefits, benefits for chil-
dren, and a lifetime pension for herself. In some societies, on the contrary, women
who are widowed are especially vulnerable to asset-raiding by the late husband’s
family, and to a lowering of social status (see Chen 1998 and Chen 2000 for an
account of the status of widows in parts of India).
Specific risks are presented by the place of work itself. There are hazards
associated with all forms of work, and formal workplaces are typically regulated by
222 Francie Lund
Men and women face specific risks over the life cycle, and additional specific risks
are attached to working in precarious work, as noted above. Specific groups of
women workers are excluded from access to social protection in specific ways. Four
such groups are considered here: migrants, sex workers, employed domestic workers
and unpaid full-time carers.
Migrants
A key feature of globalization is the mobility of labour. Increasing numbers of
people migrate in search of work, within their own countries, and across national
borders. While historically most migrants were men, sometimes accompanied by
women as family members, the numbers of women who migrate autonomously
are increasing (Kofman 2004). The highly mobile cadre of professional self-
employed migrant workers, who face uncertainty of employment and high risks,
are paid well enough to make their own provision for savings and insurance
against work-related contingencies. For the majority, though, the act of migrating
is a step into vulnerability. Women and men leave behind relationships with fam-
ily and friends, as well as neighbourhood and religious networks. They may also
leave behind or lose formal entitlements which depend on their status as citizens
in their local areas, such as access to health, welfare and other services. Many
people migrate through kinship or village-based networks, and go to places where
there is a known receiving community. Yet migrants typically assume minority
status, may be unfamiliar with the dominant language and have little knowledge
of ‘how things work’ locally. Women who migrate also leave behind a care deficit
(Chen et al. 2005: 28–30).
The access of foreign workers to social protection through work has become a
complex contemporary policy issue. For more formal workers, the access to or
exclusion from entitlements is at least transparent: a Turkish migrant who works
in a formal firm in Germany for a number of years will get access to social benefits
while there, and may be able to transport some benefits back to his or her family.
However, for millions of migrants, work comes with no social protection, and may
be unregistered. Workers are unorganized in labour terms (though they may have
Working People and Access to Social Protection 223
vibrant informal networks and associations), and are too dependent on the job to
be able to express demands. They have migrated precisely to work for an income,
and that income may be higher than they could ever earn in their home country.
Domestic workers
Worldwide, most domestic workers are women. They may live away from their
own homes, and in the homes of their employers (especially likely if they are
cross-border migrants). Working conditions are characterized by long hours, low
pay, and lack of autonomy. Working in someone else’s private domain renders
them both invisible and dependent on the employer. There is usually a large class
difference between employer and employee, and this may be accompanied by a
language difference as well, making communication difficult.
The relationship between (woman) employer and (woman) domestic worker is
complex. As women, they have shared interests in, for example, child development
and education. The career prospects of the working woman employer are contin-
gent on the (typically poorly paid) work of the domestic; the employer’s children’s
day-to-day security depends on the employee, who has to be away from her own
children to do this work. While the employer–employee relationship offers the
worker little autonomy or voice, there may be informal measures of assistance
such as with health costs, or school fees of children. However these are not con-
tractual obligations, and as such depend on the whim of the employer, and this is
not a sound basis for building long-term security.
society. Health and welfare policies are premised on some model of the family, typ-
ically one in which there is an infinite supply of women’s caring time. Decisions
to allocate more, or less, to the development of health and welfare services will
have direct effects on the time that has to be spent in families, and in communi-
ties, in unpaid care work. This work is overwhelmingly done by women, and is not
classified as work or employment in the systems of national accounts.
Concepts and practices such as ‘community care’ and ‘home-based care’ conceal
much of this work; it is people living in families and neighbourhoods who do the
caring. There can of course be a positive value-base to community care – people in
need of care may live a qualitatively better life when cared for at home and in the
community, as opposed to in an institution such as a hospital or home for elderly
people. Many people get deep emotional satisfaction from the caring role.
However, community care is too often introduced in a cost-cutting environment
as part of a neoliberal regime which also carries conservative messages about the
primacy of the family, and women’s place in the home.
AIDS is having a catastrophic effect in many countries of the world, and where
resources for formal health and welfare services are scarce, there will be increasing
calls for home care and community care. Though AIDS is no respecter of social
class, in developing countries the burden of AIDS falls most heavily on poorer peo-
ple, and in sub-Saharan African countries, on poorer women. It is contradictory for
international agencies and national governments to say that they are pursuing
gender-sensitive policies, and pro-poor policies, at the same time as they advocate
cutbacks in social spending, privatization of health and welfare services, and
community-based care.
with low and erratic incomes. In many countries, the implementation of labour
laws on basic conditions of employment and social protection is not properly
monitored even for those in formal employment. Bearing these difficulties in
mind, however, there are examples of innovation which show that progress can be
made. The first example below comes from India, and describes an integrated
insurance scheme, built by poor women, for poor women and their families, in
partnership with government and with external support. The second comes from
Costa Rica, which designed a voluntary scheme for independent and unremuner-
ated workers, which over time will become compulsory for independent workers.
The final example is the South African elderly pension, which provides basic
security for millions of older and poorer men and women, and assists also the
members of the households in which they live.
repaid to the member. Those paying the fixed deposit receive a wider package of
benefits, in particular access to a maternity grant.
The life insurance covers natural and accidental deaths of the member (and
more recently her husband). Asset insurance covers losses experienced by the
woman due to fire, floods and riots, and reflects SEWA’s focus on responding to
disasters in such a way that members are able to restore their income-earning abil-
ity as fast as possible. The health component includes hospitalization up to a max-
imum amount per year; those paying premiums through the fixed deposit get
additional coverage for cataract operations, hearing aids and dentures. The scheme
allows members to choose their own healthcare provider, private or public, though
SEWA gives advice about reliable providers.
Husbands can get access to limited benefits; they receive the insurance against
wives’ natural and accidental death, and women get coverage for their husbands’
death. Children have recently been included in the scheme as well. The SEWA
member, though, retains sole control of the SEWA Bank account through which
these benefits operate.
Vimo SEWA is a most interesting partnership between government, the insur-
ance industry (which until recently was government-run) and SEWA. Essentially,
the Indian government subsidized two large insurance corporations to offer some
of their services to poor people, and SEWA was one of a few large NGOs or trade
unions to take advantage of this opportunity. Vimo SEWA started in a partnership
with government insurance; SEWA then took the health scheme over themselves
for a while; they have recently again entered a partnership in which they purchase
insurance from a government general insurance company, but SEWA keeps control
of enrolment, approval and processing.
SEWA attributes some of the success of the insurance to the interaction between
its different programmes: the SEWA Bank, into which annual premiums are paid;
the health education by SEWA Health, which heightens members’ awareness of
health problems; and the literacy training provided by SEWA Academy, which
enables members to interact with the literate world of formal finance. It also ben-
efited from a large grant from Germany’s GTZ which allowed it to secure a finan-
cial base, cover administrative expenses, and also to innovate and experiment.
SEWA’s documentation of the scheme shows a real alertness to emerging problems,
and a process of continual innovation and experimentation, in response both to
objectively changing risks, and to needs expressed by members. As an ILO study
put it, it is a story of ‘a process of continuous adjustment sustained by great
creativity and tenacity’ (ILO 2001: 79).
Costa Rica’s population of some four million is about 50 per cent urbanized, and
it scores high in the region in terms of its per capita income, which was $3,960 in
2000. It has a long history of extensive social-security coverage for its relatively
small and homogeneous population. Yet demographic changes and changes in the
labour-market are presenting new challenges to social protection. The last two
decades have seen a rapid increase in women’s labour market participation, but on
poor terms relative to men. Most women work in the services sector where the
gender earnings gap is pronounced; for those with little education, the 1999 aver-
age wage of urban women was 59 per cent of the average wage of urban men. The
unemployment rate in 1999 was 6 per cent, much lower for men (4.9 per cent)
than for women (8.2 per cent) – but it should be noted how low these unemploy-
ment rates are compared to many low-and middle-income countries.
An increasing proportion of the Costa Rican labour force is not covered by occu-
pationally related social insurance, partly because of the increase in numbers
working in the informal economy. Some transnational firms and factories have
imposed labour practices which are illegal in the country: it is common knowledge
that people work for less than the minimum wage and for illegally long hours, and
that the right of women to breastfeed during working hours is not respected.
The World Bank tried to introduce pension reforms in 1994. The unpopularity
of these reforms led to widespread unrest, and a ‘Forum of National Concertation’
was then set up to work towards consensus about restructuring the social benefit
system. One of the components of the new agreement was the introduction of a
voluntary insurance scheme. This is available for independent workers – own-
account workers and non-remunerated workers, such as family workers and house-
wives, as well as students. This income-tested scheme is aimed at those who have
never contributed to a health or pension plan, or who did so for too short a period
or at too low rates to accumulate adequate benefits. Introduced as a voluntary
scheme, it is to be made statutory by 2005, when independent workers will be
required to join. It is funded by contributions of the state and those who join, with
the state contributing 0.25 per cent and the independent worker contributing
7.25 per cent of reference income.
By 2003 it was estimated that nearly three-quarters (74 per cent) of indepen-
dent workers contributed to the health insurance, while nearly one-quarter
(24 per cent) contributed to the pension insurance. It is remarkable that so many
independent workers are reached by the health insurance component, and it
makes the comparison with the much lower numbers opting for an old-age bene-
fit very striking. Probably, people opt not to join because old age seems far away,
there are conflicting immediate demands on available resources, and the returns
to present contributions seem uncertain. In addition, there is a disincentive to
joining because there is a non-contributory pension scheme for poor adult Costa
Ricans who are not covered by another regime.
This extension of social protection is an interesting example where a country
with a good history of social provision is attempting to adjust to changes in the
labour market, as well as to demographic changes, in flexible ways. It is trying to
set up seamless links between contributory and non-contributory schemes. It does
228 Francie Lund
not yet, however, address social protection for those who are waged workers but
who are not properly covered.
Informal men and women workers thus effectively have a guarantee of partial
economic security in their elderly years, affording them an earned place in the
household.
There are of course problems with any system of this nature. Some are con-
cerned that a benefit targeted to one group – the elderly – is consumed by the
household at large. There is some evidence that it may reduce the amount of
remittances sent from migrant household members. Pensioners themselves com-
plain about administrative problems, particularly with delays in receiving pay-
ments, and with the review system (eligibility has to be reassessed from time to
time, to check whether people are still poor enough to receive the pension, though
escaping poverty in the elderly years is highly unlikely). There are relatively minor
Working People and Access to Social Protection 229
concerns about leakage in the system to the better-off. Generally though the sys-
tem delivers pensions monthly to between 70 and 80 per cent of all elderly South
Africans, at an amount of approximately US$3 a day at present levels. Pension-
receiving households (which are larger than non-receiving households) are still
poor, and households in the very poorest income deciles are there because they
have no elderly person, and no disabled person (there is a similar social assistance
system for people with disabilities).
The numbers of people in the older years is a small fraction of the overall popu-
lation, and the OAP has been judged to be sustainable and affordable. The AIDS
epidemic has now reduced longevity drastically, and in future proportionally
fewer of the population will reach eligible age. Older people have already taken on
more responsibilities for caring for children whose parents have died of AIDS. The
OAP will continue to make a vital contribution to household security.
systems. SEWA found that an obstacle in its health insurance is that ‘it has
attracted a membership that is older and more likely to experience illness’ than the
general population of SEWA members (Chatterjee and Ranson 2003: 294).
Chatterjee and Ranson (2003: 294) identify another lesson of Vimo SEWA:
that improving access to health care may be ‘a mixed blessing where the qual-
ity of care is variable and, at worst, dangerous’. Where poorer people can only
get access to poor quality health services, and facilities in which for example
there may be little in the way of reproductive health services for women and for
men, then it is not surprising that people may avoid going to those health
services.
None of the case studies focused on employer contributions, and there is a need
for further exploration as to how to secure employer participation in cases where
there is an employer–employee relationship. India has a system of welfare funds
for workers in certain economic sectors, such as tobacco, building and construc-
tion, the film industry, certain mining sectors including limestone and dolomite,
and iron ore, manganese ore and chrome ore. Levies or cesses are placed on the
production or sale of goods in these sectors, and the welfare funds are used for col-
lective benefits in housing and education for workers and their children
(Subrahmanya 2000: 70). This overcomes the problem of identifying and securing
contributions from individual employers. Subrahmanya draws attention, though,
to significant problems, among which are administrative inefficiency, under
spending of the funds, and the fact that in general they do not see to the social
security needs of individual workers.
A number of examples show the importance of the role of the state in being
able to deliver to large numbers of people, through existing and new institu-
tions. This is clear in the South African OAP and the Costa Rican scheme and
points to the need for greater acknowledgement of this role, and for seeking
ways of making state interventions even more effective. Thailand has recently
introduced the ‘30 baht’ health insurance scheme which is available to informal
workers (as it is to all Thai citizens). India has agreed to pilot a new scheme of
social security specifically for ‘the unorganized sector’ or informal workers. Both
of these new programmes should be monitored closely; both carry the potential
to reach previously uncovered categories of worker, but possibly not the very
poorest.
None of the case study examples involving informal workers addressed the issue
of AIDS directly. In countries with a high AIDS prevalence, the epidemic is eroding
social protection for all people. Formal insurance companies are raising premiums
for policy-holders. The actuarial basis of smaller and informal savings and insur-
ance mechanisms is affected, as members themselves become sick, or have to use
their savings and insurance on health and funeral costs for family members.
Healthcare schemes are capping the value or number of services. Even in countries
with relatively sophisticated health services, people in the advanced stages of AIDS
are sent home from hospital to die, as hospital wards are full and priority is given
to assisting those who are not yet in the terminal stage. This increases the need for
home-based care.
Working People and Access to Social Protection 231
Conclusion
The world is facing a growing crisis with the problem of the increasing insecurity
of poorer workers. Millions of people work all their lives yet remain poor, and
whole age cohorts are entering their elderly years with no savings for retirement.
This crisis will not be solved by patchwork responses. Some of the policy pres-
criptions that are put forward to deal with this problem are destined to fail,
because they do not take into account the reality that poorer people face greater
risks, through their poverty and through the kinds of work that they do. Nor do
they take into account that women face specific risks of their own. In line with the
conclusion of Benería and Floro (Chapter 9 of this volume), this chapter points to
the need for substantial and sustained state intervention, and a role for significant
donor funding as well.
Examples have been given of innovations in social protection that have reached
large numbers of people, including large numbers of women. They show that the
state has a constructive role to play in extending existing measures of social pro-
tection to more categories of worker, and also in working in partnership with orga-
nizations of informal workers. Through the SEWA example, we see an organization
that both promotes self-reliance and advocates an active role for the state. Employ-
ers and owners of capital are, however, uniformly hard to identify and to reach.
It has been argued that the informal sector and workers within it are in fact cov-
ered by some aspects of labour standards (Trebilcock 2004). In addition to the call
for extension of labour legislation, though, a multi-pronged strategy is needed
which considers different conditions governing different industrial sectors, the
institutional capacity to implement labour codes, and the potential effectiveness
of different kinds of alliances between different interest groups, at different times.
In addition, there is a need to understand how both international labour codes
and national labour legislation can be mutually reinforcing, bearing in mind that
some of the fair-trade initiatives have voluntary codes of conduct which have in
fact fewer protections for workers than existing national labour legislation. A core
strategic question is: what combinations of local, national, regional and supra-
national alliances and interventions would give access to people in different
employment relationships?
The access of informal workers to better (or any) measures of social protection is
contingent on their building strong and effective organizations. Local-level orga-
nizations are the vehicle through which women informal workers especially can
practise empowerment in day-to-day ways. Organizations are also the vehicle
through which to express their voice and interests. Building organizations of
informal workers is a long struggle, and social-protection benefits can be a focus
around which to organize. Informal workers need to find institutional spaces
where they can directly represent their own interests, and influence the way poli-
cies are formulated and implemented.3 Homenet Thailand, an organization pro-
moting the interests of industrial outworkers in the garment industry, worked
actively in the Thai Parliamentary Commission on Universal Health Care Coverage,
which resulted in the introduction in 2002 of the ‘30 baht’ universal healthcare
232 Francie Lund
Notes
1 This case study draws primarily from Chatterjee and Ranson (2003) and ILO (2001).
2 This case study relies exclusively on Martinez Franzoni and Mesa-Lago (2003).
3 There have been good examples in the last few years; for a comprehensive account of
examples of good practice, see Chen et al. (2005): 67–77.
References
Ardington, Elisabeth and Frances Lund (1995) ‘Pensions and Development: Social Security as
Complementary to Programmes of Reconstruction and Development’. Development
Southern Africa, Vol. 12, No. 4: 557–77.
Bandyopadhyay, Nandinee, with Swapna Gayen, Rama Debnath, Kajol Bose, Sikha Das,
Geeta Das, M. Das, Manju Biswas, Pushpa Sarkar, Putul Singh, Rashoba Bibi, Rekha Mitra
and Sudipta Biswas (2004) ‘ “Street Walkers Show the Way”: Reframing the Debate on
Trafficking from the Sex Workers’ Perspective’. IDS Bulletin, Vol. 35, No. 4: 104–11.
Barrientos, Armando and Stephanie Ware Barrientos (2003) ‘Social Protection for Workers in
the Horticulture Industry’. In Frances Lund and Jillian Nicholson (eds) Chains of
Production, Ladders of Protection: Social Protection for Workers in the Informal Economy.
Durban: School of Development Studies, University of Natal.
Case, Anne (2001) ‘Health, Income and Economic Development’. Paper presented at the
ABCDE Conference, World Bank, Washington DC, May.
Case, Anne and Angus Deaton (1998) ‘Large Cash Transfers to the Elderly in South Africa’.
Economic Journal, Vol. 108: 1330–61.
Working People and Access to Social Protection 233
Chatterjee, Mirai and M. Kent Ranson (2003) ‘Livelihood Security through Community-
based Health Insurance in India’. In Lincoln Chen, Jennifer Leaning and Vasant
Narasimhan (eds) Global Health Challenges to Human Security. Cambridge MA: Global
Equity Initiative, Asia Center, Harvard University.
Chen, Martha (1998) Widows in India: Social Neglect and Public Action. New Delhi: Sage.
Chen, Martha (2000) Perpetual Mourning: Widowhood in Rural India. New Delhi: Oxford
University Press.
Chen, Martha, Joann Vanek and Marilyn Carr (2004) Mainstreaming Informal Employment and
Gender in Poverty Reduction: A Handbook for Policy-makers and Other Stakeholders. London:
Commonwealth Secretariat.
Chen, Martha, Joann Vanek, Frances Lund and James Heintz, with Chris Bonner and Renana
Jhabvala (2005) Progress of the World’s Women 2005: Women, Work and Poverty. New York:
UNIFEM.
Doane, Donna, Rosalinda Ofreneo and Paonoi Srikajon (2003) ‘Social Protection for Workers
in the Garment Industry’. In Frances Lund and Jillian Nicholson (eds), Chains of Production,
Ladders of Protection: Social Protection for Workers in the Informal Economy. Durban: School of
Development Studies, University of Natal.
Heintz, James (2005) ‘The Growing Informality of Labour Markets’. Paper presented at the
UNRISD/CGGS Conference on Gender and Social Policy, Marstrand, Sweden, 28–29 May.
ILO (2001) Women Organizing for Social Protection: The Self-employed Women’s Association’s
Integrated Insurance Scheme, India. Geneva: ILO
ILO (2002) Women and Men in the Informal Economy: A Statistical Picture. Geneva: ILO.
Kofman, Eleonore (2004) ‘Gendered Global Migrations’. International Feminist Journal of
Politics, Vol. 6, No. 4.
Lund, Frances (2002) ‘ “Crowding in” Care, Security and Micro-enterprise Formation:
Revisiting the Role of the State in Poverty Reduction, and in Development’. Journal of
International Development, Vol. 14, No. 6: 681–94.
Lund, Frances and Jillian Nicholson (eds) (2003) Chains of Production, Ladders of Protection:
Social Protection for Workers in the Informal Economy. Durban: School of Development
Studies, University of Natal.
Lund, Frances and Smita Srinivas (2000) Learning from Experience: A Gendered Approach to
Social Protection for Workers in the Informal Economy. Geneva: ILO.
Martínez Franzoni, Juliana y Carmelo Mesa-Lago (2003) La reforma de la seguridad social en
costa rica en pensiones y salud: Avances, problemas pendientes y recomendaciones. San José,
Costa Rica: Fundación Friedrich Ebert.
Orloff, S (2000) ‘The Significance of Changing Gender Relations and Family Forms for
Systems of Social Protection.’ Background paper prepared for the World Labour Report
1999–2000. Geneva: ILO.
Subrahmanya, R. K. A (2000) ‘Welfare Funds: An Indian Model for Workers in the
Unorganised Sector.’ In Renana Jhabvala and R. K. A Subrahmanya (eds) The Unorganised
Sector: Work Security and Social Protection. New Delhi, Thousand Oaks and London: Sage.
Trebilcock, Anne (2004) International Labour Standards and the Informal Economy. Geneva: ILO.
UNRISD (2005) Gender Equality: Striving for Justice in an Unequal World. Geneva: UNRISD.
Von Ginnekin, Wouter (ed.) (1999) Social Security for the Excluded Majority. Geneva: ILO.
Vosko, Leah (2005) ‘Precarious Employment: Towards an Improved Understanding of Labour
Market Insecurity’. In Leah Vosko (ed.) Precarious Employment: Understanding Labour Market
Insecurity in Canada. Montreal: McGill-Queens University Press.
Part III
Gender Dimensions of
Social-sector Restructuring
11
Gender and Health Sector Reform:
Analytical Perspectives on African
Experience
Maureen Mackintosh and Paula Tibandebage
Vignette 1
A small private dispensary in the Tanzanian Southern Highlands was run in 1998/9 on a
day-to-day basis by a trained nurse, a woman, and overseen by her husband, a medical
doctor, who was not often there since he was employed in a mission-run hospital. The
nurse saw patients, managed other staff, and did the books; unusually in a private dis-
pensary she provided some preventative care such as vaccinations and ante-natal care.
Most of the clients we met were women, many bringing young children or other sick
relatives.
Health systems are gendered institutions. That is, their organization reflects and
responds to gender inequalities in the wider society. The hierarchy among health care
staff places doctors, policy-makers and administrators – predominantly male –
above nurses, paramedical staff and orderlies who are more likely to be female.
Furthermore, in Africa as in many parts of the world, day-to-day working relations
in health care are predominantly between women: nurses and lower-level staff
dealing with women accompanying sick children and the elderly, women caring
for relatives who are in-patients, or coming for care themselves (Vignette 1).
Furthermore, mother and child health care (MCH) is known to be important to
society’s health outcomes (WHO 2003a, LaFond 1995).
We would therefore expect, when we turn to the prescriptive and analytical lit-
erature on the health-sector reforms that have been proposed and implemented
across sub-Saharan Africa, that gender would be a key tool of analysis, and that the
literature would explain how the reforms address women’s needs.
Yet this is not so. The influential multilateral policy documents of the 1990s are
largely silent on gender. The World Development Report 1993: Investing in Health
(World Bank 1993) mentions ante-natal, delivery and post-natal care and family
237
238 Maureen Mackintosh and Paula Tibandebage
planning as ‘clinical interventions’ that are ‘highly cost effective’; and the
importance of schooling for girls, of ‘educated mothers’, and of women’s access to
income. It discusses the gendered pattern of disease burden and the problem of
insensitivity of health care to women’s broader health needs. Yet the moment we
turn to proposed reforms gender largely disappears. Discussion of the core policy
recommendation to introduce user fees, and the sections on insurance, market
liberalization and structural reform do not refer to gender differences in need or
ability to pay.
Similarly, the World Development Report 1997: The State in a Changing World
(World Bank 1997) outlines ‘new public management’ reforms without reference
to differential effects on men and women. The World Development Report 2004:
Making Services Work for Poor People (World Bank 2004) discusses clients and
providers, citizenship and health services in a largely ungendered manner, exam-
ining ‘co-production’ of health between communities and health sector staff with-
out recognition of its reliance on low-income women’s unpaid work. The silence
also runs through the Millennium Goals agreements, which identify gender equity
as an objective but are mute on the institutional changes that might be needed to
eradicate gendered inequity from key public-service institutions.
The silence has now been broken however by the World Health Organization, in
the World Health Report 2003: Shaping the Future (WHO 2003a) which identifies
gender discrimination in the health professions as a serious problem for the deliv-
ery of services to poor and disadvantaged populations. The report also promotes
women’s participation in health care governance and management.
This chapter argues that despite the silences, health sector reform models are
implicitly gendered: that is, they have gender built inaudibly into their assump-
tions. Section 2 explores analytical approaches to the gendered nature and impacts
of health sector reform: gender equity, women’s health needs and gendered health
systems frameworks. Our objective is conceptual: to examine health-sector reform
through a gender ‘lens’, considering how the reform framework is gendered and
the extent to which that gendered process may operate to the detriment of
women, especially poor women (Sections 3–6). We illustrate our arguments with
empirical evidence drawn largely but not exclusively from Africa. Section 6
considers implications of the analysis particularly for the African context.
Health sector reform models are themselves not without internal contradictions,
and contain gendered assumptions that may be problematic.
Health sector reform as promoted in sub-Saharan Africa has particularly encom-
passed the following (WHO 2000a, Mills et al. 2001):
● What are the assumptions about household and family structure that underlie
health sector reform and are they justified? How do health financing strategies
respond to changing social understandings and experience of women’s finan-
cial dependence or autonomy within the household?
● What are the assumptions about women’s labour market participation – including
participation in the health-sector labour force? How, for example, do wider
work opportunities for women influence the nursing labour market?
● What are the assumptions about women’s capacity to undertake unpaid care? Does
health care reform make unjustified assumptions about women’s availability to
care for sick children and the elderly?
● How do changes in health care funding and expenditure differentially affect
women and men in different social groups? What are the gender-differentiated
effects of changes in the pattern of public expenditure on health care, the struc-
ture of the tax system, and the balance between public and private health
finance?
● To what extent do health and social services influence women’s circumstances as
well as responding to them? How do these services (re)construct the nature of
women’s citizenship and women’s rights, and how these are differentiated
according to income, social class, ethnicity or location?
● What are the assumptions about health sector governance in reformed frame-
works: who will participate and how? How can users from different social
groups exert leverage over resources through governance structures?
back into social patterns of disadvantage. The standards and principles of evalua-
tion of health systems and the reform models themselves reflect patterns of estab-
lished disadvantage including gender hierarchies (Twigg 2000; Mackintosh 1998).
Conversely, health systems’ immense political and social importance, and the
association of health care reform with political and economic crisis, implies that
health systems can be sites of challenge to gender disadvantage. We seek in this
chapter to contribute to a relevant set of analytical questions about gender and
health sector reform for the African context.
Key questions:
How do changes in health care funding affect differentially women and men in
different social groups?
What are the underlying assumptions in HSR models about women’s access to
cash and are they correct?
Table 11.1 Selected African countries: percentage of the labour force that is female, and
percentage of the male and female labour force receiving wages or salaries (most recent
year available)
Ghana 1990 51 33 9
Ethiopia 2000 41 10 7
Kenya 1995 46 32 12
Nigeria 1995 36 49 33
Tanzania 2000 42 n/a n/a
Uganda 1995 48 21 7
Zimbabwe 2000 45 51 22
Zambia 1995 45 31 9
However most community-based MHI schemes face a problem of very low par-
ticipation rates. In Tanzania, many rural Community Health Fund (CHF) schemes
were found to have participation rates below 10 per cent of eligible households
(Tibandebage 2004).1 Such low enrolment is not unusual in developing countries;
reasons cited by non-members include inability to pay, lack of transparency and a
decision-making process that does not effectively involve members (Jutting and
Tine 2000; Tibandebage 2004). Implementation of exemption for poor households
has been weak.
The low enrolment may disproportionately affect women since they are generally
less able than men to afford payment at the time of illness, and are likely to have
more health needs. In one rural scheme in Tanzania, members also lacked knowl-
edge of some of the benefits they were entitled to, such as referral to a hospital,
which could be disastrous for women with, for example, complicated pregnancies in
need of hospital-level care but with no out-of-pocket cash (Tibandebage 2004).
A strong assumption behind the financing reforms in HSR was thus that
women’s incomes and mutual support networks were robust enough to withstand
the payments demanded. The sharp decline in usage and low levels of participa-
tion in community insurance schemes documented above show how unrealistic
that assumption was in contexts of generalized poverty.
Key questions:
How has HSR affected the conditions of women as health care workers?
How have those effects in turn influenced women’s experience as health care users?
What are the assumptions underlying HSR about women’s interactions as workers,
users and citizens and are they correct?
Gender and Health Sector Reform 245
As the ILO noted in 2004 in its social dialogue on gender and health services, ‘in
most countries the workforce in the health sector is predominantly female’, yet
‘they fall at the bottom of the hierarchy in terms of authority, remuneration and
qualification’ (www.ilo.org/dialogue, accessed on 27 April 2004). Case studies
repeatedly confirm this, and it is likely to hold across Africa despite a tradition in
some countries, including Mozambique, of men going into nursing (WHO 2003b;
Ngufor 1999; Corkery 2000). Therefore downward pressure on wages, working
conditions, employment and professionalism fall particularly strongly on women
health workers in all sectors.
Vignette 2
This kind of thing [bad behaviour] happens, and it is because of poor morale, low com-
mitment, severe overwork and low salaries. Imagine, you are a nurse on duty … You may
have a ward of 40–60 seriously ill patients … You are two trained people at best. How will
you divide yourself? You are constantly over-working and under pressure. You are wor-
ried about family problems and commitments. For twelve hours you do not know what
is happening to your children. And you may not have as much as a cup of tea. Then there
is the problem of the commitment of other staff. You are a nurse by profession. The doc-
tor, who is supposed to be responsible, works his official hours and goes away, he waits to
be called. You are there, someone is bleeding and you cannot help them. She needs to be
operated on, no one is there, there are no facilities … . Relatives rush to send the sick per-
son to hospital, then we are not in a position to help the patient.
Public-sector decline and health sector liberalization tend to widen the gap in
wages and working conditions between doctors and nurses. More doctors than
nurses do additional private practice, and the ‘going rate’ for informal payments
to doctors was substantially higher than for nurses.3 Nurses have far more contact
with patients than doctors, and when the service falls apart they take most of the
strain. The matron quoted above described an acute shortage of trained nurses;
very low pay associated with the loss of allowances (such as for transport and uni-
forms, which they now had to pay for) as a result of reform; and lack of access to
training opportunities since specialist nurse training courses now had to be paid
for by the nurses themselves ‘and if people have a family they simply don’t see this
as a possibility’.
The result, as this matron confirmed, was that nurses had started to go abroad,
initially from Tanzania to Zimbabwe. In the five years since then, this flow
has accelerated from many African countries, as high-income countries have
expanded international recruitment (Mensah 2005). Doctors migrate too and in
Ghana, for example, most incentives aimed at retaining health care staff have
been focused on doctors, increasing nurses’ resentment. Given that, as Mensah
puts it, nurses are the ‘backbone’ of African health services, the rising out-migra-
tion is having extremely serious effects on services, and very specifically on pri-
mary provision on which low income women predominantly rely.
Gender and Health Sector Reform 247
Key question:
Has HSR improved health systems’ capacity to help redress women’s vulnerability
to reproductive ill health? If not, why not?
Vignette 3
A 16-year-old girl from a poor rural family was about to complete primary school, but had
often been sent home for lack of school supplies. One day a man offered her gifts and
bought her exercise books. In return he wanted a relationship. Because she felt she
needed his help, the girl accepted and knowing nothing about contraception she soon
got pregnant. She hid her pregnancy until an advanced stage and then had to quit school.
She went through almost the whole pregnancy without attending an ante-natal clinic. At
delivery she was taken to a nearby health centre and there was a delay as the mother
rushed back home to borrow the money. The girl had complications during delivery and
because of her young age and the delay she lost the baby. Luckily the girl survived.
Globally, 20 per cent of the burden of disease among women of reproductive age
is connected with sex and reproduction; in sub-Saharan Africa it is 40 per cent
(UNFPA 2002). Women and adolescent girls continue to face uncontrolled fertility,
maternal morbidity and mortality, unwanted pregnancy, early childbearing,
sexually transmitted diseases including HIV/AIDS, sexual violence, and female
genital mutilation (FGM) which often results in reproductive-health-related
complications. The average maternal mortality rate in sub-Saharan Africa (SSA) is
1,000 deaths per 100,000 live births as compared to a world average of 400 (UNFPA
2002) Africa has one of the highest teenage pregnancy rates in the world, often
resulting in unsafe abortion (WHO 2000b).
The patriarchal ideology which treats women as inferior to men and allows
them few decision-making powers also continues to bear heavily on women’s
sexual and reproductive health. Women lack freedom over their sexual lives,
putting them at risk of contracting STDs as unwilling partners, and lack resources
for reproductive and other health needs. Adolescent girls are immensely vul-
nerable, especially girls struggling to cope with poverty in ways that puts their
health at risk in a health system that cannot assure their protection nor meet their
health needs.
Contrast this situation with the essential elements of reproductive health agreed
at the International Conference on Population and Development (ICPD) held in
Cairo 1994 including: having a satisfying, safe sexual life; access to appropriate,
safe, effective, affordable and acceptable methods of family planning based on
informed choice and dignity; prevention, diagnosis and treatment of STIs/
STDs/HIV; services for safe pregnancy and childbirth; elimination of harmful prac-
tices (FGM, domestic violence, sexual trafficking); and emphasis on poverty
alleviation, girls’ education, women’s empowerment and reproductive rights. Also
Gender and Health Sector Reform 249
emphasized was the role of civil society and community in promoting sexual and
reproductive health.
HSR should be judged on whether it has focused or diverted health systems from
tackling this huge burden of reproductive ill health in the context of the HIV/AIDS
pandemic. In 2001, SSA had 3.5 million people newly infected with HIV/AIDS
(UNAIDS 2002). There, women have generally higher infection rates than men.
Women accounted for 59 per cent of the 2.23 million adults estimated to be living
with HIV in Tanzania in 2001, and young women aged 15–24 had twice the preva-
lence rate of young men (8.06 per cent compared to 3.55 per cent) (World Bank
2002). In Uganda, HIV infection rates were rising fastest among Ugandan women
aged 15–25 years (Ochieng 2003).
Key questions:
To what extent is the context of HSR one of unmanageable burdens of paid and
unpaid work for some categories of impoverished women?
To what extent is HSR exacerbating or relieving those burdens?
Vignette 4
A middle-aged woman runs a fruit-and-vegetable stall near her home. Her husband is a
clerk in a government office, and she helps in buying food items and paying for her
health expenses and sometimes the children’s. She also does all the caring for her three
young children, cooking, cleaning and fetching water from a nearby well. She has to run
her petty business close to home because there is no one else to care for the children,
since all adults are now trying to earn some money. She is usually exhausted by the end
of the day, while her husband hardly does any housework. She says however that she
must struggle on with the business because everything now requires money, including
health care and education. Overwork is taking its toll on her health.
African women bear disproportionately the burden of unpaid work, and have long
needed to earn cash for their own and their children’s needs. The economic crisis
and subsequent economic and social-sector reform programmes have escalated
costs of basic needs such as food, health care and children’s education, making the
combined pressures of paid and unpaid work more severe. Women survive by
working longer hours and years in low-return income-generating activities in the
informal sector (Whitehead 2001). This survival strategy in turn generates overwork,
affecting women’s health and increasing their health care needs – yet ironically it
may not pay enough to cater for increased health care, creating another vicious
circle of ill-health. Because of multiple working and caring roles at home and in
the workplace, women find it difficult to allocate time to seeking health care when
they are sick; long distance, long waits and staff absent from government facilities
make the vicious circle worse.
It is in the plight of elderly women that these pressures are exposed at their very
worst. AIDS death rates and the numbers of AIDS orphans in Africa are staggering.
By 2001, 83 per cent of world AIDS deaths had occurred in sub-Saharan Africa,
including 90 per cent of those dying before 15 years. In 2000 92 per cent of the
global total of 13.2 million AIDS orphans were in SSA (UNAIDS 2002; Africa Action
2000). As women and men aged 15–49 succumb to the epidemic, the elderly are
forced to take care of their terminally ill adult children and then to care for their
orphaned grandchildren. In Tanzania an assessment of the economic and social
impact of HIV/AIDS found 34 per cent of carers of orphans interviewed in schools
were grandparents, some taking care of as many as eight grandchildren (ESRF
2004). In South Africa and Uganda 40 per cent of orphaned children were found
to be living with their grandparents; in Zimbabwe the proportion was more than
half (HelpAge International and the International Alliance on HIV/AIDS 2003,
citing Ainsworth and Filmer 2002; Subbarao and Coury 2003)
The gendered division of labour continues to ascribe domestic work and caring
for children and the sick to women, in this case grandmothers, and women also live
longer than men. In Zimbabwe, two-thirds of 685 older care givers interviewed were
women. Often in ill-health and unable to take care of themselves adequately,
elderly women are a vulnerable group needing care and financial support, yet the
epidemic robs them of the support of their adult children. Doubly disadvantaged –
as female and elderly – they are among the poorest groups in society, now forced to
take on the additional burden of caring and raising their grandchildren within the
context of decreasing economic means and disintegrating family networks. Taking on
the additional responsibility of caring for their orphaned grandchildren means strug-
gling to meet food, clothing, health and education expenses for the grandchildren.
Elderly women’s poor health is exacerbated by physical and emotional stress
when taking care of their terminally ill adult children, and they risk infection
because they lack information on ways to protect themselves (HelpAge International
and The International Alliance on HIV/AIDS 2003). After their children die, the
worsening strain and hardship is illustrated by an interview with an elderly
woman in a South Western town of Tanzania (Vignette 5).
Vignette 5
This elderly lady is a widow who has seen her two sons die and is caring for the children
of both sons. Her only way of making an income is to buy vegetables, split them up into
small parcels and resell them retail. The money is not enough to feed the children
properly, nor herself. The children now go to school (since primary school fees were
abolished) but cannot come home for lunch because there is no food for them. She
herself is weak and sick from poor diet. There will be cumulative effects of poor diet for
the children.
HSR has in no way alleviated this cumulative burden of caring for the sick and
dying, raising children in old age and struggling to make ends meet to pay for
food, school books and clothes. Rather, it has added to needs for cash and unpaid
252 Maureen Mackintosh and Paula Tibandebage
work in the context of an eroding family support system and inadequate or total
lack of a formal social protection system for the elderly.
Conclusions
The social context of unequal gender relations adversely affects women’s health
and overall well-being, both directly and through its health-service effects
(Cuttingham and Mynttri 2002). The health sector reflects established gender
hierarchies within its staffing and priority-setting; when it responds poorly to
women’s needs it reinforces women’s relative social and economic disadvantage,
undermining women’s capacity to earn and care for themselves and others.
A discriminatory and inequitable health sector reinforces other aspects of women’s
relative lack of rights and dignity in societies that are highly unequal in terms of
both gender and social class.
Health sector reform, far from challenging these mutually reinforcing forms
of gender disadvantage, appears to have made some of them worse. This chap-
ter has argued that these failures appear rooted in invalid assumptions about
women’s responsibilities and capabilities to cope with rising work burdens. A con-
ceptualization of HSR as gendered institutional change would have highlighted
the dangers – for women and for population health – of exacerbating women’s
expenditure and work burdens to unmanageable proportions and of reinforcing
women’s relative disadvantage at the lower end of health systems’ work and care
hierarchies. A gender ‘lens’ on the proposed reforms would have questioned
the desirability, for example, of user fees and refocused attention on improving the
relationships and integration of services within primary care.
So what would it take to build instead a more gender-equitable set of health
sector institutions? Here are some propositions.
1. Design reform of health care finance in such a way that it does not increase the bur-
den on the incomes of poor women, nor the relative economic disadvantage of women
in general. This requires more information on women’s control over resources,
integrated into the design of health-financing mechanisms.
2. Design health sector reform to promote equality between male and female health
workers and to improve the working conditions of lower-level staff. This requires
close attention to the gender-differentiated and interacting needs of both staff
and patients.
3. Involve women equally with men, both as staff and citizens, in health sector reform
from the most local to the highest level. Women’s knowledge and perspectives are
essential to achieve policy design that is not discriminatory on gender lines,
basing itself rather on concepts of equal citizenship: draw on and support
women’s organizations for this purpose.
4. Do not undertake reforms that increase the work burden of women, especially poor
women; instead, design approaches that reduce those burdens. This may also require
men to carry a greater share of the burden of unpaid care.
Gender and Health Sector Reform 253
5. Build links between reproductive health care and provision for the other health needs
of women in ways that reduce the strain upon women of accessing care. This requires
investigating women’s points of view on how to integrate services.
6. Introduce reforms outside the health sector that strengthen women’s rights and status,
hence their ability to deal with the health sector on more equal terms with men. This
requires strengthening women’s rights and dignity through supportive laws,
policies and organizations.
Acknowledgement
We thank Shahra Razavi, Debbie Budlender, Susan Himmelweit, Dzodzi Tsikata and all
participants in the UNRISD workshop on Gender and Social Policy in Maarstrand, Sweden,
May 2005, for helpful comments on an earlier draft. The chapter is the sole responsibility of
the authors.
Notes
1 One scheme has subsequently achieved higher rates: P. Mujinja, personal communication.
2 Anecdotally, this seems to be the case in Accra, Ghana, where highly skilled nurses have
moved from the government hospitals to private specialist clinics.
3 Authors’ fieldwork.
References
Afford, C. (2003) Corrosive Reform: Failing Health Systems in Eastern Europe. Geneva: ILO.
Africa Action (2000) ‘Africa HIV/AIDS update’ at http://www.africaction.org/docs00/
hiv0007//htm, accessed on 20 April 2004.
Ainsworth, M. and D. Filmer (2002) Poverty, Children, Schooling and HIV/AIDS: A Targeting
Dilemma. Policy Working Paper No. 2885. Washington, DC: World Bank.
Alubo, O. (2001) ‘The Promise and Limits of Private Medicine: Health Policy Dilemmas in
Nigeria’. Health Policy and Planning 16(3): 313–21.
Atim, C. (1999) ‘Social Movements and Health Insurance: A Critical Evaluation of Voluntary,
Non-profit Insurance Schemes with Case Studies from Ghana and the Cameroons’. Social
Science and Medicine 48: 881–96.
Bassett, M, L. Bijlmakers and D. Sanders (1997) ‘Professionalism, Patient Satisfaction and
Quality of Health Care: Experience During Zimbabwe’s Structural Adjustment
Programme’. Social Science and Medicine 45(12): 1845–52.
Bennett, S., B. McPake and A. Mills (eds) (1997) Private Health Providers in Developing
Countries: Serving the Public Interest? London: Zed Books.
Berer, M. (2002) ‘Health Sector Reforms: Implications for Sexual and Reproductive Health
Services’. Reproductive Health Matters 10 (20): 6–15.
Booth, D, J. Milimo, G. Bond and S. Chimula (1995) ‘Coping with Cost Recovery’. Report to
SIDA, Development Studies Unit, Department of Social Anthropology, Stockholm
University.
Booth, D. and H. Lucas (2002) ‘Good Practice in the Development of PRSP Indicators and
Monitoring Systems’. Working Paper No. 172, London: Overseas Development Institute, July.
Bridge (2004) ‘Gender and Budgets.’ Gender and Development In Brief, Issue 12, www.ids.ac.uk/
bridge, accessed on 4 May 2004.
Buchan, J. (2000) ‘Health Sector Reform and Human Resources: Lessons from the United
Kingdom’. Health Policy and Planning 15 (30): 319–25.
254 Maureen Mackintosh and Paula Tibandebage
Rifkin, S. and G. Walt (1986) ‘Why Health Improves: Defining the Issues Concerning
“Comprehensive Primary Health Care” and “Selective Primary Health Care” ’ Social Science
and Medicine 23: 559–66.
Sauerborn, R., A. Nougtara and E. Latimer (1994) ‘The Elasticity of Demand for Health Care
in Burkina Faso: Differences across Age and Income Groups’. Health Policy and Planning
9(2): 185–92.
Sen, G., A. George and P. Ostlin (2002a) ‘Engendering Health Equity: A Review of Research
and Policy’. In G. Sen, A. George and P. Ostlin (eds) Engendering International Health.
Cambridge MA: MIT Press.
Sen, G., A. George and P. Ostlin (eds) (2002b) Engendering International Health. Cambridge
MA: MIT Press.
Sen, K. (2003) Restructuring Health Services: Changing Contexts and Comparative Perspectives.
London: Zed Books.
Standing, H. (1997) ‘Gender and Equity in Health Sector Reform Programmes: A Review’.
Health Policy and Planning 12 (1): 1–18.
Standing, H. (2002a) ‘Frameworks for Understanding Health Sector Reform’. In G. Sen,
A. George and P. Ostlin (eds) Engendering International Health. Cambridge MA: MIT Press.
Standing, H. (2002b) ‘An Overview of Changing Agendas in Health Sector Reform’.
Reproductive Health Matters 10 (20): 19–28.
Standing, H. (2002c) ‘Towards Equitable Financing Strategies for Reproductive Health’.
Sussex UK: Institute of Development Studies.
Subbarao, K. and D. Coury (2003) Orphans in Sub-Saharan Countries: A Framework for Public
Action. Washington DC: Africa Region (Human Development) and Human Development
Network (Social Protection), World Bank.
Tangcharoensathien, V. (2002) ‘ICPD Goals and Targets Worth Revisiting in the Context of
Health Sector Reforms’. Reproductive Health Matters 10 (20): 138–40.
Tanzania Gender Networking Programme (TGNP) (1999) Budgeting with a Gender Focus.
Dar es Salaam: TGNP.
Tibandebage, P. (2004) ‘Health Sector Reforms and Gender Implications: A Case Study of
Mutual Health Insurance Schemes in Tanzania’. Research paper for UNRISD, Geneva at
www.unrisd.org, accessed on November 2004.
Tibandebage, P. and M. Mackintosh (2002) ‘Institutional Reform and Regulatory
Relationships in a Liberalising Health Care System: a Tanzanian Case Study’. In J. Heyer,
R. Thorp and F. Stewart (eds) Group Behaviour and Development. Oxford: Oxford University
Press.
Tibandebage, P., H.H. Semboja, P. Mujinja and H. Ngonyani (2001) ‘Private Sector
Development: The Case of Private Health Facilities’. ESRF Discussion Paper, Dar es Salaam.
Towers, B., P. Smith and M. Mackintosh (2000) ‘Dimensions of Class in the Integration of
Health and Social Care’. Journal of Interprofessional Care 13/3: 219–28.
Twigg, J. (2000) Bathing, the Body and Community Care. London: Routledge.
UNAIDS (2002) Report on the Global HIV/AIDS Epidemic, UNAIDS, Geneva. July.
United Nations Population Fund (UNFPA) (2002) State of the World Population 2002: People,
Poverty and Possibilities. New York: United Nations Population Fund.
Walraven, G. (1996) ‘Willingness to Pay for District Hospital Services in Rural Tanzania’.
Health Policy and Planning 11(4): 428–37.
Van der Westhuizen, C. (2004) ‘Monitoring Provincial Budgets from a Pro-poor Perspective’.
Budget Information Service, Cape Town IDASA: February, at http://www.idasa.org.za/
accessed March 2005.
Whitehead, A. (2001) ‘Trade, Trade Liberalisation and Rural Poverty in Low Income Africa:
A Gendered Account’. Background Paper for the UNCTAD 2001 Least Developed Countries
Report.
Williams, F. (1989) Social Policy: A Critical Introduction. Cambridge: Polity.
World Bank (1993) World Development Report 1993: Investing in Health. New York: Oxford
University Press.
Gender and Health Sector Reform 257
World Bank (1997) World Development Report 1997: The State in a Changing World. New York:
Oxford University Press.
World Bank (2000) Advancing Gender Equity: World Bank Action since Beijing. Washington DC:
World Bank.
World Bank (2002) Tanzania at the Turn of the Century: Background Papers and Statistics.
A World Bank country study, Washington, DC: World Bank.
World Bank (2004). World Development Report 2004: Making Services Work for the Poor.
New York: Oxford University Press.
World Health Organization (WHO) (1978) Primary Health Care: Report of the International
Conference on Primary Health Care, Alma Ata 1987. Geneva: WHO.
World Health Organization (WHO) (2000a) World Health Report 2000. Health Systems:
Improving Performance. Geneva: WHO.
World Health Organization (WHO) (2000b) Abortion: A Listing of Available Data on the
Frequency and Mortality of Unsafe Abortion. WHO/RHT/MSM/97.4, Geneva: WHO.
World Health Organization (WHO) (2002) Impact of AIDS on Older People in Africa: Zimbabwe
Case Study. WHO/NMH//NPH/ALC/02.12, Geneva: WHO.
World Health Organization (WHO) (2003a) The World Health Report 2003: Shaping the Future.
Geneva: WHO.
World Health Organization (WHO) (2003b) Strengthening Nursing and Midwifery. Report by
the Secretariat, 56th World Health Assembly, Geneva.
12
Health Sector Reform in China:
Gender Equality and Social Justice
Jufen Wang1
Introduction
In any country the most important aspects of social policy are the social security
and social welfare systems. These systems are not only the results of social and eco-
nomic development but also, through the redistribution of national income, can
actively enable the harmonious, stable and healthy development of society.
Moreover, these systems can strengthen people’s capacity to identify and avoid
risks and accommodate unexpected events. The Constitution of China (PRC)
states that the ‘citizens of PRC are entitled to obtain physical assistance from state
and society in old age and ill health’.
There is insufficient theoretical research on social security in China. For a long
time, Chinese social security theory lacked a clear definition. Social security was
understood in terms of social insurance, which in turn was equated with labour
insurance. For example, social security rights were defined as
the rights enabling workers to access necessary help from the nation and
society when they lose their ability or opportunity to work temporarily or per-
manently through [factors such as] old age, disease, disability, unemployment
or other difficulties
Guo and Li 2001: 173
258
Health Sector Reform in China 259
direct relatives based in state-owned enterprises. Insurance covered all the workers’
own medical expenses and 50 per cent of their relatives’ medical expenses. The
following year, in August 1952, the government released a set of detailed regula-
tions for a second healthcare insurance system that became known as the Gongfei
system. The latter applied to workers in government institutions, the medical and
education sectors and other public-sector-oriented services. In the post-Mao
economic reform era serious drawbacks in the Gongfei and Laobao systems have
appeared and the main problems are discussed here with particular reference to
the structural inequalities. The issue of geography is also pertinent as is the process
of devolvement of responsibility by the central government.
One of the first casualties of economic transition has been the two main
providers of healthcare within the social security system of Maoist, socialist urban
China: the Gongfei and Laobao systems. Neither system is now supported by the
realities of the market-oriented economy. Previously, state ownership had domi-
nated the Chinese economic structure and the dominant state-owned economic
structure ensured that all of the urban population had access to health insurance.
At the end of the 1970s, China began its economic reforms by restructuring the
patterns of ownership. The new policy accepted state ownership as the main type
of ownership, while all other patterns of ownership were also to be allowed.
Private enterprises, joint ventures and other forms of companies and institutions
were henceforth permitted, ushering in a new era of economic flexibility. With the
introduction of this major structural reform and the accompanying social change,
the urban population found that they had other options to add to the once abbre-
viated list of employment with either stated-owned or collective companies.
However, while non-state-owned enterprises and companies (SOEs) proliferated
they were also excluded from the established system of healthcare, something the
population rapidly became aware of. The continued benefits of the Gongfei and
Laobao systems not only affected the way in which the newly created labour
market operated but also the nature of uptake of care. A survey conducted in 1998
by the State Medical Service determined that the highest hospitalization rate in
the urban population was amongst those covered by the Gongfei (89.97 per cent)
and Laobao (60.32 per cent) insurance systems. Those who have to pay at their
own expense exhibited the lowest rate of hospitalization, at only 29.49 per cent.
The main explanation for this differential was the nature of entitlement to healthcare
insurance (Zhang et al. 2003: 148).
Furthermore, the original medical insurance system as it operated through the
Gongfei and Laobao systems led to conflicts concerning the schemes’ financial
solvency in the post-1978 reform period. Neither system had established its own
funding source: the Gongfei scheme was funded by governmental appropriations
while the Laobao scheme was funded from the business welfare funds within each
individual enterprise. In the post-1978 market-oriented economic system, the
Laobao insurance system had to rely heavily on the new concept of business
profits. In contrast the Gongfei system was affected by the capacity of local gov-
ernments to raise their own revenues rather than rely on centrally distributed
funds. A further problem lay in the way in which the compensation mechanism
Health Sector Reform in China 261
Reform of the Chinese urban healthcare system started with the delivery of the
document on ‘Decisions on the establishment of basic healthcare insurance for
urban employees’ by the State Council in December 1996. There had been earlier
attempts to reform the health insurance system. In the early 1980s, reforms aimed
to reduce expenditure on medical services. These State Council-led experimental
strategies considered the establishment of special insurance funds for particular
diseases, cost-sharing strategies and careful accounting of the population covered
by insurance schemes (Liang et al. 2003: 40). A variety of healthcare models were
piloted at the local-area level, notably the Liangjiang Model named after the cities
of Zhengjiang, in Jiangsu province, and Qiujiang, in Jiangxi province. The three
central features of the Liangjiang model were: a public fund, personal accounts,
and a three-stage payment system. By 1996, the experiment had been expanded to
nearly forty cities.
At the end of 1998, the State Council promulgated a further document entitled
‘Urban employees’ basic healthcare insurance scheme’ which stated that basic
healthcare insurance schemes should be managed at the local authority level and
reflect current levels of economic development. The implication of this was that
areas with higher rates of economic growth would be expected to establish more
comprehensive forms of healthcare insurance for their population, while govern-
ments in low-growth areas would not be expected to achieve the same standards of
healthcare insurance, thereby resulting in a patchwork of differing health insurance
regimes across the country and with restrictions on the transfer of policies
between the cities. Despite this flexibility with respect to economic context, all
urban residents are required to participate in the scheme. The document also
requires that under this new national insurance system costs be jointly borne by
employers and employees (http://www.em800.com).
been granted ‘green cards’ for residence in Shanghai.2 Basic insurance is also
granted to retired workers, mainly from the city of Shanghai. It excludes those
retired workers who worked in other cities and returned to Shanghai upon retire-
ment. This exclusion results from the way in which healthcare insurance schemes
are tied to individual cities and reflects the local and non-transferable nature of
accumulation and social responsibility of devolved governments. Moreover, as
employees’ healthcare insurance is funded by both employers and employees
themselves, healthcare provision is situated within the workplace and is not trans-
ferable if or when an employee moves to another job. Participants in the basic
healthcare insurance programme can choose to access medical services anywhere
within a city, except private hospitals. Should workers move to another city, their
right to a health insurance scheme remains with their original workplace, and
they must start a new scheme with their new employers. Such health insurance
schemes exclude the unemployed who only have rights to unemployment
insurance. However, the self-employed can be included if they choose to pay the
insurance premiums themselves.
Five characteristics of the post-reform urban basic healthcare insurance, as it
operates in Shanghai, can be differentiated. Firstly, the new scheme not only
combines aspects of both the Gongfei and Laobao systems but has been extended
to employees of all sectors. Secondly, the establishment of public funds has been
an innovation. Public funds improve risk-management capacities and allow funds
to be used across broader sections of society, acting as a mechanism for the reallo-
cation of national income. Additionally, public funds support socioeconomic
development in a long-term and sustainable manner. Thirdly, the post-reform
scheme has emphasized the shared responsibilities of the state, employers and
employees. The public fund is financed both by employers and employees
through contributions from employers equivalent to 12 per cent of an employee’s
salary. Employees meanwhile contribute 2 per cent of their wages. The first 50 per
cent of the payment made by the employer goes into the public funds, henceforth
referred to as social funds, while the second 50 per cent goes to an employee’s
individual account along with the employee’s contribution. Fourthly, the new
basic healthcare scheme aims to be more financially efficient. It emphasizes indi-
vidual responsibility as well as the benefits of healthcare insurance. The amounts
transferred into a personal account from the social funds vary according to the age
and the level of individual contributions. Individuals do have a certain degree of
individual control over the funds. For example, since there is no time limitation
on the use of the funds in each personal account, should the fund not be used it
can be accessed during older age. It is hoped that this aspect of the policy will
motivate people to save into the funds as the monies are never lost to the individual
and can be transferred to relatives or children.
Fifthly, healthcare insurance is utilized via sequential stages of entitlement: per-
sonal account, individual payment and shared responsibility. With diagnosis of a
disease or injury, patients commence using the funds in their personal accounts to
pay for their healthcare expenses. After using up the money in this account, they
reach the second stage of individual payments whereby the costs must be borne by
264 Jufen Wang
themselves or their family. Once their payments for health care reach a sum
equivalent to 10 per cent of the average wage in Shanghai for the previous year,
the system moves into the third stage. At this stage, the costs are recouped from
individuals and the social fund. The level of individual financial responsibility is
correlated with both the grade of the hospital used by the patient and the age of
the patient. Individuals bear the largest portion of medical expenses when in
receipt of treatment from the most prestigious hospitals; individuals bear an inter-
mediate cost of medical expenses if they seek treatment in second-grade hospitals
while they pay the least amount if they use third-grade or community hospitals.
These sums are also calculated according to age, with older people paying a smaller
proportion of the incurred expenses compared to younger age groups. When
expenditure exceeds 400 per cent of the average wage of the preceding year public
insurance funds are closed to a particular patient.
scheme for urban workers fails to consider this gendered wage differential. As
contributions to the health insurance scheme are based on wage levels and as
men’s average wages are higher than women’s, men’s contribution to the social
funds will be greater than those of women. This means that women are unable to
build up the same levels of insurance premiums as men, which in turn means that
they will not be able to access the same levels of professional care as men.
In addition, the average age of retirement for women is five years earlier than
that of men. Thus, men will contribute more to insurance funds over their
working lifetime compared to women. Furthermore, as the regulations governing
medical insurance mean that the retired population can utilize the funds on retire-
ment age, women will access their funds sooner than men and over a longer
period as female life expectancy is higher. According to the Shanghai Statistical
Yearbook, the average life expectancy of women in Shanghai reached 81.8 years in
2003, while that of men reached 77.8 years (SSB 2004). If it is assumed that women
will retire five years earlier and die four years later than men, they will need to use,
potentially, some nine years more of funds compared to their male counterparts.
The insurance system therefore effectively discriminates against both men, who
make greater contributions, and women, who do not have access to the same level
of funds.
Age is also an important factor with regard to the basic health care insurance
program for urban workers in Shanghai. Age is important for a number of reasons.
Older people are more susceptible to ill-health. Different age cohorts experience
differing exposure to environments, diseases and events which may compromise
health and well-being. In general, the older population survive on their pensions
and rarely have access to surplus income which they can invest in insurance
schemes. Additionally, since the reforms’ principles mean that funds in individual
accounts accrue over a lifetime the older population have been excluded from the
long-term accumulation process. However, provision has been made to credit
the insurance premiums of the older population: retired workers are not required
to make contributions to health care insurance, rather their accounts are credited
using the social funds, and the payments they are required to make, for outpatient
care for example, have been reduced in comparison to the working population.
Considerable research underpins the identification of differences in life
expectancy for women and men. Women experience longer durations of ill-health
at older ages compared with men. In Spain, in the over-65 age group, women expe-
rience, on average, some 12 years of health problems at a level sufficient to require
bed-based rest; the average for men is nine years (United Nations 2000). Analyses
on the healthy-life expectancy of Shanghai citizens conducted by the Shanghai
Disease Prevention and Control Centre indicate that over 50 per cent of women’s
excess average life expectancy is lived in a non-healthy state (Zhou et al. 2001).
Chronic diseases result in medical expenses and high nursing fees, which can
easily become a heavy burden for older women. The current basic health care
insurance scheme stipulates that once the stage of shared responsibility has been
reached, pensioners must pay between 10 per cent and 20 per cent of their total
medical expenses, regardless of gender. Owing to women’s lower wages and
266 Jufen Wang
reduced pensions, the same share of total medical expenses imposes a heavier
burden on female pensioners.
The threshold set for individual payments is relatively higher for the female
population than the male. As outlined above, insurance scheme regulations sepa-
rate medical payments for health care into three sequential phases: a personal
account stage; an individual payment stage; and a stage in which costs are shared.
The third stage is only encountered once individual payments for health care
have exceeded a sum equivalent to 10 per cent of the average wage in the preced-
ing year. Since the amount of money in the individual payment stage is not
directly related to individual income, actual income will affect and restrict access
to medical services when needed. This is because 10 per cent of the average wage
of the whole city in the previous year is a lower sum for high-income earners
compared to those on low incomes. As women’s wages are generally lower than
men’s, this aspect of the insurance system has important implications for women
through restricting their access to medical services. Given that the average male
wage equals the average wage of the city as a whole, an individual payment rep-
resents one month’s salary for a man. For the average woman, whose wages tend
to be less than the city average, an individual payment represents a month and
one quarter’s salary. Consequently, with respect to proportional income, female
workers are required to pay more for the same medical services. The 10 per cent
threshold was originally intended to ensure that insurance participants did not
abuse the shared-responsibility aspect of the system by requiring patients to first
pay for a certain share of their own medical expenses. The unequal gender
impacts are an unfortunate outcome of this attempt to counteract abuse of the
system.
Finally, because of the exclusion of workers’ spouses and children the current
basic health care scheme for urban workers augments the burden on single moth-
ers. Though no document has ever announced the exclusion of spouses and chil-
dren to medical benefits (there were different provisions in the Gongfei and
Laobao systems), the majority of units have abandoned this original broader
aspect to medical health care. Some units still follow the previous Gongfei and
Laobao regulations on the inclusion of spouses. Since children of divorced families
tend to live with their mothers, any medical expenses will add to their mothers’
financial burdens. This is not the only kind of bias that adversely affects women in
a strictly individualistic system such as this; the other exclusion is of women who
are full-time unpaid carers (taking care of young children, the elderly and so on)
who will not be able to receive any coverage as ‘dependants’ (if their husbands are
covered).
Table 12.1 Shanghai social healthcare scheme, contributions by percentage and age
In order to explore this further it is assumed that consumption and needs can be
divided into three categories in a model based on Maslow’s hierarchy of needs
(Maslow 1970). The first of these categories of demand is basic subsistence (food,
shelter and transportation), the second, health (medical care), and the last
category is leisure (culture, recreation and travel). These three categories differ in
income elasticity, the first category having the least income elasticity, the third the
greatest. Given gender income inequity (in both wages and pensions) women tend
to sacrifice leisure for the sake of health, and health is sacrificed for basic subsistence.
Research conducted by the author has revealed that, having used up the funds
in their individual accounts, some elderly women did not access necessary health
care until monies were credited to their accounts the following financial year;
some women postponed necessary treatments and operations through poverty.
Although this situation may also apply to men, it occurs relatively more
frequently in the female population especially in the lowest-income groups,
clearly revealing the problematic aspects of this medical insurance scheme.
Table 12.2 shows the relationship between age, hospital capacity and the percent-
age of costs paid by individuals and from social funds. Generally, the younger an
individual is and the further into the future their predicted retirement the lower
the percentage of medical costs paid for through social funds and the greater the
individual contribution.
Current practices embedded within the provision of basic medical insurance for
urban workers overlook physical differences, different morbidity patterns and con-
sequently different medical demands between genders. These patterns are particu-
larly evident in the older population. In 2003, elderly citizens aged over 65 years
accounted for 14.8 per cent of the total population of Shanghai (CPN 2005).
A recent survey conducted by Ministry of Public Health indicates that 71.3 per
Table 12.2 Standard scale for the shared-responsibility stage of urban insurance, Shanghai
cent of elderly citizens suffer from chronic diseases, the percentage being more
than twice that of the general adult population. Additionally, 42 per cent of
elderly citizens suffer from more than two chronic diseases (Ou 2002). In an
ageing society, older female citizens are subject to more health problems than
older male citizens. ‘Despite longer average life spans, the health circumstances
and quality of life of female elderly citizens are worse than those of the male’ (Zeng
and Gu 2002: 59–66).
The basic health care insurance system currently offers the greatest coverage and
best policies for urban employees in Shanghai. It also covers most of the demands
on the medical services. Compared with other cities, Shanghai’s citizens enjoy
one of the best insurance schemes in the country. In comparison, elsewhere
policyholders only receive help with medical expenses once hospitalized.
Maternity insurance
Under the Gongfei and Laobao programmes, a woman’s entire medical expenses
during pregnancy and while on maternity leave were covered, with salaries
provided by companies and businesses. With health care insurance reform, the
restructuring of ownership of businesses and industry and especially the introduc-
tion of profit-focused, market-oriented managerial methods, companies started to
consider expenses related to child bearing as an unwelcome and additional
burden.
To provide salaries and medical expenses during pregnancy and the post-natal
period is to both acknowledge women’s particular social contributions and
promote women’s equal employment rights. With the support of the All China
270 Jufen Wang
Women’s Federation (ACWF) at national and local level, the State Council issued
documents establishing maternity insurance. This was part of a promise made at
the Fourth World Conference on Women in Beijing in 1995. As a part of the
reform of the whole of the Chinese urban medical insurance programme,
maternity insurance will play an important role in enhancing women’s health.
Maternity insurance guarantees the following two basic rights. Firstly, ante-natal
care and expenses related to childbirth is covered. Secondly, it guarantees the con-
tinuation of salaries during maternity leave. It could be argued that this insurance
represents progress in the protection of urban women’s rights to social security.
However, although restrictions based on household registration do not exist
theoretically – meaning that all working women of childbearing age can participate –
some migrant women are excluded because they are not covered by the basic
urban health care insurance system.
The Shanghai municipal government initiated maternity insurance in October
2001. This scheme aims to improve women’s employment circumstances and
secure their entitlement to maternity benefits. Prior to the implementation of this
insurance, employers paid medical expenses during pregnancy and the salaries of
women on maternity leave. Now, these expenses are funded through the new
maternity insurance system. However, monies are not specifically collected for this
system; rather they are transferred from the basic urban health insurance schemes
and from pension funds. The advantage is that there is no need for society to raise
money separately. However, coverage is limited to those female workers already
enrolled under the basic health care insurance system. Therefore, a large number
of migrant women, who work outside the formal employment sector, are excluded
from this insurance scheme although they too are working in Shanghai. Under the
terms of this new plan, eligible women are entitled to four and a half months’
maternity leave while receiving 100 per cent of their wages and near-free medical
care during pregnancy and birth if their child is delivered in mid-grade hospitals.
The levels of insurance guarantees for participants are higher in Shanghai than
elsewhere in terms of maternity wage levels, length of maternity leave, as well as
coverage of care.
In China as a whole, however, maternity insurance has developed comparatively
slowly. By the end of 2003, some 36,550,000 women had participated in maternity
insurance schemes throughout the country (Xinhua News Net 2005). Some 16
provincial capitals have formulated regulations on maternity insurance.3
Moreover, maternity insurance in many places only targets registered permanent
residents, a residual from the former planned economy, and thus excludes migrant
women.
Shanghai. The new policy favours those who are low-income earners and who
have been diagnosed with chronic diseases because the proportion of the medical
expenses borne by the individual is closely related to his or her wage or pension.
The supplementary policy was subsequently revised in April 2004 and its main
items are outlined in Table 12.3 below.
Table 12.3 demonstrates the way in which women, who tend to be low-income
earners, are more likely to be able to reduce the costs of their personal medical-
care expenses. People who for example earn a sum equivalent to, or less than, the
average minimum wage in Shanghai in the year preceding the onset of medical
expenses receive a 90 per cent reduction in their medical costs once those costs
have surpassed a figure equivalent to 30 per cent of the average minimum wage in
Shanghai in the preceding year.
The supplementary aspects of the basic health care insurance system have had
marked benefits for women as they challenge the gender inequalities of the previous
system. Thus, to some extent these subsidiary aspects have meant that gender
equality has been reached in Shanghai’s urban health care scheme.
Examination of the current basic medical insurance for urban workers shows that
it fails to consider gender throughout its structure including in the scope of insur-
ance and in the share of payment. Based on age alone, the scheme appears to be
gender neutral but, in effect, brings out inequities in rights and interests by ignor-
ing the differences between genders in social status, employment, income, age at
retirement and health status. Fortunately, the supplementary policy demonstrates
the gains which can be made in the implementation of national gender equality.
Table 12.3 Main aspects of the 2004 supplementary policy on reducing medical care
expenses, Shanghai
Expanding participation
The criteria for judging a social insurance scheme should highlight coverage and
benefit levels, although national and local levels of economic development, and
the redistribution of socioeconomic resources must also be taken into considera-
tion. Currently, basic health care insurance for urban workers, the principal part of
Chinese urban health care security, is based on employment and permanent resi-
dence permits and thus results in the exclusion of the lowest stratum of society
which encompasses the unemployed, informal workers and migrant labourers.
These aspects of the social security entitlement system not only hamper coverage,
but also compromise the size of available funds. For example, by the end of June
2003, the basic health care scheme covered 42 per cent of the total number of
urban workers in China (Chou 2005). Currently, this means that some 109.02 million
members of the urban population are included within the scheme, of which 79.75
million are workers and 29.27 million are pensioners (PRC Year Book 2005: 414).
Pensioners account for some 26.8 per cent of the population entitled to the
health care insurance system. Pensioners pay no contributions into the scheme
but are heavy users, which places a drain on insurance funds. Statistics from
the Medical Insurance Clearing Centre of Zhenjiang City reveal that in 2001 the
average sum paid out for medical expenses per worker was 673.05 RMB, while
that per pensioner was 1651.66 RMB, some 2.45 times higher (Li 2003: 14).
‘In 2001, retirees accounted for 27.40 per cent of the total participants [of the
health insurance scheme] but consumed 44.84 per cent of the total medical
expenses and 48.42 per cent of the total medical funds’ (Jiang et al. 2004: 18). Yet,
health care insurance policies that have favoured retirees still fail to meet the large
medical demands of the older population. If current practices of allowing pen-
sioners to consume the majority of social funds continue, not only will younger
participants’ interests be violated, but also the operation of the whole health care
insurance system will be compromised.
To solve this problem effectively, two measures have been identified. First, the
authorities should formulate a special programme offering compensatory benefits
to those pensioners who have made due contributions during the planned-
economy era, or establish a subsidiary health care insurance scheme for pensioners,
to alleviate the pressure on the basic health care insurance (which will require sig-
nificant subsidies from the state, both local and central). Secondly, manipulation
of the age structure of participants should take place by allowing more young
migrant labourers working in cities to participate in basic health care insurance.
health care insurance rights were largely secure. However, the rate of women’s
participation in the formal workforce has decreased sharply. A 2003 survey in
Shanghai indicated that the employment rate of women in the 16–54 age group had
declined to 52.6 per cent, and their unemployment rate had reached 22.5 per cent
(Xu and Yuan 2004). Urban basic health care insurance schemes based on employ-
ment therefore presented an institutional barrier to women’s health care insurance
rights. At present, differences between genders in medical insurance rights have
not became obvious because contracts between laid-off women, early retirees and
their former work units require those units to continue to pay contributions
towards women’s health care insurance and pension schemes, thus fundamentally
reserving their rights to basic health care. In the near future when the levels of
unemployment among young women, the increase in the proportion of female
informal workers and the trend towards short-term employment become significant,
the gender differences in urban health care insurance rights will become a
more visible form of social injustice in China. Apart from taking measures to
secure women’s equal employment rights, the government should also further
improve the regulation of basic health insurance, through the inclusion of part-time
workers and informal workers in the urban basic health care insurance scheme.
For a significant period, the health care security system in China was binary in
nature, being divided between urban and rural locations with city residents
enjoying higher standards of living compared with rural residents. Before health
care insurance reform, the high employment rate in urban areas guaranteed that
nearly every registered permanent urban resident could access health care under
the Gongfei and Laobao systems while the rural population relied on a less
comprehensive cooperative medical service. Entitlement to this latter system
was based on access to agricultural land. In some areas the cooperative medical
services did not survive the agricultural reforms, which led to the financial collapse
of rural health care institutions. This has severely reduced access to any form of
health care in rural areas.
According to incomplete estimations, there are some 120 million migrants from
rural areas working in China’s large cities. Part of this population has been living
and working in the cities for up to ten years and can be considered at the least semi-
permanent residents. This shift of rural labourers to cities, as we have seen, has
brought with it medical security problems. Despite reforms, entitlement to the
basic health care insurance in urban areas still rests on registered permanent
residence or ‘green cards’ which grants access to urban privileges for a select
population. Chinese health care insurance reform is still bound by traditional
thinking, which formulates social or public policies based on the inflexible
residence-registration system. As a result the differences in urban and rural social-
security systems intensify and social equity problems crystallize. The form that any
social security system for rural migrants may take has aroused heated discussion. In
Shanghai, while a comprehensive insurance for rural migrants has been estab-
lished, only a minority of rural migrant workers can participate. Due to the high
mobility of rural migrant labourers and the desire to reduce costs, employers have
not made contributions. Additionally, a large number of rural migrant labourers
Health Sector Reform in China 275
work within the informal employment environment and are excluded from this
insurance system.
More attention should be paid to the increasing number of women migrating
into cities who are excluded from medical insurance systems. In 1988, the gender
ratio of the migrant population in Shanghai was 241 men to 100 women. By 2003
it had fallen to 134 men for every 100 women. More female migrants are excluded
from the medical insurance system because more undertake independent, informal
work for which there are no insurance provisions. The most significant health
problems experienced by female migrant workers are those related to the repro-
ductive health system and are also an important barrier to the further improve-
ment of maternal health standards in Shanghai. Migrant women are less likely to
participate in ante-natal care (Zhan et al. 1999) and experience problematic
outcomes. In Beijing, the perinatal death rate for migrant pregnancies is 5.7 times
higher than that of the local population (Huang 1999). Different levels and
patterns of health care for women with different residences have affected the
health of mothers and foetuses. Some pregnant migrant women, excluded from
urban health care, return home to deliver their babies and to search for under-
qualified doctors whose fees are cheaper. Thus, they are further marginalized with
respect to their maternal health compared with women holding Shanghai
residence permits. The main solution to the problem of the provision of health
care insurance for migrant women would be to break the link between registered
residence and access to health care and admit migrant women workers into the
basic health insurance system as well as maternity insurance. An alternative would
be to establish migrant childbearing insurance via an individual payment system
(but without government subsidies this will not be viable given migrants’ irregular
incomes and the unwillingness of employers especially in the informal economy
to make contributions). Relevant government departments should also consider
subsidizing the migrant childbearing insurance system. The migrant women’s
maternity insurance could be run by insurance companies under government
supervision. This would also aid the government’s aim to establish a multi-level
health insurance programme for the population of China.
Conclusion
China is a developing country with a huge population and had experienced nearly
twenty years of intense reform and change by the end of 1990s. By this stage there
were serious fiscal concerns with regard to the provision of social security. This
chapter has outlined the development and progress of health care reform from
1998 to date. Bearing in mind the nature of the needed change and the size of
the population, the new policy has been a marked success.
Yet, detailed examination reveals that the current basic medical insurance for
urban workers failed to consider gender at the design stage, whether in terms of
the scope of basic insurance or in the share of payment. Based on age, the scheme
appeared to be neutral and equal but in practice exacerbated inequalities. Shanghai’s
health insurance scheme overlooked initial differences between genders in social
276 Jufen Wang
status, employment, income, retirement age, physical features and health status at
the inception of the scheme, resulting in gender inequity in rights and interests.
Fortunately, the supplementary policy has brought about reductions in medical
expenses based on individual wage and pension levels and presents a national
example in the implementation of China’s national policy of gender equality.
Results have shown that there is still room for improvement as the current scheme
excludes informal workers, migrant workers and worker’s spouses. However, it repre-
sents a significant move forward especially when considering that the supplementary
policy was not designed specifically to resolve gender inequalities. It is also true that
the solution to certain social issues also requires economic development and time.
The lessons we can learn from China’s urban health care reform are: firstly, the
realization that mainstreaming gender issues is still a challenging process for all of
society; and secondly that the design of any social policy should include gender
considerations from the start and any potential negative impacts on vulnerable
populations should be avoided.
Notes
1 This chapter has benefited from Caroline Hoy’s excellent editing.
2 ‘Green cards’ are normally granted to individuals who originate from outside the city of
Shanghai but who have obtained a post in the formal employment sector in the city, or
contributed to the development of Shanghai’s economy in some way.
3 Namely, Chengdu, Nanjing, Changchun, Ji’nan, Hefei, Yinchuan, Guangzhou, Hangzhou,
Shijiazhuang, Taiyuan, Xi’ning, Kunming, Nanning, and Wulumuqi.
References
Chen, Yani and Yi Chou (2002) ‘Record on Shanghai Health Care Reform’, at
www.people.com.cn, accessed 10 August 2002.
China Statistics Bureau (2004) People’s Republic of China Year Book. Beijing: People’s Republic
of China Year Book Press: 414.
Chou, Yulin (2005) ‘The Circumstances and Improvement of the Health Care Insurance
System in China’. Monthly Journal of Social Security 2, 39. Information Center for Social
Science, People’s University, Beijing.
China Population Net (CPN) (2005) at www.chinapop.gov.cn, accessed 13 January 2005.
Department of Propagation of Shanghai Municipality and Association of Social Science
(DoP) (1998) The Development Report of Shanghai Philosophy and Social Science Research.
Shanghai: Shanghai People’s Press: 401.
Ge, Shouchang (1999) Social Security Economics. Shanghai: Shanghai University of Economics
and Finance Press.
Guo, Weihua and Fucheng Li (2001) The Legal Protection of Workers’ Rights. Beijing Xiyuan
Publishing House.
Hu, Shanlian (2002) ‘Experiences and Issues in Shanghai Health Reform’. Journal of China
Health Resources 5, 5: 202.
Huang, Xiaoling (1999) ‘Study on Perinatal Infant Death in the Migrant Population’. Journal
of Maternal and Child Hygiene 14, 4: 242.
Jiang, Licheng, Linping Xiong, Yiping Li and Xiuqiang Ma (2004) ‘An Analysis of the Medical
Expenses of Insured Government Employees and Workers in Zhenjiang City’. Journal of
Chinese Health Economics 23, 6: 18.
Health Sector Reform in China 277
Li, Qiufang, Simei Wang and Xiuhua Jiang (2003) ‘The Circumstances of and Measures for
Chinese Women’s Employment’. China Women’s Express Vol. No. 6.
Li, Yiping (2003) ‘A Study of the Aging of Health Care Insurance Participants’. Journal of
Chinese Health Economics 22, 8: 14.
Liang, Hong, Yuan Cheng, Ying Zhou and Ying Zhu (2003) ‘Inequality Problems with
Medical Services in Cities from the point of View of the Health Care System’. Journal of
Chinese Population Science 1: 40.
Maslow, Abraham (1970) Motivation and Personality. Second edition, New York and London:
Harper & Row.
Ou, Shuisheng (2002) ‘Concerns about the Health Care Security of the Urban Poor’. Journal
of Soft Science of Health 16, 4: 11.
Shanghai Health Care Insurance Bureau (2000) Hand Book of Shanghai Basic Health Care
Insurance for Urban Workers. Shanghai: Shanghai Health Care Insurance Bureau 19.
Shanghai Statistics Bureau (SSB) (2004) Shanghai Statistical Yearbook 3.3 at www.
stats-sh.gov.cn, accessed 5 August 2005.
State Council document on ‘Decisions on Setting up Urban Basic Health Care Insurance for
Urban Employees’, at http://www.em800.com, accessed 14 October 2001.
United Nations (2000) The World’s Women, Trend and Statistics. New York: United Nations: 75.
Xinhua News Net (2005) at www.xinhuanet.com/, accessed on 15 February 2005.
Xu, Ke and Jing Yuan (2004) ‘Investigation into Shanghai Women’s Employment’. Shanghai
Women 10, October: 8–13.
Ying, Xiaohua, Wen Chen, Lijun Huang, Qin Jiang and ShanLian Hu (2004) ‘Equity of
Health Care and Health Finance’. Journal of Chinese Health Economics 23,1: 52.
Zeng, Yi and Da’nan Gu (2002) ‘Dynamics in International Research on the Quality of Life of
the Elderly Population’. Journal of Chinese Population Science 5: 60–1.
Zhan, Shaokang, Lingyu Xie, Changqin Liu and Yangguang Wang (1999) Research on the
Maternal and Prenatal Care of Migrant Women in Xuhui District, Shanghai’. Journal of
HaiXia Health Protection 5: 1–2.
Zhang, Taihai, Maojin Cheng, Hong Wang, Yun Peng and Fang Wang (2003) ‘Thoughts
and Measures for Health Care Insurance of the Urban Poor’. Journal of Soft Science of
Health 17, 1: 7.
Zheng, Gongcheng (2002) Evaluation of China’s Social Security System. Beijing: China People’s
University Press: 120–2.
Zhou, Feng, Guixiang Song, Songbai Lin, Yaling Sun and ZhiYing Wang (2001) ‘Analysis on
Shanghai Population’s Health Expectancy’. Journal of China Health Statistics 18, 2: 86.
13
Secondary Education in the Indian
State of Uttar Pradesh: Gender
Dimensions of State Policy
and Practice
Jyotsna Jha and Ramya Subrahmanian
Introduction
Linking gender and education to the social policy agenda is not straightforward.
Many questions arise, most of which hinge on the kinds of impact education is
expected to have on human behaviour, capacities and skills, and on gender iden-
tities and relations. In that sense, education within the wider social policy debate
is linked to questions of society and citizenship, understandings about modernity
and national identity, and the recasting of notions of masculinity and femininity
at different historical moments of time. The content of education, the impact of
policy choices adopted in the process of meeting universal-education agendas,
and the ways in which equity and social-inclusion concerns are addressed through
public institutions, are among the several issues that arise in relation to the
reframing of social policy from a gender perspective.
A central proposition of this chapter is that the focus on minimum ‘thresholds’
for public investment, in turn derived from the analysis of rates of return to
education, has contributed to the neglect of female post-primary education.
Influenced by Human Capital theory (HCT), ‘gender’ and female education have
been central framing discourses of education policy, resulting in substantial policy
rhetoric and concern about women’s and girls’ education as a lever of develop-
ment and progress. In India, acceptance of this global rhetoric has been mediated
by particular policy choices, which have resulted in the neglect of the secondary
sector, the rise of for-profit schooling at all levels of education, and a fragmented
formal elementary-education system, with particular implications for achieving
gender parity and equality. This has resulted in a range of issues relating to female
well-being being erased from the policy map. Girls disappear off the formal educa-
tion policy agenda past the age of 14, at a crucial age when aspirations can be
channelled into opportunities. In this chapter, we focus on secondary schooling,
which we believe best serves the interests of girls, especially if supported by poli-
cies that expand its availability, address socio-cultural constraints that exclude
278
Secondary Education in the Indian State of Uttar Pradesh 279
girls (both within society and within the school), and keep its costs low. We argue
that the lack of policy focus on secondary schooling for girls is linked to the
curious contradictions between policy rhetoric, on the one hand, and policy
prescriptions on the other, where development visions are not matched by
policy decision-making processes that can realize these visions.
Our choice of Uttar Pradesh (UP) as a case study is guided by three factors. One,
UP is the most populous state in the country, and also considered one of the most
socially ‘backward’ in terms of development indicators. Two, there is recent and
high-quality research material on education in the state, on the political economy
of education (Kingdon and Muzammil 2003), on sociology, gender and education
(Jeffery et al. 2004), and more generally on education provision (Dreze and Gazdar
1996). Three, there has been a concerted attempt in the state to focus on
reproductive health and fertility decline.1 There are also innovative programmes
for empowering women through education that operate outside the formal
schooling system. Given the overall correlations drawn between fertility decline
and female education, exploring secondary schooling for girls against the backdrop
of this orientation towards fertility decline, on the one hand, and empowerment
on the other, provides an opportunity to delve deeper into the gender politics of
investment in females.
In this chapter, we report primarily on material garnered through secondary
research, as well as field work undertaken to explore the status of secondary
schooling in UP, particularly in relation to patterns of financial investment and
provisioning of single-sex schools for girls. Our empirical research uncovered a
vital consideration for gender-policy analysis in education – the ways in which
wider discourses get played out through particular policy processes at state level,
which in turn are dictated by the compulsions of democratic politics as they are
played out in India. In that sense, UP offers both an interesting case study but also
an impossibly complex one, given the multiple political actors who inform the
education policy agenda (see, for instance, Kingdon and Muzammil’s (2003)
fascinating account of teacher politics in Uttar Pradesh). Gender seems almost
irrelevant in this tableau, although its very irrelevance is in itself a revealing
insight into the ways in which rhetoric on gender equality that aims to please
diverse publics, from vote-banks (perhaps) to union government and donor
agencies, can mask actual practice. In that sense, the account that follows
illustrates the ‘reality’ of policy-making in a developing-country context, not least
the opaqueness of the concept of policy, and the difficulty of tracking what is or is
not evidence of ‘policy’ in an intensely political policy-making environment.
The political economy of market and commerce, on the one hand, and complex
socio-cultural norms, on the other, find their own ways to influence the realm of
policy and education system, often leading to inconsistencies between policy
intents and practices adopted at ground level. An analysis of policy remains
incomplete unless traced to its translation into implementation and practices at all
levels. Policy practices need to be traced through penetrations and informal
interactions at various levels, making it difficult to collect evidence following
‘scientific’ methods. Much of the empirical information presented in this chapter
280 Jyotsna Jha and Ramya Subrahmanian
Notes: GPI (Gender Parity Index) ⫽ GER for girls divided by GER for Boys
These are actual data. Data available for later periods are provisional.
Source: A Handbook of School Education and Allied Statistics (1996), Ministry of Human Resource
Development (MHRD), Government of India (GOI), 1996 for 1980/81 and 1990/91 figures; Education in
India (1994/95, 1995/96 and 1996/97), MHRD, GOI, 2001, 2002, 2003 for 1994/95, 1995/96 and 1996/97
Drop out rates (Grades 56.51 52.93 62.11 40.67 39.71 41.90
I to V) 2000/01
Drop out rates (Grades 61.02 56.26 68.54 53.67 50.33 57.95
I to VIII) 2000/01
Drop-out rates 62.11 56.22 73.17 68.58 66.41 71.51
(Grades I–X), 2000/01
Repeaters (%) in
Grades IX and X, 13.84 14.88 10.68 6.49 7.02 5.57
1996/97
Source: Annual Report, MHRD 2001/02 for drop-out rates and education in India (1996/97) MHRD;
Government of India 2003 for proportion of repeaters.
trend changes when it comes to the transition rate between secondary and senior
secondary (Table 13.3). A significantly smaller proportion of boys studying in
class X continue with their senior secondary schooling as compared to the proportion
of girls studying at the same level. Pass percentages are also higher for girls at both
grades X and XII, explaining to some extent the higher transition rate at that level
(Table 13.4). A high proportion of boys join the labour force at this age, which also
is partially responsible for their discontinuation of schooling after grade X.
Relatively low transition rates from primary to upper primary, and upper
primary to secondary, for girls indicate that the secondary-schooling participation
patterns cannot be understood in complete isolation. This is especially true for the
fact that the largest drop-out incidence takes place within primary level and only
38 per cent of girls enrolled in grade I reach grade V (Table 13.2). As a corollary to
282 Jyotsna Jha and Ramya Subrahmanian
Table 13.3 Crude transition rates for different levels of school education (1998/99 to
1999/2000)
Note: This has been calculated by taking enrolment in the entry year as a percentage of enrolment in the
terminal year of the previous level during the previous year. This does not take contributions of migration
from other states, repetition, etc. into consideration. Also, the fact that a large percentage of children
never reach Grade V, the terminal grade for primary education, is not reflected in this table.
Source: Selected Educational Statistics, MHRD, Government of India.
this, the GERs at upper primary level are significantly lower than that at the
primary level (Table 13.5). A combination of high drop out within primary stage,
better academic performances at secondary level and a high transition rate from
secondary to senior secondary indicates that though a relatively small proportion
of girls continue with their post-primary schooling, those who do continue
perform better in examinations and a greater proportion is likely to complete the
senior secondary level. However, higher transition rates for post-secondary stage
might be indicative of gender differentiation taking a different shape where girls
do not have equal opportunities to join the labour market.
Secondary Education in the Indian State of Uttar Pradesh 283
Table 13.5 Gross Enrolment Ratio (GER) at different stages of school education
Total Boys Girls GPI Total Boys Girls GPI Total Boys Girls GPI
1980/81 62.50 81.80 40.30 0.49 32.10 56.30 18.00 0.32 32.10 39.40 11.30 0.28
1990/91 71.06 89.13 50.98 0.57 27.33 63.13 25.62 0.40 27.33 34.79 19.69 0.56
1994/95 61.20 72.03 48.86 0.68 25.51 55.52 31.28 0.56 25.51 33.31 15.64 0.47
1995/96 61.19 71.95 48.92 0.68 25.45 53.24 29.96 0.56 25.45 33.36 15.48 0.46
1996/97 61.13 71.71 49.02 0.68 25.12 51.24 28.91 0.56 25.12 33.04 15.17 0.46
GPI (Gender Parity Index) ⫽ GER for girls divided by GER for Boys
Source: A Handbook of School Education and Allied Statistics (1996), MHRD, Government of India, 1996 for
1980/81 and 1990/91 figures; Education in India (1994/95, 1995/96 and 1996/97), MHRD, Government of
India, 2001, 2002, 2003 for 1994/95, 1995/96 and 1996/97.
Elementary Secondary
Year education education Total educationa
a
including higher and other education
b
Budget estimates
Notes:
Figures in parenthesis show the percentage share of respective level in total
education expenditure.
These figures reveal expenditure on Revenue Account of the Education
Department. In addition to these, a small amount of expenditure is incurred
on Capital Account of the department, by other departments and ministries.
However, together they account for less than one to two per cent of the total
expenditure on revenue account of the department.
Source: Budgeted Resources for Education (1951/52 to 1993/94), Department of
Education, MHRD, Government of India, New Delhi, 1995 and Analysis of
Budgeted Expenditure on Education, Government of India (different years).
covered primary grades (I–V) only. Both these projects in UP were funded by the
World Bank, the former being the first basic education project being financed by
them in India. Although the contribution of these externally-aided projects to total
expenses of the sector has not been considerable, they formed a significant propor-
tion of the new investments. The total amount coming from external sources
amounted to nearly Rs.20,000.00 million during 1993–2002,2 which was about
82 per cent of total new investments on elementary education in UP during the
period. The project design for both UPBEP and DPEP demanded that the state
government provide about 15 per cent of the total project cost. Taking both exter-
nal and internal allocations together these two projects accounted for more than
96 per cent of total plan expenditure or the new investment during the period. This
means that the high allocation for new investment during this period was mainly
due to the presence of externally aided projects and only about 4 per cent of the
total allocations was spent on areas/items not covered by these two projects.
Secondary Education in the Indian State of Uttar Pradesh 285
a
Budget Estimates
Source: Based on figures provided in Budgeted Resources for Education (1951/52 to
1993/94), Department of Education, MHRD, Government of India, New Delhi,
1995 and Analysis of Budgeted Expenditure on Education, Government of India (different
years).
Table 13.8 Per student expenditure at elementary and secondary education levels,
Uttar Pradesh, India
1980/81 1 714.47 12 762 037 134.34 419.81 1 097.20 18 53 459 591.97 1 849.90
1990/91 12 109.42 16 466 813 735.24 1 007.17 6 179.82 29 55 073 2 081.11 2 850.83
1993/94 13 335.88 17 593 289 758.00 758.00 8 034.28 30 15 817 2 664.04 2 664.04
1994/95 16 101.16 17 835 564 902.75 828.21 9 505.75 30 45 167 3 071.11 2 817.53
1995/96 18 621.84 18 055 640 1 031.35 866.68 10 881.83 31 54 719 3 449.38 2 898.63
1996/97 21 306.51 16 602 422 1 283.33 1 002.60 11 976.85 32 05 232 3 736.65 2 919.25
1997/98 22 521.59 18 481 210 1 218.62 889.50 13 498.13 32 39 513 4 166.71 3 041.39
1998/99 33 168.06 18 680 569 1 775.53 1 175.84 18 145.52 32 74 449 5 541.54 3 669.89
1999/00 26 892.21 19 019 535 1 413.92 912.20 17 546.05 33 34 369 5 262.18 3 394.95
2000/01 34 446.74 18 049 991 1 908.40 1 142.75 18 041.06 33 18 222 5 436.97 3 255.67
a
At 1993/94 prices (taken from Economic Survey, Government of India)
Note: While total expenditure is given in terms of million rupees, per student expenditure is in terms of
rupees.
Source: Table VII for total expenditure, Shikhsha Ki Pragiti (different years), Government of Uttar Pradesh
for enrolment except for 1993/94 and 1994/95 and Education in India, MHRD, Government of India for
1993/94 and 1994/95.
school was brought under the grants-in-aid scheme. The growth of private
unaided secondary schools has been spectacular during the 1980s and 1990s. The
number increased from a modest 845 in 1994/95 to 6,541 in 2001/02, registering
a nearly eight-fold increase. The proportion of private schools, especially unaided
ones, has gone up to more than half of total secondary schools in UP after the
bifurcation of the state, reflecting the fact that the parts of UP that constituted the
newly carved state, Uttaranchal, had a higher number of government and aided
schools (Table 13.9). Near-stagnation in the number of government and aided
schools which charge only nominal fees and expansion only through fee-charging
private unaided schools reflects the state’s policy of promoting commercialization.
This is despite the fact that there exists apparently a greater demand for admis-
sions in government school as compared to private, including aided, schools
because of a common perception that these are better-resourced and higher-
quality.10 The results of the terminal stages at grade X and XII of all schools affiliated
to the UP board indicate that this perception is not ill-founded (Table 13.10).
The increased focus on commercializing secondary stage of education has negative
implications for girls’ participation. In a situation where girls’ education is not
valued, the demand for a fee-charging school system is bound to be low as parents
would not be willing to pay high charges for their daughters’ education. As mentioned
earlier, the increase in schooling participation at primary level witnessed during the
1990s is largely on account of demand from poorer and disadvantaged sections,
whose capacity to pay is also limited. The continuation of schooling for girls from
these sections depends crucially on state support and increased commercialization
Private
Government ⫹
Year local bodies Aided Unaided Total
a
2001/02 distribution denotes the position of schools that
remained in UP after bifurcation of the state into UP and
Uttaranchal.
N/A: Not available
Source: Education in India (1994–95, 1995–96 and 1996–97),
MHRD, Government of India, 2001, 2002, 2003 for 1994/95,
1995/96 and 1996/97, and Directorate of Secondary Education,
Uttar Pradesh for 2001/02.
Secondary Education in the Indian State of Uttar Pradesh 289
Table 13.10 Number of examinees and pass percentages in different types of schools (Grade
X and Grade XII examinations – 2002), Uttar Pradesh, India
works against that. But ironically, the only two schemes started to support girls’
education during the late 1990s focused on promoting unaided, fee-charging
schools. These two schemes were meant to direct private investments towards the
opening of single-sex schools for girls, but even these were subverted with the
outcome of further discouraging girls’ access to secondary schooling. How political
considerations successfully subverted the policy goal is discussed in detail in the
next section.
The massive presence of privately managed, aided and unaided schools makes
the role of the regulatory framework crucial. A regulatory framework is important
for directing resources and provisioning, maintaining essential physical facilities
and an adequate number of teachers, and improving the quality of teaching–
learning processes, in order to establish a system of accountability. Perusal of the
existing norms, guidelines and directives indicates a near-absence of accountabil-
ity mechanisms not only for unaided schools but also for aided schools that
survive on substantial grants from the state exchequer. Gender focus even in terms
of ensuring facilities and creating an enabling environment for girls is entirely
lacking.
Successive committees and national policy statements have highlighted the
need for an increase in the number of women teachers in the light of parental inse-
curity and the resultant restricted mobility of girls as well as to create an enabling
environment by providing positive role models in an otherwise non-inspiring
environment. This is especially true for rural areas. However, the state does not
have any policy of reservation for hiring women teachers at secondary level either
in government or aided schools. Though the government funds the entire salary
bill for aided schools, making this item the single largest constituent of the
290 Jyotsna Jha and Ramya Subrahmanian
government’s expenditure on this sub-sector, and the teachers enjoy all the
advantages of government service, no regulatory frame for their recruitment and
conduct exists in the state. Teachers from government and aided schools enjoy
substantial political clout and resist any effort to have accountability measures
in place. Kingdon and Muzzamil’s (2003) research shows that UP is one among
only four states that has an upper chamber of the state legislature, and teachers
have guaranteed representation in this body. They note two kinds of political
influence:
Political influence has been of two types: one, from above which has been
instrumental in shaping the education system; and two, the lobbying and
pressure groups from within the system originating at the local levels (and uniting
at the state-level) in the form of organizations of teachers. Education-related
legislation in UP has often been framed under immense lobbying pressure
from teachers, particularly at the primary and secondary levels. Teachers
in school (as opposed to higher education) have been instrumental in
determining the local base of political parties in the state.
2003:6, our emphasis
% of % of % of
Girls girls Girls girls Girls girls
(single- to (single- to (single- to
Total sex) total Total sex) total Total sex) total
1950/51 31 979 2520 7.88 2 854 468 16.3 987 154 15.6
1960/61 40 083 4927 7.93 4 335 661 15.2 1 771 282 15.9
1970/71 62 127 11624 18.71 8 787 2 008 22.8 3 415 581 17.0
1980/81 70 606 13 555 3 200 23.6 5 178 758 14.6
1985/86 73 490 14 728 3 377 22.9 5 709 848 14.8
1990/91 77 111 15 072 3 319 22.0 5 999 886 14.7
1991/92 78 085 15 328 3 447 22.4 6 060 889 14.6
1992/93a 74 889 17 730 3 610 20.4 6 115 560 9.2
1993/94 79 522 15 546 3 509 22.5 6 637 962 14.4
1994/95 82 023 15 976 3 817 23.8 6 637 962 14.4
1995/96a 88 817 19 516 2 651 13.6 6 977 1 042 14.9
1996/97 91 093 19 917 2 746 13.7 7 003 1 057 15.0
1997/98 92 554 20 436 2 960 14.4 7 135 1 127 15.7
1998/99 94 476 20 675 3 051 14.7 8 339 1 364 16.3
1999/00 96 764 21 678 3 237 14.9 8 549 1 427 16.6
2000/1b 86 361 19 639 3 021 15.3 8 459 1 501 17.7
2001/2 88 927 20 429 3 102 15.1 9 063 1 608 17.7
Notes:
a
The particular year issues of the Government of Uttar Pradesh publication were not available to the
researchers.
b
The state was bifurcated in 2000 into Uttar Pradesh and Uttaranchal. Hence, the number of schools
shown in 2000/01 and 2001/02 does not include schools located in Uttaranchal. For this reason all trend
analysis is up to 1999/2000.
Sources: (i) Shiksha kee Pragati (Education’s Progress) (Government of Uttar Pradesh), Directorate of
Education, Allahabad, different years.
(ii) Growth of Schools 1950–1992 (State-Wise), MHRD, Department of Education, 1997 for 1992/93 data.
(iii) Education in India 1995–96, MHRD, Department of Education, 2002 for 1995/96 data.
The first of these two schemes was introduced by the state government of UP in
1997 making any new private single-sex school for girls opened in a hitherto
uncovered block headquarter eligible for a one-time infrastructure grant of
Rs.1 million.11 Once all the block headquarters were covered, a new scheme was
brought to provide Rs.2 million to any new girls’ school being opened in an
uncovered Nyaya Panchayat.12 The fact that a good number of unaided private
girls’ secondary schools were opened reflects a positive response to these schemes.
However, these are fee-charging schools with no aid from the government for pay-
ing salaries or incurring any other recurrent expenditure. This means these schools
have to survive on their own revenue, the fees collected from parents being the
292 Jyotsna Jha and Ramya Subrahmanian
Table 13.12 Annual growth rate for the number of institutions at different school levels in
Uttar Pradesh, India
Girls Girls
Year Total Total (single-sex) Total (single-sex)
main source. The demand for fee-charging private schools for girls in these interior
areas did not seem to take off. As mentioned earlier, girls’ education not being
highly valued and the paying capacity of poorer households being low together
act to keep down demand for fee-charging schools for girls. The owners of the
schools opened under these schemes demanded that they be allowed to admit
boys as well. Initially, in 1999, all schools opened under these schemes were appar-
ently allowed to admit boys by an executive office order, which was later revised
after a gap of two years. Under the revised order, only those schools that are
located in rural areas were allowed to admit boys along with girls.13 It implied that
schools opened under this scheme and located in urban areas could not admit
boys. Separate statistics for enrolment are not available in the public domain to
indicate the proportion of boys and girls in these schools.
This experience shows that girls’ schooling participation in rural areas cannot be
promoted by the expansion of provisioning through fee-charging schools. It also
reveals the role of political considerations in policy-making and their implemen-
tation. While the larger concern for girls’ education at the national and interna-
tional levels made it a good political decision to have directed schemes to improve
provisioning of girls’ schools, narrow political considerations at the local level
demanded measures which subverted the very basis of allowing subsidy for a
private enterprise. Most of these schools were opened by local politicians who first
used the scheme to access public money for their private initiative and later
exercised their political influence to get an order passed which allowed them to
make their enterprise initially viable and later profitable. These facts are not easily
traceable as the available aggregate statistics do not provide any pointers and only
a deeper analysis of executive office orders and other such directives help reveal
such subversive attempts.
The rate of growth in enrolment slowed down during the 1990s, especially
the later part of the decade, for all stages of education. The growth rate for girls’
enrolment, however, has been higher as compared to boys at all stages. This is natural
Secondary Education in the Indian State of Uttar Pradesh 293
more students including girls are encouraged to opt for those subjects the shortage
of women teachers is going to continue. The fact that these practices have been
retained in the latest curriculum changes reflects the lack of a guiding vision based
on notions of gender equality to guide policy-planning. These practices perpetuate
gender disadvantage and the school system becomes an agent of reinforcement
rather than of change in the process.
Conclusion
In this chapter we have argued that education policy initiatives remain divorced
from broader visions of gender justice and social policy, partly on account of the
continuing dominance of discourses of investment in female education, which
rely on identifying thresholds for minimum public action. Although there are
multiple discourses influenced by diverse actors in the education policy arena,
these do not seem to translate into policies and practices that have an impact on
female schooling in a way that reflects the reality of girls’ education today.
We make three specific points. First, we argue that, at a symbolic level, the
continuing sway of the idea of the ‘educated person’ purely in terms of human
capital has resulted in mechanistic definitions of what it is people are assumed to
gain from education. This, arguably, emphasizes elementary education invest-
ment over subsequent levels of education, reduces the focus on developing poli-
cies appropriate to girls and boys at different stages of the life cycle, and places
issues of quality and equity at the fringes of policy action. Combined with insti-
tutional biases and the political challenges of promoting gender equality, claims
for investment in women that continually emphasize women’s role in helping
their families, communities and nations end up reinforcing the core messages of
human capital.
In our attempt to map the operation of discourses and practices in the secondary
schooling arena, we found that at all levels the neglect of secondary education for
girls was reinforced, thus appearing to be a direct translation of assumptions
related to human capital. However, we also argue that ‘gender’ does not operate
only in the symbolic arena and that symbolic discourses are not all-powerful.
Discourses can and do get renegotiated in the realm of practice, driven by varied
motivations and compulsions. State policies for equity, such as in the UP case, are
often implemented through market-mediated processes. While we do not wish to
suggest that this is the result of a linear process of translation of a particular (albeit
hegemonic) policy discourse at different levels of policy decision-making and exe-
cution, our case study has shown how political actors at the local level can subvert
policies enacted in favour of women, resulting in the dilution of state commit-
ments to gender equality, even if these were weak at the outset.
In the case of the policy enacted in UP to encourage female secondary school-
ing, we argue that this was weak to start with because it omitted to build up the
demand-side of girls’ secondary schooling, resulting in a low uptake of the
schools made available. The neglect of a focus on the demand for girls’ school-
ing also meant that when the policy was reversed, through the actions of
296 Jyotsna Jha and Ramya Subrahmanian
private providers, girls and parents affected by this move had barely any voice
to challenge the change in policy. This leads us to argue strongly for the active
role of states, even where private providers are in play, not just to subsidize the
costs of schooling, but also to build up demand to ensure that education provi-
sion is seen as a matter of right, and not a privilege to be bestowed and then
taken away.
Our third and final point refers to the issue of the outcomes of policy. Given the
emphasis on gender parity in secondary education as a Millennium Development
Goal (MDG) and Education For All (EFA) target, as well as the emphasis on gen-
der equality in the EFA goals, the issue of gender and education cannot rest at the
level of grand policy pronouncements on the importance of investment in
women. This point has global relevance, but here we speak to the Indian context.
While elementary education is finally getting the impetus it has been promised
for decades, care has to be taken that the policy directions followed for this sub-
sector do not have negative implications for the sub-sector that follows. If all
schools are not improved in terms of their quality, transition to secondary
schools in terms of access of the poorest, and of girls, will continue to be difficult
to achieve. This fact, combined with the neglect of the secondary sub-sector in
terms of providing girl-friendly single-sex schools that are relatively cost-free, will
mean that girls are pushed out of education at a time when they need resources
to develop the capacities, skills and aptitudes that will support their transition to
adulthood.
Notes
1 Notably, through a large USAID funded programme in the state, Innovations in Family
Planning Services (IFPS).
2 Based on estimates provided in the Implementation Completion Reports (World Bank)
of the two programmes.
3 See Vaidyanathan and Nair (2000) and Jha and Jhingran (2002)
4 These statistics have not been included in this chapter.
5 See Jha and Bhardwaj (2001) for detailed analysis on women teachers in rural India.
6 Office orders are executive measures often based on decisions taken by the bureaucracy,
which do not require to be approved by elected legislative bodies.
7 The Board of High School and Intermediate Examinations was established as an
autonomous body functioning under the Directorate of Secondary Education in 1922
in UP.
8 A special Secondary Education Commission (known as Acharya Narenda Dev Committee)
was appointed in UP which submitted its report in 1952.
9 Based on Sharadindu (2001)
10 Based on personal interactions with people and functionaries belonging to government
and private school systems.
11 Block is a sub-district administrative unit in India. It usually covers about 50 to 100 primary
schools.
12 Nyaya panchayat is a smaller administrative unit covering several villages and about
15–20 primary schools.
13 The revised order of the state government (order number 2501–8-2001–3009 | 5 | / 94) is
dated August 17, 2001. It revised the order issued on September 29, 1999 (order no.
15–8–99–3009 | 5 | / 94).
Secondary Education in the Indian State of Uttar Pradesh 297
References
In addition to the references cited below, a number of Acts, unpublished reports, internal
analyses papers, handouts, office orders, etc. were referred to.
Board of High School and Intermediate Examinations, Uttar Pradesh (different years),
Analyses of Examination Results. Allahabad Board of High School and Intermediate
Examinations.
Council of Boards of School Education In India (COBSE) (2002). Sahodaya, Newsletter,
Special Number, January 2002, Vol. XXXXI. New Delhi: COBSE.
Dreze, J. and H. Gazdar (1996) ‘Uttar Pradesh: The Burden of Inertia’. In J. Dreze and A. Sen
(eds) Indian Development: Selected Regional Perspectives. Oxford and New Delhi: Oxford
University Press.
Government of India, Ministry of Human Resource Development (different years).
A Handbook of School Education and Allied Statistics (1996). New Delhi: Department of
Education.
Government of India, Ministry of Human Resource Development (different years). Annual
Report. New Delhi: Department of Education.
Government of India, Ministry of Human Resource Development (different years). Analysis
of Budgeted Expenditure in Education. New Delhi: Department of Education.
Government of India, Ministry of Human Resource Development (different years). Education
in India. New Delhi: Department of Education.
Government of India, Ministry of Human Resource Development (1996) Growth of Schools
1950–1992. New Delhi: State-Wise.
Government of India, Ministry of Human Resource Development (1997) Growth of Enrolment
in School Education 1950–51 to 1993–94. New Delhi: State-Wise.
Government of India, Ministry of Human Resource Development (different years) Selected
Educational Statistics. New Delhi: Department of Education.
Government of Uttar Pradesh (2002) Vittiya Akhya, Mahyamik Shiksha (Financial Statement
Secondary Education) (2002–03). Lucknow.
Government of Uttar Pradesh (different years) Shiksha Ki Pragati (Progress of Education).
Allahabad: Directorate of School Education.
Jeffery, P., R. Jeffery and C. Jeffrey (2004) ‘Islamisation, Gentrification and Domestication:
“A Girls’ Islamic Course” and Rural Muslims in Western Uttar Pradesh’. Modern Asian
Studies, Vol. 38, No. 1: 1–53.
Jha, Jyotsna and Dhir Jhingran (2005) Elementary Education for the Poorest and Most Deprived
Groups: The Real Challenge of Universalisation. New Delhi: Manohar.
Jha, Jyotsna and Meenakshi Bhardwaj (2001) Women Teachers in Rural India. New Delhi:
UNESCO and Ministry of Human Resource Development (MHRD).
Kingdon, G. M. Muzammil (2003) The Political Economy of Education in India: Teacher Politics
in Uttar Pradesh. Oxford: Oxford University Press.
Ministry of Human Resource Development (MHRD) (1997) Remodeling of School Education
Board. Report of the Task Force on the Role and Status of Boards of Secondary Education.
New Delhi: Department of Education, Government of India.
Ministry of Human Resource Development (MHRD), Programme of Action 1992 (National
Policy on Education 1986). New Delhi: Department of Education, Government of India.
Ministry of Human Resource Development (MHRD), Report of the CABE Committee on Policy
1992. New Delhi: Department of Education, Government of India.
Ministry of Human Resource Development (MHRD), Towards an Enlightened and Humane
Society, NPE- A Review. Ramamurti Committee Report. New Delhi: Department of Education,
Government of India.
National Council of Educational Research and Training (NCERT) (1998). Sixth All India
Education Survey Report, New Delhi: Government of India.
Planning Commission (2002) Tenth Five Year Plan 2002–2007. Volume II, Sectoral Profiles and
Programmes. New Delhi: Government of India.
298 Jyotsna Jha and Ramya Subrahmanian
Sharadindu. (2001) School Education in U.P., Status, Issues and Future Perspectives. State Council
of Educational Research, Lucknow; mimeo of draft submitted to the National Council of
Educational Research and Training, New Delhi.
Stromquist, N. (1999) ‘The Impact of Structural Adjustment Programmes in Africa and Latin
America’. In C. Heward and S. Bunwaree (eds) Gender, Education and Development: Beyond
Access to Empowerment. London: Zed Books.
Subrahmanian, R. (2002a) ‘Engendering Education: Prospects for a Rights-Based Approach to
Female Education Deprivation in India’. In M. Molyneux and S. Razavi (eds) Gender, Justice
and Development. Oxford: Oxford University Press.
Subrahmanian, R. (2002b) ‘Gender and Education: Review of Issues for Social Policy’. Social
Policy and Development, Paper No. 9. Geneva: UNRISD.
Subrahmanian, R. (2003) ‘Gender Equality in Education: Definitions and Measurements’.
Background paper for the Global Monitoring Report on Education for All 2003–04. Paris:
UNESCO.
Tyagi, R. S. and Sharadindu (1999) Educational Administration in Uttar Pradesh. National
Institute of Educational Planning and Administration. New Delhi: Vikas Publishing House.
United Nations Educational, Scientific and Cultural Organization (UNESCO) (2002)
Education for All: Is the World on Track? Global Monitoring Report 2002/03. Paris: UNESCO.
United Nations Educational, Scientific and Cultural Organization (UNESCO) (2003) The Leap
to Equality: Gender and Education for All. Global Monitoring Report 2003/04, Paris:
UNESCO.
United Nations Children’s Fund (UNICEF) (2003) The State of the World’s Children 2004: Girls,
Education and Development. New York: UNICEF.
United Nations Development Programme (UNDP) (2003) ‘Millennium Development Goals:
A Compact among Nations to End Human Poverty’. Human Development Report 2003.
New York: United Nations Development Programme.
Unterhalter, E. (2003) Education, Capabilities and Social Justice. Paper prepared for the EFA
Global Monitoring Report 2003/09. Paris: UNESCO.
Unterhalter, E. and S. Dutt (2001) ‘Gender, Education and Women’s Power: Indian State and
Civil Society Intersections in DPEP (District Primary Education Programme) and Mahila
Samakhya’. Compare, Vol. 31, No. 1.
Vaidyanathan, A. and P. R. Gopinathan Nair (eds) (2001) Elementary Education in Rural India.
New Delhi: Sage.
Verghese, N. V. (2003) ‘Universalisation of Higher Education and Equity Concerns’. Paper
presented at the Oxford Conference on Education and Development, Oxford, UK.
Wazir, Rekha (2000) ‘Profiling the Problem’. In Rekha Wazir (ed.) The Gender Gap in Basic
Education: NGOs as Change Agents. New Delhi: Sage Publications.
Woodhall, M. (1987) ‘Human Capital Concepts’. In G. Psacharapoulos (ed.) Human Capital
Concepts. Page references relate to the reprint in A. H. Halsey et al. (eds) (1997) Education:
Culture, Economy and Society. Oxford and New York: Oxford University Press.
World Bank (2003a) Secondary Education in India. South Asia Human Development Sector
Report No. 2, New Delhi.
World Bank (2003b) Opening Doors: Education and the World Bank. Washington DC: World Bank.
World Bank (2003c) A Policy Note on the Grant-in-Aid System in Indian Education. South Asia
Human Development Sector Report No. 3, New Delhi.
Part IV
Financing Social Provisioning and
Counting in Women
14
Gendered Implications of Tax
Reform in Latin America: Argentina,
Chile, Costa Rica and Jamaica
Evelyne Huber
In Latin American and Caribbean countries, poverty and inequality have been
longstanding problems, and the momentous economic and social policy changes
over the past two decades have done little to ameliorate them. The most
effective means for reducing class- and gender-based poverty and inequality would
be citizenship-based entitlements to basic (i.e. allowing basic subsistence) income
support, healthcare, and education. In advanced industrial societies, public spending
is an extremely important instrument for the alleviation of class- and gender-
based poverty and inequality (Moller et al. 2003; Bradley et al. 2003; Huber and
Stephens 2001), and it could potentially play a similar role in Latin America and
the Caribbean. However, responsible, that is non-inflationary, financing of such
programs requires a sound system of taxation, something that is scarce in
developing countries, including in Latin America and the Caribbean. Systems of
taxation on their part have important implications for class and gender equity.
This chapter explores changes in the systems of taxation in four Latin American
and Caribbean countries – Argentina, Chile, Costa Rica, and Jamaica – from the
point of view of their gendered impact.
The starting point of the argument is that effective tax collection is a necessary,
though not sufficient, condition for the amelioration of gender-based poverty and
inequality. Low-aggregate tax collection hurts women because it prevents the
establishment of programmes that counteract market distribution of income, in
which women are generally disadvantaged. They are disadvantaged in market
income because they provide the bulk of the non-paid care work, because their
paid work takes place in the informal sector to a greater extent than men’s paid
work, and because – if they work in the formal sector – they tend to be employed
in smaller enterprises and to earn less than men.
Latin American countries as a whole have been undertaxing their populations,
with an average tax burden of 14 per cent of GDP in the first half of the 1990s,
compared to 17 per cent of GDP in a group of East and Southeast Asian countries
(IADB 1996: 128). Direct taxes amount to about 25 per cent of tax revenue only,
and of this amount some 60–80 per cent typically comes from corporate tax
301
302 Evelyne Huber
payments, while only 10–15 per cent comes from private individuals (ECLAC
1998: 72). Interestingly, the situation in the English-speaking Caribbean has been
very different, with an average tax burden in the first half of the 1990s of
27–28 per cent of GDP, essentially double the rate of Latin America, and direct
taxation on individuals and corporations accounting for some 40 per cent of tax
revenue (ECLAC 1998: 66–72). This contrast suggests that the fundamental reasons
for the poor tax collection performance in Latin America are poor policy choices,
rather than low levels of economic development and technological capacity.
The dominant pattern of tax reforms over the past three decades in Latin America
and the Caribbean, which – with some exceptions – holds true for our four cases,
has consisted of reductions of marginal tax rates on corporate and individual
incomes and attempts to broaden the tax base, but these have not succeeded in
significantly increasing income-tax collection. At the same time, revenue from
imports and exports and other international transactions declined significantly in
many cases due to the lowering of tariff rates and other forms of deregulation.
To make up for this decline and to facilitate tax collection, most countries have
shifted more weight to value-added taxes (VAT). There has been more variation in
the treatment of social-security taxes, corresponding to the different approaches
with regards to social policy reforms.1
Advocates of this pattern of tax reform, prominent among them representatives
of the World Bank, have argued that reduction of marginal tax rates reduces the
incentive for tax evasion (not declaring income or transactions) and tax avoidance
(manoeuvreing within the tax code to reduce tax liability) and that broadening of
the tax base reduces non-neutral tax treatment that encourages resource shifts to
relatively low tax activities and thus distorts resource allocation. Moreover, reducing
the complexity of the tax structure makes tax collection less costly and more effective
Tax category 1980–81 1999–2000 1980–81 1999–2000 1980–81 1999–2000 1980–81 1999–2000
Income, profits, 0.0%a 2.7% 5.5% 3.8% 2.7% 2.9% 10.8%b 9.2%b
capital gains
& property
Goods and services 5.4% 5.9% 12.0% 10.0% 5.0% 8.0% 14.5% 9.3%
International trade 0.9% 0.7% 1.5% 1.4% 4.1% 1.1% 1.3% 2.2%
Social-security 2.7% 3.4% 4.8% 1.4% 4.8% 6.2% 1.2% 0.0%
contributions
Other 4.4%a 0.0% 1.8% 0.8% 0.2% 0.0% 1.0% 2.0%
Total 13.5% 12.7% 25.7% 17.4% 16.8% 18.2% 28.8% 22.8%
a
Carciofi and Cetrángolo (1995) show a different figure in ‘Income, profits, capital gains & property’, and
a lower in ‘Other’.
b
Includes some payroll taxes that are social policy schemes.
Source: IMF Government Finance Statistics, and IMF International Financial Statistics.
Tax category 1980–81 1999–2000 1980–81 1999–2000 1980–81 1999–2000 1980–81 1999–2000
Income, profits, 0.0%a 20.9% 21.5% 21.8% 15.8% 15.8% 37.6%b 40.6%b
capital gains &
property
Goods and services 41.1% 46.7% 46.8% 57.3% 29.8% 44.0% 50.3% 40.9%
International trade 7.1% 5.8% 6.0% 8.8% 24.4% 5.9% 4.5% 9.7%
Social-security 19.6% 26.5% 18.7% 8.2% 28.7% 34.3% 4.1% 0.0%
contributions
Other 32.1%a 0.1% 7.1% 4.7% 1.4% 0.0% 3.5% 8.9%
Total 100.0% 100.0% 100.0% 100.7% 100.0% 100.0% 100.0% 100.0%
a
Carciofi and Cetrángolo (1995) show a different figure in ‘Income, Profits, Capital Gains & Property’, and
a lower in ‘Other’.
b
Includes some payroll taxes that are social policy schemes.
Source: IMF Government Finance Statistics, and IMF International Financial Statistics.
304 Evelyne Huber
come from the IMF International Financial Statistics (1991, 1997, 2003) and
Government Finance Statistics (1988, 1991, 1995, 2003). Other authors cite data that
may vary quite considerably from those presented here. For instance, Tanzi (2000)
cites ECLAC data that show significant discrepancies to our data for Chile and Costa
Rica in the earlier period and for Argentina and Costa Rica in the later period.
Carciofi and Centrángolo (1994), citing Tanzi (1987), present data for 1980–82
for Argentina that are much higher than ours, with total tax revenue accounting
for 19.9 per cent of GDP, and Gómez Sabaini et al. (2002) present figures for
Argentina for 1997 that show central government tax revenue as 19.7 per cent of
GDP, provincial tax revenue as 3.7 per cent of GDP and municipal revenue as
1.2 per cent. Thus, one would have to add about 25 per cent to our total central
government tax-revenue figures to arrive at the total tax burden. In addition, different
authors and institutions may classify certain taxes differently.
Having acknowledged these data problems, we can make three points to affirm
the usefulness of the IMF data for the comparisons to be made here. First, our data
all come from the same source and thus are more comparable than data pieced
together from different sources. Second, IMF figures are important because they are
used in setting targets for governments in the formulation of fiscal policy. Third, if
we look at the pattern of change in composition of tax revenue over time, and if we
are attentive to and correct for different classification decisions, the various sources
and authors paint a similar picture: They show a decrease in reliance on foreign-
trade taxes and an increase in reliance on indirect taxes, with the share of direct
taxes remaining roughly constant. The important exception is Jamaica.
government tax revenue accounting for 25.7 per cent of GDP and 28.8 per cent
respectively, whereas Costa Rica with 16.8 per cent of GDP and Argentina with
13.5 per cent of GDP were closer to the mean. It is important to note that Chile
had already undergone radical economic reforms and a tax reform by this point.
Indeed, Chile already relied on indirect taxes for close to half of total tax revenue,
with 47 per cent. Direct taxes, that is, income and capital-gains taxes on individuals
and corporations, together with property taxes, amounted to between 16 per cent
of total tax revenue (in Costa Rica) and 38 per cent (in Jamaica).3 The figure for
Jamaica is inflated because some earmarked taxes for social policy are added to the
income tax, which in other countries would be classified as social-security taxes.
Chile and Argentina had already lowered taxes on foreign trade by the early
1980s, whereas they continued to play a very important role in Costa Rica, more
important than income taxes. For Jamaica, we are dealing with a classification
problem. Bahl (1989), for instance, shows 25.5 per cent of total government revenue
in the early 1980s coming from export duties. In addition, he shows 20 per cent
coming from import duties, so the total contribution of foreign-trade taxes to gov-
ernment revenue in his data is 45.5 per cent, as in Costa Rica clearly higher than
income taxes. In the IMF data, stamp duties on imported goods must be classified
under taxes on goods and services and thus greatly inflate that figure. If we count
them as taxes on foreign trade, then Jamaica is in an intermediate position with
regard to the relative share of foreign trade and domestic indirect taxes, compared
to Chile and Argentina on the one hand and Costa Rica on the other. The former
two countries have a low share of total taxes coming from foreign trade and a high
share from domestic taxes on goods and services, and the latter has a comparatively
higher share coming from foreign trade and a lower one from domestic taxes on
goods and services.
slightly. Nevertheless, this category of taxes is by far the biggest source of tax
revenue, bringing in more than twice the revenue from income taxes. As a
reminder, we should add tax revenue at the provincial level and at the municipal
level amounting to roughly 25 per cent of tax revenue of the central government,
so that by the late 1990s total tax revenue for all levels of government reached
roughly 16 per cent of GDP. This is still lower than tax revenue in Chile, Costa Rica
and Jamaica.
A study of the distributive impact of the tax system, based on the 1997 house-
hold survey in urban areas and a survey of the consumption structure, shows that
it is slightly regressive (Gómez Sabaini et al. 2002). Since 88 per cent of the popu-
lation reside in urban areas (World Bank 2003), these data characterize the tax
system as it affects close to nine-tenths of the population. The GINI index for
income distribution among deciles increases from 0.55 before taxes to 0.58 after
taxes. This is mainly due to the VAT burden that weighs particularly heavily on the
lowest decile and least on the highest. The structure of social-security contribu-
tions is equally regressive, though it has a somewhat lower weight on the lowest
decile than VAT. The income-tax burden is progressive. However, since VAT has
such a central role in the tax structure, the system as a whole is shaped by its
incidence. Thus, looking at the growing role of VAT and of social-security contri-
butions from the early 1980s to 1997, we can conclude that it has become more
regressive. Indeed, the authors analysed data from 1986 with the same methodology
as the 1997 study and found that the degree to which the tax system was regressive
was more marked in the later period (Gómez Sabaini et al. 2002: 67).
over the original budget for 1990, and an increase in real expenditures on social
investment, mainly in healthcare, education and housing, of 12.2 per cent from
1990 to 1991 (Schkolnik 1992: 27–8). Tax revenue as a percentage of GDP
increased from 14.5 per cent in 1990 to 18 per cent in 1993.
Similar to Argentina, the tax structure in the second half of the 1990s
emphasized indirect taxes, particularly VAT, and de-emphasized direct taxation,
particularly individual income taxes (see Table 14.3). One of the main reasons for
the small contribution made by personal income taxes is the fact that declared
incomes are low, even in the top decile. Wealthy Chileans have the option of
forming a company, transforming part of their personal incomes into business
income, computing various expenses as costs, buying assets that are used in
personal consumption, and paying the flat 15 per cent business tax on profits. This
is a perfectly legal form of tax avoidance (Engel et al. 1998).
It is no surprise, then, that the tax system as a whole is somewhat regressive, just
like in Argentina. The Gini index before taxes in 1996 was 0.49, and after taxes
0.50 (Engel et al. 1998). Again, the income tax is clearly progressive and VAT clearly
regressive, but since VAT accounts for a much larger share of taxes, the system as a
whole is regressive. Gómez Sabaini et al. (2002) cite results from studies of tax
incidence in Chile in 1969 (Foxley et al. 1980) and 1997 (Engel et al. 1998), which
clearly show that the system was somewhat progressive in 1969 and became
significantly more regressive with the series of reforms since the 1970s. What is
important to note is that public expenditure in Chile in the 1990s did ameliorate
inequality; Engel et al. (1998) estimate that the Gini index fell from 0.5 after taxes
but before spending to 0.43 after taxes and spending.
In contrast to Chile and Jamaica, where total tax revenue as a percentage of GDP
fell rather markedly between 1980/81 and 1999/2000, Costa Rica managed to
increase revenue slightly in this period, from 16.8 per cent to 18.2 per cent of GDP,
which put it second only to Jamaica among our four countries. What makes Costa
Rica distinctive is the high share of total tax revenue accounted for by social-
security taxes. These increased their share from 29 per cent to 34 per cent of the
total in this period, whereas they increased also, but at a lower level, from 20 per cent
to 26 per cent in Argentina but declined from 19 per cent to 8 per cent in Chile.
The large decline in Chile is heavily due to the fact that employer contributions
were abolished when the pension system was privatized and employers contribute
nothing for health insurance for their employees either.5 As noted above, the data
for Jamaica are not comparable because taxes designated for social insurance and
services are included under income taxes. As in Chile, direct taxes on income,
profits and property retained a roughly constant share of the total in Costa Rica,
with 16 per cent.
I was not able to find any studies of the incidence of taxation in Costa Rica.
However, if we look at the composition of taxation in Costa Rica and Argentina in
1999/2000, we find a remarkable similarity (see Table 14.3). Thus, we might
assume that the distributive impact of the tax system is similar. However, this
assumption would ignore the differences in the structure of indirect taxes in the
two cases shaped by exemptions, as well as the differences in social-security taxes.
In Costa Rica, a basket of basic goods, veterinary products and agricultural
inputs are exempt from sales tax. In Argentina, a wide range of goods and services
is exempted from VAT, but these exemptions tend to benefit all income groups,
rather than being focused on low-income groups (CIPPEC 2002). Since the basket
of basic goods and agricultural inputs absorbs a large part of the income of low-
income groups but not of high-income groups, these exemptions make the Costa
Rican sales tax less regressive than the Argentine VAT. As noted, Gómez Sabaini
et al. (2002) found social-security taxes to be regressive in Argentina. These are
high and they have caps. Social-security taxes in Costa Rica are much lower
and there are no caps (Social Security Administration 1995). Thus, these taxes are
proportional to income.
Given the similarity in weight but differences in the structure of indirect taxes
and social-security taxes between Costa Rica and Argentina, with exemptions from
the sales tax more focused on goods consumed by low-income groups and social-
security taxes proportional to income in Costa Rica, we would expect the Costa
Rican tax system as a whole to be distributionally neutral or slightly progressive.
In addition, we need to keep in mind that redistribution happens mostly on the
expenditure side, and thus the fact that Costa Rica has the second highest level of
total taxation, behind Jamaica, puts the country in a comparatively favourable
position to effect redistribution. Indeed, in studies of household income, Costa
Rica along with Uruguay consistently shows the lowest Gini indices for income
distribution among 17 Latin American and Caribbean countries in the 1980s and
1990s (Morley 2001).
Gendered Implications of Tax Reform in Latin America 309
indirect taxes were decreased; the exemption from the general consumption tax of
food stuffs, health supplies, and books and materials for education, among other
categories, cushioned the burden on lower-income earners.
Looking at the composition of tax revenue as a whole at the end of the 1990s
and comparing it with the composition in the early 1980s, we can venture the
assessment that the tax system developed in a slightly more progressive direction,
with direct taxes increasing their share and taxes on goods and services decreasing.
These developments stand in contrast to the other three countries, where the
share of taxes on goods and services increased, sometimes dramatically so,
whereas the share of direct taxes remained stagnant. If we look at the sum of direct
taxes and social-security taxes as a percentage of GDP, a pronounced difference
emerges between Jamaica and Costa Rica on the one hand and Argentina and
Chile on the other (see Table 14.1). In Jamaica they accounted for 9.2 per cent
of GDP at the end of the 1990s and in Costa Rica for 9.1 per cent, compared to
5.2 per cent in Chile and 6.1 per cent in Argentina. Thus, the Jamaican and Costa
Rican tax systems were clearly more effective in collecting direct and social
security taxes.
Despite the tax reforms, all four countries continue to suffer from widespread
tax evasion and avoidance. The tax administrations all lack sufficient qualified
personnel and equipment for effective monitoring of tax compliance and for
auditing tax returns. There are many potential taxpayers, both individuals and
corporations, who are not even registered. Of those who are registered, only a frac-
tion actually file returns, and most of them under-report their income. For
instance, Bahl (1997: 220) points out that in Jamaica in 1992 there were about
80,000 registered self-employed taxpayers, and of those only about 25 per cent
filed returns. In a random sample of self-employed people from nine occupations,
the per cent of income tax filers was 17 per cent in 1982 and 11 per cent in 1983
(Alm et al. 1991: 196). Among large enterprises, compliance with the general con-
sumption tax was around 85 per cent but among all other enterprises it was in the
low 60 per cent range (Bahl 1997). For Chile, Engel et al. (1998) estimate that
around 23 per cent of the total tax base is not reported. In Costa Rica, a report
issued by the Controller General’s Office found that sales tax evasion in 2002
amounted to 46 per cent of the total, or roughly 2 per cent of GDP (La Nación,
8 March 2004). The evasion rate for VAT on domestic transactions in Argentina in
1993 was estimated at 32 per cent, a huge improvement from the 62 per cent in
1989, but still a large gap (Rosenwurcel 2000: 45).
An important part of the reason for continued tax evasion is the reluctance of
governments to punish it, particularly among major taxpayers. The example
of Jamaica shows that low to intermediate levels of development are not insur-
mountable barriers to tax collection due to insufficient technical capacity. In 1980/81,
Jamaica collected 29 per cent of GDP in tax revenue, with GDP per capita at
US$3,452 in 1980. In 1999/2000, Chile collected 17 per cent and its GDP per
Gendered Implications of Tax Reform in Latin America 311
capita in 1999 stood at US$9,568, and Argentina collected 13 per cent with a 1999
GDP per capita of US$11,181.6 It is clearly not technically feasible to enforce total
compliance (this is true for any of the advanced industrial democracies also), but
it is feasible in all four countries to improve tax collection capacity, if governments
develop the political commitment and support to do it.
If we look at the politics behind the tax changes, we find that these changes
were for the most part treated as a necessary component of general economic
reforms. In Costa Rica and Jamaica, they were explicitly and openly tied to donor
conditionality. In Argentina and Chile, the influence of the IFIs was exerted more
through technocratic networks of the sort analysed by Teichman (2001). She traces
the establishment of networks among officials of IFIs, particularly the World Bank
and the Inter-American Development Bank, and high-level government officials
in Argentina, Chile and Mexico. Through regular interaction over long periods of
time these networks became the carriers of neoliberal ideas and thus shaped
reform designs in a neoliberal direction even in the absence of explicit conditionality.
The fact that many of the high-level government officials shared educational
backgrounds with the IFI officials, such as PhD degrees in economics from US
universities, made them more susceptible to the transmissions and reinforcement
of neoliberal ideas.
There are two different paths to tax reform, exemplified by Argentina and Jamaica.
In Argentina, tax changes were generally ad hoc reactions to economic changes,
whereas in Jamaica they were part of a major coherent reform of the entire system of
taxation. In all four countries, the debates about tax reform were generally confined
to a relatively small circle of technocrats in the Ministry of Finance or Economics,
with minimal public involvement. Parliamentary debates tended to be comparatively
short also. Compared to other economic reforms, such as reduction of subsidies for
basic goods, energy, and the like, or privatization of state-owned enterprises, tax
reforms stimulated little popular protest. These other reforms had much more direct,
sudden and visible impacts on the population than the tax reforms that were phased
in more gradually. In some cases, such as the forced-savings scheme in Argentina
introduced in 1985, there were protests, but mainly after the fact.
Of course, one should also ask why Jamaica started from such a different
pre-reform position and remained more concerned about protecting lower-income
earners than Chile and Argentina. The key is the strength of the People’s National
Party (PNP), a democratic socialist/social democratic party that alternated in office
with the Jamaica Labour Party (JLP). The PNP has had a clear commitment to
reduction of poverty and inequality and when in office oriented both tax and
expenditure policies towards those goals (e.g. Manley 1982; Stephens and
Stephens 1986).7 Moreover, both the PNP and the JLP had close links to labour
unions, so the JLP governments were constrained in rolling back programmes that
the PNP had established. Even the tax reform discussed here, under JLP Prime
Minister Edward Seaga, had as one of its goals to avoid making the lower classes
worse off (Bahl 1997: 186). The National Liberation Party (PLN) in Costa Rica used
to play a similar role in promoting social programmes and a mode of financing
these programmes that would protect lower-income earners.
312 Evelyne Huber
In Chile, in contrast, major reforms of social policy and the tax system were
carried out under the dictatorship of Pinochet and were oriented towards stimu-
lating economic growth in the private sector and reducing the responsibilities of
the state for social welfare. Since unions and parties of the left were ruthlessly
repressed, these reforms could be carried out without having to deal with real
opposition. In Argentina, the major neoliberal reforms of economic and social
policy were carried out by President Menem, ironically from the Peronist party, the
traditional pro-labour party in the country. The party had, however, always been
highly personalistic and had turned into a non-programmatic, clientelistic party.
Major changes in the tax system towards increasing reliance on indirect taxes
had already been effected by the military government, and the various steps going
further in this direction with complicated packages did not evoke much public
debate.
Given the general absence of public debate about tax reform, it is little surprise
that concerns about the gender-specific impacts of the changes were hardly raised at
all. Effects of tax changes were not near the top of the agenda of women’s
movements in these countries either. In Argentina and Chile, human rights, democ-
ratization and cooperative economic-survival strategies dominated the agenda. To
the extent that women protested against economic policy in all of the four coun-
tries, it was mostly against the general austerity measures, visible in higher unem-
ployment and higher prices. Deteriorating pensions and deteriorating healthcare
were additional issues that elicited protests from women (as well as men). In short,
there were so many policy decisions taken to transform these economies in a neolib-
eral direction, in addition to general austerity measures, that tax-reform decisions
did not stand out in any major way in the public mind in general and in the mind
of women activists in particular. This may be in part cause and in part effect of the
lack of studies of the gender-specific effects of tax reforms in Latin America.
… the skepticism of the 1980s and early 1990s with respect to the redistributive
capacity of taxation has resulted in an almost total lack of up-to-date, detailed
empirical studies on the implications and distributional effects of various taxes
or combinations thereof in countries of the region. After almost two decades of
intensive and far-reaching reforms of the tax, tariff and social security systems,
there is not enough data to permit an informed assessment of such issues
ECLAC 1998: 96–7.
The main results of the few studies that exist for our countries were discussed
above. All of them looked at incidence on income groups, not a single one contained
Gendered Implications of Tax Reform in Latin America 313
any data on gender. Thus, the attempt to improve our understanding of the
gender-specific implications of these changes is more speculative, based on a
combination of theoretical and comparative knowledge and reasoning.
Many analysts would argue that the most important issue for married women is
whether the taxes of married couples have to be filed jointly or can be filed
separately. Joint taxation in the presence of a progressive income tax works as a
disincentive for married women to join the labour force.8 From that point of view,
all four cases are gender-egalitarian. In Chile, the wage tax and general income tax
are personal, levied on individuals and not on households (Engel et al. 1998).
In Argentina and Costa Rica married couples are taxed separately on their income,
but only one of them (in Argentina the husband) is taxed on property income and
property, and only one of them can take deductions for dependants (in Costa Rica
they can be claimed by either spouse). Where there is a flat-rate income tax, as in
Jamaica, individual or joint filing does not matter.
In order to assess the gendered implications of the changes in the tax structure, we
have to begin by situating women in the occupational and earnings structure of their
countries. In every one of our four countries, female labour-force participation is lower
than male. Though female labour-force participation increased significantly between
1990 and 2000 in Chile and Costa Rica, from 35 per cent and 39 per cent to
42 per cent and 43 per cent, respectively, and in Argentina slightly from 44 per cent
to 45 per cent, this rate remained clearly below the male rate, which varied from
73 per cent to 77 per cent (ECLAC 2002). The women who are in the labour force are
more likely to work in the informal economy, that is, in small unregistered enterprises,
own-account and non-remunerated labour, and as domestic workers (Table 14.4).
Women who work in the formal sector are over-represented in the public sector,
most probably as teachers and nurses. In the private sector, they have roughly
equal representation in professional and technical jobs to men but are greatly
under-represented in medium and large enterprises, that is, enterprises that employ
more than five workers (ECLAC 2002: 179–82). Finally, open unemployment
is higher among females than among males (ECLAC 2002: 207). Women earn
less than men, and the differences are greater if we look at the entire employed
population than if we consider just wage-earners. In 1999/2000, the ratio of average
female to average male incomes among the entire employed population in urban
areas was 61 in Chile, 65 in Argentina and 70 in Costa Rica (ECLAC 2002).
Given the lower labour-force participation rates and lower earnings of employed
women, the low share of revenue that comes from direct taxes clearly favours
males, because they are higher-income earners and more likely to be owners of
corporations or shares in corporations than women. Accordingly, men constitute
a stronger tax base and would have to pay more if the rate structure was more
progressive and exemptions lower, and – most of all – if the laws were enforced.
The same is true for the low share of total taxes coming from property taxes.
Social-security taxes are tied to earnings of employees and of those self-employed
who are covered, so their incidence on males and females should be similar to the
incidence of income taxes. However, to the extent that social-security taxes have a
cap and men tend to be higher-income earners, the effect of social-security taxes is
314 Evelyne Huber
Wage
earners in Own-account and Domestic
Year Country Total businesses non-remunerateda servants
Argentinab
1990 Men 37.3 12.4 23.1 1.8
Women 45.4 10.2 22.7 12.5
1994 Men 35.7 14.5 20.8 0.4
Women 42.5 11.5 18.7 12.3
1997 Men 34.1 17.5 16.2 0.4
Women 39.8 9.6 17.5 12.7
1999 Men 34.2 15.9 18.1 0.2
Women 38.3 10.3 15.3 12.7
2000 Men 35.1 16.4 18.6 0.1
Women 40.6 12 15.7 12.9
Chile
1990 Men 32.2 10 22 0.2
Women 45.5 8.2 18.2 19.1
1994 Men 27.5 9.1 18.3 0.1
Women 40.3 7.7 15.8 16.8
1996 Men 26.9 9.7 17 0.2
Women 39 8.2 14.5 16.3
1998 Men 26.2 9.7 16.4 0.1
Women 38.1 9.7 13.2 15.2
2000 Men 23.8 7.9 15.8 0.1
Women 36.7 7.4 13.3 16
Costa Rica
1990 Men 28.6 10.3 18.1 0.2
Women 37.2 8.6 16.6 12
1994 Men 28.6 11.6 16.7 0.3
Women 36.5 10.3 16.1 10.1
1997 Men 29.7 12.4 17.1 0.2
Women 37.1 9.2 18.7 9.2
1999 Men 30.4 13.3 16.7 0.4
Women 40.1 9.4 18.1 12.6
2000 Men 31.2 12.4 18.5 0.3
Women 38 10.9 15.7 11.4
a
Does not include professional and technical jobs.
b
Covers Greater Buenos Aires only.
Source: Tables 4.1 and 4.2 in Panorama Social de America Latina 2001/02 – ECLAC 2002.
The implications of the changes in tax legislation regarding direct taxes are
partly contradictory. Clearly, the reduction of marginal tax rates and of payroll
taxes benefited mostly men, because they are the high-income earners and formal-
sector employers and employees. On the other hand, to the extent that exemp-
tions and loopholes were closed, such as car allowances, these changes affected
men more negatively than women, as men were more able to take advantage of
those means before the reforms to reduce their tax payments. If we look at the
overall picture, we can assume that the burdens of direct taxes on men and women
remained roughly unchanged. If we introduce social-security taxes into the picture,
we see an increase in the share of these taxes in Argentina and in Costa Rica, and
a decrease in Chile. Accordingly, we would conclude that the total tax burden on
men, from direct taxes and social security taxes, was increased in Argentina and in
Costa Rica to a greater extent than on women, and that men experienced a greater
decrease in their tax burden than women in Chile.
We can begin an assessment of the gendered effects of the shift to greater
reliance on indirect taxes with some general observations. To the extent that
indirect taxes are regressive, which they clearly are in Argentina and Chile (Engel et al.
1998; Gómez Sabaini et al. 2002), we can assume that women as lower-income
earners are more disadvantaged. Where the degree to which they are regressive is
tempered through exclusions focused on basic goods and services in health and
education, as in Costa Rica and Jamaica, the incidence is presumably more gender-
neutral. To the extent that a large share of indirect taxes comes from excise taxes
on alcohol and tobacco, women as lower consumers of these goods would enjoy
an advantage – assuming that expenses for alcohol and tobacco do not come out
of the shared household budget. This forces us to think more about women’s
position in the household and about the determination of household budgets, a
topic to which I shall return in a moment.
The final incidence of taxes on foreign trade depends on the share that is passed
on to the consumer. In the case of import taxes on consumer goods, they simply
increase the price of those goods in the domestic market, and the incidence depends
on whether these goods are essential to the poor. Import taxes on intermediate and
capital goods increase the price of goods manufactured with the imported ones,
and the extent to which those increases are passed on to consumers depends on the
competitiveness of markets. In more competitive markets, presumably a greater
share of the import taxes for intermediate and capital goods would be absorbed by
the manufacturers, and in less competitive markets a greater share would be passed
on to consumers. In the latter case, we would again need to know whether those
goods are part of the necessary consumption of poor households. Everything else
being equal, a lowering of taxes on imports should have lowered prices to the con-
sumer and thus partially offset an increase in VAT, sales or consumption taxes.
In order to advance further in assessing the gendered impact of changes in the
tax systems towards greater reliance on indirect taxes, we need to situate women
not only in the occupational and earnings structure, but also in the structure
of households. We can think of at least eight combinations of marital status
(married/cohabiting in stable relationship or single/widowed/divorced), labour-force
316 Evelyne Huber
regressive taxes of all kinds, social security and indirect. Single women with
children are generally in the economically weakest positions; not only are they
likely to be disadvantaged in the labour market but they also bear the responsibil-
ity of providing for their children. Thus, they tend to have low incomes and have
to spend more time and energy to provide for the needs of their families if prices
increase due to tax increases. Unless the exemptions from the indirect taxes are
heavily directed towards basic foods, transportation, health and education services,
single women with children will shoulder the greatest burden from increases in
indirect taxes, that is, they will pay the largest percentage of their income for these
tax increases.
We could detect two clear tendencies as a result of the tax reforms effected in
Argentina, Chile, Costa Rica and Jamaica, over the last two decades of the twentieth
century. First, in three of the countries overall tax receipts as a percentage of GDP
fell; the exception is Costa Rica. Second, three of them came to rely more heavily
on indirect taxes levied on domestic goods and services; the exception here is
Jamaica. Both of these developments are negative from the point of view of reduction
of poverty, inequality and gender-based disadvantages through state intervention.
Lower tax revenue means a reduced capacity of the state to correct market outcomes
that disadvantage the poor and women by financing health and education services
and anti-poverty programmes. Indirect taxes are generally regressive, though the
extent to which they are regressive depends on exemptions of basic goods and
services. Women as lower-income earners and as mothers responsible for the needs
of the family, particularly if on fixed budgets, are more negatively affected than
men by changes that increase regressive indirect taxes.
In this light, Chile and Argentina have clearly done worse by their low-income
earners and by women than Costa Rica and Jamaica. They both lowered their tax
burden, which now is lower than in both Costa Rica and Jamaica, even though
their GDP per capita is much higher. They both increased their reliance on indirect
taxes on domestic goods and services, and their share from these taxes is higher
than the shares in Costa Rica and Jamaica. Their structure of exemptions does not
focus on basic goods and services to the same extent as that in Costa Rica and
Jamaica. Accordingly, their tax systems are regressive, whereas the Jamaican tax
system is progressive. Studies cited above have shown that they are regressive
across income classes, and we can conclude that they are also regressive in gender
terms. The lesser reliance on indirect taxes on domestic goods and services and the
more focused exemptions from these taxes prevent the indirect taxes in Costa Rica
and Jamaica from making the whole tax systems regressive, from both a class and
gender point of view.
Two major policy lessons emerge from this analysis. First, it is possible to
construct overall progressive tax systems that are effective in raising at least moderate
levels of tax revenue in Latin American and Caribbean countries, even at com-
paratively low levels of development, as demonstrated by the example of Jamaica.
318 Evelyne Huber
The keys are reliance on direct taxes with an adequate standard deduction and few
special deductions for higher-income earners for an important share of the total,
a structure of indirect taxes that exempts only basic goods and services, and social-
security taxes without a cap. In Argentina and Chile, income taxes are progressive
but the tax systems on the whole are regressive, because of the heavy reliance on
and the regressive structure of VAT. However, the extent of redistribution that can
be achieved through the tax system in practice seems to be limited. In particular,
the very top incomes are difficult to tax for political reasons. The wealthiest
(potential) taxpayers are also investors who need to be cultivated by governments
in order to generate economic growth, and they may also be donors to political
campaigns who need to be cultivated by politicians to win elections. Moreover, in
this era of liberalized capital markets and internationalized production it has
become more difficult to tax corporations. Finally, powerful international actors
oppose high tax rates. A World Bank study argues that, ‘most economists and
policymakers consider that high tax rates not only significantly discourage
and distort economic activity but are largely ineffective in redistributing income and
wealth’ (World Bank 2004: 130).
Second, it may be as important to enforce the tax laws that are already on the
books as to spend energies and political capital on changing the tax systems in a
major way, such as by introducing a VAT or increasing its importance on the
grounds that VAT is easier to collect than other kinds of taxes. This by no means
implies abandoning efforts to improve the distributive profile of tax systems, for
instance by changing the structure of the VAT to focus exemptions on basic goods
and services or by eliminating caps on social-security taxes, but it does mean to
make enforcement a priority. Arguably, the low capacity to enforce tax legislation
remains the key problem for Latin American tax systems, including those in our
four countries. It is not a new problem, but the reforms since 1980 have not been
able to solve it. Poor enforcement of legislation and consequent low tax collection
clearly are detrimental to the poorer sectors and to women, because redistribution
is in practice accomplished mostly on the expenditure side, and solid tax revenue
is needed to finance expenditures on transfers and social services.
The degree to which social expenditures in Latin America are redistributive
depends on the category of expenditures. Expenditures on health and education
tend to benefit lower income groups more than upper income groups in relation
to their incomes, whereas social-security expenditures tend to favour middle and
upper income groups (World Bank 2004, ch.4). The same argument can be made
for women and men; if women on a fixed budget are predominantly responsible
for providing for the health and education of their children (and for their own
healthcare), better public health services and educational facilities will benefit
them. In contrast, women are less likely to be covered by social-security schemes
as they are constructed at present because they are more heavily represented in the
informal sector. Since benefits are generally earnings-related, even those women
who are covered are disadvantaged as lower-income earners. Thus, increased tax
revenues devoted to high-quality and highly accessible education and health
services and to anti-poverty policies such as non-contributory flat-rate pensions,
Gendered Implications of Tax Reform in Latin America 319
Acknowledgement
I would like to thank Inés Valdez for excellent research assistance, and Debbie Budlender,
Shahra Razavi and John D. Stephens for comments on earlier drafts.
Notes
1 One should also note that most countries have introduced user fees for a variety of social
services, most prominently health services. Strictly speaking, these user fees constitute a
kind of tax, but they are not captured in any figures for tax burdens. To the extent that
these user fees are flat rate, they are regressive.
2 In liberalized economies without price controls, it is possible that the benefits of such
exemptions could partly be captured by producers or retailers rather than consumers,
which requires vigilance and potentially corrective regulatory action on the part of the
authorities.
3 In Table 14.1, property taxes are included in the category of direct taxes, along with
individual and corporate income taxes. They are low everywhere, between 4 per cent of
total tax revenue of the central government in Chile (Engel et al. 1998), 2 per cent in
Jamaica (Bahl 1997: 212), and less than 1 per cent in Argentina. In addition, in Argentina
provincial and municipal real-estate taxes amounted to 1 per cent of GDP (Gómez Sabaini
et al. 2002).
4 This section relies heavily on information provided by Carciofi and Centrángolo (1994),
Gaggero and Gómez Sabaini (1999), and Gómez Sabaini, Santieri, and Rossignolo (2002).
5 The Chilean pension system was completely privatized by the Pinochet dictatorship
and turned into a compulsory private savings scheme where issues of redistribution are
eliminated.
6 Data for GDP per capita are from the Penn World Tables, in constant dollars, adjusted for
purchasing power parity.
7 We know from comparative studies of advanced industrial societies that incumbency
of social democratic and Christian democratic parties is a strong predictor of generous
welfare states (Huber and Stephens 2001).
8 One might argue that this factor is less important in countries where a relatively large
proportion of lower-class women never marry even when they have children, which is
the case in Jamaica. However, since most of the direct income taxes are paid by middle- and
higher-income earners to begin with, who are also more likely to marry, this issue remains
relevant.
9 This is important to point out from the point of view of policy lessons; it may be better
to keep high excise taxes on gasoline and subsidize bus transport and transport of basic
foodstuffs, both from a class and gender point of view. Again, we have to consider the
expenditure side along with the tax side.
Bibliography
Alm, James, Roy Bahl and Matthew Murray (1991) ‘Income Tax Evasion’. In Roy Bahl (ed.)
The Jamaican Tax Reform. Cambridge MA: Lincoln Institute of Land Policy.
Bahl, Roy (1989) ‘The Political Economy of the Jamaican Tax Reform’. In Malcolm Gillis (ed.)
Tax Reform in Developing Countries. Durham NC: Duke University Press: 115–76.
320 Evelyne Huber
Bahl, Roy (1991) ‘The Economics and Politics of the Jamaican Tax Reform’. In Roy Bahl (ed.)
The Jamaican Tax Reform. Cambridge MA: Lincoln Institute of Land Policy: 1–62.
Bahl, Roy (1997) ‘The Jamaican Tax Reform: Its Design and Performance.’ In Wayne Thirsk
(ed.) Tax Reform in Developing Countries. Washington DC: World Bank: 167–232.
Bird, Richard M. and Barbara Diane Miller (1989) ‘The Incidence of Indirect Taxes on Low
Income Households in Jamaica’. Economic Development and Cultural Change 37(2) (January):
393–409.
Bradley, David, Evelyne Huber, Stephanie Moller, Francois Nielson and John D. Stephens
(2003) ‘Distribution and Redistribution in Post-Industrial Democracies’. World Politics,
Vol. 55, No. 2, January: 193–228.
Carciofi, Ricardo and Oscar Cetrángolo (1994) ‘Tax Reforms and Equity in Latin America:
A Review of the 1980s and Proposals for the 1990s’. Innocenti Occasional Papers; Economic
Policy Series, No. 39. Florence, Italy: UNICEF International Child Development Centre.
Centro de Implementación de Políticas Públicas para la Equidad y el Crecimiento (CIPPEC).
(2002) ‘Hacia un sistema tributario permanente para Argentina’. Buenos Aires: CIPPEC.
Economic Commission for Latin America and the Caribbean (ECLAC) (1998) The Fiscal
Covenant: Strengths, Weaknesses, Challenges. Santiago, Chile: ECLAC.
Economic Commission for Latin America and the Caribbean (ECLAC) (2002) Social Panorama
of Latin America 2001–2002. Santiago, Chile: ECLAC.
Engel, Eduardo M.R.A., Alexander Galetovic and Claudio E. Raddatz (1998) ‘Taxes and
Income Distribution in Chile: Some Unpleasant Redistributive Arithmetic’. Working Paper
6828. Cambridge MA: National Bureau of Economic Research.
Foxley, Alejandro, Eduardo Aninat and Jose Pablo Arellano (1980) ‘Las Desigualdades
Economicas y la Accion del Estado’. Mexico City: Fondo de Cultura Economica.
Gaggero, Jorge and Juan Carlos Gómez Sabaini (1999) ‘El Sistema Tributario Federal:
Diagnóstico y Propuestas’. Unpublished manuscript, Universidad Nacional de Buenos
Aires, Argentina.
Gaggero, Jorge and Juan Carlos Gómez Sabaini (2002) ‘Argentina: Cuestiones Macrofiscales y
Reforma Tributaria’. Buenos Aires: Fundación OSDE-CIEPP.
Gómez Sabaini, Juan Carlos, Juan José Santieri and Darío Alejandro Rossignolo (2002)
‘La equidad distributiva y el sistema tributario: un análisis para el caso argentino.’ Serie
Gestión Pública. Santiago de Chile: Comisión Económica para América Latina.
Hossain, Shahabuddin M. (2003) ‘Poverty and Social Impact Analysis: A Suggested
Framework’. IMF Working Paper WP/03/195, International Monetary Fund.
Huber, Evelyne and John D. Stephens (2001) Development and Crisis of the Welfare State:
Parties and Policies in Global Markets. Chicago: University of Chicago Press.
Huber, Evelyne, David Bradley, Stephanie Moller, Francois Nielsen and John D. Stephens
(2001) ‘The Welfare State and Gender Equality’. Luxembourg Income Study Working Paper
No. 279, September.
Inter-American Development Bank (IADB) (1996) Economic and Social Progress in Latin
America: 1996 Report. Washington DC: IADB.
International Monetary Fund (IMF) (1988, 1991, 1995, 2003) Government Finance Statistics.
Washington DC: IMF.
International Monetary Fund (IMF) (1991, 1995, 2003) International Financial Statistics.
Washington DC: IMF.
La Nacion, at www.nacion.com, accessed 8 March 2004.
La Nacion, at www.nacion.com, accessed 1 December 2003.
Manley, Michael (1982) Struggle in the Periphery. London: Third World Media.
Moller Stephanie, David Bradley, Evelyne Huber, Francois Nielsen and John D. Stephens
(2003) ‘Determinants of Relative Poverty in Advanced Industrial Societies’. American
Sociological Review, Vol. 68, No.1, February: 22–51.
Morley, Samuel (2001) The Income Distribution Problem in Latin America and the Caribbean.
Santiago, Chile: ECLAC.
Gendered Implications of Tax Reform in Latin America 321
Quesada Camacho, Juan Rafael, Danile Masís Iverson, Manuel Barahona Montero, Tobías
Meza Ocampo, Rafael Cuevas Molina and Jorge Rhenán Segura (1999) Costa Rica
Contemporánea: Raíces del Estado de la Nación. San José, Costa Rica: Editorial de la
Universidad de Costa Rica.
Rosenwurcel, Guillermo (2000) ‘Fiscal Reform and Macroeconomic Stabilization in
Argentina’. In Guillermo Perry, John Whalley and Gary McMahon (eds) Fiscal Reform and
Structural Change in Developing Countries, Vol. 1: 26–58. Basingstoke: Palgrave.
Santiere, Juan J., Juan C. Gómez Sabaini and Darío A. Rossignolo (2000) ‘Impacto de los
Impuestos sobre la Distribución del Ingreso en la Argentina en 1997’. Banco Mundial,
Secretaría de Programación Económica y Regional.
Schkolnik, Mariana (1992) ‘The Distributive Impact of Fiscal and Labour Market Policies:
Chile’s 1990–1 Reforms’. Innocenti Occasional Papers; Economic Policy Series, No. 33.
Florence, Italy: UNICEF International Child Development Centre.
Sjoquist, David L. and David Green (1992) ‘Distribution of Tax Burdens’. Jamaica Tax Review
Project, Report No. 3. Atlanta GA: Georgia State University, Policy Research Center, College
of Business Administration.
Social Security Administration (1995) Social Security Programs Throughout the World.
Washington DC: SSA.
Stephens, Evelyne Huber and John D. Stephens (1986) Democratic Socialism in Jamaica:
The Political Movement and Social Transformation in Dependent Capitalism. London:
Macmillan and Princeton: Princeton University Press.
Tanzi, Vito (1987) ‘Quantitative Characteristics of Tax Systems in Developing Countries’. In
Newbery and Stern (eds) The Theory of Taxation for Developing Countries. New York: Oxford
University Press: 205–41.
Tanzi, Vito (2000) ‘Taxation in Latin America in the Last Decade’. Paper prepared for
the Conference on Fiscal and Financial Reforms in Latin America, Stanford University,
9–10 November.
Teichman, Judith (2001) The Politics of Freeing Markets in Latin America: Chile, Argentina, and
Mexico. Chapel Hill NC: University of North Carolina Press.
Thirsk, Wayne (1997) ‘Overview: The Substance and Process of Tax Reform in Eight
Developing Countries’. In Wayne Thirsk (ed.) Tax Reform in Developing Countries.
Washington DC: World Bank: 1–32.
Wasylenko, Michael (1991) ‘Tax Burden Before and After Reform’. In Roy Bahl (ed.)
The Jamaican Tax Reform. Cambridge MA: Lincoln Institute of Land Policy: 751–92.
World Bank (2003) World Development Indicators. Washington DC: World Bank.
World Bank (2004) Inequality in Latin America and the Caribbean: Breaking with History?
Washington DC: World Bank.
15
Expectations versus Realities in
Gender-responsive Budget
Initiatives
Debbie Budlender
Introduction
The quote above and similar quotes from other reports give either implicit or explicit
indications of the ambitious expectations that the gender-responsive-budgeting
(GRB) idea evokes. The key question addressed in this chapter is: How does what
GRB initiatives have done in practice compare with the claims and expectations
about what they can achieve? In asking this question, the chapter does not aim to
detract from what has been achieved. Instead, it attempts to bring some realism
into the discussion, planning and assessment of these initiatives. The chapter also
stresses that different initiatives have different objectives and different outcomes
which depend on context, who is involved, and a host of other things. There is
therefore no single ‘correct’ approach.
The chapter does not explain in any detail what GRBs are. There is a range of
other sources which provide such a description and discussion (see, for example,
Reeves and Sever 2003; Budlender et al. 2002; Budlender and Hewitt 2003). The
working definition of a GRB for this chapter is that it involves an analysis of the
government budget in terms of its reach and impact on women and men, girls and
boys. A GRB is thus, in effect, a form of policy analysis from a gender perspective.
GRBs do not focus only on the numbers contained in the budget. They focus as
much – if not more – on the policy and programmes underlying those numbers.
Ideally, they also focus on what happens when the policies and programmes are
implemented. The ‘added value’ of GRBs in terms of policy analysis is that they
recognize that any other government policy or programme will not be effective
unless adequate resources are allocated to implement it.
The chapter is divided into three sections. The first section deals with issues
related to budgets and their relationship to conceptualizations of the economy
322
Expectations versus Realities in Gender-responsive Budget Initiatives 323
and economic and social policy. The second deals with issues related to gender as
a critical variable that structures the economy and society, alongside other axes of
difference such as race, class and age. The third deals with issues related to policy-
and budget-making as a process. The different sets of issues are often related to
each other, and there is thus some overlap between the sections.
Macroeconomic policy
GRB initiatives are often described as being about engagement with economic
policy. This is true because the budget constitutes a major part of the fiscal macro-
economic policy of the country. However, the assertion can be misleading if it
relies on too narrow a notion of the ‘economic’.
First, while the budget is part of economic policy, the way it is distributed affects
the effectiveness of the policies of all parts of government – economic, social and
protective. The budget is, in effect, the monetary reflection of all the policies of a
government. Second, talking about the budget as ‘economic’ policy can detract
from the aim of using gender budget work to integrate social and economic con-
siderations. Third, the understanding of budgets as part of economic policy high-
lights the very limited progress that has been made in addressing other parts of
economic policy. Broad statements about the gender impacts of macroeconomic
policy are common, and relatively easy to make. Translating these into concrete
analysis and proposals is a harder nut to crack.
If GRB initiatives are serious about influencing macroeconomic policy, they will
need to engage with the Ministry of Finance beyond the budget division which is
usually involved in GRB work. They might well discover at this stage that the
divisions responsible for macroeconomic issues have limited power and that it is,
instead, the international financial institutions (IFIs) who call the shots. The
World Bank has shown its support for GRB work to the extent of attempting to
establish initiatives in several countries. The International Monetary Fund has
published a working paper (Sarraf 2003) on the topic. However, both institutions
might balk at allowing GRB work to ‘interfere’ with macroeconomic strategies,
policies and models.
Enhancing rights
GRB is sometimes seen as a way of monitoring the Convention on the Elimination
of All Forms of Discrimination Against Women (CEDAW) and socioeconomic
rights more generally. However, the area of rights, like that of macroeconomic policy
beyond budgets, has proved one of the more difficult areas of GRB work. One chal-
lenge is that those who get involved in budget work usually have a different
approach to those who get involved in rights work. In particular, those who work
on budgets tend to have a more technicist approach, while the rights advocates
adopt a more idealist approach. Stated crudely, the technicists tend to argue in
terms of ‘prioritization’ and choices within a given budget ceiling, while the rights
advocates are more inclined to argue for ‘more for everyone’.
A related potential obstacle is that the rights discourse is sometimes not persua-
sive when talking to bureaucrats, especially those from technical ministries such as
Finance. In addition, reference to international instruments will often have little
sway with technocrats, especially where they already feel themselves under con-
siderable other pressures from IFIs and donors who wield power over their policies
and budgets.
Impact
GRBs are political interventions which aim to change the nature of budgets and
the policies and programmes that underlie them so as to change the situation of
(poor) women and men, girls and boys in a particular country. As such, the ques-
tion naturally arises as to their impact – whether they have effected any of the
intended changes. Unfortunately, impact is not easy to measure or to attribute
to particular interventions, whether for GRBs or any other political and social
interventions.
Some initiatives can claim some budget changes. In Mexico, for example, the
Commission on Gender Equality of the Chamber of Deputies succeeded, on the
basis of GRB analysis conducted by civil society, in its arguments for increased
allocations for reproductive health. Additional funding was also allocated for a
programme to address maternal mortality and for programmes for women in agri-
culture and immigrant women. In the United Kingdom, the work of the Women’s
Budget Group convinced the Treasury to redesign the Child Tax Credit so that it
was payable to the main caregiver (likely to be a woman) rather than to the main
breadwinner (likely to be a man). In South Africa, the GRB can claim some credit
for the introduction of the child support grant which is given to the primary
caregivers of young children from poor households.
Some GRBs can point to statements by finance ministers and others in which
they acknowledge the importance of gender. Some can point to budget documents
which now more clearly report on gender issues. All these achievements are impor-
tant, at least symbolically. Overall, however, the more than fifty GRB initiatives
around the world have probably produced relatively few budget changes.
One reason for the limited changes in budget figures is that policy does not
usually change because of a single initiative, but rather because of the combina-
tion of a range of forces. It is thus difficult for a single group to claim credit for a
Expectations versus Realities in Gender-responsive Budget Initiatives 325
change in policy. Another reason for the limited claims that GRB initiatives can
make in respect of budget changes is that policies seldom change simply because new
and better ‘facts’ are presented, much as we would all like to believe in evidence-based
policy-making.
A third reason for the limited claims that can be made about changes is that
those involved often do not have the power themselves to change budgets. This
lack of power is clearly the case for civil-society actors. I argue below that it is also
the case for most parliaments. And I argue further that it is often the case for civil
servants.
The limitations on the power of civil servants to change budgets is particularly
acute if the GRB initiatives involve only gender focal points or gender ministries.
Gender ministries should, in theory, have more power to influence others as their
mandate is usually stated as being to ensure mainstreaming through all other min-
istries. However, the weakness of these ministries in most countries is well-known.
Further, the staff involved often do not engage very energetically in GRBs. Often
the officials fear the technical nature of the venture. Many feel more comfortable
doing gender-awareness training within government and in civil society. Many
feel more comfortable focusing on women’s projects which, while outside the real
mandate of most gender ministries nowadays, continue to be funded by donors
who want concrete ‘deliverables’.
half of the population in most countries of the world. The second objection is that
if the GRB truly focuses on gender, it concerns every individual in that every
individual is part of the gendered make-up of society. Thus while ‘special interest’
budget initiatives tend to argue for ‘more’ for children, people with disabilities, or
whatever group is involved, theoretically GRBs should be advocating for ‘more
equitable’ programmes and allocations – ones which address the different needs
and interests of individual from different (overlapping) social groups.
‘More equitable’ is, of course, not a straightforward criterion to apply. Many
would see a 50–50 allocation between male and female as ‘equal’, and therefore
‘equitable’. A more sophisticated – and more valid – conception sees ‘equitable’ as
requiring recognition of the different situations of women and men, girls and
boys, and other social groups which give rise to their potentially different needs
and interests.
I wrote above that ‘theoretically’ GRBs should be advocating for more equitable
programmes and allocations. In practice, the situation is more complicated.
Several later initiatives in other countries used similar categories. Ideally, the
three-way categorization should highlight the necessity for focusing on the third
category given that this accounts for the overwhelming majority of the funds
allocated in the budget as a whole.
In Philippines and Indonesia (women) politicians have, perhaps unwittingly,
encouraged an approach of focusing on ‘special’ allocations for women or gender.
In Philippines in 1995 Senator Shahani proposed and pushed through a law which
Expectations versus Realities in Gender-responsive Budget Initiatives 327
stipulated that every national agency should allocate 5 per cent of its budget for
‘gender and development’ (GAD). This allocation soon became known as the
‘GAD budget’. The ruling was later extended to all local government units.
The position of the National Commission on the Role of Filipino Women
(NCRFW), which is the Philippines’ national gender machinery, is that the 5 per cent
GAD budget should be used to promote gender mainstreaming in the budget as a
whole. It asks that every government agency and local government unit draw up a
GAD plan, and use at least 5 per cent of the budget to implement this plan. Women
activists in the Philippines defend the approach strongly. They and the NCRFW
have put significant energy into training of officials and other activities to assist in
the drawing up of GAD plans and budgets. The energy has not always paid off. The
most notorious example of misallocation is the agency that included money for
ballroom dancing lessons for female civil servants in its GAD allocation.
One can understand the move to specify a percentage of the budget as a strategy
to have something which is measurable and thus more easily monitored. One
wonders, too, to what extent the percentage approach was influenced by the quota
debate in respect of representation of women.
Unpaid labour
A key aspect that distinguishes (or should distinguish) gender budget work from
other budget work is the focus on unpaid labour.
Both government and non-government budget work is often dominated by
economists. And in virtually all schools and all countries, economists are trained
to analyse markets and production that goes through the market. Feminist econo-
mists have taken the lead in pointing out the blindness of an approach which
ignores the non-market production which occurs in all economies and societies,
and which is necessary to ensure the ‘production’ and ‘maintenance’ of the people
who do the monetary production.
While unpaid labour is unseen in traditional economics, in my experience it is
very easily and quickly grasped by diverse audiences. Ordinary women tend to
understand the issue immediately, as they have daily experience of what it entails.
Middle-class women – and civil servants in particular – sometimes do not ‘feel’ the
issue in the same way because they have the resources to employ other women to
carry some of the burden of unpaid labour for them. However, they will have
usually experienced the difference in expectations of male and female when children,
as well as when adult in respect of childbearing and childrearing. In addition, most
men will – after initial puzzlement – start nodding their heads in agreement when
the issue is discussed.
The inclusion of unpaid labour in one’s approach to economics is in many
ways revolutionary. However, as one senior economist in a Ministry of Finance
remarked after a short presentation on the topic, it is ‘completely logical’ that
unpaid labour should be included in macroeconomic and other models. However,
despite its importance, and the fact that it is ‘obvious’ once pointed out, the
unpaid-labour aspect is not present in all GRB work. And when it is present, it is
often not strongly developed.
328 Debbie Budlender
One obstacle is the lack of available data. Relatively few countries in the world
have undertaken the time-use studies which provide estimates which can be con-
verted into monetary equivalents. Detailed data are, however, not necessary for
unpaid labour to be considered in GRB initiatives. If – as is usually the case – everyone
acknowledges that women bear the main burden of most types of unpaid labour,
exact data strengthen the case and make it appear more scientific, but are not
required for appropriate policy-making.
Unpaid labour is completely absent from some GRB initiatives. In others, it is
mentioned as part of the situation analysis, but not necessarily taken further when
it comes to discussing policy and associated budgets. Given the importance of
unpaid labour as a basis for gender inequality, there are surprisingly few concrete
suggestions as to how it should be recognized in policy- and budget-making
beyond broad calls for increased provision of safe water, electricity and childcare.
The unpaid-labour focus probably emerges most strongly in the work of
the United Kingdom’s Women’s Budget Group around tax credits. The UK example
is, however, not immediately useful to developing countries because few have
revenue/benefits systems. Unpaid labour should, in theory, be even more of an
issue in developing countries than in developed because there is less public provision
of relevant services by the state or by the market. The HIV/Aids pandemic and
its attendant need for additional (unpaid) care should increase the importance of
the issue.
Pro-poorness
The World Bank has suggested that GRBs can ‘engage in a process of transformation
to take into account the needs of the poorest and the powerless’ (Blackden and
Bhanu 1999: 64–5). However, GRBs do not all necessarily have a pro-poor agenda.
Some groups get involved in GRB work out of a concern with disadvantage and, in
particular, the disadvantage that female people usually experience when compared
with male people. These groups will usually also be concerned about other aspects
of disadvantage. Other groups have a narrower focus on ‘gender equality’. This
often translates, in practice, into the promotion of women’s interests. These
groups would be as interested in promoting the interests of women entrepreneurs
as promoting the interests of poor women (and men).
There are several examples from gender analysis of revenue which illustrate the
tension between a pro-women and an equity approach. The first example comes
from the South African GRB’s second attempt to analyse taxation from a gender
perspective. During the apartheid years, the South African personal tax system
provided separate schedules for ‘married persons’ (actually married men), married
women and unmarried people. ‘Married persons’ were taxed at a lower rate on the
assumption that they would be supporting others. Married women were taxed at a
higher rate on the assumption that theirs was a second income in a household.
In subsequent years this was changed so that personal tax is now determined on an
individual basis, irrespective of the sex or marital status of the person concerned.
However, Smith (2002) points out that while the change has done away with
formal discrimination, it has probably left – and perhaps introduced – substantive
Expectations versus Realities in Gender-responsive Budget Initiatives 329
focused on the interests of middle-class women rather than the poorest women
(and men). They raised, for example, the high taxes placed on imported khangas
and kitenge, baby food and sanitary pads. None of these are goods which are likely
to be of concern to poorer women. Similarly, their call for the increase of the
amount at which people start paying tax would benefit more men than women,
while simultaneously decreasing the amount of money available to government to
provide services to those who earn too little to pay direct tax with the current
cut-offs. In contrast, the UK Women’s Budget Group has consistently argued
against tax cuts or raising tax thresholds even at the very bottom of the income
scale on these grounds.
Participation
There is often slippage in discussions of ‘progressive’ budget work in that
GRB work is assumed, by definition, to involve a more participatory approach. The
slippage can be partly explained by the fact that GRB work is often taken up
by organizations and actors with an interest in women’s political participation.
The slippage is, however, unfortunate. First, greater political participation by
women does not, in itself, always result in greater attention to gender issues or
women’s interests, although – overall – it is likely to do so. Second, there is a
chicken-and-egg question – does greater political participation by women result in
GRBs, or does GRB work tend to result in greater political participation by women?
Third, we need to be clear whether we are talking about participation in the
bureaucracy, in parliament, or within civil society. Finally, we need to avoid the
limited understanding of some governments who seem to feel that participation
should be by invitation and that any participation by non-governmental actors is
sufficient, no matter who the actors are. In these cases, participation is often
restricted to academics, think-tanks and/or government-aligned organizations
who are not necessarily those who might raise the issues which should be at the
heart of a GRB approach.
in parliament for women and other ‘marginalized groups’ (the military, workers,
youth and people with disabilities). At the time FOWODE established the GRB in
the late 1990s, the women’s caucus had more power and impact than similar bodies
in many other countries because of the absence of parties. Indeed, by uniting with
representatives of other ‘marginalized’ groups, the women’s caucus was, in effect,
the single largest caucus in the Ugandan parliament. The GRB thus had an immediate
audience and advocate.
In the first years, FOWODE’s GRB activities focused on the national-level budget.
FOWODE later extended activities to the local level as a direct result of its concerns
about participation, as well as out of concern as to what decentralization might
mean in terms of gender equality. The GRB initiative was established around the
time of the Constituent Assembly, which drew up a new constitution for the coun-
try. The women parliamentarians put a lot of effort into the Constituent Assembly
and managed to win a quota for women at the local level. FOWODE followed up
on this victory by providing support and training to women to stand for and win
local elections. They then went a step further by providing training to the success-
ful women to ensure that the victory went ‘beyond numbers’ to bring gender-sen-
sitive policy into local government. FOWODE’s local-level GRB was part of the
training and other activities intended to take the local-level quota a step further.
In South Africa, the GRB initiative was established one year after the first demo-
cratic elections which brought an end to apartheid. Over a quarter (27 per cent) of
the membership of the national parliament in the first post-apartheid parliament
were women – a significant increase from the 3 per cent of the final apartheid
years. Most of the new women representatives had been part of the struggle
against apartheid. Many had participated in the Women’s National Coalition
(WNC), a cross-party and cross-race grouping which, among others, ensured that
the interim constitution of the country included strong clauses on gender equity.
The WNC had also drawn up a Charter on Effective Equality for presentation to
the Constitutional Assembly. This Charter was drawn up through a fairly extensive
process of consultation around the country and represented a wish-list of what
women wanted to see in the ‘new South Africa’. Pregs Govender, previously the
national organizer of the WNC, was one of the new parliamentarians. She recog-
nized that women parliamentarians would need some way to ensure and monitor
that the demands of the Charter were being met, and that the budget could be a
powerful tool to do so. She therefore met with two policy-oriented research NGOs
to discuss how this could be done. This saw the birth of South Africa’s GRB.
In both the Ugandan and South African cases, then, the establishment of the
GRB resulted fairly directly from increased participation of women in parliament.
GRBs have, however, not been established in all countries with high levels of
participation. Some quota systems might, in fact, discourage the establishment of a
GRB. This could happen if the GRB is seen as a too effective means of oversight,
and if the women elected through a quota system feel beholden to government
and fear losing their seats if they are too critical.
The Ugandan and South African GRBs were established at a time when women
representatives were largely supportive of government policy. In both countries,
332 Debbie Budlender
the executive, on its side, had a strong commitment to gender equality. Thus while
both GRBs styled themselves as exercising oversight over government, they were
largely advocating what the executive itself was committed to doing, but was
perhaps not doing forcefully enough, or with enough resources. In both countries
the leaders of government had no problems with the GRB initiatives, and
sometimes openly supported them. Indeed, Pregs Govender was rewarded by
being made chairperson of a new Parliamentary Committee on the Improvement
of the Status and Quality of Life of Women. In Uganda, FOWODE was invited to
participate in the sector committees which meet with the Ministry of Finance in
the early months of the budget cycle each year.
In later years the leaders of the GRB initiatives in both countries increasingly
criticized government policy. In Uganda, the situation was complicated by the fact
that Winnie Byanyima’s husband stood for the presidency (and lost) against Yoweri
Museveni. In South Africa, tensions developed when Pregs Govender started
criticizing the stance of the government on issues such as large-scale military
expenditure and HIV/Aids policies, and took many of the members of her
committee along with her. Both women leaders were prepared to stand up to gov-
ernment and pay the consequences. Winnie Byanyima suffered ongoing harassment.
For Pregs Govender, the consequences were leaving parliament. Many other
women parliamentarians might not be prepared to take this strong a stand.
potential of GRB work, and are prepared to put in the effort needed to sustain it,
that are crucial. Further, all three countries are ones where there has been a recent
transition away from authoritarian government, and where women’s groups have
been very active and mobilized within this transition. One must therefore be wary
of generalizing from these examples.
Nevertheless, what the cases suggest is that GRB work can be an important
supplement to work around increasing women’s political participation. In particular,
GRB initiatives can help to ensure that women’s participation means more than a
change in the profile of political decision-makers. It can do so by increasing the
representatives’ understanding of what gender-sensitive policy means and helping
them focus on a key policy without which any other policy will not be effective,
namely the budget.
of definite figures for the following year, with indicative figures for the following
two or four years. This approach to budgeting should provide greater opportuni-
ties for parliamentarians and other players to have an impact on the actual budget
figures, as there is a longer time in which to advocate for or against government
plans. In practice, the opportunities do not seem to have been utilized, either by
GRB players or others.
Finally, parliaments in developing countries usually have little, if any, research
capacity. At first glance this provides an obstacle to GRBs. However, it can also
provide opportunities as parliamentarians who lack their own research resources
might be more prepared to receive input from outside and, in particular, from
NGOs and other gender advocates. Several non-governmental organizations
(NGOs) have taken advantage of this gap. Perhaps the most surprising of these is
Zimbabwe, where, despite the political turmoil, the Zimbabwe Women’s Resource
Centre Network (ZWRCN) has on several occasions been invited by cross-party
parliamentary committees to provide input based on its GRB research into key
ministries.
While some might be concerned about people who are poorer and more
disadvantaged than themselves, they are less likely than leaders from less advantaged
backgrounds to have an understanding of the real needs of those involved. Similar
regulations about who can stand for election are less common at the local level.
A focus on the local level is also important when – as is often the case – key
services such as basic health and education services are decentralized. In these
cases the national level’s responsibility is usually restricted primarily to policy-
making. It is at the local level that the impact of the implementation – or
non-implementation – of the policies will be felt. It is thus at the local level that
citizens – and particularly women – will often be most open to mobilization
around the needs of their children and other family members.
Demystification of budgets
The section on participation above focuses on participation at the official decision-
making level. Many donors are primarily interested in initiatives which focus on
official decision-makers, whether in the bureaucracy or parliament. However,
some GRB initiatives (for example, South Africa, Philippines, Indonesia, India)
have also been interested in extending participation among civil society, and in
demystifying budgets.
the lead organization contracted researchers from the university to do the bulk of
the research. In Bacolod City, in contrast, after initial hesitation on the basis that they
were not researchers, DAWN itself conducted the research.
DAWN’s approach was the most effective both in terms of producing a gender-
analysis of the city’s budget and in terms of providing a conducive environment
for implementation of recommendations. What was particularly interesting was
the extent to which the GRB research exposed Celia Flor, an energetic councillor
of several years’ standing, to new information and understanding of how the
council worked both in gender terms and more generally. Flor and her co-researchers
were also visibly empowered and proud of what they had achieved in doing the
research themselves. And the fact that they had done so facilitated the subsequent
dissemination of the approach to other cities in Negros Occidental.
In South Africa, the non-governmental Women’s Budget Initiative (WBI)
was conceived from the outset as a multi-stakeholder exercise. The core group was
made up of parliamentarians and two policy-research NGOs. From the start, this
core group involved people outside the core as central players. In the first three
years, the initiative produced research chapters on each of the departments
(equivalent to ministries in other countries) of the national government, as well as
on several other topics such as public-service employment, taxation, the budget
process, and intergovernmental fiscal relations (the allocation of money raised at
one level of government to other levels for spending). Members of the core group
coordinated the research and provided back-up to the researchers, but did very
little of the research and writing themselves. Instead, they drew on people from
other NGOs as well as academics to do the research in their areas of expertise.
In addition, they set up a reference group consisting of at least one government or
non-government person knowledgeable about each sector being researched. These
were generally people who were too busy to do the research themselves, but who
were considered important sources of knowledge or access to information, as well as
key people to whom the WBI wanted to spread its new approach to understanding
budgets.
There were several reasons why the inclusive approach was adopted by the WBI.
First, the initiators recognized that analysis of budgets from a gender perspective
requires a multidisciplinary approach and knowledge of several different fields.
They felt that it would be easier for people who already understood the gender
issues in a particular area to acquire the necessary skills to analyse budgets than for
people with budget analysis skills to acquire adequate knowledge of gender issues
in each sector. Second, and related, the initiators hoped to spread the relatively
uncommon skill of budget analysis so that NGO workers in different areas would
have this extra ‘tool’ to strengthen their advocacy and engagement with govern-
ment. Third, the approach had the by-product of generating relatively widespread
knowledge of, interest in, and buy-in to the project. The interest and buy-in was
increased by a large public launch of the first year’s research which was opened
by the then Deputy Minister of Finance. Fourth, the two core NGOs did not see
themselves primarily as advocacy organizations. Instead, they saw themselves as
assisting others to obtain budget-related skills and information which they could
Expectations versus Realities in Gender-responsive Budget Initiatives 337
use in advocacy. The parliamentary partners were the first target for such skills and
information. The researchers and their organizations were a second target which
could increase spin-offs.
The inclusive approach was successful in spreading awareness and buy-in. It also
resulted in more spin-off work than in most other countries. Many of those who
acted as researchers commented on how empowered they felt in having been able
to do the budget analysis. A few continued to do similar work outside of the WBI.
Several other NGOs which had not been involved themselves took up the idea of
GRBs. Some of these came back to the core WBI and asked them to do the research.
A few took the work forward themselves, perhaps with back-up from the core WBI.
Most successful among these was an NGO which had previously engaged in
extensive research and advocacy on gender violence, and saw the potential of the
GRB approach as an additional tool. Still others both inside and outside government
spoke about the importance of looking at budgets, although this often did not go
beyond talk. Several universities included GRB in their gender and development
studies courses.
In South Africa the composition of the core group, which included researchers
in the NGOs and advocates in the parliamentarians, appeared to provide an ideal
mix in that the researchers could provide the parliamentarians with information
to strengthen their advocacy. In practice, however, the parliamentarians did not
use the research to the extent hoped for. The Tanzanian initiative provides further
lessons about the dangers in separating the research and advocacy roles. TGNP is
the central GRB organization in Tanzania. When TGNP initiated their gender
budget initiative (GBI) they did so in collaboration with the broader Feminist
Activism grouping of which they form part. Nevertheless, they played the main
role from the very start.
TGNP’s approach mimicked that of South Africa in many respects in the first
year. In particular, one of the first activities was to commission research into the
budgets of key ministries, as well as into the budget process. For each of the papers
TGNP commissioned a team of at least two people. Typically, one was an academic,
while the second was someone from inside the relevant ministry. Instead of
coordinating and editing the work themselves, TGNP commissioned a feminist
political scientist at the university to do so.
TGNP’s choices reflected its own perception of itself – as an activist organization
rather than a research or academic one. The choice of academics reflected the
smaller number of other civil-society organizations and individuals who might
be persuaded to do this work. One weakness of the approach was that TGNP had
less knowledge of the detail of the research. This is an important weakness in
advocacy with ‘opponents’ such as government technocrats as the latter use tech-
nicalities as one of their key weapons. In subsequent research and related work,
TGNP has increasingly tried to participate more in the research process.
Conclusion
The rapid growth in the number of GRB initiatives over the last few years testifies
to the potential which many see in these ventures. Yet the discussion above
Expectations versus Realities in Gender-responsive Budget Initiatives 339
suggests that often the initiatives do not fulfil all the roles which are attributed to
them. The chapter has tried to illustrate that one of the reasons for this apparent
failure is the great diversity in GRB initiatives. As a corollary, it implies that
non-fulfilment of each and every goal should not be seen as a failure. Whether an
initiative can play a particular role depends on the actors, their goals, their under-
standing, and the activities they undertake, as well as the political and social
context in a particular country. Those promoting GRB work would be more helpful
if they promoted it as a tool that can be used at many different stages in the policy-
making process, by many different players in many different ways to advance
many different causes in addition to the broad cause of gender equality. A single
initiative cannot, and should not, aim to be all things to all people. GRB work is
not a panacea for anything.
Bibliography
Blackden, C.M. and C. Bhanu (1999) ‘Gender, Growth and Poverty Reduction’. Technical
Paper No. 428, Washington DC: World Bank.
Budlender, D. and G. Hewitt (eds) (2002) Gender Budgets Make More Cents: Country Studies and
Good Practice. London: Commonwealth Secretariat.
Budlender, D. and G. Hewitt (2003) Engendering Budgets: A Practitioners’ Guide to Understanding
and Implementing Gender-responsive Budgets. London: Commonwealth Secretariat.
Budlender, D., D. Elson, G. Hewitt and T. Mukhopadhyay (2002) Gender Budgets Make Cents:
Understanding Gender Responsive Budgets. London: Commonwealth Secretariat.
Fontana M. (2002) ‘Modelling the Effects of Trade on Women: The Case of Zambia.’ Working
Paper No. 155, March 2002. Sussex: Institute of Development Studies.
Fontana, M. and A. Wood (2000) ‘Modelling the Effects of Trade on Women, at Work and at
Home’. World Development 28(7): 1173–90.
Krafchik, W. and J. Wehner (1999) The Role of Parliament in the Budget Process. Cape Town:
Institute for Democracy in South Africa.
Meena, R. and R. Mhamba (2003) A Gender Study on the Tanzania’s Government Revenue and
Taxation. Dar es Salaam: Tanzania Gender Networking Project on behalf of Gender Macro
Policy Working Group.
Pillay, K., R. Manjoo and E. Paulus (2002) Rights, Roles and Resources: An analysis of Women’s
Housing Rights – Implications of the Grootboom Case. Cape Town: Women’s Budget Initiative.
Reeves, H. and C. Sever (2003) Gender and Budgets: Key Texts, Case Studies, Tools, Guides and
Organisations. Sussex: BRIDGE Support Resources Collection, Institute of Development
Studies.
Sarraf, F. (2003) ‘Gender-Responsive Government Budgeting’. Working Paper No. 03/83,
Washington DC: International Monetary Fund.
Sharp, R. (2003) Budgeting for Equity: Gender Budget Initiatives within a Framework of
Performance Oriented Budgeting. New York: United Nations Development Fund for Women.
Smith T. (2002) Women and tax in South Africa. Cape Town: Women’s Budget Initiative.
United Nations Development Fund for Women (UNIFEM) (2001) Annual Report. New York:
UNIFEM.
Index
abortion, 77, 78, 79–80, 83n20 Basic Income Grant (BIG), 116, 120–6
Adato, M., 57, 63n28 impacts, 123–4
Adjustment with a Human Face (Cornia et al. Bates, R., 28
1987), 47 Baude, Annika, 168n5
adult literacy, 2, 4, 121 Beijing, China, 275
adult-worker family model, 173–87, 219 Beijing Platform for Action (PFA), 32, 53
in EU policy-making, 174, 175–8, 180 benefits see social benefits
Africa, 19, 28, 30, 33, 245, 247 Berer, M., 249
health sector reform, 237, 238–9 biological determinism, 159
see also sub-Saharan Africa; and individual see also maternalism
states Bojer, H., 186
AIDS pandemic, 26, 116, 121, 218, 229, 249 Bolivia, 2, 3, 4, 5, 6, 11, 45, 194, 196, 207
care implications, 218, 224, 230, 250–1, 328 1980s economic crisis, 200
orphans, in Africa, 251 kin networks, 209–10
and social protection, 230 labour informalization, 200, 201, 202–4,
in sub-Saharan Africa, 249 212
in women, 121, 249 poverty, 201
Albertyn, C., 127n6 structural adjustment, 200, 201
alimony benefit, 70, 73, 76, 80, 83n12 tin mining, 200
All China Women’s Federation (ACWF), 270 unemployment, 200
Alma Ata declaration, 1977, 249 urban, 201, 209–10, 211
Angeles City, Philippines, 332, 335–6 Bradshaw, S., 59
An-Naim, 101 Brazil, 8, 44, 45, 51
anti-poverty programmes, 58–62, 198, 213 1988 constitution, 51
and gender, 58–60 Ceará province, 53
see also poverty alleviation; and individual Bremer, Fredrika, 155
programmes Britain see United Kingdom
Antonopoulos, R., 206, 211 Brock, K., 48–9
Arat, Y., 106 brokering, 153, 162–4
Argentina, 2, 3, 4, 5, 6, 8, 43, 44, 62n8, 224, boundary-making, 153, 155, 156–8
312 budgeting, 29–30, 316, 323, 333–4
neoliberal reforms, 312 analysis, 337–8
taxation (reform), 304, 305–6, 307, 308, demystification, 335–6, 338
310, 311, 313, 315, 317 and economists, 337–8
Asia, 8, 28 gender-responsive see gender-responsive
1997 economic crisis, 14, 132, 144 budgeting
see also East Asia; and individual Asian medium-term expenditure frameworks
states (MTEFs), 333–4
Asia Foundation (TAF), 332, 335 performance-based, 325, 333
asset pawning, 211, 212 policy changes, 324
Australia, 11, 326 research, 334
Austria, 11 Byanyima, Winnie, 330, 332
340
Index 341
pensions – continued poverty alleviation, 1, 15, 19, 21, 28, 48, 50,
racialization, 113 52, 54–5, 117, 123, 198, 218, 311, 317
reforms, 18–19, 227 Attacking Poverty (World Development
widow’s, 161–2 Report 2000/1), 50
see also social benefits; and individual states see also anti-poverty programmes
Personal Status Code in North Africa and poverty line, 45, 72, 119, 123, 201, 205, 213n4
the Levant, 92 Povey, E. R., 98
see also Islamic law privatization, 2, 15, 72
Peru, 45 of education, 288–9, 291–3
Peter, F., 184 of health care, 245
Philippines, 97, 232, 326–7, 332, 335 of pensions, 18–19
Development through Active Women of social services, 2, 48, 113–14, 119, 198
Networking (DAWN) Foundation, PROGRESA/Oportunidades programme,
332, 336 Mexico, 20, 25, 53–8, 61–4
piece-work, 11, 202, 206, 222 accountability, 56
Pichetpongsa, A., 206 concept of co-responsibility, 55, 56
Poland, 2, 3, 4, 5, 6, 22, 26, 30, 34, 68–9, evaluations of, 57–8
70–1, 73–80, 83n6 gender impacts, 56–7
economic reform post-socialism, 71–2 success of, 55
family benefits, 70–2, 73: compared with targeted benefits, 54–6
Czech Republic, 75–82 public expenditure, 29, 322, 329
Plenipotentiary for the Equal Status of see also social spending
Women and Men, 80 public sector
Solidarity movement, 71, 78 Africa, 245–6
Polanyi, K., 1 as care provider, 249: see also care(work)
political liberalization, 30 in China, 260
see also democratization employment, 19, 29, 92, 165, 200, 245:
political participation, 2, 5, 6, 31, 32–4, 36, decline, 194, 246; low morale, 246;
80, 90 preferred by women, 165–6, 315
by women, 2, 5, 31, 32–4, 36, 80, 99, 110, funding, 29, 246
145–6, 152–3, 155, 167, 168n12, 244, gendered hierarchy, 245–6
332–7: as MPs, 110, 130, 143, 145–6, institutions, 2, 29, 252–3
244, 326, 330, 331 public works, 115, 120, 121, 133
political parties, 32–4, 36, 311
see also under individual states Qavam, Ahmad, 95
population growth, 44 Quirós Viquez, A., 59
postcolonial states, 28 Quito, Ecuador, 210
post-socialist countries, 68–82, 83n1, 160
see also Central Europe; Eastern Europe racism, 111–12
poverty, 33, 45, 198, 199 apartheid regime see under South Africa
alleviation see poverty alleviation Ranson, M. K., 225, 230
children in, 119 recession, 1980s, 47
coping strategies, 208 religion
definitions, 213n4 Islam see Islam
discourse on, 1, 15, 26–7 Roman Catholicism see Catholicism
as enduring phenomenon, 48 and the state, 92
extent of, 62n15 ultra-conservative, 249
family, 76–7: see also under family rentier economies, 89–90
and gender, 33, 121 reproductive health see under healthcare
increase, 10 reproductive rights, 78, 79, 248, 249
of single parents, 166 see also abortion
urban, 208–9 rights, social and political, 32, 44, 46–7, 51,
variable see earnings variability 87, 100, 110, 117
see also inequalities legislation, 43, 46
350 Index
state, role of the, 1, 14–15, 28–9, 48, 51, 52 Tanzania, 2, 3, 4, 5, 6, 239, 246, 251
state institutions, 28–30 feminist organizations, 337
state socialism, 7, 30, 33–4, 79, 81, 83n1 health care, 239, 241, 245: AIDS, 249, 251
Standing, H., 240, 241, 249 poverty, 239, 251: reduction, 244
Stephens, J., 159 public expenditure review, 244
structural adjustment, 9, 14, 48, 305, 309 public-sector workers, 246: low morale,
impact of, 14, 28, 47, 53, 198, 201 246
subcontracting, 202, 206 social health insurance, 242, 243, 244:
Subrahmanya, R. K. A., 230 community-based, 16, 242, 243
sub-Saharan Africa (SSA), 10–11, 224, 248 women politicians, 244
AIDS, 224, 248, 251 Tanzania Gender Networking Programme
democratization, 29 (TGNP), 329, 337
health sector reform, 16, 248 Tanzi, V., 304, 305
see also individual states targeted social benefits, 14, 15, 19–21, 50,
subsidies, 47, 112, 125, 127n2, 319n9 54, 200, 244, 249
Sweden, 2, 3, 4, 5, 6, 7, 10, 34 effectiveness, 19
Commission on Family and Marriage, 162 as a means of redistribution, 47
feminism, 152–3, 154, 155, 156–8, for women, 20, 53, 325
167 taxation, 23, 54, 138, 160, 199, 301, 329
folkhemmet, 154–5 avoidance/evasion, 302–3, 307, 310
gender equality, 152, 161, 164, 168n8 and development, 310–11
Group 222, 159–60 direct, 301–2, 304, 307, 308, 313, 315,
immigrants, 163, 166 318, 319n3
marginalization, 153 domestic, 305
Morgonbris journal, 154, 156 enforcement, 310–11, 318
parent-worker-citizen model, 161–2 and gender, 312–17, 329
political history, 153–64: Cossack on incomes, 307–8, 309, 313
Election, 154; Slatsjöbaden indirect, 304, 305, 307, 308, 309, 310,
agreement, 158 315, 316, 317–18: see also value-
political participation by women, 155, added tax (VAT)
167, 168n12 legislation, 315, 318
political parties, 160, 167–8: Centre Party, for married couples, 313
162; Conservative Party, 156; progressive, 317–18
Farmers’ Party, 156; Liberal Party, rates, 309, 315
156, 157, 160, 162; Social Democratic reform, 301, 305–19
Party, 153–5, 156, 160, 161, 163 regressive, 317
social policies, 152–3: Women’s Life and social security, 304, 308, 310, 313–15,
Work, 161 318–19
trade unions, 156, 163–4 structures, 302–5
unemployment, 155–6, 166 tax credits, 324, 329
wage structures, 151 and total revenue, 302, 303, 305, 306, 307
welfare state, 151–68: expansion, 153, on trade, 305, 309, 315
159–60; retrenchment, 168 under-, 301–2
as women-friendly state, 151–3, 163, 164 Taylor, Viviene, 122
women’s organizations, 154, 156, 162, teenage pregnancy, 248
167: Fogelstad Group, 155; Fredrika Teichman, J., 311
Bremer Society, 168n1; National Tendler, J., 30, 53
Association for Housewives, 156, Thailand, 3, 4, 5, 6, 194, 205–6, 211
168n1; Professional Women’s ‘30 baht’ health insurance scheme, 230,
Organization, 157; Redstockings, 167; 231–2
Social Democratic Women’s Homenet Thailand, 231
Organization, 168n1 trade unions, 9, 33, 46, 95, 121
women’s right to work, 155–7 health workers, 247
worker state, 156 industrial action, 199
Index 353