This book is based on research supported by the International Development Research Centre (IDRC), Canada. IDRC provided not just the funding, but the space and flexibility for the project to evolve in many directions. The editors would especially like to thank Rohinton Medhora of IDRC, whose interest and encouragement ensured the success of this initiative. Thanks also to our partners – both individuals and instructions that have contributed to this book – for ther support in this endeavour.

Tracking Gender Equity
Under Economic Reforms

Continuity and Change in South Asia


International Development Research Centre

Ottawa • Cairo • Dakar • Montevideo • Nairobi • New Delhi • Singapore





Non-Conventional Indicators of Gender Disparities Under Structural Reforms



Household Response to Gender Issues: A Survey on Households of Female EPZ Workers in Bangladesh



Status of Women Under Economic Reforms: The Indian Case



Economic Reform and the Status of Women in Nepal



Gender and Adjustment Policies: Evidence from Pakistan



Continuity and Change: Women Workers in Garment and Textile Industries in Sri Lanka



Family Structure, Women’s Education and Work: Re-examining the High Status of Women in Kerala



Trade, Gender and Employment Issues



Gender in a macroeconomic Framework: A CGE Model Analysis



Towards Integration? Gender and Economic Policy








What is New About ‘Gender Bias’?

It is unlikely that there will be much difference of opinion about the fact that along with class, gender continues to be a basic criterion that structures most societies around the globe. In South Asia, the region which provides the main context of the research presented in this book, caste divisions combine with class and gender to generate what must be among the most complex and highly structured, and among the most hierarchical of societies in the world today, patriarchy being only one of the major principles of social stratification in them. Yet these are far from being stagnant societies or polities, sheltered and secluded as it were, from economic and political influences from the rest of the world. For over half a century now, the subcontinent has harboured the largest democracy in the world. And over the last two decades or more, inward-looking autistic growth policies and central planning practices of the ‘socialist’ era in most of the countries of the region have been gradually giving ground to more open and liberal market-oriented economic reforms. It is legitimate to expect that these changes would have had some influence on gender relations in these societies. However, unlike sex, which is biological, ‘gender’ is a socially constructed category. Depending on the context, it may manifest itself along different dimensions in a whole range of different ways. In the language of econometrics, one could say that ‘gender’ is a ‘latent’ variable, in principle unobserved and unobservable. What one can at best hope to capture in terms of observed categories, are some of its overt manifestations. The challenge therefore is to figure out how best to identify and monitor ‘gender bias’, especially under such complex dynamic conditions.

Our modest attempt here has been to clear up some cobwebs that we feel have gathered over the years around the issue of the identification of gender bias. We examine issues that depend critically on the analysis of the quantitative indicators of such biases. UNDP did a seminal service to the community of gender researchers and those working on human development issues in general by putting forward the concepts of ‘gender development’ and ‘gender empowerment’, and developing the corresponding indices. This served as the first step in putting in place a system of assessment of the gender situation across countries and over time. The inherent limitation of the exercise was acknowledged from the beginning by the group of experts who helped develop these indicators: this is a limitation that is common to all indices. There is, for example, the question of which variables to choose, how to decide on the weight assigned to each, and what procedure to choose for aggregation.

However the easy availability of large data sets and of computer technology to analyse them, has spawned a considerable body of gender research that has gone on to mechanically expand the UNDP exercise in every conceivable direction, without paying much attention to the context within which such exercises are being carried out. Although the very obvious forms of gender discrimination may not be difficult to identify under any situation, the more subtle manifestations of bias may vary considerably from one context to another, and new mutations might well emerge under changing conditions. Any analysis of gender bias, especially in complex socio-economic situations, must therefore pay close attention to the context. We attempt to do this exercise within a context that has both a relatively static and a relatively dynamic dimension. The static aspect is that of South Asian societies, stratified by class, caste, ethnicity, religion and location. The relatively dynamic aspect within which these structures are located for the purpose of this set of research is defined by the processes of change that have been put in place in the macro economic policy regimes in these countries through recent economic reforms programmes.

Research from around the world seems to suggest that one set of indicators of women’s secondary position in society that seems to be more universal, in the sense of cutting across all contexts, is violence against women, or more appropriately perhaps, credible threats of such violence, along with its actual perpetration. In general, research on violence has been more in the nature of being

qualitative, oftentimes even anecdotal. Much of it has been carried out by activists, or by feminist researchers from disciplines such as sociology or social anthropology. For the most part, this set of gender research has generally not had much interaction with GDI-induced quantitative gender research, which is a pity since both groups of researchers are interested in the same set of issues, i.e., the phenomenon of gender bias in society.

One of the things that this book attempts is to provide a forum of interaction between these two genres of gender researchers through the introduction of some ‘non-conventional’ indicators of gender bias designed to measure gender-related stress, anxiety and violence, which can be then be mapped onto the other, more conventional ones. Apart from bringing in the dimension of gender-based mental health issues within a tractable form of inquiry, which is important in itself, we feel this will serve to provide a common platform for a wider group of people interested in gender issues to interact with one another.

Considering that the dynamic framework of the inquiry is defined by the context of recent economic policy reforms, the proposed analysis of gender indicators has to be couched within that macro-economic framework. This leaves open the question of how best to trace the impact of changes in economic policy initiatives at the macro level onto gender relations in society: something that is played out mainly in the arena of the household and the community, and to a lesser extent perhaps, in the workplace. The direct line of transmission of such policy changes has to go through changes in the economic environment of the family, and secondarily through the social ramifications of these changes on the family, the community and the workplace. Ideally a macro model with a non-unitary model of the household sector reflecting a situation of intra-family inequalities and co-operative conflict embedded within it, could have provided the framework of such an inquiry. However in the current state of knowledge, there is no such model available, nor indeed are the requisite data or estimation techniques. The option that has been adopted here, thus, is a somewhat piecemeal one. We start by identifying areas or sectors of the economy which have been known to be directly impacted by reforms, and design household surveys to do the analysis within such contexts. Our book therefore focuses on the results of coordinated household surveys that were carried out in

four countries of South Asia in the households of women workers working in Export Processing Zones or Export Oriented Units. Each of the countries in our study is known to have undergone expansion under the new policy regimes. In the second phase of the project, a further set of surveys were initiated in households of workers who have been retrenched under recent reforms. It is hoped that by piecing together such evidence from diverse contexts, we will be able to understand the nature of changes in gender relations that are taking place in these countries and to see what impact economic factors may have on changing basic social conditions. The household level analysis is supplemented by research at the sectoral level, as well as research on gender modelling at the household and macro levels.

Overall then, this book hopes to supplement the already rich literature on gender by attempting to put in place a consistent framework for gender analysis, by demonstrating the importance of identifying the context of such analysis, and by highlighting the necessity of differentiating ‘gender’ per se from its various ‘indicators’. It also seeks to put in place a new agenda of gender research by expanding the existing set of gender indicators to include those such as gender-related stress, anxiety and violence. The viability of such an approach is demonstrated through a coordinated set of household surveys designed for inter-country comparisons. The meso and macro level studies serve to complement the household surveys in an attempt to close the loop. Although it has not been possible to do this here, it is our hope that together, the various components of this body of research will provide the ingredients for integrated household and macro level gender analyses in the future.

Gender: The Indicators and the Context

‘Gender’ as separate from its ‘indicators’

Any indicator, however good and appropriate it may be, can at best be an approximation of what it intends to measure. A set of indicators combined in an index has the advantage of representing the multiple dimensions of the phenomenon, but involves additional problems concerned with the choice of appropriate weights and methods of aggregation. Identification of the instrument(s)

of measurement with what is being intended to be ‘measured’, can create problems. In the case of a complex and latent variable such as gender, the problems are restricted not just to the set of limitations that are common to all index numbers, but can in fact go much deeper. This may result in serious and very inappropriate conclusions.

Gender is a socially constructed phenomenon, the ramifications of which are primarily played out in the arena of inter-personal relationships, which indeed have powerful social and economic implications. Although overt discrimination is not difficult to identify, depending on the time, place and the context, second order or covert gender bias can have multifarious and complex manifestations. As a category, it is unobserved and unobservable, and like a chameleon, it might manifest itself differently under different contexts. Thus whatever indicator, or set of indicators one might choose to characterize gender relations in a particular society, unless properly contextualized, these could end up sending the wrong signals.

There are numerous instances of how things can go wrong. A study by the Helen Keller Foundation in Bangladesh showed that, in a sample of five thousand poor households in rural Bangladesh, nutritional levels of young girls were lower than corresponding figures for young boys, but that as household incomes and/or the size of land holdings increased, the differences became smaller. Projecting this trend, one could argue that beyond a certain level of household incomes, there may be no observable gender differences in nutritional attainments of young boys and girls. However even if it were actually so, there is no reason to take this as an indication that there is no gender bias in relatively more affluent households in rural Bangladesh. The bias could very well show up in many other forms, in more stringent restrictions on the mobility of young girls in such households, for instance.

Nor is it necessarily the case that all indicators of gender development will move together under all contexts. Not merely that they need not necessarily be collinear; under some circumstances, there may even be significant non-linearities in their relationships. An example of the inappropriateness of using non-contextualized indicators of gender development can be cited from the area of economics. Measured female labour force participation rates (FLFPRs) are, on an average, low in the South Asian region.

Although there are regional variations, cross sectional data from India show that FLFPRs are relatively high for illiterate women from poor households. They are also high for women with high enough literacy levels. But there is a dip in the curve in between the two extreme levels of female literacy. This means that if female literacy rates are monotonically related with household incomes, which indeed they are for good reasons, the relationship between household income and female FLFPR will also be U-shaped. FLFPR will decrease as household incomes rise for initial ranges of incomes, but after a certain level of income is reached, the two rise together.

In the context of the caste-ridden society of India, this non-linearity has been explained by a phenomenon termed the Sanskritization process which is defined as the tendency, among lower caste households, to emulate upper caste social norms and practices, one facet of which is ‘protecting’ their women from paid work outside the home. Very poor households, which also have a disproportionately high representation of households from lower social strata, cannot afford to not send their women for paid work. But as households move up the income scale, there is an urge to move up the social ladder as well, and one of the ways in which they can establish their upward social mobility is by emulating the social practices of higher caste households. When income levels rise sufficiently high, and women become relatively highly educated, the relative benefits of women working outside the home supercede the perceived social costs and FLFPRs go up once again. Without understanding the social context, using one or the other indicator, or a combination of both as a measure of women’s status especially for households that are to the left of the turning point in the curve, could clearly be problematic.

Thus the conventional indicators of gender bias such as gender differences in health and nutrition, education and skill development, political participation and decision-making powers, need to be put in their proper contexts before they can be used sensibly as indicators of gender bias.

The ‘static’ context: social systems and gender roles in South Asia

South Asia is believed to be one of the most gender-insensitive regions in the world. The overt manifestation of this insensitivity,

or bias, often starts at birth, and sometimes even before that, and can be traced through a whole range of indicators in diverse areas through the life cycle of the child. An average girl child in the Indian subcontinent may be discriminated against in the natal family in terms of nutrition and health care, as well as of educational and skill development opportunities. The socialization of gender roles starts very early in South Asian culture. This is evident in the very different and distinctive time-use pattern of young girls and boys in South Asian families. Girls are taught domestic skills from a very young age, while young boys are generally spared this.

A young girl in a representative household in the region will be conditioned from her childhood by rigidly enforced norms of womanly behaviour and bound by a hundred and one restrictions designed to prepare her for life after marriage. Marriage is seen as being virtually universal for young women, and once a woman is married, she is likely to be placed at the bottom rung of the internal power structure within the marital household. Insufficient dowry has been the cause of much harassment, mental and physical torture, even death, for an appallingly large number of Indian brides. Girls are often married off at an early age, and the primacy of the reproductive role that is ordained for them by widely prevalent social norms, translates into their being pushed into child-bearing from a very young age, and at frequent intervals. Preventable maternal mortality in the region is among the highest in the world.

Distinctive gender roles and responsibilities for men and women are ingrained in the South Asian psyche in a manner which is probably far stronger than in the West. The dominant Hindu tradition in this respect is based on the precepts of the Manusmriti which looks upon the female as a complement to the male in the social order, but one who is distinctly subservient to him. According to the precepts of Manu, a woman has to be under the father in her childhood, the husband in her adulthood and the son in her old age. Similarly Islam, the other major religion in South Asia, places women squarely in the domestic sphere, under the protection of the male. The birth of a male child is usually greeted with much jubilation, while the birth of a girl child rarely evokes much happiness. After all, a girl child will go away to another household, she will need dowry to be married off and it is no wonder that she

is basically looked upon as a burden. The male child, on the other hand, will stay with the family, look after parents in their old age, and according to Hindu beliefs, perform the last rites of the parents to deliver them from hell. It is immaterial how much these beliefs and expectations are realized in real life. The beliefs persist and in many communities, the birth of a girl is deemed a minor disaster, sometimes leading to female infanticide.

Much of South Asian society, especially in India, Nepal, and to a certain extent Sri Lanka, is also ridden with caste hierarchies. It has been argued that the caste system had evolved from the inner logic and dialectics of Vedic philosophy and it is meant to ensure a holistic organization of the social system. In practice, this has hardly been the case. Caste hierarchies have split South Asian society in terrible ways, and the results of this can be seen in the sectarian politics of the region today. Although, strictly speaking, caste is a Hindu phenomenon, the hierarchical social structure it postulates has influenced all other major religious groups in the subcontinent. Historically, caste hierarchies have, to a large extent, been tied up with class hierarchies, insofar as upper castes have always had easier access to resources, opportunities and assets.

The high degree of social stratification along class and caste lines has had important implications for the gender dimension of society as well. Gender roles are rigid in the region, just as caste and class hierarchies are deeply entrenched in the social fabric. As well, being a region which houses the majority of the world’s poor, endemic poverty is a fact of life for a large fragment of the South Asian economy. Gender discrimination and poverty are the two major factors that determine the context within which the average South Asian woman spends her life.

While the picture drawn above is not universal, and there are significant variations between different socio-economic groups within each country as well as between different countries, the above situation does characterize the attitudes and behaviour patterns of uncomfortably large segments of the population in the region. In this context, one question that continues to crop up is the question of what, if any, has been the gender impact of economic reforms that all countries in the region have recently embarked on.

The ‘dynamic’ context: economic reforms and gender

Ever since structural economic reforms were adopted in the region, apprehensions have been expressed regarding the impact of such reforms on poor and marginalized sections of society, particularly women. A number of reasons have been cited for such views. These include arguments such as those regarding an inherent bias against women in the operationalization of structural adjustment programmes, insofar as such market-centered programmes take a partial view of the total labour allocation process by ignoring altogether women’s reproductive labour within the household (Elson 1994). There are also other general reasons cited that are likely to hurt both men and women, such as the probability of downplaying labour rights concerns and job security requirements, privatization of publicly owned concerns leading to higher insecurities of private sector jobs, globalization leading to erosion of level playing fields of domestic concerns etc. Because women are concentrated at the lower ends of the job spectra in the labour market, these developments are likely to hurt them more than men. Apart from this, there are other likely fallouts which may do the same. These include adverse changes in the household economic environment, leading to greater penury and more hardships at the household level which women, as home managers, have to face, reduction in state responsibility in social sectors such as health care and care of the aged leading to an increased workload for women within the family, and so on. There is also the question of making adjustments of various kinds such as in the increased involvement in wage labour to bridge the gap in household economic needs, budgetary adjustments brought about by increases in prices, and livelihood restructuring adjustments induced by instabilities propelled by adjustment processes, and so on (Ghosh 1993, Mukhopadhyay 1998).

Most of these arguments are based on a priori reasoning. Not much information exists in the way of standardized evidence to either refute or corroborate the hypothesis that structural adjustment programmes have had a significantly gendered impact, and if so, what has been the precise nature of that impact. The important thing to do would be to clearly specify the macro framework and the structure of relationships at the macro level, to identify the channels of transmission, and to trace the gender dimensions of the impact

of change at the macro level to the household and the community in terms of the standard indicators of gender bias under identified contexts.

The ‘Non-Conventional’ Indicators of Gender Bias

Gender-based stress, anxiety and violence

We have argued earlier that gender discrimination can have very culture-specific manifestations, and unless one is familiar with the cultural context, standard indicators of gender bias may miss out on a lot of subtle nuances. However a whole range of micro studies from around the globe suggests that there is one set of potential ‘indicators’ of gender bias that seems to cut across social and cultural contexts. These have to do with the stress and anxiety of various kinds that women go through to a greater or lesser extent, as a consequence of being born women. In South Asian societies, the region which these studies relate to, an average young girl grows up with an abiding fear of unwanted sexual attention, or worse, molestation or rape. The fear and the anxiety of losing her sexual ‘purity’ is ingrained in young girls since early years through a process of social conditioning that puts a heavy premium on it. The important thing to note is that in one way or another, this phenomenon cuts across all social and economic strata of society, reflecting to a certain extent, the objective conditions of personal safety that women live with in this region. This is especially so for those women who are seen to be without a ‘protector’ (an adult male — such as a father, a husband or a son, for example, single women who are young and live alone, divorced women or widows), and/or the ‘fallen’ woman (i.e., someone who is perceived to have broken the strict boundaries of ‘womanly’ behaviour by going against the prescribed moral norms of society; such a woman becomes easy prey). Many methods are deployed to protect the ‘purity’ of women inside the family. Strict restrictions are imposed on their mobility, and on outside exposure, as for instance in the case of Muslim women who are forced to wear ‘purdah.’ In many communities in rural India it is common practice to marry girls off at a young age, and until they are safely handed over, like a piece of property, to their rightful owners, (i.e., the husband and the in-laws) their virginity must be protected. A woman victim of rape

is seen as someone who has lost the most valued thing in her life; her purity and her honour, and is now destined to a fate worse than death. She is doubly victimized by the insensitive law enforcement machinery and by society at large.1 Even if a woman does not personally experience such extreme forms of violence, she can count herself lucky if she can manage to go through life in this part of the world without experiencing some kind of harassment on account of her gender. The unpleasantness and the anxiety of having to go through such experiences is enough to create an endemic condition of stress as Taslima Nasrin, the Bangladeshi writer, has shown in her novel, Lajja.

The second aspect of this range of issues is domestic violence, and there is enough evidence now to suggest that violence against women, and the credible threat of such violence within the family, is far more widespread than was believed to be the case earlier. Nothing brings out the secondary status of women within the family more than unilateral violence, and a society which condones it and resorts to it as a method of establishing the superiority of the male, is clearly gender-biased even if the women in such societies are well taken care of in terms of nutrition, health care and education. It stands to reason therefore that if it were possible to get a set of indicators that could measure the extent of stress, anxiety and violence that women go through qua women in different situations, it would be useful to have them analysed along with the standard, or conventional indicators of gender bias in social, economic and political domains.

However anybody who has worked in the area of domestic violence knows that reliable information on the subject is very hard to come by and that even when they are available, such data are likely to have serious reporting biases. Apart from this, one must remember that actual perpetrated violence is only the tip of the iceberg. If one is looking at violence primarily because of one’s

1The Home Minister of India, while debating proposed amendments to rape laws in the Indian Parliament, had recently recommended the death penalty for rapists by citing the reason that a raped woman suffers a ‘fate worse than death’. Under the Hudood laws of Pakistan, if a woman wants the legal system to punish a rapist, she has to furnish two male eye-witnesses who would testify in her favour. If she fails to do that, she will be charged with perjury and will face the death penalty. (Mukhopadhyay: In the Name of Justice)

interest in issues centring around women’s status, then an equally potent indicator should be not just the actual recorded cases of perpetrated violence, however correctly that reporting may have been done, but also the credible threat of it, and that is something that may be perhaps harder to monitor.

The GPN project sought to address this set of issues by bringing into the fold of conventional gender research, the psychological dimension of gender as well. Two measures, GHQ (arising out of the General Health Questionnaire) and SUBI (the Subjective Well Being Indicator), were developed by extracting sections from already existing and widely tested questionnaires used by psychologists to measure levels of stress and mental well-being in varied population groups to suit the requirements of the GPN research agenda. Although it has not been possible to decompose stress and anxiety levels attributable to separate causal factors such as gender, since these questionnaires have been canvassed to the entire sample of respondents, women as well as men, in our household surveys, standard multivariate analysis has been used to identify the effect of gender on stress levels. To highlight the newness of this category of indicators in the context of conventional gender research, and admittedly for want of a better terminology, we have termed them ‘non-conventional’ indicators of gender bias, although there is hardly anything particularly non-conventional about them.

The links between conventional and non-conventional indicators of gender bias

It may generally be true that situations that are characterized by high values of conventional gender indicators, are also those that are marked by no violence and low gender-based anxieties. But this need not always be the case. Apart from the reasons cited above about why it is important to include gender related stress, anxiety and violence for their own sake in analyses of women’s status, it is also important to note that there are situations when these indicators may suggest strikingly different conclusions from those that are suggested by conventional indicators. It is when these two sets of measures tend to pull in different directions that it becomes doubly important to pay heed to the insights provided by the

non-conventional indicators. One may put up a hypothetical scenario to highlight this issue.

Consider a stylized situation where all conventional indicators move together and there is no problem of contextual complications as outlined above. Assume that there are two levels each for the conventional as well as the non-conventional indicators, i.e., ‘high’ and ‘low’, where high and low values correspond to the ‘good’ and ‘bad’ status of women respectively. There can be four possible scenarios under these conditions as outlined in the table below:



For situations characterized by options I and IV in the above table, i.e. where both the conventional and the non-conventional indicators point towards the same direction, good or bad, there is no contradiction in the diagnosis being suggested by the two, and hence no problem. But in the off-diagonal cases, say option II, where conventional indicators are good (good education, good health care etc., for women) but non-conventional ones are bad (women are subjected to many restrictions, strict rules of propriety are enforced, deviation from socially ordained norms attracts strong retribution, including violence), how does one read the situation? Can one, for instance, say unequivocally that women’s status is good? Note that if the non-conventional indicators were not considered at all, on the basis of conventional indicators alone, one would indeed have to do that, and do that unequivocally! As we will argue later, similar mistakes are committed although perhaps not in such a blatant manner, when people hold up the case of Kerala in India or indeed even Sri Lanka, as epitomes of gender development.

Similarly, consider option III characterizing a situation where there is not much stress or anxiety among women but they have far less education as well as lower nutrition and health care, than men. This can happen in very traditional societies where the norms of patriarchy are so deeply entrenched and where the social

construction of gender is so deeply ingrained in people’s minds that they are not questioned by anybody. Both men and women subscribe to them unquestioningly. Women may have much less stress in such societies as compared with others where different sets of people within the family and the community, hold different views on gender roles and on issues such as women’s place in society. In a survey of two thousand rural households spread over eight districts in the states of Uttar Pradesh (UP) and Karnataka in India, it was found that the women from UP were much less literate, had much lower exposure to the outside world in terms of labour force participation rates, and were worse off than the women from Karnataka in terms of several other conventional indicators of gender bias. However they reported much less friction within the family than the latter. It is interesting to note that the opinions of women on a range of issues, including their views on education and upbringing of girl children as compared to boys, child marriage, dowry etc., — when mapped against those of their husbands on these matters — were far closer in UP as compared to the situation in Karnataka. This suggests that the values and perceptions of women and men are more similar in UP as compared to Karnataka. It also suggests that for the very same reason, we can assume that there may be a higher degree of stress in the Karnataka households in the sample as compared to those from UP. The fact that the Karnataka women have had higher levels of education, that they have been more involved in paid work outside the home, and in general have had much greater exposure to the outside world than the UP women, in other words the very fact that they have better levels of conventional indicators, may be the main contributory factor to a higher level of stress. It is very likely that their greater exposure has led them to question traditional norms of patriarchy: a view that may not be shared by other members of their family, thereby causing tension in the family. It is little wonder that these women also report higher levels of domestic violence as compared to the women from UP.2

One can only hope that such tension is transitory, that men in these families will come round to the views of their wives, but it is not easy to figure out how soon, if at all, this may happen. The

2Mukhopadhyay and R. Savitri.

thing to note is that at the point of conducting the survey, women were found to be more stressed out in one situation as compared to another and that it is imperative to look at both conventional as well as the non-conventional indicators before one passes judgement on the status of women in a particular context.

Providing a common forum of interaction

The inclusion of non-conventional indicators of gender bias does not merely expand the scope of the analysis to encompass the psychological dimension of gender which has so far been largely excluded from the ambit of standard gender analysis, it has other potential benefits as well. One of the major fallouts of bringing in the psychological dimension of well-being in a manner in which it could fit in with the methodology adopted by conventional gender analysis, has been to provide a common platform for different strands of gender analysis, hopefully towards greater enrichment of all. Until now, research on gender violence, although quite widespread, had mostly been of a descriptive anecdotal kind which, while being valuable in itself, has generally been carried out by different sets of researchers, using different kinds of methodologies as compared to the more quantitative GDI-induced research. The use of psychological indicators of gender-based stress and mental well-being within the broad framework of standard gender analysis opens up new possibilities of multi-disciplinary research.

The Macro Framework

The inefficiency of the efficiency argument

We have aragued so far that in order to investigate the impact of macro economic policy reforms on gender at the micro level, one needs to be able to interpret various conventional and non-conventional indicators of gender bias in a sensible and co-ordinated fashion. The flip side of this requirement is that there needs to be a broader macro framework for such analysis than is generally used. The rationale for looking at a broader macro perspective for a sensible gender analysis stems from the inadequacy of the standard macro framework from the point of view of gender.

Much of the rationale for deregulation and privatization under

economic reforms stems from the argument that reducing barriers to free markets is good from the point of view of allocational efficiency. However, considering that social norms make it imperative that the burden of reproductive labour be borne by women outside the sphere of markets, the much-hyped allocational efficiency properties of markets are at best incomplete. To the extent the argument is valid, and there are many caveats to such validity, it is so only for a fraction of human labour use. It does not take into account the reproductive labour of women that takes place within the domestic sphere. The argument that domestic labour is not considered ‘economically productive’ does not absolve it from the fact that even so, the efficiency argument for marketization is essentially untenable.

Most available studies on women’s work under the new economic regimes in the region, and indeed elsewhere as well, concentrate on the low-productivity and insecurity of much of women’s employment. Our studies go beyond that dimension to see what happens within households when women avail of these market opportunities. By and large what we found is that when the external environment of the family changes, adjustments are indeed made in the overall living arrangements and labour use patterns of various members of the households. Since reproduction and home management continue to be the primary responsibility of women, even when women have full time, officially recognized market jobs, there is usually little sharing of home responsibilities across gender, and ‘working women’ generally end up carrying the double burden of work at home and outside.

Although the nature and extent of labour market attachment of men and women are very different in these countries, with official statistics showing much lower rates of market participation by women, recent evidence suggests that there is an increasing trend towards feminization of the labour force especially in certain segments of the manufacturing and services sectors, even by official counts, which generally undercount women workers. The undercounting can be explained by a number of factors ranging from the invisibility of women workers who are often home-based and/or unpaid family labour, to the fact that a number of activities women in these countries routinely carry out, and which should be seen as economically productive, are not perceived as such because

they are deemed to be mere extensions of domestic activities and therefore are not recorded as productive activities.

In the area of women’s work therefore, the challenge is to try to capture the impact of changes in the external economic environment of the family, not merely on the market involvement of women, but also the effect it has on non-market productive work which by all accounts is very high in this part of the world, and to register the changes that it brings about on the burden of domestic work on women. Unless this can be done, the efficiency argument for greater marketization is faulty at best.

Questions of data and models

In order for this to be done, one needs to have a sensible model incorporating women’s non-market productive work and their reproductive labour. There are many barriers to that possibility. The most obvious of them is the paucity of data. Most women in South Asia work in the agricultural sector and a very large majority work as unpaid family labour. They also are engaged in multiple activities, some of which are market-based, some purely domestic and some fall in an indeterminable zone in between the two. Many of these latter activities are technically considered to be economically productive if one goes by the revised definition of productive work as defined by the UN System of National Accounts (1993). However, estimates of women’s work using such definitions are hard to come by.3 There is also very little information on the nature of women’s multiple activities. Apart from this, the quality of gender disaggregated data on various other dimensions of women’s lives is also quite inadequate.

More problematic is to have a macro model that is capable of accommodating the non-market and home production activities of South Asian women. There are hardly any models that incorporate the domestic sector within a broader macro economic structure. A notable exception to this is the simple but elegant model developed by Marzia Fontana and Adrian Wood in the context of the

3Recently the Central Statistical Organization of India has produced a document based on a time use survey of women’s work in six states of the country. However it would be difficult to use this data set in conjunction with other variables needed for estimating an economy-wide macro model.

Bangladesh economy.4 It has not been possible to take advantage of this work in this phase of the research, but it is hoped that in the second phase of GPN work, it will be possible to utilize this framework.

The other aspect of the problem is how best to incorporate the household sector within the macro framework. Most models of household behaviour assume that household decision-making is a unitary process, that there are indeed no serious intra-household distributional issues. If this assumption is not valid, and there is sufficient evidence from South Asian countries that it may not be, then the household decision-making process has to incorporate the non-unitary nature of the process, reflecting a situation of co-operative conflict.5 This project included a survey of non-unitary models of household decision-making which is available on the project website.

The attempt therefore has been to approach the problem from both the micro as well as the macro end and it is hoped one has made some advance towards closing the proverbial gap between the two.

A Preview of Results

The essays included in this book constitute a subset of studies prepared under the GPN project. The country papers included in the second part of the book were specially written for this publication. They are a combination of vastly abridged versions of original country reports prepared for the project, as well as substantive new material that did not get included in the original reports. Others have been revised, in some cases quite thoroughly, for inclusion in the book.

This introductory chapter lays out the questions being explored in the GPN research agenda and the methodology that has been adopted for finding answers to them. It explains the rationale behind structuring the design of the project in the manner it has been done here, so that individual studies carried out under the project can be looked upon as components of an integrally linked

4World Development, 2000.

5Sen, 1990.

system. Together, the studies provide some of the basic building blocks, admittedly somewhat incomplete, of that structure.

The second paper in this section, written by Shobna Sonpar and Ravi Kapur, lays down the foundations of what, for want of a more imaginative vocabulary, we have called the ‘non-conventional’ indicators of gender bias. It provides an insight into the enormous complexities associated with interpreting psychological measures of stress and anxiety, more so if they have to be identified by source of causation such as gender. It reviews existing literature to show how in many societies around the world, women’s status is defined in terms of their sexual relations with men, and how ‘the ideology of power and subordination’ structures the construction of gender identity in a society. The paper dwells upon the culture-specificity of gender roles and underscores the fact that control over women’s sexuality is a key indicator of women’s status in South Asian societies. The writers emphasize the need for rigorous gender analysis, not only for women but also for men, in order that ‘egalitarian feminist goals’ can be reached. The essay explains the basis of construction of GHQ and SUBI, the two key ‘non-conventional’ indicators of gender bias that have been used in the household surveys along with other, more ‘conventional’, indicators in the country studies prepared under the GPN project.

Chapters III, IV, VII and VIII look at the work done in Bangladesh, India, Pakistan and Sri Lanka respectively. Among other things, they report condensed results of coordinated household surveys carried out in the households of women workers in Export Processing Zones and Export-Oriented Units in these countries. The survey in Sri Lanka incorporates additional data from some other kinds of households as well. These surveys and the supplementary research involving secondary data sources were carried out by research teams from the Bangladesh Institute of Development Studies (BIDS) in Dhaka, Bangladesh, the Institute of Social Studies Trust (ISST) in New Delhi, India, the Pakistan Institute of Development Economics (PIDE) in Islamabad, Pakistan and the Centre for Women’s Research (CENWOR) in Colombo, Sri Lanka, respectively. Chapter VI on Nepal is based on a year’s work done under the project by APROSC in Kathmandu, Nepal. No household surveys could be carried out in Nepal, and Chapter VI is an edited version of the report prepared by

APROSC and submitted to ISST at the end of the first year of the project. Chapter V on Kerala condenses the findings of a much larger report prepared by a team of researchers from the Centre for Development Studies in Tiruvananthapuram in Kerala, which is designed to understand why the average Kerala woman with remarkably high GDI levels continues to live under conditions of pronounced patriarchal subordination.

One of the new points characterizing the essays in Section II is the incorporation in the survey questionnaires of modules on GHQ and SUBI, the two indicators of stress and mental well-being. These were canvassed along with other standard questions to adult men and women in the sampled households. The data thus generated have then been used to analyse the incidence of gender-disaggregated stress and anxiety along with other standard indicators of gender bias in order to understand the impact of changes in the external economic environment of sampled households brought about by reforms-induced new job opportunities for women in the export sector.

It must be kept in mind that the results reported in this section of the book are objective assessments of gender conditions that happen to be very context-specific. It reflects the situation in small samples of households in four South Asian countries in the Network under what would in all likelihood be among the more positive features of trade liberalization regimes in these countries, i.e., where new jobs have been created for women in EPZs or EPUs. As mentioned earlier, the second phase of the project will look into situations which have had a negative impact of macro economic reforms on the households. In particular, we will investigate gender conditions in households where workers, men or women, have lost their jobs due to reforms-related factors.

It would be fallacious to read more into these results than they are designed to convey. If, for instance, one has found that the women in some of the sites have been happy to have had the opportunity to work in factories, it only suggests that the other options open to them at that point of time were deemed by the women themselves to be worse. For instance in the surveyed households in Sonepat in India, women who did not have factory work, almost uniformly said they would have been happier if they had also had the opportunity of doing a paid factory job. This does not mean that the conditions of work in these factories had in some

normative sense been ‘ideal’, or that women’s entry into the paid workforce had not necessarily been ‘distress driven’. Nor should it be construed to suggest anything on the future stability of the employment being generated in these units. These jobs may indeed disappear in the face of the volatility of export demand, as has happened elsewhere in the world. To derive generalizations from these micro level results for analysing issues for which they were not designed in the first place would be unwarranted.

This of course does not answer the question how the micro evidence fits into the macro picture that is portrayed by some of the secondary data that has been reported in the papers in this section. Our answer to that is that it does not, and the whole point of this set of studies has been to admit upfront that at the current state of model building and data availability, it simply cannot. The project had started with the purpose of laying out a sensible framework of gender analysis simultaneously at the micro, meso and macro levels. We see these micro level results as building blocks that will in due course hopefully help one to complete the jigsaw puzzle of linking micro evidence with macro level data systematically, block by block.

Section III compiles a couple of the studies that were commissioned under the project to explore the macro and meso-level dimensions of the problem. Chapter IX by Manju Senapaty reviews the prospects of female employment in the region under the new trade regime. The paper by Anushree Sinha and N. Sangeeta explores the feasibility of extracting gender based information using a standard Computable General Equilibrium modeling framework (Chapter X). These two studies are part of the Regional Component of the GPN project designed to investigate problems that are not country specific, but may be of common concern to the region and beyond.6 Chapter XI by Ratna Sudarshan pulls together the common concerns and the conclusions that bind the individual studies together.

6In a companion paper within the regional component of the project, Anyck Dauphin has reviewed the growing literature on models of household decision-making in situations of co-operative conflict under unequal intra-household distribution of resources. This paper is forthcoming in the GPN Working Paper Series. Also see MIMAP and ISST websites.

Concluding Observations

The essays and reports included in this volume comprise a subset of those that were prepared for the entire project. In particular, the country papers included in Section II are considerably abridged versions of much larger volumes that include a great deal of other pertinent material including available secondary data on relevant variables and case study details from all the five countries. (See ISST website, and the website of MIMAP for the detailed reports and working papers). One has to keep in mind that this is a report of work in progress. In particular, one needs to remember that the results of the household surveys are applicable to the specific context within which they were designed. These need to be supplemented with results of other surveys characterized by very different contexts before they can be cited with any degree of generality.

What seems clear though is that the process of economic reforms has been instrumental in changing the external economic environment of households affected by them in one manner or the other, and may have done it at a pace that has been faster than anything before. These in their turn, have brought about changes in the manner in which different members of the households adjust to the changed situation, and how they perceive such changes. New market roles for women need not necessarily bring about greater empowerment. The old order of hierarchies may persist and new forms of subordination may surface, reinforcing the unequal power equations between the sexes.

However with more and more women getting exposed to the external world, through schooling and the labour market, it is possible that subtle changes are indeed taking place in this bastion of patriarchy. One may just be on the threshold of a revolution where questioning of traditional gender equations by women within families may usher in a new era of gender equality.


Central Statistical Organization (2000) Report of the Time Use Survey, Ministry of Statistics and Programme Implementation, New Delhi: Government of India

Doniger Wendy (1992) The Laws of Manu, (Translation of ‘Manu Smriti’ from Sanskrit) London: Penguin

Elson, D. (1994) “Micro, Meso, Macro: Gender and Economic Analysis in the Context of Policy Reform” in Bakker, I. (ed.) The Strategic Silence: Gender and Economic Policy, London: Zed Books

Ghosh, Jayati (1994) ‘Gender Concerns in Macro Economic Policy’, Economic and Political Weekly V29 (18)

Mukhopadhyay, Swapna (1998) In the Name of Justice: Women & Law in Society, New Delhi: Manohar Publishers

Mukhpadhyay, Swapna and Savithri, R. (1998) Poverty, Gender and Reproductive Choice: An Analysis of Linkages, New Delhi: Manohar Publishers

Sen, Amartya (1990) ‘Gender and Cooperative Conflict’ in Irene Tinker, Persistent Inequalities, New York: Oxford University Press

System of National Accounts (1993) Inter-Secretariat Working group on National Accounts, Commission of the European Communities-Eurostat, International Monetary Fund, Organisation for Economic Co-operation and Development, United Nations, World Bank, Brussels/Luxembourg, New York, Paris, Washington D.C

World Development, Growth, Trade, Finance and Gender Inequality, Special Issue, 2000

Non-Conventional Indicators of
Gender Disparities Under
Structural Reforms



Structural adjustment refers to economic reforms undertaken in countries with heavy burdens of international debt. Typically they involve lifting subsidies on food and other basic commodities, deregulation of local currencies, decreased investment in social services like health and education, denationalization of state-sponsored production activities and shifting from production for domestic use to production for export. Diverse projections have been made about the outcomes and differential effects by gender of such reforms. According to the Economic Agenda document prepared by the Coordination Unit for Women NGOs (1995), a futuristic nightmare is visualized for women where widespread unemployment will lead to economic insecurity and an increase in crime, where cuts in government spending and employment will mean worsening health services, less access to education, and deterioration in civic services. As jobs disappear and the cost of living rises, women will be forced to seek employment in insecure, poorly paid jobs and since gender relations at the household level governing the sexual division of labour tend to remain rigidly in place, women will be forced to work harder and longer hours. Girl children will be forced to give up their education or prospects for gainful employment in order to help their mothers with household maintenance. Rising food prices will mean poorer nutrition, especially for the female members of the family. The fallout will ultimately be on their health and well-being. It is thus held that economic reforms, by ignoring the crucial structural category of

gender, interact with existing gender asymmetries to affect women in negative ways.

On the other hand, some have held that this doomsday scenario is, in fact, present reality and argue that some aspects of economic reforms hold promise for healthy change (Kishwar, 1996). It is true that the process has brought more women into the workforce. Whether this will have liberating and empowering effects for women ultimately, especially for those women whose options have thus far been limited by their confinement to the household sphere, remains to be seen. Certainly their well-being will be affected if their working outside triggers conflict in the home, and also if their work burden has to stretch to include income generation in addition to their traditional tasks of household and family maintenance. Although the overall effects of SAP are difficult to distinguish because women in different sectors are affected differentially, it is clear that women are working longer hours in an effort to reduce the effects of the rising costs of living and cuts in social services (Elson, 1994).

Economic changes interact in complex, far-reaching and sometimes unforeseen ways with gender relations. On the one hand, the effects of economic changes are mediated by and assimilated into the prevailing sociocultural constructions of gender. On the other, they may actually create altogether new gender-related configurations for that culture. For instance, Vlassoff (1994) shows how economic development associated with agricultural growth in Maharashtra has enabled families to withdraw female labour from the fields thus enhancing their prestige and the woman’s status within the home. This is consistent with traditional seclusion practices for women, especially those belonging to higher castes. But women’s confinement to the domestic arena actually reduces their autonomy and power in society. Bulow (1992), on the other hand, found that among the Kipsigis in Kenya, a previously classless, decentralized society with autonomous spheres of male and female economic and political activity, has undergone a transformation into patriarchy in which men control the means of production and are the formal heads of households. Respect and reciprocity between men and women in the past has been replaced by women’s economic dependence on men and men’s fear that women may ‘try to be bigger than men’.

A USAID study (Altomore, 1991) concludes that the gender

differentiating impact of SAP follows three different scenarios. Theoretically, in countries where mobility, education and access to resources are uniform across gender, the impact on men and women will be equal. In those countries with practically no economic participation of women, like Pakistan, adjustment will not result in increased differentiation along gender lines. But in most African and Asian countries, the relative position of women will slide down despite their significant economic participation because this participation is invariably based on unequal terms. The gender gap in indices such as life expectancy, literacy, and earnings has in fact increased from the sixties to the eighties in South Asia and sub-Saharan Africa.

Assessing Gender Disparity

In 1995, the UNDP brought out the Gender Development Index (GDI) as a tool to capture gender inequality. The GDI notes the inequalities between men and women on the same variables that make up the Human Development Index (HDI) and which are related to overall achievement in that society. These are the variables of education, health and income. The GDI has been criticized on the grounds that by considering only three variables, it fails to take into account important dimensions such as the quality of community life, access to basic amenities and human rights. Dimensions which are particularly salient for women such as safety and security, household allocation of resources, unpaid labour, constraints imposed on mobility and on sexual and reproductive freedom by patriarchal ideology do not find any place (Hirway and Mahadevia, 1996). It is argued that rising incomes may not indicate improved status for women. Other variables that reflect women’s subordinate status such as son preference, nutritional status or the sex ratio may not show positive change. Also some indicators may not differentiate at all income levels. For instance, while education and income may be high, there may still be a great degree of violence against women. Or in the urban middle classes, while there may not be overt gender disparities in provisioning for basic needs, dowry demands may show escalation. Similarly, the process of Sanskritization may lead to greater female seclusion and more

rigidly defined gender roles even as the family prospers (Mukhopadhyay and Sudarshan, 1997).

In recognition of the centrality of gender-based power imbalances, UNDP has clearly identified the movement towards gender equality as a political process. Its Gender Empowerment Measure (GEM) attempts to reflect this and consists of variables that reflect women’s political participation, their access to professional positions and their earning power. A comparison of GDI and GEM yields interesting results. For example, countries like France and Japan are high on the GDI but lag far behind on the GEM. Some developing countries like Barbados, Cuba, China have higher GEM rankings than developed countries like Japan, France and Spain. The GEM too is criticized on the grounds that it is based on only three variables and that it is built on a very narrow conception of empowerment. It does not include legal and human rights, nor does it reflect the ways in which cultural constructions of gender identity and the practices these entail, disempower women. In discussing the lacunae of the GEM, Wieringa (1999) cites the example of the Caribbean country of Barbados which fares better than the UK, Switzerland, Japan and France on the GDI and the GEM. Yet these ratings have not changed the gender ideology that views women as subordinate to men and which has become increasingly and overtly misogynist as women have advanced in education and work.

Gender issues related to power and subordination cannot be ignored. The distinction between women’s practical and strategic interests is relevant here. Kabeer (1994) draws attention to the fact that this distinction helps to uncover the tensions between policies that attempt to meet women’s practical needs by improving the concrete material conditions of their lives, and those that seek to transform women’s position in an unequal society. While the former could serve as a starting point for challenging inequalities in some instances, in many others they act to preserve and reinforce existing inequalities. The distinction also helps to understand the resistance to and violent reactions against changes to the gender status quo.

From what has been said above it may be seen that if we want to assess the effect of structural adjustment reforms differentially on men and women, the conventional parameters used to measure gender disparity may not be enough.

Mental well-being, life quality and SAP

Human well-being at both physical and mental levels is the ultimate goal of development. While there have been attempts to track physical well-being through measuring aspects like health and nutrition status, anthropometric data, and mortality rates, mental well-being has been relatively neglected as an important aspect of the quality of life. The empirical literature does show it to be sensitive to psychosocial stress and of relevance to the concerns of this paper. There is clear evidence of gender differences in the nature and extent of psychological distress experiences.

According to the World Bank (1993), mental health problems and behaviour related illnesses, many of which are the result of conditions like poverty, exploitation, discrimination, unemployment and violence, account for about 42 per cent of the disability-adjusted life years lost. Desjarlais et al (1995) in their exhaustive coverage of the state of world’s mental health, note that in many parts of the world, economic progress and gains in longevity have been accompanied by an increase in social, psychiatric and behavioural pathology.

The economic reforms under structural adjustment have been known to result in increasing poverty for some peoples, mainly the poor and for women. Secondly, there is evidence that the gap between the rich and the poor has widened. These two factors are known to have important implications for mental health.

Decades of research find poverty to be a correlate of psychological distress and diagnosable mental disorder. In their authoritative review, Neuegebauer et al (1980) found that averaging across studies, psychopathology is at least two and a half times more prevalent in the lowest social class than in the highest. The association between poverty and mental health is not surprising when one considers that poverty imposes considerable stress on people while at the same time undermining many potential sources for social support.

It also appears that absolute levels of income may not be critical to mental well-being except at the lowest levels. Factors such as social status and power and the effects of social comparison may be more significant. ln this connection, it is relevant to note that it is income distribution and not just mean levels of income in a society that is important. Seidman and Rapkin (1983) show that although

the prevalence of mental disorder increases in economic downturns, this effect is greatest in heterogeneous communities where recession does not affect everyone equally. Other studies cited by Diener (1984) show that communities with less equal incomes are less happy than those with more equal incomes. Thus feelings of distributive justice and relative deprivation, as well as rising aspirations when one sees people immensely better off than oneself, mediate the effects of income on well-being.

Gender and psychopathology

There is clear evidence that women across all socio-economic levels, and across diverse societies, have a greater prevalence of psychological distress and clinical symptomatology than men do. In their review for the WHO, Dennerstein et al (1993) write of the greater prevalence of depression, phobias, somatization disorders and anxiety-related disorders among women. Men, on the other hand, are more likely to suffer from alcohol and drug related disorders and more likely to have a diagnosis of anti-social personalities. From their review of Indian studies, Kapur and Shah (1991) and Davar (1999a) conclude that women have a significantly higher rate of psychiatric morbidity. This higher psychiatric morbidity is found to be consistent across urban/rural backgrounds, religious and caste affiliations and socio-economic class. Depression has been found to be the most prevalent disorder in women all over the world and in India too (Nandi et al 1980; Kapur and Singh, 1983; Nolen-Hoeksema, 1987). Reviewing studies from across the world, Desjarlais et al (1995) find comparable results in Nigeria, Mexico, Uganda and Brazil.

There are strong arguments for a psycho-social etiology in accounting for the higher prevalence of mental distress in women. First, the so-called ‘severe’ mental illnesses such as schizophrenia and bipolar affective disorder show no significant gender difference in prevalence. Biogenetic factors are clearly implicated in these unlike the case of the depressive and anxiety related disorders in which prevalence rates are higher among women (Kapur and Shah, 1991; Nolen-Hoeksema,1987).

Secondly, while prevalence rates for disorders are similar for boys and girls in childhood, the rates are differentially higher for females after puberty especially during the reproductive years (Davar,1999).

This is probably related to the fact that these years are dominated for women by the need to follow culturally prescribed codes of conduct which are burdensome, oppressive, discriminatory, disempowering and restrictive. Interestingly, community surveys that include elderly women show a marked decrease in disorders after the age of 50 for women whereas for men the rates of depression increase. Venkobarao (1989) in his study of geropsychiatric morbidity in a region near Madurai, explains this on the basis of the loss of self-esteem and personal power that accompanies lack of employment in elderly men. For women, the reverse may be true as they now have more control over family life.

Thirdly, there are many factors that characterize the everyday lives of women that, taken together, can account for their higher levels of mental distress. These factors include poverty, violence, sexual discrimination, women’s lack of power and autonomy, their sex role socialization, the nature of roles that women are expected to fulfil, and the stressful effects of role overload and role conflicts on them.

As described earlier, poverty is not only an independent risk factor for mental disorder but is associated with other factors that compound the risk. Poverty affects women more than it does men because of women’s weaker fall-back status arising from their lack of access to and control of resources, and also because of their responsibilities to their children which are central to their identity and self-esteem. In her discussion of the mental health consequences of poverty, Belle (1990) points out that poverty undermines the ability to fulfil important social roles and such failure may be the conduit between experiences of poverty and depression. Thus for men, the financial strain of unemployment is compounded by their failure in the role of breadwinner leading to elevated symptom levels (Ross and Huber,1985). Through extensive interviewing, Wolf (1987, cited in Belle, 1990) found that negative social role identification such as ‘bad mother’, ‘bad provider’, as well as breaches of conscience that were necessitated by poverty (’I am a thief’), constituted many of the specific experiences that women reported as having precipitated their depression. Poverty also exacts a toll on social support systems including on marriage bonds. Economically insecure women are also often trapped in unhappy domestic situations since they rely on these for services they cannot afford to buy such as childcare.

A growing body of evidence has implicated violence as a grave risk to mental and physical health. Violence is related to the development of a wide range of psychiatric problems including post-traumatic stress disorder and other anxiety-related disorders, depression, suicidal attempts, substance abuse and eating disorders (Brannon, 1999). Violence may be an important factor that accounts for the gender differences in depression (Cutler and Nolen-Hoeksema, 1991). Childhood victimization is especially harmful and the research points to the fact that child sexual abuse is far more common for girls than it is for boys. Data collected in India among school-going girls and middle class women reveal shockingly high rates of childhood sexual abuse (Ganesh,1994; RAHI,1998; Purewal et al, 1999). Harris and Landis (1997) report that in the US, 35-51 per cent of mentally ill women report a history of child physical abuse and 20-54 per cent report a history of childhood sexual abuse. Purewal et al (1999) report that in their study of women attending psychiatric services, over 60 per cent reported experiencing violence prior to the onset of symptoms. Of the depressed women in the sample, 20 per cent had experienced child sexual abuse, 42 per cent had experienced domestic violence and 22 per cent had experienced sexual coercion. There is evidence too that everyday experiences of sexism such as sexual harassment are related to psychological distress in women and predict some symptoms better than do more general measures of stress (Brannon, 1999).

The role of power in household relations is an important issue in mediating psychological distress. In their study in Kota, Carstairs and Kapur (1976) found that the change from the traditional matrilineal to a patrilineal system was associated with stress for women compared to men because of the disempowering effects of the change for women. Davar (1999) concludes from the evidence that the role of ‘housewife’ and ‘motherhood’ have been associated with high levels of distress primarily because they are high in demand and low in control.

Women’s sex-role socialization is also such that certain qualities that make for healthy adulthood (Broverman et al, 1970) such as independence, autonomy, assertiveness, and the development of instrumental competencies are actively discouraged. In reviewing the ‘Iearned helplessness’ explanation for women’s higher rates of depression, Nolen-Hoeksema (1987) points out that women receive

more helplessness training over their lifetime than men do and cites cross-cultural evidence that indicates that in many cultures girls are encouraged to be dependent on others, whereas boys are taught to be more self-reliant and active. From childhood through adulthood, females have many opportunities to learn that their responses do not control outcomes with obvious implications for their sense of self-efficacy and personal agency. Further, traditional femininity and success outside the domestic arena are often seen as opposed.

As a gender class, women are evaluated more favourably than men because of their nurturing qualities (Eagly et al, 1991) but these same qualities may also put them at risk. Davar (1999b) argues that women are at risk for mental distress because of their socialization as selfless ‘carers’, especially in situations where their subordinate status easily permits ‘caring’ to be exploited and abused. Women assume the burden of caring not only for children but for their husbands, elderly relatives and the sick, and typically provide more support than they receive.

The mental health effect of women’s multiple roles is not a straightforward one but appears to be mediated by a woman’s social support network, her coping style, and how significant each role is to her self-esteem. Many studies attest to the beneficial mental health effects of multiple roles. Many working women report a great sense of satisfaction from having an independent income and a work environment of some autonomy and control. However, power is a significant variable here. In India, Chakraborty (1990) found in her field study that women in higher positions of power experienced less distress than did those in lower positions. But for women in low level positions, the distress was four times higher than for men in similar positions. This may be because their employment causes friction in the home, interferes with their household responsibilities, or overburdens them where domestic help and labour saving devices either cannot be afforded or are not used for other reasons. Another angle on household power relations comes from the interesting finding in studies in the west that while women who are employed may have fewer mental health problems, their husbands may not. The shift in power that accompanies wives’ employment may be a problem for husbands (McBride, 1990; Brannon, 1999). The high rates of depression among housewives are partly accounted for by the fact that the housewife role usually combines high responsibility and work overload with low power

which is a potent combination for depression and anxiety (Rosenfield, 1989).

Gendered effects on mental health are seen also in the salience of certain roles. For men, stress is related to work roles and the threat to their role as provider, whereas for women symptomatology is contributed to more by marital than by job-related stress, and their parenting role. There is general agreement that parenting contributes significantly to role overload especially since women feel responsible for the continuing well-being and success of their children and are often blamed when things go wrong (McBride, 1990).

In this connection it should be recognized that the burden of work and family maintenance for many women in the developing world is indeed arduous. The productive work that women do is usually unrecognized and the new economic policies have made women’s labour even more invisible and increased their work burden to the point of affecting their health and nutritional status. According to Elson (1994), it is women who have had to be the shock absorbers of the SAP. The rigid sexual division of labour means that women continue to perform their traditional roles while also earning an income, thus working 12-18 hours a day. It falls to daughters to give up schooling in order to share the domestic responsibilities with their mothers and sisters.

Women’s lack of authority, autonomy and ‘bargaining power’ also renders them without redress in the face of discrimination, exploitation and injustice. Further, as Connell (1995) describes it, the cultural construction of femininity functions as a sort of ‘cultural disarmament’ of women’s fighting-back potentials. The expression of distress and the negotiation of a better deal for themselves may therefore be quite indirect. In this connection, it may be useful to consider two specific types of disorder that are especially common among women compared to men in the developing countries, namely, dissociative disorder and somatization disorder. Cultural perspectives on dissociative disorders — trancing and possession being common manifestations — suggest that these may actually be a kind of protest behaviour. Typically, when ‘possessed’, norms of proper modest female behaviour are violated, hierarchies are turned upside down in the demand to be treated as divine and criticism and abuse are voiced against those it would be

impossible to criticize openly in normal circumstances (Lewis, 1986; Desjarlais et al, 1995).

Similarly, explanatory models of somatization disorders suggest that it is a way of conveying distress among those with limited support and limited opportunity to ventilate their feelings (Nichter, 1982). Shah (1996) draws attention to the problem of somatization among women in India and cites Isaac et al’s (1995) study that found that 65 per cent of their patients with unexplained somatic complaints were women. Ethnographic studies too reveal that where psychological pain is not realized as depression or anxiety there may be local ‘idioms of distress’ such as ‘nerves’, ‘heaviness of heart’ and so on. Such studies of ‘nerves’ in North and South America, the Mediterranean region, and the Middle East consistently showed a higher prevalence among women (Davis and Low,1989). Careful attention to the social and cultural meanings associated with these complaints often points to power conflicts in families and communities, and the experience of oppression and deprivation (Good and Good, 1982; Jenkins 1991).


Violence towards women is found to be etiologically related to psychological disorders as described above. On the one hand, it reflects gender asymmetries in power relations, and on the other, it is frequently a response to challenges to gender ideology. Also, it is a pervasive feature of women’s lives as is shown below. Violence towards women may therefore be considered to be a sensitive index of their well-being and status.

Severe and ongoing domestic violence has been documented in almost every country in the past decade. In the western data, 21–28 per cent of adult women are battered in marital relationships (Maynard,1993). Desjarlais et al, (1995) provide statistics from a range of studies showing shocking levels of domestic violence in many countries. Thus 60 per cent of a random sample of women from a low-income neighbourhood in Sri Lanka had been beaten, husbands killing wives accounted for 50 per cent of all murders in Bangladesh, one out of three women in Mexico is a victim of family violence, 50 per cent of women in Bangkok’s largest slum, and about 60 per cent of women from both poor and elite groups in

Papua New Guinea are beaten regularly. In India, Sriram (1991) reports that 35-60 per cent of women were battered by their husbands in.Gujarat. Agnes (1988) reports that up to 3 per cent of women suffer gross assault. She also points out that domestic violence cuts across education and income levels, it occurs in both joint and nuclear families, and is not necessarily related to alcohol abuse. In their community study of rural women and men in Karnataka, Batliwala et al (1998) found that far more women (61 per cent) than men (39 per cent) reported experiences of harassment or violence for various reasons. For women the greatest number of incidents of abuse occurred within the home at the hands of the husband or in-laws. Men, on the other hand, were more often involved in, or were victims of violent conflict with their own kin over property or money disputes. While much of the abuse was nonphysical, women were substantially more often the victims of physical violence than were men.

Domestic violence resulting in death such as dowry deaths in India, and female infanticide in lndia and China deserves particular mention. Data based on national crime statistics and police records grossly underestimate the prevalence of violence. Even so, they present a shocking picture. For instance, dowry deaths in lndia were 999 in 1985, 1786 in 1987 and 5157 in 1991. Harassed women are also driven to suicide. In Gujarat alone, it is estimated that 2000 women committed suicide on account of domestic violence in 1989-90 (Mehta and Dighe, 1991). Other studies in India such as those of Banerjee et al (1990) in West Bengal, Shukla et al (1990) in UP and Kodandaraman et al (1985) in Bangalore identify domestic strife and violence as the most significant reason for female suicides. While unemployment and other financial problems, and discord with parents are significant reasons for male suicides, family ill-treatment and violence are more significant in the case of female suicides. Desjarlais et al (1995) quote ethnographic data from Oceania, South America and China that corroborate the findings that domestic violence is directly related to depression and suicide.

According to Agnes (1988) the most common causes for domestic violence were arguments over money, alcohol and the wife’s employment, jealousy and suspiciousness, rows over housework, disputes over children and over the husband’s extramarital affairs, or at the instigation of the in-laws. In Batliwala

et al’s (1998) community study of rural women and men in Karnataka, the majority of women reported the trigger to violence to be any act of theirs which the husband construed as disrespectful or disobedient, or a challenge to his authority or to the primacy of his family. Rows over the husband’s use of alcohol also triggered off violence. Women were beaten for being childless or for only giving birth to daughters. Sociological perspectives on violence such as resource theory and exchange theory have shown that power is based on resources and that violence is the ultimate resource for securing compliance. In his exchange/social control model of family violence, Gelles (1983) holds that violence will be used when the rewards are higher than the costs. The private nature of abuse, the reluctance of outsiders to intervene, and the low risk of any other intervention reduce the costs of abuse. The cultural approval of violence as both expressive and instrumental behaviour (for instance, in disciplining wives) raises the potential for it.

But these theories are limited in that they do not take into account that domestic violence occurs in a context of institutionalized power asymmetry. Yllo (1993) argues that without a feminist lens in understanding violence, explanations are incomplete. The feminist coercive control model of violence identifies it as a tactic of entitlement and power that is deeply gendered. Support for this is sound. Gelles and Strauss (1988) found in the course of their extensive interviews that events around power and control were most frequently at the core of events leading up to the use of violence in domestic situations in the U.S. Ethnographic studies suggest that in egalitarian societies such as in the Vanatinai islands, attacks on women are rare (Lepowsky, 1994 cited in Brannon, 1999). Gelles (1993) quotes research that shows how researchers using different methodologies on different populations all found that gender inequality explains variations in the prevalence of violence towards women. He cites a study in Papua New Guinea that contrasts the explanatory power of competing theoretical models of violence, namely, the social disorganization model and the gender inequality model. Using in-depth interviews and questionnaires, it was found that although modernization in Papua New Guinea did produce new pressures, expectations and changes in women’s support systems, the underlying explanation for the abuse of women was the husband’s perceived right to control his wife and a social structure that

legitimized the exercise of this right. Parallel results were found in an analysis of data from 50 U.S. states. Although social disorganization does contribute to domestic violence, the greater the inequality between men and women, the greater the violence. Similarly, using an ethnogender approach in her study of marital violence among South Asian immigrant women in the U.S., Abraham (1998) found that the high levels of such violence were related to the need of South Asian men to uphold traditional values of male dominance and female subordination and to enforcing traditional scripts of female roles in terms of perfect and self-sacrificing home-makers and care-givers.

Sexual violence against women has also assumed frightening proportions. It is estimated that some 20 per cent of women across the globe have been subjected to rage (Koss, 1993; Goodman et al, 1993). Rape is grossly under-reported for obvious reasons. Yet in discussing the mental health of Indian women, Mane (1993) reports that in 1990 there were 7,856 cases of rape reported, 16,683 cases of molestation, 9,625 cases of ‘eve-teasing’. Desjarlais et al (1995) give an account of the female sexual slavery widespread in Asian countries including India, Nepal, Thailand, Philippines and Burma. Blaming and stigmatization are particularly common in cases of rape and in countries where the ideology of purdah and izzat are extreme, victims of rape may be killed by male family members for being ‘spoiled’ and thus bringing dishonour to the family. Nearly half the rural women in Batliwala et al’s study (1998) reported a sense of insecurity and fear of sexual assault. They also feared having their reputations sullied by gossip about their sexual morality. The powerful effects of the ideology of shame and honour are evident.

Brannon (1999) reviews recent research on sexual violence and concludes that rape is not the act of pathological men but is the action of men behaving in ways that their culture allows and maybe even encourages. Acceptance of rape myths (rape is provoked by women, women secretly enjoy rape, etc), acceptance of interpersonal violence, the desire for dominance, and hostility towards women are all factors known to be related to the appeal of sexual aggression. She cites Malamuth (1996) who predicts sexual aggression based on two variables, namely, high levels of uncommitted, impersonal sex, and hostile masculinity (hostility toward and desire to dominate women).

It is important to also recognize that since women’s chastity and sexual exclusivity are symbols of male status and family/comrnunity honour in patriarchal societies, women became inevitable targets of sexual violence in conflict situations (Das, 1997). Sweetman (1998) draws attention to the increase in male violence towards women in times of conflict, in militarized societies and upon challenges to gender ideologies of female subordination. Fifty three per cent of her respondents in Lesotho were beaten by their husbands and all reported an increase in such violence after their husbands’ retrenchment. As the proportion of income women brought into the home increased, so did the level of domestic violence. She cites the work of Rowlands (1997) who showed that in Honduras, challenging the traditional sexual division of labour and gender ideology, without recognizing the strength of male resistance to such change, can have tragically violent consequences for women.

In summarizing the evidence, Davar (1999a) concludes that violence towards women is not a clinical problem but a social problem arising from a context where unequal power relationships and entitlements are legitimized and used to enforce subordination. Male violence generates a climate of fear which then regulates female behaviour in ways that perpetuate the status quo. An important point she also makes is that in the Indian context, violence towards women is more aptly described as patriarchal violence rather than as male violence towards women. This is because women are victimized by men as well as by other women for patriarchal ends. Thus dowry harassment and murder frequently involves mothers-in-law. Other examples are the practice of sati, son preference, and the devdasi systems. In examining attitudes toward wife beating, Batliwala et al (1998) found that in this area women seemed to be more ardent upholders of male rights than men themselves with 70 per cent of women stating this to be the husband’s prerogative while only 44 per cent of men did so.

To summarize the above, it appears that:

  1. there is a significantly greater prevalence of psychological distress and psychiatric morbidity, particularly of depression and anxiety-related disorders, among women compared to men.
  2. given the social origins of much of this distress, this higher prevalence may be taken as an indicator of the status of women in society.
  3. the prevalence of violence towards women is clearly reflective of their status in society. It not only occurs as a symptom of gender inequality, but is also a response when male domination is threatened.

It is important to note that the relationship between prevalence rates of psychiatric disorder and the mental health of a group is a complex one. The literature on subjective well-being testifies to this. The empirical literature shows only a very small correlation, if at all, between ‘ill-being’ and ‘well-being’ indicating that these are not the two poles of a single dimension but are two separate dimensions (Diener, 1984; Sell and Nagpal, 1992). Secondly, it is to be noted that the process of psychiatric assessment and diagnosis is not itself without gender bias. This is partly because mental health professionals may not be free of gender stereotyping in their practice (Addlakha, 1996), and secondly, because the psychiatric classification systems, i e, the ICD and DSM, are not free of gender bias either. Davar (1999a) and Caplan (1995) have exhaustively discussed these issues. Thirdly, there is some literature that suggests that men too suffer comparable degrees of distress but deal with their distress in a different style which is manifest in higher rates of alcohol and substance abuse. This is discussed in a later section.

The assessment of psychological well-being and distress

A number of instruments ranging from elaborate semi-structured interview schedules to brief self-administered questionnaires are available which assess different aspects of mental health and distress. These have been used in psychiatric epidemiology and in research on stress and well-being. A widely used tool in survey research is the General Health Questionnaire (GHQ) developed by Goldberg (1972), a questionnaire that has found application in a variety of psychiatric and nonpsychiatric contexts, and has been translated, revalidated and used in many countries including India. In their review of survey instruments that would enable public health specialists to measure the psychological health and well-being of a population, the GHQ was found to be highly valid and sensitive when the focus was on nonpsychotic and nonorganic mental disturbance (Bartlett and Coles, 1998a). It has been used in many studies related to the economic situation, occupation and unemployment (Strandh, 2000; Warr et al, 1988; Banks et al,

1980). The GHQ is available in several versions of varying length. The 12-item GHQ has the advantage of brevity and hence convenience of administration and scoring. It is also as sensitive and valid as are the longer versions. It consists of the 12 “best items” of the original 60-item questionnaire, that is, those items that were most discriminating in determining clinical status and were not endorsed by a physically ill control group. In India, the 12-item GHQ has been validated against the Indian Psychiatric Survey Schedule and has been found to demonstrate excellent sensitivity and specificity, that is, it is able to accurately identify the presence of psychological disturbance with few false negatives and false positives (Shamsunder et al, 1986). Each of the 12 items of the questionnaire has four possible response choices. The scoring is simple, the total score being a summation of the score on each item. In surveys, the GHQ can be used as a dimensional measure of psychological disturbance and can also be used to identify a “case”. In the former situation, the total score is used to indicate the degree of psychological distress. In the latter situation, a threshold score is used so that persons scoring above the threshold are identified as potential psychiatric cases. Thus, for the 12-item GHQ, the recommended threshold score is two and above.

In order to assess the psychological well-being of a population, it is useful to include both positive and negative aspects; these help to get a complete picture. As mentioned earlier, this is a concern since there is some evidence that psychological distress and well-being are two separate dimensions and not the two ends of a single scale. However, as Bartlett and Coles (1998b) point out in their assessment of well-being instruments, the positive scales are not as well validated as are the negative (ill-being) scales. It is also of interest that the differences in psychological health of different socioeconomic groups pertain to differences in ill-being or mental distress. Nagpal and Sell (1985) found that there is little difference among socioeconomic groups on the positive dimension of well-being. They posit that the negative aspect of distress or ill-being is more sensitive to the effect of situational adversity. Further, they predict that a significant impact on the positive factors would herald a more serious problem like chronic demoralization or anhedonia (Sell and Nagpal, 1992).

The Subjective Well-being Inventory (SUBI) developed by Sell and Nagpal (1992) attempts to address this issue by including both

positive and negative aspects. It consists of a 40-item questionnaire, each item having three response options. It yields a total score as well as sub-scores for the positive (well-being) and negative aspects (ill-being). Its validity has been established through factor analysis. Factor analysis over different samples, in different languages, and in different parts of India shows a good degree of stability of factorial structure. There are 11 factors, namely, general well-being: positive affect, expectation-achievement congruence, confidence in coping, transcendence, family group support, social support, primary group concern, inadequate mental mastery, perceived ill-health, deficiency in social contacts, and general well-being: negative affect.

We now come to discuss those variables that concern the power relations between the sexes and the social mechanisms that uphold and enforce gender inequalities.


Since gender inequality is inextricably bound up with power relations, the empowerment of women has been seen to be the goal of efforts to reduce such inequality. Although empowerment has been a buzzword in developmental circles, the concept has been used in so many different ways by different agents that it remains ambiguous. Definitions of what constitutes empowerment have ranged from the development of personal instrumental competencies and skills, to the process of challenging existing power relations, to household decision-making, to gaining access and control over resources like credit, income, land, knowledge, etc, as well as to subjective variables like the sense of personal power or self-efficacy. Conceptualizing power and gender adequately entails understanding its multidimensional nature, the complex ways in which women experience subordination, and the ways in which they negotiate or manage this state of affairs. The multidimensional nature of power suggests that empowerment must take place at different levels and in various spheres.

In a useful analysis of women’s empowerment, Kabeer (1994) draws on Lukes’ (1986) three dimensions of power to discuss the issue. The first dimension, the ‘power to’, is closely associated with decision-making over issues over which there is observable conflict.

Here power is seen as the capacity to affect outcomes. This is the kind of power that is examined in studies of household decision-making or the effects on such decision-making for women who earn an income. The second dimension shifts from the interpersonal to the institutionalized basis of power. This refers to the ‘power over’, whereby one group manages to suppress certain conflicts by denying their validity. Hence they are not even put on the decision-making agenda. For example, the assignment of household and childcare responsibilities is so deeply institutionalized that it appears non-negotiable. The third dimension of power recognizes that conflicts of interest may be suppressed not only from the decision-making agenda, but also from the consciousness of the parties involved. Here both the dominant and the subordinate parties subscribe to accounts of social reality that deny that any inequality exists. This formulation of power is concerned with ideology, and with the sociocultural constructions and patterning of behaviours. It prevents conflicts between dominant and subordinate groups by shaping attitudes, wants, needs and preferences in such a way that both accept the existing order either because they can imagine no alternative to it or because they value it as divinely ordained or beneficial. Feminist analysis has shown how social rules, norms, values and practices play a crucial role in concealing the reality and pervasiveness of gender inequality and in defusing gender conflict. For instance, the reverence of mothers and motherhood conceals, and also serves as a rebuttal, of the evidence of the exploitation, devaluation and oppression that women face.

Women’s experience of gender subordination is therefore complex. Women may not necessarily be aware of, or desire to realize their ‘strategic’ gender interests. This may arise out of women’s socialization into patriarchy thus creating in them a ‘false consciousness’. Or it may be that the restrictions on their mobility and experience prevent them from knowing other ways of being. Or they may be aware of the circumscribed nature of their lives without knowing what to do about it. Thus in South Asia women often describe themselves as frogs in a well or like oxen blindly turning the grindstone.

Understanding power and gender also means having to recognize the trade-offs that women make in order to cope with their dependent, subordinate status. They may subordinate their personal well-being to that of male authority in order to ensure long-term

security. Jeffery and Jeffery (1996) discuss the issues of empowerment and agency through extensive narratives of women’s everyday lives in rural north India. They suggest that women have stakes in the system and may prefer to make ‘patriarchal bargains’ (Kandiyoti,1988). The obedient wife could usually expect her husband to provide lifelong economic support and the compliant daughter-in-law could expect her situation to change with her seniority. They saw women evidence agency in various guises, including in ways that upheld the status quo rather than challenged it. Acts of resistance were indirect, for instance in their songs. Fear of shaming, gossip and ostracism and of losing home and children deterred them from open defiance. The authors also note the difficulty they had in finding satisfactory translation of the ideas bound up in ‘agency’ and ‘autonomy’. They found that terms such as ikhtiyar (power, right, authority), zimmedar and mukhtar (free agent, empowered) were not necessarily positively valued by women. It was not that women could not visualize having these qualities. Rather, such attributes could entail costs for those without much power since their initiatives could easily be blamed, punished or seen as shameless or immodest. Taking the argument a step further in her discussion of women of the Mukkuvar fishing community, Ram (1992) suggests that we need to consider how cultures also produce a subjectivity and a positive identity which have a stake in the current system despite experiences of subordination, devaluation and exploitation.

Two dimensions of gender inequality warrant further elaboration following from the above. One is that of gender identity and gender roles especially as these contribute to the maintenance of gender inequality. The second area, which has not been addressed thus far, is that of the self, and of personal power or self-efficacy.

Self, Self-Efficacy and Empowerment

Relational self and the ethic of care

A number of workers attest to the importance of the sense of self to empowerment. For instance, Schuler and Hashemi (1993), from their observations of, and interviews with, women in empowerment-related projects in Bangladesh delineate a list of indicators beginning with the personal and subjective elements of the sense of

self to more externalized and collective aspects. Yet there is a paucity of theorizing of women as subjects, of their sense of themselves and their sense of agency. Two aspects related to the sense of self that characterize women, namely, the relational self and the ethic of care, have implications for empowerment.

Kapadia (1999) inquires into the self-experience of women in India and the implications for empowerment. Her inquiry is concerned with the notion of the relational self (as contrasted with the autonomous self) which describes the predominant pathway of self-development in many non-western cultures, and in women across cultures. The self of women is defined through their relations with others and their roles as daughters, wives, mothers, daughters-in-law (Parikh and Garg, 1989). Using narrative methodology in a longitudinal study based in Orissa, Seymour (1999) interviewed many mothers, daughters and grandmothers of the same family and found that the women saw themselves through the prism of changing roles and ‘responsibility’ so that their identity and personal development could not be understood separately from roles and role transitions. Kapadia (1999) found that the self-esteem of women is located in the achievements of their significant others and in their success in carrying out their role prescriptions. She notes that negotiating for personal power in the family can be fraught with danger for women: it could lead to conflict, evoke allegations of irresponsibility and selfishness, threaten to damage or destroy their relationships and thereby their self-worth.

Interwoven into the relational self is the ethic of care, a moral orientation that predominates among women. Men, on the other hand, predominantly use the justice orientation. Drawing on Gilligan’s (1982) work on this gender difference in moral development, Davar (1999b) argues that while the care orientation may represent a superior morality for human society as some feminists have maintained, it costs women dearly in an unequal society. Where gender inequality prevails, it is economically and otherwise profitable for men to have a class of people to serve them under the rubric of care. Women are not only socialized into the care ethic but discouraged from negotiating for justice. They thus compromise on issues of self-interest, rights and entitlements. They are also more vulnerable to the moral affects of shame and guilt that belong to the care ethic, and consequently, perhaps, to depression.

Kapadia (1999) concludes that empowering women may entail helping them to redefine the sense of self and include a caring niche that creates space for the self as well as others. She cites Kagitcibasi’s (1996) Turkish work on the autonomous-relational self. According to this model, each individual has both a relational and an autonomous self. This then means kindling and unravelling a woman’s autonomous self in order to empower her. The above discussion raises questions as to how women define themselves outside a social role-definition or outside the caring function, whether there exist spaces where women can exercise the autonomous self and which give scope for the play of personal desire.


Personal power is related to self-efficacy in that people who have personal power actively engage in their world and know both how to get what they want and believe that their actions can lead to desired outcomes. Albert Bandura, who first formulated and researched this construct, holds that among the mechanisms of agency, none is more central or pervasive than people’s beliefs in their capabilities to produce given levels of attainment. Perceived personal efficacy influences the choices that people make, their aspirations, how much effort they mobilize on a given task, how long they persevere in the face of difficulties and setbacks, whether their thought patterns are self-aiding or hindering, the amount of anxiety and stress they experience in coping with threatening or taxing situations, their vulnerability to depression and their resilience in adversity (Bandura, 1992). Self-efficacy operates as a resource factor in the appraisal of stress so that people with high self-efficacy are buffered by this in their experience of stress whereas those with a low self-efficacy appraise such situations as highly threatening (Jerusalem and Schwarzer, 1992).

Bandura and Busey, (undated), in their discussion of gender role development and functioning, review research that shows that gender role identity and self-efficacy are linked. Thus, women who take a more egalitarian view of the role of women display a higher sense of efficacy for traditionally male occupations and pursue these careers more often. They thus construct different identities and futures for themselves. Conversely, women who see themselves as

highly feminine also see themselves as less efficacious in mastering the educational and job requirements of traditionally male gendered occupations even though they do not differ in actual tested ability. Thus, gender identity can be an impediment to achievement. They also discuss research on stress that showed that women in employment who had a strong sense of coping efficacy (in terms of being able to manage the multiple demands of family and work, to exert influence over their work schedules, and to get their husbands to help with child care) experienced a lower level of physical and emotional strain, good health and a more positive sense of well-being. This held true across income levels, occupational workload, and division of childcare responsibilities. Other studies show that what is experienced as an occupational stressor depends partly on the level of perceived efficacy. Women employees who have a low sense of personal efficacy are stressed by heavy work demands and role responsibilities. In contrast, those with a high sense of efficacy are frustrated and stressed by limited opportunities to make full use of their talents.

People’s beliefs in their efficacy develop out of experiences of mastery, that is, from experiences where their actions have led to desired outcomes. It has been discussed in an earlier section how the socialization of women provides ample opportunity to learn to be helpless. In its extreme form, the learned helplessness model helps explain what is called the battered woman syndrome (Walker, 1993). Vicarious learning through social modelling — seeing others similar to oneself succeed by sustained effort — raises one’s beliefs in one’s own efficacy. This is one important reason for the success of collective action and participation in women’s groups in the empowering of women.

Gender Identity

Various strands in the above discussion have suggested the significance of gender identity. This is because the construction of masculinity and femininity, and the ideology of gender-relations characteristic of a culture, act as powerful organizers of behaviour. Failure to observe culturally valued and gendered prescriptions for behaviour can lead to social disapproval and even punitive social reaction. It also leads to self-doubt, guilt and shame since these

values are deeply internalized and become core regulators of self-esteem. This helps to understand how women are co-opted into systems that oppress them. A case in point is Batliwala et al’s (1999) finding that endorsement of wife-beating as a husband’s prerogative was present in a substantial number of rural women in Karnataka, more so than among men. Similarly, Datar’s study (1995) of women tobacco workers in Nipani presents a revealing account of how gender identity is constructed around the idea of the ‘good’ woman, that is, the ‘sowbhagyavathi ’ who remains devoted to her husband despite ill-treatment, and the ‘bad’ woman who is husbandless, independent and sexually free. This despite the fact that Nipani is a town where most women are employed, whereas few men find employment and hence migrate out. Women continued to determine the status of other women in terms of their sexual relations with men, and divisions among women-workers were most damaging to their union-related activities when such discord was sparked off by sexually derogatory comments on morals. These women also chose to marry off their daughters as early as possible and demonstrated a preference for having sons. This was so despite their own experience of ill-treatment at the hands of husbands and sons.

Cultures differ in what specifically constitutes the ideal gender roles for women and men. For instance, in India, Bangladesh and Pakistan, the ideology of shame, honour and purdah prevail, and control of women’s sexuality is a major issue. In Thailand, however, a very different situation obtains. Both are oppressive to women but the oppression arises from different facets of how gender roles are constructed in these cultures. Pyne (1999) discusses how the traditional Thai woman possessed substantial autonomy; she participated in agriculture and in trade, oversaw household finances, selected her own spouse and brought him to live with her family, and inherited property equally with her brothers. Economic value was attached to the birth of daughters since it implied a bride-price and security in old age for parents. The Thai woman is both valued for, and burdened by, her strong sense of duty to her parents, husband and children. Pyne argues that the economic transformation of Thai society in the past 20-30 years has, in fact, built on women’s traditional mobility, greater economic responsibility for family care and maintenance, and less stringent attitudes toward sexuality. Their autonomy and relative sexual freedom have been

exploited without efforts to help them meet their traditional responsibilities to their family in alternative ways. Thus the disempowering aspects of gender roles may vary from culture to culture.

Control of mobility, sexuality and reproduction

In countries like India, Pakistan and Bangladesh, the degree of control women have over their sexuality and reproduction, and the degree to which they are subjected to seclusion practices may be seen as core indicators of their status since these are fundamental to their subordination in these patriarchal societies. Anthropological work has shown how women’s bodies and their sexuality are given meaning in the cultural ideologies of purity and pollution, shame and honour and purdah (Mandelbaum, 1988). Central to these ideas is the notion of good women as mothers (less so as wives) and of female sexuality as being easily aroused, uncontrolled, insatiable and dangerous and likely to bring shame to the family if not tightly reined in. Women’s mobility is therefore restricted because they may be in danger of upsetting the patriarchal order. To a greater or lesser degree and with some local variations, this seems to be true in India, Pakistan, Bangladesh and Sri Lanka (Mandelbaum, 1988; Ahmed and Ahmed, 1981; Nath, 1981; Risseeuw, 1991). Examining the basis of the subordination of women in early Indian society, Chakravarti (1993) argues that the central factor was the need to have sexual control over women to maintain patrilineal succession and caste purity. The sexuality of higher status women is subject to more restrictions since they are points of entrance for social climbers and hence their sexuality is especially to guarded.

The ideology of shame and honour are important ways whereby women’s sexuality is controlled and women are co-opted into these same values. For instance, an ethnographic study in a Karnataka village shows how women’s gossip serves as a powerful social control mechanism regulating female conduct and sexuality (Niranjana, 1999). It reinforces socio-moral boundaries by emphasizing the spaces and behaviour appropriate to a ‘good woman’ even at the cost of blaming women for the lapses of men.

Guzder and Krishna (1991), using case studies from therapy practice discuss how traditional cultural ideals reinforce a feminine

identity which serves familial and social ends. Women are identified primarily in the wife and mother roles, with Sita being the ideal of Indian womanhood. Singh and Uberoi (1994) in their analysis of the writing in a popular women’s magazine, show how such writings are ideological productions that serve to reinforce women’s position of structural and emotional subordination and dependence. These writings extol the virtues of adjustment and compromise on the part of wives as the key to marital felicity and warn that if adjustments are not made in time, marriages will fail, signifying the woman’s failure as a wife and consequently, her lack of worth.

Regarding rape, Vishwanath (1997) discusses how women’s feelings of shame vis-a-vis their bodies are located within a certain discourse that allows men to violate women’s bodies and throw the blame back at women. Male sexuality is portrayed as lust-driven and easily tempted, so that the blame invariably must lie with the woman — she is bad, and by not conforming to the behaviour and roles that patriarchy prescribes for her, she is legitimate prey for sexual violence. Feminists have tried to show that there is a link between shame and sexual violence. For instance, the Hindi work for rape, ‘balaatkar’, is rarely used, the more common phrase being ‘izzat lootna’ which places it within the discourse of shame and lost honour. This is why feminist activists teach women to reject the feeling of shame and guilt that accompanies sexual assault and to assert instead that a crime has been committed against them.

Such gender ideology has led to various practices. First, it accounts in large measure for early marriages leading to high fertility with consequent deleterious effects on infant and maternal health and morbidity. The early marriage of girls is seen as imperative so as to guard their izzat and ensure that they do not stray sexually. Second, fertility rates are increased by the need to prove fertility and to acquire status as mothers. Third, the preference for sons leads to pressures to reproduce until male offspring appear, and leads to the practice of female foeticide and infanticide. Many studies have shown that women are most vulnerable if not fertile or if they have no sons. Son preference is the single most common reason for not using birth control. It is also the most powerful reason for the sex ratio. A fourth reason is that women have little control over determining the size of their family or the use of birth control (Batliwala et al, 1999; Mukhopadhyay and Savithri,1997;

Jeffery and Jeffery, 1996). Also women who are unable to reproduce are stigmatized and ill-treated for their failure to perform their reproductive duty even if it is the husband who is infertile. Much of the time, they have no rights or control over sexual relations with their husbands. Their right to satisfying sexual lives simply does not arise and they are expected to perform their conjugal duties without complaint. Ironically, while women cannot decide the number of their children nor whether to use birth control, once they have satisfied their husbands and in-laws as to the number of children, it is they who have to take the onus of preventing further births. Finally, women’s physical mobility is restricted to varying degrees in accordance with seclusion norms. As described earlier, these practices embody the ideology surrounding shame, honour and pollution. These restrictions are the major cause for girls leaving school and for women being unable to work outside the home. They limit their access to health care, to social, informational and business networks and prevent them from seeking redress in the face of injustice or to fight for their entitlements and rights (Agarwal, 1994; Mies, 1982).

Household work and the sexual division of labour

Developmental efforts rarely challenge the idea that domestic work and the day-to-day tasks of family maintenance must be done by women nor do they promote men’s participation in household chores. Industrialized nations have seen a shift in men’s participation in housework and in childcare but the acceptance of this shift differs widely among both men and women, as does the perception of the amount of such work actually performed by men. Men’s participation in these chores is also associated with the development of modern technology. It is difficult to promote the idea that men can and should do ‘women’s work’ since the work itself is seen as having low value and women are seen as being inferior.

Risseeuw (1991) elaborates on this point in describing how women in Sri Lanka do the lowest, most repetitive and dirty tasks in the home as well as in the coir industry. When men participate at all in household work, this would not be on a regular basis nor would it involve tasks that were a drudgery to perform, nor would they do the tasks that restricted them to the compound. Thus

gender hierarchies clearly operate in the sexual division of household labour so that women do the menial work, work that is done by servants should a family be able to afford this. Men seen to do ‘women’s work’ risk ridicule and may be seen as wanting in manliness (Sweetman, 1998). The rigidity with which gender roles concerning housework and childcare are maintained, and the tolerance for deviations, may be useful indicators of gender inequality to consider.

While there is some recognition of the centrality of the above areas in studying gender disparities, the difficulty lies in operationalizing them for empirical study. Structured questionnaires may not reveal the subtle workings of gendered power within families. As Kabeer (1994) points out, it may be the silences and absences within the research encounter, the information that is withheld rather than volunteered, that signal the presence of disempowering relations. The use of qualitative methods of research may be significant here. A household survey may not be able to capture these elusive variables. Yet, three possibilities suggest themselves in this regard. First, it may be possible to identify the behavioural practices that reflect those aspects of gender ideology and identity that make women subordinate (for example, early marriage, fertility, son preference, purdah practices, restrictions on mobility, decision-making within the household, rigidity in the sexual division of labour, and so on). Researchers have usually worked along these lines. Secondly, it may be possible to assess attitudes among women and men such as attitudes towards female education, towards widows, towards wife-beating, towards those who transgress gender boundaries of behaviour and role such as single women, and so on. Thirdly, differences in socialization practices and in visions for the future of their sons and daughters could also throw light on this.

Assessing women’s status: examples

There have been attempts to assess women’s status as outlined above and, in order to illustrate, two such attempts are described below. Wieringa (1999) describes a measure called the Gender Equality Index that is in the process of being developed by researchers at and around the Institute of Social Studies at the Hague. The measure aims to provide a global index of the status of women

based on the power differences between men and women in any society. It consists of eight dimensions, each of which is to have indicators that are culturally relevant. The dimensions are gender identity (reflective of gender ideology such as rigidity of sexual division of labour and tolerance of transgender practices), autonomy of the body (as reflected in incidence of gender based violence, and control over sexuality and reproduction), autonomy within the family and household (as reflected in freedom to marry and divorce, decision-making power within the household), political power (in terms of decision-making in municipalities, unions, government, and the proportion of women in managerial positions), access to social resources (such as health and education), access to material resources (such as lands, houses, credit), employment and income dimension (in terms of gendered wage differentials, relative distribution of paid and unpaid labour), and time use (particularly relative access to sleep and to leisure).

The measure is promising and especially valuable in that it takes cognizance of the fact that the experience of inequality and disempowerment is multidimensional, and also of the significant role played by gender ideology in maintaining gender inequality. This dimension can include indicators that are culturally apt. Secondly, it recognizes the centrality of the body and sexuality in issues of empowerment. Thirdly, it is concemed with the household, an arena that has been seen as the last bastion of gender inequality, and includes resource allocation and domestic violence. Indicators on these dimensions, when considered along with the more conventional indicators, may help to diagnose where the crux of gender inequality lies in different societies.

In assessing the status of rural women in Karnataka, Batliwala et al (1998) use a conceptual framework that focuses on women’s access to, and control over, resources that confer power at individual, household and societal levels. Seven broad parameters are studied: women’s control over their labour and income, women’s access to public resources, women’s control over private assets and resources, women’s control over their bodies, women’s control over their physical mobility, their access to and control over political spaces, and their access to rights and legal redress. Gender identity is not directly assessed but is reflected in some behavioural practices as well as through studying some attitudes.

Gender disparities and men

Emerging from critiques that examine the gendered effects of structural adjustment, and the recent focus on men as equally gendered beings, gender relations have come into focus. The inclusion of men and masculinities in gender studies is argued for on the grounds that by recognizing their gendered identities we dethrone men from claims to represent normative humanity. Also there is a gender analytical need to understand male gender identity in more satisfactory terms than the crude stereotypes of pampered sons and tyrannical patriarchs and, of late, of irresponsible layabouts (Jackson, 1998). In order to work towards egalitarian feminist goals, gender analysis has to be rigorously applied to both men and women so as to understand men’s participation in daily life, the sexual division of labour, and the enduring puzzle of why women put up with male domination, or of how women, as agents, come to perceive their interests as partly served by male domination (Sweetman, 1998; Jackson 1998). It is necessary to understand, for instance, why women and men may have a vested interest in keeping up the illusion of traditional gender roles when the reality is otherwise. Thus among Lesotho women in Africa, predominantly wives of migrant miners, Sweetman (1998) found that they concealed their participation in income generation in order to maintain their relationship with the husband since this offered the best chance of survival and stability in the long term. Further, following retrenchment of the miners, domestic violence increased as women began to bring in more income. Not all beaten women condemned the violence; some explained it as an understandable reaction to unemployment. While it is clear that women need to be aware of the unacceptability of violence against them, the author argues that social policy must also respect the wish and need of women to live safely alongside men and that development must promote changes in male gender identity and in the construction of male gender roles. Taking another relevant example, the need for greater freedom of reproductive choice for women has led to the exclusion of men from the processes of reproductive choice and has in fact promoted gender inequality in reproductive responsibilities and costs. It is also to be recognized that men have practical and strategic gender interests which may, or may not, coincide with

those of women. Since men have interests, and are powerful, gender planning must be as explicit about men’s interests as it is about women’s interests (Thin, 1995).

The burdens of the male gender identity have of late come to the fore with the growing focus on men and masculinities. It is argued that the difference in the prevalence of psychopathology among men and women is a methodological artifact of the way that symptomatology has been classified in psychiatric classification systems. They contend that both men and women are equally stressed but that men tend to externalize their suffering through substance abuse and aggressive behaviour, resulting in an under-reporting of psychological distress in the form of ‘felt’ disorders such as depression and anxiety. The disability-adjusted life years data tabulated by the the World Bank reflect these differences. While depressive disorders account for about 26 per cent of the disability from neuropsychiatric disorders among women but only for about 10 per cent among men, alcohol and drug dependence account for about 26 per cent of the disability for men but only for about 6 per cent for women (World Bank, 1993). Brannon (1999) reviews research evidence and reports on the association between drinking and depression among men who are social drinkers as well as those who are problem drinkers. These studies indicate that many men drink to relieve negative mood states. She also cites a study by Horowitz and White (1991) which found that depression at age 21 was significantly related to alcohol problems at age 24 for men but not for women. Thus men may tend to use the strategy of drinking to cope with stresses which then puts them at risk for problem drinking.

Connell (1995) in his discussion of hegemonic masculinity also discusses its oppression of men who do not fit the mould. It is seen too that patriarchal systerns, while favouring men over women, also favour senior men over younger ones in a hierarchy that demands subservience, obedience and respect from those lower in the hierarchy. Seymour (1999) describes several longitudinal ethnographic studies in India which indicate that men are increasingly demanding more autonomy and control in the family. Thus conflicts between sons and senior male family members are becoming more overt. In some instances, this ‘rebellion’ is accompanied by more zealous control of women in an attempt to compensate for the challenge to tradition.

A central aspect of masculine gender identity is that of being the provider. Men do indeed experience shame at their failure to protect and provide for their families and depression is high among unemployed men. A high commitment to the breadwinner role can be a source of strain for men. Working in a Kenyan community, Silberschmidt (1991) found increasing alienation and mental illness as men lost their positions as breadwinners and community decision-makers thus experiencing a lack of fit in their conception of manliness and the external reality. In his study of the lives of labourers in India, Breman (1996) found that poor working men were clearly concerned about their obligations as fathers and as providers; he documents their shame and pain at not being able to fulfil this aspect of their gender role. Secondly, the work that most of these men had to do was back-breaking — stone quarrying, road construction, salt making, brick making — many could not keep up and felt burned out. The codes of manliness within such work prescribe that the men take up the work that is particularly onerous and hazardous while women act as helpers, even though women’s work is very demanding too. Consuming alcohol is seen as necessary to numb the exhaustion of extremely hard work. Thus among the working poor in India, it is possible that the responsibility of provider inherent in the ideas of manliness, entails both possibilities for domination as well as risks for self-exploitation, and high mortality and morbidity. These ideas are echoed in Jackson’s (1998) discussion of male gender identity where she shows how ideas of manliness link up with the nature of work. She points out that while one indicator of effort in work is the time it takes to perform, the indicator that looms large in local discourses is that of the strength and physical burden that the work entails. Among the lower socioeconomic groups, a code of masculinity exists where men are judged by their physical size, strength and capacity for hard manual work. She therefore calls for ‘including the sweat’, not as a means to validate male strength since women’s work also involves considerable physical effort, but as an approach that offers a fuller understanding of the gendered experience of work and the vulnerability it imposes. In her study of landless labourers in India, she used the Body Mass Index as an indicator of well-being, and finds that it is men in their reproductive years who are most nutritionally challenged and face early death.

Ahmad (1998) reporting on a seminar on men’s issues held in Pakistan, notes that the primary problem facing men was the social pressure to perform according to a predefined gender role. Men faced a tremendous pressure to prove their manliness in terms of providing economically for their families as well as in being protectors of the family honour. They felt under pressure to prove their virility among their male peers. As well, the absence of emotional outlets and support and the bar on the expression of soft, weak feelings permitted only one outlet — that of anger and violence. Finally, it was also noted that men felt that women’s dependence on them posed a substantial additional burden. Similar themes emerge from a male gender workshop in India (Seshadri, 1994) with men’s roles revolving around earning to support a family, of having to be the ones to handle the difficult and crisis situations at work, of having to be the initiator and active partner in sexual behaviour while at the same time feeling anxious about virility and, consequently, self-esteem.


Nonconventional indicators of gender disparity may be found both in measures that tap mental distress and violence among men and women, as well as in measures that tap into the disempowering aspects of the way gender is constructed and enforced in particular societies. The prevalence of mental distress and of abuse and violence are important indicators of the well-being of a community and subordination and it is shown that these are significantly differentiated by gender. Further, the ideology governing gender relations of power in a society structures the construction of gender identity and consequently of self-identity too. The socioeconomic changes wrought by structural reforms have the potential to disrupt existing notions of gender in ways that could be threatening, demoralizing and oppressive for men and women in some contexts and empowering in others. Some of these factors, especially those that concern gender ideology, may indeed be difficult to ‘measure’. It is therefore necessary to consider research methodologies that go beyond the quantitative in order to do justice to the complexity of these phenomena.


Abraham, M. (1998) ‘Speaking the Unspeakable: Marital Violence against South Asian Immigrant Women in the US’, Indian Journal of Gender Studies, 5, 215–41

Addlakha, R. (1997) ‘The Marginalization of Gender in Mainstream Psychiatric Theory and Practice’, Indian Psychologist, 2(1), 9–15

Agarwal, B. (1994) A Field of One’s Own: Gender and Land Rights in South Asia, Cambridge: Cambridge University Press

Agnes, F. (1988) ‘Violence in the Family: Wife Beating’ in R. Ghadially (ed.) Women in Indian Society: A Reader, New Delhi: Sage

Ahmad, K. (1998) Report on the Other Gender: A Seminar on Men’s Issues, Gender Unit, Islamabad: UNDP

Ahmed, A.S. and Z. Ahmed (1981) “Mor’ and ‘Tor’:Binary and Opposing Models of Pukhtun Womanhood’ in T.S. Epstein and R.A. Watts (eds.) The Endless Day: Some Case Material on Asian Rural Women, Oxford: Pergamon Press

Altomore, M. (1991) Gender and Adjustment. USAID, Office of Women in Development, Washington, DC. Abstracted in Gender, Society and Development (1995), Abstract No. 107, Amsterdam: Royal Tropical Institute

Bandura, A. (1992) ‘Self-Efficacy Mechanism in Psychological Functioning’ in R. Schwarzer (ed.) Self-Efficacy: Thought Control in Action, Washington: Hemisphere Publishing

Bandura, A. and K. Bussey (undated manuscript) ‘Social Cognitive Theory of Gender Development and Differentiation’

Banerjee. G., D.N. Nandi, S. Nandi, S. Sarkar, G.C. Boral and A. Ghosh (1990) ‘The Vulnerability of Indian Women to Suicide: A Field Study’, Indian Journal of Psychiatry, 32, 305–308

Banks, M.H., C.W. Clegg, P.R. Jackson, N.J. Kemp, E.M. Stafford and T.D. Wall (1980).’The Use of the GHQ as an Indicator of Mental Health in Occupational Studies’. Journal of Occupational Psychology, 53, 187–94

Bartlett, C.J. and E.C. Coles (1998a) ‘Psychological Health and Well-being: Why and How should Public Health Specialists Measure it? Part I: Rationale and Methods of Investigation, and Review of Psychiatric Epidemiology’, Journal of Public Health Medicine, 20(3), 281–87

Bartlett C.J. and E.C. Coles (1998b) ‘Psychological Health and Well-being: Why and How should Public Health Specialists Measure it? Part II: Stress, Subjective Well-being and Overall Conclusions.’ Journal of Public Health Medicine, 20(3), 288–94

Batliwala, S., B.K. Anitha, A. Gurumurthy and C.S. Wali (1998) Status of Rural Women in Karnataka, Bangalore: National Institute for Advanced Studies

Belle, D. (1990) ‘Poverty and Women’s Mental Health’, American Psychologist, 45(3), 385–89

Brannon, L. (1999) Gender: Psychological Perspectives, Boston, Allyn and Bacon

Breman, J. (1996) Footloose Labour: Working in India’s Informal Economy, Cambridge: Cambridge University Press

Broverman, T.H., D.M. Broverman, F.E. Clarksen, P.S. Rosenkrantz and S.R. Vogel (1970) ‘Sex Role Stereotypes and Clinical Judgment of Mental Health’, Journal of Consulting and Clinical Psychology, 34, 1–7

Bulow, D. von (1992) ‘Bigger than Men? Gender Relations and their Changing Meaning in Kipsigis Society’, Kenya, Africa, 62, 523–46. Abstracted in Gender, Society and Development (1995), Abstract No 7, Amsterdam: Royal Tropical Institute

Caplan, P. (1995) They Say You’re Crazy, New York, Addison-Wesley

Carstairs, G.M. and R.L. Kapur (1976) The Great Universe of Kota: Stress, Change and Mental Disorder in an Indian Village, London: Hogarth Press

Chakraborty, A. (1990) Social Stress and Mental Health: A Social-Psychiatric Field Study of Calcutta, New Delhi: Sage

Chakravarti, U. (1998) ‘Conceptualizing Brahminical Patriarchy in Early India: Gender, Caste, Class and State’, Economic and Political Weekly, 28(14)

Connell, R.W. (1995) Masculinities, Oxford, Polity Press

Cutler, S.E. and S. Nolen-Hocksema (1991) ‘Accounting for Sex Differences in Depression through Female Victimization: Child Sexual Abuse’, Sex Roles, 24, 425–38

Das, V. (1997) ‘Sexual Violence, Discursive Formations and the State’, Paper presented at the International Conference on Gender Perspectives in Population, Health and Development in India, New Delhi: NCAER

Datar, C. (1995) ‘Deterrents in Organizing Women Tobacco Workers in Nipani’ in S. Wieringa (ed.) Subversive Women: Historical Experiences of Gender and Resistance, New Delhi: Kali for Women

Davar, B.V. (1999a) Mental Health of Indian Women: A Feminist Agenda, New Delhi: Sage

— (1999b) ‘Moral Development and Mental Distress: A Feminist Proposal’ in T.S. Saraswathi (ed.) Culture, Socialization and Human Development: The Research and Application in India, New Delhi: Sage

Davis, D.L. and S.M. Low (1989) Gender, Health and Illness: The Case of Nerves, New York: Hemisphere Publishing

Dennerstein, L., J. Astbury and C. Morse (1993) ‘Psychosocial and Mental Health Aspects of Women’s Health’, Geneva: WHO

Desjarlais, R., L. Eisenberg, B. Good and A. Kleinman (1995) World Mental Health: Problems and Priorities in Low-Income Countries, New York: Oxford University Press

Diener, E. (1984) ‘Subjective Well-Being’, Psychological Bulletin, 95, 542-75

Eagly, A. (1995) ‘The Science and Politics of Comparing Women and Men’, American Psychologist, 50, 145–58

Elson, D. (1994) ‘Structural Adjustment with Gender Awareness?’ Indian Journal of Gender Studies, 1:2, 149–67

Ganesh, A. (1994) ‘A Preliminary Report of a Workshop Series and Survey on Child Sexual Abuse’, Bangalore: Samvada

Gelles, R.J. (1983) ‘An Exchange/Social Control Theory’ in D. Finkelhor, R.J. Gelles, G.T. Hotaling and M.A. Straus (eds.) The Dark Side of Families: Current Family Violence Research, New Delhi: Sage

Gelles, R.J. (1993) ‘Through a Sociological Lens: Social Structure and Family Violence’ in R.J. Gelles and D.R. Loseke (eds.) Current Controversies on Family Violence, New Delhi: Sage

Gelles, R.J. and M.A. Straus (1988) Intimate Violence: The Causes and Consequences of Abuse in the American Family. New York: Simon and Schuster

Gilligan, C. (1982) In a Different Voice: Psychological Theory and Women’s Development, Cambridge, MA: Harvard University Press

Goldberg, D.P. (1972) The Detecting of Psychiatric Illness by Questionnaire: A Technique for the Identification and Assessment of Non-Psychotic Psychiatric Disease. Oxford: Oxford University Press

Good, B. and M. Good (1982) ‘Toward a Meaning-Centred Analysis of Popular Illness Categories: “Fright Illness” and “Heart Distress” in Iran’ in A. Marsella and G. White (eds.) Cultural Conceptions of Mental Health and Therapy, Reidel, Dordrecht

Goodman, L.A., M.P. Koss, L.F. Fitzgerald, N.F. Russo and G.P. Keita (1993) ‘Male Violence Against Women: Current Research and Future Directions’, American Psychologist, 48, 1054–58

Guzder, J. and M. Krishna (1991) ‘Sita-Shakti: Cultural Paradigms for Indian Women’, Transcultural Research Review, 28, 257–301

Harris, M. and C.L. Landis (1997) Sexual Abuse in the Lives of Women Diagnosed with Serious Mental Illness, Amsterdam: Harwood Academic Publishers

Hirway, I. and D. Mahadevia (1996) ‘Critique of the Gender Development Index: Towards An Alternative’, Economic and Political Weekly, October, 87–96

Horowitz, A.V. and H.R. White (1991) ‘Becoming Married, Depression and Alcohol Problems among Young Adults’, Journal of Health and Social Behaviour, 32, 221–37

Isaac, M., A. Janca, K.C. Burke, J. deSilva, S.W. Acunda, A.C. Altman, J.D. Burke, C.R. Chandrasekhar, C. T. Miranda and G. Tacchini (1995) ‘Medically Unexplained Somatic Symptoms in Different Cultures: A Preliminary Report from Phase 1 of the WHO International Study of Somatoform Disorders’, Psychosomatics, 64, 88–93

Jackson, C. (1998) ‘Men’s Work, Masculinities and Gender Division of Labour’, ESRC Seminar Series, Seminar on Men, Masculinities and Gender Relations, University of Bradford

Jeffery, P. and R. Jeffery (1996) ‘Don’t Marry Me to a Plowman! Women’s Everyday Lives in Rural North India, New Delhi: Vistaar Publications

Jenkins, J.H. (1991) ‘The State Construction of Affect: Political Ethos and Mental
Health among-Salvadoran Refugees’, Culture, Medicine and Psychiatry, 14, 134–65

Jerusalem, M. and R. Schwarzer (1992) ‘Self-efficacy as a Resource Factor in Stress Appraisal Processes’ in R. Schwarzer (ed.) Self-efficacy: Thought Control in Action, Washington: Hemisphere Publishing

Kabeer, N. (1994) Reversed Realities: Gender Hierarchies in Developmental Thought, New Delhi: Kali for Women

Kagitcibasi, C. (1996) Family and Human Development across Cultures: A View from the Other Side, Hillsdale, NJ, Earlbaum

Kandiyoti, D. (1988) ‘Bargaining with Patriarchy’, Gender and Society, 2, 274–90

Kapadia, S. (1999) ‘Self, Women and Empowerment: A Conceptual Inquiry’ in T.S. Saraswathi (ed.) Culture, Socialization and Human Development: The Research and Application in India, New Delhi: Sage, 255–77

Kapoor, R. and G. Singh (1983) ‘An Epidemiological Study of the Prevalence of Depressive Illness in Rural Punjab’, Indian Journal of Psychiatry, 25, 110–14

Kapur, R.L. and A. Shah (1992) ‘A Psychosocial Perspective of Women’s Mental Health’ in Women in Development III: Gender Trainer’s Manual, Bangalore: Sakti

Kishwar, M. (1996) ‘A Half Step Forward: The Thwarting of Economic Reforms in India’, Manushi, 92–93, 51–80

Kodandaraman, P., K. Sathyavathy and H.N. Murthy (1983) ‘A Comparative Study of Suicides and Attempted Suicides among Women’, Indian Journal of Psychological Medicine, 6, 44–49

Koss, M.P. (1993) ‘Rape: Scope, Impact, Interventions and Public Policy Responses’, American Psychologist, 48, 1062–69

Lepowsky, M. (1994) ‘Women, Men and Aggression in an Egalitarian Society’, Sex Roles, 30, 199–211

Lewis, I.M. (1986) Religion in Context: Cults and Charisma, Cambridge: Cambridge University Press

Lukes, S. (1974) Power: A Radical View, London, Macmillan

Malamuth, N.M. (1996) ‘The Confluence Model of Sexual Aggression: Feminist and Evolutionary Perspectives’ in D.M. Buss and N.M. Malamuth (eds.), Sex, Power, Conflict: Evolutionary and Feminist Perspectives, New York: Oxford University Press

Mandelbaum, D.G. (1988) Women’s Seclusion and Men’s Honour: Sex Roles in North India, Bangladesh and Pakistan, Tucson: University of Arizona Press

Mane, P. (1993) ‘Mental Health of lndian Women: Realities and Needed Response’ in P. Mane and K. Gandevia (eds.), Mental Health in India: Issues and Concerns, Mumbai: TISS Publications

Maynard, M. (1993) ‘Violence Towards Women’ in D. Richardson and V. Robinson (eds.), Introducing Women’s Studies, New York: Macmillan

McBride, A.B. (1990) ‘Mental Health Effects of Women’s Multiple Roles’, American Psychologist, 45(3), 381–84

Mehta, S. and U. Dighe (1991); ‘Review of Journals in Regional Languages’ in M. Krishnaraj (ed.), Women and Violence: A Country Report, A Study Sponsored by UNESCO, Research Centre for Women’s Studies, Mumbai, SNDT Women’s University

Mies, M. (1982) The Lace Makers of Narsapur: Indian Housewives Produce for the World Market, London: Zed Books

Mukhopadhyay, S. and R. Savithri (1999) Poverty, Gender and Reproductive Choice: An Analysis of Linkages, New Delhi: ISST

Mukhopadhyay, S. and R. Sudarshan (1997) ‘Gender Discrimination in an Age of Liberalization: Some Issues,’ paper presented at a workshop on Gender Discrimination under Structural Reforms, New Delhi: ISST

Nagpal, R. and H. Sell (1985) Subjective Well-being. SEARO Regional Health Paper No. 7, New Delhi: WHO

Nandi, D.N., N.N. Das, A. Chaudhuri, G. Banerjec, P. Datta, A. Ghosh and G.C. Boral (1980) ‘Mental Morbidity and Urban Life-An Epidemiological Study’, Indian Journal of Psychiatry, 22, 324–30

Nath, J. (1981) ‘Beliefs and Customs.Observed by Muslim Rural Women During Their Life Cycle’ in T.S. Epstein and R.A. Watts (eds.), The Endless Day: Some Case Material on Asian Rural Women, Oxford: Pergamon Press, 13–28

Neuegebauer, D.D., B.P. Dohrenwend and B.S. Dohrenwend (1980) ‘The Formulation of Hypotheses About the True Prevalence of Functional Psychiatric Disorders Among Adults in the US’ in B.P. Dohrenwend, B.S. Dohrenwend, M.S. Gould, B. Link, R. Neuegebauer and R. Wunsch-Hitzig (eds.), Mental Illness in the US, New York: Praeger, 45–94

Nichter, M. (1982) ‘Idioms of Distress: Alternatives in the Expression of Psychological Distress: A Case Study from South India’, Culture, Medicine and Psychiatry, 5, 379–408

Niranjana, S. (1999) ‘Of the Body. Further Considerations on Women, Sexuality and Agency’, Indian Journal of Gender Studies, 6, 1–20

Nolen-Hoeksema, S. (1987) ‘Sex Differences in Depression: Evidence and Theory’, Psychological Bulletin, 101, 259–82

Parikh, I. and P. Garg (1989) Indian Women: An Inner Dialogue, New Delhi: Sage

Purewal, J., D. Bhuyan, I.M. Ganesh and S. Sanyal (1999) In Search of Her Spirit: A Report on Women, Violence and Mental Health, New Delhi: IFSHA

Pyne, H.H. (1999) ‘Reproductive Experiences and Needs of Thai Women: Where Has Development taken Us?’ in G. Sen and R.C. Snow (eds.), Power and Decision: The Social Control of Reproduction, Harvard University Press, 19–41

RAHI (1998) Voices from the Silent Zone: Women’s Experiences of Incest and Child Sexual Abuse, New Delhi, RAHI

Ram, K. (1992) Mukkuvar Women, Gender Hegemony, Capitalist Transformation in a South Indian Fishing Community, New Delhi: Kali for Women

Risseeuw, C. (1991) Gender Transformation, Power and Resistance among Women in Sri Lanka, New Delhi: Manohar

Rosenfield, S. (1989) ‘The Effects of Women’s Employment: Personal Control and Sex Differences in Mental Health’, Journal of Health and Social Behaviour, 30, 77–91

Ross, C.E. and J. Huber (1985) ‘Hardship and Depression’, Journal of Health and Social Behaviour, 26, 312–27

Rowlands, J. (1997) Questioning Empowerment, Oxford, OXFAM

Schuler, S.R. and S.M. Hashemi (1993) Defining and Studying Empowerment of Women: A Research Note from Bangladesh, JSI Working Paper No 3, Arlington, Virginia, 1

Seidman, E. and B. Rapkin (1983) ‘Economics and Psychosocial Dysfunction: Towards a Conceptual Framework and Preventive Strategies’ in R.D. Felner, L.A. Jason, J.N. Moritsugu and S.S. Farber (eds.), Preventive Psychology, New York: Pergamon Press, 175–98

Sell, H. and R. Nagpal (1992) The Assessment of Subjective Well-Being, WHO Regional Health Paper, SEARO, No. 24

Seshadri, S. (1994) Report on Male Gender Workshop, Unpublished document, Department of Psychiatry, Bangalore: NIMHANS

Seymour, S.C. (1999) Women, Family and Child Care in India: A World in Transition, Cambridge: Cambridge University Press

Shah, A. (1996) ‘Gender Issues in Mental Health: A Clinical Psychology Perspective’, Paper presented at the National Seminar on ‘Women in India and their Mental Health,’ Hyderabad

Shamsunder C., S. Krishna Murthy, O. Prakash, P. Prabhakar and D.K. Subbakrishna (1986) ‘Psychiatric morbidity in a general practice in an Indian city,’ British Medical Journal, 292, 1713–1715

Shukla, G.D., B.L. Verma and D.N. Mishra (1990) ‘Suicide in Jhansi City’, Indian Journal of Psychiatry, 32, 45–51

Silberschmidt, M. (1991) ‘Rethinking Men and, Gender Relations: An Investigation of Men, their Changing Roles within the Household and the Implications for Gender Relations in Kisii District, Kenya’, CDR Research Report No 16, Abstracted in Gender, Society and Development (1995), Abstract No. 55, Royal Tropical Institute, Amsterdam

Singh, A.T. and P. Uberoi (1994) ‘Learning to ‘Adjust’: Conjugal Relations in Indian Popular Fiction’, Indian Journal of Gender Studies, 1, 93–120

Sriram, R. (1991) ‘Research on Marital Violence: Some Trends, Implications and Emerging Issues’ in Unit for Family Studies, TISS (eds) Research on Families with Problems in lndia, Vol. II, 483–95, Mumbai: TISS Publications

Strandh, M. (2000) ‘Different Exit Routes from Unemployment and their Impact on Mental Well-being: the Role of the Economic Situation and the Predictability of the Life Course,’ Work, Employment and Society, 14(3), 459–79

Straus, M.A. (cited in press in Gelles, 1993) ‘State to State Differences in Social Inequality and Social Bonds in Relation to Assaults on Wives in the US’ in
R.J. Gelles (ed.), Family Violence, Special Issue, Journal of Comparative Family Studies

Sweetman, C. (1998) Sitting on a Rock: Integrating Men and Masculinities into Gender and Development, ESRC Seminar Series, Seminar on Men, Masculinities and Gender Relations, University of Bradford

Thin, N. (1995) ‘A Critical Review of Women’s Status and Rights’ in Gender, Society and Development, Advancing Women’s Status: Women and Men Together? Royal Tropical Institute, Amsterdam, 22–36

Venkobarao, A. (1989) Psychiatry of Old Age in India, Madras: Macmillan India Press

Vishwanath, K. (1997) ‘Shame and Control: Sexuality and Power in Feminist Discourse in India’ in M Thapan (ed), Embodiment: Essays on Gender and Identity, New Delhi: Oxford University Press, 313–33

Vlassoff, C. (1994) ‘From Rags to Riches: The Impact of Rural Development on Women’s Status in an Indian Village’, World Development, 22(5), 707–19

Walker, L. (1993) ‘The Battered Woman Syndrome is a Psychological Consequence of Abuse’ in R.J. Gelles and D.A. Loseke (eds.), Current Controversies on Family Violence, New Delhi: Sage, 133–53

Warr, P., P. Jackson and M. Banks (1988) ‘Unemployment and Mental Health: Some British Studies’. Journal of Social Issues, 44, 47–68

Wieringa, S. (1999) ‘Women’s Empowerment in Japan: Towards an Alternative Index on Gender Equality,’ Unpublished, Institute of Social Studies, The Hague

Wolf, B. (1987) ‘Low-income Mothers at Risk: The Psychological Effects of Poverty Related Stress,’ Unpublished Doctoral Dissertation, Harvard Graduate School of Education

World Bank (1993) World Development Report, 1993: Investing in Health, New York: Oxford University Press

Yllo, K. (1993) ‘Through a Feminist Lens: Gender, Power and Violence’ in R.J. Gelles and D.R. Loseke (eds.), Current Controversies on Family Violence, New Delhi: Sage, 47–62

Household Response to Gender Issues:
A Survey on Households of Female EPZ
Workers in Bangladesh




Background of the study

Export-oriented industrialization has been encouraged in Bangladesh through bonded-warehouse facilities and back-to-back letters of credit since the late 1970s. One of the instruments used for this was the setting up of Export Processing Zones (EPZs). Female employment has increased because of the expansion of labour-intensive export industries both inside and outside the EPZs. Although a wider definition of EPZs would include instances where there is no physical zone as such, for the purpose of this study, an EPZ will be taken to be an industrial estate, which constitutes an enclave within the trade and customs regime of a country and in which free trade applies.

*The author is grateful to International Development Research Centre, Canada for funding the project and to ISST, India for making it possible. She is grateful to Mr Mohammad Yunus for his contribution in drafting section 3. This research would not have been possible without the cooperation of the household members of the female workers of Dhaka EPZ. They deserve special thanks. Thanks are also due to the survey team for conducting the survey and to Mr Razaul Hoque Mondal and Mr Abdus Samad Akhand for data processing. The responsibility for any errors rests with the author.

There are two EPZs currently operating in Chittagong and Dhaka, which were set up in 1983 and 1993 respectively. This study is based on a survey of members of the households of female workers working in the EPZ located in Dhaka (Savar). The Dhaka EPZ (DEPZ) is about 35 kilometers to the north of Dhaka city.

The analysis presented here is based on data drawn from an in-depth semi-structured questionnaire canvassed to a total of 112 households and 246 members of households. The sample included members of the selected households who were above twelve years of age. A maximum of four members was interviewed from each household. The survey was conducted between June and October 2000 in different locations within 12 kilometers around the DEPZ. The analysis compares the responses of 98 male and 166 female members of the households.

The main problem encountered was in trying to meet all the members of the households in order to interview them, as they were not always available at the same time. However, most of the workers of the DEPZ were available on Fridays, the weekly holiday. In many cases, several visits had to be made to the same household.

In this essay I begin, first, by analysing the household structure of the female workers working in the DEPZ. I then go on to describe the structure of employment and control over work-related decisions. This is followed by an examination of intra-household changes. Social perceptions on education, health, physical mobility, marriage and dowry and reproduction and sex are dealt with after this, and I then go on to look at violence and mental health.

Household Structure

The growth of EPZs in Bangladesh had several spillover effects. Factories are located within the EPZs but there are no residential facilities. Nor do employers provide workers with accommodation. Many local families have stepped in here and providing housing for EPZ workers has become a good source of income for many of them. Most of the houses constructed for this purpose are made of

brick with a tinned roof. As these were especially designed for EPZ workers, care was taken to provide adequate toilet and bathing facilities, and there seemed to be no problem as far as queuing for the use of toilets, or cooking was concerned. The workers felt secure in the new communities which have developed in this way at different locations around the DEPZ.

Survey findings revealed the use of health care facilities to be as follows: private/NGO clinic (37 per cent), Gono Shasthyo Kendro, an NGO hospital (22 per cent), local hospitals (9 per cent), thana hospitals (5 per cent), and the hospital of the Bangladesh Export Processing Zone Authority (BEPZA) (5 per cent). The average distance to healthcare facilities was 2 kilometers. Only a-third of the households sent their children to school: some 59 per cent had children in primary school and 22 per cent in secondary school. The average distance from school was 1.5 kilometers.


It was widely believed that women’s migration in the early eighties was low due to purdah. In fact, it was the growth of the labour intensive readymade garment industry that led to the migration of women from rural to urban areas, and their earnings became central to the livelihoods of many households. About 16 per cent of the members in our survey were local to Savar where they had lived since birth. Of the migrant workers, about 55 per cent were from Dhaka Division, 17 per cent from Barisal Division, 12 per cent from Khulna Division, 10 per cent from Rajshahi Division and 6 per cent from Chittagong Division.1 This has had far-reaching implications for the lifestyles of women in Bangladesh. About 68 per cent of the female migrants came either with their husbands or with their parents or brothers/sisters.

Information about the availability of work (for women) in the garment industry reaches rural areas through the kinship network which consists of family, friends, neighbours etc. Employers have used this type of job information through contacts as a strategy for recruiting workers. The major districts sending female workers to

1As mentioned earlier, there is another EPZ in Chittagong, due to which migration from this Division is relatively low.

the DEPZ include Dhaka, Faridpur, Gazipur, Jamalpur, Manikganj, Mymensingh, Tangail, Kushtia, Magura, Barisal, Perojpur, Bogra, and Rangpur. Most of the migrant workers return to their villages once a year.

Household profile

A household has been defined to include a group of individuals who eat together on a daily basis. This definition gave an average household size as 3.3, a figure that is much lower than the urban national average of 5.6 found in the Household Expenditure Survey, 1991. About 28 per cent of the households were female-headed. This is about four times higher than the national figure for urban areas of 7 per cent found in the Labour Force Survey, 1989.

Most of the households had young members: with only 26 per cent of the members being above the age of 30 and about 20 per cent being below the age of 14. About half the adult members were currently married and 40 per cent were unmarried. It is important to note that more women (61 per cent) were regular earners than men (49 per cent). Most of the members are educated: 15 per cent of men and 24 per cent of women had no education, while 44 per cent of men and 24 per cent women had education above secondary level.

Economic activities

The traditional concept of the sexual division of labour in our society considerably limits the outside activities of women. The rising participation of women in the labour force shows that there has been a major change in social norms. Of the regular workers, most men (67 per cent) and women (94 per cent) were employed in the manufacturing sector. The rest of the men were employed in the retail trade and the service sectors. In terms of labour status, more women (98 per cent) than men (79 per cent) were salaried workers. About 10 per cent men were self-employed and another 8 per cent were wage earners. The average monthly income of men in these households was Tk. 3416 while that of women was Tk. 2352 (including all sources of income).

Employment and Control Over Work

Structure of employment

The increasing flows of investment have generated an opportunity of employment for both unskilled and semi-skilled workers in the DEPZ. The high incidence of female employment may be explained by the mushroom growth of readymade garment factories (RMG) both in and outside the DEPZ. Around 95 per cent of the female workers reported having been employed in the garment factories compared to 43 per cent of their male counterparts. More than 12 per cent of the male workers reported having been employed in the textile factories. Around 20 per cent of the male workers appeared to have been employed in transport and construction activities.

The workers received information about job vacancies from several sources. Around half of them reported that they had obtained information about their present jobs from friends/relatives and/or neighbours. Around a third of the respondents reported having entered the current job by self-searching on their own or looking at the vacancy notices posted at the gate of the DEPZ.

Mostly, workers had made independent decisions to take up one or other job — this figure was as high as 86 per cent for men and more than 50 per cent for women. Some of the female respondents also said they had sought the guidance of their husbands and parents. Thus, there seems to have been some kind of change in the mindset of these families regarding female employment.

The source of women’s skill acquisition may be better understood by tracing the history of their previous occupations. Interestingly, around 50 per cent of those women who were previously not in the labour force had found a job during this period. More than 90 per cent of those previously employed female workers were employed in the RMG sector. The situation is somewhat different for male workers: there appears to have been a displacement of male workers from some other sector of the economy to the RMG sector; around 10 per cent of the male workers who were employed elsewhere took up garment-related jobs.

Working conditions

Although several factors could have played a significant role in causing the respondents to shift away from previous occupations, the insecurity of temporary jobs appears to have been the main driving force, especially for men. More than half of the male workers were employed in temporary positions in their previous occupations. The change of job of around half of the female workers, however, cannot be explained by the insecurity of the previous job as they were permanently employed earlier. The dominant reason for leaving their previous jobs for both men and women turned out to be the low wage received (40 per cent). Less than 20 per cent of the workers reported long working hours and/or night duty and/or unsatisfactory rules and regulations at the workplace as reasons for leaving the previous job. More importantly, more than 25 per cent of the female workers had to leave their previous occupation due to illness or family reasons such as pregnancy, marriage etc. These problems did not seem to have affected the male workers to quite the same extent. Conversely, a small percentage of male workers had to quit their previous jobs as they were either thrown out by the management or could not tolerate their abusive behaviour.

There is a gender gap in the average length of service in the current job: female workers were employed for 27 months and male workers for 36 months. Workers reported several other benefits apart from salary and overtime compensation. The female workers appeared to have been employed in low paid jobs, but received benefits more frequently than their male counterparts-except medical coverage (31 per cent for female worker vs. 41 per cent for male worker). While more than 90 per cent of the female workers were able to get a weekly rest by rotation, 66 per cent of their male colleagues did not enjoy such a facility. Similarly, 86 per cent of the female workers received a festival bonus, as against 66 per cent of their male co-workers. It may be noted that there is hardly any female worker who did not receive some type of ‘other’ benefits whereas the incidence of this is 23 per cent for male workers.

Almost all the female workers and more than two-thirds of the male workers received their wages on a monthly basis. In the majority of cases, the management fixes the salary without any

negotiation with the worker. Most of the female workers reported having fixed working hours, this figure was less than 50 per cent for male workers. More than 60 per cent of the female workers have to work eight or more hours a day. The incidence is much lower for male workers; a little over 25 per cent of male workers work more than eight hours a day.

Around 90 per cent of the workers do not work in shifts. The monthly salary, including all perks, stood at Tk. 2332 for female workers and Tk. 3106 for male workers. This included an overtime payment of around 20 per cent. The incidence of overtime duty is more pronounced among female workers than their male counterparts. Whereas more than 90 per cent of the female workers reported working overtime, only 55 per cent of their male counterparts did so. However, the male workers reported having worked overtime longer (in a month) than their female co-workers, 49 hours a month for an average male worker compared to 46 hours a month for an average female worker. Most workers do not object to overtime duty till it becomes too stressful as it means getting some extra money at the end of the month.

Almost all of the workers said that they were allowed a break for lunch. One of the stringent conditions imposed on the workers during lunch/tea breaks is that they are not allowed to leave the factory premises. Only 15 per cent of the respondents seemed to be aware of trade union activities in the enterprises in the EPZ. The female workers showed a lower awareness than their male co-workers; while about one-third of the male workers have been aware of trade union activities, hardly any female worker was aware of such activities. The low incidence of trade union activities in the EPZs may be attributed to the government ban on such activities in the EPZ in order to attract foreign investment.

Control over changing occupation

A significant proportion of the workers reported that it is necessary to consult others before entering a new job or changing jobs. The person consulted is generally the husband in the case of female workers, and the parents and wife in the case of male workers.

Less than 20 per cent of the workers reported having encountered barriers in changing jobs. The main barrier

encountered by a typical female worker was the lack of requisite skill. But the main problem facing male workers in job mobility is the lack of adequate money. However, due to the mushrooming of factories, particularly RMG factories, it is not very difficult to change jobs frequently, especially for women.

Around one-third of the female workers faced disapproval if they had to work night shifts. More than 40 per cent of the male respondents faced disapproval for entering jobs that bring disrepute to the individual worker himself or to the family.

Intra-Household Changes

Control over assets

Contrary to expectations, the survey findings showed that 25 per cent men and 9 per cent women owned land and 34 per cent men and 8 per cent women owned houses. A small proportion of household members owned a bicycle. The households surveyed were also found to be good at saving for a rainy day; they saved in different forms. Women put their savings into jewellery, while men put their money into banks.

These households possess a number of consumer durables. A large number of women (40 per cent) do not own any consumer durable, but the proportion of such men was only 10 per cent. About 50 per cent of men and women possessed furniture, electric fans and TVs. Consumer durables and jewellery are the main assets over which household members could exercise some personal autonomy.

About 17 per cent of the men and 35 per cent of women do not have control over their assets, that is, they cannot dispose of any assets on their own. While men can dispose of consumer durables, jewellery, house/flat and bank deposits, women can only do so for jewellery and consumer durables. Not only is there a gender difference in selling assets, but more women (75 per cent) than men (63 per cent) also have to take permission to buy assets. Of those who have to seek permission, surprisingly, about 70 per cent of men seek permission from their wives, while 58 per cent of women do so from their husbands. Thus there is cooperation between the spouses in buying assets. This is likely to reduce conflicts in the

household. More interestingly, similar proportions of men and women have to take permission from their parents/in-laws.

About one-fourth of female respondents and 5 per cent of men said they could not purchase anything on their own without taking permission from anybody. Hence, even though most women were earning money, there are greater restrictions on their spending than on that of men. Most of the women were able to exercise choice only in the purchase of clothes, cosmetics and household essentials. For purchases involving a considerable amount of money, permission had to be sought from husbands or elders. Hence, these women are still bound by traditional expectations and roles.

Control over income

It is important to note that similar numbers (more than 80 per cent) of men and women contributed at least half their earnings for household expenditure. Hence, women are becoming the breadwinners of their families, although they still lack control over their income.

About 61 per cent of the male earners had full control over their earnings. The proportion for the female earners was only 36 per cent. Women have to hand over their earnings either to their parents or to their spouse.

The contribution to family income of unmarried girls has often resulted in a delay in their marriage. In some cases the whole family is dependent on the woman’s income. Although women contribute to household income, many of the unmarried migrant girls did not tell their families in the village that they were working in a factory in town. This deliberate concealment of facts prevails because working in factories is still looked down upon by people in the rural areas.

Control over labour use pattern

Employment opportunities at the EPZ have increased the work burden for most women. Factory work has also made women more visible. Their unpaid work in the household is now gaining more recognition. Married women, however, still have to manage household chores along with their factory work. In contrast to the norms, some husbands have started to share the household

workload. This is definitely a positive change. During the interviews, it was even observed that some husbands were helping in chopping meat, cooking rice, etc. They also washed clothes, provided nobody was around while they were doing so. In these households 40 per cent of men did at least some cleaning and 22 per cent did some cooking. About half of the men reported that they never did any cooking. This unusual sharing of household chores by men has become possible as they are living away from their traditional family environment.

If any of the female members became ill the male members often did the household chores. They did the cooking, washing of clothes and also took care of the ailing members of the family. About one-fifth of men and one-third of women reported that men would do everything if necessary. Only 5 per cent of men and 10 per cent of women said that the man does nothing. The activities that men would never do included washing women’s dirty clothes, washing dirty utensils and cleaning toilets.

Control over household decision-making

The most discernible impact earning an income has had on women’s lives is that they have acquired decision making power. Quite interestingly, there was not much gender difference in responses with regard to household decision-making. About 98 per cent of men and 87 per cent of women responded that they participate in household decision-making. Participation in decision-making was higher for men than for women for all the indicators used. Women had relatively more autonomy in making decisions relating to daily provisioning, employment-related matters and buying and selling of jewellery. Around 30 per cent of women reported that they participated in decision-making, but that their decisions were not final.

Relative to women (72 per cent) fewer men (31 per cent) suggested that women in their family ought to have more decision-making power than they have now. The main reasons cited were women should have equal say because they manage the household well and they too can make good decisions. The response by men was higher on all these issues than by women. On the issue of whether women should have more decision-making power in the community, about two-thirds of the women and a third of the men

thought that they should do so. Interestingly, the main reason cited by all men and 85 per cent of the women was that women should have equal rights and opportunities. The above observation suggests that women in these households are quite aware of equal rights and opportunities. These findings also suggest that many men do not want women to have more power in the community but those who do, also support the idea of gender equality. According to some men, women’s decision-making in the community does not need to increase. Those who do not favour women’s empowerment argue that it is the responsibility of men to take decisions. Some of the other reasons cited include women being less knowledgeable and less experienced and hence ill equipped to participate in the public sphere. Men also felt that more decision making by women would affect man’s dignity.

Attitude to female employment

It is important to note that both men and women felt that positive changes do take place when women begin to work outside the home. In most cases the response was that employment provides more economic security for the women. On the other hand, when a woman starts to work outside the home, more men (28.6 per cent) than women (17.5 per cent) felt that no negative change takes place. The most important negative changes perceived by both men and women were conflict over domestic chores and the neglect of children. More importantly, a few male and female respondents reported ‘threat to joint family’ as a negative change.

Moreover, when a woman starts working outside the home, there are perceptible changes in her lifestyle. In general her total workload, standing in the family and influence over household decision-making increases more than her mobility and her self-esteem. This pattern was reflected in the responses of all members of the household. Relative to women, more male members feel that women’s mobility and self-esteem decrease when they start working outside. According to them, because of the pressure on time, women’s mobility is limited to going to work and back. As work in the RMG factory is usually looked down upon in society, so her self-esteem also declines when she works in the DEPZ.

Changes in Perceptions

Attitude toward education

The families involved in the study looked upon higher education positively. Women in the DEPZ were among the first generation to begin reaping the benefits of expanding educational opportunities for females. Across all income groups, the higher education of daughters has been linked with prestige. Women working in the DEPZ were often unable to move up in their jobs because of their lack of higher education. Hence, they wanted their sons and daughters to have higher levels of education so that they would not have to work in the garment factories as they themselves have done. In most cases (more than 93 per cent) the response was that males and females should be given same level of education. In their families similar educational opportunities have not been offered to boys and girls in general. Hence the attitude towards education has become gender neutral in this generation of workers’ households.

When we posed the question about who would be the first to be withdrawn from school if there were financial or other constraints there seemed to be no gender difference. That is, both father and mother agreed that they would withdraw the child who was not doing well in school. However, if all children were doing equally well, both felt that they would first withdraw the girl child. Quite surprisingly, relative to female members more male members reported that they would continue the education of children in any event.

Health and nutrition

Contrary to the cultural norms that men and women do not eat together, it was noted that both men and women in most of the households did eat together. Among those who do not a few said that this was due to tradition. Most often it became necessary because household members had different schedules. Boys and girls in these households eat together.

The findings suggest that there was no gender difference in the incidence of illness during the last six months: 64 per cent for men and 66 per cent for women respectively. The most frequent illnesses reported by family members were fever, cough/cold and

gastro-intestinal ulcers. But more women suffer from fever and more men suffer from cold/cough and gastro-intestinal ulcers. Men are also prone to injuries. Both men and women use local dispensaries for seasonal and common diseases such as fever and colds/coughs. Qualified physicians such as MBBS doctors were consulted only when the illness was serious — for example, waterborne or airborne diseases, or typhoid, cholera, malaria, tuberculosis, and chicken pox.

Physical mobility

Women’s mobility was found to be very limited. A higher proportion of women (60 per cent) than men (46 per cent) felt that girls have less freedom of movement than boys do. But the reasons put toward for lower mobility differed by gender: 69 per cent of men felt that this was so because women’s mobility sends bad signals, 55 per cent felt that women are less mobile because they fear harassment, and 53 per cent felt that family prestige was also an important factor. Women, on the other hand, thought differently: 61 per cent of them put the reason at sending bad signals, 47 per cent felt it was against tradition/custom and 23 per cent listed fear of harassment. Hence, the perception of social norms by men and women is different. What women regard as maintaining tradition/custom may seem like fear of losing family prestige for men.

Regarding the mobility of women outside the household, it was noted that going to banks (0.4 per cent), the market (0.8 per cent) and the parental home (2 per cent) was very limited. Most women did not go to the mosque, restaurant/cinema or the market. But going to work did not seem to be a problem as most women travel in a group. This in-group mobility provides the required security while travelling to and from work, including returning home after dusk. In most cases women could also go alone to work — 74 per cent reported that this was possible. They did not need permission to do so, or to stay out late when the reason was work-related. But they did need permission if they were out on a visit to their parental home; if they wished to go to an educational institution, about half the women said they would need to seek permission.

Marriage and dowry

Most of the male and female workers agreed that a good age to get married was 18 years (or above). The mean age at marriage for men was found to be 23 and for women, 16.

Despite the fact that Anti-Dowry Act makes dowry a punishable offence, the taking and giving of dowry has been flourishing to an alarming degree. Dowry is demanded not only at the time of marriage, in many cases grooms demand dowry several times after marriage. It can be paid in cash, but also in the form of gold ornaments and furniture, etc. Of those who were ever-married, 62 per cent of men and 72 per cent of women reported that no dowry was paid during their marriages.

Most members preferred their sons to marry as soon as they started earning and preferably between 20-25 years of age. On the other hand, daughters are generally married off at the age of eighteen. More women (65 per cent) than men (36 per cent) wanted to give dowry for their daughter’s marriage. A few men (10 per cent) and women (12 per cent) wanted to demand dowry for their son’s marriage. About 70 per cent of men and women felt that dowry demands had increased in their community. The main reason given for this increase by men was that the number of girls has increased (23 per cent) while women felt that it has become a custom (27 per cent). The amount of ‘dower’ has also increased. But claiming the rights of ‘dower’ is almost non-existent in our society.

Interestingly, an overwhelming majority of women (61 per cent) think that if a woman is unhappy in her marriage she can move out to live separately. But only 20 per cent of men think the same. There were gender differences in the circumstances under which it was seen as acceptable for a man to leave his wife. The men suggested that they could do so if women were unfaithful (82 per cent), or disloyal (46 per cent) or if they were poor housewives (43 per cent). But the women suggested that men who leave do so if women are unfaithful (60 per cent), or if their families do not give the promised dowry (44 per cent) or if women are infertile (23 per cent). The gender difference reflects the different situation of men and women in society i.e. women have to adapt more in their married lives than men.

Reproduction and sex

Of those ever-married, about 71 per cent men and 55 per cent women have adopted family planning methods. There was a clear preference for a son by male members. About 36 per cent men and 54 per cent women did not express any sex preference. The reason for son preference also varied by gender. Men wanted sons as the latter are believed to look after parents during old age (56 per cent), to keep the family heritage (20 per cent) and women giving birth to sons get more respect (22 per cent). On the other hand the reason given by women was mainly that sons look after their parents when they grow old (26 per cent). Household members were aware of amniocentesis but none had used it to abort a female foteus.

More than 75 per cent men and women felt that it was very important for the husband to find sexual satisfaction. But there was a gender difference regarding women’s sexual satisfaction. Although 82 per cent men thought it to be very important, only 38 per cent of women thought so. More women (76 per cent) than men (70 per cent) reported that they could say ‘no’ to their spouse if they did not desire sex. But a larger proportion of women (15 per cent) than men (6 per cent) reported having been subjected to verbal/sexual abuse when they refused to have sex.


It is important to find out the level of stress within these households. Most of the women members work, and it is important to ask whether employment for women results in or leads to an increase or reduction in violence and sexual abuse, as well as how it impacts on power relations and self-identity.

Violence at the workplace

The survey noted that a large number of employed men and women suffered from work-related harassment in their respective workplaces: 44 per cent of men and 47 per cent of women. The most common types of work-related harassments were: heavy, additional workloads to handle (34 per cent and 43 per cent respectively for men and women), being made to work extra hours without extra pay (20 per cent and 27 per cent respectively). In

addition men suffered from being given work that requires them to stand for long hours (34 per cent), while women suffered from being shifted repeatedly from their normal place of work (35 per cent). Four women (3 per cent) have faced sexual harassment at their workplace. This was mainly touching or brushing against parts of the body, and molestation. Moreover, about 4 per cent men and 17 per cent women reported that someone in their workplace had complained of sexual harassment. Most workers were not satisfied with the response of the management and fellow workers to such incidents.

Violence in the family

Most household members reported that there was no violence within their own household, but they reported violence among the neighbouring households. About 67 per cent men and 84 per cent women reported that they noticed conflicts in families in the neighbourhood. The reason for such conflicts differed by gender. Men reported quarrels with the spouse and her family (64 per cent), fights over property (44 per cent), suspicion (32 per cent), extra marital affairs (29 per cent), talking back (27 per cent) and dowry demands (21 per cent). Women emphasized fights with the spouse and his family (68 per cent), talking back (34 per cent) and fights over property (31 per cent).

Coping strategies adopted by household members to deal with violence in the family also differed by gender: the men mostly went for compromise (30 per cent), or talking things out or walking away from the situation (19 per cent), or going out (13 per cent). But women preferred to keep quiet (39 per cent) and talk it out (25 per cent). Quite surprisingly, about three-fourth of men and half of women reported that they do not face physical abuse in the family. This appears to confirm the belief that people tend to conceal violence even when they are subjected to it. Both men and women reported the frequent incidence of verbal abuse. Women are beaten and verbally abused more often than men. A small proportion of women (6 per cent) has also been threatened with being sent back to their maternal homes. Such violence also often results in the destruction of various household items. In this context, it was found that men destroy household items more than women do.

If subjected to abuse, 50 per cent of men and women reported that they would tolerate it. About 41 per cent of men and 27 per cent of women reported that they would keep it to themselves. The third strategy was to let family members know about it. Some women reported that they would go back to their maternal homes if such incidents occur.

About 26 per cent men and 8 per cent women have faced violence outside the domestic sphere. The context for men was differences over property (30 per cent) and political quarrels (30 per cent). On the other hand the context for women was harassment of girls (53 per cent) and fights over property (20 per cent).

Mental Health

Human well-being includes physical as well as mental well-being and these are interrelated. Even though past studies have mostly looked at physical levels, a few have looked at mental levels, but these were piecemeal analyses. The extent of stress can also be related to violence and sexual abuse as violence is a cause for many mental and physical health problems.

Mental distress and well-being

Two measures of mental health and well-being were used in the survey. One is the shortened version of the Goldberg’s General Health Questionnaire (GHQ) to measure mental distress or stress level and the second is the Subjective Well-Being Inventory (SUBI) to measure mental well-being. Mental disorders result from the combined effect of a number of stress stimuli. The GHQ comprises 12 questions. Scoring is done on a bimodal scale for each question. The four options for each question are scored 0,0,1,1 respectively. Hence, the total score is the sum of scores on each question. The maximum possible score is 12, while the minimum possible score is zero. In general, a score of 2 and above is suggestive of significant distress. On the other hand, the SUBI had nine questions. The response on each question was graded, with the level of mental well-being going down as one moves down the options. Scoring for SUBI is done on a three-point scale for each question. Lower scores indicate greater well-being. The total score is the sum of scores on

each question. The maximum possible score is 27 and the minimum possible score is 9. In general, a score of 16 and above suggests significantly low well-being. There are nine questions which comprise three factors: (a) general well-being (positive effect); (b) expectation-achievement congruence and (c) confidence in coping. Separate scores were calculated for each of these factors. The maximum possible score for each factor is 9 while the minimum possible score is 3.

It is likely that compared to men, women have a greater prevalence of psychological distress. This higher psychiatric morbidity was also found in households across different subgroups. Some stress stimuli affected women more than men. More importantly, about 29 per cent of the male members and 35 per cent of the female members did not face any mental distress: they had an aggregate value of zero. But, those women who were stressed, scored higher than men. Thus at very low levels of stress, women were shy in reporting it.

Studies on workers in export oriented industries have looked at physical well-being through measuring aspects like health and nutrition status, but mental well-being has been relatively neglected as an important measure of the quality of life. The nine items were used for analysing mental well-being. For all the indicators there are fewer women than men scoring ‘one’, which indicates greater well-being. The reverse is true for score ‘three’ which indicates poorer well-being. The scores on each item have been summed to arrive at the total score. The combined effect was shown in the cumulative score: more women score 17 to 23 than men. Hence, the gender gap in mental well-being is high. More women than men have greater general well-being scores of ‘three’ and then again a higher score of ‘nine’ and hence men are more evenly distributed than women. But on expectation achievement factors, more men scored ‘three’ than women. The gender gap is high for this indicator. However, the gender gap in the confidence in coping factor is worse than expectation achievement.

Multivariate analyses of mental distress and
subjective well-being inventory

It is interesting to look into the impact of different stress stimuli on the GHQ and the SUBI. For this purpose both logistic and

OLS regressions were run. In the logistic regression the cut-off mark for GHQ was taken as 2, i.e., a value of more than or equal to 2 was assigned number 1, otherwise it was assigned zero. It has, thus, become a binary choice variable. Similarly the cut-off mark for SUBI was taken as 16, i.e., a value of more than or equal to 16 was assigned number 1, otherwise it was assigned zero. It has, thus, become a binary choice variable. Then a proximate list of explanatory variables was used to explain the variation in the respective dependent variable. At the same time OLS regressions were run involving the same variables in order to compare the results. It may be noted that the respective dependent variables are real scores instead of binary choice variables as in the case of logistic regression. The sample characteristics are given in Table 1. The two indices are not different as they were significantly correlated (.5519) but they have a negative relationship i.e. as stress level increases, well-being decreases.

Table 2 and Table 3 show the results of logistic and OLS regressions respectively for GHQ and SUBI indices. Some of the findings which emerge from these tables are as follows:


Sample Characteristics of GHQ and SUBI for Selected Indicators


Note: Figures in column 5 without parentheses are ‘t’ statistics while those parenthesized are probability levels. Figures with one asterisk indicate significant at 10 per cent error probability level, two asterisks indicate significant at 5 per cent error probability level, and three asterisks indicate significant at 1 per cent error probability level.


Logit Regression for Mental Distress (GHQ2) and Subjective Mental Well-being Inventory (SUBI16)


Note: The figures are the values of the coefficients. Figures in the parentheses show significance level. Figures with one asterisk indicate significant at 10 per cent error probability level, two asterisks indicate significant at 5 per cent error probability level, and three asterisks indicate significant at 1 per cent error probability level.


OLS Regression for Mental Distress (GHQ) and Subjective Mental
Well-being Inventory (SUBI)


The figures are the values of the coefficient and the t-statistics are given in parentheses. Figures with one asterisk indicate significant at 10 per cent, two asterisks indicate significant at 5 per cent, and three asterisks indicate significant at one per cent level.

Concluding Remarks

This study is based on a survey of household members (female workers) who were working at the EPZ located at Savar, Dhaka. The survey focuses on the households of the female workers in the EPZ. The analysis compares the responses of 98 male and 166 female members of these households.

Most of the respondents were migrants from different districts. Hence providing housing for the workers has become a good source of income for many families. These houses were built for these workers and hence have adequate toilet and bathing facilities. The workers felt secure in the new community life. Besides the hospital at the EPZ, most of the healthcare facilities were provided by the NGOs.

Most of the workers were at the productive period of their lives and were employed in the manufacturing sector. An increasing level of female labour force participation was observed among them. Most of them took up the current jobs on their own, that is, deciding for themselves. Even though they prefer permanent positions the current jobs are also mostly temporary. Despite this, they preferred these jobs to the previous ones because of the relatively high pay and increased security. Female workers were employed in the low paid jobs relative to their male counterparts, though they reported having various additional allowances which their male counterparts did not have. A large number of them had to take permission from their spouses or parents if they wanted to change their jobs.

Men had more access to formal banking than women did. A large number of women do not own any consumer durables. A small percentage of men and women have control over their assets though a large number of them contribute their entire earnings for the family expenses. Employment opportunities at the EPZ have increased the work burden for most women.

As most women have not managed to get highly-paid jobs because of their low level of education, they are keen to educate their sons and daughters so that they may not face the same disadvantage, and this is not gender specific. In most of the households where the survey was carried out, families ate together and the overall atmosphere seemed to be congenial. However, social norms differ in the perceptions of men and women. Women’s

outward mobility was very limited; they could go to educational institutions and the workplace by themselves but not to other areas. Most of the married respondents were satisfied with the dowry they gave/received. However, most of the unmarried workers did not want their parents to take/give dowry for their marriage. Despite that more women than men wanted to give dowry for the marriage of their daughters. Dowry is still ubiquitous because of an unequal balance of power between men and women in the household.

A large section of men and women felt that it was very important for husbands to have sexual satisfaction. But there were gender differences regarding women’s sexual satisfaction. Although most men thought it to be very important, but only 38 per cent women thought so. Factory work has also made women more visible and their unpaid work in the household is gaining more recognition. Most women suggested that women in their family ought to have more decision-making power than they have now.

Factory employment appears to have different effects on the perception of men and women in the household. It is important to note that both men and women perceived that positive changes do take place when women begin to work outside the home because it gives more experience and knowledge to women. However, few male and female workers reported threats to the joint family as a negative change.

A large number of men and women faced harassment in their workplace. Both men and women commonly reported the incidence of verbal abuse. Even though no violence was reported in their families they had heard of such incidents in the neighbourhood. Around one third of male and female members did not face any mental distress. Of these, women who were stressed scored higher than men did. Women have a significantly higher level of mental distress than men; also women have a lower level of well-being than men. An important policy implication from the regression results is that education above secondary level as well as an increase in household income reduces stress and increases well-being. Migrants have higher levels of stress than the locals and also have lower well-being.


BBS (1995) Report on the Household Expenditure Survey, 1991–92, Dhaka: Bureau of Statistics

BBS (1992) Report on the Labour Force Survey, 1989, Dhaka: Bureau of Statistics

Zohir, S.C. and P. Paul-Majumdar (1996) Garment Workers in Bangladesh: Economic, Social and Helath Condition, Research Monograph No. 18, Bangladesh Institute of Development Studies, Dhaka

Paul-Majumdar, P. (1998) ‘Health Status of the Garment Workers in Bangladesh: Findings from a Survey of Employers,’ Mimeo, BIDS, Dhaka

Status of Women Under Economic
Reforms: The Indian Case*




This essay seeks to pull together the insights from a larger study on the impact of recent economic policy reforms on the situation of women in India.* To the extent possible, the attempt has been to compare the situation before and after the reforms, in absolute terms and in relation to that of men. The methodology adopted has been similar to the common one adopted in the other countries: i.e. analysis of secondary data on various standard indicators of women’s status in India over time and across regions,1

* The paper draws on the two-volume Indian report prepared by the Institute of Social Studies Trust on the gendered impact of economic reforms in the country in recent years. The material contained in the bigger report has been supplemented by information that became available in the intervening period, between the time work was undertaken for the report and the preparation of this paper, as for instance the NSS 55th round data on which section II.1 is based. Among those who contributed in various manners to the preparation of the bigger report at ISST are Sheela Saravanan, Rina Bhattacharya, Jyotsna Shivaramaya, Madhurima Nundy, Vibha Arora, Amita Joshi, Nidhi Arora and Vishal Goyal. Rina Bhattacharya and Sheela Saravanan have been associated with the project from the beginning. A special note of thanks goes out to Sheela Saravanan who has provided valuable assistance for the preparation of this paper.

1This is clearly not the ideal method. However, the alternative(s) may be too intractable, as noted in the first chapter of this book. One can only take cover under the fact that a large array of studies on the impact of economic policy reforms on poverty in India for instance has depended solely on scanning of secondary data for the purpose (See Gupta, Tendulkar).

supplemented by results from a household survey. The surveys in India were done in a sample of households of women workers in Export Processing zones (EPZs) and Export Oriented Units (EOUs) which have been marked by rapid absorption of female labour in the wake of economic reforms.

Much has been written on the adverse impact of economic reforms on women in India although few systematic studies exist to date on the issue.2 What makes the problem doubly intractable for India is the enormous diversity of its people, in terms of ethnicity, religion, social norms and cultural practices: factors that largely shape the nature of gender construction in society. Given this, the issue at hand is to seek answers to questions such as the following: whether changes induced by recent policy reforms in the economic environment of the households have been manifested in changes in differential behaviour and characteristics of labour demand and supply of men and women, and if so, what are these? Are there significant variations in such factors across groups and regions? Have these had any impact on gender differentiated labour use patterns within households? Have these been in the domestic sphere, or in unpaid work? Additionally, have they induced any changes in women’s agency and/or in their decision-making powers and abilities, in women’s access to resources, in their self-esteem? Have they affected women’s status in the eyes of other members of the family and the community? Have they impacted upon women’s ability to resist force, both physical and psychological? Broadly, we might ask, are women of India better or worse off as a result of economic reforms?

Many of these questions cannot be adequately answered from quantitative secondary data. One has to look for supplementary insights from micro studies and surveys. However, unless the studies are carefully designed, such insights can be very context-specific, and not necessarily generalizable. Nonetheless, a range of such studies under varying contexts can be used to piece together a more complete picture. It is expected that between the secondary data on labour market changes and the results emanating from the

2For the larger issue of the impact of economic reforms on women, and the problems of systematic analysis of these, see chapter 1. For India-specific studies on the impact of economic reforms on women, see Ghosh, Shah & Shah etc.

micro studies, one would get better insights into the impact of economic policy changes on the condition of women in India.3

The impact of reforms

The current set of economic reforms are generally traced to the early nineties when an unprecedented crisis in the foreign exchange balance precipitated the onset of liberalization and adjustment measures. However, even before the 1991 economic crisis, from the mid-eighties onwards, winds of change had started blowing and the country had been gearing up towards more openness in the trade regime and towards decontrolling of the domestic production sector, which had been regulated, and ‘protected’ through an intricate system of bureaucratic controls for decades. This meant, among other things, a steady withdrawal of state presence, primarily from the production sector, and to some extent from the service sector, especially from those areas which could be legitimately privatized. While in principle the move for greater privatization and deregulation was supported by arguments of greater efficiency and accountability (Rangarajan, Little & Judge), there were fears that a withdrawal of state presence would be associated with an erosion of labour rights. There have also been apprehensions that increasing prices of essentials under private sector profit-driven motives are going to be disproportionately disadvantageous to the lower income groups who have virtually no cushion against adverse changes in economic conditions, even in the short run (Bhaduri & Nayyar). Since women are seen to be crowded in towards the bottom ends

3The household survey for the study was conducted in the households of women workers working in the Export Processing Zone in NOIDA in Uttar Pradesh and some Export Oriented Units in Sonepat in the state of Haryana in Northern India. A sample of the information collected from the men and women of such households especially in the area of gender dimensions of access to and control over resources within the households is reported in the Appendix to the paper. For the full set of information on the household survey data as well as the questionnaire used to elicit the information, one should refer to the larger report. Volume II of the report also contains a whole range of statistical tables based on secondary data from various sources on the economic, social, demographic and political dimensions of ‘conventional’ indicators of gender bias in the country.

of the labour market, and since they are also believed to bear the brunt of intra-household inequalities in terms of access to resources and work-loads in the aggregate in these households, any changes that lead to adverse consequences for poorer households have been seen to affect women more adversely (Ghosh, Mukhopadhyay, Shah & Shah). To the extent this characterization is valid, economic policy changes, broadly speaking, are likely to result in gender differentiated impact both directly through this impact on the labour market, and indirectly through changes in the economic and social environment within which the reproductive work of women is carried out in the domestic sphere and in the social sector.

This essay, which is based on a larger study done in the context of India, is organized as follows: this introductory section is followed by a section in two parts. The first part looks at changes in the labour market characteristics of women and men in the last couple of decades using secondary data, while the second traces changes in gender dimensions in the social sector from national level statistics, supplemented by information collected from micro studies. Section III presents the results of the household survey carried out to analyse changes, if any, in intra-household gender equations and labour use patterns that may have been brought about by women entering the labour force as new employment opportunities are created under reforms policy. It also analyses gender differentiated stresses and anxieties that may be associated in the context of such change. The final section brings together the conclusions that emerge from the study.

Gender Impact of Economic Policy Reforms: The
‘Conventional’ Indicators

The reforms started in right earnest in the early nineties. A process of trade liberalization, gradual decontrol and privatization of the economy would have had some impact on the labour market. While it is difficult to separate the gendered impact of such processes from the secular trend, a comparison of the pre and post reform scenario would provide some indication of how things may have changed. The next section provides such a comparison using latest available data on the labour market and comparing it with earlier scenarios.

This is followed by a comparison of the pre and post reforms situation in terms of social sector differences.

Gender differences in the labour market

The major data source used for this section is the recently published information from the 55th round (1999-2000) of the National Sample Survey Organization (NSSO) which coincides with the Sixth Quinquennial Employment Unemployment Survey in India.4

The general picture that emerges from available statistical information is that while gender differences in the structure of industrial and occupational distributions and the distribution of labour status categories seem to be lessening over the years, especially in urban areas, they have nonetheless continued, and in some cases even increased, in rural areas. Also, differences in various other characteristics of male and female employment in India continue to persist at the end of the nineties.

To some extent the bridging of the differences where they have occurred may indeed be ascribed to changes in economic policies. Higher demand for female labour in some sectors, for instance, can be linked to a greater thrust towards export-orientation and/or deregulation in the domestic production sector. But the persistence of gender differences in some other areas of the labour market, such as gender differences in earnings and in unemployment rates, suggest that changes in the structure of job opportunities have not really translated into overall reduction in the degree of market segmentation along gender lines. Table 1 below provides a snapshot of the distribution of male and female workers by broad industry groups over the last couple of decades of NSS rounds

4The 2001 Population Census data on some dimensions of employment have been recently released. However these are not being reported here for there are well-known problems in comparing Census information on female employment related data with those from the NSS. For the sake of brevity, we stick here to the NSS data as these are believed to be more reliable than Census information in so far as female employment related issues are concerned. See Mukhopadhyay (1999) for details.


Changes in the Distribution of Male and Female Workers between Broad Industry Groups between 1983 and 1999–2000


(↓): Falling; (↑): Rising; (↔): Less than 2 per cent change

It can be seen from Table 1 that the direction of change in the industrial distribution of the workforce has been almost identical for men and women over the last decade and a half, although the rates of change have varied considerably. The decline in the absorption of the rural male workforce in the primary sector has been very slow and the corresponding rate for women workers has been even slower, although in urban areas this decline has been much faster in general, and that for women has been even faster than that of men. Similarly the rise in the share of the service sector workforce in both rural and urban areas has been sharp for all categories, especially so for urban women workers. Thus male-female differences in the industrial distribution are closing down more quickly in urban areas, but in rural areas, the gap if anything has been increasing although the directions of change may have been similar.

In terms of labour status categories, the scenario is as follows:


Changes in the Distribution of Labour Status Categories Over Time


(↓): Falling; (↑): Rising; (↔): Less than 2 per cent change

Table 2 shows that in terms of labour status categories, the pattern of changes over the last fifteen years has been nearly similar for men and women in rural areas. There has been a reduction in self-employment and an increase in casualization for both men and women workers, while the incidence of “regular” employment has marginally decreased for men and marginally gone up for women in urban India. Labour status distribution has not changed very much for the urban males, but that for urban females seems to have undergone substantial changes. In particular, there is a sharp reduction in the incidence of casual labour — from 28.4 per cent in 1983 to 21.4 per cent in 1999–2000, and a sharp increase in the incidence of regular employment from 25.8 per cent in 1983 to 33.3 per cent in 1999–2000, thereby bringing the female distribution far closer to the distribution for urban males. However, there is a need to temper this conclusion with one caveat. For ‘usual status’ figures for the female workforce, the importance of marginal employment continues to be fairly high. As a result the difference between ‘principal status’ figures and ‘usual status’ figures is quite marked, much more so than for men, although the difference shows a tendency to go down over the years. While the extent and direction of change are broadly unaltered in both urban and rural areas, the major difference between ‘principal status’ and ‘usual status’ female workers appears to be rooted in the category of ‘self-employed’ women, some of whom clearly get categorized as casual labour and “regular” employees when their marginal activities are taken into account. In the case of industrial distribution of the workforce given in Table 1, the industry group that corresponds to the category of ‘self-employment’ is the ‘primary sector’. Table 3 provides a snapshot view of the various unemployment rates reported for the two rounds of NSS.

The relative picture on unemployment suggested by Table 3 is that in rural areas, usual status unemployment rates which are very low to start with, have gone up for both men and women, while CWS and CDS rates have changed in such a way that the corresponding male and female rates have come closer. In urban areas however, the rates for women are higher than those for men in every category.


Unemployment Rates


(↓): Falling; (↑): Rising; (↔): Less than 2 per cent change

These unemployment rates are for all workers. Very few among the women who are in the workforce are educated. Only 3.7 per cent of the rural females in employment and 28.9 per cent of urban females in employment are educated. The corresponding figures for men are 16.0 per cent and 42.5 per cent respectively. However, curiously enough, if one compares the percentage of educated women who are in the labour market but unemployed, with the corresponding percentage of men, it is clear that educated women have a far lower probability of getting absorbed in the job market compared to their male counterparts. As many as 62.7 per cent of unemployed women in rural areas are educated as compared to 55.2 per cent of men. The corresponding figures in urban areas are 74.0 per cent for women and 56.9 per cent for men. (Table 15, p. 29, NSS Report No. 455 (55/10/1), Dec. 2000). Another way of looking at this anomaly is to look at the unemployment rates of educated male and female workers. The following table provides this information.


Unemployment Rates (usual status, 1999–2000) by Educational Levels


Note: Figures in brackets are corresponding figures for 1993–94.

It is clear from Table 4 above that in every category, educated women are much worse off than educated males in the workforce although, barring rural graduate women, unemployment rates are generally coming down for both male and female educated workers. To some extent the recent expansion in IT related activities must have had some bearing on the increased demand for educated workers in the economy, and although unemployment figures continue to be higher for women as compared to men in all categories, the rates of decline in these are much higher for women, except in the rural graduate category, where they have increased. From this we may deduce that the expansion in labour demand in IT-enabled sectors in the economy in recent years may have benefited the educated female workforce relatively more than the male.

One area where things have not changed much is the difference in male-female wage earnings. Although the nation-wide survey data do not provide enough information for calculating equivalent wages, persistently low wages of women as compared to those of men, to the tune of anywhere between 50 per cent to 80 per cent, suggest systemic job as well as wage discrimination. There is little evidence to suggest that on an average things have changed substantially for the better in this area after the reforms.

Wage earnings of female casual workers in 1999–2000 were only 64.70 per cent of corresponding male earnings in rural India and even lower at 60.57 per cent in urban India. The former percentage has in fact declined marginally from a figure of 66.39 per cent in 1993–94 although in urban areas the figure in 1993–94 was lower at 56.81 per cent. On an average, women earned only 59.66 per cent of what men earned in regular salaried/wage employment in rural areas and 79.76 per cent in urban area in 1993–94.

What has indeed emerged in the wake of economic reforms is the phenomenon of a small group of highly qualified and well paid young women in the area of finance and management. This has been instrumental in pulling up the wage equation in that small segment. The impact of this phenomenon can be seen if one disaggregates male-female wage differences by education-levels of workers and across industry groups. While in sectors like agriculture or manufacturing, the wage differences are not systematically linked to increases in the educational levels of the workers, for the service sector, the links are more pronounced and systematic, with female

earnings as a percentage of male earning showing a positive correlation with rising educational levels.

By and large one could say that except for a small but rising sector of highly paid professionals in the service sector, and the public sector where the letter of the law against wage discrimination on grounds of sex is observed, on an average, women continue to have substantially lower earnings then men do. The extent of the difference may vary across sectors, age groups and skill levels of workers (See Vol. I), but the general picture is one of continuing wage, and job discrimination. The demand for female labour may have gone up in some sectors such as in IT-enabled activities in the wake of economic reforms, and this may have had some impact on their recorded unemployment rates, relatively speaking. However, there is no indication that apart from a small segment of relatively educated women in urban areas, the rising demand has translated itself into better wages. Rather, what is becoming available to the average Indian woman worker is more of low-paying low-productivity jobs. This may be true also in factories and in IT-enabled sectors, where some schooling is being used more as a screening device than a factor determining the wage rates.

More information is needed in this area to enable a more informed analysis. But, by and large, for the vast majority of Indian women in the labour force, gender discrimination in the Indian labour market is alive and well, even in the post-reform era.

Gender bias in conventional non-economic indicators

Economic policy changes have gendered implications for the labour market in a direct and straightforward manner. They will have gender-differentiated implications for other market and non-market spheres as well. Changes in the external economic environment may induce changes in labour use patterns of men and women within households. Changing structures of employment opportunities may induce changes in the differential demand for education for boys and girls. Changes in the level and structure of prices may induce gender differentiated health seeking behaviours. Many of these linkages are difficult to delineate clearly. However, there is indeed a case for looking at changes in some non-economic indicators of gender bias before and after the reforms, because some of this change may have been triggered by the reforms.


Female literacy has always lagged behind male literacy in India. But the difference between the two has been more pronounced in rural areas as compared to urban. A secular trend in bridging the gap is observed only in the latter. Thus any attempt to reduce the male-female literacy gap must specially target rural female literacy.


Gender Differences in Educational Attainments: Literacy Rates (Population 7 & above)


Source: Census of India, Various Years.

Any number of researchers have pointed out the external benefits of female literacy, in terms of reduced infant mortality, higher work participation of women and bolstering women’s agency in general in a society which continues to be strongly patriarchal. In the various surveys carried out by ISST in recent years, including the one for this particular project, it has been repeatedly made apparent that parental opposition or apathy to the education of the girl child is fast receding, even in the traditionally male-dominated states of northern India. Given the right infrastructure — schools located in the neighbourhood, preferably with female teachers — parents would allow their daughters to study “as long as they would like to”.

A number of factors may have contributed to this change of heart, and reforms may not be the most important of them. Many state governments have been providing added incentives towards girls’ education. The age of marriage is going up. And girls themselves are keen to get educated, often much more so than their brothers. However, higher education need not necessarily get translated into easing the process of labour market entry. In some communities, sending their unmarried girls for paid work is as yet not acceptable, although there may not be such a strong taboo on

sending the wife for factory work. The extra money brought in by her helps. In some other communities, however, especially in urban and semi-urban areas, young girls are getting into paid work of all varieties, and some degree of education is needed for landing a job.

All in all, the mindset on girls’ education appears to be undergoing a transformation, even in the traditional Hindi-speaking north Indian states, but the extent to which it is translated into true empowerment is a different issue altogether. As one found in the course of our interactions with the surveyed families, the same families who are willing to see their daughters go through college education, may react violently if the girls decide to choose their marriage partners, or to challenge the set norms of feminine behaviour. The links between female literacy and empowerment are not as simple and uncomplicated as it is sometimes believed. Even in a state like Kerala, which has boasted of very high female literacy for decades now, and which has been held up as the ‘model’ of women’s emancipation, it is amazing how little of this literacy has translated into the strengthening of women’s agency in different spheres of life (see chapter — in the book)


The link between standard health indicators and economic reforms is once again indirect at best. The set of reform-related policies that are believed to have some direct effect on health are basically those that result in greater privatization of health services, the institution of user fees and reduction in the role of the state in the provision of health care. It is felt that these measures are going to hurt the poorer sections of society in a disproportionate manner. A contraction in the provision of public sector health services through pricing, quantity restrictions or both is likely to result in a reduction in the consumption of such services and/or in an increase in the need for health care provision within the household. This will be associated with a corresponding increase in the workloads and responsibilities of women in households as they are the ones who look after the sick and the aged. In either case, there is a welfare loss for the population in general and for women in particular.

A secular trend suggesting slow improvement in the health status of the population has existed long before the reforms started, and

it is difficult to estimate how much of the change in these (if any), can be legitimately ascribed to changes in economic policy. It is even more difficult to differentiate the impact by gender. However, based on the recent changes in some of the key indicators, some hypotheses can be framed about the matter.

Expectation of life at birth has slowly gone up over the years, and currently women’s life expectation stands at a level higher than that of men, something that was not so two decades back and something that is a universal phenomenon almost everywhere, except in some countries in South Asia which are ridden with gender discrimination. This clearly is a development that is notable.

India had been a signatory to the 1978 Alma Ata Declaration, proclaiming ‘health for all’ by the year 2000. Yet achievements made in the area of health have been anything but spectacular. The public health scenario has been dominated by several centrally sponsored vertical programmes dealing with the control of diseases like malaria, leprosy, tuberculosis, and more recently HIV/AIDS. However provision of primary health care has not been the strong point of the government. Inadequate sanitation and hygiene coupled with lack of adequate drinking water, pervasive poverty and illiteracy of the population have continued to thwart the realization of the Alma Ata vision. Under these circumstances, the abrogation of state responsibility for the provision of basic health facilities can only worsen the situation of the poorer sections of the population, and of women in particular in these households.

Also, over the years, a lot of resources that could have gone into the provision of health have actually gone into measures that were felt to be needed for population control through the government’s Family Planning Programme which was put in place as early as the fifties. The programme has been strongly criticized for being gender-insensitive, and a targets-and-incentive-driven exercise. In recent years there have been attempts to impart a more inclusive and women-centred focus to this under the garb of the Reproductive and Child Health Programme. A range of government programmes have also been started to counter the bias in society against women and girl children; these are expected to have positive health outcomes for them. None of these is directly linked with economic policy reforms.

However, even the combined effect of all these measures does not seem to have been wholly successful in countering gender bias.

Infant Mortality Rates have indeed gone down over the years but gender differences in such rates continue to persist. The same is true of the Under-Five Mortality Rates. Although there is a secular decline in all these rates over time, girl children continue to suffer with respect to their male counterparts, which suggests a continuing bias in provision in nutrition and health care.


Infant and Under-5 Child Mortality Rates in India by Gender


Source: Sample Registration System, various years.

If gender bias in treatment seeking behaviour is indeed the primary reason behind these differences between the mortality rates of young boys and girls, then such differences would show up also in the morbidity patterns, if the data were available and if they were free from reporting bias. Unfortunately such information is not readily available.

Sex Ratio

A demographic indicator that is related to gender-differentiated mortality rates is the sex ratio, i.e. the number of females per thousand males in the population. Given the sex ratio at birth, which under neutral conditions should be the same everywhere, this indicator encapsulates the aggregate effect of differential mortality in all the age cohorts. In India, after a drop in the aggregate sex ratio in 1991, the recent Census figures show a marginal increase in the ratio to a figure of 933. What is alarming however is the steady decline in the sex ratio of the population below 6 years of age, from 976 in 1961 to a low of 927 in 2001. This reinforces the suspicion that there is persistent bias against the girl child in large sections of the population, which is strong enough to be reflected in systematic inequalities in the probability of her chances of survival in life.


Sex Ratio in India


Source: Census of India, various years.

The evidence presented in this and the earlier section suggests that gender bias has persisted in the population in a manner and to an extent that has shown up in the differential rates of survival of young girls. Persistently high maternal mortality rates in the country in spite of a long-standing family welfare programme suggests that adequate efforts and resources are not being allocated for reducing avoidable deaths on account of pregnancy and childbirth. The fear is that reduction in state resources in the social sectors will, if anything, exacerbate the existing gender bias in society.

Changes Outside the Labour Market: Survey Results

The gendered effect of changes in the external economic environment are not confined only to the labour market. In fact, the labour market behaviour of women is strongly determined by the fact that reproductive responsibilities continue to be primarily a female concern within households. One of the major issues to contend with is to find out what happens to the labour use patterns within the households when women join the labour force. Is there any sharing of household work? Of what kind, to what extent, and under what conditions? Apart from the question of sheer workload, there are other, less tangible, but equally important questions: on women’s decision-making and their autonomy, freedom and choice: issues which determine the status of an individual in society. Some of these questions were explored in the context of a household survey carried out in two locations in the northern states of Haryana and Uttar Pradesh.


In all 114 households comprising 594 individuals were surveyed. All the households in both locations had at least one female member working in an Export-Processing Zone or in an Export-Oriented Unit.


Some Basic Features of the Sample


There is a certain difference between the two locations where the survey was carried out in India. The study site in Sonepat is situated within a prosperous agricultural region. The infusion of industry in the region has been of relatively recent origin, and has also been quite gradual as compared to Noida. Noida at this point of time is

a major industrial belt, encompassing one of the seven EPZs in the country located within it and has several Export Oriented Units (EOUs) around the Noida Export Processing Zone (NEPZ). A major upshot of this difference is that while Noida has attracted a lot of migrant worker families as well as single migrant women (specially from the south), the female factory workforce in Sonepat largely consists of women from locally settled agricultural families or from families who have lost their land either due to fragmentation of land holdings or deaths in the family, or due to property disputes. There are also some migrant families, especially from Bihar and Uttar Pradesh who had originally migrated to this place to work as farm labour in the green revolution era, and had subsequently settled down in the area. These are primarily landless families.

These differences in the background may explain why there is a fair amount of difference in attitudes towards women working in factories among the locals in the two places. There is little doubt that both localities have benefited from industrialization. The local Noida population has prospered by constructing and renting out small rooms in matchbox-like structures that have cropped up all over the surrounding villages, to the migrant worker families (Kirayadars) who have flocked to the area from all over the country, or by selling land to the government in lieu of hefty compensation. They would rather do that than send their own women to work at the factories. Given a choice, they would much rather procure jobs for their sons in the factories. They would not consider sending their own women for factory work as they feel it is below their dignity to do so, although they are willing to educate them “as long as the girls want to”.

In Sonepat, in contrast, women from agricultural families have been gradually entering the workforce in the local factories. There is also a sense, albeit slow, of changing mindsets in this regard within the community. There are cases where the women are not willing to continue working but are being persuaded by their husbands not to drop out ‘at least for a few more years’, because the additional money is useful. In some cases sudden financial problems in the family may have initiated the move to send women to this new kind of work. In other cases, dire economic necessity — death of the main earner, property disputes, indebtedness etc. — have left them with no option. The point to note is that over

time, factory work for women has become more and more acceptable socially. In fact most non-working women who were interviewed in Sonepat said that they would have been happier if they had the opportunity to work in factories.

In Noida the situation was somewhat different. The families were mostly nuclear families who have migrated to the area basically because of the job opportunities. The wives and daughters here are all factory workers. Social constraints and sanctions imposed on women’s mobility or paid work are not directly operative here. The exigencies of the situation, such as the lack of fall-back support in nuclear family set-ups, make it imperative that compromises are made in various matters. Men of the households surveyed in Noida share in household work to a much larger extent than in Sonepat. But this is more because they are away from their traditional family environments. It is interesting to note that while all the women surveyed in this study were working in the local factories, many of them have not divulged this information to their families back home. Most of these families are from UP and Bihar.

It is clear that the family situation one finds in Noida is unlikely to reflect the true situation of women, be it in terms of sharing of housework or in terms of observed indicators of women’s autonomy. One could argue that in such matters, the Sonepat situation is likely to be closer to reflecting the true scenario. The fact that many families in the Noida sample have not divulged to relatives back home that that their women have joined the labour force, shows the strength of custom and convention in determining the contours of womanly behaviour. The following sections provide some results from the survey in terms of women’s work both in and out of the labour market, and other, less tangible indicators of women’s situation in the sampled households.

Results of the survey

Conditions of work

By and large, the conditions of work for the women in the sample were found to be reasonable by local standards. In general, working conditions were better for women than for male workers in the factories. This is contrary to the findings of some studies, which suggest excessively exploitative working conditions for women in

EPZs. The majority of women, as compared to men in both locations, were found to work on fixed timings and not on shifts, although more women as compared to men were found to be working overtime, sometimes not out of their own volition. Nevertheless, the survey results show that by and large women enjoyed better benefits than men and greater job security.

However one must emphasize that this could be a result of the fact that women were found to be more docile and non-complaining than their male counterparts. This could also explain why factory owners were found to have a strong preference for migrant as opposed to local labour, and for women as compared to men. One other feature of work conditions that emerged was that women workers were generally reluctant to talk about harassment of any kind, especially harassment of a sexual nature. As expected therefore, these dimensions of work conditions did not come out very well through answers to the structured questionnaires, although intensive interviews and case studies suggest that they do exist in good measure both inside and outside the workplace.

Labour use patterns within households

It is interesting to compare the labour use patterns of men and women within the households in the two study locations. As noted above, the Noida sample consists mostly of migrant nuclear households in all of which women are working full time in the local factories. Some of their men are unemployed. Some others have periodic, casual work. In Sonepat, there are many families, where there are one or two non-working women in the households. Thus while in Noida 17.5 per cent of the men do ‘no cooking’ and 10.5 per cent of them do ‘no cleaning’ work whatsoever, in Sonepat the corresponding percentages stand much higher – at 61.1 per cent and 47.2 per cent respectively. This is clearly explained by the fact that in Sonepat there are other women in the households to take over work which is considered to be women’s work. This general perception is corroborated by the figures at the other end of the spectrum which reveal that even in Noida where all the women are involved in full time ‘productive’ work, 90 per cent of ‘all’ or ‘most’ of cooking and cleaning work in the household is done by the women.

These figures show the resilience of traditional gender roles even

under radically changed family situations such as those encountered by the migrant nuclear families in Noida. Irrespective of the fact that the women are spending a large chunk of their day in earning money for the family — more so in fact than the Noida men — cooking and cleaning continue to be their primary responsibility. Seventy per cent of the Noida men do little or no cooking whatsoever and 67 per cent do no cleaning. The corresponding figures for Sonepat men are 75 per cent and 70 per cent respectively.

There are other kinds of household duties in which men share to some extent, such as ‘taking care of children and the old’. But even here, women bear the lion’s share, with 71 per cent of ‘all’ or ‘most’ of child care responsibilities and 52 per cent of the responsibility of ‘taking care of the aged and the sick’ being borne by them in the two locations.

Interesting differences emerged between the perceptions of men and women when they were asked whether the pattern of household labour put in by men and women was significantly different now as compared to the time when women were not working. Fewer men (62.8 per cent) than women (81.1 per cent) felt that it was indeed so, reflecting, perhaps, the guilt and unease felt by many of the women. Many talked about the neglect of children as a result of women taking up paid jobs. Equally interesting is the perceptional difference between men and women when confronted by the question about whether men will take up some activities normally done by women if and when the woman falls sick. While 92 per cent of the men felt they would, the corresponding figure for women was only 73 per cent. However, when asked specifically if there are some activities ‘men will never do’ under any circumstances, it was clear that some activities normally done by women, more than some others, are considered beneath their dignity by men. Eighty per cent of the men would never consider washing women’s clothes, and 57 per cent would never clean toilets. At the same time almost all men felt that they would take care of the children, the aged and the sick, if need be, and do household shopping. But even here as elsewhere, women’s perception of what men would be prepared to do differs from what men say they would, suggesting the existence of a gender divide even in the understanding of the contours of gender roles within households.

Positive and negative changes brought about through women’s entry into the labour force

Most men and women in both locations were of the opinion that the phenomenon of women working outside the household has had an overall positive impact. An overwhelmingly large number of women felt that it had increased their self-esteem (85.3 per cent), their standing in the family (81.1 per cent) and their decision-making power (62.2 per cent). The corresponding figures for men are 76.7 per cent, 69.8 per cent and 43.4 per cent respectively. Only 6.2 per cent of men thought that there was no positive affect (‘none’) and 3.9 per cent were non-committal (‘can’t say’). Women were even more certain about the overall positive impact of their labour market insertion, with corresponding figures for them being a startling 0.0 per cent and only 1.4 per cent respectively. Also, far more women in both locations cited more than one way in which the change has been beneficial. While for men, the percentage of double or triple citations was 51.9 per cent, the corresponding figure for women was three times higher at 167.1 per cent. Clearly women saw many more benefits coming out of their labour market entry than men did.

By and large, economic security was cited as a major benefit by both men and women. But women also cited factors such as more experience (91.6 per cent) more enlarged social networks for them (48.3 per cent) and stronger personality (32.2 per cent) as other fall-out benefits. Fewer men felt so, the corresponding figures for men being only 42.6 per cent, 20.2 per cent and a meagre 3.9 per cent respectively, something that signals the emergence of a gender divide in perceptions.

At the same time, women were also more conscious of the negative impact of their entry into the labour market. It is significant that while 59.7 per cent of the men thought there was no negative impact of women working outside the home, women themselves were less positive, with only 32.9 per cent of them saying that there had indeed been no negative impact. Also, that many more women in Sonepat as compared to Noida, recorded negative factors, such as ‘conflict over domestic chores’ (44.7 per cent of Sonepat women compared to only 1.7 per cent of Noida women) and ‘neglect of children’ (54.1 per cent of Sonepat women as opposed to only 24.6 per cent of the Noida women). It appears

that such conflicts have been perceived by Sonepat women as being between women who work and those who stay in, with the latter group having to shoulder the burden of the additional work. The men in Sonepat are largely unaware of such conflict — with only 15.3 per cent of them reporting conflict over domestic chores and 20.8 per cent reporting neglect of children. In Noida, since hardly any family has non-working adult females in it, it is the working women who bear the double burden of work. What is interesting is that under no situation had the men pitched in to share in household work in any substantive way. In Sonepat, it is very likely the additional workload that triggered off disharmony between working and non-working women, while in Noida, women have simply doubled their work burden. One way or the other, housework continued to be primarily the responsibility of women.

Decision-making and control

Working outside the home in a paid job did not translate into appreciably greater autonomy within the household for most women. Far fewer of the working females, for instance, could take a decision on changing jobs as compared to men in both locations. Sixty six per cent of the women in the total sample said that they need to consult somebody — presumably a male in the family — and take permission before they would change jobs, while only 21 per cent of the males said so. While most women seemed to think that they enjoy joint ownership of household assets, very few reported the ability to take the decision on buying or selling assets of most kinds on their own — without the knowledge and concurrence of a male in the family. The only things that some women seem to be able to buy on their own are clothes, provided they are not very expensive, and costume jewellery; 27.6 per cent of the women in Noida and 35.3 per cent of the women in Sonepat said that they are allowed to buy ‘nothing at all’. Considering that all the women in our sample in Noida are ‘working’ women, this result is quite startling.

A similar picture emerges in the matter of control of earned income, especially in Sonepat. While most women in Sonepat give away all the income they earn to their spouses and parents-in-law, most men in Sonepat retain a large part of their incomes to themselves or hand it over to the parents. Hardly anything goes to the wife. In Noida, the picture is somewhat different. It appears

very likely that here men and women pool their salaries together. But even here, while the majority of women — 39 out of 58 — hand over their total salary to their husbands, only 17 out of 57 of the men prefer to give their total salary to their wives. So even in Noida, control over income is mostly in the hands of men.

Most women, both in Sonepat and in Noida, reported that their decision-making powers have increased as a result of working outside the home. However closer scrutiny reveals that this is not borne out by the objective state of affairs. Women’s decision-making largely continues to be concentrated in making small purchases for daily household needs. In major economic matters, such as buying and selling of assets, they have hardly any say at all. Yet the general view among women is that in this respect, things are getting better for them, although the men do not think so.5

Family discord and violence

One fallout of being in a nuclear migrant family is perhaps that there is less chance of family discord. As compared to Noida, women in Sonepat report a higher incidence of family discord, with problems with the spouse’s family (45.9 per cent), discord triggered by disobedience to husband (31.8 per cent) and talking back (36.5 per cent) being some of the major reasons cited. However in both locations, the phenomenon that tops the list of factors leading to family discord is alcoholism, with 64.7 per cent of women is Sonepat and 31 per cent in Noida citing it as the major cause.

It is interesting to compare the differences between the answers given by men and women in the two study locations to some of the questions asked in this area. When asked whether they have experienced tension within the family, 59.6 per cent of men and 51.7 per cent of women in Noida answered in the affirmative. However the same question in Sonepat evoked widely different responses from men and women, with 60 per cent of Sonepat men saying there was no tension. To what extent are Sonepat men insulated from such household tensions and to what extent this is a matter of false reporting by men is a moot question. Since

5For a sample of the kind of response one received from men and women of the households in the two study sites on these and related issues, see the Appendix to this paper.

Sonepat women have also reported that there are various different factors, which triggered tensions in the family, it is unlikely that men are unaware of such tensions. It is more likely that men do not give much importance to women’s problems and are ready to ignore them altogether. The fact that Sonepat men exhibit the lowest level of stress suggests that the factors that cause so much stress for Sonepat women are perhaps not felt as stress-inducing by Sonepat men at all. In this regard Noida men appear to be closer to the Noida women, sharing not merely the incomes but also the stress levels of their wives.

Gender related stress, anxiety and violence

Stress and anxiety are part of human existence. External circumstances can trigger them off, and the individual’s psychological make-up determines how and to what extent these affect one’s mental health and well-being. While in principle one can separate external proximate reasons and the internal psychological make-up of the individual reacting on such circumstances to produce a certain level of stress, in practice the two sets of factors are so intricately interlinked, given the current state of psychological research, that it is not possible to separate their impact. However if one can measure levels of stress, and identify other relevant parameters to locate an individual within a certain framework of analysis, one could use standard methods of multi-variate analysis to get some understanding of the factors causing stress and anxiety in a community of individuals. If such a community is subjected to unusually stressful circumstances, such as war, riots or endemic economic insecurity and poverty, one would expect a general heightening of stress levels in all individuals, although there will be differences between individuals depending on other parameters of their objective situation as well as their psychological make-up and coping abilities.

This study has attempted different methods of analysing the question of whether the status of being the ‘second sex’ creates additional levels of stress for the women in the sampled households. Such stress could be caused by a number of factors that may be shared by all women, a key one being that working outside the house, if anything, increases the total workload of all women, working and non-working. The very fact of being born a woman in a highly patriarchal and hierarchical society is another. As regards

the second issue, however, one needs to be careful about coming to a hasty conclusion. If the woman concerned has fully internalized the dominant patriarchal norms and values, then playing second fiddle may not create additional stress or anxiety. Such women may even accept occasional beatings and verbal abuse from the husband and the in-laws as the natural order of things. However, for those women who are going out to work for pay for the first time, the phenomenon itself may have implications which could very well be more than purely economic. It exposes them to a world they have hitherto been insulated from. The exposure has the potential of being instrumental in raising questions in their minds that were probably not there earlier. It is significant that a very large percentage of the sampled women have said that apart from providing their families with greater economic security, the out-of-home work experience has increased their self-esteem. Perhaps what one sees here is the emergence of a new confidence and a changing mindset about women’s agency among women themselves. It is interesting to note that far fewer of the men, especially in Sonepat, think along the same lines. A similar picture emerged when we recorded responses to identical questions by husbands and wives, with the presumption that differences in the answers given by the two can be taken as a measure of perceptional differences. As expected, couple responses from Sonepat show a greater degree of discord between husbands and wives. Such differences in perception can indeed trigger additional stress within households, the genesis of which can be traced to gender-related factors.

Two methods were used to explore stress and violence in the household. On the one hand a number of direct questions were asked on a range of issues. On the other, multivariate analysis was carried out to explain variations in measured levels of stress and mental well being for men and women in the sample. The difference between Noida and Sonepat samples is fairly striking here here as well. While 60 per cent of the men and 52 per cent of the women in Noida said that there are indeed tensions in their own families, barely 11 per cent of the Sonepat men and a staggering 60 per cent of the Sonepat women report such tensions. Many more women than men in both Noida and Sonepat report physical and verbal abuse in the family. Understandably, property and politics related tensions are much higher in Sonepat than in Noida, but sexual harassment outside the home is reported from both locations.

Results of a multivariate analysis

A set of twelve questions selected from the General Health Questionnaire (GHQ) and nine questions from the Subjective Well-being Inventory (SUBI) were canvassed to all the 271 men and women in the two locations in order to assess the level of mental stress and well-being of these individuals. OLS and Logit regressions were run on the scores obtained from the interviews to explore the explanatory factors causing variations in individual scores. The results are reported in Table 9.


Results of Logit and OLS Regressions for the Combined Sample from Noida and Sonepat


It is clear from Table 9 that by all counts, the stress levels of women are much higher than those of men. But this, one is told, is a near universal phenomenon, reported from different countries and under different circumstances. The moot question is: to what extent can this result be ascribed to the specific circumstances of the women and men interviewed for this study?

Stress and mental ill being appear to be inversely related with household size. This is perhaps because a larger household offers the cushion to absorb additional workloads caused by women working outside the home. This explanation seems to be a reasonable one since the linkage becomes less significant if one introduces any variable reflecting the presence of small children in the household. Also, the introduction of a slope dummy suggests that the stress levels associated with these factors are much more significant for the women of the households than for men, especially in Sonepat.

Variable names

AgeAge in years
Hh size Household size
Childpro Ratio of children below 12 to hhsize
Edu Education dummy 0 for < Secondary
1 for > Secondary
Sex Male = 0, Female = 1
MS Marital status dummy.Unmarried = 0,
Ever married = 1
WS Work status dummy. Working = 0,
Not working = 1
Inc Household income
Inc sq Square of household income
GHQ2 Binary variable, O for GHQ < 2
1 for GHQ >= 2
SUBI16 Binary variable, 0 for SUBI < 16
1 for SUBI >= 16
GHQ Posted scores of GHQ
SUBI Posted scores of SUBI

One interesting feature of the equations here is that given other things ’age’ and ’marriage’ increase stress levels, as does the level of education. Once again, the introduction of a slope dummy shows that the first two factors are much more stress-inducing for women than they are for men. Interestingly, working persons have lower stress than those who do not work. The reasons for this could be different for men and women, but this needs to be explored further.

Also, overall levels of stress and mental ill-being are lower in Sonepat as compared to Noida, perhaps because families in Sonepat have fallback family support while Noida families are primarily migrants uprooted from their moorings. Sonepat men are the least stressed of all categories. By and large the factors that seem to create the most stress in women are somewhat different from those that cause stress in men, highlighting the importance of gender-related factors in inducing stress.

Concluding Remarks

A decade of reforms may not have touched many segments of the Indian population in any substantive manner as yet, especially those involved in the farm sector in the rural hinterlands of the country, or the casual workforce which depends primarily on unskilled wage labour for survival. However, there are segments of the economy which have indeed been affected by reforms. Wherever reforms have brought in changes in the external economic environment of households, be it terms of changes in opportunities or in constraints, it is likely that their impact will be differentiated along gender lines.

Our results suggest that a number of changes have indeed set in. The new jobs for women have been looked upon favourably by both men and by women who are working. Both like the increased economic security they bring in. By and large in our sample, working conditions are also not worse than average, or worse than what they are for the men in comparable conditions. Women in particular are happy also for the non-economic fall-outs of paid work: higher self esteem, larger social network, and a passage to higher status in the family and the community as perceived by the women themselves.

Unfortunately, there is little evidence of such views being shared

equally by the men in the family. For them, the benefits are primarily economic. The possibility that having a paid job might confer higher status or greater decision-making powers on women is something that simply does not figure in the perception of most men. This phenomenon of non-congruence of perceptions is much more evident in the settled agricultural location of Sonepat where most families are joint families, as compared to Noida where they are nuclear families which have migrated from other parts of the country, away from their traditional moorings. Many among them have kept the fact of their women working in factories a closely guarded secret from their families back home. What is striking is that there is not much substantive sharing of housework in either location across the sexes, showing the resilience of traditional gender roles within the domestic sphere even when women put in a full day’s paid work outside the home. Noida women bear the major brunt of domestic work by putting in additional hours, while in Sonepat, non-working women of the household pitch in. Better economic opportunities do not seem to have absolved women from their reproductive responsibilities. Earning in the marketplace has not automatically translated into more flexible gender roles or greater overall empowerment.

However there are clear signs that at least for the women themselves, especially those from Sonepat, the benefits go beyond purely economic ones. They report a higher sense of self worth, but at the same time they also report much family discord. The Sonepat men appear to be blissfully unaware of the stress building up in their own backyards. The perceptional differences that are emerging between the men and the women in Sonepat seem to indicate that while gender differences in ‘conventional’ indicators may be closing up, the non-conventional indicators of gender bias may be poised for a worsening phase.


Selected results on actual and perceived differences between men and women on decision-making, access over resources and control over life situations of women in the surveyed households in the two study sites. All answers are percentages of total respondents.

Control over assets: disposal

Which of these assets can you dispose off on your own?


Question # III.2.c.4. In Sonepat vehicles and jewellery are combined

Control over assets: acquisition

Which of these assets can you buy without takingpermission from anybody?


Question # III.2.c.4.5 This question was asked only to women

Control of labour

Have you ever been harassed in the workplace by anybody?


Question # III.2.c.2.

Type of harassment


Question # c.2.12.

Changes with employment

Do you think that the opportunity of working at the factory and earning has brought about changes in your life situation? Are these changes positive?


Question # c.7.2

Or are these negative?


Question # III.2.c.7

Decision-making and control

In which of these areas do you participate the most in decision-making processes within your household?


Question #III.2.c.13.

In which of these areas is your decision final?


Question #19.1.2.


Gupta, S.P. (1995) ‘Economic Reform and its Impact on the Poor,’ Economic and Political Weekly, Vol. 33 (22), June 3

Ghosh, J. (1994) ‘Gender Concerns in Macro Economic Policy,’ Economic and Political Weekly, Vol. 29 (18)

__. (1996) ‘Gender and Macro Economic Policy in India since 1991,’ Women’s Link Vol. 3 (4)

Sundaram, K. and Tendulkar, S.D. (2001) ‘NAS-NSS Estimates of Private Consumption for Poverty Estimation: A Disaggregated Comparison for 1993–94,’ Economic and Political Weekly, Vol. 36 No. 2

Tendulkar, Suresh D. (1998) ‘Indian Policy Reforms and Poverty: An Assessment,’ in Ahluwalia I.J. and I.M.D Little (eds.) India’s Economic Reforms and Development: Essays for Manmohan Singh . Delhi: Oxford University Press

Rangarajan, C. (2000) Structural Reforms in Industry, Banking and Finance: A case study of India, New Delhi: Capital Publishing Company

Bhaduri, Amit and Nayyar, Deepak (1996) Intelligent Person’s Guide to Liberalization, New Delhi: Penguin Books

Institute of Social Studies Trust (2001) India Country Report Vol. 1 & 2 . Submitted to IDRC under the Gender Planning Network Project

National Sample Survey Organization (2000) Employment and Unemployment in India 1999-2000, Key Results, NSS 55th Round

Shah, Nandita and Nandita Gandhi (1992) ‘Shadow Workers: Women in Home Based Production,’ (Mimeo), Bombay

Registrar General & Census Commissioner (2001) Provisional Population Totals, Paper 1 of 2001, India

Census of India 1991, Series I, India Paper 2 of 1992, Final Population Total, Brief Analysis of Primary Census Abstract, Registrar General and Census Commissioner, India

‘District Level Estimate of Fertility and Child Mortality for 1991 and their inter relations with other variables’, Occasional Paper no. 1 of 1997. Registrar General of India

Registrar General of India (1999) SRS Bulletin Vol. 33 (1)

Oxford University Press (2000) Human Development in South Asia 2000: The Gender Question

WHO (1999) The World Health Report 1999 Geneva

WHO (2000) Women of South East Asia: A Health Profile: India

Mukhopadhyay, Swapna (1999) Women’s Work in South Asia: A Situational Analysis and Policy Concerns, New Delhi: International Labour Organization

Economic Reform and the Status of Women in Nepal



Economic Liberalization

Nepal’s economy went through a difficult time in the early 1980s. Real GDP growth rates during the period fluctuated erratically. Development expenditure rose so much relative to revenues that the share of such expenditures financed by revenue surpluses (total revenue less regular expenditure) dropped to below 17 per cent. Borrowing from the banking system largely financed the enlarged budget deficit. The pace of expansion of domestic credit persisted due mainly to increased banking sector credit to the government.

To counter various disturbing trends in the macro economic scenario the government initiated a reform programme including currency devaluation, financial deregulation, trade liberalization, reduction in budget deficits and policy reforms, removing input subsidies and input taxes. The gradual removal of subsidies from power, agriculture, and industry also forms part of the reform programmes launched in the country.

The recent economic reforms programme in Nepal can be split into a few distinct phases, beginning from 1985 when the economic stabilization programme was put in place.

The economic stabilization programme (1985-86–1989-90)

The economic stabilization programme which was started in 1985

1This paper is largely based on a report prepared by Bhanu Niraula and Samira Luitel of the Agricultural Projects Services Centre for the Gender Planning Network. Sheela Saravanan has contributed to it, and editorial revisions have been made by Ratna M. Sudarshan and Swapna Mukhopadhyay.

sought to correct distortions in the real exchange rate by adjusting the nominal rate and restraining budget deficits, primarily through the controlling of public expenditures. This encompassed a Structural Adjustment Programme (1987-88 to 1989-90) which was designed to

  1. sustain the real GDP growth at 4 to 5 per cent per year,
  2. contain external current account deficit (excluding grants) to about eight per cent GDP on average, and
  3. achieve an overall surplus in the BOP.

The full implementation of SAP was prematurely aborted owing to Nepal’s prolonged trade impasses with India in 1989 and the subsequent people’s movement for the restoration of democracy. However, between 1987-88 and part of 1989-90 SAP was in force. Over this period key macroeconomic indicators were as shown below.


Macroeconomic Indicators of Nepal, 1985/86–1989/90


Source: Policy Framework Paper of IMF and various tables of Economic Survey 1995/96 of MOF, HMG/N. Government’s borrowings from the banking system (Economic Survey Table 8.1)
(Partially adopted from Acharya and Karki, 1996).

Since agriculture remains the mainstay of the Nepali economy with almost 90 per cent of its population dependent on agriculture that generates more than 41 per cent of GDP, the growth stimulus has to originate from the agricultural sector. As agriculture responds more to weather and climatic conditions than to the macroeconomic policy environment, the credit attributed to SAP needs rethinking.

Post SAP development, 1989-90–1991-92

A coalition government of all political parties jointly struggling for the restoration of democracy was formed in 1991. The new popularly elected government began liberalizing the economy by widening the OGL list, rationalizing tariffs, revising the prices of government controlled public utilities, deregulating the financial sector by freeing various interest rates, privatization of state enterprises and by allowing exchange rates to be determined by market forces. Economic indicators for this period are given in Table 2.


Post SAP Macroeconomic Scenario of Nepal, 1989-90 to 1991-92


Enhanced structural adjustment facility (ESAF), 1991-92–1995-96

The major macroeconomic targets of ESAF were as follows. Real GDP was targeted to grow at the rate of 5 per cent a year initially which was later revised to 6 per cent in 1993-94 and 4.2 to 4.8 per

cent every year thereafter. Consumer price inflation was set to increase at the rate of 8 per cent a year, which was later lowered to 5 per cent. Net domestic financing of budget deficit was to be lowered to 0.7 per cent of GDP and ultimately down to 0.6 per cent by 1995-96. Similarly, the fiscal deficit (after grants) would have to be brought down to 6 per cent of GDP in 1993-94 and was set to increase to 8 per cent by 1995-96. Revenue was targeted to increase by 0.5 percentage points of GDP every year.


Macroeconomic Developments, 1991-92–1995-96


Source: HMG/N, MOF Economic Survey 1995-96 (various tables)

IMF’s Policy Framework Paper.

Total budgetary expenditures, which had declined from 18.7 per cent of GDP in 1992-93 to 17.5 per cent in 1993-94, increased to

18.6 per cent in 1994-95 and further up to 19.5 per cent in 1995- 96 due mainly to mounting pressure from the regular budget. If this trend continues, development expenditures will fall behind regular expenditures in the near future.

At the policy level, import licensing was completely withdrawn and the Nepali rupee was made fully convertible in current account transactions in July 1993. Customs duty and sales tax have been rationalized by compressing the rate as well as streamlining the rate structures. In line with the structural adjustment programme, 13 public enterprises were privatized by 1994. Selling prices of government controlled commodities and services such as chemical fertilizers, sugar, dairy products, petroleum products, electricity and drinking water have already been revised upward in stages. The licensing procedure for banks and financial institutions has been simplified. The entry conditions have been eased considerably. Capital adequacy and other prudential norms now govern their entry. The statutory liquidity ratio, other than cash reserves has been withdrawn. The interest rate is governed by the treasury bill auction rate and the central bank’s rediscount rate.

The ninth plan and macro-economic policy, 1997-2002

Prepared in the light of the 20 year long-term perspective plan, the major long-term development objective is to create a society that is cultured, modern, development-oriented and endowed with skill through alleviating the prevailing widespread poverty in the country. It gives emphasis to higher economic growth rates, pro-poor development processes and equitable distribution of income. To attain rapid and sustainable development, plans, policies and programmes are geared towards poverty alleviation, employment generation and promotion, and reduction of regional disparities. Under the plan, the liberal, open and market oriented economic policies pursued by the government will be further strengthened and their implementation will be made more effective.

It is estimated that the GDP at producer’s prices will grow by 6 per cent per annum in the Ninth Plan and by 8.3 per cent per annum in the 12th Plan, thus making the average growth rate of 7.2 per cent during the 20 year period.

Projections 1997–2015

The following table provides some projections for major macro economic indicators provided at the beginning of the Ninth Plan.


Macro Economic Development Indicators of Long Term Plan, Nepal 1997–2015


Source: Unofficial English Version of the Ninth Plan.

To sum up,

Nepal has an open border with India, the country’s major trading partner. Macro economic reforms initiated in India have direct implications for Nepal. For example, following the removal of the subsidy in chemical fertilizers in India, Nepal also had to reduce its subsidy. But the chemical fertilizer market in the country has become unreliable with low quality products. Farmers are averse to buying these. The government has made it a policy to increase agricultural productivity in the following ways:

Some issues in economic reforms in Nepal

Nepal started planned development efforts in the mid-1950s. Guided by the idea of a welfare state, several public enterprises were created to undertake productive and distributive functions. Over time, they become ‘white elephants’ incurring heavy losses and imposing a heavy burden on the treasury.

Nepal entered into a new era of democratic government in 1990. Since then, it has been pursuing macro-economic reform programmes to promote market-led development in the country with renewed vigour. Privatization of state-owned industries was one of the initial steps to encourage the private sector. In recent years, the government has been attempting to join WTO. Notwithstanding the likely benefits of joining the WTO, the necessary infrastructure needed to gain from its membership remains at the minimum in Nepal. For example, the country has a small industrial base and a weak human resource base. In such a situation, the impact of WTO membership with free flow of goods and services may further jeopardize the interest of weaker sections driving them away from subsistence agriculture, with an increased cost of living.

Poverty stricken and with an under-developed industrial and market base Nepal has nonetheless attempted to mobilize market forces to stimulate production. Because of this, questions are being raised about the viability of this approach to development.

The Status of Women

Women make up about half of Nepal’s population, which crossed 22 million in 1999. But the female population is not evenly distributed across the country. Topographically, the country can be divided into three ecological zones, the mountains, the hills and terai, each accounting for 7.8 per cent, 45.5 per cent and 46.7 per cent of national population respectively. In recent years, the terai has become a ‘safety valve’ for out-migrants from the hills and mountains. As a result, the share of terai population to total population has increased from 37.8 per cent in 1971 to 46.7 per cent in 1991, an increment of almost 10 per cent over this period.

For planning purposes, Nepal is divided into five development regions. The distribution of population across development regions and the corresponding sex ratios are shown in Table 5. More than one-third of the people live in the central development region (33.4) followed by the eastern (24.1) and western (20.4) development regions. The share of the mid-western and far-western development regions is 13 per cent and 9 per cent respectively. The sex ratio is skewed in favour of females in the hills and mountains, reflecting male out-migration. In contrast, the male population, in the terai exceeds the female population. Male migration to urban areas and the terai has the consequence that women’s labour is needed both for work on the farm as well as within the household in the rural/mountain areas. The sex wise analysis of the working age group (15-59 yrs) shows that, there were 52 women to every 100 persons in the rural areas as against only 48 women to every 100 persons in the urban areas (Table 6).


Percentage Distribution of Total Population and Sex Ratio in Different Ecological Regions of Nepal


Source: Population Monograph of Nepal, Central Bureau of Statistics, 1995, p. 29. Preliminary Results of Population Census 2001.


Percentage Breakdown of Population of 15–59 Age Group in Rural and Urban Areas by Sex, 1991


Source: Statistical Profile on Women in Nepal, 1995, Shavitri Singh, p. 14.

At the level of the national government there is a lack of explicit policy and programme formulation to reduce gender disparities; at the community level, social and religious norms and taboos continue to prevail and constrain the activities of women; and at the household level there is often an overt discrimination against women. The consequence is that the status of women in Nepal continues to be low. They are disadvantaged in many ways: as poor people they live under the same harsh conditions as their male counterparts, as women they suffer from social, cultural and political biases, and as heads of households they have to carry out the full traditional roles with the added responsibility of household and production management. In contrast, men have benefited from expanded educational and employment opportunities, so that women’s relative status has probably been declining.

Nepal has more than 60 caste/ethnic groups and women’s status varies among these groups. The life of the Nepali people in most of these communities is governed by traditional cultural values. In many ways, these stand as a barrier in the holistic development of women. The state religion of Nepal is Hinduism, (Nepal is the only Hindu country in the world), and about 87 per cent of the population was Hindu in 1991, followed by Buddhists (7.8 per cent) and Muslims (3.6 per cent) (CBS, Nepal 1995). Women’s status may also vary by religion: women of Buddhist religion enjoy more freedom than Hindu women.

The irony of having powerful goddesses and dis-empowered women suggests that a closer look is needed at the religious, cultural and legal framework of society. The framework of rights is provided by national policies relating to such matters as the right to property, divorce and marriage. In practice, despite legal freedom, women remain vulnerable and dependent on men, with identities largely derived from the roles they play in relation to men—wife/mother/daughter. Men are expected to be caretakers, protectors and

masters of the household and exercise considerable control over women in every walk of life. Children are identified by the father irrespective of whether he is alive or dead. A widow still needs to produce the identification of her husband in any written document. Given the socio-cultural values there is a high degree of son preference.

The nature of gender bias gets mediated through the extent of household poverty. By and large, poverty accentuates the hardship of women’s lives. Overall 40 per cent of the population lives below the poverty line. In these households women are often found to bear a disproportionate burden of work, and yet receive little recognition since much of the family and farm related work is classified as economically unproductive. The incidence of male violence and domination, resulting in psychological depression and dependency syndrome among women is reportedly high.

The Nepal marriage act defines the legal age of marriage with parental consent as 18 years for boys and 16 years for girls. Without parental consent, these ages are 21 years for boys and 18 years for girls (Shavitri Singh 1995). Almost half the girls in urban areas are married in the age group 15 to 19 years (Table 7).


Percentage Married in Age Groups 10 to 14, 15 to 19 and 20 to 24 Years—1991


Source: Population Monograph of Nepal, Central Bureau of Statistics, 1995, p. 187.

Marriage represents a major transition, from a life of relative liberty to one of control. An unmarried girl enjoys more freedom and privileges at her parents’ home than in her husband’s home where her life is limited to the household. Her prime duty will be to please her in-laws, to adjust to the norms of her new home, and it is this that will determine how secure her future will be. A woman’s position in her marital home is influenced by the strength or weakness of her natal home. An economically and socially strong parental background and parental support raises her position in her

husband’s home, and vice versa. Acharya and Bennett (1981) explain the position and power of a woman in her husband’s home as being influenced by

  1. Women’s continuing relationship with their maiti or natal home;
  2. Women’s sexuality which encompasses both their ability to bear children and to give (or withhold) pleasure to their husband;
  3. Women’s ability to uphold or tarnish the honour/reputation (izzat) of the family in the community at large;
  4. Women’s own labour and ability to contribute to the productivity of the affinal family.

Thus, a woman in a small nuclear agricultural family where her contribution is more visible enjoys more freedom than in a large economically better off family.

As far as kinship practices are concerned, these vary across different ethnic groups. The Sherpas of the Himalayas practice polyandry (a woman is married to all the brothers in the family). Studies show that women in some communities, including the Newar women of Bulu (Pradhan 1981), the Tharu women of Dang (Rajaure 1981), the Kham Magar of Thabang (Molnar 1980), the women of Baragaon (Schuler 1987) have been presented and treated as equals to men within the community. Similarly, equality and freedom among the Rai women have been described by McDougal (1968); among the Limbus by Jones and Jones (1976); among the sherpas by Holmberge (1989). Women from these communities have traditionally been very hard working and enjoy inheritance rights in property, although national law denies property rights to women. The national law of Nepal provides inheritance right in property only to sons. Married daughters can claim a share only if there is no grandson. An unmarried daughter over the age of 35 is entitled to a share in the parental property but if she subsequently marries she has to return the property to her brothers.

Conventional Indicators of Gender Inequality

Women lag far behind men in all spheres — education, personal property, health status, legal, political and social rights, showing a high sex disparity in these ‘conventional’ indicators.


The health status of Nepal’s people is one of the lowest in the South Asian region and this is particularly true for females. According to the 1991 census, female life expectancy is 54 years as against 55 years for men. High birth rates, low life expectancy, high infant and maternal mortality rates and high death rate indicate the poor health status of women. Infant mortality rates and under 5 mortality is higher for the female child (see Table 8). Discrimination in care and nurture, differential socialization, low levels of nutritional intake, poor sanitation and health facilities are some of the leading factors which contribute to higher mortality among female children. Mortality levels (neo-natal, post-natal, infant and child) are higher in rural areas than in urban areas and in the western part of the country as compared to the eastern part.


Selected Development Indicators 1991



Maternal Mortality Rate/1000 live births


*Source: CBS Population Monograph of Nepal 1995, p. 80, 95.

Women’s high work burden and illiteracy have a direct impact on their health. The limited spread of the modern health care system and low utilization of health services contribute to high maternal and infant mortality rates. In addition, low consumption of nutritional food, repeated and closely spaced pregnancies and inadequate birthing services contribute to complications during

childbirth, post natal problems, infant deaths and maternal mortality. Every year an estimated 927,000 pregnancies take place in Nepal of which 40 per cent are considered high risk for the mother.

While poverty is a major part of the explanation, gender discrimination in the care and nurture of children contributes to higher death rates among girls. A low level of nutrition, poor sanitation and general ill health all contribute to child mortality and morbidity. Studies indicate that among the children who survive, the female child is nutritionally worse off than the male. Girls under 5 are at twice the risk of being under-nourished than boys (RIDA 1991). A study on the status of the girl child (RIDA 1991) found more girls (50 per cent) than boys (40 per cent) to be malnourished, anaemic and stunted.


Sex Differentials in Children’s Nutritional Status


Source: Pradhan A., Aryal R.H., Regmi G., Ban B., Govindsamy P., Nepal Family Health Survey, 1996. Kathmandu and Calverton, MD: Ministry of Health, New Era and Macro International Inc.; 1997.

Ref: WHO 2000, Women of South East Asia, A Health Profile, p. 280.

Protein Energy Malnutrition (PEM) is a chronic problem in Nepal. Women are frequently found to lack Vitamin A, iron, iodine and protein in their diet. Nutritional anaemia among women during pregnancy is another common problem. The situation is worse in the western rural areas of the country. Women are found to be the major users of traditional medicine. It is estimated that around 91 per cent of deliveries take place at home. Low maternal height and weight can cause considerable health hazards for women. The height below which a woman is considered to be at risk is in the range of 140-150 cm. It was found (MOH 1997) that about 15 per cent of women are below 145 centimetres. The Nepal

Family Health Survey conducted in 1996 showed that a growing risk factor for women in recent years is the increasing prevalence of sexually transmitted diseases (STD) and HIV\AIDS (Table 10). Diseases like syphilis and gonorrhoea in early pregnancies can lead to stillbirths, neonatal deaths and infected infants (UNICEF 1996).


AIDS and HIV Infections


Note: *Population estimates for 1996 based on UN population figures for mid-1994 with annual growth rate applied as appropriate.

Source: WHO, HIV-AIDS and Sexually Transmitted Diseases: An Update. New Delhi: WHO Regional Office of South East Asia, 2000 (unpublished report).

Ref: Women in South East Asia, A Health Profile, p. 42, WHO 2000.

Another high risk factor for women in Nepal is the practice of early marriage and young motherhood. According to Ministry of Health estimates about 60 per cent of girls marry at an age below 18 years. The age at motherhood has declined since 1991 with 42.4 per cent girls being mothers at the age of 19. Problems of malnutrition and small pelvic bones and birth canals also result in stunting among children. Among the older women close-spaced births and a high birth rate keep both mother and child at risk. The 1991 census reports an average of 5.1 children of Nepali women of reproductive age, which reportedly declined to 4.6 in 1996 (MOH 1997). According to the MOH survey (1993) the median birth interval was 34 months.

Quality and accessibility of health care services

The utilization of health care services is low in Nepal. Data collected from the NMIS shows that a large majority of pregnant women do not visit the ANC. The medical council directory (1994) presents a figure of 1547 total doctors out of which 288 are women whereas the economic survey reports a total of 1497 doctors. The national coverage stands at 12,611 persons per doctor. It is estimated that 60 per cent of the doctors are working in Kathmandu. This means that the doctor population ratio is very

low in rural areas. Nurses and Auxiliary Nurse Midwives (ANMs) are the trained practitioners who provide services to the people. But there is a shortage of nurses and health practitioners in hospitals. The ratio of nurses to the population in hospitals is 1: 6,295. (MOF 1994). The number of ANMs is low (2,126) and many positions remain unfilled.


Maternal Health Status


Ref: WHO, 2000, Women of South East Asia, A Health Profile, p. 188, 281. Nepal Demographic and Health Survey 2001.

The use of contraceptives has increased in recent years showing a growing consciousness about reproductive burdens. However most of the contraceptives and sterilization are directed at women. Female sterilization increased more than 358 per cent while the male sterilization rate increased by only 134 per cent from 1981 to 1991 (Acharya and Bennett 1981).


The total literacy rate in the rural areas in 1991 was 36.8 per cent compared to 66.9 per cent in the urban areas. In the rural areas, 51.9 per cent of males and 22.0 per cent of females were literate in 1991. In urban areas, 78 per cent of the males and 54.8 per cent of the females were literate. There is evidently a wide disparity in the literacy rates in the rural and urban areas. The higher literacy rate in the urban areas is attributable to the greater necessity of being literate in these areas, the greater access to educational facilities and the tendency of educated people to migrate to the towns.

Concerted efforts to raise literacy have resulted in an average overall level of 38 per cent in 1995-96, with female literacy standing at 24 per cent and male literacy at 52 per cent (Table 12). The lower literacy rates among females are a result of a variety of historical, economic and social reasons. Social prejudices against female education, restrictions on mobility, low social status granted to the females, the system of early marriage and low participation of females in formal sector too result in lower literacy for females. School enrollment at the primary level among girls is lower than among boys. Although the literacy rate of men and women is increasing, the gender gap is also increasing. The gender gap in 1991 was more in the rural areas (29.9) than in the urban areas (23.2).


Literacy Rates in Rural and Urban Areas by Sex (1981–1997)


Source: *CBS Population Monograph of Nepal 1991, pp. 379–81

**Nepal Living Standards Survey Report 1996 — Main Findings, Vol. I, p. 57

M-F: Male Female Gap

The content of schooling continues to present conventional gender stereotypes with textbooks showing women in the private/household sphere and men in public affairs (CERID, 1997). The media, by highlighting a woman’s body as a commodity and a source of income presents a challenge to the educated elite (Thapa 1995). The physical environment of the school and the teaching/learning environment is often not congenial for young girls and contributes to high dropout rates. Many public schools lack toilet facilities and there is a shortage of female teachers in schools which contributes to lower motivation for girls’ attendance.

Although women are the main contributors to domestic activities and are often solely responsible for family maintenance, their access to knowledge, skill, resources and power is very limited.


A large part of women’s work is not considered as economic activity. As a result, in the 1991 census only 45.2 per cent of women as compared to 68.2 per cent of men are classified as economically active (Table 13).


Economic Participation Rates by Sex


Source: CBS Population Monograph 1995.

Ref: Gender Planning Network Report, October 1999, pp. 30.

The occupational structure reveals that 81 per cent of the population was employed in the agricultural sector in 1991. While the per cent working in agriculture is greater (92.7 per cent) for females than males (80.2 per cent), the per cent working in service, production and sales is higher for males than females.

The urban-employed work in more diverse occupations than the rural-employed. What is interesting is the fact that around eight per cent of the urban-employed females were found in professional and technical jobs compared to their male counterparts who constituted slightly over five per cent (5.13) (Table 14). However, this data does not tell us where these women are located in the hierarchy. As discussed later, few women have reached top decision making positions.

Self-employment in agriculture is the predominant form of employment in the country (Table 15). More then 92 per cent of the employed labour force is in rural areas and 81 per cent is in agriculture. Only 19 per cent of the rural workers and 21 per cent of all workers work as wage labourers. A majority of the wage labour is in the unorganized sectors. Only between 7 and 10 per cent of all workers are employed in the organized segment of the economy. The employment status of the rural population shows that own account workers or self-employed workers account for almost 69.5


Percentage Distribution of the Employed Population (Aged 10+ years) by Major Occupational Groups, Nepal 1991


Source: CBS Population Monograph of Nepal 1995, p. 211.

per cent of population (83.7 per cent for males and 75.3 per cent for females). About 11 per cent of females are employees in rural area while only 0.4 per cent hold the status of employer. Corresponding figures for males are 18.25 per cent and 0.7 per cent respectively. In urban areas, own account workers constitute almost half of the female work force followed by employees (46 per cent). Corresponding figures for males in urban area are 39 per cent and 58 per cent respectively.


Employment Status of Economically Active Population by Sex 1991


Source: Population Census, 1991, CBS.

Ref: Statistical Profile on Nepalese Women, Meena Acharya, p. 63.

Despite diversity in urban employment agriculture still provides employment to one-fourth of the urban population including 38 per cent for women and 20 per cent for men. The personal and community service sector is the largest employer in urban areas in which almost 31 per cent of women and 36 per cent of men are employed. This is due to slow industrial growth and less opportunity in the secondary employment sectors. This pattern has not changed over time though a marginal increase is shown from 1971 to 1991 in the secondary and tertiary sectors. Women’s labour participation in secondary sectors in the urban areas shows near parity with that of men in 1991. This is due to the growing carpet and garment industries in urban areas, which absorb more female labour.

Among the workforce the level of education is low. The available evidence suggests that 30 per cent of the workforce in 1991 have no schooling at all and 56.94 per cent had a school leaving certificate, and about 7 per cent had a higher degree. Thus the level of education of the workforce is abysmally low even by the standards of other South Asian countries, and the level of education of female workers is substantially lower than that of male workers.

According to ILO-SAAT the historically inherited regional inequalities and the uneven pattern of development have made Nepali labour highly mobile. Labour flows between Nepal and India have given rise to a special problem. Unskilled labour from the hills and the mountains migrates to India, seasonally as well as for extended periods; semi-skilled and skilled labour from India migrates to Nepal, usually for extended periods. The net migration is probably close to zero.

Women’s work

Data collected by the Central Department of Population Studies on employment shows that 7 per cent of women are employed in paid jobs, 30 per cent in various income earning activities while a large chunk of women (63 per cent) are involved as unpaid family workers (CDPS, 1997). This partly explains their low socio-economic status.

Overall, the participation rate is higher for males than females, the percentage of female workers in rural and agricultural

occupations is higher than the corresponding percentage for male workers, the percentage of female workers employed in the organized sector is much lower than the corresponding percentage for male workers, and underemployment affects female workers more than male workers. Thus not only are the female workers confined to less productive and less remunerative jobs, they also have fewer employment opportunities than males. It is possible to argue that the labour market disadvantages for female workers arise partly because they are less skilled and less educated than male workers. But this reflects discrimination against women in education and skill development programmes. In other words, gender based discrimination in the labour market in part reflects the generally low social status of women.

It has been noted above that about 19 per cent of wage labourers in rural areas are in regular wage employment. Detailed facts about their conditions of employment are not known, but it is generally believed that many of them are in a bonded or quasi-bonded state. For example, what is known about one group of such labourers, known as Kamaiyas, leaves little doubt that they can be classified as bonded labourers.

In Nepal, women are more vulnerable than men to poverty and unemployment. Available data indicate that women face higher levels of unemployment and underemployment in both rural and urban areas. Women are the ones left behind to take care of the family when men temporarily migrate elsewhere for jobs during slack seasons.

Bonded labour

It is unfortunate that a system of bonded labour is still prevalent in the western terai districts of Nepal, including Dang, Banke, Bardiya, Kalali and Kanchanpur districts. The bonded labourers under the system are called Kamaiyas. The Eighth Plan introduced some policy measures to get rid of this age-old exploitation, through extension of educational and vocational training opportunities. The government has been attempting to implement certain programmes to help the Kamaiyas and programmes are also being launched by certain NGOs for empowering the Kamaiyas. So far, however, these programmes have not shown any meaningful results.

Wage rates

Wage rates in large parts of the organized sector are not markedly different from those in the unorganized sector. These facts suggest the low bargaining power of all workers. Discrimination against women in job opportunities and wage rates is a common pattern in both the rural and urban areas. Female wage rates as proportion of male wage rates in agriculture have come down from 1980-81 to 1991. In recent years wage rates seem to have declined for both males and females which is also indicative of increased poverty levels. However, reliable data on wages are scarce. Analysis is made more difficult because there are many different modes of wage payment in Nepal. At least a part of the wage is paid in kind and labourers often enter into a variety of credit relations with their employers.


Female Wage Rates as Proportion of Male Wage Rates (in Rs.)


Source: Rashtra Bank Quarterly bulletins for various years

— denotes not available

Ref: Statistical Profile on Nepalese Women 1994, Meena Acharya, p. 79.

The present employment situation is more of a problem for women than for men in the sense that higher proportions of women work long hours in poor working conditions without compensation or insurance. Lack of private property and control over productive resources or land property has barred women from accessing credit for productive works. The CBS (1997a) and NLSS reports show that only 4 per cent women against 11 per cent men are engaged in self-employment outside agriculture, while 81 per cent women are still confined to self-employment in the agricultural sector. This

is a low productivity and high labour intensive area. A study conducted by Strii Shakti (1995) reports that rural women’s workload has increased since 1981.

Women in professions

There are very few women working in the legal field. The Supreme Court report of 1994 shows that there are only 251 persons working in this field. Not a single woman has served as a judge of the Supreme Court until the present day. Only two women are serving as judges in the Court of Appeals. Women also do not own senior advocate positions. They may study the law, but few are able to enter the profession. Of those who do, most are found to be working for the support of women as pleaders in different courts.

Women entered formal politics along with men in 1950 with the reshuffle of the Rana regime in the country. Since then women are represented in both national and local bodies but their participation is very low. Women’s representation in the bureaucracy is also very low. Only one woman so far has served as ambassador. After 1991 not a single woman has been placed in the diplomatic corps. Limited participation in politics, bureaucracy and judiciary does not stop women from making a remarkable contribution in the decision-making process at the household level. Women serve as decision-makers in farm management, domestic expenditure (food items, clothes and other expenses), the children’s education, religious and social travel, household maintenance and also capital transactions. However, women’s decision-making roles seem to have declined in recent years.

Legal rights

Nepal’s Constitution states that there will be no discrimination of any kind on grounds of sex, caste, religion and race. Yet there are certain national codes which privilege men. The rights to inheritance to property, marriage and divorce, tenancy and transaction rights still favour men. The social and human aspects also do not seem congenial for women, as there still remain certain clauses in public law (Mulki Ain) that put women in a vulnerable position. For example property laws allow a daughter to inherit the deceased parent’s property only when there is no surviving male (son or son’s

son) of either parent. The law does not recognize any transaction carried out by a wife without the consent of her husband. The Army Act 2016 bans the recruitment of women into the Royal Nepal Army. A family with only one son is prohibited from giving him up for adoption. In the case of a single daughter, her adoption would be legal. There are no legal safeguards against sexual harassment of domestic violence targeted at women.

Nepal became a signatory of several international conventions after the restoration of democracy in 1990. It has ratified CEDAW (Convention on the Elimination of all forms of Discrimination against Women) in 1991, which reinforces the rights of girls and women to decisively eradicate inequality and discrimination. Others conventions include

Awareness of such commitments is low, both among men and women.

Non-conventional Indicators of Gender Inequality

The cultural construction of Nepali society is shaped by Hindu theology, which advocated and promoted patriarchal attitudes and practices that exclude women from a variety of functions. This has resulted in gender-stereotypes where women’s role as unpaid productive and reproductive labour is assigned negligible power or control over economic resources.

Women’s autonomy

The data reported in tables 17–19 below are from a survey conducted in 1998 to assess women’s autonomy and reproductive behaviour from two urban areas of Nepal (Niraula and Lawoti 1998) and provide some indication of the extent to which conventional roles are accepted, and shape interactions between men and women.


Wives’ Perceptions Towards Gender Roles and Duties


P-value *0.001

Source: B.B. Niraula and Dovan Lawoti, 1998.

Attitude towards husband-wife relation


Wives’ Response to Conditions under which Husband is justified in Leaving his Wife



P-value *.05**, .005*** .001

Source: B.B. Niraula and Dovan Lawoti, 1998.


Women’s Responses to Household Income, Ownership of Resources and Attitude Towards Self-support


Trafficking of women/children and prostitution

Trafficking and prostitution in Nepal have historical, religious and traditional roots. These have become a major concern today. Although exact estimates are lacking, it is generally agreed that about 200,000 women and children have been trafficked to India, and forced into prostitution in red light districts in different cities and towns (CWIN, 1998; Asmita, 1998).

The various factors that contribute to trafficking and commercial sex work can be summed up as follows:

In addition to poverty and low status of women, other causes of women/girls trafficking are tradition and culture, open borders, a lack of political commitment, lack of effective enforcement of existing laws, the inadequacy of existing laws, and the lucrative nature of the business. Reported cases of trafficking suggest that often the offenders go scot free because of intricate networking with

officials and politicians. Trafficking and CSW is common among all caste-ethnic groups in the county.

Violence against women and children

Like child labour and sex work, violence is a common feature in Nepali society. Strongly entrenched patriarchal norms and traditions, fortified by selective adherence to ritualistic versions of Hinduism, provide a conducive environment for women’s subordination. Poverty, ignorance, illiteracy and isolation are other traps that help to sustain women’s subordination to men. One of the several manifestations of women’s subordination is violence against women. Women and children are victims of several types of violence: domestic, rape, polygamy and customary violence such as practices of Badi2 and Deuki3 system in the country.

Liberalization and gender concerns

The government’s explicit concern with women’s welfare began with the celebration of the International Women’s Year in 1975. The Women Services Co-ordination Committee (WSCC) was asked to formulate a national action plan for women’s development. Subsequently, a task force was formed, which brought into being a national plan of Action for Women’s Development in 1981 recommending programmes in seven sectors, i.e. education, health,

2Badis are a tribe found in the mid western and far western regions of Nepal. They are migratory people moving south from the foothills of the Himalayas under moderate climatic conditions. The Badi community is dependent on females for economic sustenance of the community through the sex trade. They remain migratory today, families continue to travel as a group and family bonds and ties remain strong. The Nepali ‘Badi’s seem to be very similar to the ‘Bedia’ community who are found in Rajasthan. Very little on the Nepali ‘Badi’s was available on the net. Cf. International Journal of STD and AIDS 1993; 4; 280–283. Also

3Deukis are young girls between five and nine years of age who are offered to temples. They generally end up searching for livelihood as servants and are also compelled to sell their bodies. Often men from rich households buy up young girls from poor families to be offered to temples before they reach menarche and are subsequently used for sexual gratification. See and http://www.ce/

agriculture, employment, co-operatives, forestry and law. Since the sixth plan the National Planning Commission has incorporated special policies and strategies in their periodic development plans, which mainly focus on increasing women’s access to education and training, institutional credit and wage employment. In 1990, Nepal ratified the 1979 UN Convention on the Elimination of all forms of Discrimination Against Women (CEDAW), and the Constitution of the Kingdom of Nepal 1990 guaranteed that no ‘discrimination in respect of remuneration shall be made between men and women for the same work’. Over the years, gender concerns have been given attention in general projects and programmes while some targeted programmes have also been launched to promote women’s participation in gainful work. In 1993, the Ministry of Education made it obligatory to recruit at least one female teacher in all public primary schools of the country. The National Council for Women and Child Development was constituted in 1995 under the chairmanship of the Prime Minister.

Targeted credit programmes such as Production Credit for Rural Women (PCRW), Women Development Programme under the Small Farmer Development Project (WDP/SFDP) and the Micro Credit Project for Women (MCPW) have been the important programmes which directly address the problems of women workers. Some of the programmes have had low coverage, replicability and viability, thus constraining their impact on women’s employment.

The status of women in Nepal is the outcome of culturally accepted norms and attitudes. The review above shows that there are several areas of concern, including the prevalence of illiteracy and large gender gaps in levels of literacy; high birth rates, poor health facilities, high mortality; high work burdens on women as a consequence of the high rates of male migration. Poverty affects women more sharply than men. Apart from the material deprivations that follow as a result of being poor, women face greater social, cultural and political biases and consequently have a low say in decisions affecting their lives within the household, community or nation. They are expected to contribute to production and income earning but without any exemption from traditional and ritual obligations. Reports that there is a high incidence of male violence resulting in depression among women provide further confirmation of high stress levels.

With economic growth and liberalization, there has been an expansion of educational and employment opportunities. However, it appears that men have been able to benefit from this more than women. The country remains largely agricultural, in parts precapitalist, with a large subsistence economy, and women are still mainly confined to this kind of work and life. There is a contradiction between ‘modern’ thinking that has gained prominence with the opening up of the economy, and the ‘traditional’ outlook on life that is still largely prevalent over many parts of the country.

There are positive influences at work: recognition of the value of equality, as seen in the ratification of international conventions, a legal, political and constitutional framework that guarantees rights and creates new opportunities, and the entry of new social images that could contribute to autonomy and empowerment. On the other hand, there are negative outcomes, some of which are direct consequences of the process of liberalization. These include the increased work burdens of women in areas from where male migration has been stepped up; the alleged increase in violence and depression and in the number of women going into commercial sex work due to a lack of opportunities. There is some evidence of increased employment opportunities for women in the trades that have been stimulated by a growing export demand such as garments and carpets, although more research will be needed to understand what the quality of this employment is like.


Acharya, Meena (1994) The Statistical Profile on Nepalese Women: An Update in the Policy Context, Institute for Integrated Development Studies, Kathmandu, 1994

Acharya M. and L. Bennett (1981) Status of Women in Nepal (The Rural Women of Nepal: An Aggregate Analysis and Summary of Eight Village Studies, Vol. 2, part 9. Kathmandu: Center for Economic Development and Administration, Tribhuvan University

Central Bureau of Statistics (1996) Nepal Living Standards Survey Report Vol. I, National Planning Commission Secretariat

Central Bureau of Statistics (1996) Nepal Living Standards Survey Report Vol. II, National Planning Commission Secretariat

Central Bureau of Statistics (1996) Nepal Statistical Pocket Book, Nepal National Planning Commission Secretariat

Central Bureau of Statistics (1997) Statistical Year Book of Nepal

Central Planning Commission Secretariat (1995) Kathmandu

CERID (1997) Gender and Secondary Education, Kathmandu: Research Center for Educational Innovation and Development. Tribhuvan University

Holmberge, David (1989) Order in Paradox: Myth, Ritual and Exchange Among Nepal’s Tamangs, Ithaca: Cornell University Press

Jones, R.L. and S.K. Jones (1976) The Himalyan Women: A Study of Limbu Women in Marriage and Divorce, California: Mayfield Publishing Company

McDougal, C. (1968) Village and Household Economy in Far Western Nepal, Kathmandu: Tribhuvan University

MInistry of Finance (1994) Economic Survey 1993-94, Kathmandu, Ministry of Finance

MOH (Ministry of Health) (1997) Nepal Family Health Survey 1996, Kathmandu: Family Health Division, Ministry of Health

Molnar, A.M. (1980) Flexibility and Options A study of the Dynamics of Women’s Participation Among the Kham Magars of Nepal . Ph.D Thesis, Madision: University of Wisconsin

Niraula Bhanu B. and Lawoti, Dovan (1998) ‘Women’s Autonomy and Reproductive Behaviour in Two Urban Areas of Nepal’, Contributions to Nepalese Studies, Vol. 25 (Special Issue). Tribhuvan University

Pradhan, Bina (1981) Institutions Concerning Women in Nepal, Status of Women in Nepal, Kathmandu (Volume I, Part 3): Center for Economic Development and Administration, Tribhuvan University

Rajaure, D. (1981) Tharus of Dang. Status of Women in Nepal, Kathmandu: Center for Economic Development and Administration, Tribhuvan University

Schuler, S. (1987) The Other Side of the Polyandry, Boulder: Colorado West View Press

Singh Durbar (1997) Nepal-Population & Development Journal, Ministry of Population & Environment, Kathmandu

Singh Durbar (1997) Status of Population in Nepal, Ministry of Population & Environment, Kathmandu

Singh Shanitri (1995) Statistical Profile on Women of Nepal, Shtrii Shakti, Kathmandu

Thapa, 1995

UNICEF (1996) Children and Women of Nepal: A Situational Analysis, Kathmandu

WHO (2000) Women of S.E. Asia, A Health Profile, New Delhi: WHO Regional Office of SE Asia

Gender and Adjustment Policies:
Evidence from Pakistan1




Since the 1980s many developing countries, including Pakistan, have adopted two major economic reform programmes, namely the stabilization and structural adjustment programmes. The stabilization programme, adopted to deal with fiscal deficit and persistent balance of payments crises, is believed to be a short-term one whereas the structural adjustment programme is a medium/long-term plan designed to deal with structural imbalances in the

1This research is conducted under the project “Gender Planning Network” for the region of South Asia. The funding was provided by the IDRC, Canada. The ISST, India is the coordinator for the region. We are thankful to IDRC, Canada for providing financial support and ISST for providing all the support and help for the study. Randy Spence from IDRC, Canada and Swapna Mukhopadhyay from ISST, India deserve special thanks for supporting the work on this important topic in the region. We are grateful to Dr A R Kemal, Director PIDE for providing all the support and encouragement for completing this research. Thanks to the Owners of all the industrial units in Faisalabad, Sialkot and Karachi, and a number of other persons in the three cities for making it possible to complete the survey. Thanks to Ms Ayesha Salam for helping us in the tedious work of data editing and reading the first draft of the report. Thanks to the team of Computer Centre for data entry and to Mr Afsar Khan and others for providing excellent typing support. Last but not the least we thank all the factory workers for making the survey a really joyful and learning experience for all of us. Thanks and apologies to all of those whom we may have forgotten to acknowledge. All the remaining errors are the sole responsibility of the authors.

economy in order to make it internally and externally competitive. A generally held view is that these adjustment policies have led to unemployment, low investment, decline in real wages, capital flight, rise in inequality and poverty, and deterioration in the living conditions of the poor, at least in the short run.

Most of these macroeconomic policies claim to be gender neutral. But a number of studies, conducted in the late 1980s and the 1990s, argue that ignoring the gender dimensions of the cost in terms of time, effort and resources in production and maintenance of human resources results in miscalculation of the effects of macroeconomic policies on productivity and efficiency (see for example, Ali 2000, Ghosh 1994, Khan 1986, Shah 1994 and Standing 1989). These studies argue that adoption of an export-led industrialization strategy and labour market deregulation has resulted in the post adjustment era, in reliance on low wage labour and informalization of jobs to reduce production cost. Since females are concentrated in low paid jobs, this informalization of the job market has led to marginalization of the female workforce. Empirical studies also reveal that in the post adjustment period poverty is pushing male workers to migrate from rural areas to urban areas to search for jobs. As a result, the burden of agricultural activities is increasingly falling upon women. They are becoming the defacto heads of households and are coping with heavier work burdens. The studies have also found that even if employment is an important factor determining female empowerment in the long run, there is an increasing trend towards feminization of agricultural labour, and feminization of poverty in the post adjustment period, at least in the short run (see for example, Ali 2000, Brown 1992, and Khan 1999). However, the overall effect of the adjustment policies, particularly with a gender perspective, is difficult to measure as it varies across countries, across sectors, and across individuals within a household.

In Pakistan, as in other developing countries, the implementation of stabilization and structural adjustment programmes since the mid-eighties has affected economic activity adversely. Studies show that the adjustment programme has led to increased poverty levels both in urban and rural areas, as a result of higher unemployment rates, downsizing in the banking sector and also in other labour intensive projects and reduction in remittances coupled with high inflation rates that surfaced in the early 1990s (see for example

Amjad and Kemal 1997, and Mahmood 1998). How have these adjustment policies and the consequent decline in economic activity affected males and females in Pakistan? This is an important issue in a country like Pakistan, where all the conventional indicators like education and health show a wide but declining gender gap, as well as lower female involvement in productive activities mainly due to limited access to productive inputs, lower investment in human capital, discrimination in the labour market, underestimation of females’ contribution and other social and cultural factors. It needs to be recognized that despite the lack of institutional support and the existence of widespread discrimination, women play an important role in economic development2 (see for example, Ali and Ahmed 1983, Freedman and Wai 1988, Khan and Bilquees 1976, and Sarwar, Saleem and Khan 1993).

In this essay, we look at a survey of industrial workers in Pakistan in order to analyse the gender dimensions of various conventional and non-conventional indicators in the post adjustment period. Conventional indicators include education, health, labour market involvement, control of assets and income, mobility and decision-making, while non-conventional ones include violence and psychological well being. In the post-adjustment period in Pakistan, the main emphasis of macroeconomic policies has been on export-based industrialization and market deregulation. We have thus focused our survey on workers in formal and informal industrial units which produce Pakistan’s major exportable goods (i.e., textile and ready-made garments, sports goods, surgical goods and fisheries).

Our study begins with a discussion of the impact of adjustment policies in Pakistan. We then go on to look at the gender dimensions of labour force participation in the industrial sector. This analysis is based on secondary data of export-based industries. The sample, survey methodology and questionnaire are discussed in the next section and, based on survey data, the gender dimensions of socio-economic (conventional) indicators, for example, age structure of population, education, health, and labour market participation

2According to these studies livestock related activities account for a large proportion of females’ time per day, i.e., 3 to 5 hours per day. Similarly, Ali, M.M. and N. Ahmed (1983), report that in rice growing area of Punjab females spend at least 5-hours a day in farm related activities.

are then examined. We also look at the issue of female autonomy and empowerment as well as the question of gender violence and the mental well-being of the respondents, and we conclude with some policy directions.

Adjustment Policies

Like other developing countries, Pakistan implemented its stabilization and structural adjustment programme with support from multilateral institutions such as the International Monetary Fund (IMF) and the World Bank (WB) in the late eighties and nineties. These adjustment programmes are adopted to reduce the fiscal deficit, to improve the balance of payments, to make the economic structure more productive and to promote internal and external competitiveness. Most studies evaluating the performance of the adjustment programme have indicated limited success, with serious consequences for the poor in developing countries, and Pakistan is no exception. The adjustment programmes did not succeed in improving the economic indicators like output growth, employment generation, poverty and others.

The efforts to reduce the fiscal deficit focused on broadening the tax base and on reducing public expenditure. The fiscal deficit declined initially as a result of widening the tax base, and an increase in prices of public utilities but later due to a decline in public expenditure. The decline in fiscal deficit resulted in worsening of income distribution and a rise in poverty. According to Kemal (2000), from 1987–88 to 1990–91, the increase in tax incidence on the poor and a reduction in the public expenditure on subsidies, primary education and health, accentuated the problem of poverty in Pakistan. The effect of the decline in public expenditure on social sectors is expected to be disproportionately higher on the users of these services, that is, the poor and vulnerable, particularly women. Furthermore, the withdrawal of incentives, a poor law and order situation, and frequent changes in government policies resulted in poor investment trends despite the availability of credit.

As mentioned earlier, the overall impact of the adjustment policies is a slower growth rate of output, declining employment and a rise in poverty. In fact, employment elasticity was negative

in the 1990s, in the agriculture sector (–0.06) as well as in the manufacturing sector (–0.37) (Kemal et al 2000). This decline in employment elasticity indicates a rise in capital intensity in the two major sectors of Pakistan’s economy. Due to the rise in inflation rates real wages declined. This, coupled with a rise in food prices, affected the population adversely and the effect was disproportionately higher for the poor.

Trade liberalization, with a reduction in tariff and non-tariff barriers and market deregulation opened the protective markets particularly for agriculture and industrial products. This has particular implication for women in Pakistan as it is expected to affect the labour market significantly. Policy changes have a differential impact on the employment of females and males, on conditions of work, unpaid labour and on the distribution of work. On the one hand, the low-cost labour force may become the main beneficiary under trade liberalization if labour intensive policies are adopted in the post liberalization period. On the other hand, privatization may increase reliance on capital intensive technology reducing the demand for low skill workers. Consequently, the share of females in employment may go down, despite feminization of the labour force. In Pakistan, the expansion of female employment in manufacturing has taken place outside the regular factory workforce and mainly in the form of temporary and contract workers. This trend may not be welfare promoting for the females if it is not only a reflection of the low absorptive capacity of the large-scale industrial sector but also indicates a deliberate policy by employers to exploit females as a cheaper and a more pliable form of labour.

Earlier studies indicate that a majority of female workers are in a severely disadvantaged position in the labour market because they are concentrated in a few occupations and industrial groups and this depresses their wages. Due to low skill and less mobility their participation in the modern, highly productive, sector of the economy has not increased during the course of liberalization.

In this study, the emphasis is on labour market issues, responsibilities within households, decision making, violence and mental health. In order to fully understand the impact of these adjustment policies on female employment, it is necessary to have an analysis of existing inequalities in factor endowment, existing employment status, access to credit and markets, control of income

and assets and the physical and mental health of the population. Based on a survey of the workers in formal and informal industrial establishments producing exportable products, we examine these issues below.

Export Orientation and Employment in the Manufacturing Sector

In this section we briefly analyse the trend in major exports and employment in the manufacturing sector, particularly in the export oriented industries of Pakistan. The composition of exports as a percentage of manufactured exports has not changed significantly over time. Exports are concentrated in a few products like textiles, readymade garments, leather goods and carpets and rugs. Textiles and related products, which primarily depend on the cotton crop, readymade garments, and sports goods maintained their share in total exports equalling 31.36 per cent in 1986-87 and 34.12 per cent in 1995-96. The share of surgical instruments declined marginally from 2.89 per cent in 1986-87 to 2.07 per cent in 1997-98. Although the share of petroleum products has decreased from 1.33 per cent in 1986-87 to 0.58 per cent in 1995-96, on average, the share has declined drastically from 5.90 per cent in 1980-87 (the pre-adjustment period) to 1.42 per cent in 1986-98 (the post-adjustment period). Textile manufactures, as mentioned earlier, have the highest share in total exports and in manufactured exports and this share has increased significantly over time. The share of readymade garments, sports goods, as well as chemicals and drugs increased from 12.2 per cent to 15.5 per cent, 1.5 per cent to 2.9 per cent, and 0.35 per cent to 0.41 per cent, respectively, between 1986-87 and 1995-96. The share of fish preparations, carpets and rugs and leather goods declined as these products lost competitiveness in the world market.

The share of export-oriented industries in employment has changed over time. While the employment share of food, textile and manufactures decreased, the employment share of the remaining products listed in Table 1, increased. According to Kemal (1994), sports and ginning factories, handicrafts, and apparel industries employed very few skilled professional workers perhaps

as a result of rising capital intensity in these sectors in the post adjustment period.

Gender patterns in manufacturing employment

This section explores the gender pattern of employment in the manufacturing industries for the period 1987 and 1992 (latest available).3 Table 1 shows that manufacturing industries employed very low percentages of female workers in 1987 and 1992, respectively. The highest percentage of female workers is employed in chemicals, petroleum and plastic products industries (8.24 per cent and 12.96 per cent of total workers, respectively) and the lowest is in non-metallic mineral products, (0.07 per cent and 0.87 per cent of total workers during the respective periods). Table 1 shows the structure of employment and growth in the manufacturing industries for the years 1987 and 1992. The highest percentage of male workers is also found in textile industries followed by food, beverages, chemicals, and fabricated metal products. The highest percentage of female workers is in textiles (45.8 per cent) followed by chemicals (41.35 per cent). Over the period 1987 to 1992 the share of female workers in textiles has declined to 32.4 per cent and their share in the chemicals industry increased to 47.61 per cent. Wood and wood products accounted for the lowest percentage of females (0.03 per cent in 1987 and 0.05 per cent in 1992). The reasons for changes in employment structure may include a change in the demand for high skilled workforce after liberalization, initially a high concentration of females in a few sectors, and a rise in informal employment activity in the post adjustment period.

Data from the Enquiry on Labour Welfare covers only permanent workers in formal employment, while a significant proportion of female workers in the manufacturing sector, are employed as casual/contract informal workers. For example, Hafeez (1989) found that of the total female workers, 55 per cent were employed as temporary, casual and seasonal workers. Only 1.8 per cent females were found to be permanent workers against 53.32 per

3Due to non-availability of gender breakdown at the disaggregated level in the manufacturing sector, we analyse the gender patterns of employment in major industries at an aggregate level. Data source is Enquiry on Labour Welfare.


Structure of Employment in Manufacturing Industries


Source: Enquiry on Labour Welfare, 1987, 1992.

cent male workers. Furthermore, according to Khan (1986), a large number of females are not employed in factories, but they work at home to produce goods or components on a contract basis for large-scale manufacturing industries.

This discussion reveals that the formal sector employment data does not cover female employment adequately. While female employment patterns shows some changes, the marginal increase in diversification could be due to a rise in female education and skill. Another reason could be the replacement of male workers by low paid female workers as a way of reducing the cost of production. Keeping these trends in view, we discuss our survey-based findings below.

Survey Methodology and Coverage

We have discussed in the last section that the share of textiles, sports, surgical and fishery products in the total manufactured exports of Pakistan changed during the pre to post adjustment

period. The share of garments and textiles increased from 54.63 per cent in 1986-87 to 58 per cent in 1997-98. The share of fisheries, sports goods and surgical instruments was 5.79 per cent, 3.0 per cent, and 2.87 per cent in 1986-87, respectively, The shares of these products in total manufactured exports changed to 2.82 per cent, 6.35 per cent, and 2.07 per cent, respectively in 1997-98, showing a rising share of sports goods and decline in fisheries and surgical goods. The discussion also reveals that the employment structure has changed in the post adjustment era, particularly in the export oriented industrial units. Therefore, we have selected the workers in the export based industrial units for our survey. The respondents for the survey are workers, males and females, both, in the formal and informal industrial units producing textile and garments, surgical goods, sports goods and fisheries.

The questionnaire for the survey is in the form of two modules. The first relates to household level information and the second is for individuals, above the age of 12 (for males/females separately) in the household. The first module collects information on the migration status of the household, the availability and distance to facilities like schooling, hospital, doctor, water and sanitation and quality of the living environment. It also collects information on household composition, schooling, labour market, and sources of income for each member of the household. The second module includes information about labour markets, education, mobility, decision-making, control of income and assets, reproduction and sex, health, domestic and community violence and the psychological profile of the respondents.

The survey was carried out during May-July 2000 in three major industrial cities of Pakistan, i.e. Faisalabad, Sialkot and Karachi. Faisalabad is the main producer of textiles and clothing. Sialkot is a major producer of sports goods and surgical goods, and Karachi is a major port of Pakistan and an exporter of readymade garments and fisheries. All these products face various forms of rules and regulatory frameworks after the Uruguay Round. The textile goods industry is facing the scenario of elimination of MFA and imposition of environmental standards, the sports goods industry faced a problem related to labour standards, particularly the child labour issue, and surgical goods and agriculture are facing regulations regarding quality standards.

In total 298 workers of 250 households, belonging to various

formal and informal industrial establishments, were interviewed. The distribution of establishments covered by product groups is given in Table 2. Workers from 31 industrial establishments, 10 from Faisalabad, 10 from Sialkot and 11 from Karachi, were interviewed. The sample included 94, 90, 107 workers from Faisalabad, Sialkot, and Karachi, respectively. However, the sample of only three cities of two provinces may not be representative for the entire population. But for a gender-based analysis the selected areas provide a good representation of the female involvement in export based industries.


Distribution of Establishments by Product Group


Socioeconomic Profile

Over time the ratio of male to female population has decreased from 110.4 in 1981 to 108.1 in 1998, but there is no significant improvement in the quality of life for females, and in accounting of their contribution to economic development. Female labour force participation rates, the literacy rate, and their access to credit and health facilities are also very low. This results in lower productivity and limited choice of occupations for women. In this section we discuss the socio-economic profile of the factory workers and their household members in the cities of Faisalabad, Sialkot and Karachi.

Demographic indicators

Economic slow down in the post adjustment period has affected socio-demographic factors like the marriage rate, fertility decisions, family planning, health, education, and labour force participation. For example, recent data show a decline in fertility rates — which could be the result of rising poverty in Pakistan — in the post adjustment period. In addition, rising female education and labour

force participation, which increases the cost of raising children, are important determinants of decline in fertility in the country (see Siddiqui 1996). A threefold increase in the use of most effective contraceptive methods that provided longer protection against pregnancy was also indicative of a changed approach towards the reproductive process.


Age Distribution of Population of Households


N = Total number of observations

Table 3 reports the age distribution of the sample population. The table shows that 78.8 per cent of males and 76.9 per cent of the female population is in the economically active age group, which is higher than the national average of 66.5 per cent for males and 66.3 per cent for females. Similarly, the proportion of female population in the reproductive age group is 74.9 per cent, higher than the national average of 62.3 per cent. The city-wise

distribution of population shows that old age dependency is higher in Faisalabad whereas young age dependency is higher in Sialkot and Karachi.

Furthermore, about 31 per cent females and 34 per cent males, above the age of 10, are unmarried in our sample. The preferred age at marriage of the unmarried respondents is 20-22 years for females and above 25 years for males whereas for the currently married the age at marriage was between 15-18 years. This is in conformity with the view that the age at marriage has increased in Pakistan.

The responses to the questions about health and reproductive health indicate that the population has access to health services like hospital, private doctors and others. An interesting finding of the survey is that more than 90 per cent of the females who reported falling sick — a higher fraction than of males who fell sick (85 per cent) — during the last six months, reported consulting a private doctor. The reason is that the majority of these females are working as temporary/piece rate workers, so that the opportunity cost of a day off is very high for them. They consult private doctors to save time and to get better treatment so that they can go back to work.

About a third of married males and females report that they use family planning methods. However, the non-availability, lack of information and disagreement of husbands and in-laws, particularly mother-in-law, are important factors determining the use of family planning methods. This shows that both demand and supply factors, affect the efforts to control population.


The role of education in raising productivity and the living standard of the population is well established. In Pakistan, the literacy rate has increased over time from 35.1 per cent and 16.0 per cent in 1981 to 56.5 and 32.6 per cent in 1998, for males and females, respectively. This shows that the literacy rate, particularly for females, is considerably lower than the literacy rate in most developing countries. The reason could be that there are fewer educational institutions and women have less access to them. If so, then any decline in public expenditure affecting the supply of social sector services will have a disproportionately higher effect on females, who are the main users of public schools.


Educational Attainment of the Sample (percentage)


More than 80 per cent males and 74 per cent females in our sample are literate. This percentage is considerably higher than the national average (see Table 4). The reasons could be that our respondents are from urban areas where educational institutions are available to more than 90 per cent within two kilometres of their residence.

The survey finds that children leave school mainly for economic reasons, as 23 per cent males and 10.8 per cent females left school because they ‘started working’ and 15.6 per cent males and 21.2 per cent females left school because it was ‘expensive’.4 Interestingly, in response to the question ‘who will be withdrawn from school’ in case of financial hardships, in general the response was as expected — that it will be a female child, but a significant proportion of respondents reported that schooling for the female child would be continued. The reasons for wanting the girl to continue were as follows:

4Quality could be another reason for leaving school but a small fraction of population, i.e., 1.7 percent, reported leaving school because the education provided is not useful or up to standard.

  1. At times of economic hardship a boy can work as a labourer on the street and earn some money;
  2. A girl cannot work on the streets. If she needs to work, an educated girl can find a job in a decent and protective environment.

Considering the importance of education in raising productivity, the quality of life of the population and inter-generational socio-economic mobility, it is critical to identify the factors that affect the demand for schooling. To examine this issue, we have tested if there is a causal relationship between the current demand for basic education and the socio-economic characteristics of families. A probit model is applied to examine this causal relationship. The model is specified as:

S = F (x)

where S = 1 if a child in the age group 6–10 reports going to school and S = 0 if the child is not going to school.5 X is a set of the following socio-economic indicators:

  1. Distance to school,
  2. Household income,
  3. Asset ownership (own a house or not),
  4. Mother’s education,
  5. Mother’s work status, and
  6. Household size.

Distance to school is included as a price variable because an increase in distance adds transport cost and other cultural factors (particularly for girls) to the cost of schooling. Thus, an increase in distance is expected to have a negative effect on the probability of a child going to school. Income and asset ownership are expected to have a positive effect on the demand for schooling and an increase in household size is expected to have a negative effect on the demand for schooling. A mother’s education is expected to have a positive effect on the child’s schooling. However, her work status may have a positive effect if the income effect dominates and

5We are restricting the model to basic education. The sample of children is in the age-group 6–10 in order to eliminate the effects of availability of market work and over-age students.

negative (particularly for girls) if the substitution effect dominates. In order to capture the effect of differences in socio-cultural factors across cities, dummy variables for the cities of Faisalabad and Sialkot are also included.

The results of the estimated probit model are reported in Table 5. Since all the boys in the selected age group reported going to school the equation is estimated for the girl child only. The results show that increase in distance has a negative effect on demand for schooling for girls but the effect is not statistically significant. Similarly the impact of mothers’ work and education is positive but not statistically significant. Income is the primary factor determining the demand for girls’ education. The effect of city dummy is significant for Sialkot implying that the probability of a girl going to school is higher if she is from Sialkot. The main reason for this result could be that in Sialkot a few employers provide either a schooling facility for the children of factory workers or give cash benefits to those workers who send their children to school.6 Furthermore, recent efforts to eliminate child labour may have discouraged children working in the sports goods industry and, as a result, parents are sending their children to school.


Demand for Schooling


6The strict implementation of laws against child labour may have motivated employers to provide for child schooling.

The inherent gender bias is obvious from the fact that all the boys in the schoolgoing age attend school but all the girls do not. The girls’ schooling depends significantly on economic factors like family income. Thus, a decline in economic activity even if the Social Action Plan (SAP) is implemented to ensure supply of funds for social sectors, is expected to have a negative effect on girls’ education.7

Labour market participation

Adjustment policies have affected the labour market also. The macro level data show that the refined activity rate declined in all provinces except in Punjab, in the post adjustment period (see Table 6). The female labour force participation rate increased whereas the male activity rate declined. The rise in female refined activity rate indicates feminization of labour in Pakistan in the post adjustment period in almost all the provinces. However, the rise in female activity rate has been accompanied by a more than proportionate rise in female unemployment rate, equalling 18 per cent. The rise in unemployment rate among urban females is about 35 per cent.

The labour force participation rates (LFPR) computed from survey data differ from the national average (see Table 7). The LFPR is higher for females as compared to the national average of 13.9 for females, but lower for males as compared to national average of 70.7 per cent. Table 7 also shows significant differences in employment status in the three cities. In Karachi, a higher fraction of males and females are regular employees but in Faisalabad and Sialkot the majority of the males are regular employees and the majority of females are casual/piece rate temporary workers.

Furthermore, the number of days and hours worked also show differences between males and females. About 20.4 per cent females and 14 per cent males reported working more then 26 days during the last month and 21 per cent females and 15 per cent males reported working more then 10 hours per day during the last week. This shows relatively longer working months and days for a larger fraction of working females (see Table 7).

7The implementation of SAP is under criticism for corruption and below target achievements. In addition, the poor quality of public schools also discourages parents to send their children to public schools.


Refined Activity Rate: Pakistan and Provinces


Source: Government of Pakistan, Labour Force Survey, (1987/88, 1997/98).

Another labour market indicator of gender differences is the higher concentration of female workers in a few industries and occupations, which keeps their wages low despite the increase in demand for female labour. For example, according to the Labour Force Survey of 1997-98, at the national level, about 95.45 per cent of the employed females were in agriculture, services (domestic) and in manufacturing (small-scale) industries. Similarly, about 90 per cent of employed females are concentrated in occupations like farming, production and community services. This shows that industrial and occupational choices for females are limited. This has important implications for changing labour markets in the post adjustment period.

In our sample also, female production activity is mainly concentrated in stitching and related activities, whereas male respondents show occupational diversity. The comparison of earnings reveals that, in general, females earn less then males but


Labour Force Participation Rate and Employment Status of Population (for population aged 10 years and above (percentage))


Note: Employer-a: Employer with 10 or less workers; Employer-b: Employer with greater then 10 workers.

the difference is not very high. However, despite similar monthly earnings, significant gender based differences exist in the labour market. For example, we have seen earlier that a higher fraction of females work more (in terms of day and hours per day) to earn the same amount of money. In addition, gender differences prevail in terms of hiring practices. A larger fraction of females work as casual or temporary workers, fringe benefits are almost non-existent for them and a lower fraction of females enjoy medical benefits. Furthermore, most females changed jobs because they were ‘thrown out’ from the previous job, paid less wages and worked higher number of days in a month. The difference in payment for overtime work also reveals significant gender-based differences. The majority of the females report that they are either paid less or same as men as payment for work in regular hours. About 51.8 per cent males

and 21 per cent females reported receiving more then double the rate for overtime. In Faisalabad, the majority of the males are either paid double or more than double for overtime work. Thus, the same monthly income is not a sufficient indicator for the absence of gender discrimination in the labour market. Furthermore, in the post liberalization scenario when the employment cost is reduced by increasing the capital intensity of the production process or by changing the nature of the job contract, i.e., to temporary jobs with no fringe benefits, the female employment rate may increase. However, as the rise in poverty results in a feminization of the labour force and concentration of females in few industries and few occupations, the rise in demand for female labour will not result in higher rewards for them. Thus, the adverse impact of changes in the labour market in the post adjustment period is expected to be disproportionately greater for females.

Determinants of earnings

Is there a causal relationship between various socio-economic indicators and the earnings of workers? To examine this relationship we estimate an earnings function for all workers and for males and females separately. The function is specified as:

Ln (y) = f (x)

where Ln (y) = natural log of individuals’ earnings
X = set of socio-economic indicators

The socio-economic indicators include standard human capital variables, i.e., education and age (a proxy for experience). These variables are expected to have a positive effect on earnings. Marital status is also included in the equation as a social indicator. Employment status is included to capture the effect of changes in the labour market. The variable is defined as ‘0’ if the worker is either a regular employee or employer and it equals ‘1’ if the respondent is a temporary (casual/piece rate/contract) worker. In order to capture the city-specific and industry-specific factors, dummy variables for cities and industries are added.

The results of the estimated earning function are reported in Table 8. The results show that human capital contributes positively to earnings. However, the rate of return to schooling for females is lower than for males. Marital status has a positive effect on earnings

but it is statistically significant only for males. Employment status does not affect earnings of male workers significantly, but the earnings of temporary female workers are significantly lower. The industry specific factors do not affect earnings significantly. However, city-specific factors seem to play an important role. In Faisalabad and Sialkot, females’ earnings are significantly lower as compared to Karachi. Thus, we can say that human capital and city specific differences in the labour market affect earnings of the workers significantly. Furthermore, the informalization of the labour market is expected to have significant negative impact on female earnings.


Earning Functions


Autonomy and Empowerment

The issue of female autonomy and empowerment is critical in determining the gender relations within a household and in a

society. While gender differences in human capital, particularly education, involvement in labour market activities and income are often used as indicators of gender discrimination, there is a need to elaborate on gender differences within the household to understand the issue of female autonomy and empowerment. The changing macro economic situation is expected to have an impact on intra household gender relations also. Based on survey data, this section explores various issues like the control of assets and income within household, the burden of work when females are actively involved in market production, mobility, and decision-making. All these issues, discussed below, are closely linked with female autonomy and empowerment.

Control over assets and income

The previous section indicates that females are actively involved in market production and they contribute to household income. Interestingly, contrary to popular belief, our data show that whereas all the males contribute to household income, about 2.1 per cent, 5.8 per cent and 5.1 per cent females in Faisalabad, Sialkot and Karachi, respectively, do not do so. The reason may be that our sample is based on workers in industrial units in the low-income group and some women have reported working for their dowry because of the increased demand for it in society. Our data also show that whereas more than 90 per cent males and females contribute to household income, the mother/mother-in-law controls this income. A small fraction of working females report that they are able to keep their income. Does this mean that females have more control over assets as well as income?

In most empirical studies, women’s lack of access and control over physical and financial assets are blamed for their lower involvement in paid productive work, their lower status and autonomy in society. This section examines the ownership and control of various forms of assets. i.e., land, house, vehicle, consumer durable, jewellery and financial assets (bank account).

House ownership shows an interesting pattern. In Faisalabad 40 per cent of the females report owning a house whereas only 9.6 per cent females in Sialkot and 18.6 per cent females in Karachi own a house. However, 62.9 per cent, 95.2 per cent and 48.9 per cent males report owning a house in Faisalabad, Sialkot and Karachi,

respectively. For vehicles and other consumer durables, it is difficult to determine the ownership as in most cases they are owned by the head of household, irrespective of who bought them, and used by everyone in the household. Surprisingly, about 67 per cent females report that they do not own jewellery. Out of the remaining 33 per cent who own jewellery, only 10.7 per cent report its value as being greater than Rs. 20,000. The reasons for no jewellery or reported low value could be either poverty and/or mis-reporting as women were reluctant to talk about the value of their jewellery. Similarly, more then 90 per cent males and females report that they do not have a bank account in their own name due to insufficient savings.

A related issue is the buying and selling of assets. Table 9 shows that the majority of males (51.7 per cent) and females (65.4 per cent) report that they need to get permission to buy assets. Females, in general, need permission either from the spouse (23.3 per cent) or from the mother/mother-in-law (35.8 per cent) to buy an asset. A higher fraction of males (90 per cent) report that they can buy everything on their own. It means that even if males need permission ‘to buy/sell’ it is not binding. Furthermore, the majority of the females (44.3 per cent) report that they can buy their own clothes without permission. Similarly women cannot sell anything without prior approval. About 15 per cent females, the largest fraction, can sell jewellery on their own. As with income, control is in the hands of mothers/mothers-in-law. It is interesting to note that change in status, i.e., being mother/mother-in-law within the household, can give more autonomy to females. However, this control is exercised over females only. For example, females have to get permission from the mother/mother-in-law, in the household to buy or sell assets. But this constraint is not effective for males as the majority of them can buy/sell on their own. Thus, it is not just the gender difference which limits female autonomy, age and hierarchy in relationship within a household are also important.

Similarly though a higher fraction of males is contributing their entire income to household income, females control the income of the household. Across cities this proportion is about 70 per cent and it is highest in Faisalabad equalling 62 per cent. Interestingly, a higher percentage of males report that the wife has control of income. This is confirmed by females as a very small percentage of females reports that husbands control the income.

Our study also finds that, as expected, after entering into paid


Control over Assets


work the responsibilities of the women increase. However, it has also resulted in higher female mobility, higher self esteem and an increased role for females in decision-making (see Table 10). Thus, paid/market work seems to have a positive effect on female autonomy and empowerment.


Change in Females after Work



Empirical studies report that restricted female mobility is the main characteristic of Pakistani society. However, as mentioned in the last section, involvement in work activity brings higher self-esteem, mobility and decision-making among females. The issue of mobility is multidimensional. It is not just whether females can or cannot go anywhere. The issues are: can they go alone? If not; can they go with someone? Where can they go alone? How long can they stay out? Whose permission is needed to go out? There is no unique answer for each of these questions. All these questions were asked




from male and female respondents. Table 18 shows that the majority of males and females reported that there is less freedom for girls in their households. The main reason is tradition/custom. Another reason is the fear of harassment. Though the majority of females reported they have to get permission to go anywhere, they face fewer restrictions for going to an educational institution or to the workplace. The data show that a higher number of females can go to school (84.9 per cent), work (89.35 per cent) and market (66.7 per cent) alone. However, a very small fraction can go to the

cinema or parental house alone. The response varies across cities: a lower fraction of females are allowed to go to the restaurant in Faisalabad relative to Sialkot and Karachi. Furthermore, a majority of respondents report that they can go anywhere with other family members or friends. Whereas the time schedule for males is liberal, the majority of the females are not allowed to stay out after dark.


How far are females involved in household decision-making? What are the major areas of female decision-making? Where is the female’s decision final? These are important issues. The extent of female involvement in decision-making is another indicator of female autonomy and empowerment within the household and in the community.

Table 12 shows that more than 80 per cent males and females report participating in household decision-making. However, the extent of female decision-making is limited. Among the females a higher fraction (48.6 per cent) can decide about household provision and the decision of 23.9 per cent females is final. Similarly, the education of a child is another major area where the decision of females is final. However, the role of females declines when decisions regarding major expenses, except for buying/selling of jewellery, are taken.

The possibility of giving females more autonomy is limited as the majority of males (64.7 per cent) think that there is no need to give more decision-making power to females at the household level (see Table 12). Furthermore, 78.2 per cent males feel that there is no need to give more decision-making power to females in community affairs mainly because females have less knowledge. Surprisingly, a high fraction of females, i.e., 46.5 per cent, supports this view.

Thus, we can conclude that major areas of females’ decision-making are providing for the daily needs of the family, the education of children, and their marriages. However, females feel that they should have more decision-making power within and outside the households but males feel that their decision-making should be restricted to the household. This discussion shows that decision-making is limited among females and there is little support even within the female community for changing this situation.




Violence and Mental Well-being

The psychological and mental abuse of females comes in many guises and pervades all areas of their lives. In general, cultural practices and age–old traditions deny the majority of females the right to make decisions about their lives. There has been no systematic analysis of violence against females in Pakistan, except for a few studies (see, for example, Sathar and Kazi 1997). Lower female autonomy along with less protection from society, from law and law-enforcing agencies increases the vulnerability of females, particularly during economic downturns and with rising poverty. Based on secondary data we can see that violence takes many forms and the incidence of violence increases with the rise in poverty. In our survey, we have also collected information on violence faced by the respondents within and outside the household, as well as its causes and the coping mechanisms of those who face violence. Poverty, violence, discrimination, lack of autonomy and rising workloads affects the mental health of the population also (see Sonpar and Kapur, this volume). The issues of mental distress and mental well-being of the respondent males and females are also discussed in this section.


We posed questions about whether respondents have faced abuse within or outside the family, what the reasons for this were and what kind of coping mechanisms they had turned to. Whereas a higher number of females (71.7 per cent) report that they have heard about violence in the neighbourhood, the number of males responding positively (52.1 per cent) is low (see Table 13). According to the male respondents the main reasons that trigger off violence include property disputes, economic reasons, and neglect of household duties. Females report economic reasons, also if they refuse to obey the husband, or answer back, or neglect their duties as major reasons for domestic violence. This shows that economic factors play an important role in the incidence of violence, particularly domestic violence, in a society.

In the domestic sphere, females face verbal and physical abuse whereas males report receiving threats. However, a majority of the


Causes of Violence


respondents report that they do not face violence outside the domestic sphere. Whereas the male respondents reported that they have faced/heard about communal and political violence, females report fights over property and harassment of girls as the major forms of violence they have faced outside the home (see Table 14).


Type of Abuse


Interestingly, the majority of respondents report that taking up a job by females has not resulted in higher tension in the family. Only 10 per cent of females and 6 per cent of males report rising tension after females start working. For most of the females in Faisalabad it is social pressure, for females in Sialkot family pressure, that leads to tension within the family. For example, a brother disapproves of his sisters’ working, so he stops going to school.

The coping mechanism for domestic violence is reported to be either, ‘no response’ or/and ‘discussion’ (see Table 15). Reporting violence to the police or parents and friends is not seen as an appropriate way of dealing with it. Most of the respondents, particularly females, reported that they do not speak about violence because they are embarrassed to tell any one.


Violence Coping Strategies


Mental distress (GHQ)8

The rise in tension within the family due to internal or external socio-economic factors, can affect the mental health of individuals. A number of socio-economic factors may cause mental distress. Here, we discuss mental distress among individuals based on their work status, age, education and income. Surprisingly, our survey reveals that mental distress among the non-working respondents is higher than among those who are working. However, among the working population mental distress is higher among females (see Graph 1). Lower mental distress among working females, as compared to non-working females, supports our earlier finding that involvement in market work raises self-esteem, mobility and awareness among females.

Graph 2 shows that mental distress is higher among females in the younger age groups. It is highest among the age group of 25-29 years. However, mental distress among males is significantly higher in the age-group of 50 and above. There seems to be an inverse relationship between mental distress of males and females below the

8A 12-item version of original 60-item general health questionnaire (GHQ) was included in the individual questionnaire to determine psychiatric illness. The GHQ is widely used in community surveys. Each item has four possible responses. However, each item is coded as 0,0,1,1 for the 4-point scale on each item. The individual’s score is the sum of the scores on each item. The minimum score of an individual could be 0 and the maximum score could be 12. A higher score indicates higher mental distress.

age of 50 and above the age of 50. The reason could be socio-cultural. We have seen earlier that female mobility and status in a traditional society are strongly related to the age of women. Furthermore, decision-making within the household is the responsibility of the mother/mother-in-law. Thus, lower stress among females in the higher age groups could be a result of more mobility and higher involvement in decision-making.

The mental distress among males and females seems to vary with the level of education also. Surprisingly, Graph 3 shows that mental distress is higher among males at the lower education level, i.e., less than primary education, but it is higher among females with an education level greater or equal to secondary education. However, the gender differences in mental distress by education level do not seem to be very high.

Another indicator of the socio-economic set up is income level. We can see from Graph 4 that mental distress declines visibly among males as income levels go up (except for the respondents in the income group of Rs. 6000–7000). Among females, mental distress is lower at the higher income level but the difference does not seem to be very significant. Thus, work related decline in mental distress among males and females may not be strictly related to income; other factors mentioned above are also important.


Graph 1 Work Status and Mental Distress


Graph 2 Age and Mental Distress


Graph 3 Education and Mental Distress


Graph 4 Income and Mental Distress

Mental well-being (SUBI)9

A closely related concept to that of psychological health is the ‘mental well being’ of the respondents. A higher code for the SUBI means lower mental well being. Graph 5 confirms our earlier finding that female involvement in market work has a positive effect on women. Surprisingly, the mental well being of non-working females is worse than that of non-working males.

Graph 6 shows, in general, poor mental well being of females as compared to males below the age of 50. Among males the mental well being improves as the age goes up but after the age of 30, the score on SUBI rises showing a decline in mental well being. However, among females above the age of 39, mental health is better. This confirms our earlier finding that the psychological condition of females improves with age and vice versa for males.

Surprisingly, education level does not give a clear picture.

9A set of 9-items was included in the questionnaire. The questions are drawn from the Subjective Well Being Inventory (SUBI). The three responses on each item are scored 1, 2, 3 with lower scores indicating greater well being. The scores on each item are summed to arrive at the total score.

Graph 7 shows poorer mental well-being of males with lower education levels but as the education level increases the mental well-being improves. However, for females, higher education seems to result in lower mental well-being.

Economic security, in terms of higher income, is expected to improve mental well being. Graph 8 shows that mental well-being


Graph 5 Work Status and Mental Well-being


Graph 6 Age and Mental Well-being

improves significantly as income increases. Thus, any decline in economic indicators is expected to have a strong effect on the mental well-being of the population.


Graph 7 Education and Mental Well-being


Graph 8 Income and Mental Well-being

In order to determine the extent and causal relationship between socio-economic factors and mental well-being we have estimated the following regression model.

SUBI = f (X)

The dependent variable (SUBI) is defined as the sum of scores for each individual. A number of socio-economic indicators (X) are included in the model. For example education, marital status, work status, income, quality of infrastructure, migration of the family and city and industry dummy variables were included in the analysis. Income variables are expected to have a negative impact as the mental well-being is expected to improve with income. This variable also captures the effect of the work status. The migration variable is defined as the length of stay. As the length of stay increases, mental well-being is expected to improve. Thus, the coefficient is expected to have a negative coefficient. As the quality of house improves mental well-being is expected to improve. Since the quality of house is defined as the ‘house grading’, increasing as the grading declines, the sign of the coefficient is expected to be positive. The house quality may capture the wealth effect and the effect of the living environment. Lower assets and a poor living environment are expected to have a negative effect on mental well-being.


Factors Affecting Mental Well-Being


The results of the estimated equation are reported in Table 16. In the final selection, only income, migration, and grading of the house are retained. These results are according to expectations. For

males and females, both, the effect of income and migration is negative and of the living environment positive. For both males and females as income goes up the mental well being improves and the effect is statistically significant. The increase in the duration of stay also increases mental well being, however, the effect is not statistically significant for males. The living environment seems to be very important for both males and females as the coefficient has the expected sign and is statistically significant.

Conclusions and Policy Recommendations

Pakistan, like other developing countries, implemented the stabilization and structural adjustment programme to overcome fiscal and balance of payment problems and to correct the structural imbalances in the economy. It is claimed, in a number of empirical studies, that these programmes have increased economic hardships in most developing countries, including Pakistan, at least in the short run because of rising unemployment, low investment, decline in real wages, capital flight and rise in inequality in the post adjustment period. All this, in turn, has resulted in deterioration in living conditions of the poor, at least in the short run.

Analysis based on secondary and survey data suggests that during the last fifteen years the ratio of female to male population and the labour force participation rate has increased. However, no significant achievements have been made to improve the quality of life, particularly of females, and to improve their contribution to economic development. Based on this study, we can draw some conclusions and make some policy suggestions as detailed below.


The gender disparities in socio-economic indicators have important implications for the role of females in economic development and on the material and mental well being of the population. The survey-based evidence shows that the female literacy rate, as expected, is lower than the male literacy rate. The cost of education and family income are major determinants of girls’ education. Furthermore, the cost of education, the decision to work, and ‘no interest’ are important reasons for leaving school.

The secondary data shows that in the post adjustment period,

female labour force participation rate increased supporting the view of the feminization of labour in Pakistan. But there has also been a sharp rise in the female unemployment rate relative to the male unemployment rate. Furthermore, females are working either as farm workers, production workers or community workers in agriculture, in manufacturing (small-scale) industries and in the services (domestic) sector. This affects their wages and their ability to change occupation. The majority of the female respondents in our survey are from the formal and informal industrial units producing export-oriented goods, viz., textiles and garments, sports goods, surgical goods and fisheries. These females are concentrated in the stitching and related jobs in the garments and sports goods industries. Similarly, in surgical and fisheries they are employed in low paying jobs. The distribution of work months and working hours shows that, on average, females’ workdays were higher and the work hours are longer. This suggests that the load of market work is higher among females. Unlike the majority of the males, most working females are not eligible for fringe benefits as they are not permanent workers. Thus, the impact of adjustment policies on the informalization of the labour market has important gender-based implications and the negative fallout is expected to be higher for females. The survey of these establishments reveals that the provision of transport facilities plays an important role in females’ decision to work. It also provides the employers an assured supply of labour services when they need it. Structural differences exist between labour markets in various cities. These differences should be taken into account while formulating industrial and employment policies.

The extent of female involvement in the decision-making process determines female autonomy and empowerment within the household and in the community. The finding of the survey is that the majority of males and females participate in household decision-making. However, the extent of female decision-making is limited, particularly in the major buying and selling decisions and in community affairs. Among females, a high proportion can take decisions regarding household provision and for the education of children. But the role of females declines when decisions regarding major expenses like buying and selling of assets are to be taken, except for the buying and selling of jewellery. The possibility of giving females more autonomy is limited as the majority of males

think that there is no need to give more decision-making power to females at the household level and in community affairs, mainly because they perceive females to have less knowledge. Surprisingly, a high percentage of females support this view, making it difficult to change the situation. An interesting finding of the study is in terms of the control of income. Our results seem to suggest that hierarchy within the family relationship is an important determinant of control of income. It is not the daughter/daughter-in-law who controls the family income but the mother/mother-in-law who has control over family income.

The results of the study show that mental distress and the mental well being of respondents is closely linked to work status for both males and females. Mental well being is not only linked to economic indicators but it is significantly affected by the living environment.

The discussion in this study supports the argument that the impact of adjustment policies leading to liberalization and change in the labour market has a disproportionately higher negative effect on females. Thus, adjustment policies are not gender blind as claimed by the policy makers.

Policy implications

Based on these conclusions, what policies can improve the gender relationship within and outside households? We propose the following few directions for policy formulation:

Since the cost of education and rising poverty affect the household decision to send children to school, a programme linking schooling with the provision of a food subsidy can be helpful in reducing the direct and opportunity cost of children’s education. There is some evidence that the provision of education facilities for the children of workers by an employer plays an important role in the current demand for education. Thus, employers can be encouraged through fiscal and other incentives to provide schooling and childcare near the workplace of mothers/fathers.

The results show that industrial and occupational choices for females are limited, and can be expanded by increasing female education and by providing extra facilities to female workers like skill training facilities at subsidized rates and transport facilities. This will have important implications keeping in view the increased

informalization and feminization of the labour market in the post adjustment period.

In order to increase female autonomy and empowerment, there is a need to encourage the role of females in decision-making and to discourage the stereotyped thinking among males and females, both, within the household and in society. The media can play an important role in this regard.

The study also shows that economic factors are important determinants of violence. Since in most cases females depend emotionally and financially on the abusers, it becomes critical to know how females experience violence and how to intervene to reduce gender-based violence effectively. A lack of technical competence and resources, cultural stereotypes, negative social attitudes, institutional constraints, and females’ reluctance to disclose violence are the major barriers to controlling domestic violence. Thus, in order to reduce gender-based violence, there is a need to empower females, raise the cost of violence to abusers, provide for the needs of the victim, reach out to abusers and other males, provide effective legal cover and create awareness in the community.

To bring change in the lives of the population, particularly females, it is important to improve the levels of education and other economic indicators. However, this may not be enough. There is a need to bring change in the thinking of males and females both. The media and community participation can play an important role in this regard.

Future directions

Based on these conclusions and the proposed policy framework, we can identify the following future research directions for more in-depth understanding of the issues. The findings of this study indicate that there is a need to conduct an in-depth study of the workers who have lost their jobs as a result of adjustment policies. There is a need not only to formulate gender sensitive policies but also to propose mechanisms for their effective implementation. For this purpose it is important to identify the role of individuals, the community and government to bring about change. However, to be able to suggest broad-based macro policies, much more data is

needed, and hence there is an urgent need to extend this survey-based analysis to other cities and sectors of the economy.


Ali, K. (2000) ‘Structural Adjustment Policies and Women in the Labour Market’, Third World Policy Review, vol. 22, No. 1, pp. 1–21

Ali, M.M. and N. Ahmed (1983) Role of Women in Rice Farming in Pakistan, Social Science Division, Pakistan Agriculture Research Council

Amjad, R. and A.R. Kemal (l997) ‘Macroeconomic Policies and their Impact on Poverty in Pakistan’, The Pakistan Development Review, vol. 36, No. 1, 1997, pp. 39–68

Brown, J.K. (1992) ‘Introduction’, in Sanctions and Sanctuary: Cultural Perspectives on the Beating of Wives, by D.A. Counts and J.K. Brown (eds.) Boulder, Colarado: Westview Press

Freedman, J., and L. Wai (1988) Gender and Development in Barani Areas of Pakistan, Canada

Ghosh J. (1994) ‘Gender Concerns in Macro-Economic Policy’, Economic and Political Weekly, vol. XXIX, No. 18

Government of Pakistan (1987, 1992) Enquiry on Labour Welfare, Islamabad: Ministry of Labour

Government of Pakistan, Labour Force Survey, (various issues), Islamabad: Federal Bureau of Statistics

Hafeez, Sabiha (1989) Women in Industry in Pakistan, Women’s Division, Islamabad: Government of Pakistan

Kemal, A.R. (1994) ‘Structural Adjustment, Employment, Income Distribution and Poverty’, The Pakistan Development Review, Vol. 33, No. 4

Kemal, A.R, F. Bilquees, S. Rauf, R. Siddiqui, F. Hussain and Z.M. Nasir (2000) Macroeconomics of Pakistan’s Economy, Manual prepared for the training programme of the SBP Officers, Islamabad: Pakistan Institute of Development Economics

Khan, N.S. (1986) Women’s Involvement in the Industrial Sector in Punjab, Lahore: Applied Socio Economic Research Centre

Khan, S.A., and F. Bilquees (1976) The Attitudes, Environment and Activities of Rural Economy: A Study of Jhok Sayal, Islamabad: PIDE

Khan, S.R. (ed.) (1999) ‘Gender and Structural Adjustment’, in ‘Do World Bank And IMF Policies Work?’ New York: St. Martin’s Press Inc.

Mahmmood Z. (1998) A Social Strategy for Pakistan. Pakistan Institute Of Development Economics, Research Report 163

Mahmmood Z. (1999) ‘Pakistan: Impact of Economic Liberation on Poverty ’, SDPI Working Paper

Sarwar, M., M.A. Saleem and M.J. Khan (1993) ‘Baseline Study of Punjab, Small
Holders Dairy Development Project Gujranwala’, publication No. 296, Punjab Economic Research Institute

Sather, Z. and S. Kazi (1997) ‘Women’s Autonomy, Livelihood and Fertility: A Case of Rural Punjab’. Islamabad: Pakistan Institute of Development Economics

Shah Nandita, S. Gothoskar, N. Gandhi and A. Chhachhi (1994) et. al. ‘Structural Adjustment, Feminization of Labour Force and Organizational Strategies’, Economic and Political Weekly, Volume. XXIX, No. 18

Siddiqui R. (1996) The Impact of Socio-economic Factor on Fertility Behaviour: A Cross-Country Analysis, The Pakistan Development Review, Vol. 35, No. 2

Sonpar, S. and R. Kapur (2000) ‘Non-conventional Indicators of Gender Disparities under Structural reforms’, Paper presented at Gender Planning Network, workshop at ISST, November 2000

Standing, G. (1989) ‘Global Feminization Through Flexible Labour’, World Development, Vol. 17, No. 1.

Continuity and Change: Women
Workers in Garment and Textile
Industries in Sri Lanka



Macroeconomic Reforms

Sri Lanka’s economy remained largely unchanged for three decades after independence was regained in 1948. Deteriorating terms of trade and slow economic growth led to the introduction of import substitution industries, and eventually, by the 1970s, to a controlled economy. In 1977 a radical shift took place in macroeconomic policies to market liberalization and an open economy, concomitantly with the adoption of a ‘stabilization and structural adjustment package’ promoted by the IMF and World Bank.

The main components of the new policies were:

  1. liberalization of trade and payments
  2. devaluation of the currency
  3. decontrol of imports and prices
  4. reduction in consumer and producer subsidies
  5. reduction in public expenditure and especially curtailing social sector expenditure
  6. export promotion with incentives for local and foreign investors, and
  7. privatization of public enterprises and reduction in the size of the public sector.

The private sector was to be the ‘engine of growth’ and the state was to confine itself to providing infrastructure, such as the

establishment of the Export Development Board and Export Processing Zones (EPZs).

The second phase of structural adjustment programmes introduced in 1989 extended these policies, such as further devaluation of the currency, elimination of consumer and producer subsidies, promotion of private enterprise and privatization of public enterprises and ‘rationalization’ of the public service. The Industrial Promotion Policy and Act of 1989 and 1990 directed the expansion of export-oriented industries to the rural sector outside the EPZs. The Board of Investment was established as a ‘one stop Investment Promotion Centre’ and the whole country was declared an Investment Promotion Zone in 1992. Rural garment factories were established by entrepreneurs and the state organized industrial estates in different parts of the country in the 1990s. These policies are ongoing and public enterprises have been divested and privatized at a more rapid pace in the last decade.

Assessments in international and cross country studies of structural adjustment programmes noted that these programmes have failed to promote sustained development while their social costs have had adverse consequences for the poor. (Harrigan, Mosley and Toye, 1981, UNICEF, 1987). Studies in Sri Lanka have also underscored the failure of the reforms to achieve sustained growth and have pointed to increasing income disparities and the exclusion of the poor from the benefits of economic development. (Lakshman, 1997; Indraratne, 1998).

As women have not been a specific target of macro-economic policies and adequate gender disaggregated data is not available, it is difficult to identify clearly their impact on women. However, international and cross country studies (Commonwealth Secretariat, 1989; Heyzer, 1987; Joekes, 1989; Sparr 1994; Elson, 1998;) and the few Sri Lankan studies available (CENWOR, 1995; Jayaweera, 1998; Ariyabandu et al., 1995), indicate that the quality of employment available to women as a consequence of these macro-economic reforms has deteriorated and that women in low income families have borne a disproportionate share of the burden of ‘adjustment’ to reforms.

Labour force data and studies have indicated that these reforms have had a differential impact on women and men in employment. Female labour force participation rates have increased more rapidly than those of men. Although female unemployment rates have

continued to be consistently double those of men, overall unemployment rates have declined. Apart from this, labour force data indicates that there has been a trend towards the casualization of labour, with the percentage of women in regular employment in the formal sector declining from 80 per cent to around 50 per cent in the 1990s. While men have moved to self employment as “own account workers”, the percentage of women unpaid family workers which had declined to 6.5 per cent in 1981, increased to 20 per cent to 25 per cent in the 1980s and to 30 per cent by 1998 (Labour Force Surveys).

These trends are reflected in changes in employment in the different sectors. Examples are the use of women’s unpaid family labour to meet the rise in production costs caused by the removal of producer subsidies, such as fertilizer, in the agricultural sector; the reduction in the public service through ‘golden handshakes’ and the closure or privatization of public enterprises in the 1990s; and the rapid increase in the temporary migration of women in low income families to meet the demand for domestic labour in West, East and South East Asia and Western Europe.

The most visible impact of macroeconomic reforms has been the export orientation in the industry sector. The share of the manufacturing sector in exports increased from 5 per cent in 1979 to 75 per cent in 1995 as export-oriented industries expanded within the three Export Processing Zones established in Katunayake in 1978, in Biyagama in 1985 and in Koggala in 1990, and outside the zones. The lead industry was and is garment manufacture in which 80 per cent of the labour force in the EPZs and 90 per cent in the rural garment factories have been women, chiefly young women between the ages of 18 and 30. The comparative advantage of low cost but literate female labour was used to attract foreign investment, resulting in the incorporation of women in the international division of labour. Subcontracting operations proliferated with the efforts of transnational and local entrepreneurs to reduce production costs and have access to a flexible peripheral labour market outside the ambit of labour legislation (Jayaweera and Sanmugam, 2000).

One of the immediate consequences of market liberalization was the collapse of small local industries which had been protected earlier and could not compete now with cheaper and better quality imports. The worst affected was the handloom industry which had been ‘feminized’ in the 1960s and 1970s and had depended on

imported and subsidized imputs. The result was the closure of many centres and the loss of livelihood of around 40,000 women (Atukorale, 1986). Other small industries such as coir, pottery, carpentry and brassware survived because they used local materials. Attempts at reviving small local industries have failed as macro-economic polices have been biased towards largescale export-oriented industries. While self employment was encouraged in the open economy, it was chiefly women and men with family resources and initiative who were able to utilize the incentives offered and become successful entrepreneurs in the domestic and export markets. The majority of women without resources were engaged in often unviable self employment in the informal sector.

While these developments in the employment scene changed the life chances of many women, the social costs of the sharp shift in policies, such as the continuing escalation in the costs of living and consequent decline in real incomes, the change of the food ration into food stamps that were not indexed to costs of living, and the increase in income disparities and poverty in the 1980s created additional burdens for women in low income families whose labour inputs make a crucial contribution to family survival and maintenance. Women workers in factories, overseas domestic service and plantations have helped to sustain the economy but have not received adequate economic and social rewards. Overall, these economic and social trends are the consequences of a process of globalization and one way adjustment in which economically developing countries are compelled to ‘adjust’ to global pressures.

Background of the Study

This study, which is part of a larger South Asian study, is focused on the impact of changes in the garment and textile industries on women’s lives. The garment industry was in the vanguard of the rapid expansion of export-oriented industries. The textile industry, a traditional avocation of women, underwent vicissitudes of fortune during the last two decades.

There were two phases in the study (CENWOR, 2001). Phase 1 examined the employment experiences of women in six groups of women garment and textile workers and a smaller number of men

workers (370 women and 75 men) who were affected by changes in macroeconomic policies—

  1. Export Processing Zone workers
  2. rural garment factory workers
  3. home-based garment workers in the informal sector
  4. textile factory workers
  5. handloom workers, and
  6. retrenched textile workers.

The study was confined chiefly to a non affluent socio-economic group but a few entrepreneurs from economically advantaged families were also interviewed from a comparative perspective. Phase 2 explored changes in gender roles and relations in the families of a sub sample of 40 per cent of the households of the women participants in Phase 1 — 140 women, 150 other women and 118 men in their families (408 in all). Numbers of women and men making up purposive samples were interviewed in their houses or lodgings and a few case studies were undertaken. While it is not possible to relate all the experiences of these women directly to changes in macroeconomic policies, the study provided useful insights into the economic and gender specific factors that contributed to continuity or change in the lives of women and their families.

The employment experiences of workers are discussed in the next section of the chapter.

Garment Factory Workers

These factory workers were employed in garment factories in the three Export Processing Zones, two located in the periphery of the metropolitan district, Colombo, and one in the Southern Province not far from the provincial capital, and in three rural garment factories, in the Kalutara district, in the Kurunegala district and in an industrial estate in the Gampaha district.

The majority of the workers were young, belonging to the age group 16 to 29 years: 81.6 per cent women and 80.0 per cent men in the EPZ factories and 73.3 per cent women and 60 per cent men in the rural garment factories. Workers in two of the rural garment factories had the same age profile as the workers in the EPZ factories but the factory in the Kurunegala district also

employed older women and men. The majority were also single women and men, 83.4 per cent women and 70.0 per cent men in the EPZ factories and 65.6 per cent men and 66.7 per cent women in the rural garment factories. Around 75 per cent of these workers were dropouts from secondary schools and 20 per cent to 25 per cent had had a complete secondary education of 10 to 12 years.

The women were part of the large labour reserve who have been moving in and out, or have been moved out of the factories. A high percentage, 71.4 per cent women and 76.6 per cent men in EPZ factories and 83.3 per cent women and 73.3 per cent men in the rural garment factories had been employed for less than five years in their present workplace. Their motives for seeking employment in these factories were: the need for a job and income and the absence of alternative employment opportunities. The spouses of married workers welcomed the economic participation of women as a contribution to family income. A small number of women workers had been previously employed, mainly in garment factories, while the men had been in a wide range of jobs.

It is evident that the EPZ factories have opened up new employment opportunities for women who were willing to move out of their homes and live in the vicinity of the zones. Subsequently rural garment factories have offered new employment opportunities for women who preferred to work nearer their homes. In the context of the high incidence of unemployment among secondary school leavers and secondary school dropouts, these factory jobs were an alternative to unemployment. The macro-economic reforms have therefore provided access to employment to a large number of women, particularly young women. These workers came chiefly from low income families in which the men were cultivators with unstable incomes, artisan or low skill workers or unemployed, and few women were economically active. The regular income of women factory workers has brought economic stability to these families. The women too have access to an independent cash income.

Among the women factory workers, 96.9 per cent in the EPZ factories and 92.2 per cent in the rural garment factories were machine operators or helpers or were engaged in related tasks and were classified as semi-skilled assembly line workers. Only 2.3 per cent among the EPZ workers and 3.3 per cent among the rural factory workers had reached positions of line leader, supervisor or

quality controller. Others were labourers. Among the men, 83.4 per cent in EPZ factories and 53. 3 per cent in rural factories were machine operators, while 16.6 per cent in EPZ factories were line leaders, supervisors, quality controllers and storekeeper, and 40 per cent men in the rural factories were employed as manager, quality controller, supervisor, maintenance worker and mechanic, and one man was a labourer. Over half the workers in the EPZs and a lesser proportion in rural garment factories saw no prospect of promotion even in the distant future. It appears that women were incorporated into a gender-based hierarchy in the workplace. The only manager was a man, and men had access to technical jobs.

Most workers considered themselves to be in the permanent work force — 93.1 per cent women and all the men in EPZ factories and 78.9 per cent women and 86.7 per cent men in the rural factories. However, these women workers and the majority of men workers had not received official contracts, contrary to regulations. With the exception of a few men, all the women and most men believed that they could be dismissed without much formal notice. Some of the case study subjects had been victims of arbitrary dismissal. The casualization of this workforce is reflected in the lack of job security.

Labour legislation provided for a nine hour working day inclusive of a one hour interval but 40.8 per cent of women and 36.7 per cent of men in the EPZs and 34.4 per cent women and 66.7 per cent men worked more than nine hours in compulsory overtime, and around 70 per cent women and 70 per cent to 100 per cent men were employed in night shifts. Only the higher level workers could leave the workplace during the lunch and tea breaks. The majority of workers in the EPZ factories and a smaller proportion in the rural factories claimed exposure to occupational health hazards such as dust pollution, continuously standing for long hours and hand injuries, and many complained of the absence of safeguards and compensation. Workers in the Kurunegala factory had the least complaints. Around 70 per cent of workers in all factories found the work monotonous or tiring, but not more than 30 per cent wished to change their jobs as they had no alternative employment opportunities. Trade unions could not operate in EPZ factories and around half the workers were members of the organizations established by employers. There were trade unions functioning in two of the rural factories and worker unrest has

compelled the entrepreneurs to close the factory in the Kalutara district soon after the study, leaving an unemployed labour force in search of employment. Working conditions were not, therefore in consonance with workers’ rights as spelled out in national labour laws or in international instruments such as ILO Conventions and CEDAW.

Wages were low but were in conformity with or above the minimum wages specified for garment workers. Around 90 per cent received Rs 2,500 to Rs 5,000 a month, which is just above the ‘poverty line.’ In the EPZ factories, statutory minimum workers’ benefits such as EPF, ETF, and overtime payments were given, also medical benefits and tea, but only half had meals and rest room facilities and few had transport facilities. Both facilities and amenities were better in rural factories as all received not only EPF, ETF and overtime payments but also medical benefits, tea, meals, incentives and bonuses while one third had access to rest rooms and transport. These facilities were the result of the intervention by the state laying down minimum conditions for entrepreneurs at the inception of the rural garment factories. Clearly a laissez faire stance by the state in promoting private enterprise is not conducive to the interests of workers.

The establishment of factories in zones led to an influx of workers seeking accommodation in the vicinity of the zones. Among the EPZ workers in the sample, 53.3 per cent women and 16.7 per cent men lived in boarding houses. The majority of women complained of overcrowding in houses improvised to accommodate 11 to 100 workers (83.3 per cent) and in rooms shared, with five to ten other workers (74.8 per cent), especially in the Katunayake EPZ. Fees were low as rooms were shared, but poor transport facilities, especially after night shifts, exposed workers to sexual harassment and abuse. Among rural garment workers, 26.7 per cent women and 40 per cent men lived in boarding houses but these were largely rural houses where there was hardly any congestion in houses or rooms. In fact, the rationale for establishing these rural factories was to enable workers to travel to work from their homes as far as possible. Hence they did not spawn agglomerations of boarding houses as happened near the Katunayake and Biyagama EPZs.

The situation of women workers in export-oriented garment factories that surfaced in this study confirms the findings of

international and cross country studies such as those by Elson and Pearson, (1981), Heyzer, (1987); Joekes, (1989), and by local studies (Voice of Women, 1983; Goonetilake, 1987; Weerasinghe, 1989; CENWOR, 1994). Employment opportunities for women have increased significantly, but the absence of opportunities for upward mobility has reinforced the gendered hierarchy in the labour market and working conditions have created new forms of gender subordination to largely male employers and managers in a system in which women are seen as low cost and dispensable labour. Men machine operators shared the tribulations of women workers but men had more opportunities for upward mobility and greater access to professional and technical employment in factories. According to their own perceptions, both women and men workers felt that their access to employment had increased women’s self confidence and self esteem as income earners, but the majority did not feel that they had improved their ‘status’ in the community as society tended to look with disfavour on the low status of factory workers in the occupational and social structure.

Home-based garment workers in the informal sector

Home-based production of garments for sale is a popular occupation of women in the informal sector. This study examined developments in a locality where the cut piece sewing industry, a spin off from the garment factory industry, is concentrated. Households in some villages in the suburbs of Colombo have, since the heyday of import substitution industries in the 1970s, purchased unutilized cloth from garment factories and produced women’s and children’s clothes and household linen, particularly for less affluent buyers. With market liberalization, their enterprises have expanded and cloth and ancillary materials are purchased now from factories and from shops. Traders have opened shops, and subcontractors use home-based female labour for production of garments.

The spouses of the women in the study sample were engaged in business or were urban workers. Unlike the factory workers these women belonged to an older age group and 86.7 per cent were between 30 and 59 years, and 83.3 per cent were married or were widows. Several of the men engaged in this occupation (and their overall numbers were small) were young and married. Two thirds of the women and 60 per cent of the men were secondary school

dropouts and 33.3 per cent women and 40 per cent men had a complete secondary education. The women had taken to this activity because they wished to increase family income, but they needed also to combine child care and domestic responsibilities with economic activities. Their inevitable option has been the main activity in the locality – the cut piece sewing industry. Around 25 per cent had worked earlier in garment factories or in other self employment ventures. Around 30 per cent had been engaged in this occupation for over 10 years. Most of the men, however, had been involved for less than five years.

These home-based workers were either self employed — 75 per cent women and 60 per cent men — or were subcontracted workers — 16.7 per cent women and 40 per cent men. The majority (70 per cent) were engaged in cut piece sewing and the rest in conventional sewing or in both. Most of them, over 90 per cent women and around 80 per cent men had acquired their own sewing machines and block machines. They rarely worked more than eight hours, but worked into the night when there were deadlines to meet. The self employed sold their products to shops, local traders or itinerent traders, and subcontracted workers executed the orders channelled to them by middlemen or subcontractors.

They were spared the hardships of factory routine and excessive labour control, but the subcontracted workers were at the mercy of the intermediaries who garnered an inequitable share of the profits, and the self employed had to compete in the open market and to contend with the malpractices of traders. Consequently their incomes were not substantial, with 36.7 per cent of the women earning less than Rs 2,000 (below the poverty line) 33.3 per cent between Rs 2,500 and Rs 3,500 and 30 per cent between Rs 4,000 and Rs 6,000. The men earned between Rs 2,500 and Rs 3,500. With few exceptions they were not more prosperous than factory workers but over 75 per cent enjoyed their work. Over 80 per cent wished to continue these activities. Nearly all the women and men had increased self confidence and self esteem, and most women felt that they had enhanced their status in the community as income earners and not ‘housewives.’

These women workers have responded to new employment opportunities that have opened up in the market economy, without detriment to their child care responsibilities. Their profits have not been large but they are not as economically depressed as many other

low income women in the informal sector. The case study of the home-based garment workers showed how a widow started this sewing enterprise after her husband’s death and developed it with confidence and skill into a family enterprise with her daughters. Overall, their relative autonomy as self employed women competing in the market has enhanced their self respect and they contribute significantly to family survival and upward mobility.

Textile factory workers

The power loom textile industry was set up before the macro-economic reforms introduced in 1977. With the reforms the state established a few large power loom textile factories in 1979-80 which, however, were subsequently privatized in the 1980s. This study examined the situation of women and men textile workers in a large factory which had been a small private enterprise in the 1970s and had been acquired and expanded by the state as a large textile mill. Losses incurred by the state led to its privatization. The ownership of the factory changed hands twice in the 1980s and, after some vicissitudes, has settled down to stability under the present ownership.

The factory was located in the North Western Province and the workers were drawn, as in the case of garment factory workers, from the families of small farmers and industry workers. Age wise, 63.3 per cent of the women workers were between 20 and 29 years and 53.3 per cent of the men belonged to an older age group; 40 to 54 years; 56.7 per cent women and 41.7 per cent men were married. Their educational levels were higher than those of garment factory workers as 76.7 per cent of the women and 73.7 per cent of the men had a complete secondary education and only 20 per cent of women and men were secondary school dropouts. One woman and one man had a university degree. A few of the women and half the men had been employed previously, chiefly in garment and other factories. While 63.3 per cent of the women and 46.7 per cent of the men had been less than five years in their current employment, 20 per cent of the women and 53.3 per cent of the men had been employed in the same establishment for over 10 years, a few since the inception of the factory. The profile of these workers was therefore slightly different from that of garment workers. Two thirds of the women and half the men had sought employment

because they were in need of a job and income. The rest had failed to find alternative employment.

While the majority of the women workers (80 per cent) and men workers (73.3 per cent) were machine operators, one fifth of the women workers had jobs such as lecturer, trainer, laboratory assistant and supervisor and there was a male executive and men in positions of trainer, electrician and clerk. The establishment had therefore offered women space to access at least middle level employment, although the only highly paid executive was a man. However 60 per cent of the women and 33.3 per cent of the men did not anticipate promotion. Monthly wages were higher than in garment factories, perhaps as a consequence of long service, ranging from Rs 3,500 to Rs 7,000, middle level employees received between Rs 8,000 and Rs 10,000 and the male executive enjoyed a salary of Rs 25,000. Workers were provided with all benefits, those prescribed by labour legislation as well as tea, meals, uniforms, medical facilities, restrooms, incentives and bonuses. Some had transport facilities and even accommodation.

Although the majority liked their work and 83.3 per cent women and 86.7 per cent men had job satisfaction, there were nuances of job insecurity. All the men and 90 per cent of the women were permanent employees, but 26.7 per cent of the women and 33.3 per cent of the men did not have contracts, and half the workers believed that they could be summarily dismissed. None of the workers reported excessive working hours but 83.3 per cent and 73.3 per cent men were allotted night shifts. Most of them — 90 per cent women and 80 per cent men — complained of occupational health hazards such as dust, chemicals, noise, machines and standing long hours, and absence of safeguards.

While 43.3 per cent of the women and 26.7 per cent of the men travelled to work from their homes, only 20 per cent of the women and 26.7 per cent of the men were in boarding houses as the rest had found accommodation with families. Boarding houses were not overcrowded and workers had no transport problems but some of the women were fearful of sexual harassment. Married workers did not find it easy to combine their jobs and childcare responsibilities. The spouses, however, welcomed their economic activities as inputs to family incomes. Both women and men workers had positive perceptions regarding the impact of their employment. They were confident and assured because they were recognized as income

earners and, unlike the garment workers, they felt that their status in the community had improved.

It is apparent that textile factories are perceived as distinct entities from garment factories. Workers’ unrest has been prevalent in large textile factories but the management of this particular factory has provided amenities that have contributed to job satisfaction among employees. At least some women have had access to non-manual employment. But as in all factory establishments, workers appear to be vulnerable to job insecurity in an ethos dominated by the demand by employers for deregulation of labour.

Handloom workers

As discussed earlier, one of the negative consequences of the macro-economic reforms was the collapse of local industries and their continuing underdevelopment at the village level. In the 1960s and 1970s the protected and subsidized handloom industry had thrived in villages providing income earning opportunities for women. After their debacle in 1978-80, many centres all over the country were closed and these displaced women sought other means of livelihood, often with little success. The industry survived, however, in some villages as a consequence of state efforts to establish Janasalu centres to revive the industry.

The women handloom workers in villages in the Kurunegala district in the study had taken to this occupation as a survival strategy because of a lack of access to better employment opportunities. No men were employed in this industry. These women belonged to agricultural families and their spouses were farmers or labourers. The majority (60 per cent) were over 40 years of age and 66.7 per cent were married. Their educational levels were not low, as 36.7 per cent had a complete secondary education and 50 per cent were secondary school drop-outs. The case study of the handloom workers found that the women workers had sought state and other avenues of employment on the strength of their secondary school certificate qualifications but after repeated failures had fallen back on the handloom industry in the village.

Half these women workers were attached to the State Janasalu Centre. The Department of Textiles purchases yarn, dyes it, distributes it to the centres, and workers are paid for their products on a piece rate basis. These products are sold by the Department

at shops, fairs and exhibitions. The other half were members of the Textile Co-operative Societies which are co-operative societies of former weavers. These Societies followed the same procedure as the Department in distributing materials and making piece rate payments for the products. Both the Department and the Societies found it difficult to sell the products turned out by these women in the face of competition in the market. Consequently the incomes of the women were low, between Rs 1,500 and Rs 2,500.

These women had no problems in their working environment such as those faced by factory workers. They had no marketing problems like the home-based garment workers, as their products were purchased by the Department or Societies. Their problem was their low economic rewards as their income was necessary for family survival. They were dissatisfied with their occupation but had no alternative avenues of employment. A malaise caused by many years of frustration seemed to overpower them. As ‘residual workers’ in an industry that had an uncertain future, few had self confidence (43.3 per cent), self esteem (10 per cent) or recognition as producers (9.3 per cent), while no one felt that their economic activities improved their status in the community.

It is clear that the village based handloom industry has yet to recover from its collapse two decades earlier or to develop new directions. Handloom workers in the rural environment have remained in the industry because they have no alternative employment opportunities. Their survival is an outcome of the subsidies that are still offered by the state in spite of losses, as a legacy of the past.

Retrenched textile workers

The accelerated pace of privatization and reduction in the size of the public sector in the 1990s affected both men and women employees in the services sector and in state economic enterprises. Among the establishments closed down or sold to the private sector were the smaller power loom textile centres distributed throughout the island. This study examined the consequences of this facet of the macro-economic reforms for workers in textile centres in one district in the North Central Province and two districts in the Southern Province.

These workers had been retrenched between 1991 and 1995.

Two thirds of the women and 40 per cent of the men had been employed in centres that had closed down and one-third of the women and 60 per cent of the men in centres that had been privatized. At the time of retrenchment, 15 per cent of the women and 13.3 per cent of the men were between 28 and 35 years of age, 80 per cent of the women and 86.7 per cent of the men between 35 and 49 years, and 3.3 per cent of the women were between 50 and 54 years. The majority therefore had many more years of service ahead and were in their economically productive years. All the men had been married, 12 per cent of the women were single, 83.3 per cent married and 3.3 per cent were widows. They were secondary school products as 45 per cent of the women and 40 per cent of the men had had a complete secondary education, and 51.7 per cent of the women and 60 per cent of the men were secondary school dropouts. One woman was a university graduate.

The majority of the women (90 per cent) had been machine operators, 5 per cent had been Assistant Supervisors, Supervisors or Instructors and the woman university graduate had been an Assistant Manager. Among the men, 40 per cent had been machine operators, 26.7 per cent welders, fitters and cleaners, and 33.8 per cent labourers. The majority had been employed for many years — 20 per cent of the women and 13.5 per cent of the men for 20 to 27 years, 73.3 per cent of the women and 60 per cent of the men for 10 to 20 years, and 6.7 per cent of the women and 26.7 per cent of the men for less than 10 years. The majority (75 per cent women and 66.7 per cent men) had had salaries between Rs 2,500 and Rs 4,000, and 22 per cent women and 26.7 men had earned less than Rs 2,500.

The majority of the workers, 60 per cent women and 66.7 per cent men had been caught unawares by the measures taken to retrench them. They were given a compensation package, ranging from Rs 2,000 to Rs 70,000, — 66.7 per cent women and 53.3 per cent men receiving between Rs 18,000 and Rs 38,000 — although 6.7 per cent of the women and 26.7 per cent of the men claimed that they had still to receive this package. They also received their consolidated pensions and continue to receive their monthly pension.

Their experiences after retrenchment illustrate the plight of relatively low income workers displaced suddenly from jobs which

they had been engaged in for several years, in a country with a high incidence of unemployment and escalating costs of living since the macro-economic reforms. Their compensation package and consolidated pensions had been used to settle loans and for house construction and repairs, the education of children, illnesses and other family expenses. Only three of the women had invested some of the money in opening a boutique/small shop, purchasing a sewing machine and assisting in the spouse’s business.

Two thirds of the men and 21.3 per cent of the women had sought other avenues of employment; 10 per cent of the women and 13.3 per cent of the men had found jobs in other textile factories which had been closed down soon after, displacing them again. At the time of the study, 8.5 per cent of the women were self employed in sewing, weaving door mats and petty trade and earning between Rs 600 and Rs 2,500 a month; 3.3 per cent were cultivating their own land and earning around Rs 1,000 and one woman was a daily paid casual labourer (1.6 per cent). Among the men 53.3 per cent were employed as security officers, labourers and in self employment and cultivation, earning between Rs 500 and Rs 4,000. Nearly all had no income from other sources such as land, and most of them could not depend on assistance from the extended family.

These women and men were strongly affected by their retrenchment. Nearly all said that their pensions and current incomes were inadequate to meet the high costs of living and 46.7 per cent among both women and men were in debt. Their quality of life had deteriorated. With the exception of around 20 per cent of the women who felt they had more time now for childcare, the majority of women and all the men had been emotionally upset when they were retrenched and faced an uncertain future. They were anxious regarding their inability to meet expenses and to give their children a good education, and 70 per cent of the women and 86.7 per cent of the men had been mentally depressed by their loss of status as income earners. They were uncertain regarding their future prospects. A few had plans and some men hoped to be self employed. These women did not belong to the age group in demand as factory labour. In their circumstances and in the macro-economic context they faced unemployment, unviable self employment or domestic service overseas.


The three women and two men entrepreneurs who related their life histories all belonged to middle class families with economic resources. The women were married to men in the public service or in business, and two of the women were themselves professionals. One man had been a public servant and the other was born into a thriving family business. They all had family support and connections that enabled them to visit enterprises abroad. Four had commenced their activities on a small scale during the 1970s but their enterprises had expanded and flourished with the incentives offered by the new policies for private enterprise in the 1980s and 1990s.

The entrepreneurs were themselves creative and had the initiative to meet market or upmarket needs. They also had the stamina and resources to cope with inconsistencies in state policy especially with respect to handlooms, and the absence of adequate infrastructure. These successful handlooom and garment exporters and batik artists lived in a different world from the village based handloom workers and the home based garment workers. They were themselves employers of factory workers. These different experiences mirror the increasing income disparities in the 1980s and 1990s.

Employment, underemployment and unemployment

The employment experiences of these different groups of women illustrate the impact of three strands of policies stemming from the macro-economic reforms and structural adjustment programmes introduced over two decades ago. The export orientation of the economy has been seen to create new employment opportunities especially in garment factories in Export Processing Zones, and in rural garment factories, particularly for young women. To many of these women in low income families without regular incomes, their jobs brought some temporary economic stability to their families and were for them an alternative to unemployment or to elementary economic activities in the informal sector. The large privatized textile mills were another avenue of opportunity. The home-based cut piece sewing industry was a spin-off of the garment industry in the informal sector as distinct from outsourcing through direct subcontracting by firms which is the subject of another CENWOR study (2000). Women (and men) utilized this space to create

enterprises that facilitated the generation of income to meet the needs of low income families.

The second policy was the withdrawal of producer subsidies to small local industries and the decontrol of prices that led to the collapse of these industries in a competitive market, as in the case of the feminized handloom industry that had depended heavily on subsidized imported inputs. The village-based handloom industry continues to be depressed in the context of unsupportive macro-economic policies. The handloom weavers in this study are the underemployed survivors who subsist on state support sans opportunities for economic advancement.

The third policy of privatization of public enterprises and reduction in the size of the public sector resulted in the closure of textile centres which provided employment opportunities for low income families for many years. Many of these retrenched workers had many more years ahead in employment. Their sudden displacement in a context of high incidence of unemployment relegated them to unemployment or to self employment, to a decline in economic resources and to deterioration in their quality of life.

The experiences recounted by entrepreneurs from middle class families underscore the futility of the farcical concept of a ‘level playing field’ in economically developing countries, and its implications for widening income disparities and perpetuating poverty.

The women who gained access to new employment opportunities in factories have been incorporated into the global market. As documented in international and local studies, these women too have been vulnerable to forms of labour control that include ‘deregulation’ in terms of employment, excessive working hours and workload, exposure to occupational health hazards, and absence of job security and opportunities for upward mobility. They have moved into productive employment that reinforces gender inequality and new forms of gender subordination. The more ‘autonomous’ home-based workers have inadequate resources to contend with market forces and with exploitative or intermediaries in order to enhance their prospects.

Gender Roles and Relations

The second phase of the study explored gender roles and relations

in a sub sample of the households identified for study in Phase I. Each woman worker, one or two other women and a man in each family in the sub sample of 140 households spoke of their roles, their relationships and their perceptions. It is not possible to relate manifestations of or changes in gender roles and relations directly to the impact of macroeconomic reforms but the study provided insights into the many facets of change or continuity in gender specific behaviour in families.

Gender roles

The social construction of gender and the socialization of women and men from infancy into ascribed gender roles have determined to a large extent the overt demarcation of the reproductive roles of women and the productive and community roles of men and the concomitant gender division of labour. Women’s unequal responsibility for domestic work and the underestimation and low visibility of their productive role are seen in all societies to have disadvantaged them in their access to the labour market, extended their ‘day’ and limited their participation in the ‘public sphere.’

As in most societies, women’s labour force participation rates have increased in Sri Lanka in recent decades. As a consequence of macro-economic reforms women workers in garment and large textile factories and home-based workers in the cut piece sewing industry in the informal sector had extended their productive roles even as low cost dispensable labour, while handloom workers and retrenched workers had been marginalized or displaced. All these women however have made a crucial contribution to family income and to the economy.

While women’s extension of their productive role has necessitated adjustment of activities in some families, overall the gender division of labour in the household was not found to have changed significantly. Six ‘household’ tasks were identified for analysis of the division of labour and responsibilities — cooking, washing utensils, washing clothes, fetching water, child care and care of the old and sick. In the six groups of households in the study 50 per cent to 100 per cent of women workers and other women in their families, both economically active and non active, responded that ‘almost all’ or ‘most’ of these tasks were performed by them. Only a few women had low levels of participation. On

the other hand, 50 per cent to 70 per cent of the men said that they had ‘little’ or ‘no’ involvement in these tasks and 2 per cent to 33 per cent claimed that they undertook a substantial share of these tasks. The study confirmed the findings of most research that the gender division of labour continues to be inequitable despite the increasing responsibilities and work of women as income earners.

Any shift in the division of labour was often at the cost of the increasing workload of other women in families in which there were economically inactive women rather than in more participation by men. There were however nuances of role conflict in sharing domestic chores in the perceptions of both women and men. The maternal role also surfaced as a contentious issue, with men perceiving the neglect of children as a necessary corollary to women’s economic activities outside the home and women’s own ambivalence and sense of guilt regarding the implications for children. The ideology that underpins the allocation of the reproductive role to women appeared to condition the perceptions of both men and women.

Gender relations

Asymmetrical gender relations are largely conditioned by patriarchal social norms, values, traditions and customs which have been internalized also by both women and men. As the primary social unit the family is seen to be a central location of unequal gender relations and social control. The study explored gender relations in the family in the allocation of resources, control of income and assets, decision-making, physical mobility, marriage, reproduction and sexuality and gender-based violence.

Allocation of resources

The study was limited to the relative access of women and men in families to education, health, food and employment. The social philosophy of equity in access to education and health underlying national policies of free education and health services without gender differentiation since the 1940s evidently pervades public perceptions. There was consequently no overt gender discrimination in these families in access to education or health care. Almost all women and men (98 per cent) believed that girls

and boys should have the same education opportunities. In these families 91 per cent of the women and men said that girls and boys had had similar educational opportunities and equal family support. Aspirations for the higher education of girls were in fact slightly higher than those for boys. Again, 91 per cent of women and men resolved not to withdraw any child from school in the event of severe financial constraints or only to withdraw the child — whether male or female — who was not doing well in school. Only 3.4 per cent preferred to withdraw a girl child and 1.2 per cent to withdraw a boy child. The influence of gender role stereotypes was evident in other aspects, in the identification by a minority of different tasks assigned to boys and girls in schools and in the perception of different behavioural expectations, and overtly in their own gendered vocational aspirations while in school.

There was a similar acceptance of gender equality in access to health care by 91 per cent of the women and 90 per cent of the men. The often controversial issue of the distribution of food was a non-issue. The majority agreed that male and female members of families do not often eat together, but attributed this fact to different time schedules and changing life styles. Very few, from 2.2 per cent to 8.8 per cent in the different groups, subscribed to the view that such behaviour reflected a tradition of gender inequality in access to food.

Contrary to these egalitarian perceptions, and despite the fact that these women workers had obtained some form of employment, gender differences were seen clearly in access to employment. The interface of lack of employment opportunities and family opposition to women seeking employment outside the home or in unwelcome occupations appears to have limited the options particularly of workers in the informal sector and several of those who were economically inactive. Only 27.1 per cent women workers and 26.0 per cent other women in their families have made independent choices of employment. It is significant that 73.7 per cent of the men have made their own decisions, indicating gender inequality in access to economic resources.

Control of economic resources

Around half to one-third of men in these families were found to own land and houses but few women, particularly in families of

factory workers, had such assets. Men, too, appeared to have greater power to acquire and dispose of these assets.

However, 93.1 per cent of women workers and 72.9 per cent of other women had control of their wages, and 86.2 per cent and 68.9 per cent respectively had their independent bank accounts while half of the women workers also had informal savings. Surprisingly only 45.3 per cent of the men retained their wages, 44.2 per cent handing their pay over to their wives presumably for ‘safe keeping,’ while only 3 per cent of the women gave their wages to their spouses. Only 57.9 per cent of these men had separate bank accounts.

It appears therefore that while men tended to have control over traditional assets such as land and houses, women have acquired control of the income they generate. It was also seen that women workers controlled their income whether they were in the formal sector or were home-based workers, and that any type of economic activity has enabled women in these families to have more control of economic resources. In this respect, the macro-economic reforms appear to have only reinforced a process of power sharing that already prevailed in these families.

Decision-making in the household

The pattern of joint decision-making in the family that emerged from the responses of the women and men confirmed the findings of many other studies in Sri Lanka. There was no perception that these decision-making powers had increased significantly with access to employment, contradicting an observation made at another point that employment outside the family increased women’s decision-making powers. There was however no hint of male dominance in decision-making.

It is interesting to note that while 62 per cent of the women and half the men wanted the decision-making powers of the women in the family increased, a much higher proportion, 87 per cent and 84 per cent favoured an increase in decision-making by women in community matters. It could be that women’s low level of participation in political and community life motivated this concern, or that increase in decision-making outside the family was less problematic than an increase within the family.

There was general agreement (96 per cent women and 92 per

cent men) that women had less physical mobility than men. Around two-thirds of both women and men associated this situation with women’s vulnerability to sexual harassment in public places, and only around one-fourth saw it as a restriction imposed by traditional social norms. There was a difference seen in freedom of movement to educational institutions, workplaces and kinship and social networks on the one hand, and to cinemas and restaurants on the other.

Overall, around 70–100 per cent women and men in the six research samples had positive perceptions regarding changes in the personalities and life styles of women (particularly those employed outside the home), brought about by their greater economic security and the knowledge and experience they acquired in their ‘public’ activities. Among the changes noted were greater physical mobility, improved self esteem, and to a lesser extent, self confidence, and increased standing in the family, all of which could modify gender relations in the family.

Marriage, reproduction and sexuality

There was no evidence of an unyielding patriarchal control of marital relations but issues were inevitably complex. Family control in the selection of spouses was seen only in the families of village based handloom workers. Overall, over 65 per cent of women and 75 per cent of men had made independent decisions in their choice of spouses. There was only one child marriage, at 12 years, and a small number of women and even men were 16–18 years old when they married. The majority of women were between 18 and 25 years and a small proportion were older, and men were over 25 years. Their hopes for the marriage of their offspring were on the same lines — 18 to 25 years for daughters and over 25 years for sons. Attitudes to marriage outside caste, religion and ethnic group were seen to be moving towards greater flexibility, except again in the houses of village handloom workers.

A majority of women workers (53.1 per cent to 91.7 per cent) and other women in their families (41.7 per cent to 87.5 per cent) and around 42.9 per cent to 83.3 per cent of men felt they could move out of an unhappy marriage if they had adequate financial resources. There were, however, double standards in acceptance of reasons for dissolution of a marriage, as a woman’s unfaithfulness

was considered to be a more important reason than a man’s infidelity or wife beating. Only around 40 per cent of families had used contraceptives, largely through preference for ‘natural methods’ or lack of information. Sexual satisfaction was considered to be more important for men than for women but male control of sexuality did not surface as a major problem in marital relations as 77.8 per cent of women workers and 71.7 per cent of other women said they could refuse sex if they so desired.

Despite these relatively egalitarian trends, gendered norms underlying social practices that fostered unequal gender relations influenced the attitudes of women and men. The dowry system did not appear to be as oppressive as in some societies as less than half the women (and men) had been given dowries in the context of the economic constraints of low income families and the relatively low percentage of ‘arranged’ marriages. While over half the women and men hoped to give their daughters dowries, less than 10 per cent intended to ask formally for dowries for their sons — a positive trend reflected in the distinction between ensuring the economic security of their daughters and the unseemly bargaining for dowry for sons that takes place in more affluent households.

On the other hand, 69 per cent of women workers and 59 per cent of other women had undergone the virginity test on the night of their marriage. Many women justified this test as essential to ensure the ‘purity’ of a woman before marriage, and very few women questioned its unscientific assumptions or its relevance. There appears to be a colossal degree of insensitivity to the ‘double standards’ and the gender subordination inbuilt in this test, and to its violation of the rights of women, largely as a result of an unchallenged process of socialization in families. ‘Son preference’ was also articulated by two thirds of the women and over half the men, in consonance with gendered norms that ascribe to men the role of providers and protectors of families and transmitters of the family name. The girl child, however, was not disadvantaged as seen earlier, and none of the 3 per cent to 4 per cent of women and men who had used the scanning process to ascertain the sex of the child before birth, had used this information to abort the female foetus.

Women in these families were not therefore seen to be subservient in marital relations and in their reproductive roles, but their acceptance of gendered myths and practices erodes the equality manifest in other aspects of gender relations.

Gender-based violence

The response of both women and men indicate that domestic violence was rife in families in their neighbourhood. There was considerable reluctance to provide comprehensive information regarding domestic violence in their own families, in their concern for family prestige and privacy. Nevertheless around half the women and men admitted to domestic violence and abuse in their families but downplayed the level of violence to largely verbal abuse which was tolerated by the majority of victims. Few spoke of battering and very few (around 3 per cent) had taken concrete action such as moving out of the marriage or seeking the assistance of the police. Some of the ‘case studies’ had rejected alcoholic spouses and physical violence and had taken decisive steps to establish their individual households while others had decided to live with domestic violence.

The factors contributing to domestic violence in the neighbourhood identified by women and men point to tensions within families. It was agreed by almost all (95 per cent) that addiction to alcohol was the most important predisposing factor followed by economic constraints (around 75 per cent). Another factor in the perceptions of around 60 per cent of women and men was the repudiation by women of patriarchal power relations and control, resulting in accusations by men of neglect of household work by women, ‘talking back’ to husbands and disobedience to husbands and elders, and the resistance by men to this challenge to male power. Extra marital affairs were also causes of tension, but dowry demands, birth of girl children and childlessness have had little effect on spousal relations.

Few spoke of violence in the public sphere. Only 2 per cent to 3 per cent reported rape or verbal sexual harassment in the workplace. In an environment that is conducive to under reporting, the figure of a mere 10 per cent in the incidence of rape among EPZ factory workers bodes ill for the human rights of women.

Gender specific aspirations

In a religious environment in which the majority attach importance to re-birth, the impact of gender inequality in the family, economy and society is seen vividly in the aspirations of 73 per cent women workers and 78 per cent other women in their families to be reborn as men, 15 per cent and 13 per cent as men or women and only

10 per cent and 6 per cent as women. Among the men, 90 per cent wished to be reborn as men and 8.5 per cent as men or women.

Mental distress and well-being

The two standard measures — Goldberg. General Health Questionnaire (GHQ) and the Subjective Well-Being Inventory (SUBI) — used in phase 2 of the study explained the mental health stress of the participants in the study. The results indicated that coping with economic constraints, and multiple roles to meet individual and family needs had a negative impact on the women and men in these low income families in the lowest strata of the occupational hierarchy.

Nearly half the participants (46.9 per cent) manifested some degree of mental distress and around 88 per cent had low levels of well-being. Inevitably women with their multiple roles had a higher level of mental distress than men (50.8 per cent women and 37.4 per cent men), but had similar low levels of well-being as men (88.6 per cent women and 86.9 per cent men). It has to be noted that the majority of women were as much family income earners as the men. Involvement in economic activities has clearly had a salutary effect on mental health. Both women and men who were not employed experienced a higher level of mental distress (63.4 per cent women and 65 per cent men) than those who were employed. (43 per cent women and 31 per cent men). Likewise more women and men who were not employed had lower levels of well-being (91.1 per cent women and 95 per cent men) than those who were employed (87.1 per cent women and 85.1 per cent men).

Multi-variable analysis indicated that education (at least secondary education) and marriage reduced mental distress and increased the level of well-being, while age tended to increase the levels of mental distress and reduce levels of well-being. Overall increasing employment opportunities appears to be a strategy to relieve mental distress but these stresses experienced by women and men engaged in efforts to achieve sustainable livelihoods are a matter of concern.

Continuity or change

This exploration of gender roles and relations in the families of women workers affected positively or negatively by macroeconomic

reforms has provided useful insights into family relations although evidence of a direct relationship between the impact of economic reforms and changing roles and relations cannot be adduced.

Women in EPZ garment factories, rural garment factories and large textile factories and home based garment workers in the cut piece sewing industry have extended their economic roles and have become visible and crucial economic producers as a consequence of new employment opportunities. Women handloom weavers and retrenched textile workers have been denied employment opportunities as an outcome of the same reforms but still contributed to family income.

Women’s productive and reproductive roles were seen to be interrelated. Women factory workers have been incorporated in the international division of labour as low cost dispensable labour because as Elson and Pearson (1981) have argued, women’s reproductive roles are seen as justification for perceiving them as secondary earners, and patriarchal family relations are transposed to gender subordination in the workplace. Women’s reproductive roles were seen also to limit their options in employment and to confine them to home-based economic activities or to ‘voluntary’ unemployment. Women’s ascribed nurturing and servicing roles were reinforced through gender based curriculum differentiation and behavioural expectations in the education process and in stereotypical vocational aspirations.

As documented in studies the world over, women’s expanding economic roles have not significantly changed the inequitable gender division of labour in the household. However, conflict over the sharing of domestic chores and women’s sense of guilt regarding their child care responsibilities and their ambivalence regarding the juxtaposition of economic and domestic roles are symptoms of a transition that could be accelerated by the movement of labour in the context of rapid globalization.

Gender relations have presented more complex issues. Macro social policies have ensured gender equality in access to education and health care in the families of these women workers. Allocation of food was not seen widely as a gender issue. Women’s access to employment outside the home and to specific occupations was seen to be limited to some extent by family control or influence in contrast to more autonomy by men as natural ‘providers’ of families.

Men appeared to have more control of assets such as land and houses. But women had control of the income they generated from employment, and the savings they accumulated in the form of bank accounts, jewellery or informal savings. Their income was not subsumed in family income while the men tended to entrust their wages to wives as ‘keepers of the purse.’

The fact that this practice prevailed in all families, irrespective of whether women workers were beneficiaries or victims of macro-economic reforms indicates that many women have been always empowered in the control of the resources they generate. Inevitably, increasing access to employment and income would enhance this power and facilitate more equitable gender relations.

There was no clear evidence that women’s decision-making powers had increased substantially with economic participation. Women claimed that they share in decision-making and the pattern of power sharing that emerged was not male dominance but joint decision-making. Women were disadvantaged, in physical mobility, not by a tradition of seclusion, but through increasing awareness of their vulnerability to sexual violence.

Patriarchal relations in marriage, reproduction and sexuality were leavened by several positive aspects of marital relations. Child marriages were not envisaged. The majority of women had selected their spouses without family intervention, were able to move out of an unhappy marriage if they had adequate financial resources, and had the capacity to refuse to have sex with the spouse. While oppressive male dominance was not seen in many families, there was implicit gender inequality in practices determined by gendered norms. While the dowry system was not oppressive, and son preference did not result in female foeticide, the majority of women had undergone the unscientific virginity test on marriage and accepted it unquestioningly, unaware of its reinforcement of gender subordination and double standards. As there was no significant difference between the attitudes and practices of employed and non-employed women these different manifestations are perhaps part of the gender ethos and social perceptions in these families. The families of the handloom workers were more conservative largely because of their social exclusion from progress and concomitant changes in life styles.

The strongest evidence of patriarchal control was seen in the incidence of domestic violence. Although responses regarding

violence and abuse in their own families were muted, there was adequate information from the neighbourhood to identify male alcoholism, poverty and patriarchal relations as major causes of domestic violence. It was seen that women’s resistance to male authority and demands provoked men to resort to violence to retain and reinforce their control. In the public sphere, sexual harassment appeared to be tolerated but the few incidents reported of rape, particularly of garment workers, underscored extreme forms of male control of sexuality.

Gender relations in these families did not conform to the classic model of patriarchy discussed by Kandyotti (1988) and others. These multifaceted relations were not in consonance with the concept of power defined simply in binary terms of dominance and subordination. There were both asymmetrical and egalitarian relations in the control of resources, power sharing and marital relations and practices.

This study endorses the rejection by Beneria and Sen (1997) and others of the Marxist thesis that engaging in productive activities outside the home automatically liberates women. However, the fact that women employed outside their homes were perceived by both women and men to have made individual gains in self esteem, standing in the family, mobility, some self confidence and decision making capacity is consistent with Lim’s view that some modification in patriarchal relations does take place (Lim, 1997).

Wolf (1987) in her study of factory workers in Indonesia and Taiwan noted that young Indonesian women made their independent decision to seek employment in factories and had control of their incomes while in Taiwan families exercised full control of the entry of young women to employment in factories as well as their wages. She observed that employment outside the home had reinforced existing gender relations in the two societies. The absence in this Sri Lankan study of marked differences between the situation of women in factory employment, home based economic activities and exclusively domestic work indicates that increased economic participation has reinforced existing gender relations while at the same time creating new demands for role flexibility and power sharing.

The interface of macro policies, the social construction of gender and elements of egalitarian gender relations have contributed to continuity and change in the situation of women and men in the

relatively low income families of women workers in this study. Macro-economic reforms have had a differential impact on the employment opportunities of different groups, incorporating some in the global market on unequal terms and marginalizing others. They have not brought about radical changes in existing gender roles and relations. Macro social policies have reduced gender inequalities in access to resources but have not countered adequately the impact of the social construction of gender on the lives of women and men. Gender roles and relations in the families of these women workers were not seen to be oppressively patriarchal except for a few obscurantist social practices and gender-based violence in the domestic environment and the public sphere. Irrespective of economic changes these families have yet to achieve the total transformation envisaged by Szinovecz (1984) from ‘sex role segregated’ families to ‘sex role transcendent’ families in which role flexibility and equality in gender relations are ensured.


Ariyabandu M. and Hiddlege V. (1995) “Structural Adjustment: The Small Producers’ Dilemma”, Economic Review Vol. 20. Nos. 10–11

Atukorale P. (1986) The Impact of 1977: Policy Reforms in Domestic Industry, Upanath, 1:1

Beneria, Lourdes and Sen Gita (1997) ‘Accumulation, Reproduction and Women’s Role In Economic Development’, in N. Visvanathan (ed.) The Women; Gender and Development Reader, London: Zed Books

CENWOR (1994) Export Processing Zones in Sri Lanka: Economic Impact and Social Issues. Geneva: International Labour Organization

CENWOR (1995) Structural Adjustment and Women; The Sri Lanka Experience, Colombo: Centre for Women’s Research

CENWOR (2001) Gender Dimensions of the Impact of Macro Economic Reforms on Women, Parts 1 & 2, Colombo: Centre for Women’s Research

Commonwealth Secretariat (1989) Engendering Adjustment for the 1990s . London

Department of Census and Statistics (1981) Census 1990 — Labour Force Surveys, Colombo 2000

Elson, Diane (1991) Gender and Adjustment in the 1990s, London: Commonwealth Secretariet

Elson, Diane and Pearson, Ruth (1981) “Nimble Fingers make Changed workers: An Analysis of Women’s Employment in the Third World Export Manufacturing.” Feminist Review No. 7

Goonetileke, Hema et. al. (1987) ‘Industrialization and Women Workers in Sri
Lanka’, in Daughters in Industry, Kuala Lumpur: Asian and Pacific Development Centre

Harrigan J., Mosley P. and Toye J. (1981) Aid and Power. The World Bank and Policy Lending in the 1990s, London: Routledge

Heyzer, Noeleen (1987) Women Workers in South East Asia: Problems and Strategies, ILO/ARTEP, New Delhi

Indraratne, A.D.V. de S. (1998) Economic Development and Policy: An Overview in Fifty Years of Sri Lanka’s Independence. A Socio-Economic Review, Colombo: Institute of Social and Economic Studies

Jayaweera, Swarna (1998) Impact of Structural Adjustment Programmes on the Economic Activities of Women, Colombo: Centre for Women’s Research

Jayaweera, Swarna, Sanmugam Thana and Rodrigo, Chandra (2000) Invisible Women: Workers in Subcontracted Industries in Sri Lanka, Colombo: Centre for Women’s Research Study Services no. 15

Joekes, Susan (1986) Industrialization, Trade and Female Employment in Developing Countries, INSTRAW

Kandiyotti, Deniz (1988) Bargaining with Patriarchy, Gender and Society 2 (3), pp. 274–90

Lakshman, W.D. (1997) “Income Distribution and Poverty” in W.D. Lakshman (ed.) Dilemmas of Development, Sri Lanka Association of Economists

Lim, Linda Y.C. (1997) “Capitalism, Imperialism and Patriarchy: The Dilemma of Third World Women Workers in Multinational Factories” in Women, Gender and Development, Reader, London: Zed Books

Sparr, Pamela (1994) Mortgaging Women’s Lives, London & New Jersey: Zed Books

Szinovacz, M.E. (1984) “Changing Family Roles and Interactions” in Beth B. Hess and Marion B. Sussman (eds.) Women and the Family, New York: The Hawarth Press

UNICEF (1987) The Invisible Adjustment: Poor Women and the Economic Crisis, Santiago

Voice of Women (1983) Women Workers in the Free Trade Zone in Sri Lanka, Colombo

Weerasinghe, Rohini (1989) “Women Workers in the Katunayake Investment Zone of Sri Lanka” in Women and Development In South Asia, New Delhi: Macmillan, India Ltd.

Wolf, Diane L. (1997) “Daughters, Decisions and Domination. An Empirical and Conceptual Critique of Household Strategies” in N. Visvanathan (ed.) The Women, Gender and Development, Reader, London: Zed Books

Family Structure, Women’s Education
and Work: Re-examining the High
Status of Women in Kerala




Kerala has drawn considerable attention in recent years due to its paradoxical pattern of growth, characterized by high social achievements on a weak economic base, often referred to as the ‘Kerala model of development’.2 The dramatic decline in fertility since the seventies and the process of demographic transition in the state is one such achievement. Attempts to understand the determinants of fertility decline in Kerala, as also in other regions

1We would like to thank Vanita Nayak Mukherjee and K.Saradamoni for helpful comments on an earlier draft of this paper. N. Meera for research assistance.

2A comprehensive case study of Kerala’s development experience (UN 1975) highlighting the fertility decline since the early seventies, drew a more general and significant policy inference for countries in the early stages of industrialization: “It could be that the more proximate factors responsible for decline in fertility in the now advanced countries were educational and health developments which happened to take place together with increases in per capita incomes, industrialization and urbanization. Much the same results could be achieved in less industrialized societies by giving higher priority and precedence to the development of these services.” While this study itself did not project this as a Kerala model, pointing out in fact that such development should also lead fairly rapidly to all-round advances in agriculture and manufacturing industry, it has come to be popularly described as the Kerala model of development (see Raj 1994).

of the developing world, yielded a strong negative association between female literacy and fertility rates. ‘Women’, during this time (the ’70s), were emerging as a recognized constituency in the development effort and this relationship helped in strengthening the conceptual links between women’s issues and economic development (Kabeer 1999). Literacy, together with non-domestic employment, which gave women access to independent sources of income, came to be regarded as important components of women’s ‘status’, which affected fertility and mortality outcomes (Mason 1985). Since Kerala women have on average, been among the most literate compared to women in other states of India (though the same can not be said of female work-participation rates), much was written about the ‘high status’ of women in Kerala (see Table 1) and their central role, historically, in social development (Jeffery, 1992). Later research questioned this straightforward relationship between girls’ education and fertility, emphasizing the need to focus on the social context within which women make decisions (see Heward and Bunwaree 1999). For instance, how is a macro outcome like the decline in fertility negotiated at the micro level of the family, riddled by gender differentiated authority, roles and responsibilities, giving younger women, particularly, very little power to make decisions?

The growing uneasiness with Kerala’s social development outcomes, with the rising visibility of gender-based violence, particularly domestic violence, mental ill-health manifested increasingly as suicide, and the rapid growth and spread of dowry and related crimes, reinforced the need to study family structures and practices. It might be suggested that the growing dowry demands and the visibility of domestic violence, indicated that women had internalized subordination, sustaining inequitous gender relations within the household. We attempt to understand the contradictions in social development in contemporary Kerala by focusing on families (used interchangably with households) as microcosmic sites of such contradictions, where unequal relations of power between men and women are shaped. Changes in the structure and practices of families in Kerala in the past century, themselves shaped by wider social processes, had wide-ranging implications for gender relations. We will attempt to show that the dominant persuasion of families today, particularly in terms of their role in regulating access to material and social resources is patrifocal,


Development Indicators: Kerala and India


Source: Parayil, Govindan (ed.) 2000.

(one that gives precedence to men over women).3 While on the one hand alterations in marriage, inheritance and succession practices, changing dramatically the practices of erstwhile matrilineal groups, have weakened women’s access to and control over inherited resources, on the other, the changing levels of female employment and persistence of a gendered work structure have limited their claims to “self-acquired” or independent sources of wealth. At the same time, it is in the norms of masculinity and femininity taking

3This is a term used by Mukhopadhyay and Seymour (1994). They distinguish “patrifocal” from “patriarchy” which tends to imply a monolithic system in which males always predominate in all settings and socio-economic contexts and at all stages of the life-cycle. The patrifocal concept is more “flexible” and adaptable creatively to both internal/external pressures for change.

shape in the context of emerging consumer practices, that the details of a patrifocal ideology are being consolidated and reinforced. Importantly, this consolidation of ‘conservative’ change also invests men with forms of control over property/resources and over women’s sexuality.

This essay is divided into five sections. In Section 1 we discuss the concepts of patrifocality, bargaining power and fallback position, status, autonomy, agency and empowerment, used to understand gender relations and discern the need to focus on families and their social contexts. Changing family structures and practices and their implications for gender relations are analysed in Section 2. The indications emerging from the non-conventional indicators, domestic violence and mental distress are presented in Section 3. In Section 4 we explore the less known aspects of the relationship between literacy/education, conventional indicators of the high status of women in Kerala, and employment linked to patrifocal family practices. We conclude with the suggestion that education and employment have not played the transformative role so generally expected of them, a ‘discontinuity’ that is shaped by their mediation by patrifocal families on the one hand and that is evident in non-conventional indicators on the other. The study is based on secondary, published material and is exploratory in nature and underlines the need for more research. It also draws insights from ‘reconstructing the past’ through life histories of 15 ‘ordinary’ women, that is, women not marked by any level of social visibility in the state. We have categorized these respondents into three age groups: five were elderly (60 years and above), two were middle aged (in their 40s) and eight were young (roughly between 25 and 40).4

4The life histories were documented and compiled by Dr T.K. Anandi, a sociologist based in Kozhikode, with contributions from Sandhya Chandrasekharan, Geetha, and Bindu in Trichur; Deepa Shankar and N. Meera in Trivandrum. The names of our respondents have been changed. Not all of them were able to recall their age accurately. The elderly respondents, P. S. Nair, 72, and Uma Antarjanam, 76, had not worked outside the home and lived in middle class households; Mariamma, 60, (Christian) is a nurse; Karthiyayini amma, 65, (Nair) stopped working as a domestic help recently and lives with her daughter and husband; Pathumma, 65, (matrilineal Muslim), manages her household in Malappuram district, while her husband and three sons (of four sons and four daughters) run


Families regulate gender differentiated access to and control over resources, both material and social, including education, health and property rights and are an important arena where gender relations are structured. Mukhopadyay and Seymour (1994: 3) use the term patrifocal to understand a family that is in important aspects focussed on the interests of men and boys. ‘As in most intensive agricultural, socially stratified, state-level societies, there have evolved in India a set of predominant kinship and family structures and beliefs that give precedence to men over women — sons over daughters, fathers over mothers, husbands over wives, and so on’. These male oriented structures and beliefs, they argue, constitute a socio-cultural complex that profoundly affects women’s lives. The structural features of patrifocal families include patrifocal residence; patrilineal descent; patrilineal inheritance and succession, all of which emphasize the centrality of males to the continuity and well being of families. In association with this is the relative marginality of females centered in the expectation that upon marriage they will shift residence and affiliation to the family of their husband. Importantly, when speaking of Kerala, these structures were not so clearly given even as far as the early twentieth century. The different regions of Kerala sustained diverse forms of matrilineal and patrilineal families, that with the exception of the Nambudiris

a provision store in Mysore. The middle aged respondents Fathima, 40, (patrilineal Muslim) whose husband sells fish for a living doesn’t go out to work; Chandrikakumari, 43, is from a middle class Namboodiri household. Of the younger respondents only two are from middle class households, Kala, 33, (matrilineal Nair) employed in a government welfare scheme and Arifa, (matrilineal Muslim) in her thirties, unemployed and married to a Gulf migrant; Shanti, 35, (matrilineal Nair) recently joined a self employed women’s scheme and depends largely on the proceeds of coconut and cashew on 50 cents of land and livestock; Nafissa, 26, (patrilineal Muslim) works irregularly as a domestic help; Anitha, 28, (patrilineal Ezhava), is a wage labourer; Lakshmi, 38 (Ganaka, a backward caste) is a domestic help, Omana, 35, who converted recently to Christianity, works as a sweeper; Sindhu (patrilineal Hindu) in her twenties and the only never married woman among our respondents, works on a daily wage in a tile factory. (Life Histories).

(Brahmins, mostly patrilineal), clearly excluded some of the more extreme forms of discrimination against females.

Following Sen’s (1990) exposition of the household, it is suggested here that gender relations within the family are characterized both by cooperation and conflict and their hierarchical character is maintained or changed through a process of bargaining between men and women with differential access to and control over resources. We will also draw upon his analysis of the household in terms of bargaining power, fallback position and perceived interest response. Gender disparity is maintained in the household through the association of men with ‘productive’ work and the ‘outside’ and women with the ‘inside’ or ‘reproductive’ work. A member’s bargaining power is defined by a range of factors, in particular the strength of his/her fallback position (outside options which determine how well off he/she would be if cooperation ceased) and the degree to which his/her claim is seen to be legitimate. Since women’s perceived interest is so intimately linked to the family’s welfare, it could influence bargaining outcomes such as making a perceived interest choice, weakening their individual well-being.

Given such a social context, defining women’s ‘status’ in terms of schooling and labour force participation, is at best partial. No doubt the two are important variables as potential sources of ‘autonomy’ for women. Together they enhance choice and opportunity in women’s lives; provide an independent source of income, strengthen the fallback position, the perception of individual interest and help to raise their perceived contribution to the household. However, indications are that education alone does not enable women to challenge gender relations; much depends on channeling education towards engendering critical attitudes. Crucially, domestic violence and dowry deaths went alongside rising levels of education. Nor did work by itself ensure women’s control over earnings or their ability to take ‘self interested’ decisions. Importantly, ‘status’ here, was not necessarily distinguished from women’s position in society reflecting the values of the community and evoking some idea of esteem (Dyson and Moore, 1983; Mason, 1985; Mason, 1993; Jeffrey and Basu, 1994). In this sense, it could also go against women’s ability to make independent choices, against societal values (Kabeer, 1999). There may sometimes be a strong rationale for women to make choices which are

disempowering. Deeply entrenched social rules, norms and practices, which shape social relations within households as one arena, influence behaviour and shape choice (Kabeer, 1999).

Clearly, there was a need to focus more broadly on socio-cultural institutions such as family and kinship, which regulate gender relations. Sociological studies using kinship systems across India as proxies for autonomy, found that south India, including Kerala, represented ‘greater freedom for women’ (Karve, 1953) or greater female autonomy defined as ‘the ability to manipulate one’s personal environment’ (Dyson and Moore, 1983). The prevalence of matrilineal kinship among sections of the population in Kerala with its patterns of inheritance, marriage and post-marital residence seemed to indicate greater decision-making power for women vis-a-vis women in patrilineal families of north India. It appeared therefore that Kerala women enjoyed not only a high status but also a more egalitarian gender regime. However, studies, which use access to land (women’s rights to land under specific kinship systems) as a measure of autonomy or empowerment, reflected a simplistic relationship (Kabeer, 1999). It was seldom demonstrated how such access translated into actual control and hence the power to make decisions. Moreover, that autonomy, not easily measurable, could be severally constituted and indications that more direct measures of autonomy could yield different results, were being thrown up (Visaria, 1996; Rajan et al., 1996). Visaria’s (1996) measure of economic autonomy in terms of women’s access to and control over household income, suggested that women in Gujarat had higher levels of autonomy than those in Kerala despite much lower levels of literacy. The recent National Family Health Survey, 1998-99 (IIPS 2000), which incorporated measures of autonomy for ever-married women for the first time, also revealed that Kerala trailed Gujarat in terms of all the measures of autonomy — household decision-making, freedom of movement and access to money. However, these studies relied on understanding empowerment through a number of questions on decision-making by women, some strategic to their lives and some not so relevant. Such ‘statistical’ perspectives on decision-making should be taken for what they are: simple windows on complex realities, revealing very little about the subtle negotiations that go on between men and women (Kabeer, 1999). In fact gender empowerment measures (GEM), like the UNDP’s measure (1995) and the alternative

measures developed by Hirway and Mahadevia (1996) for the Indian states, which did not include the household dimension due to non-availability of data on gender inequality within the household, continued to place Kerala women at or near the top.

The concepts of ‘autonomy’ and ‘empowerment’ were expected, nevertheless, to shed critical insights on power relations (Batliwala, 1994, Kabeer, 1999). Differential access to and control over resources, reflecting constraints imposed by the extant social order, deprives women of the ability to make decisions/choices and exercise agency. Agency or the ability to define one’s goals and act upon them tends to be operationalized as decision-making (Kabeer 1999), but it is important to note that it may take other forms like bargaining, negotiation or manipulation or the more intangible cognitive process of reflection. Empowerment alludes to a recognition by women of the ideology that legitimizes and sustains male domination (Batliwala 1994). However, if rights in property customarily enjoyed by women get eroded over time and women’s work expands only slowly or into selective and relatively lower paying occupations, their control over resources is weakened and so also their relative position within the household.

Changing Family Structures and Practices

In the last few decades there has been a gradual but concerted shift in the understanding of women’s property rights in Kerala. The shift, which has involved the major social groups in the state — patrilineal and matrilineal — has been towards a convergence on property practices usually associated with patrilineal forms of families. Conventionally and very generally women’s property rights in patrilineal societies tended to be organized around marriage, in a range of practices including the transfer of women and change in their kin identity, residence, dowry, exchange of gifts, and obligations at childbirth. These practices framed the denial virtually of inheritance rights for women. Though not uniformly, this combination of practices went along with a preference to transfer to women movable rather than immovable property. This is perhaps most sharply delineated in north and north western India where village exogamy was observed and close kin marriages prohibited. In sharp contrast to this picture, women’s property rights in

matrilineal societies in Kerala were clearly delinked from marriage, emerging instead from a birthright in the family property not unlike (though distinct from) the manner of males in the Hindu Undivided Family. However it is notable that the convergence on a set of practices usually associated with patrilineal groups has involved changes in women’s property rights among the dominant patrilineal groups as well. That is to take one instance, if dowry (otherwise associated with patrilineal groups) is becoming a very general practice, it has failed to retain its characteristics (linked to social sanction and regulation) of the mid-twentieth century even among the patrilineal groups. It is in this sense of a convergence of practices that we speak of the dominant persuasion of families in Kerala today being patrifocal. However, this section is limited to examining the basis of changes in gender and property relations in families through alterations, broadly, in the organization of marriage and linked to it, of the contours of masculinity and femininity, in the latter part of the twentieth century. Though it is conceivable that family practices have seen important changes across social groups, this section will focus on the major matrilineal social groups, the Nairs and Ezhavas5 and the major patrilineal groups, the Christians and Muslims. This choice is dictated as much by availability/scarcity of material as the visibility of these groups.

Social and legal reform

Processes of reform of matrilineal and patrilineal families leading to legislation in the first half of the twentieth century established the basis of patrifocal families in Kerala.6 Until 1986 when the Supreme

5The Nairs are a forward and the Ezhavas a backward caste and both were conventionally associated with land holding and cultivation. Besides the Ezhavas were associated, widely in colonial narratives, with toddy tapping as an occupation.

6As many as 20 legislations were enacted between 1896 and 1976 which marked the gradual revocation of a legal framework for matriliny. The processes of reform however were of two kinds — (a) legislative reform and (b) reform of caste/social practices involving local level mobilization and campaigns — and were crucial in terms of their implications for gender relations within the family. Legislation in the early twentieth century also set forth property rights among patrilineal social groups, the Christians and the Nambudiris.

Court held that the Christians of Travancore and Cochin were to be governed by the provisions of the Indian Succession Act, 1925, they were governed by highly gender discriminatory laws, which came into force in the early twentieth century.7 In fact, the Travancore Christian Succession Act, 1916 has been described as an outcome of the expression of fear and anxiety on the part of the Christian community over certain decisions by the courts in Travancore applying the British Indian law for Christians to adjudicate on the rights of widows. The denial of women’s rights to property rested on ‘fears’ of domestic disharmony and ruin arising from frequent litigation and fragmentation of property.8 Under the Travancore Act, women were eligible to receive one fourth the son’s share or five thousand rupees whichever was less as stridhanam (dowry) and did not inherit paternal property. The Indian Succession Act, 1925 does not discriminate between the sexes in matters of intestate succession.

A series of legislations in the early twentieth century introduced measures ‘recognizing’ the conjugal family (as against the matrilineal family which did not centre conjugality) and defining relations of protection and dependence between husband and wife and father and children, facilitated a patrifocal family among the matrilineal Hindu groups. Hindu women now had individual rights over their share of family property but this right was achieved within a legal framework of dependence on men as husbands. The Hindu matrilineal social groups today come within the ambit of the Hindu civil code, with some special provisions such as their exclusion from the mitakshara (Hindu Undivided Family) coparcenary. The Kerala Joint Hindu Family (Abolition) Act, 1976 eliminated the legal conception of joint family property among the Hindus by replacing joint tenancies with ’tenancies in common’ as if partition had taken place among members on a per capita basis.

7The state of Kerala was constituted in 1956 of territories from the princely states of Travancore, Cochin and the erstwhile British district of Malabar.

8“If daughters are given share along with their brothers and the widow is allowed to have any claim whatever, except maintenance in the property of their husband, it would destroy the domestic tranquility, throw open the flood-gates of litigation, bringing all sorts of calamities and eventually ruin the community” (Tharakan, 1997: 125 citing ‘Original Appendix No. 1’, Dissenting Minute to the Christian Committee, c 1916).

Legal changes went alongside ritual and material reorganization of marriage. Marriage was streamlined and consolidated in one rite, where earlier there were two, that established its sacred aspect.9 It is not without import that marriages among the matrilineal groups today emphasize the tali rite (rite investing women with a small gold ornament worn around the neck and considered symbolic of marriage) and considerable thought has gone into the fashioning of the tali (the Ezhavas have also incorporated the caste Hindu rite of kanyadanam which symbolizes the gift of the virgin) (Osella and Osella, 2000: 105).

In phases Islamic law was made to apply to descent of property among the Mappillas of north Malabar. The Mappilla Marumakkatayam Act, of the late 1930s provided individual rights to partition of the taravad (matrilineal joint family) and a Kerala amendment in 1963 brought the share of any member of a Mappilla taravad under the purview of Islamic law, while also substituting Muslim for Mappilla (Agarwal, 1994: 233). Muslim women’s inheritance rights among the patrilineal groups of south Malabar, Cochin and Travancore were generally on the lines of Islamic law, moderated by local custom as anywhere else in India.10

Matriliny and women’s rights

A question that inevitably arises in the context of discussion of matriliny is whether the system allowed women effective rights (as in control) to property. Matriliny under colonial law was unabashedly patriarchal, investing the right to manage and regulate property in the senior male. This has led to the easy dismissal of matriliny as affording little by way of effective property rights to

9Among the Nairs and Tiyas and Ezhavas, it was in a pre-pubertal rite called the tali kettu kalyanam that the tali was tied. While this rite was in several aspects symbolic of marriage, it did not lead to cohabitation. After attaining puberty women could enter into sambandham (one of several terms used for conjugal relations) following customary ceremony. The tali rite was the more elaborate and expensive one as also the rite targeted for elimination by social reformers. Gradually the ‘tying of the tali’ came to be observed as part of one consolidated marriage rite, and to be invested with the sense of ‘wifehood’.

10A large majority of Muslims in Kerala are Sunni and are governed by the provisions of the Hanafi school of Islamic law.

women. In this context, it is imperative to ask whether property rights could ever be complete without descent of property and lineage. There has also been no engagement with the cleavage or ‘tension’ emerging from vesting descent/lineage in the female members of the taravad and significant management rights over property in the male members. Even at the risk of a degree of generalization, it needs to be stated that property rights, characterized by descent/lineage on the one hand and managerial powers on the other, flowing among different sets of people — senior and junior male members and mothers and daughters respectively — should not be confused with patrilineal forms of family in terms of their implications for women. The implications of the cleavage or ‘tension’ for the distribution of authority were completely ignored in colonial law, more concerned with identifying and enforcing a rigid set of rules. Besides, the tendency to dismiss matriliny, on the grounds that authority was gendered in favour of males, masks from view the less dramatic but equally important gains resting in greater sexual choice, positive attitudes towards girl children, and social support and security from residence in the natal home. Positive factors emerging from matriliny may be grouped into three: (a) emerging from differences in regulation of women’s sexuality, (b) from positive attitudes towards girl children, and (c) in terms of kinship (social) identity.

Greater sexual choice If matrilineal societies in Kerala had come in for shocked, surprised and even exaggerated comment historically for affording an unusual degree of sexual freedom to women (Fuller, 1976), its distinct organization of marriage and property made for very real differences in social attitudes towards women. It has been pointed out that Nair and other matrilineal women seemed to have had greater space for making decisions on marriage and sexual relations (Menon, 1996, Ramachandran, 1997: 279). However pressure could be brought to bear upon women to establish or continue marriages that were beneficial to the taravad or to discontinue those that were frowned upon (Puthenkulam, 1976, Gough 1961, 1993). This was evident from one of our elderly Nair respondents who felt forced to enter into a sambandham at the age of 15 but was able subsequently to get a divorce. “After four years, I took him and went to the Registrar’s office and nullified the marriage”. Given this constraint, the institution of marriage provided women with greater security. Notably, women’s ability to

walk out of a marriage was shaped by their permanent and uncontested right to subsistence in their natal home. Besides, women could remarry on termination of a prior connection or on the death of their husbands.11 Importantly, in comparison with patrilineal societies the oppressive edge of widowhood was absent. An absence of the more stringent forms of sexual control made possible greater mobility and greater exposure of girls to locally available forms of literacy among women of matrilineal castes (Gough, 1961).

Security of girl children The birth of a girl was a welcome event in matrilineal families. This eliminated at least the more extreme forms of discrimination (Alexander, 2000, Jeffery, 1992). It also made available to girls even in the mid-nineteenth century, a level of education that was not to be found elsewhere (Jeffery 1992). However in recent years there are indications that the earlier advantages, reflected for instance in Kerala’s favourable female sex ratio, cannot be taken for granted anymore (Rajan, 2000).

Familial (kinship) identities Inheritance and lineage were through women, which underlined their importance to family or kin identity. Hence women were members of their property group by survivorship, their maintenance and residence rights in their natal home were achieved directly, i.e., they were not mediated by marriage or derived from their husbands. More importantly, these rights marked a sense of continuity and security rather than rupture and vulnerability. It followed that senior women, particularly of competence, in the bigger taravads had an important role in making and/or influencing decisions regarding the household and property (Gough, 1961).

On the other hand, among patrilineal groups marriage (expressed through several ritual observations) marked the severing of a woman’s ties with the natal house/family.12 Among the Christians

11Menon (1996) gives more recent instances of women who chose break off their marriages and to live with men of their ‘choice’.

12Even as post marital residence and property transfers are central indicators in a material sense of a woman’s disinheritance and fractured identity; their symbolic expression is to be found in some of the key rituals of the marriage service (Visvanathan, 1993: 108). Visvanathan’s study is based on ethnographic work in

of Kerala, this severing of ties (and sense of fractured identity) is expressed in the payment of stridhanam, affiliation to husband’s family and church, and the specific character of her incorporation in the husband’s family. That is, upon marriage though a woman’s affiliation is to her husband’s family, she is not ever ‘fully’ incorporated, by which we mean that she is not incorporated as an individual with rights equal to her husband. Rather she is incorporated as a wife, which is to say that she does not even have control over her stridhanam, not to speak of a substantial right in her husband’s family property. A wife’s rights are restricted to the right of maintenance from her husband’s estate (Visvanthan, 1993). A separated woman had no place in Syrian Christian society (ibid, 112, Roy, 1999: 210, 213).

Family practices

Changes in the organization of marriage provide indications of changes in gender relations and erosion of property rights. Particularly since the 1970s these changes have been shaped in important ways by consumer practices and identities. With outmigration, specially to the Gulf countries and the inflow of remittances, Kerala has been in the vanguard of consumer trends (Osella and Osella, 2000, Kurien, 1994).13 Osella and Osella (1999: 992) argue that the characterization of consumption as an empowering and ultimately egalitarian act is severely limited. In our

the village of Puthenangadi, known for its ‘old Christian families’ near Kottayam in former Travancore.

13It has been estimated that there were around 1.7 million Indian migrants in the middle eastern region in 1983, more than 50 per cent of whom were in all likelihood from Kerala. (P R Gopinathan Nair cited in Kurien, 1994) Importantly, the turn-over of migrants is quite rapid and the number of people who have been involved over a period of time is much higher. Besides, in the state, Gulf migrants are concentrated in certain pockets reflected in the fact that remittances contributed up to 50 per cent of the gross domestic product of high migrant districts (ESCAP, 1987: 74 in Kurien, 1994). While the ripple effect on spending in non migrant households (communities) is quite evident in villages with large numbers of outmigrants and is documented, Kurien (1994) suggests that consumption patterns of non migrants (earning comparable income) in villages with little or no migration tended to be different from that of migrants.

context they point to caste status and income levels as constraints; that lower castes cannot hope to attain an entirely new identity by adopting new consumption patterns and that income levels exclude participation in specific kinds of consumption. Hence inequality is built into the politics of consumption. However, inequality can also be transmitted through the very act of ‘enforced’ consumption as against the process of exclusion. Gender-based inequality in the context of marriage practices emerges not by excluding women but by norming specific kinds of consumption — jewellery, consumer durables most visibly. By generating expectations on a wide scale through lavish marriages, expensive jewellery and large dowries, women are objectified in dangerous ways.

Marriage, female roles and family status Two recent studies focusing on consumer practices in the context of ‘Gulf migration’ found that marriage formed a priority item of spending of remittances (Ibid, 1999, 2000, Kurien, 1994).14 ‘Marriages are occasions for dramatic staging of public performances of a family’s wealth, status and style They were also occasions when consumer goods change hands’ (Osella and Osella, 1999). This emphasis on marriage, as a consumer practice that provides access to social mobility, has had two kinds of effects. One, it effects changes in the idiom of marriage, both in terms of celebration — through ritual and feasting — and in terms of organization that is, preference to avoid pre-existing practices of matrilineal social groups such as village endogamy and cross cousin marriages. Linked to sustaining ties that already exist by marriage, these practices helped keep marriages at close distances, giving women and men constant recourse to their natal kin; called for less formality and expense (Aiyappan cited in Puthenkulam, 1977). It is significant then that cross-cousin marriages do not find favour anymore. Puthenkulam (1977: 93) also notes that the growing practice of demanding a dowry at marriage has led to a decline in cross-cousin marriages ‘as it is delicate to demand or receive a dowry from the mother’s brother’. If Puthenkulam (1977) records this trend among the Nairs, Osella and Osella (1999: 1010) note a ‘recent wholesale

14Osella and Osella (1999, 2000) are concerned focally with the Ezhavas, a backward caste, and their data is based on study of a village in the Kuttanad area of south Kerala. Kurien (1994) has studied three villages, a predominantly Muslim village in north Kerala, Christian in central Kerala and Ezhava in south Kerala.

disdain for village endogamy, and an increasing preference for marriage conducted outside the village’. Large dowries, dominant norms of femininity and extravagant celebration, which characterize the preferred forms of marriage make them inaccessible to poorer families (ibid, 2000: 89-97).

Two, marriage, raised to the level of a social imperative, mediates gendered interest and identity. Notably, families ‘husband’ scarce resources so as to achieve the best possible marriage for women, hence ‘family decisions’ regarding education and employment of women are specifically targeted towards marriage. Here dominant norms of femininity dictate that women use their education in the interests of marriage — as accomplished wives and better mothers (ibid, 2000: 41-46). In contrast while marriage is not unimportant for men, masculinity is not centered on it as it is on work. When a man goes to the Gulf it is to work, earn and gain social mobility, even if Gulf migrants are reckoned as preferred bridegrooms. One of our respondents indicated the dimensions that ‘masculinity’ could take in this context. She was considered old at the time of her second marriage at 23 years to a Gulf migrant, who already had a wife and three children. The marriage was arranged at his behest, because he thought he could ‘afford’ it. He told her that, ‘earlier, I never had money to buy even a beedi. But now I am in the Gulf and have got money. I can take care of two families’.

Post-marital residence of women In Kerala, perhaps more than anywhere else, changes in family practices are underwritten by comprehensive change in family structure across the spectrum of communities. Kerala has a very low incidence of joint families.15 It is in this context that we have to understand changes in post marital residence over the last half century. The general trend is towards adoption of nucleated residence (where a married couple sets up a household). We will denote this form of residence as patrifocal, informed as it is by gendered power relations. Mencher (1965)

15Kolenda (1987) estimated from Census 1961 data for all states and territories from a 20 per cent sample of households, that Kerala’s 9 districts came within the 62 districts showing the lowest incidence of joint families. Kerala was also a state with a low proportion of married sons and/or other married relatives in the sample households. Kolenda also points out that higher proportions of joint family could be associated with land ownership and forms of kinship and that most of the joint families in Kerala were of the Nairs.

found in her south Malabar village that 50 per cent of Nair households were small matrilineal branches (an average of 3.5 persons per household), 15 per cent were nuclear family units living in houses that were received by women from their taravad, 15 per cent lived in houses established by men and women after marriage and 20 per cent comprised an assortment of related persons. She noted that there was a great deal of flexibility in residence patterns among Nairs, whether poor or wealthy, in the villages. From a sample survey of 403 matrilineal households across Kerala, Puthenkulam (1977: 107) found that the general pattern was patrifocal residence (where a married couple moves to a new house).16 However this pattern was more dominant in south Kerala (85.7 per cent) followed by central Kerala (61.8 per cent) and north Kerala (59.7 per cent). The largest proportion of matrilocal residence was in central Kerala (29 per cent) and residence in either the natal or husband’s home while waiting to move to patrifocality was highest in north Kerala (16.9 per cent).

A casual comment made by Puthenkulam (ibid: 109) is insightful regarding gendered power relations and the division of roles. ‘[T]he common residence pattern now is generally virilocal. Today no self-respecting person [male] attaches himself to his wife’s house and lives on her wealth like a drone. Such husbands are derisively described as “Koil Thampurans”, the consorts of royal ladies who lived by their wives. There are cases however of the husband shifting to the wife’s home to assume the management of her and the children’s property’ (emphasis ours). The dice, it would seem, is heavily loaded against the woman-subject! Yet, rather than resorting to extreme characterization Puthenkulam’s (109) findings indicate a state of continued mediation of earlier forms of matriliny with the contemporary emphasis on conjugal residence. It is important here that the natal home continues to provide refuge and security greatly to women in north and central and considerably in south Kerala. On the death of the husband, more than 90 per cent (of 398 respondents) of wives

16Puthenkulam (1977) uses virilocal. We have preferred to use patrifocal for two reasons — to prevent confusing this with our earlier use of virilocal to mean ‘residence in the matrilineal home of the husband’ and because in the post reforms period (social reform and land reforms of the 1960s) the male-headed conjugal household has gained visibility.

returned to their natal homes in north and central Kerala and about 50 per cent in south Kerala. In south and central Kerala more than 70 per cent and in north Kerala more than 50 per cent of widows lived with married sons.

Inheritance rights Mid-twentieth century, Gough (1952) found that the taravad (matrilineal joint family) houses were inherited matrilineally but sons and daughters inherited other property. In Fuller’s (1976) study village in central Travancore, a distinction continued to be made between (i) taravad land inherited matrilineally, the alienation of which required the consent of all adult matrilineal descendents of the person holding it and (ii) separate land which was freely alienable. Recent research in central Travancore has shown that women continue to inherit a house but are less likely to receive agricultural land (Osella and Osella cited in Agarwal 1994: 177). More importantly, Osella and Osella (2000: 106), while not commenting directly on the inheritance rights of women, note that the transfer of their share of land is recorded in the community register of the Nair Service Society (NSS) and Sree Narayana Dharma Paripalana Yogam (SNDP) for the Nairs and Ezhavas respectively, suggesting that inheritance has been replaced in a substantial way by transfer at marriage or dowry. It is also significant that this land is often sold and the cash equivalent given to the husband and that the dowry is not usually under the control of the girl. ‘While a newly wed bride living with her husband and his relatives is in no position to refuse to relinquish control over her dowry, her contribution may give her some leverage in the family’ (Osella and Osella, 2000: 102). It is possibly an effect of cumulative change that they (Ibid, 101) note that many women no longer have land to pass on to their daughters and mother-daughter inheritance is becoming rare.

Before the 1986 decision of the Supreme Court on Syrian Christian inheritance, women’s rights to paternal property were exhausted by the stridhanam. Given the evidence of resistance from entrenched social interests, church and community, there does not seem to have been a dramatic departure from existing custom, (Agarwal, 1994, Roy, 1999) though high rates of stridhanam go alongside the exercise of testation to safeguard patrimonial interests (Roy, 1999, Visvanathan, 1993: 146).

Stridhanam, dowry or disinheritance of women? An important indicator of the direction of change of women’s property rights is the very general visibility of dowry and the signs of its growing presence, including among groups that did not conventionally observe dowry. Yet empirical evidence of this is limited to a few micro studies. The practice of giving stridhanam at the marriage of a girl was customary among the patrilineal communities — the Christians, Muslims, Ezhavas and Nambudiris — and has been recorded among specific matrilineal groups as well — the Tiyas, Mappillas, and Ezhavas (Gough 1961). Importantly in the case of the latter, dowry did not exhaust women’s inheritance rights. However the custom varied widely among different groups and regions. Given this, village studies indicate that, across the spectrum of communities, the customary understanding of these practices is giving way to what Visvanathan (1993) terms a more ‘market’ approach. Among the matrilineal Hindu groups, there has been over the past century a very general shift to dowried virilocal monogamous marriages, conventional of the Christians (Osella and Osella, 2000: 85). Puthenkulam’s (1977: 104) survey recorded a fairly even presence of dowry in north (32.7 per cent), central (29.8 per cent) and south (24 per cent) Kerala among the matrilineal groups, which is somewhat contrary to the prevailing view that dowry is less prevalent in north Kerala.

To get a sense of change, customary regulation is fast giving way to ‘competition’. An agreement on the stridhanam ’due’ to the husband and/or his family was an essential part of the conventional arrangement of marriage among the Christians. Customary regulation was evident in its ‘public’ character. At a ritual event before the marriage, the amount of stridhanam was announced publicly among other details of the marriage. Also the marriage was solemnized only after the girl’s family gifts four per cent of the stridhanam amount to their church and a share to the bridegroom’s church (Visvanathan, 1993: 112).17 If stridhanam was understood

17If Visvanathan’s informants pointed out that it was not unusual for people to ‘trick’ the church by stating a much smaller amount than is actually paid, Kurien (1994) notes complaints by the church that with the influx of Gulf money people have tended to turn away from it. Could this be an indication of the ‘tightness’ of the marriage market, the need to concentrate all resources on ‘getting the best’ even at the risk of former community ties?

as a woman’s share of her father’s property, indications are that it is lending itself to a process of disinheritance of women. Visvanathan (1993: 111) argues that its manipulative aspect has become dominant empirically and money is used to contract marriages with desirable families. Hence, the resemblance to a form of groom price in that a) only on payment of stridhanam is it possible to agree on a marriage, b) the money or property that changes hands is not controlled by the woman but by her husband and/or his kin, prominently his father, c) consequently a woman has only a right to maintenance in her conjugal home (Ibid: 113). The need to pay stridhanam is frequently a financial strain on a girl’s parents leading sometimes, particularly in middle class families, to sale of property and the pressure is such that the stridhanam sometimes (though not usually) exceeds what the son/s receive. The rate of stridhanam varied according to socio-cultural factors including educational qualifications and the employment status of men and women, and factors of considerable importance such as a woman’s complexion and ‘beauty’ (Visvanathan 1993: 111, Osella and Osella 2000: 101).18

Kurien (1994) found that dowry was a major head of expenditure in two of three villages, studied in the context of migration-induced spending. In a Muslim village, where migrants were from the lower income groups, she finds that ‘the value placed on the purity and seclusion of women manifested itself in several ways in the expenditure patterns of this area’. This had led to a tremendous increase in dowry rates as well as the use of taxis, considered the more appropriate mode of travel for women. In the second, an Ezhava dominated village, while dowry is not mentioned, the major heads of expenditure were life cycle rituals and festivals. ‘Marriages were the biggest of such celebrations and migrants spent a good proportion of their Gulf money on the weddings of their sisters, daughters and close relatives’. In a relatively affluent Christian village, the largest heads of expenditure were education (donations to professional colleges) and dowries. ‘Status in this community accrued from having a large bank balance, professionally educated family members (the large dowries were often ways of securing such

18Notably dowry and education (or employment) are posed as competing demands for cash within families (Morrison, 1997).

sons-in-law) . . .’. Our elderly Christian respondent revealed that at her marriage in 1970 (between a nurse and a government employee) her dowry was Rs 3,000 ‘a big amount then’. More recently she paid Rs 3 lakhs as her elder daughter’s dowry.19

There are indications of significant escalation in dowry rates in the state. Dowries for mid-status middle class marriages (as between children of Ezhava primary school teachers or local factory workers) were up to Rs 200,000 in 1996 and rising. Besides, dowries include a combination of cash, gold, land and consumer durables. A high prestige, high wealth dowry could include up to 101 sovereigns of gold, a preferred form of dowry (Osella and Osella 2000: 101). Among the Ezhavas, the bulk of the dowry consisted of land given by the father, cash and gold. Though some notional distinction was made between land and gold to remain in the bride’s name and cash and goods going to the husband and his family, in practice most women lost control over the entire dowry, which is used to support the needs of the husband’s family. Osella and Osella (2000: 106) seem to suggest that the Nairs and Christians share these practices, for the Ezhavas are described as adopting the formers’ practices in a bid towards upward mobility. One point of difference cited, however, is that of public registration of the dowry paid and cash gifts received at the marriage of a girl with the SNDP. Ezhavas consider this important to guard against loss in case of a break up of the marriage while Nairs and Christians are noted to consider this shameful. Hence a Nair bride has proof of only her share of land via the NSS register.

It is important to note that property transfers are being made increasingly at the time of marriage of a girl and that there is an element of force associated with such demands. Two of our younger respondents, who are Nair, reported that dowry was not demanded or given at their marriages. However one of them mentions that her husband who later deserted her, would point out that she had not ‘brought’ anything, despite the fact that they lived in her natal home. While expressing herself against dowry, she added that she would be forced to pay dowry for her girls, if demanded, for fear

19Mariamma views dowry as a woman’s rightful share, “because we won’t give them money anymore”. Besides, “I have given my land and house to my son. I am not partial. She can come to my house at any time but she is not supposed to show any authority over my house”. (Life Histories)

that otherwise they may be harassed. The unmarried respondent pointed out that dowry was a factor blocking her marriage — very recently a construction worker and wage labourer like herself demanded Rs 50,000 and 50 sovereigns of gold as dowry for his marriage.20 Among the matrilineal Muslims too there seems to be a gradual shift towards dowry. Our younger respondent from this group pointed out that while dowry was not paid at her marriage, they may not be able to stop their son (now 13) from taking it, ‘because they are the new generation’.21 It was clear from the patrilineal Muslim respondents that they did not have control over dowry, specially cash and sometimes gold and land.22

Non Conventional Indicators: Violence and Mental Distress

Analysis of reported crime in the state shows a four-fold increase between 1991 and 1996, of which growth in rape and domestic violence was the highest (National Crime Records Bureau, various issues). On the basis of the 1995-1997 average, it is seen that in the ascending order of crime, Kerala was ranked 25th among the Indian states in molestation and domestic violence, which includes dowry-related crime, 18th in sexual harassment and 10th in terms of dowry deaths (Mukherjee et al., 1999). While it must be remembered that dowry deaths were little heard of even in the recent past, it is possible that some dowry deaths were disguised as suicides. In a micro level study of 133 survivors of attempted suicides in 1994-95, more than half were women (Jayasree, 1997).23

20Sindhu pointed out that in the past it was possible for girls to marry without dowry. Expressing herself against the practice, she added that she would receive 10 cents of land and 5 sovereigns as her share of property. (Life Histories)

21Arifa like the elderly matrilineal Muslim respondent, Pathumma, received mehr, (given to the wife by the husband at marriage). (Life Histories)

22Fathima points out that she was unaware that her husband had used her dowry to renovate his parent’s house which was to go to his sister. She also had to give some of her gold to her husband’s sisters at their marriages. Nafissa’s parents had paid Rs 10,000 and 8 sovereigns of gold to her first husband at their marriage about 10 years ago, which was not returned on divorce. (Life Histories)

23In general the incidence of completed suicides is higher for men and attempted suicides are higher for females (World Health Assembly Report, 1998, Vadakumcherry, 1994).

A major factor associated with suicide attempts by women was marital disharmony — 36 of 75 women. It was found also that one third of the women were suffering from domestic violence. Men, however, related economic problems followed by family problems. While harassment of women was not linked directly to dowry in this study, an autopsy study conducted in a district in Kerala revealed that more than one third of the women who had died had a dowry problem preceding suicide (Ibid).

A study of domestic violence undertaken between 1997–99 in seven sites (cities) in India, revealed that Thiruvananthapuram (Trivandrum) had the highest prevalence of overall violence (to some extent due to higher reported rates of husband’s infidelity). Levels of psychological violence were very high — over two thirds of the sample women in rural and a little lower in urban non-slums (INCLEN/ICRW, 2000). Key causes of violence were: perceived lapses in fulfilling household responsibilities, infidelity and alcoholism. ‘Disobedience’, including any act construed as disrespectful or disobedient or a challenge to male authority, has been seen as a critical precipitating factor (Batliwala, 1998, Jejeebhoy, 1998). Four of our younger and one middle aged respondents experienced violence from their husbands. As factors precipitating violence they cited suspicion of their fidelity, ‘disobedience’ and ‘disrespect’. 24

It is understood that the more severe mental disorders such as schizophrenia show no significant gender difference in prevalence. However there is a greater prevalence of the more common type of conditions such as depression, and anxiety among women across socioeconomic levels and in diverse societies (Sonpar and Kapur, 1999). Research seems to indicate the social (rather than entirely psychological) influences on common mental distress in women, in particular due to domestic problems, are quite strong. A study of cases brought before the Family Court in Trissur between 1995-98,

24In all these cases the husbands were also given to consuming alcohol. Fathima claimed that her husband had grown more violent since his bout of mental illness two years ago. Two younger respondents ended their first marriages on account of violence. Besides three of the younger respondents reported that their fathers used to abuse their mothers. Of these however only one experienced violence from her husband; another was not married. Our discussions with psychiatrists in Trivandrum, particularly Dr Suraraj Mani of the Mental Health Authority and Dr S. Jayaram, Mental Health Care, corroborated factors outlined.

indicates that petitions filed increased from 477 to 860 — almost two thirds were filed by women for divorce and maintenance induced by protracted marital disharmony. Through case studies the author suggests that women suffered from greater stress (James, 2000). Based on his experience with psychiatric patients in Trissur district, a doctor asserted that the most common cause of psychological stress among women who are educated is lack of employment and the roles they are expected to assume after marriage (cited in Halliburton, 1998). Hence despite data limitations, there are clear indications that violence and mental distress are growing to be a serious problem in Kerala, warranting social concern and intervention.

Women, Education and Employment

As was stated earlier, male-oriented structures and beliefs profoundly affect women’s access to education and educational achievements (Mukhopadhyay and Seymour, 1994). While the 19th century reform movements had established the case for educating girls, this also had to be done without jeopardizing the interests of the patrifocal family.25 Over time arguments had been built up for imparting higher education to girls but couched in ‘marriageability- enhancing’ terms; later this argument was augmented by the increasing value placed on their potential earning capability and contribution to the economic well-being of the patrifocal family. If this posed a challenge to gender differentiated roles within the family, the resilience of the social division of labour is evident in

25Research on the negotiation of the issue of western education for girls (specially at the secondary and college levels) in the late nineteeth century brings to the fore the ‘value’ question. This could inform the contemporary debate as much for its difference in terms of policy as the continuity it marks in terms of socialization of girls and familial perceptions. Writing of Bengal and Calcutta in particular, Kerkhoff (1998) points out that high schools at this point were not intended to bring about equity in society, gender, class or otherwise. Girls high schools were recognized as socializing institutions established to better equip adolescent middle class girls to changing demands of the colonial urban and patriarchal society. It was hoped then that these high schools would reproduce the ideological and cultural hierarchies of the metropolis on which colonial rule depended.

that women still (had to) do the cooking (Sen 1990) and shoulder principal responsibility of child care and other household chores.

In fact tensions between the value of formal education for women and the disruptive potential for the patrifocal family have influenced the very system of education. (Mukhopadhyay and Seymour, 1994). While certain historical specificities do mark out Kerala from the compulsions of a patrifocal family; broadly the changing strategies and practices of families in shaping educational achievements towards their ‘collective well-being’ remain. The high levels of female literacy in Kerala have been well documented. In relation to men too, the achievements of women have been laudatory: over the decadal periods since 1961, gender-based disparity (ratio of male literacy rate to female literacy rate) has narrowed rather sharply from 1.39 to 1.07 by 1991. Some of the more notable achievements have been the near universalization of primary education for both girls and boys, and the very low (compared to all-India) school drop-out rates for girls which in fact are higher for boys since at least the early seventies, at each level of school education (Ambili, 1996). However, what gets obscured in the context of the very high aggregate literacy levels in the state is that gender disparities (ratio of male/female percentages at each level) at higher than primary or middle levels of schooling existed and continue to exist in college education, particularly technical education. It is well-known that in the state, macro level pressures that increase the desirability of education for girls have been strong historically and have been maintained over time (Jeffrey, 1992). As argued above such pressures create tensions at the micro-level of the family. In the context of Kerala it has been argued that matriliny, in this respect, would have been less restraining since even prior to the 1860s, the period in which the centralized education systems were established, girls often attended local schools (Jeffrey, 1992). The fact that Malabar showed much higher rates of female literacy compared to the other districts of Madras Presidency, gave this added support. In contrast, in Namboodiri households in the 30s and 40s, girls did not attend schools but were taught at home. However, it appears that this freedom to girls for attending schools did not hold for higher than primary/middle levels of schooling even among the matrilineal households.

The attainment of puberty seems to have been a significant barrier to further education for girls cutting across family structures,

communities and socio-economic groups. A respondent from a middle class Nair family pointed out that (around the 1940s) her father tried to stop her education after middle school, relenting only due to her insistence. In this context her mother had argued that she (mother) had only studied up to fourth standard and ‘that was the norm for girls’.26 The experience of three Muslim women (40 years and below) suggests that girls study up to the primary level at least. In contrast, the elderly Muslim respondent reported that three of her four daughters were teachers, having studied entirely out of their own interest. Two younger respondents from poorer Hindu matrilineal families indicated that economic hardships were instrumental in pushing them out of school. For if education was ‘free’, it nevertheless involved certain costs, especially as higher schools tended to be at longer distances.27 Gender differentiated roles implied that girls often had to combine household chores, including care of younger siblings with schooling, while boys mostly ran outside errands. This constrains the educational options of girls more than boys.

In contemporary Kerala, even though the disparity (at higher than primary/middle levels of education) is extremely low up to the 10th standard, and in fact reversed to some extent at the intermediate/pre-University and non-technical diploma levels, considerable difference still exists at the college levels, particularly in the technical fields. What is more interesting is that even in the field of higher education, particularly at the intermediate, pre-university level, it is courses which will lead to ‘suitable’ professions for women, from the point of view of their familial roles/responsibilities, that have a larger intake of girls. In teaching, the percentage of girls outnumbered the percentage of boys among the graduates and above category (Table 2). Some data on trade-wise intake in government ITIs and private ITCs (one year course)

26Surprisingly, even one of our younger respondents, Anita, only in her twenties, cited puberty and economic hardship as reasons for her education being stopped at class seven. Almost all our respondents pointed out that puberty spelt restrictions on their mobility. Only two respondents, Kala and Chandrikakumari stated that puberty did not bring any restrictions. (Life Histories)

27Proximity to primary school was there but schools for higher levels were not very near. Currently the state has one lower primary school for every sq km and one secondary school for every 4 sq km.


Gender Disparity (M/F) in the Different Levels of Educational Achievements of Kerala: 1971–1991




Note: NTDip: Non Technical Diploma not equal to degree

TDip: Technical Diploma not equal to degree

I: Graduation other than technical degree

II: Post Graduation degree

III: Engineering and Technology

IV: Medicine

V: Agriculture, Dairying and Veterinary

VI: Teaching

This disaggregation is given only for urban areas.

2. Gender disparity is estimated as the ratio of male to female percentage of literates in each educational category

*Percentage to literates

Source: Census of India, Social and Cultural Tables: Kerala (various issues)

shows a preponderance of girls in stenography, dress-making, cutting and tailoring, secretarial practice and data preparation (Table 3); girls’ intake in the two-year technical courses is negligible (Table 4) except in civil draftsman and radio mechanic trades. Hence, the narrowing of gender disparities in education have equipped women to acquire earning power in ‘suitable’ occupations


Tradewise Intake in Government ITIs and Private ITCs during 1995–96 (one year course)


Source: Kerala State Planning Board (1997) Report of the Steering Committee. Ninth Five Year Plan: 1997–2002.


Tradewise Intake in Government ITIs and Private ITCs during 1994–95 (two year course)


Source: Kerala State Planning Board (1997) Report of the Steering Committee. Ninth Five Year Plan: 1997–2002.

generally non-technical in nature as we shall see from the employment pattern. The persistence of gender differentiated family roles, with primary responsibility of domestic chores falling on women, in turn perpetuates this sexual division of labour through an asymmetry of opportunities offered for acquiring ‘untraditional’ skills. While girls have made remarkable advances in professional courses such as engineering, medicine, agriculture, dairy development and veterinary science, their achievements are still low compared to boys. However, in the Kerala context, there is a certain sense in which girls are overeducated, being encouraged to study further while waiting to get a suitable job.

This is related to the nature of the labour market. A reason cited often for the lower (than all-India) and falling levels of female work-participation rates in Kerala is the longer years spent in schooling/higher education. Almost 32 per cent of males and 26 per cent of females were recorded as students in rural Kerala in 1987-88; the figures for rural India were 19 per cent and 11 per cent respectively (NSSO 1990). However (Kumar, 1992) points out that in a situation of slow growth of desired employment opportunities, commensurate with the levels of education, the causation may be the other way round: that the girls continue in the educational stream due to the lack of suitable employment avenues. Thus the growing proportion of students may well be a reflection of the falling levels of participation rather than the reverse. At the micro level this fits in with the patrifocal family, as it could be seen to further the ‘marriageability’ of girls, by enabling them to make better wives and mothers and/or be a potential contributor to the economic well-being of the family. However, higher levels of education of girls in a situation of high overall unemployment rates, are also manifested in poorly educated men with good jobs marrying better educated women, observed particularly among the Gulf migrants (Rajan et al., 1996).28 While this does not seem to have affected gender relations in any visible way; there are indications that it is one of the factors shaping domestic violence. The INCLEN/ICRW (2000) study revealed a strong association between violence, physical and psychological, and female favourable gender gap in education and employment.

Levels of participation

Female work participation rates (WPRs) in Kerala have been among the lowest in India and declining, (Gulati and Rajan, 1991; Eapen, 1992; Kumar, 1992). The 1991 Census ranks Kerala 22nd among the states with respect to female participation. Using the NSSO

28This is also evident in a recent study of migrant households in several districts that there is a ‘premium’ attached to the Gulf migrant as a ‘desirable’ bridegroom. In cases of migrant bridegrooms, the age difference between husband and wife tended to be wider and on an average the wife was better educated than the husband. Only in government jobs were there a higher proportion of women married to Gulf migrants as compared to all women (Zachariah et al., 2000).

data which adopts a more extended definition of work, we find that female WPRs, in terms of the usual principal and subsidiary status, hover between 20 per cent (urban) and 23 per cent (rural) according to the 50th Round of the NSSO for 1993-94, compared to male WPRs of 56 per cent (urban) and 54 per cent (rural). While male work participation rates have remained steady (in rural areas) since 1977-78 (32nd Round) or turned mildly upwards in urban areas, female WPRs have declined consistently, more so in rural areas, and it is only between 1987-88 (43rd Round) and 1993-94 that female urban WPR has increased; however the rural WPRs declined further.

That the female WPR is low does not mean that the supply of female labour is low since a certain percentage of women would be unemployed. Indeed female unemployment rates are very high in Kerala, in particular among educated women in rural areas. Table 5 attempts to relate labour market indicators by level of education for women and men, highlighting the inferior position of women in the labour market, both in terms of employment/unemployment. Almost a quarter of women graduates in rural areas are unable to procure employment compared to 13 per cent for men. While the problem is less serious in urban areas, among matriculates it is very severe. Although unemployment increases with the level of education, the possibility of securing regular work is also higher. This is reflected in the sharp rise in the proportion of women in regular employment with graduate level education, a reason we mentioned earlier for women continuing to study.

It is also a matter of concern that even in the prime working age groups, 20-34, female work force participation rates (44 per cent in rural and 25 per cent in urban areas) are less than half that for men (96 per cent and 95 per cent respectively); and while male WPRs in the age groups of up to 54 years have increased between 1987-88 and 1993-94, female WPRs have declined (data not shown here). Is it possible then that in the wake of increasing male participation rates, and high female unemployment rates, women are withdrawing from the labour market and it is probable that the withdrawal is among the educated. A recent Migration Survey in Kerala (Zachariah et al., 2000) shows not only that wives of emigrant husbands were on average better educated but a significantly higher proportion of them (84 per cent), than the proportion of all women (60.9 per cent) reported being housewives.


Various Aspects of Employment and Unemployment of Persons of Age 15 years and above Across Educational Categories; Genderwise 1987-88 (usual principal and subsidiary status)


Note: I: Not Literate

II: Literate upto Primary

III: Middle

IV: Secondary

V: Graduate and above

Unemployment rate is per 100 population; the base is the educational categories.

Source: Sarvekshana, Vol. 26(2) Oct.–Dec. 1992: Results of the Fourth Quinquennial

Survey on Employment and Unemployment (43rd Round).

Patterns of work

Manufacturing, trade and services are growing areas of female employment in Kerala. While a more detailed break-up of industry groups is not available for 1993-94, between 1977-78 and 1987-88, we find that within manufacturing the growth in employment was in food processing industries including canning/processing of fish, beedi making, garment making, and wood products within which basket/mat weaving was important, and non-metallic mineral products, primarily brick making. The growth of employment in non-traditional sectors was marginal except in ‘electrical machinery’, largely on account of a number of labour intensive sub-contracting units which came up in the electronics industry. In trade most of the growth in employment was in the retail trade, primarily as sales girls/assistants (Eapen, 1994). While there was an increase of female employment in public administration, education and health under Social, Community and Personal Services, which are largely organized sector activities, it must be remembered that for women almost 60 per cent of organized sector employment is accounted for by the private sector while for men it is the reverse (Government of Kerala, 1989).

However, this growth in organized sector employment has to be probed deeper since looking at changes in the status of employment of women, we find that the share of casual work has been increasing while that of regular employment shows a decline. For instance, between 1987–88 and 1993–94 casual employment for females grew from 48 per cent to 50 per cent in rural areas and much more sharply, from 23 to 32 per cent in urban areas. Regular employment in urban areas declined very sharply. Since it is in urban areas that female participation rates show an increase, is it that a substantial part of the so-called newly emerging formal sector employment for women is not of a permanent, regular nature and hence being recorded as casual work? In other words does it reflect the increasing informalization of activities in a bid to keep the earnings low?

All this evidence points to an assymetrical position for women in terms of occupational distribution. We have seen how family structures channel women’s education to specific areas, facilitating occupational segregation, in areas generally less capital-using and less productive. This is confirmed by the occupational distribution of women, relating to the Census year 1981. The differential

occupational distribution by gender, captured by the Occupational Segregation Index, is rather high, the index of dissimilarity (which measures the extent of dissimilarity in the distribution of men and women across occupation taking a value between 0 and 100) being 52 (Table 6). It is interesting to highlight the type of jobs women are engaged in: even in the field of professional/technical activities where the proportion of women is relatively high (female to male ratio being 0.64) most of the women are engaged in the lower rungs


Occupational Classification of Workers other than Cultivators and Agricultural Labourers, Kerala: 1981


Dissimilarity Index 0.515579 Calculated as (1/2) Σ (Fj/F) – (Mj/M)

Source: 981: Census of India, Kerala, Series 10, Part III-A&B[iii], General Economic Tables (B21 & 22).

of the professional hierarchy — teaching but largely in schools, especially at the primary and nursery levels. In the medical profession the larger number is in nursing. Other professions are clerical, sales assistants, maids, sweepers, cooks and very few in managerial, administrative occupations. That even for the same levels of professional education, women’s earnings are lower, establishes the fact of both occupational segregation and discrimination in the labour market. A study shows that the gender gap in earnings of highly qualified persons, measured by the ratio of female to male earnings is 0.78 in Kerala in the science and technology fields. About 18 per cent of the differential can be explained in terms of the occupational segregation, 50–60 per cent by productivity characteristics, which suggests that almost a quarter is in terms of discrimination (Duraisamy and Duraisamy, 1997).29

We have attempted to highlight certain dimensions of the labour market, which suggest a certain continuity in terms of the gendering of employment, reflecting women’s weaker position. Nor has there been any marked improvement in recent years. On the contrary, the informalization of formal sector activity and a tendency towards withdrawal from the workforce on the part of the educated women, indicate a worsening of their access to this ‘self-acquired’ income. This only weakens women’s position within the family, since any attempt to have a greater ‘voice in the family’ could be misconstrued as an attempt to challenge the gender differentiated family authority and lends itself to domestic violence, particularly because of the wife’s greater economic dependence on the husband. Increasing levels of domestic violence in the state and a norming of the male working subject and a domestic woman (Osella and Osella, 2000) suggests that education, even higher education, does not appear to have motivated large numbers of women to challenge gender role assumptions. In this context an observation by a well educated woman (a qualified lawyer) from our life histories, is instructive. ‘Though there is not much open discrimination against women in Kerala, there is a sharp distinction in the roles of men

29Gender disparity in earnings of casual workers in agriculture and non agriculture clear from secondary information was confirmed by three of our respondents. Whether agriculture, construction or tile factory work, women receive ten to fifteen rupees less than men “for the same work”. (Life Histories).

and women. If women enter their roles, they have to face all kinds of abuse’.


From the preceding analysis it appears that ‘status’ of women as conventionally defined is inadequate for capturing the relations of power between men and women, which systematically place women in an inferior position in the household and outside. Although measures of literacy have been improved, in terms of enrolment rates and retention rates, to highlight the structural constraints on women’s education, its snowballing effect in terms of occupational rigidities and women’s own perceptions and aspirations for adult life need to be addressed. Our study throws up the need to understand decision-making at the household level, in the context of norms and practices that influence behaviour and shape choices. There are clear indications that families (whether natal or conjugal) mediate education and employment decisions of women, channeling them towards the ‘marriageability’ of girls. Alongside are the indications of the decline of women’s property rights in erstwhile matrilineal families as well as women’s lack of control over property transferred at marriage among matrilineal and patrilineal families. Greater access and resort to consumer practices have left their stamp on the organization of marriage as well as gendered decisions on education and employment. More importantly they have added new dimensions to earlier images of masculinity and femininity in the direction of the male ‘working’ subject and ‘domestic’ women. It is perhaps in this context of the ‘discontinuity’ between education and employment of women and ‘autonomy’ that we need to place the emerging picture of declining property rights, violence and the mental ill health of women. Here we need to reiterate that Kerala leads other states in the number of reported suicides, the links between dowry, violence and suicides on the one hand and that women lead men (as elsewhere in the country) in ‘common’ mental conditions such as stress on the other hand. Also important are the male-female differences in reasons advanced for common mental conditions, a larger number of women attributing it to marital disharmony as against the larger number of men citing economic factors.


Agarwal, Bina (1994) A Field of One’s Own: Gender and Land Rights in South Asia, Cambridge: Cambridge University Press

Alexander, William (2000) ‘Normal Kerala and Abnormal India: Reflections on Gender and Sustainability’, in Govindan Parayil (ed.) Kerala: The Development Experience: Reflections on Sustainability and Replicability, London: Zed Books

Ambili, C.S. (1999) ‘Growth of School Education in Kerala: Patterns and Differentials’, M.Phil Dissertation, Trivandrum: Centre for Development Studies

Ammal, R.S. (1980) ‘Relationship of Gynecological Problems to Neuroticism’, Psychological Studies, 25(1), 26-30

Batliwala, Srilatha (1994) ‘The Meaning of Women’s Empowerment: New Concepts from Action’, in Gita Sen et al (eds.) Population Policies Reconsidered: Health, Empowerment and Rights, Cambridge MA: Harvard University Press

Davar, Bhargavi V. (1998) Mental Health Of Women: A Feminist Agenda, New Delhi: Sage Publications

Director of Census Operations; Kerala, Primary Census Abstract: 1961, 1971, 1981 & 1991

Director of Census Operations; Kerala, Social and Cultural Tables: 1961, 1971, 1981 & 1991

Director of Public Instruction, Government of Kerala: various issues

Duriaswamy, P and Malathy Duraiswamy (1997) ‘Gender Bias in the Scientific and Technical Labour Market: A Comparative Study of Tamil Nadu and Kerala’, Paper Presented at Dr. T.N. Krishnan Memorial Seminar on ‘Development Experience of South Indian States in a Comparative Setting’, Trivandrum: Centre for Development Studies

Dyson, Tim and Mick Moore (1983) ‘Kinship Structure, Female Autonomy and Demographic Behaviour in India’, Population and Development Review, 9(1): 35-60

Eapen, Mridul (1994) ‘Rural Non-agricultural Employment in Kerala: Some Emerging Tendencies’, Economic and Political Weekly, Vol. 29 (21)

— (1992) ‘Fertility and Female Labour force participation in Kerala’, Economic and Political Weekly, Vol.27, p.2179

Fuller, C.J. (1976) The Nayars Today, Cambridge: Cambridge University Press

Gough Kathleen (1993) ‘The Nayars and their Definition of Marriage’, in Patricia Uberoi (ed.) Family, Kinship and Marriage in India, Delhi: Oxford University Press. (first published 1959)

— (1961) ‘Nayars: Central Kerala; North Kerala; Tiyas: North Kerala; Mappillas: North Kerala’, in Schneider and Gough (eds.) Matrilineal Kinship, University of California Press

— (1952) ‘Changing Kinship Usages in the Setting of Political and Economic
Change Among the Nayars of Malabar’, Journal of the Royal Anthropological Institute, Vol. 82, Part I

Government of Kerala (1997) ‘Report of the Steering Committee on Housing, Labour, Social Security and Committee on Housing, Labour, Social Security and Welfare: Ninth Five Year Plan 1997-2002’, Trivandrum: State Planning Board

Government of Kerala, (1989), Economic Review, Department of Economics and Statistics

Gulati, L. and S.I. Rajan (1991) Population Aspects of Aging in Kerala, Their Economic and Social Context, New York: United Nations

Halliburton, Murphy (1998) ‘Suicide: A Paradox of Development in Kerala’, Economic and Political Weekly, 33(36&37): 2341–2345

Heward, Christina and Sheila Bunwaree (eds.) (1999) Gender, Education and Development: Beyond Access to Empowerment, London: Zed Books

Hirway and Mahadevia (1996) ‘Critique of Gender Development Index: Towards an Alternative’, Economic and Political Weekly, Oct 26

International Institute of Population Sciences (2000) National Family Health Survey, 1998-99, Mumbai

INCLEN/ICRW (2000) ‘Indiasafe: Studies of Abuse in the Family Environment in India—A Summary Report’, International Clinical Epidemiologists Network, India

James, Lizzy (1999) ‘Family Counselling in Family Court: An Analysis of Psycho-Social Dynamics of Families of Litigants’, (Thrissur Family Court) Discussion Paper No.13, Trivandrum: Kerala Research Project for Local Level Development

Jayasree, A.K. (1997) ‘Gender Issues in Suicides: A Study Among Suicide Attempters in Kerala State’ (source unknown)

Jayaweera, Swarna (1999) ‘Gender, Education and Development in Sri Lanka’, in Heward and Bunwaree (ed.) Gender, Education and Development: Beyond Access to Empowerment, London: Zed Books

Jejeebhoy, Shireen J. (1998) ‘Wife-Beating in Rural India: A Husband’s Right?’ Evidence from Survey Data, Economic and Political Weekly, 33(15): 855–862

Jeffrey, Robin (1992) Politics, Women and Well being: How Kerala became ‘a Model’, London: Macmillan

Jeffrey, Roger and Alaka M. Basu (1996) (eds.) Girls Schooling, Women’s Autonomy and Fertility Change in South Asia, New Delhi: Sage

Kabeer, Naila (1999) Resources, Agency, Achievements: Reflections on the Measurement of Women’s Empowerment, Development and Change, Vol. 30, pp. 435–464

Karve, Irawati (1953) Kinship Organization in India, Poona: Deccan College

Kerkhoff, K. R., (1998) ‘Production and Reproduction of Girlhood in High

Schools: The State, Family and Schooling in Colonial Calcutta’, in C. Risseeuw and Kamala Ganesh (eds.) Negotiation and Social Space: A Gendered Analysis of
Changing Kinship and Security Networks in South Asia and Subsaharan Africa, New Delhi: Sage

Kolenda, Pauline (1987) Regional Differences in Family Structure in India, Jaipur: Rawat

Kumar, Rachel (1992) ‘Women, Work and Development: Issues in Female Labour Force Participation in Kerala’, Centre for Development Studies, M. Phil Dissertation

Kurien, Prema Ann (1994) ‘Non Economic Bases of Economic Behaviour: The Consumption, Investment and Exchange Patterns of Three Emigrant Communities in Kerala’, Development and Change, Vol. 25, 753–783

Mani, Suraraj (2000) Report of the ‘Need Assessment of Severe Mental Morbidity of Kerala State’, Trivandrum: Kerala State Mental Health Authority

Mason K.O. (1993) ‘The Impact of Women’s Position on Demographic Change during the Course of Development’, in Nora Federici (eds.) Women’s Position and Demographic Change, Oxford: Clarendon Press

— (1985) The Status of Women: A Review of its Relationship to Fertility and Mortality, New York: The Rockefeller Foundation

Mencher, Joan (1965) ‘The Nayars of South Malabar’, in M.F. Nimkoff (ed.) Comparative Family Systems, Boston: Houghton Mifflin Company

Menon, Shanti (1996) ‘Male Authority and Female Autonomy: A Study of the Matrilineal Nayars of Kerala, South India’, in Mary Jo Maynes, (eds.) Gender, Kinship and Power: A Comparative and Interdisciplinary History, New York: Routledge

Morrison, Barrie M. (1997) ‘The Embourgeoisement of the Kerala Farmer’, Modern Asian Studies, Vol. 31, No. 1

Mukherjee, Chandan et. al. (1999) ‘Crime Against Women: A Preliminary Statistical and Spatial Analysis of National Crime Records Bureau Data’, Presented at the Centre for Development Studies, Thiruvananthapuram

Mukhopadhyay, Carol Chapnick and Susan Seymour (1994) (eds.) Women, Education and Family Structure in India, Boulder: Westview Press

National Crime Records Bureau, Crime in India, various issues

National Sample Survey Organization, (1990), Key Results of Employment and Unemployment Survey: All India, NSS 43rd Round, (July 1987–June 1988), New Delhi

Osella, Filippo and Caroline Osella (2000) Social Mobility in Kerala: Modernity and Identity in Conflict, London: Pluto Press

— (1999) ‘From Transience to Immanence: Consumption, Life Cycle and Social Mobility in Kerala, South India’, Modern Asian Studies, 33(4)

Parayil, Govindan (ed.) (2000) Kerala: The Development Experience—Reflections on Sustainability and Replicability, London: Zed Books

Puthenkulam, S.J. (1977) Marriage and Family in Kerala, Calgary

Rajan, Irudaya, S., S. Sudha and P. Mohananchandran (2000) ‘Fertility Decline
and Worsening Gender Bias in India: Is Kerala No Longer an Exception’, Development and Change, Vol. 31

Raj K.N. (1994) ‘Has there been a Kerala Model?’ In A.K.G. Centre for Research and Studies, International Congress, Abstracts, Vol. I

Ramachandran, V.K. (1997) ‘On Kerala’s Development Achievements’ in Dreze, J. and A.K. Sen (eds.) Indian Development: Selected Regional Perspectives, New Delhi: Oxford University Press

Roy, Mary (1999) ‘Three Generations of Women’, Indian Journal of Gender Studies, 6(2)

Saradamoni K. (1983) ‘Changing Land Relations and Women: A Case Study of Palghat District, Kerala’, in Vina Mazumdar (ed.) Women and Rural Transformation: Two Studies, New Delhi: Concept Publishing Co

Sen, A.K. (1990) ‘Gender and Cooperative Conflict’, in Irene Tinker (ed.) Persistent Inequalities, Oxford: Oxford University Press

Sonpar, Shobana and Ravi Kapur (1999) ‘Non-Conventional Indicators of Gender Disparities Under Structural Reforms’, Paper Presented at the Workshop on Gender Discrimination Under Structural Reform, New Delhi: Institute for Social Studies Trust

Tharakan, Michael P.K. (1997) ‘History as Development Experience: Desegregated and Deconstructed Analysis of Kerala’, unpublished Ph D dissertation, School of Social Sciences, Mahatama Gandhi University, Kottayam

United Nations Development Programme (1975) Human Development Report, New Delhi: Oxford University Press

Vadakumchery, James (1994) ‘Suicides: A Study about the Victims of Unsuccessful Attempts in Kerala State’, unpublished paper

Visaria, Leela (1996) ‘Regional Variations in Female Autonomy and Fertility and Contraception in India’, in Jeffrey and Basu (eds.) Girl’s Schooling, Women’s Autonomy and Fertility Change in South Asia, New Delhi: Sage Publications

Visvanathan, Susan (1993) The Christians of Kerala: History, Belief and Ritual among the Yakoba, New Delhi: Oxford University Press

Zachariah, K.C. et al., (2000) Dynamics of Migration in Kerala: Dimension, Differentials and Consequences, Trivandrum: Centre for Development Studies

Trade, Gender and Employment Issues



Globalization and World Trade

The last two decades have witnessed the adoption of uniform macro economic strategies as part of the Structural Adjustment Programme (SAP) and economic liberalization undertaken by the developing world. These strategies bring the focus of policies on the external trade sector and international finance issues. The period since the 1980s has, accordingly, witnessed a significant growth in global trade. World exports increased from US$ 314 billion in 1970–71 to US$ 2,000 billion in 1980–81, US$ 3,306 billion in 1990–91 and to US$ 5,325 billion in 1997–98. During 1990–97, the volume of merchandise trade grew at an average annual rate of growth of 6.5 per cent, whereas the world merchandise production grew by 2 per cent during the same period. Trade growth, as usual, outpaced domestic output growth by a significant margin, ‘attesting once again to the onward pace of globalization’ (WTO, 1997, Vol. 1:2).

Further, the composition of world trade has changed. The share of manufacture exports in value of world merchandise exports increased significantly during 1980–97, while that of agricultural products and mining products declined (Table 1). The impact of global trade has also not been uniform across various regions of the world. Western Europe, Asia and North America contribute to large shares in world merchandise exports and these three regions taken together contributed to 86 per cent of the total merchandise exports. The shares of central/eastern Europe, North America, Latin

*Views expressed are personal views of the author, and are in no way attributable to her organization.

America and Asia in world merchandise exports increased significantly between 1980 and 1997, while that of Africa, the Middle East and Western Europe declined (Annexe 1).


Product Share in Value of World Merchandise Exports, 1980–97


*Includes minerals other than fuels

Source: WTO, 1998:73

Trade in Asia

By 1997, Asia accounted for over one-fourth (26 per cent) of the total value of world merchandise exports. Within Asia, there is a large variation in the shares of individual countries in the total value of Asia’s merchandise exports. Shares of exports of South Asian countries are low, with India contributing 2.5 per cent in 1997, and other countries such as Bangladesh, Pakistan, Sri Lanka contributing less than one per cent each. In comparison, shares of Japan, China, Republic of Korea, Singapore, Malaysia etc., are much higher than countries in South Asia. The shares for most countries, except those of Japan and Pakistan, have increased significantly between 1990 and 1997. India’s share has increased marginally. As for composition of Asia’s exports by major sectors, in keeping with the world trend, the share of manufactures is high at 83 per cent in 1997, which increased from 79 per cent in 1990 (Table 2).

The crucial issue is whether the developing countries (in Asia) have witnessed economic growth and reduction in poverty levels during the globalization phase. A recent Trade and Development Report (UNCTAD, 1999) admits that the experience of developing countries in their efforts to liberalize and integrate into the world economy has not been as expected. Their expectations that such integration would bring gains in terms of faster growth, greater employment opportunities and reduced levels of poverty, have not been borne out. The report concludes that:


Share in Asia’s Merchandise Exports by Countries and Product Type, 1990–97


Source: WTO 1998: 65, 147.

 . . .after more than a decade of liberal reforms in developing countries, their payment disorders, which had earlier ushered in a rethinking of policies, remain as acute as ever, and their economies depend even more on external financial resources for the achievement of growth rates sufficient to tackle the deep-rooted problems of poverty and under development. (UNCTAD, 1999: vi)

Growth in developing countries recovered in the 1990s from levels achieved in the 1980s, but these were much below the average growth (5.7 per cent) achieved during the 1970s (UNCTAD, 1999). The recovery was accompanied by a significant worsening of external deficits. Further, it is also increasingly being argued that the Uruguay Round implementation process has done little to improve the market access for the exports of goods and services of developing countries. The south-east Asian currency crisis during 1997–98 points to the greater vulnerability of developing countries to financial shocks and crises and suggests that future multilateral trade negotiations have to take place against the backdrop of the increasing vulnerability of developing countries.

Issues and Evidence in Globalization, Trade and Employment of Women

Issues in globalization, trade and gender have taken on a new meaning and dimension since January 1995, when a comprehensive round of multilateral trade agreements embodying the results of the Uruguay Round of Multilateral Trade Negotiations was concluded and the World Trade Organization (WTO) was formed. Out of the present 140 members of the WTO, some 98 are from the developing world. China, till recently, was not a member and its decision to join has immense implications for the developing world especially for the South Asian countries.

The Uruguay Round of GATT (finally signed in April 1994) marks a major change in the rules of the international trading system. At this stage an assessment of the gender-differentiated impact of WTO agreements on employment can at best be speculative primarily because it will depend on the degree of implementation of the WTO Agreements and information on this is not so readily available at the moment. Also, lack of evidence in developing countries on the gender-disaggregated composition of the labour force by sectors and on the responses of the labour force to economic reforms limits any attempts at a comprehensive analysis of this issue.

The study of gender and international trade is relatively new. Some feminist economists and academics (see for example, Afshar and Dennis 1992; Cagatay, Elson and Grown, 1995; Commonwealth Secretariat, 1989; Moser, 1989) have focused on this issue in the context of their work on structural adjustment and the associated expansion of export-oriented industries since the mid-1980s. Fontana, et. al. (1998) have argued that the benefits of trade liberalization and expansion are differentiated between women and men and also between different groups of women. Evidence of the impact of structural adjustment in developing countries shows that macro economic adjustment is not gender-neutral and it tends to disadvantage women in relation to men, with adverse social and economic implications for women (Elson, 1991; Palmer, 1991).

Two basic premises on which the research on gender and international trade is based are: (i) trade liberalization brings different costs and benefits to men and women, and this gender bias cuts across all economic and social categories; and (ii) the

impact of trade liberalization is mediated by gender relations and gendered social, economic and political structures. These structures may be in the form of gender gaps in education and health; patterns of labour market discrimination and labour force participation levels; gendered patterns of rights and resources; and other socio-cultural factors. In other words, gender and trade is a two-way relationship, not only does trade have differential gender impacts but gender biases and gender barriers also influence trade policy outcomes (Cagatay, Elson and Grown, 1995).

The crucial issue is, therefore, whether development policy enables a more equitable distribution of the gains (and also the costs) associated with trade liberalization and expansion. Relevant questions that need to be asked in the context of gender and equity include the following: what is the distribution of costs and benefits of increased trade between men and women (and between boys and girls)? Has globalization and increased trade meant fewer benefits and greater costs for women? Further, it raises gender and efficiency questions such as whether there are ways in which gender relations impede the process of globalization and trade, lower productivity and obstruct the achievement of long-term development goals. Gender and efficiency in the context of liberalization needs to be discussed in terms of gender barriers to trade (Palmer, 1991).

According to trade theory (Heckscher-Ohlin), trade liberalization affects the level and composition of employment and markets adjust to move labour from non-tradables to tradables and ultimately the economy achieves full employment. Assumptions of fast adjusting markets, complete mobility of labour and full employment are strong ones for developing countries and are very significant from a gender perspective. As argued by feminist literature, women are less likely to own or have access to resources. They are likely to be less mobile given the responsibilities of child care and other human resource requirements of the family and given constraints of education and training that may be required for new jobs (Fontana et al., 1998; Elson, 1997). These disadvantages may act as barriers to perfect mobility and full employment and are likely to result in crowding women into areas of the informal sector such as domestic services, petty trading etc., where entry barriers are low, and remuneration even lower.

Another aspect of the impact of globalization is that of negative employment effects induced by cheap import displacement, which

are likely to be most in small-scale subsistence agriculture and in the informal sector. As women are concentrated in these sectors, the negative employment effects may impact them disproportionately. Existing evidence on the issue primarily focuses on examining the impact of changes in export production rather than of import displacement and therefore there is a bias towards the analysis of how trade expansion affects women by impacting on their employment opportunities in export production.

Evidence of the impact of globalization and increased trade on women’s employment shows that increased manufactured exports from the south are strongly associated with the feminization of the industrial labour force (Cagatay and Ozler, 1995; Joekes and Weston, 1994; Standing, 1989). A study of formal sector employment in manufacturing in developed and developing countries during 1980–85 shows evidence of an association between increased exports and increased female employment in manufacturing; the largest increases in both appeared to be in Mauritius, Tunisia, Sri Lanka, Malaysia, and the four East Asian ‘Tigers’ (Wood 1991). However, for some other countries, including most of sub-Saharan Africa, trade liberalization has not been associated with an expansion of female-intensive export industries. Where feminization has occurred, it may have reversed with the introduction of new technologies and new organizations of production (Beneria and Lind, 1995; Ozler, 1999). A study of women’s work in Lima (Peru) suggests that women’s first response to lower wages and increased prices (during SAPs) is to adjust consumption levels and enter the labour force if consumption levels are unsustainable (Francke, 1992). The impact of trade liberalization may be differentiated among women. As one study points out, among retrenched factory workers, married women (as against single women) were much worse off as a group in terms of job opportunities available to them after retrenchment (Hirata and Humphery, 1990 cited in Baden, 1997).

The standard theory of trade also predicts that globalization reduces wage inequalities by leading to increases in unskilled wages and lowering skilled wages. Evidence is however mixed. In east Asia, wage differentials narrowed with trade liberalization during the 1960s and 1980s, whereas in Latin America, the wage inequalities increased with globalization in the late 1980s and early 1990s. Examining the reasons why globalization may increase labour

market inequalities, Wood (2000) argues that this depends on two factors, namely, differences in resource endowments and the transfer of knowhow to developing countries. East Asian cheaper labour intensive goods (imports) eroded Latin America’s comparative advantage in low skilled goods, which led to a fall in demand and the real wages of unskilled workers. Similarly, wage inequalities may increase if an effort at inward mobility of highly skilled workers with knowhow is not encouraged and indigenized in areas of comparative advantage.

The evidence on the impact of trade liberalization on women’s employment in the countries of South Asia presents a mixed picture: globalization and increased trade resulted in an increase in women’s work participation in trade related activities in Bangladesh and Sri Lanka (Standing, 1999). Similar rising trends in employment were not evidenced in Pakistan (Dev, 2000). Systematic studies providing evidence on the impact of trade liberalization in India during the 1990s are limited. Available studies based on Indian data give a mixed picture of the feminization of the labour force. Banerjee (1999) argues that since the manufacturing sector exports form a small share of the GDP (less than 10 per cent) and since women workers account for a small share in that sector, expansion of exporting industries is not likely to result in a feminization of the Indian workforce as a whole. Some micro studies, however, report the increased involvement of women in trade-related activities (Unni and Rani, 1999). But this enhanced work participation may be distress-induced according to Mukhopadhayaya (1999). A study of a western city in India (Mumbai), found a substantial increase in labour market involvement on adverse terms and a greater demand on unpaid household work resulting in much greater stress for women workers (Gandhi, 1992). Faster annual growth in female employment in the tertiary sector was evidenced in the urban areas (Deshpande and Deshpande, 1999; Visaria, 1999). Evidence based on some micro studies (Sachetana, 1997) in an eastern state of India showed that women moved from unpaid family labour to wage work in handloom, zari work, and some had found employment opportunities in modernized mills (rayon). It is argued that while some of the increase in labour force participation witnessed in most parts of Asia may be due to greater recognition and quantification of women’s work by enumerators, there has also

been a genuine process of increasing participation by women in most of these countries (Ghosh, 1999).

Another important issue is regarding the conditions of work for women in trade related activities during the outward orientation of the economy. Some case studies (reviewed in Joekes and Weston, 1994) which are primarily of Export Processing Zones (EPZs) and export oriented manufacturing units show that women are better off in EPZs than in domestic industries, especially when their total remuneration is taken into account. Evidence in Baden and Joekes (1993) suggests that young single women were the preferred workforce. Some other studies show that the pay is poor and working conditions appear to be poor. A study of two EPZs in Bangladesh (in Chittagong and in Savar) concludes that, though the women are more empowered now than when they were confined to their households, yet they are not treated at par with the male employees and that discrimination in wages between men and women remains. What is more disturbing is that most educated women were absorbed into low wage sectors such as the manufacture of clothing. Wage discrimination was more severe among married persons than among young single persons. However, there is also some evidence which points out that in some cases new transnational export industries have provided women with employment at competitive or higher wages than other forms of local employment, which has given women a sense of freedom and greater autonomy (Lim, 1990 in Beneria and Lind, 1995); empowered them and perhaps helped them escape domestic violence (Kabeer, 1995).

Available evidence on trade and employment suggests that:

However, feminist literature points out that the research on trade

and employment has gaps, as it does not include gender explicitly as an analytical category. It does not address issues of the functioning of the invisible economy where women play a crucial role. It does not tell us whether competitiveness and comparative advantage in trade can be created to benefit from globalization, where female education can play an important role. It does not tell us if it is factors other than education, such as ‘docility’, ‘amenability to discipline’ characteristics (which would be crucial for requirements such as just-in-time methods, which are likely to increase under trade liberalization) that play a crucial role in the feminization of the labour force.

Most of the literature and evidence on gender, trade and employment is based on case studies of specific export sectors of EPZs. This is partly due to a general lack of gender-disaggregated data on employment by major export sectors in developing countries, on a time series basis. Some official agencies do undertake specific labour force surveys, which give data for that year. These surveys are typically undertaken with a gap of a few years and a lack of these data on a regular basis makes it difficult to study changes in the employment of men and women in export sectors that are associated with globalization and trade policy changes.

Fontana and Wood (2001) by constructing a gendered Computable General Equilibrium (CGE) model for Bangladesh have attempted to analyse the impact of trade liberalization on women at a more macro level. They have tried to simulate the effects of changes in trade policies on the employment, wages, leisure and social reproduction activities of women and men. They found that an increase in production and the export of ready-made garments in Bangladesh during the last decade had led to a large increase in employment in the manufacturing sector for both men and women, but the increase was larger for women. Women’s wage rates increased both absolutely and relative to those of men, suggesting a narrowing of the female/male wage gap by about eight per cent. However, women’s participation in other activities namely social reproduction, agriculture, and leisure time, fell; the fall being largest (about six per cent) in the case of women’s leisure time. These results suggest that women’s well being is not necessarily improved by the expansion of female-intensive manufactured exports (as their leisure time reduces); and the reduction in hours of social reproduction also has adverse implications for children and

dependents. This suggests that even when the economic impact of trade liberalization is favourable for women, there is a need to design complementary policies ‘to reduce competing demands on women’s time’. These policies may include measures to improve water collection, food processing and the availability of market substitutes for household services, such as childcare (Fontana and Wood, 2001).

Employment of Women in Export Sectors in India

There is a large body of evidence which shows that in developing countries, women make a significant contribution to the national employment, output and trade levels. Changes in policies that affect trade and employment impact women directly as the poverty of women is closely linked to their pattern of employment and their disadvantages in the labour market (see Palmer, 1991). To provide the context for examining employment changes in various sectors, we now present a brief review of the macro economic and external sector performance of India during the first generation reform period.1

Macro scorecard

The Indian economic reforms, a process initiated in July 1991, were a response to an internal as well as an external crisis, when foreign currency reserves plummeted to about $ one billion (or two weeks of imports), export growth turned negative, industrial growth was –1.3 per cent, inflation was soaring above 16 per cent and overall economic growth was down to one per cent (in 1991–92). As in many other developing countries, especially after the adoption of the SAP, exports in India also have been conceived as an engine of economic growth.

The reform package combined stabilization with fiscal discipline and expenditure switching through structural reforms. The first

1The period 1991-99 is being referred to as the period of first generation reforms and the period since 1999 (when the present Government was re-elected) as a period of second generation reforms when reform process continued with further liberalization of the insurance sector as announced in March 2000 at the time of the presentation of the Government of India Budget of 2000-2001.

generation reform package included a strong stabilization programme and wide-ranging reforms in fiscal policy, industrial policy, foreign trade, the exchange rate system, foreign investment policy, the financial sector, and the public sector. The reduction of the fiscal deficit (from over 7–8 per cent of GDP in the early 1990s to a targeted 4 per cent of GDP) remained central to both the stabilization and the structural adjustment components. For want of space, details of the reforms measures taken in India so far have not been listed here, but they can be found in the recent Economic Survey of the Government of India (Ministry of Finance, 2002).

There are varying views on the macro economic performance of India during the reform period. The promoters of reforms (including the GOI) argue that the performance has been impressive. Except for the crisis year of 1991–92, the average real GDP growth rate during 1992–93 to 1998–99 was 6.5 per cent. The industrial growth rate varied from a low of 2.3 per cent in 1992–93 to a high of 12.8 per cent in 1995–96, averaging around 6.5 per cent for the period between 1992–93 to 1998–99. The agricultural growth rate averaged around 2.5 per cent, with a number of year-to-year variations as shown in Table 3 below. In terms of overall economic growth, India has shown better recovery


Some Macro Economic Indicators of the Indian Economy, 1990–99


Note: The base years used for reporting the rate of growth of GDP are 1980–81 for the years 1990–93 and 1993–94 thereafter; IIP-base 1993–94.

Source: Ministry of Finance (2000).


Fig. 1 Macro Indicators of the Indian Economy 1990–99
Source: Table 3.

during the reform period when compared with other developing countries undertaking post-crisis reform programmes. India’s average economic growth of 6.4 per cent in the first three years after the start of the reforms programme in mid 1991 compares favourably with an average of 2.2 per cent for 30 developing countries surveyed in one study (Thomas et al. (eds.), 1991). (Annexe 2).

The performance of the Indian external sector during the reform period was also impressive. Since 1990–91, when the structural adjustment phase was initiated in India, exports faced an initial setback in 1991–92, when they declined by 1.5 per cent in dollar

terms. The revival started from the following year, when Indian exports increased by 3.8 per cent; this was followed by an impressive export growth in the next three years, averaging about 20 per cent during 1993–94 to 1995–96. India’s export growth rate has been better than that of the world as a whole since 1993–94, except in 1998–99 (see Annexe 3). Some international factors such as the world recession in 1998 and the Southeast Asian currency crisis affected India’s export growth rate during 1998–99. During 1999–2000, however, provisional estimates show a high export growth rate of 13 per cent. The current account deficit came down from an unsustainable level of $9.7 billion (3 per cent of GDP) in 1990–91 to a manageable level of 1.0 to 1.5 per cent of GDP even after there was higher import growth since 1993–94. Foreign investment rose to $4 billion by 1993–94 and foreign currency reserves built up from two weeks import cover in 1990–91 to six months import cover by March 1995 (Acharya, 1999).

Trade liberalization aspects, which are to do with the impact of changes in import tariffs and quantitative restrictions on imports, which in turn may have a direct impact on the employment of men and women, are not so well researched and documented. The trade liberalization measures included progressive reductions in custom duties: the peak custom duties were reduced from 300 per cent in mid 1991 to 30 per cent in March 2000. Quantitative restrictions on a wide range of imports were also removed since 1993, as these could no longer be justified on Balance of Payment (BOP) grounds under Article XVIII-B of GATT. India has removed all quantitative restrictions under the WTO with effect from April 1, 2001. The implication of such liberalization on the import restrictions for the employment of men and women and their livelihoods have not been investigated in detail and there is little evidence available in developing countries on this issue.

Along with the trade liberalization measures, a multi-pronged strategy was followed to contain the external debt situation. The strategy included: reduction in short term debt and strict controls; making the medium term borrowing from private commercial sources subject to annual caps and minimum maturity periods, encouragement of non-debt creating flows (especially FDI and portfolio investment), accumulation of foreign exchange reserves for covering external sector uncertainties, etc. As a result, external debt indicators improved. A comparison of basic debt indicators of India

and some Asian countries and China shows that India had achieved a relatively high degree of comfort level by 1997 (see Annexe 4). Despite the economic and financial crisis in south-east and east Asia which dominated the global economic scene during 1998, limiting the world economic growth rate to less than 2 per cent and world trade growth to 3.7 per cent, the Indian economy exhibited a strong macro economic growth (ESCAP, 1998:3).

Some others (Chandrasekhar and Ghosh, 2002) argue that the view that the economic reforms of the nineties led India to a new, higher growth trajectory is claimed typically by the GOI. They suggest that the transition to a higher growth path occurred during the latter half of the 1980s when the liberalization was limited and not in the 1990s when the liberalization process was fast and more widespread. The macro economic performance in the latter half of the nineties was not so impressive: primary and secondary sector growth rates showed a deceleration during this period as compared with the first half. The tertiary sector growth was higher in the second half of the nineties as compared with the first half but this was helped by the increased government expenditure on salaries of government officials. The projected real GDP growth rate of 6.0 per cent for 2001–02 was revised downwards by the GOI to 5.2 per cent. They also argued that the reforms failed to meet expectations with respect to infrastructure provision and increase in employment generation.

Poverty data shows a clear evidence of a decline in poverty in India during the 1990s as compared with the 1980s. According to official estimates, the population below the poverty line declined from 36 per cent in 1993–94 to 26.1 per cent in 1999–2000. There is, however, a debate on these poverty estimates and there is no clarity on the comparability of poverty estimates for 1993–94 and 1999–2000 as the methodologies used were different (for details see Ministry of Finance, 2001:194). National Sample Survey results showed that a rapid fall in poverty between the seventies and early nineties was associated with a rise in GDP growth, however, the evidence of the mid (1993–94) and late 90s suggests a lack of poverty reduction at the same rate despite higher economic growth. Some other poverty indicators also showed an improvement during the 1990s. Real wage rates for agricultural labourers showed an improvement during the 1990s as compared with the 1980s. The employment in the organized manufacturing sector grew faster in

the 1990s as compared to the 1980s (Parikh and Radhakrishnan (eds.), 2002). However, a significant reduction in poverty in China was reportedly due to a rapid increase in the non-farm rural employment at the rate of 12 per cent in the 1980s as compared to only two per cent in India.2 World Bank study estimates poverty in India is declining at the rate of one per cent a year in the past five years and this could happen at a faster pace if the investment climate in the country is improved. According to the World Competitive Yearbook, the Indian investment climate has been judged to be poor (43 out of 47 countries ranked) and FDI attracted is only 0.5 per cent of the GDP. As compared to this, China is ranked better (at 31) in terms of the investment climate and attracts FDI worth 5 per cent of GDP.

Macro scorecard provides a mixed picture about the performance of the Indian economy. The government argues that economic reforms in India have been successful in terms of the macro economic performance and in the reduction in poverty during the 1990s. Others argue that the experience of reforms failed to deliver the promised growth and efficiency, particularly in specific areas such as infrastructure provision, employment generation and food systems. This raises an obvious question about the adequacy of the nature, speed and sequencing of reforms in India. We will not pursue this question further, since the focus of this essay is on trade and employment, except to say that some of the major problems which existed in the early years of reforms remain and these may limit potential growth and social outcomes in the medium and long term. Some of these problems include the paradox of surplus foodgrains and the decreasing availability of foodgrains per capita, the slowdown of GDP and export growth towards the late 1990s, the persistent high fiscal deficits, the inadequate physical and social infrastructure and the lack of effective governance (see discussions in Parikh and Radhakrishna (eds.), 2002)

More specifically on the exports front, the potential of export sector to serve as an engine of growth seems limited by the low absolute value of India’s exports and the composition of the export

2Quoted in a seminar by Chief Economist, World Bank in Delhi on March 27, 2001.

basket. In terms of their absolute value, India’s exports continue to be much lower than those of other important Asian countries. During a period of two decades, a number of Asian countries namely South Korea, Thailand, and China, with export values not very much higher than India, have surpassed India’s levels many times over. In 1998, India’s exports were valued at only US$ 35 billion as compared to US$ 184 billion for China; US$ 133 billion for South Korea; US$ 73 billion for Malaysia; US$ 54 billion for Thailand; US$ 49 billion for Indonesia and this was in a year when some of these economies were in a financial crisis. The composition of India’s exports also shows their vulnerability as India’s export basket is narrow. The country depends on only few sectors for our exports, which implies that any little change in the markets, or prices of these items, will affect Indian exports significantly. In 1997–98, about 1/4th (25 per cent) of India’s exports were textiles, two major items of which are readymade garments (13 per cent) and cotton yarn and fabrics (8 per cent); about 15 per cent of our exports are made up of gems and jewellery; and about 18 per cent are made up of agriculture and allied products. Therefore, the potential of exports to serve as an engine of growth depends on a pro-active diversification strategy along with other measures of improvement in infrastructure and the investment climate. Historical evidence shows that Foreign Direct Investment (FDI) has been a key factor in reforming the export and globalization efforts of countries. India has not been able to attract FDI linked to exports, whereas in China, about 47 per cent of the exports are directly generated by FDI.

Employment of women in India: various sources

In order to be able to understand the impact of the new policies on the economic status of women, a detailed understanding of the occupational pattern of women (and men) employees is necessary. This tells us about the areas and sectors which are especially relevant from the point of view of women’s employment. Thus, the desirability of a new policy may be judged well from its impact on these areas and sectors.

A major problem in assessing the (gender-disaggregated) impact of liberalization policies is the problem of the counterfactual, that is the problem of assessing what would have happened if there were

no changes in the policies. Another major problem is related to the lack of sector-wise gender disaggregated data on a time series basis. If changes in the employment of men and women have to be examined in the context of liberalization, detailed data on employment in export sectors (and other sectors) is required on a long-term basis. Unfortunately these data are lacking in most developing countries.

In this section, we present a brief review of various sources of employment data in India and some broad observations on women’s employment based on these sources. We then comment on the possibility of compiling information on the employment of men and women in major export sectors in India. We point to the general lack of such data from official sources. However an attempt is made here to arrive at estimates of employment levels and of the employment intensity of women in some major export sectors; we also analyse changes in employment patterns in the export sectors during the reform period based on these data. Employment data are available for the major industrial divisions according to National Industrial Classification (NIC) codes and export data are available as per the International Harmonized System (HS) codes. The sources for these are the National Sample Survey Organization’s (NSSO) surveys and the Ministry of Commerce (Government of India) respectively. Activities according to the NIC classification have been grouped together under various export product heads and accordingly NSSO estimates on employment have been assigned according to these product categories.

Three primary sources of employment data in India are: the Ministry of Labour’s estimates on organized sector employment; Indian Census data; and the National Sample Survey Organization employment surveys on employment which include 38th Round (reference year 1983–84), 43rd Round (reference year 1987–88), and 50th Round (reference year 1993–94). Indian Census data sources provide data on the employment of men and women in nine major industrial categories, by the rural-urban division, for India as a whole and also state-wise. Ministry of Labour data are based on surveys of establishments of a certain size in the public and private sectors. These data are available with a number of years of lag. Both the sources provide data at a fairly aggregated level. The NSSO data are based on sample surveys and are said to be more detailed and reliable. These are available at a disaggregated

three-digit level giving details of men and women employed in various activities in various sectors. The limitation of these data is that they are based on sample surveys, and therefore, are likely to underestimate total employment, and that they are available with a gap of four years. We present the main facts on the employment of women from these sources. In the next section, we use the three digit level detailed data from the 43rd and 50th Rounds of NSSO surveys to arrive at estimates of employment in major export sectors, and to analyse changes in women’s employment in these.

The services sector is another important emerging sector, which contributes to a large share of the GDP, and to the export sector in a number of Asian countries such as India. Recent world trade in software is estimated to be $ 250 billion, of which India’s share is 1.6 per cent ($ four billion), which is much higher than the share of India’s exports in world merchandise exports (0.6 per cent). There is great potential for increasing India’s exports and its share in world trade both in software and information technology enabled services. However, in this essay, our discussion is limited to the manufacturing sector.

There are two aspects of women’s employment which are important, one is their employment as a percentage of the total female population, indicating the female work participation rate and the other is women’s employment as a percentage of the total number of persons employed in a sector indicating the female intensity of production in that sector. What is more often analysed is the female work participation rate. We calculate and examine the implications of changes in the latter also as it gives the contribution of women’s employment (vis-à-vis men) in a specific sector.

Ministry of Labour data on employment

Data on the occupational pattern of women employees are provided by the Directorate General of Employment and Training (DGE&T), the Ministry of Labour, and the Indian government’s report entitled ‘Occupational educational pattern of employees in India’. This report presents the occupational pattern, its analysis and the educational profile of employees in the private sector as well as the public sector. It is based on data collected through the occupational and educational enquiry in respect of private sector

establishments coming under the purview of the Employment Market Information (EMI) programme.

The EMI programme covers the organized sector of the economy, which includes: (i) all establishments in the public sector (except defence establishments and armed forces); (ii) non-agricultural establishments in the private sector employing 25 or more persons on a compulsory basis and establishments having 10–24 workers on a voluntary basis.

Occupational studies conducted by DGE&T provide data on the occupational distribution of white collar and blue collar workers as well as owners, proprietors, working partners or directors of firms, who are engaged on a full time basis in the establishments. A major limitation of these data is that they are likely to grossly underestimate the employment of women (and men) as they do not cover self-employed, part-time employees, agriculture and allied occupations in the private sector, household establishments, establishments employing less than 10 workers in the private sector and defence forces. The data are likely to underestimate women’s employment more as women are largely in self-employment and in smaller household establishments.

Table 4 below has been compiled from the private sector issues of ‘Occupational education pattern of employees in India’ (OEPI) for the years 1985 and 1991.

The share of women in total employment in the private sector increased from 17.8 per cent in 1985 to 20 per cent in 1991. Of the 10.25 lakh women employees in 1991, most (47.9 per cent) were employed as farmers, fishermen, hunters, loggers and related workers and the next large share (32 per cent) as women professional, technical and related workers, a group dominated by teachers and nurses. Data from the OEPI issues on the public sector, from its latest issue of 1992 (reporting data collected for 1984) have been presented in Table 5 below. These data show that women employees constituted about 11.3 per cent of the total employment in the public sector in 1984. Examining the share of women in different occupational categories we find that most of the women employees were professional, technical and related workers (and among them 65.5 per cent were teachers particularly in primary and middle schools). Clerical and related workers constituted the second largest group. When looking at the occupations which accounted for a major proportion of the women


Occupational Pattern of Women Employees (private sector), 1985 and 1991


Source: Ministry of Labour (various issues): OEPI; female intensity calculated by the author.

as a percentage of the total women employed in all the occupations, we find that clerical and related workers constituted the single largest occupational category accounting for 30.8 per cent of the total of women’s employment. Professional, technical and related workers had the second largest share of 26.9 per cent of the employees. Production and related workers and transport equipment operators and labourers followed closely, constituting 24.4 per cent of the total of women’s employment.

As mentioned earlier these data have limitations in coverage and may underestimate women’s employment in both the private and public sectors. Other limitations of these data include non- response by the establishments;3 data not updated by employment exchange;

3For example, out of 1,21,468 establishments only 90,515 responded in the 1984 public sector survey and only 64.5 per cent responded to the 1991 private sector occupational enquiry.

information furnished by the establishments not being in correct form and incomplete information. However, these data do suggest useful trends and facts about women’s employment shares and confirm evidence which suggests that women are concentrated in agriculture and low paid professional services.


Occupational Pattern of Women Employees (public sector), 1984


Indian Census 1991 and the employment of women

Indian Census years include 1951, 1961, 1971, 1981 and 1991. The Census provides data on labour force participation rates, separately for men and women. As per the 1991 Census, the female work participation rate was estimated at 22.3 per cent in 1991 (increased from 19.7 per cent in 1981) and the male work participation rate was more than double (than that for female participation) at 51.6 per cent in 1991, slightly lower than 52.6 per cent in 1981 (Government of India, 1993). Sector wise Census data show that women are primarily in the agricultural sector (81 per cent of women in 1991); and within the primary sector, women are primarily agricultural labourers (about 45 per cent of female main workers), and cultivators (34 per cent of female main workers). About seven per cent of female main workers are in the manufacturing sector and 11 per cent of women workers are in the

services and trade sector. Female intensities of production have been calculated by the author based on Census Data on ‘main workers’. These data show that women form 27 per cent of main workers in the primary sector, 15 per cent in the secondary sector and 12 per cent in the tertiary sector. Within the primary sector, women constitute 38 per cent of the total number of agricultural labourers and 20 per cent of the total number of cultivators. Women contribute highly to the manufacturing sector. They constitute 16.4 per cent of the workforce of this sector where they are more concentrated in household (home) manufacturing and where they form 33 per cent of the total workforce. Women contribute to the services sector as well. They constitute 12 per cent of the workforce in this sector; and are concentrated in services other than trade and transport.

Census data has the advantage of wide coverage and a large sample size. However, these official estimates are likely to underestimate women’s work in four areas (a) subsistence production; (b) domestic production and related tasks; (c) unaccounted paid work; and (d) volunteer work (Beneria, 1993). The underestimation is a result of both practical and conceptual problems. Women’s work is likely to be undervalued at micro and macro levels due to the restricted definition of economic activity; and the low intrinsic human value attached to activities that expand human choices and women’s household, community work, and caring, are not counted (for a detailed discussion, see Senapaty, 1997:37–53).

NSSO data and women’s employment

NSSO quinquennial surveys on employment (and unemployment) in India were carried out in the following rounds: 1) 27th round (1972–73); 2) 32nd round (1977–78); 3) 38th round (1983); 4) 43rd round (1987–88); 5) 50th round (1993–94). In these surveys, estimates are generated on employment according to the ‘usual status’ based on a reference period of one year, the ‘current weekly status’ based on a reference period of one week, and the ‘current daily status’ based on each day of the previous (reference) week (NSSO, July-September 1996).4 The NSSO surveys on

4Apart from these usual employment rounds (which provide data on employment for specific reference years), which are with a gap of four years, NSS also started annual surveys for collection of annual data which are aimed at

employment aim to measure the extent of employment in quantitative terms, disaggregated by various household and population characteristics. They also provide data and more detailed industrial division of workers in a three-digit industrial classification based on the 0–9 major divisions. Approximately 384 divisions of industry are classified by the NSSO.

For an analysis of changes in the employment of women in the major nine industrial categories, published data from the last three NSS Quinquennial rounds on employment, namely the 38th, 43rd and 50th rounds have been used to calculate female work participation rates in major industrial categories and calculate female intensity of production in these categories (Table 6). Later, the detailed analysis on working out estimates of employment of women in major export sectors is based on the three digit employment data as per the NSS 43rd and 50th rounds.

Data show that in the last decade there have been marginal changes in the structure of women’s employment. The proportion of women employed in agriculture declined from 81 per cent in 1983 to 77 per cent in 1993–94. The largest proportion of women continued to be employed in agriculture. The manufacturing sector employed the next largest proportion (around 9 per cent of the female labour force). The proportion of women working in the services sector increased from about 5.6 per cent in 1983 to 7.9 per cent in 1993–94.

Female intensity of production during 1983 to 1993–94 declined in agriculture, construction, trade, transport, while it increased in the manufacturing sector, electricity, gas etc., and services. During 1993–94, women formed about 39 per cent of the workforce in

generating annual indicators of employment and unemployment. These started from the 45th round survey operations. There are five such surveys ending with the 49th round with a reference period of January-June 1993. The information collected is aimed to classify the population as belonging to different activity categories on the basis of the activities pursued by them during the specified reference periods, adopting usual status and current weekly status approach. These data are available at the one digit level. The size of the sample considered in the annual surveys is approximately one-fifth the size of the sample considered in the usual employment rounds. For example, in the 46th round, the number of sample villages = 6984, and the number of sample blocks = 7477. The corresponding figures for the 49th round were 4822 and 2791, respectively.


Employment of Women During 1983, 1987–88 and 1993–94


*This is defined a female intensity of production.

agriculture; about 29 per cent in manufacturing; and about 24.11 per cent of those employed in the services sector; men formed a minimum of 60 to 80 per cent of the workforce in various categories.

Employment of women in export sectors in India

Here, we try to specifically trace any changes in the employment of women in major export sectors. As mentioned earlier, employment data are not separately available for export sectors. We examine the official export data for the years 1993–94 to identify major export sectors5 and for estimating the employment of men and women in these export sectors, we classify and match NIC industrial codes at the three-digit level against these sectors. The matching of export products and industrial codes is given at Annexe

6. Published NSSO data on per thousand distribution of usually employed, classified according to industry groups and sex in the manufacturing sector for the 43rd and 50th employment rounds, was used to calculate employment levels for men and women for

5Major export sectors have been identified based on the value of India’s exports.

these major export sectors.6 The results from the data for 1987–88 and for 1993–94 giving employment of women in various export sectors and calculated female intensities for each sector are presented at Annexe 7 and Annexe 8.

In order to simplify the analysis we have grouped the female intensities in four categories, namely very low, low, moderate and high according to the shares of female employment. If the share of female employment in a particular export sector in the total employment in that sector is less than 10 per cent, it is categorized as having very low female intensity of employment; if the share is between 10–20 per cent, we call that low female intensity, a share of between 20–40 per cent to imply moderate female intensity and over 40 per cent share of female employment in any sector is categorized as a sector with a high female intensity.7 Sectors giving their female employment shares as shown in Annexe 7 (for 1987–88) and Annexe 8 (for 1993–94) have been grouped based on the above mentioned categories and summary results are shown in Box 1 below. Further, in order to enable easy comparison between 1987–88 and 1993–94 in Box 1, sectors underlined show no changes in female intensity between two years, sectors shown in bold show an increase in 1993–94 and sectors showed in italics show a decrease in female intensity between 1987–88 and 1993–94.

Data for 1987–88 (43rd round) and 1993–94 (50th round) show that very few sectors have a female employment share of more than 40 per cent, implying that in most sectors male employment share is 60 per cent or more. The sectors where women have high female intensity (over 40 per cent) include tobacco, zari work in both the years; and cane and wood products, oil and ghee, petroleum refined, gas and fuel products, made up textile articles, musical instruments and sports goods in the year 1987–88; and some other sectors such as wool, silk and textile fibres, jute and coir fibres, coir and jute (bleaching, dying printing), common salt in 1993–94 showed high shares of female employment of over 40 per cent.

6Since these rounds used different versions of the NIC codes (the 43rd round used NIC-70 classification, while the 50th round used NIC-87 codes), therefore we were required to take the conversion between the two versions into account.

7There is nothing sacrosanct about these categories, these have been assumed on the basis of evidence on females shares as shown by data, which tend to be below 40% in most categories.


Note: n.e.c = not elsewhere classified.

Source: Compiled from Annexe 7 and Annexe 8.

A number of major industrial/export sectors which are potentially important for their employment implications such as gems and jewellery, engineering goods; copper, brass and other metal products; leather and substitutes and footwear have low shares of female employment of less than 10 per cent in 1987–88. Export growth in these sectors has not resulted in a large increase in the share of female employment, which continues to be less than 10 per cent in 1993–94. Another set of important industrial categories such as the manufacture of cotton yarn fabrics, made up textiles, tailoring and garments, glass and ceramic products, also have moderate female intensity of 20 per cent to 40 per cent. Efforts to increase these shares would be crucial especially in the context of economic reforms and changing global trade practices.

Data given in Annexe 7 and Annexe 8 have been further used to calculate the employment of women in major export sectors and female employment intensities in these sectors in 1987–88 and 1993–94; and these are given in a summary Table 7. Given that the manufacturing sector contributes over 76 per cent of India’s exports, the share of female employment in this sector as a whole has crucial implications for poverty and livelihoods of women and their families. Estimates on employment disaggregated by export activity and domestic production are not available and therefore estimates in a particular sector taken here show total employment only in that sector. An assumption can be made that the share of employment in the export sector is a certain percentage of the total employment in that industrial sector on the basis of the share of total production that goes for exports in that sector.8 But we have not made such an assumption as the percentage of production that goes for exports varies widely among different sectors and official data for these are not available. Also, in examining changes in female intensity after reforms (1993–94) compared with the pre-reform period of 1987–88, our analysis would not be affected if a uniform assumption is made for both periods.

8For simplicity, we could have made a uniform assumption of employment due to exports in a particular sector to be 76 per cent of total employment in that sector assuming that 76 per cent of production (and therefore employment) is for exports.


Employment of Women in Export Sectors and Sector-wise Female Intensities: 1987–88, 1993–94


Source: Compiled from Annexe 7 and Annexe 8.

Data on exports show a large increase from a value of US$ 12 billion in 1987–88 to over US$ 22 billion in 1993–94; an increase of more than 83 per cent in this period. The export of manufactured goods increased by over 225 per cent (Annexe 5); against this, the increase in female employment in the manufacturing sector during this period was only 20 per cent. The female intensity of employment in the manufacturing sector showed a slight increase (almost negligible) from 30.18 per cent in 1987–88 to 30.33 per cent in 1993–94 (Table 7). In terms of absolute numbers, female employment increased from 8.35 million in 1987–88 to 10 million in 1993–94. The increase (even though negligible) in female employment corroborates the evidence in some South Asian countries on the feminization of the labour force.

However, it is not possible to conclude whether this increase was due to increased opportunities in certain sectors because of the opening up of the economy or because of women having been forced into the labour market as a result of increased poverty.9

The five major manufacturing export sectors for India are textiles, gems and jewellery, engineering goods, chemicals and related products, and leather and manufactures. A quick analysis of export data show that these sectors primarily contributed to the large increase in the value of exports from the mid-1980s to 1993–94 (Annexe 5). Textile exports account for over 33 per cent (1993–94) of India’s manufacturing sector exports, making this the most important export sector. It is also one of the most labour intensive sectors with a lot of the production being in the small-scale sector. Textile exports increased from US$ 1.7 billion in 1985–86 to US$ 4.9 billion in 1993–94. Employment in the textiles sector increased from about 9 million to 11 million, but female intensity remained constant at 37 per cent in both the periods. The export of gems and jewellery increased from US$ 1.2 billion in 1985–86 to US$ 4.0 billion in 1993–94 and the female employment intensity increased from 2 per cent to 9 per cent during this period. Engineering goods account for about 18 per cent of India’s manufacturing exports (1993–94), which showed an increase of over 288 per cent in the value of exports during 1985–86 and 1993–94. The share of female employment in this sector continued to be limited (at less than 4 per cent), despite an increase in female employment (from 0.11 million to 0.14 million) and an increase in female employment intensity (from 2.8 per cent to 3.8 per cent) during 1987–88 and 1993–94. Chemical and related products account for about 11 per cent of India’s manufacturing exports; the value of these exports increased by over three times from 1985–86 to 1993–94. The employment of women increased in this sector, and the female employment intensity also increased marginally from 24.7 per cent to 25.2 per cent during this period. Leather and footwear also contributed about 12 per cent of India’s manufacturing exports in the mid-1980s, which declined to 7.7 per cent in 1993–94; although the value of exports

9It is possible that the feminization of the labour force would be more evident if services sector is also included as a large number of women may have entered petty trading or such like informal sector services as argued earlier.

more than doubled during this period (Annexe 5). The employment of women and female employment intensity in this sector increased during this period (Table 7).

Overall, the major sectors that have contributed to the all round increase in the female employment intensity between 1987–88 and 1993–94 include leather and footwear, paper and paper products (due to printing, newsprint etc.), chemicals and products, engineering goods and gems and jewellery. The sectors where a decline in the share of employment of women was witnessed between 1987–88 and 1993–94 include agriculture and allied products, cane and wood, paper and paper products, cotton yarn, fabrics, cotton bleaching, dyeing, blankets and shawls (details in Annexe 7 and Annexe 8).

This analysis suffers from some limitations. As mentioned earlier, a complete conformity of export products and employment codes (based on NIC industrial activity three-digit codes) was not possible. For example, it was not possible to estimate the employment of women in the readymade garments sector separately as there was no separate three digit classification with respect to garments. Also it was difficult to classify all activities such as bleaching, dyeing, the manufacture of cotton yarn, tailoring etc. under the product description of ‘garments’. Another set of limitations is due to data inconsistencies with respect to some major export sectors. The first set of inconsistencies are due to the non-coverage or incomplete coverage of some major sectors in one of the two NSS rounds, which makes comparison in employment shares difficult between the two time periods. For example, for manufacture of the tea sector, employment data are available for the 50th round (1993–94), but no data are given for the 43rd round (1987–88). This is likely to be due to non-coverage of tea in the 43rd round. Of the coffee production, about 50 per cent is exported, there are no data on employment for the coffee sector as per both the surveys. Similarly, leather and substitutes, weaving apparels etc., have employment data as per the 50th round but no data for the 43rd round; coal, coke etc., have data for the 43rd round but not for the 50th round; tailoring, garments and accessories shows high female intensity for the 50th round but no data are shown against this item in the 43rd round. The complete data for manmade textiles are available for the 50th round but these are incomplete for the 43rd round; Made-up textile articles show a

large decline (67 per cent to 21 per cent) during 1987–88 and 1993–94, but this could be due to the incomplete coverage in the 50th round as no rural areas data are provided in this round. Other sets of inconsistencies occur because in some cases data are available for both the rounds for total employment, but no data are shown against female employment, as for example in the case of synthetic textile fabrics/dyeing, bleaching etc. It is unlikely that there are no women employed in this sector.

Therefore, we find that, in undertaking a comparative analysis on employment shares of women based on published results of the two NSSO surveys covering two time periods, there could be a problem of incomplete/different coverage of industrial activities. The reasons for such inconsistencies are not discussed or documented in the documents that publish the survey results. The other limitation of this analysis is that the employment data used are only for the manufacturing sector and therefore total employment would be underestimated to that extent. We could expand this analysis further by including employment in those activities which are associated with the production of intermediate goods used to make up the final exported products, for example in the category of ‘marine products’, activities such as fishing, processing and preserving of fish could be included; in the category of textiles, activities like rearing of silk worms and sheep, growing of cotton, etc., can be included. It is also not clear whether NSS data on employment covers the employment of persons in the Export Processing Zones (EPZs); employment data for these sectors would provide the employment of men and women directly contributed by exports.

We also examined the data sets published by the Central Statistical Organization as a part of the National Account Statistics (NAS) on employment of men and women based on the 50th round of the NSS series at the three digit level. We used the same categorization as earlier (Annexe 6) and calculated female employment, total employment and female intensities as per major export sectors. These data are summarized in Table 8. A comparison of Tables 7 (data for 1993–94) and 8 shows that employment levels as given by NAS based on the NSS 50th round are higher than those calculated by us on the per thousand employment data from the same source as published by the NSSO. The reasons for these are not documented in any of the sources.

However, female intensity in major export sectors as per NAS data source (Table 8) is similar to that calculated on the basis of published NSSO data source (Table 7), reconfirming our results on female intensities and conclusions on low shares of employment of women in total employment in the major export sectors during 1993–94. The reasons for differences in the two sets of data need to be investigated further especially if the need for a consistent and reliable data base is to be addressed and if this has to feed into the planning process.


Female Intensity in Export Sectors as per National Accounts Statistics: 1993–94


Source: CSO: New series on National Accounts statistics base year, 1993–94.

World Trade Organization and the Textiles Sector

The Uruguay Round of General Agreement on Tariffs and Trade (GATT) marks a major change in the rules of the international trading system. The impact of changes in trading opportunities resulting from the implementation of the Uruguay Round may be sector specific and region specific. Women employed in the manufacturing sector may benefit the most together with women employed in the service trade. The women in the agricultural sector,

where the returns on their labour are less clear than under wage-labour contracts, are likely to benefit less. Agriculture and textiles (including garments) are two key export sectors for South Asia and are also major employers of women, either in wage employment or as unorganized labour in the informal sector. Changes in these sectors following international policies and new trade patterns will thus have a direct impact on the lives and livelihoods of millions of women in this region. As per our data (Table 7), even though admittedly gross underestimates of numbers show that about 39 per cent of the total female employment was in the textiles sector in 1987–88, which declined to 33 per cent of total female employment in 1993–94. The share of female employment in the total textile sector employment (female intensity) actually showed an increase during this period and therefore an increase in exports in this sector would (ceteris paribus) be beneficial for women’s employment in countries where this occurs.

There has not been any serious investigation into how the new arrangements are going to affect the vast majority of Indian workforce. Even less is known about the impact it is going to have on women who constitute more than 50 per cent of the sector. Various studies try to estimate the impact of liberalization on the clothing sector. Most of these conclude that developing countries should gain from the abolition of Multi-Fibre Agreement quotas. Page and Davenport (1994) estimate that countries likely to gain most are China, India and Pakistan, South Korea and eastern Europe, with economies of scale considerations becoming important. Countries like Mauritius and Jamaica are expected to lose because they do not have a comparative advantage in producing clothing but instead are exploiting EU (and US) preferences in this sector (Davenport et al, 1995; 67–70).

According to a recent study (Wood and Calandrino, 2000), given low levels of education, the future concentration of India’s exports is likely to be on labour intensive goods such as clothing (and also software); but these exports will also have to face stiff competition from the rest of South Asia. However, if this is associated with new approaches to production, with new technologies and product diversification as the sector expands, it may lead to the adoption of shifts in labour composition in favour of men and this could result in a decline in female intensity in this

sector. In the short run, geographical origins and buyers preferences will remain important, but in the long run, cost and product differentiation factors will acquire greater importance. In order to be able to examine the impact of changes in various agreements under the WTO on women’s employment, the data on existing labour composition and changes in this over time need to be available and studied carefully.

Textiles and garments sector in South Asia

The textile sector10 is one of the most important export sectors in South Asia. In Bangladesh, export earnings of garments are estimated to have increased from US$ 800 million in 1987 to US$ 3500 million in 1998 and the share of export earnings from this sector increased from just 12 per cent in the mid-80s to over 73 per cent of the total export earnings in 1998. India is less dependent on this sector in terms of its share in total export value, with textiles (including garments) constituting about 25 per cent of total exports, though it still remains one of the single most important export sectors for India. For Pakistan and Sri Lanka again, this sector remains very important with contributions of 60 per cent in Pakistan’s export earnings and 52 per cent of the total export earnings of Sri Lanka (Ramaswamy and Gereffi, 1998).

Brief History of the ATC

World trade in textiles and clothing has been governed by a range of quantitative restrictions under the ATC (earlier under MFA). Countries are given exact quotas for export to certain markets (primarily the US and Europe) on an annual basis. The ATC proposes to phase out this arrangement over a ten year period ending 2005 with its integration into the WTO framework. The importing countries are allowed to pursue different policies on the timing and selection of products in the phase out, within the parameters of the general rules laid out in the Agreement.

Trade in textiles was outside the General Agreement on Tariffs

10For the purposes of this section textiles sector includes garments and clothing.

and Trade (GATT) rules from the early 1960s when the Short Term Arrangement (STA) was established to regulate such trade. STA was followed by Long Term Arrangement (LTA), which operated up to 1973. This was replaced by the Multi Fibre Arrangement (MFA) in 1974, which continued GATT inconsistent restrictive measures of STA and LTA. Under the MFA, quota measures were extended for manmade fibre and woollen products in addition to cotton products. The MFA was to operate only for a limited period of four years, to provide temporary protection to the textile industries of the developed countries to relocate their investment and manpower from the labour intensive sector to other capital intensive sectors. However, the MFA was extended again and again and it continued from MFA I (1974–77) to MFA IV (1986–94). From 1987, the MFA was expanded further to include vegetable fibres and silk blend products.

The Uruguay Round negotiations in 1994–95 proposed integration of the textile and clothing trade into GATT through the Agreement on Textiles and Clothing (ATC) which replaced the MFA in 1995. The process of integration intended to cover ‘the phasing out of restrictions under the MFA and other restrictions of textiles and clothing not consistent with GATT rules’. The process of integration was to be progressive in character.

Since the share of developing countries in trade in textiles is large, and since clothing is a labour intensive activity, liberalization of trade in this sector through the ATC was seen as principal area of benefit for developing countries from the Uruguay Round. The negotiations gave rise to expectations in the developing countries that competitiveness and comparative advantage in this sector may put them at a greater advantage. From the Uruguay Round, the volumes of world trade were estimated to increase by 6–20 per cent and more than 1/3rd of the benefits were expected to derive from liberalization of textiles and clothing, with obvious implications for the developing countries of South Asia.

Experience with ATC implementation 1995–99

The MFA is said to have been responsible for restricting potential trade in textiles and garments and for leading to inefficiencies by putting barriers to trade. This is believed to have worked against the interests of the developing countries, for whom there have also

been high administrative costs of implementation of quotas and excessive government interventions through the allocation of quotas. The slow implementation of ATC would therefore imply further accentuation of inefficiencies and protection, which continues to favour developed countries against the interests of the developing countries.

Article 1 of ATC establishes certain basic principles for the implementation of the Agreement. These relate to increasing quota access for small suppliers and LDCs; reflecting the interests of cotton producing countries with members allowing increased competition in their markets. The experience of its implementation has, however, been quite different. Contrary to the principle of meaningful increase in access for small suppliers (Article 1:2 of the ATC), additional access in quotas for small suppliers has been only marginal.11 In addition, there have been a number of safeguard actions by US involved exports from small suppliers (e.g., Columbia, Costa Rica etc.). Both Canada and the US gave the benefit of additional access in certain cases of LDCs. However, in the case of some countries of South Asia e.g., Bangladesh, similar terms have not been accorded by Canada or the US. As regards the reflection of interest of cotton producing exporting members (Article 1:4 of the ATC), the restraining countries claim that they have held consultations with these countries but no information has been given by them with respect to the manner in which the specific provision may actually have been reflected in the implementation. It is clear that no additional quota access has been granted to these countries by any restraining country. Information on the implementation of provisions (Article 1:5 of the ATC) relating to allowing increased competition in the markets of restraining countries is not readily available. According to available information Norway and Canada did adopt some specific measures, for example, the removal of quota restrictions (except on fishing nets from Indonesia, Malaysia and Thailand) to allow for increased competition in their markets. However, no action has been taken

11This is partly because different methodologies are used by the EU on the one hand and by the USA and the Canada on the other in giving effect to the provision of Article 2:18. It is estimated that annual additional access for small suppliers comes to only 1.17% by methodology used by Canada and the US and comes to about 3.5% if EU’s methodology is used.

to facilitate increased competition by the two largest restraining members, the US and the EU. The US has announced postponement of the integration process until the end of the transition period of ATC (2005) and the EU is insisting on additional reciprocal market opening commitments by the developing countries before it considers liberalization of its restrictions.

The provisions of Article 2 relate to the integration process and the additions in access by increases in quota growth rates. Paragraphs 6–8 of Article 2 establish the obligation of integration of textile products at the start of each of the three stages; 16 per cent from January 1995; 17 per cent from January 1998 and 18 per cent from January 2002. The remaining 49 per cent will automatically stand integrated at the end of the transitional period (January 2005). These integration percentages are provided as the absolute minimum required at the start of each stage. The implementation of the integration programmes for stages 1 and 2 (combined) of WTO members maintaining restriction under the MFA show that products selected for integration are less value added items such as tops, yarn and fabrics, with very small share of clothing products. An assessment of the actual elimination of the restrictions shows that out of a total of 750 quotas for USA, only 13 quotas have been eliminated. For EU, out of 219 total number of quotas only 14 have been eliminated. In the case of Canada out of 295 quotas, only 29 have been eliminated. Where Norway is concerned 51 out of 54 have been eliminated. It is also seen that the trade accounted for by-products that have been freed of quota restrictions represented only small shares in the total restrained imports of the two largest members of the WTO: about 6 per cent in the case of US and less than 5 per cent in the case of EU.

It is therefore clear that although 33 per cent of the trade has been integrated to fulfil the minimum legal requirements of the Agreement, the process has contributed little towards the objectives of the ATC i.e., progressive phasing out of quota or liberalization of trade.

Article 2 (Paras 13 and 14) provide for additional increases in quota access, according to which the countries are required to increase the existing growth rates of quotas by a minimum of the following percentages in each of three stages: 16 per cent during stage one, another 25 per cent for stage two and a further 27 per

cent for stage three. In the first two stages (7 years) of the transition period, the position of increase in additional access is seen to be minimal, especially because the pre-existing growth rates from the MFA are generally small as compared to its norm of 6 per cent. The additional access looks especially small given the healthy consumer demand in these countries. The additional access has therefore not meant lessening of the existing quotas. As per the agreement it is required that respective programmes of integration shall be notified. In most cases the countries have notified only the list of products to be integrated and these do not specify consequential action to be taken (such as conditions that export certificates will not be required any more); leaving ambiguity for exporters about the future course of action. Only Canada notified that it integrated the tailored collar shirts and that it would not require the shipment to be accompanied by export certificates. In certain cases the visa requirements are still made by the US even for products which have been integrated by the US in stage 2.

Rules of origin

The US has made changes in the rules for determining the origin of textiles and clothing products, because of which exports to intermediary countries have been adversely affected and this has disrupted the normal pattern of trade. Under the old rules, the origin of fabric was determined either by where the fabric was formed or if the fabric underwent dyeing plus printing, its origin was determined by where the fabric was thus processed. According to the modified rules, the origin of fabric is determined by where the fabric is formed irrespective of where it may have been dyed, printed or finished etc. For made up articles of HS Chapter 63 (bed linen, table linen etc.) the origin is determined by where the constituent fabric is formed. For garments (for finished clothes), the origin depends on where the sewing was done and for simple assembly items the place of cutting to shape was the origin. This system recognized the multi-processing choice prevalent in the world these days. According to the modified rules, the origin is on an ambiguous standard i.e., where the most important assembly operations are undertaken. Changes in the rules of origin have caused a lot of concern and these rules of origin are different in different countries. The restraining countries are also adopting a

number of non tariff barriers which include inspection of production and other records, frequent factory visits, special certificates of origin from certain exporting countries containing information about production capacity, machinery installed, number of workers etc. Sometimes the introduction of new restrictions is sought on the basis that the exporting countries are circumventing the quotas. These create extra hardship for the exports of developing countries. Countries are also resorting to anti-dumping actions targeted at textile products. A number of anti-dumping investigations did not result in the imposition of anti-dumping duties. However, companies in the developing countries have had to face considerable hardship and there have been other damaging effects. This has in turn affected imports from countries which have been subjected to anti-dumping action, for example, in the case of cotton fabrics the total imports during 1997 (over 1996) declined by about 7 per cent but import from some countries targeted for anti-dumping investigations declined by over 33 per cent.

The above analysis suggests that most of the products under the quota have not been liberalized. Products selected for integration are less value added items. Even for those that have been liberalized, the necessary notifications have not been issued because of which the quota liberalization is only on paper. In addition, problems are being created with respect to the rules of origin and by the imposition of other restrictive practices in terms of non-tariff barriers because of which exports to intermediary countries may have been adversely affected and may have disrupted the normal pattern of trade. Thus implementation of ATC so far has failed to come up to the expectations of the developing countries. It is also true that at the same time South Asian countries have to take action domestically to improve their competitiveness through productivity increases; improvement in infrastructure facilities, improving the investment climate and attracting foreign direct investment linked to exports in this sector, as well as diversifying in terms of products and markets.


There has been an increase in global trade activities and trade growth has outpaced domestic output growth in recent years.

Important reviews suggest that globalization has not benefited developing countries as expected in terms of growth and poverty reduction. Available evidence on trade and women’s employment suggests that the growth of export oriented manufacturing has benefited them, especially as it has created more jobs and may have increased their influence on household decisions, while empowering them in some cases, although their wages are lower than those of men. Evidence on working conditions is inconclusive; some studies suggest poor conditions and poor pay. There is a need to create a new policy tool, which can enable the assessment of the human development effect of employment in terms of both the quantity and quality of work.

The case of India shows that in terms of the macro score card, India has had a mixed performance, and the continuing reduction in poverty at the same rate as evidenced in the late 1980s before reforms, is doubtful during the reform period. India achieved an impressive growth in exports after reforms, but it has slowed down in recent years and the value of India’s exports compares poorly with some east Asian countries and China. India’s export basket is narrow and foreign direct investment linked to exports is lacking. There is a need for a radical strategy to achieve high export and employment growth in India, with other strategies and interventions in improving the investment climate and infrastructure.

An analysis of changes in the employment of women (and men) in the export sectors is limited because of the lack of availability of data on employment in various export sectors. A preliminary attempt to estimate the employment of women in the manufacturing export sectors in India and analyse changes in female intensity of employment in these sectors during the reform period suggests that the female share in total manufacturing sector employment increased between 1987–88 and 1993–94, but only marginally. In major export sectors such as gems and jewellery, leather and footwear and engineering products, females form a small share of total employment. In the most important export sector like textiles, female employment intensity has remained stagnant. In some other major sectors like agriculture and allied sectors, musical and sports goods, it has shown a decline. This analysis thus suggests limited increase in the share of female employment in major export sectors during the early reform period in India. There is a need to examine employment trends in major export sectors on a regular

basis, and document changes in the policy environment so as to suggest corrective measures. The need for developing mechanisms for a consistent data set for employment in the export sectors should be taken up with the national official agencies in South Asia. These data would also be required to examine the effects of import displacement as a result of the removal of quantitative restrictions under the WTO. This analysis also needs to be supplemented by an analysis of employment trends in the informal sector and the home-based sector because a large percentage of women in South Asia are in these sectors. Also because changes in the policy environment will impact the informal sector mediated through changes in the formal sector.

Analysis on the implementation of the ATC so far shows that during the transition phase, the major restraining countries have not kept their commitments under the ATC. On paper only about 33 per cent of the trade has been integrated but in practice only a few quota restrictions have been actually liberalized. Effective market access for developing countries is impeded by the large back loaded level of integration determined by the developed countries. The implementation of ATC in major importing countries for the destination of textiles of South Asian countries, namely the US and the EU, is particularly slow and small so far. Also the tariff protection rates in textiles are reportedly much higher than that in other manufactured products: 18 per cent in textiles as compared with six per cent on other industrial goods. Given the importance of this sector for the region and the fact that implementation has so far been limited, whatever benefits there are, will remain limited till 2005, and will continue to be so if action is not taken. Another limitation of the South Asian textile sector is that these countries compete with each other in terms of export items and in terms of the destination of their exports. There is a high concentration of a few items in their exports and more than 90 per cent of total garment exports of south Asian economies are to the US and the EU countries. So any hope that there may be potential gains from the region needs to take account of these details. The countries of South Asia need to come together, demand greater information and transparency in the implementation process and develop support for negotiating better implementation of ATC and better access for their exports in this sector.


Share of Regions in World Merchandise Exports by Region, 1980–97


Including Baltic states and CIS countries

Source: WTO (1998): 19


Comparative Economic Growth Performance in First Three Years After Start of Reform/Adjustment Process


Note: Year in parenthesis indicates year reform or adjustment programme was launched

Source: Thomas et al. (eds.) (1991) quoted in Acharya (1999) Table 3.


Growth Rate of Exports vis-à-vis World, 1992 to 1999


Source: Ministry of Commerce for Indian data and WTO (1998) for world data.


Comparison of Selected Debtor Countries, 1997


Source: World Bank (1999): Global Development Finance, Country Tables.


Composition of India’s Manufacturing Sector Exports 1985–86, 1993–94


*Including jute and coir manufactures


Matching Table of Export Sectors and NIC Industrial Codes



Source: Export Codes from DGCI&S (HS classification) and NIC Codes from CSO (1998).


Female Intensities in Major Export Sectors, 1987–88



Source: NSSO (43rd Round); Compiled and female intensities calculated by the Author.


Female Intensities in Major Export Sectors, 1993–94



Source: NSSO (1996): S-162 to S-177; Compiled and female intensities calculated by the Author.


Female Intensity in Export Sectors as Per National Accounts Statistics: 1993–94


Source: CSO: New series on National Accounts statistics base year,, 1993–94.


Acharya, S. (1999) ‘Managing External Economic Challenges in the Nineties: Lessons for the Future’, Anniversary lecture delivered at the Central Bank of Sri Lanka, Colombo on 9 September, 1999

Afshar, H. and C. Dennis (eds.) (1992) Women and Adjustment Policies in the Third World, Basingstoke: Macmillan

Baden, S. (1997) ‘Recession and Structural Adjustment’s Impact on Women’s Work in Selected Developing Regions’, in Date-Bah, E. (ed.), (1997) Promoting Gender Equality at Work: Turning Vision into Reality, London: Zed Books/ILO

Baden, S. and S. Joekes (1993) ‘Gender Issues in the Development of the Special Economic Zones and Open Areas in the People’s Republic of China’, Paper presented at a seminar on ‘Women in the Economy’, Fudan University Shanghai, April 15, IDS (Sussex) Mimeo quoted in Fontana, M. et. al. (1998) Global Trade Expansion and Liberalization: Gender Issues and Impacts, Report no. 42: DFID

Banerjee, N. (1999) ‘How Real is the Bogey of Feminization?’ in Papola, T.S. and Alakh Sharma (eds.) (1999) Gender and Employment in India, New Delhi: Vikas Publishing House

Beneria, L. and A. Lind (1995) ‘Engendering International Trade: Concepts,
Policy, and Action’, Gender and Sustainable Development Working Paper Series No. 5, Cornell University, Ithaca

Bhall, S.S. (2000) ‘Growth and Poverty in India-Myth and Reality’, Oxus Research and Investments, research papers

Cagatay, N., Elson, D. and C. Grown (eds.) (1995), World Development, Vol 23 No 11, Special issue on Gender and Macroeconomics.

Cagatay, N. and S. Özler (1995) ‘Feminization of the Labour Force: The Effects of Long-term Development and Structural Adjustment’, World Development, 23 (11).

Chandrasekhar, C.P. and J. Ghosh (2002) The Market that Failed—A Decade of Neo Liberal Economic Reforms in India, New Delhi: Leftword Books

Commonwealth Secretariat (1989) Engendering Adjustment for the 1990s, Report of a Commonwealth Expert Group on Women and Structural Adjustment, London: Commonwealth Secretariat

Davenport, M. Hewitt, A. and A. Koning (1995) ‘Europe’s Preferred Partners? The Lome Countries in World Trade’, ODI Special Report, London: ODI

Davenport, M. and S. Page (1994) ‘World Trade Reform: Do Developing Countries Gain or Lose?’ ODI Special Report, London: ODI

Deshpande S. and L.K. Deshpande (1999) ‘Gender-based Discrimination in the Urban Labour Market’ in Papola T.S. and A.N. Sharma, (eds.) Gender and Employment in India, Delhi: Vikas Publishing House

Deshpande L.K. and Gerry Roders, (eds.) (1994) The Indian Labour Market and Economic Structural Change, Delhi: B.R. Publishing Corporation

Dev Mahendra S. (2000) ‘Economic Liberalization and Employment in South Asia-I’ Economic and Political Weekly, January 8

Elson, D. (1991) ‘Male Bias in Macroeconomics: The Case of Structural Adjustment’, in Elson, D. (ed). Male Bias in Development Process, Manchester: Manchester University Press

Fontana, M. and A. Wood (2001) ‘Modelling the Impact of Trade Liberalization on Women at Work and at Home’, DFID: Research Report

Fontana, M. et. al. (1998) ‘Global Trade Expansion and Liberalization: Gender Issues and Impacts’, Report no. 42: DFID

Francke, M. (1992) ‘Women and the Labour Market in Lima, Peru: Weathering Economic Crisis’, International Centre for Research on Women.

Ghosh, J. (1999) ‘Macroo-economic Trends and Female Employment: India in the Asian Context’, in Papola, T.S. and Alakh Sharma (eds.) (1999) Gender and Employment in India, New Delhi: Vikas Publishing House

Joekes, S. and Weston, A. (1994) Women and the New Trade Agenda, New York: UNIFEM

Kundu A (1997) ‘Trends and Structure of Employment in the 1990s: Implications of Urban Growth’, Economic and Political Weekly, June 14, 1997

Kundu A. (1999) ‘Trends and Pattern of Female Employment: A Case of
Organized Informalization’ in Papola T. S. and A. N. Sharma, (eds.) Gender and Employment in India, Delhi: Vikas Publishing House

Ministry of Commerce (2000) ‘Need for a Radical Strategy’, Government of India: MOC

Ministry of Commerce (1996) ‘Foreign Trade Statistics of India’, Government of India: MOC (DGCI&S)

Ministry of Finance (2002), Economic Survey, Government of India: Ministry of Finance

Ministry of Finance (2001), Economic Survey, Government of India: Ministry of Finance

Ministry of Labour (various issues), OEPI, Government of India: Ministry of Labour

Moser, C. (1989). ‘The impact of Recession and Adjustment Policies at the Micro-level: Low Income Women and Their Households in Guayaquil, Ecuador’, in UNICEF (1989). Poor Women and the Economic Crisis: The Invisible Adjustment, Santiago, pp. 139–166

NSSO (1996) Sarvekshana, vol XX No. 1 68th issue, July–September 1996, NSSO, pp. S-162 to S-177

Palmer, I. (1991) Gender and Population in the Adjustment of African Economies, ILO, Geneva

Parikh, K.S and R. Radhahrishna (2002). India Development Report 2002, New Delhi: Oxford University Press

Senapaty, M. (1997) ‘Gender Implications of Economic Reforms in the Education Sector in India-case of Haryana and Madhya Pradesh’, PhD thesis, Manchester: School of Economic Studies, University of Manchester

Shariff, Abusaleh and Anil Gumber (1999) ‘Employment and Wages in India: Pre and Post Reform Scenario’, Indian Journal of Labour Economics, Vol. 42, No. 2, April–June

Standing, Guy (1989) ‘Global Feminization Through Flexible Labour’ World Deveolopment, Vol. 17, No. 7

— (1999), ‘Global Feminization Through Flexible Labour: A Theme Revisited’ World Development, Vol. 27, No. 3, March

Thomas, et. al. (eds.) (1991) Restructuring Economies in Distress, Policy Reforms and the World Bank, Oxford University Press

UNCTAD (1999) Trade and Development Report, Geneva

Unni, Jeemol and Uma Rani (1999) ‘Informal Sector: Women in the Emerging Labour Market’, Indian Journal of Labour Economics, Vol. 42, No. 4, October–December

Vaidyanathan A., ‘Employment Situation in India (1994): The Emerging Perspective’ Economic and Political Weekly, Vol. 29, No. 50

Visaria Pravin (1996) ‘Levels and Pattern of Female Employment, 1911–1994’, Paper presented at the seminar on Gender and Employment in India, organized
by Indian Society of Labour Economics and the Institute of Economic Growth, Delhi

Wood, A. (2000) ‘When and Why does Globalization Increase rather than Reduce Labour Market Inequalities in Developing Countries?’ Trade and Industry Monitor, December, Vol. 16

Wood, A. and M. Calandrino (2000) When the Other Giant Awakens: Trade and Human Resources in India, Brighton: University of Sussex

World Bank (1999) Global Development Finance, Country Tables

World Trade Organization (WTO) (1997) International Trade, Geneva

World Trade Organization (WTO) (1998) International Trade, Geneva

Gender in a Macroeconomic Framework: A
CGE Model Analysis




The importance of gender aware macro-economic analysis has increased with many developing countries, including India, embarking upon Structural Adjustment Programmes (SAPs), since women and men are differentially affected by such programmes. One reason for this is the nature of participation in work. Greater competition and a need to reduce production costs generally lead to informalization (Portes et al., 1989). A number of studies have revealed that the informal sector has continued to grow at a faster rate than the formal economy in many developing countries during periods of economic reform (see Meagher and Yunusa, 1996). Further, it is observed that most of the female workforce is involved in informal activities, and there are indications of an increase in the labour force participation of women in the informal sector (Cagatay, Elson and Grown, 1995). In India around 90 per cent

1This report forms part of the ISST-NCAER project titled, “Macroeconomic Analysis of Gender: An Attempt at CGE Modelling” funded by IDRC, Canada. We thankfully acknowledge Christopher Adam for his valuable contribution in developing the CGE model. We would also like to thank K.A. Siddiqui, Diane Elson, Swapna Mukhopadhyay, Renana Jhabvala, Faye Duchin and Ratna Sudarshan for their contribution. We are grateful to Randy Spence, Luc Savard, John Cockburn, Bernard Decaluwe, Veronique Robichand, M.M.Khan and Praveen Sachdeva for their support in this work. Finally we would like to thank Rakesh Mohan for his encouragement in undertaking this research.

of the female labour force, including helpers or unpaid workers, is involved in informal activities (Sinha, Sangeeta and Siddiqui, 2000). It is important to explore the different types of work carried out by women in the informal economy and the corresponding remuneration they earn. Distinguishing the database by gender would enable an assessment of the impact of alternative development strategies on both men and women and an evaluation of policies that might improve the economic situation of women in particular (Duchin and Sinha, 1998). Major macroeconomic changes also impact upon income distribution and the social sector (Stewart, 1995). The implications of SAP for poverty are also linked to work because much of the informal sector consists of work that is too poorly paid to meet basic human needs or remains excluded from the welfare or solidarity network (Hugon, 1990).

Incorporating gender in a macro framework will allow us to understand the impact of policy changes on women. With many developing countries adopting SAPs during the 1990s, there has been greater emphasis on the integration of gender as a category of analysis in economics. Gender issues have been considered not only at a micro, intra household level, but have also been integrated into macro analysis (Beneria, 1995; Bakker, 1994). Progress in methodology has been on two fronts. The first important area is the availability of statistical data to capture women’s labour force participation with greater accuracy than in the past. Another important area is methods of imputation of the worth of household work to estimate its value in national accounts (Beneria, 1995). It has been recognized that more accurate and conceptually sophisticated gender-sensitive statistics are useful to generate a quantitative measurement of women’s work, and allow better analysis of social and economic conditions as well, more specifically in countries undergoing structural adjustment.

As in other developing countries, the Government of India also initiated a range of policy reforms designed to usher in a regime of greater competition with a more open and market-oriented economic structure in the early nineties. There have been major changes in industrial policies leading to a reduction in the scope of industrial licensing. The procedural rules have been simplified. Entry and investment has been allowed in areas that were reserved

for the public sector. The major thrust of the liberalization process has been wide-ranging trade reforms, bringing about restructuring and a reduction in custom duties together with a gradual elimination of quantitative restrictions on trade. In the pre-reform regime India nurtured import duties that were among the highest in the world. The tariff duty rates were above 200 per cent for many items in the pre-reform period. Since 1991 these rates are being reduced and the maximum tariff rate was brought down from 65 per cent in 1994–95 to 45 per cent in 1997–98, 40 per cent in 1999–2000 and to 35 per cent in 2000–2001. At present there is a total of 4 custom duty rates, i.e. 35 per cent, 25 per cent, 15 per cent and 5 per cent. The reform measures undertaken so far have created a more open economy and, alongside liberal import policies, export promotion schemes have been given a further boost through the setting up of special export zones.

To understand the impact of major policy changes on women, it is essential to have a comprehensive macro framework. In this study we build a standard computable general equilibrium (CGE) model for a preliminary analysis of the impact of macro policy changes on women. Here, gender is analysed within the formal and informal sectors of the economy because a large section of women in India are involved in informal activities, as noted earlier. There is evidence that liberalization affected growth favourably in India during the 90s (Srinivasan, 1998). However, the impact of such policies on different socio-economic groups requires an assessment. The purpose of this exercise is relatively narrow. It is an attempt to study the impact of trade reforms on income distribution distinguished by informality and gender.

This essay is organized as follows. Section II describes the Social Accounting Matrix (SAM) database, which incorporates the flow of formal and informal labour, distinguished by gender. Section III describes the concept of households distinguished into different categories and examines the structure of these households in the light of activities differentiated by gender. In Section IV we briefly describe the model and then present the preliminary results from our simulation exercise in Section V. Section VI concludes the paper with a discussion of future directions and extensions.

Building of the Base SAM

The data base for a CGE model is a Social Accounting Matrix (SAM). The SAM combines information from different sources in a consistent framework to reflect the economic and social structure of an economy at a particular point in time. The SAM constructed in this work distinguishes factors of production by formal and informal parts as well as by gender. We have mapped the flow of value added from sectors to the different types of factors of production and have also mapped the flow of factor earnings ‘to different types of factor owners within households, thus determining the factor incomes of these households. The major steps in building the base SAM for the CGE model are explained in the following sub-sections. To complete the SAM we have used information from other sources as well, such as savings rates which are from MIMAP2 data analysed at the National Council of Applied Economic Research (NCAER).

Sectoral classification

The 115 sectors as given in the input-output table of the Indian economy for the year 1989-90 published by the Central Statistical Organization (CSO) India, are aggregated into 7 sectors for this exercise (see Appendix I). The purpose of the aggregation is to keep the major broad sectors which are also important in terms of the share of the formal and informal value added disaggregated by gender. For example, agriculture is mainly an informal sector activity or the construction sector employs a large number of informal female workers. Gender wise percentage shares of workers by each of the 7 sectors are presented below (see Table II.1). The other concern in respect of these broad sectors is to help analyse policy changes, and at the same time project a more macro picture of the economy. This SAM assumes that each activity produces exactly one good, which is entirely supplied to its commodity market.

2Survey of 5000 households conducted by NCAER to collect data for the research programme on Micro Impacts on Macroeconomic Adjustment Policies (MIMAP).


Percentage Share of Women Workers by 7 Sectors


Factors of production distinguished by gender

A distinction of the present exercise is to treat factors of production as separate for women and men and consequently determine factor earnings distinguished by gender. The Annual Survey of Industries (ASI) provides information on value added (both wages and returns to capital) generated by each sector at the 3-digit level of National Industrial Code (NIC) code for registered manufacturing. The National Account Statistics (NAS) provide information on value added generated from all production sectors at the 1-digit and a few at the 2-digit NIC code for both registered and unregistered parts of these sectors. To attribute formal and informal value added distinguished by gender to formal and informal workers differentiated by gender respectively, we need to have a one to one correspondence between value added and the workers. The NAS data give us the total of workers in a sector, which needs to be broken up by different types of workers. The share of each type of labour, i.e., female and male within formal and informal types of labour, in a particular sector, is generated from the National Sample Survey Organization’s (NSSO) household survey on employment/unemployment based on certain assumptions regarding informal workers.

NIC identification of each household member is available through the NSSO survey data, and information regarding the member’s status is also recorded through this survey. Such information about the ‘working status’ of each household member above 5 years of age is taken into consideration to distinguish a

worker as formal or informal. Members who have stable employment reflected through the drawing of a regular salary/wage or those who hire labour for their own household enterprises are classified as formal workers/capital owners. Members who are casual wage labourers and own account workers are informal. We make certain objective judgements to qualify these distinctions further for rural household entrepreneurs in the agricultural sector who are identified as informal even if they hire outside labour. The rest of the working class (own account workers and casual wage labourers) in agriculture are also identified as informal workers. Thus, the factors of production distinguished by gender and informal and formal categories are broken up into the following categories for each of the seven sectors (by aggregating NIC sectors) using NSSO data:

  1. Female casual (informal) labour
  2. Male casual (informal) labour
  3. Female regular (formal) labour
  4. Male regular (formal) labour
  5. Own Account Workers (Informal Capital Owners)
  6. Employers (Formal Capital Owners)

We assume that the value added generated in a sector gets distributed to female and male workers distinguished by informality in the proportion obtained from sample survey data according to average earning rates.3 Hence, though in this exercise we do not distinguish between formal and informal production sectors, the value added is distinguished by that generated by formal and informal workers differentiated by gender. The wage rates that have been used are not at the sectoral level but at the overall level. Out of the total value added the wage income is taken out to provide the total capital income. Further, capital earnings are distinguished into formal and informal earnings on the basis of information on own-account workers and employers.

In sum, the NAS provides the total number of workers in each of the three-digit NIC sectors, without any break up between labourers, employers and own account workers. There is no further information about how many of the workers belong to unregistered

3The average earning rate is weighted by shares of different types of labour (i.e., female and male within formal and informal workers) and their wage rates so that it captures the differential earning rate of the different types of labour.

and how many to the registered parts of these sectors. We impose the formal/informal proportions of the estimated female and male workers on the actual number of workers as given by National Accounts. This methodology allows us to break up workers by gender and informality adhering to the overall national accounting as given by NAS. We thus obtain a one to one correspondence between all types of workers and value added for each sector remaining within the overall information given by the NAS on value added and workers.

Structure of labour and earnings by gender

NSSO’s Employment\Unemployment Survey is used to get the information about the male-female factor ownership proportion as noted above. Moreover, factor owners distinguished by gender are distributed across the seven industry sectors using the survey data. We see that women are mostly engaged as casual labourers rather than as regular labourers and that nearly 80 per cent of the employers are male. The difference in percentage between men and women is smaller in the case of own account workers (OAW), though women still constitute less than 40 per cent of the total OAWs. Overall, on an average 78 per cent of the worker population (which includes both labour and capitalist) is male and only 22 per cent is female.

Factor owners are related to their factor earnings and since it is difficult to distinguish capital earnings by gender from the available data collected by the statistical office, we do not use the break up earnings of OAW and employers by gender. This implies that incomes originating from such activities are neutral to male/female distinction in this study.

The percentage of different types of factor owners by sector is presented in Table 2. Table 3 depicts the sectorwise distribution of factor earnings. Sectoral value added by gender is computed by using the wage rates for different types of labourers (NSSO report) differentiated by gender as noted earlier. Information on wage rates is at the overall level and is not sector specific. Casual labourers are considered to be informal workers and so casual labour earnings are the informal labour incomes. At present capital earnings are distinguished into formal and informal and by gender on the basis of information on capital ownership because of lack of data on capital earning rates.


Percentage Distribution Factor Owners by Industry Sectors



Percentage Distribution Factor Earnings by Industry Sectors


The sectorwise distribution of the labour force shows that, as much as 12 per cent of all workers in the agricultural sector are female casual workers. However, they get only about 5 per cent (see Table II.3) of the total value added generated in the sector. Female casual labourers in agriculture contribute a substantial share in work participation, but obtain 33 per cent less earnings per unit of labour compared to their male counterparts. Of the total labour force in the construction sector, 9 per cent are female casual labourers and they retain 4.5 per cent of the value added of this sector. In manufacturing the female casual labour force constitutes 6 per cent of the total workers in the sector and earns about 3 per cent of the sectoral value added.

Activities of Households

The activity of each member within a household is obtained from the NSSO’s 50th round survey (NSSO, 1993–94). We have used only labour force activities (i.e., the different factors of production such as casual female labour, casual male labour, etc.) for building the SAM as a base data set for the CGE model. However, it is important to understand the structure of a household and also the contribution household members make towards the well being of a household through domestic and other unpaid duties. In a developing country most households operate within an informal system and their number even increases with modernization (Duchin, 1998). The hypothesis we have is that households earning income through informal activities will have a different economic behaviour compared to households earning from formal activities.

The households are first distinguished as rural or urban. Next, they are classified as formal and informal by taking the NIC, NCO and the ‘Type’ codes of households as formal-informal. The ‘Type’ code formulated on the basis of composite household information, and the household NIC and NCO codes are provided by the NSSO. Households in the urban region are classified as formal with ‘Type’ codes as employers and regular wage earners. In rural regions, exceptions are households with NIC agriculture; even with formal ‘Type’ codes, such households are classified as informal. Finally, households are classified into different income levels, i.e.,

poor, middle and rich within each broad category (Appendix II). The households are then scrutinized to identify the different types of workers and factor owners distinguished by gender that make them up. Further, the inclusion of non-labour force activities of household members is important as much of women’s time is devoted to ‘reproductive’4 rather than market activities. However, all the information on household structure is not used in the SAM building exercise since only few among the worker types are involved in labour-force activities and we have still not incorporated other types of work in the SAM.

It has been observed that informal households consume more of agricultural products than formal ones. (see Appendix III). Informal households face uncertainties of income flow and are also outside the frame of security rights. It may be noted here that apart from categorizing households as merely formal or informal, it is meaningful to have other socio-economic characteristics to distinguish them. It is possible that at a certain high level of earnings the difference in behaviour of formal and informal households may be marginal. However, most of the households involved in informal activities are poor. It has been estimated that 86 per cent of households in India can be termed as informal and 39 per cent of the population residing in such households are below the poverty line. Furthermore, a large proportion of women are involved in the informal sector and as such belong to informal households (Sinha, Sangeeta and Siddiqui, 2000).

The above tables further show that the female work force is mostly informal and even in formal rural households females are engaged more as informal workers. Moreover, in all types of households, female workers contribute to household work much more than male workers do. In addition, the share of female unpaid workers is much higher than female paid workers in all households, whereas, the share of male paid workers is higher than the share of male unpaid workers in all household categories.

4The term is now familiarized by feminist economists to describe the many unpaid services provided in households mainly by women for maintaining social well being and welfare of the household (see Elson, 1995 and Fontana and Wood, 2000).


Distribution of Female Member Activity within Formal/Informal Households (per cent)



Distribution of Male Member Activity within Formal/Informal Households (per cent)


The CGE Model

We have discussed the data as a base for the modeling structure. We now characterize the economic system generating the data. One way of viewing a SAM is to consider each recorded transaction as the outcome of a (constrained) optimization problem faced by one or more agents of the economy. Thus, for example, factor income flow from firm to household reflects the firm’s labour demand given the demand for their output, the nature of technology, their capital stock and the price of labour. Similarly, a household’s consumption of items such as manufactured goods reflects the utility maximization given their income and prevailing prices.

We can think of a (general equilibrium) model as a systematic and internally consistent description of the behavioural relations, constants and market clearing conditions, which could have generated the SAM. Moreover, the specific form of the model will determine how the system reacts when perturbed. It is important to stress that any SAM will be consistent with an infinite set of rival models. Unlike econometric analysis, the process of calibration allows all models to fit the baseline data equally well.

In this study we have developed a preliminary standard CGE model of the type discussed in Devarajan, et al. (1996), which is a widely used trade focussed models for developing countries. In this model the major contribution is to characterize the consequences of changes in trade policy for the distribution of income between the formal and informal factors (Sinha and Adam, 2000) distinguished by gender and across a variety of household types. The distribution of factor income flows, both formal and informal by gender, between household types is treated as parametric, obtained from the baseline data. In future versions of the model, we would attempt to endogenize this distribution mechanism.

This is a real economy model. Therefore, real consumer demand functions and producer supply functions are homogeneous of degree zero in the aggregate price level. Here at present we abstract from macroeconomic considerations, so that the government is not restricted to a balanced budget. Thus public sector deficits or surpluses can take place. The financing of these can be carried out through a tax or rebate on private sector saving leading to crowding-out or crowding-in of private investment through a neoclassical

closure system. We have experimented with one version of the model which, in a limited sense, is recursively dynamic. The first period solution of the model defines a vector of public and private sector gross capital formation. Net capital formation is combined with exogenous depreciation evolving capital stock. This leads to capital stock build up ‘between periods’ such that the new capital stock investment is available at the beginning of the next period. We calibrate the model to a steady-state equilibrium in which net public and private investment in the baseline is zero (gross investment exactly matches depreciation) so that the capital stock is constant. This allows us to analyse the impact of trade reforms in a focused manner. Thus with no exogenous shock the model exactly replicates the baseline indefinitely.5 In this study we also use a static version of this preliminary CGE model, where capital stock is sector specific and is obtained from the base level data. The model is being written using the GAMS programming language.

The goods market

We adopt standard goods market specifications. Firms are assumed to be perfectly competitive and produce a homogeneous output which can either be sold in the domestic market or exported. In the current version of the model all firms are price takers for all imports. The model consists of seven productive sectors, each producing a single representative good or service as noted above.

Gross output is determined by fixed-coefficients Leontief production structure where intermediate inputs are combined with value added. Value added is determined according to Cobb-Douglas production functions for all sectors of the economy. In the present dynamic version of the model the capital stock evolves over time in order to equalize real sectoral rates of return. We assume ownership of a homogenous sector-specific capital. The distinction of the capital as formal and informal provides a link of capital income to household income. In the model capital prices are equalized but the sectoral decomposition of capital between formal and informal capital is defined exogenously from the calibration data. The purpose of this specification is to define the flow of

5There is no labour supply growth assumed in the model at present.

capital income to households. In future versions of the model we aim to develop a structure in which there are two capital stocks, which are no longer perfect substitutes, and there exists an explicit market for the two types of capital.

The level of government capital formation also determines the level of private sector output in this model (key equations are presented in Appendix IV). We assume that the government produces two forms of capital: a ‘sector-specific’ capital good which is required only for the production of government services (such as government offices etc.) and a ‘public good’ capital (for example in infrastructure). The latter enters the production function of all private sector firms without competition, so that higher public spending raises private output in all sectors.

Labour markets

We assume that the supply of the four labour types. i.e., formal and informal labour distinguished by gender is fixed within the period. Since firms are profit maximizers and all the four types of labour, consisting of both female and male labour, are mobile, the average wages for each broad labour type are driven towards the value of their marginal product in each sector. However, the observed sectoral wage distribution is maintained and sector-specific wage rates are not equalized across sectors (for a given labour type). The sectoral wages are distributed around the mean wage for each skill type according to a fixed wage distribution matrix. The distribution matrix may be considered as wage rigidities, which might arise from variation in skill, or union power in certain sectors.

Allocation mechanism in goods market

In this system each productive sector’s goods are distinguished between tradable and a domestic variant. The economy is assumed to be a price taker for all tradable goods, under small country assumption. However, domestic goods and factor prices are fully flexible. Output can be consumed or applied to the formation of the capital stock, and is sold either to the domestic market or exported. On the production side, the model assumes that the domestic and export variant of the goods are imperfect substitutes so that the firm cannot switch their output costlessly between the

domestic and foreign markets.6 Following the ‘1-2-3’ convention, imperfect substitutability is reflected in the following manner. Firms produce a total output which is allocated between the export and domestic markets according to a sector specific constant elasticity of transformation (CET) functions.

Consumption for each sectoral output is assumed to be regulated by the Armington assumption of imperfect substitution between domestic product and imports. The demanded composite consumption good is a constant elasticity of substitution (CES) aggregation of imports and domestically produced goods. Here consumers are assumed to have a choice for quality so that they distinguish between domestic and imported variants of good. The composition of consumers’ demand between domestic and imported goods by household as final consumption and firms (for intermediate purchases) is therefore defined analogously to the firms’ production functions.


The CGE model has a number of prices that clear the different markets defined in the model. As the model is described in entirely real terms, it provides a solution for relative prices only. As it is a real model, we need to define a numeraire, which can be any of the prices. The numeraire should be chosen depending on the type of questions the model is designed to address. As we adopt small country assumption, firms and consumers are price takers in the world market. Domestic prices for tradables are linked to world prices wedged by the domestic tax system. The model is within the neoclassical framework and all endogenously determined prices clear their relevant markets. Domestic prices for imports and exports are defined by the price-taking assumptions. As stated earlier, domestic and traded goods are considered as imperfect substitutes so that consumers and producers make decisions over composite consumption and output. On the basis of the CES/CET aggregation functions the price aggregates i.e. consumption prices and the aggregate output prices are obtained.

In this system the crucial domestic price is endogenously

6This reflects, amongst other things, differences in quality of the good, packaging or other specification issues.

determined which is the (implicit) price of domestic output. Composite demand by households and aggregate firm-level production determine the total composite supply and domestic output. The CES/CET functions define the optimal combination of tradable and domestic goods based on embodied preferences in these functions. Under small country assumption, world prices are exogenous. Therefore, relative price changes required to clear the market for domestic goods will define the equilibrium value of the domestic price. All other prices in the model are for accounting conventions (see Appendix IV). As production involves both intermediate goods and value added it is necessary to partition the total output price into the relevant prices for the two components. Intermediate goods consist of the total composite good aggregated according to input requirement which are priced at the aggregate consumption prices. The implicit price of value added is then the difference between the net price of aggregate output and the price of intermediate inputs

Composite commodity can be used for final or intermediate consumption as mentioned above or for capital formation. Capital formation in this system has two dimensions. The decision to invest is made on a destination basis: a firm in sector decides to increase its capital stock. The composition of capital goods in the sector will determine the price of one unit of this capital. One unit of capital stock in, say, manufacturing may require a certain amount of capital goods, a certain amount of services (for example building services) and a certain amount of consumer goods. All of these sectoral inputs at the consumption price are combined according to a capital composition matrix to determine the price of capital. Finally, we define two price indices for convenience, which are useful. These are the consumer price index, and the GDP deflator.


In the model the households form a distinctive feature. In this exercise thirteen different household types have been identified. They are first distinguished by the regions to which they belong (rural and urban). Further, the households are identified as formal and informal. Finally, they are classified into three or four (in the case of rural informal) per capita consumption levels denoting the highest to lowest income/wealth household categories. The baseline

factor flow relationship, which provides a mapping from factor demands by firms to households (see Table 6) are obtained from the base Social Accounting Matrix.


Household Classification and Factor Flows Distribution of Factor Incomes (per cent)


Notes: [1] See Appendix II.

[2] Capital income share equal for net profit and consumption of capital.

The exogenous allocation coefficients are defined from the baseline data, which are then used to map the factor income generated into gross household income. Gross household income is obtained by augmenting factor income by government income transfers and private remittances from abroad. Net of direct income taxes, as paid here only by formal sector households and savings, this income is allocated to consumption across the composite goods priced at consumption price.

The consumption function is a Cobb-Douglas for each household type, where a matrix describes household consumption shares across the different goods by household type (see Appendix III). At

present we have this as a first approximation: it is possible to introduce an alternative consumption function specification such as a variant on the linear expenditure system.

Savings and investment

The model has a simple neo-classical savings driven investment closure. Foreign savings are exogenously given and the level of domestic savings determines total investment. Households are assumed to have a constant propensity to save out of their net of tax gross income, however, the propensities are different across household types. Government savings are exogenous. After determining the exogenous public sector investment requirements, the sectoral allocation of the residual investment is defined by a return sensitive function where firms’ demand for (their own sector-specific) capital is a function of the differential between the sectoral real rate of return and the economy-wide average. Investment by sector of demand is translated into a demand for investment goods, which are mainly machinery and construction services, determined by the capital composition matrix.

Macroeconomic balance and dynamic specification

Three conditions determine macroeconomic balance. The first is that the goods market clears. The second is that the external balance constraint is satisfied and the third is that the labour market clears. The above constraints taken together imply that by Walras Law the savings equals investment constraint is satisfied ex post. For each time this defines a solution in terms of market-clearing prices and quantities for goods and factors and a vector of savings and investment demand. These latter vectors determine the dynamics of the model.

The dynamic version of the preliminary model is strictly recursive as noted earlier. Therefore, at the beginning of each period, agents inherit real stocks of physical capital. Firms and households update these stocks so that the new capital stocks enter the firms’ production functions the following period. The other version is static where capital is fixed across sectors, and is used to study the impact of any exogenous change in a comparative static framework (see Appendix IV for key model equations).

In this study it is meaningful to define a specific welfare-based

objective function as we attempt to study the impact of trade policy changes on the welfare of informal households. Further, we can infer welfare implications on women workers to a certain extent since the majority of them belong to informal households Therefore, we have a welfare-based objective function in the model, which provides some indication about household welfare measures (see Appendix IV). The utility function of households (see Adam and Bevan, 1998) is defined in terms of their level of consumption (of private and public goods) where private welfare is defined by real discounted value of private and public consumption.

Simulation Experiments

During the nineties there have been major changes in tariff rates in the process of economic reforms undertaken by the government. In Table V.1 we present the percentage change in custom rates for the manufacturing and the capital goods sector for the period 1991-92 through 1998-99. In this section we present the preliminary findings by examining the distributional consequences of trade reforms using the two versions of the model. We have designed three simulations taking into consideration the cumulative changes in tariff rates as shown in Table 7. We have used the static version of the model for a comparative static analysis in simulation 1 and

2. In simulation 3 we use the dynamic version so as to capture the longer run fiscal consequences of trade reforms. In Simulation 1, import tariff of the manufacturing sector is reduced by 50 per cent; in Simulation 2, import tariff of the capital goods sector reduced by 17 per cent; and in Simulation 3 we hypothesize a 50 per cent reduction in tariff rates of all tradable sectors.

We present below in Table 8, the average tariff rates by sector in the base case and for the simulation scenarios. As noted the simulations have been designed by taking into consideration the actual changes in tariff rates that have taken place in the nineties. The lower tariff and the resultant lower import prices would change the relative demand for domestic goods to imports in each sector. These changes depend on the reduction of tariff rates and the elasticities of substitution. The values chosen for behavioural parameters follows common practice in similar CGE models applied to low income developing countries. Here we have assumed

that price elasticity of substitution in consumption is less than unity. Given common Armington elasticities for all sectors, the import share and tariff rates will play the main role in variation in sectoral production levels and sectoral prices.


Percentage Change in Sectoral Custom Rates for the Period 1991–1992 to 1998–1999


Source: Economic Survey, various issues.


Application of Post Liberalization Cumulative Change in Tax Rates Custom Duties [as per cent of the world price]


Reduction in tariff reduces the distortion between domestic and world prices of tradables but this is accompanied by a loss in revenue, at least in a static sense. In this model the level of public investment is a determinant of private sector output. Therefore, in


Input-Output Coefficient Matrix and Changes in Domestic Consumption Prices



Impact of Trade Reforms on Relative Prices


Note: P1 = First simulation period; P6 = Last simulation period.

the dynamic version of the model changes in fiscal situation influences private sector behaviour. The fiscal response as assumed in the present model leads to a decline in revenue. This is translated into a decline in government savings and hence government capital formation. Since government capital formation has a positive externality for private profitability this fiscal contraction leads to a general reduction in private sector profits which squeeze domestic profitability arising from the higher level of import penetration. In the dynamic version double pressure on private profitability leads to decline in total capital stock in the economy substantially in the long run.

As consumption of domestic manufactured goods declines due to higher imports, there is a corresponding decline in intermediate capital goods. The price of composite consumption for capital goods falls more sharply in simulation 1 compared to the composite

price of manufactured goods even though the import price of manufacturing falls more sharply. The capital goods sector uses a very high share of manufacturing intermediate goods. Therefore a fall in the price of manufacturing goods would reduce the price of capital goods because of the input-output structure. Moreover, exports rise very sharply in capital goods given the elasticities and export structure. As a result domestic import prices of capital goods rise very sharply as seen from Table 9. When the import tariff rate of capital goods is reduced by 17 per cent, the consumption price of capital goods declines most sharply. The input-output structure shows that the intermediate demand for capital goods is high for its own production. The intermediate demands from other sectors are not very high for the capital goods sector. In case of simulation 3, import prices of all tradable sectors (see Table 10) fall due to an across the board tariff reduction. As a result again the consumption price of capital goods falls very sharply.

Impact of trade liberalization on production and trade

The impact of the tariff reduction on sales, production and trade are presented in Table 11 below. We see that in case of simulation 1 tariff on manufacturing is reduced by 50 per cent implying a very large absolute reduction in tariff rate of this sector (see Table 8). As a result, imports of this sector increase and exports also experience a marginal rise. So even though there is decline in domestic sales, domestic production improves very slightly. In case of simulation 2, tariffs on capital goods are reduced by 17 per cent. There is a reduction in the import price of this sector and imports rise for capital goods. The fall in domestic prices leads to an increase in exports and decline in domestic sales. As a result the domestic output of capital goods experiences a contraction. In case of simulation 3 we assume that tariff rates are reduced for all sectors uniformly. In case of the manufacturing sector imports fall, and domestic production is diverted towards the export market with a fall in domestic sales. This is a ‘classical ’ tariff reduction situation. In case of capital goods, we see (Table 11) imports of this sector decline and there is a substantial increase in exports. Tariff reduction occurs contiguously with the liberalization in other sectors, so there is a decline in input cost for the capital goods sector as prices of manufactured goods fall. At the same time exports of


Impact of Tariff Reduction on Output and Trade


capital goods sector rise drastically. The domestic price of the capital goods sector falls more sharply than the import price of capital goods. Therefore, the relative import price rises (Table 11) and this leads to the fall in imports in this sector. Viewed in the aggregate, however, the economy’s demand for imports rises and this is matched by an increase in export production. The substitution effect is sufficiently strong to offset the small aggregate income effect from the removal of the trade distortion. Hence aggregate domestic output shrinks slightly in the short run. As noted above, the level of investment declines in the long and so overall output declines more sharply in the long run.

Impact on factor remuneration

The factor intensity in the base level is presented in Table 12. We see that the manufacturing sector is less intensive in both male and female regular labour. In case of simulation 1, there is contraction in sectors like manufacturing as well as in construction which are less intensive in regular male labour and expansion in sectors that are more intensive in regular male labour. So the relative remuneration of regular male labour rise the highest in this simulation (see Table 13).


Worker Composition by Sectors


The opposite is true for casual male labour. In case of simulation 2, there is relative contraction in capital goods and construction

sector, but there is not much expansion in other sectors. As a result the average wage rate of male regular labour rises less sharply in this simulation (see Table 13). In case of simulation 3 the level of investment declines with lower government savings leading to lower private investment. Now each worker has less capital and is less productive leading to a decline in the marginal product of labour. With a built-in assumption of full-employment and competitive market, the result of the tariff change is a decline of real wage over time. This results in absolute decline in casual real wages (more so for male casual wage earners) and a moderate growth in wages for the regular workers in the long-run.


Impact of Trade Reforms on Wages and Consumption



Structure of Factor Ownership by Households


Note: Formal and Informal capitalist are not differentiated by gender in the base SAM as yet.

Household income structure and welfare

The table on income sources of each type of household is presented in Table 14. We see that the major share of income of formal households is from regular male labour. In case of informal households, the higher earning shares are from casual workers and own account workers. So as regular male wages improve in all the simulations, the formal households benefit. Together with improvement in regular wages, the prices of manufactured items decline due to tariff reduction. Hence real consumption by formal households increases reflecting improvement in their welfare. This reflects the fact that with tariff reforms in non-agricultural sectors male workers benefit more than women workers who are less in formal households. We also present an overall welfare measure of all types of households. As public consumption is not paid for at the point of consumption this does not enter household consumption in the usual consumption based welfare measures. However, we need to add back the value of public goods into the welfare function. Otherwise we would find that reduction in public expenditure leads to an increase in welfare. So, we also assume that households benefit from the level of government production. Households’ utility is defined in terms of their level of consumption of both private and public goods. (See Section IV). We find that welfare gain (see Table 13) is higher in simulation 1 as compared to simulation 2. In case of simulation 3, trade reform has an initially positive effect raising social welfare marginally by 0.13 per cent. As the fiscal distortion begins to materialize, however, this welfare effect diminishes so than by the end of the 5th year, welfare declines marginally from the baseline.

The impacts of the simulations on the female and male earnings within different types of households are reported in Table 15. In these simulations formal households gain more than informal households, because males form a higher share of such households and their earnings rise more sharply (depending on the value added distribution of the affected sectors) compared to female earnings. In simulation 1, regular wage rates for male workers rise the highest. In simulation 2 average wage rate of male labour rises less sharply and again in simulation 3 both male and female regular wage rates increase with male rates more sharply than female rates (see Table 13). We see that in all cases regular wage rates of male


Percentage Change in Real Earnings of Female and Male Workers


Note: The earnings refer to wage earnings as capital earnings have no male–female variation.

workers rise more sharply for reasons noted above. In case of casual wage rates, female casual wage rates fare better than male casual wage rates. The earnings of households change as a result of the wage rate changes. The above table reveals that the male earnings increase more than female earnings for nearly all types of households except for rural poor and non-agriculture informal households. We can explain this by studying the worker composition of these households and the change in wage rates due to these simulations. The percentage of casual male workers in rural informal poor and non-agriculture households is very high as compared to other informal households and the share of regular male labour is very small. The lower rise in the casual male wage rate than the female casual wage rate results in lesser benefit for males in rural poor and non-agriculture households. However, in all other types of households the earnings of male workers rise more in comparison to earnings of female workers as a result of the 3 simulations.

Conclusions and Future Directions

The results presented above help us to draw certain conclusions. It should, however, be recognized that the model is a preliminary model taking a first step towards a gender oriented CGE model. The preliminary findings show that tariff reduction leads to welfare gains (measured as rise in real consumption expenditure of households) in all households as prices fall. However the formal households gain relatively more, as the regular wage rates increase marginally, whereas in comparison casual wage rates suffer. This implies that men wage earners benefit more than women wage earners as they form a larger share of workers in formal households.

It is realized that in a gender-oriented CGE model, the proportion of female workers should motivate the choice of production sectors. We have scrutinized a more disaggregated sector classification so that the sectors with higher female labour intensity can be identified and used in future work. In Appendix V we present a further break up of the seven sectors used in the present model to a reclassification of ten sectors. These ten sectors are obtained by aggregating the production sectors as given in the 1989–90 input-output table. Further, these 10 sectors are disaggregated from the current seven sectors used in this study, so that certain sectors like manufacturing are broken up further to highlight the share of women workers in these sub-sectors of manufacturing.

So far our simulations are local approximations based on the assumption that factor/gender proportions do not alter (although there may be top level substitution of capital for labour). To actually take it further requires more research on economic structure. For example, we would need a theory (and hence a structural model) about the way in which formal and informal factor markets function and how gender discrimination works. We might want to model two separate labour markets for men and women that would require investigation about how these two markets work. The Indian Statistical Office has completed a pilot study on the Time Use Survey and it will be possible to use time use data to some extent from this source in the near future. In light of the impact of globalization on informalization with likely rise in sub-contracting, it has also become important to have information on home based work, such as garment manufacturing and food processing. The CGE model could be modified to incorporate supply response to the informalization of labour force. As demand

for informal labour rises there could be an increase in wage rates. The supply of labour would respond to the wage rates, however there will be supply side constraints, which will be different for females and males. Further, incorporating ‘reproduction services’ as a production sector would allow determination of demand for this service as any other market good. However, it is important to be able to determine the opportunity cost of such services. Also it will be useful to have different price elasticity of demand in the reproduction sector (see Fontana and Wood, 2000) .

More information on female and male members of a household will help in obtaining the differential earning and expenditure preferences by gender. The information on activities will determine the resources that women and men can generate. How household income is allocated by gender is a matter of research. It is important to establish any difference in household behaviour that originates due to the female/male ratio in a household. The question is how can one theoretically measure any such pattern of household behaviour. At one extreme is the ‘unitary’ household where all household income (from whichever source) is pooled. In this case, household consumption patterns are independent of the gender composition of household labour supply (and hence income). At the other extreme is what could be called the ‘partitioned’ household where there is no pooling of income. As a consequence aggregate household consumption patterns would reflect the gender composition of income and labour supply. The most probable situation is that there is some bargaining that takes place in a household that determine the household’s pattern of consumption. The bargaining power will depend, in part, on the gender composition of income and labour supply, but will also reflect other determinants of household bargaining (such as male and female reservation wages, assets brought by different members into the household, and other, possibly unmeasurable, factors such as cultural norms). Endogenizing the bargaining power of women in the model will enrich the understanding of female response to external forces during the process of globalization.


Adam, Christopher S. and David L. Bevan (1998) ‘Cost and Benefits of Incorporating Asset Markets into CGE Models: Evidence and Design. Issues,’ Applied Economics Discussion Paper Series No. 202, University of Oxford

Bakker, Isabella (1994) The Strategic Silence: Gender and Economic Policy, London: Zed Books in association with North–South Institute

Beneria, L. (1995) ‘Toward a Greater Integration of Gender in Economics,’ World Development, Vol. 23, No. 11

Cagatay, N.D. Elson and C. Grown (1995) ‘Introduction’ in Gender, Adjustment and Macroeconomics, World Development, Vol. 23, No. 11

Central Statistical Organization (1997) New Delhi: National Accounts Statistics, Government of India

Duchin, Faye (1998) ‘Structural Economics: Measuring Change in Technology, Lifestyles, and the Environment,’ published in co-operation with the United Nations University/Institute of Advanced Studies, Washington: Island Press

Duchin, Faye and Anushree Sinha (1999): ‘Structural Economics and the Quality of Life,’ Feminist Economics, Vol. 5, No. 2, July

Devarajan S., S. Robinson and J. Lewis (1996) ‘Getting the Model Right: Applied General Equilibrium Modelling,’ Cambridge: Cambridge University Press (forthcoming)

Elson, D. (1995) ‘Gender Awareness in Modeling Structural Adjustment,’ World Development, Vol. 23, No. 11

Fontana M. and A. Wood (2000) ‘Modeling the Effects of Trade on Women at Work and at Home,’ World Development, Vol. 28. No. 7

Hugon, P. (1990) ‘The Informal Sector Revisited (in Africa)’, in The Informal Sector Revisited, Davis Turnham, Bernard Salome, and Antoine Schwartz (eds), Paris: OECD

Meagher, K. and M.B. Yunusa (1996) ‘Passing the Buck: Structural Adjustment and the Nigerian Urban Informal Sector,’ Discussion Paper 75, UNRISD, May

Portes, A., M. Castells, and L. Benton (eds) (1989) The Informal Economy, Baltimore: Johns Hopkins University Press

Sinha, Anushree, Sangeeta, N. and K.A. Siddiqui (2000) ‘Study of the Impact of Alternate Policies on the Economy with Special Reference to the Informal Sectors: A Multisectoral Study,’ (Mimeo) Delhi: NCAER

Sinha Anushree and Christopher S. Adam (2000) ‘Trade Policy Reform and the Informal Sector in India,’ paper presented at the Thirteenth International Conference on Input-Output Techniques at Macerata, Italy, August

Srinivasan T.N. (1998) ‘India’s Economic Performance: A Comparative Analysis,’ paper in Isher J. Ahuluawlia and I.M.D. Little (eds) India’s Economic Reforms and Development: Essays for Manmohan Singh, Delhi: Oxford University Press

Stewart, Frances (1995) ‘Adjustment and Poverty Options and Choices,’ London: Routledge

The National Sample Survey 50th Round (1993–94): Schedule 1.0 on Consumer Expenditure and Schedule 10.0 on Employment and Unemployment .

The National Sample Survey 50th Round (1997) Employment and Unemployment in India, 1993–94—Fifth-Quinquennial Survey NSS Fiftieth Round. Report No. 409, New Delhi.


Appendix I: Aggregation Scheme for Seven Industry Sectors


Source: Input-Output Transactions Table 1989–90, Central Statistical Organization, (1997), Government of India, New Delhi.

Appendix II: List of Household Categories Identified


Appendix III: Proportion of Consumption Expenditure by Household Types for 7 Sectors


Appendix IV : Key Model Equations


Import Prices:./img/trackinggen_364_la_0.jpg
Export Prices:./img/trackinggen_364_la_1.jpg
Consumer Prices:./img/trackinggen_364_la_2.jpg
Output Prices:./img/trackinggen_364_la_3.jpg
Value added Prices:./img/trackinggen_364_la_4.jpg
Capital Prices:./img/trackinggen_364_la_5.jpg
Consumer Price Index:./img/trackinggen_364_la_6.jpg
GDP deflator:./img/trackinggen_364_la_7.jpg

Output and Factors

Aggregate production function:./img/trackinggen_364_la_8.jpg
Average wage determination:./img/trackinggen_364_la_9.jpg
Labour market equilibrium:./img/trackinggen_364_la_10.jpg
CET output aggregation:./img/trackinggen_364_la_11.jpg
Output share equation:./img/trackinggen_364_la_12.jpg
Consumer good aggregation:./img/trackinggen_365_la_0.jpg
Consumption share equation:./img/trackinggen_365_la_1.jpg

Current Demand

Intermediate goods demand:./img/trackinggen_365_la_2.jpg
Consumption Demand:./img/trackinggen_365_la_3.jpg
GDP (value added):./img/trackinggen_365_la_4.jpg
Real GDP:./img/trackinggen_365_la_5.jpg


Government value added price:./img/trackinggen_365_la_6.jpg
Government current expenditure:./img/trackinggen_365_la_7.jpg
Government revenue:./img/trackinggen_365_la_8.jpg
Import tariff revenue:./img/trackinggen_365_la_9.jpg
Export duty revenue:./img/trackinggen_366_la_0.jpg
Domestic indirect tax revenue:./img/trackinggen_366_la_1.jpg
Direct tax revenue:./img/trackinggen_366_la_2.jpg
Production tax:./img/trackinggen_366_la_3.jpg
Government current savings:./img/trackinggen_366_la_4.jpg

Income, Savings and Investment

Gross profit, net of interest costs:./img/trackinggen_366_la_5.jpg
Sectoral profit rate:./img/trackinggen_366_la_6.jpg
Average profit rate:./img/trackinggen_366_la_7.jpg
Gross household income generation:./img/trackinggen_366_la_8.jpg
Household saving:./img/trackinggen_366_la_9.jpg
Firms investment function:./img/trackinggen_367_la_0.jpg
Demand for investment goods:./img/trackinggen_367_la_1.jpg
Total savings:./img/trackinggen_367_la_3.jpg
Current Account Balance:./img/trackinggen_367_la_4.jpg
Commodity balance:./img/trackinggen_367_la_5.jpg

Objective Function


Variable List

Endogenous Variables
pmDomestic price of imports
pe Domestic price of exports
pd Price of domestic good
pc Price of composite good
pq Price of composite output
pvaValue added price
pk Destination price of capital
CPIConsumer price index
PGDPGDP deflator
XDomestic output
XDSales of domestic output to domestic market
QTotal composite supply
MTotal imports
EiTotal exports.
GDPVATotal value added
LLabour demand
WaAverage wage by skill-type
NDIntermediate demand
CDConsumption by household
YHousehold income
GXGovernment recurrent expenditure
GRGovernment recurrent revenue
GOVSAVGovernment recurrent surplus
TARIFFTariff revenue
DUTYExport duties
INDTAXIndirect taxes
DURTAXDirect taxes
PTAXProduction taxes
rReal sectoral profit rate
rReal average sectoral profit rate
RKGross sectoral profit (net of interest costs)
HHSAVHousehold saving
IDiInvestment demand by sector of origin
DEPRCDepreciation by sector
SAV INGTotal savings
FSForeign Savings
INVESTTotal investment
Exogenous Variables
erNominal exchange rate
pwmWorld price of imports
pweWorld price of export
tm Import tariff rate
tx Export duty rate
it Indirect tax rate
pt Production tax rate
t Direct tax rate
s Savings propensity
KP Sectoral capital stock
KG Public capital stock
DKG Government investment in public goods
DKpub Government investment in government capital
Trns Government transfers
debtsrv External debt service (US$)
rmit Remittances from overseas (US$)
L Labour supply
NETP Net factor payment to factor service (US$)
Technical, Behavioural and Policy Parameters
Calibration Parameters
aijinput-output coefficients
bijcapital-composition coefficients
wciweights for consumer price index
ADiProduction function shift parameter
Ψi,lcWage distribution matrix
ATiOutput CET aggregation function shift parameter
γiOutput CET share parameter
ACiConsumer good CES aggregation shift parameter
δiConsumer good CET share parameter
clesi,hhConsumption shares (by sector and household)
glesGovernment share in GDP
φiDepreciation rate
hwHousehold welfare weights
‘Exogenous’ Parameters
αliProduction function labour shares
αki Production function private capital shares
αgiProduction function public capital shares
ρtiCES transformation parameter
ρciCES substitution parameter
ΠiInvestment share parameter
θInvestment responsiveness parameter.
λiFormal profit share
ΓLhhLabour income allocation parameter
ΓfhhFormal Capital income allocation parameter
ΓinfhhInformal Capital income allocation parameter

Appendix V: Percentage Share of Women Workers by 10 Sectors


Note: Aggregation from 115 sectors from Input-Output Transactions Table 1989–90.

Towards Integration? Gender and
Economic Policy



Gender and Economic Policy

This final chapter tries to pull together the main themes that have emerged from the various contributions to this book. Overall, the volume can be seen as an exercise in integrating gender analysis and economic policy, focusing on the specific aspects of gender disparities and trade liberalization in South Asia. The formulation of economic policy is usually done in abstraction from the concerns of social policy, and without explicit reference to the socio-cultural characteristics of society. In support of this viewpoint, it is argued that macroeconomic policy has a shorter time horizon than social policy, so that the social context can be assumed to be constant over that period. For example, bringing about macro-economic stabilization in an economy needs quick measures to be implemented over a short period of time. The agenda of those concerned with social policy is necessarily a longer term one, since its implementation usually requires bringing about structural changes in gender and social relations. Consequently, it is argued that while the importance of each should be recognized as also their interdependence, the two sets of policies are best developed independently.

‘Gender’ is a category that finds a place in the economic analysis of feminist economists, but its influence on economic policy-making is weak. The result is very limited empirical information on the interplay between economics and gender, or social concerns more generally. That there is an interdependence between the two is evident both from the observable social impact of economic

change, or alternatively the social underpinnings of economic activity. The social impact is manifested in consumption patterns, lifestyles, modes of dress, behavioural norms etc. The influence of social networks on industrial organization and entrepreneurial behaviour is equally well documented. (See for example Grief 1994; Palaskas and Harris-White 1993). Gender roles and relations, like other social and cultural norms, can be expected to respond to economic changes. As Swapna Mukhopadhyay puts it in this volume, there is a ‘static dimension’ and a ‘dynamic dimension’ to the context in which gender bias is to be examined. The static context — class, caste, ethnicity, religion — is generally the domain of sociological enquiry; the dynamic context — changes set in motion by new economic policy regimes — the domain of economists. It is difficult to find a direct correlation between economic reforms and gender relations. It is even more difficult to assess whether new gender roles that are necessitated when women start working outside the home are permanent, in the sense that if the work-imposed demands were to disappear this would not mean a falling back to the original position.

Recent analysts of development have found in the concept of ‘social capital’ a useful way of emphasizing the fact that the pattern of economic growth in any country needs to be internally generated and socially embedded to be sustainable. It has been observed that economic activity tends to build on existing social norms and networks. Those concerned with gender approach the concept of social capital more warily, because clearly patriarchy and the traditional sources of gender bias form part of these norms and networks. But one can still argue that the process of changing these norms requires that we understand them. Traditional structures imposed constraints on women’s behaviour that are completely unacceptable today; in return they guaranteed a measure of security and protection. Modernization, loss of kinship and loss of traditional social supports need to be compensated by new systems of security along with the new dimensions of freedom and autonomy that they bring in their wake. In that sense, ideally we need to pick the best from the old and the new.

Policy that attempts an ‘integration’ of economic and social concerns faces a stumbling block when it comes to methodology. This book has tried to be innovative in this respect. A ‘gender sensitive economic policy’ is at the least, one that is formulated with

awareness of differences in gender roles and relations, so that existing inequalities and vulnerabilities are not intensified, and if possible are reduced. The first step is thus to understand the framework of transmission between economic policy and gender roles and relations. From a policy perspective, we need answers to several questions: what are the key indicators of gender inequalities in any given society? Can we track the pattern of change in these different indicators in response to faster economic growth, or to changes in the framework of economic policy? Can these indicators provide some kind of feedback to policy makers? How best can qualitative understanding of gender inequalities inform the formulation of policy?

The opening chapters in this book distinguish between ‘gender’ and its indicators, and present a framework for analysis of gender bias in a changing economic environment.

Gender Bias and its Measurement

Swapna Mukhopadhyay presents the specific objective behind the research presented here in the opening chapter. This is to develop a methodology for linking gender and economic policy, by tracking the behaviour of gender differentials in selected indicators over time and linking this behaviour to economic policy change. The first concern that guides the research method used in this book is the importance of interpretation with reference to context. The measurement of gender differentials is now a well-established research area, but such measurements are not always adequately sensitive to the context. Gender inequality is an ambiguous and multi-faceted concept, and its manifestations include cultural, social, legal, political and economic dimensions. The search for a small set of indicators that could chart the movement of gender differences has been pioneered at an international level by such constructions as the Gender-related Development Index (GDI) and the Gender Empowerment Measure (GEM) (UNDP 1995). However, inadequate attention is often given to the context in which gender differentials are being measured. The interpretation of any indicator calls for good understanding of the cultural context, and for nuanced interpretation. This is particularly so in regard to such concerns as mental stress and violence, but is true

also of more conventionally used measures such as work participation. (see for example EPW 1996; Saith and Harris-White 1998; Mukhopadhyay and Sudarshan 1997). Although the focus of the empirical studies presented here has been on quantification and measurement, the analysis of these emphasizes judgement and knowledge of context. A second primary concern of this study has been to expand beyond the set of indicators generally used. These are described here as ‘conventional’ indicators (health, education, employment, and income). Although it is widely accepted that violence against women is a universal phenomenon and an important indicator of unequal gender relations, it has usually been studied in a qualitative manner. It is difficult to open up debate (between qualitative and quantitative gender researchers, or between mainstream economists and gender analysts) unless there is some form of quantification. To this end, the programme of research on which this book is based has attempted the quantification of people’s levels of well being and mental stress. Bringing in these quantitative measures of well being/stress is seen as a way of capturing some dimensions of the intra-household changes that may result from changes in the external economic environment. Finally, the design of the research programme has had to deal with the fact that there is no available model structure that can adequately link gender differentials to macro policy change, even apart from the absence of adequate gender disaggregated data. The approach taken is as follows: Micro studies based on selected samples in the different countries are used to generate a picture of gender roles and relations using an expanded data set, and two macro studies present an analysis using available data.

The paper by Shobhna Sonpar and Ravi Kapoor provides ample grounds for caution in trying to link changes in ‘non-conventional’ indicators to economic reform. At the same time, there is an underlying agreement in their paper with the proposal made in the first chapter, for example the possibility of a link between economic reforms and mental health. Economic reforms that are accompanied by an increasing incidence of poverty among some groups, as well as by increasing inequality can have important consequences for mental health by generating stresses on the one hand and undermining sources of support on the other. Sonpar and Kapoor recommend the use of two measures that have been widely used in survey research. These are the GHQ (General Health Questionnaire)

and SUBI (Subjective Well-Being Inventory). These can be used to expand the data set beyond the strictly conventional measures. At the same time, the authors point to the complexity of changing gender norms and the inability of any quantitative measure to be able to capture or understand adequately these dimensions.

These two introductory papers set out the basic framework for the empirical country studies. This framework is an innovative attempt to define measurable indicators and to attempt an integration of gender and economic analysis. It also allows one to see if conventional economic analysis can throw greater light on gender issues by the use of a broader data set.

South Asian Experience: Some Common Themes

The second section of this book presents six studies that use the framework suggested above to examine the experience of Bangladesh, India, Nepal, Pakistan, Sri Lanka, and of the state of Kerala in India. The last 10-15 years have been a period of reform all over South Asia, although the year in which new economic policies were introduced varies from country to country. A Structural Adjustment Programme was introduced as early as 1977 in Sri Lanka, soon after in Bangladesh, in the mid-eighties in Nepal, the late 1980s in Pakistan and the early nineties in India. The nineties have been a decade of policy belief that trade liberalization and the globalization of production and marketing networks would be able to create an environment for more rapid economic growth. While South Asian countries have experienced positive growth rates over this period, the impact on poverty is less clear. Likewise, the impact on gender inequalities is problematic because there is such a wide variety of experience. In the first phase of the project each country team did a careful scanning of available secondary sources of data for each of the five countries (the conventional indicators). This reveals the trends over the last two decades in labour force participation of men and women, including unemployment and earnings data where available; education; health and demographic behaviour, to the extent that data are available. This statistical overview sets the context for examining the impact of the new economic regime. Since this impact is not expected to be uniform across different groups, each country team has selected in the first

phase a group of households in which women are working in EPZs or EOUs (with the exception of Sri Lanka, where six different groups were surveyed). This is justifiable since it is through changing patterns of work that many people would first experience the impact of liberalization: whether positively, through new jobs, or negatively, through the loss of jobs. The household surveys reported here largely focus on groups that have gained from new opportunities. Data on employment and conditions of work as well as a range of variables relating to intra-household matters, including labour use within households; perceived benefits of women’s workforce participation; decision-making and control over resources; family discord and violence and levels of stress have been collected. The household surveys were supplemented by a few case studies, to help in interpreting the findings. These studies are expected to provide some clues on how gender relations have been influenced by liberalization: with the caveat that the findings are not generalizable, and that they are focussed on groups that have gained new employment. Each of the papers presented here is based upon the larger studies done as a part of the project. Certain common themes run through these studies.

Female labour force participation

If new jobs are created for women in export oriented production it is possible that some women not previously in the labour force may get drawn into it. That is, there may be a net increase in female labour force participation. Macro data confirms such a trend, although the extent of change is small. Nonetheless, the fact that EPZs and export oriented units have generated a substantial number of jobs for women can be used as an argument that liberalization is good for women by giving them access to income earning opportunities. Even to the extent that new jobs have been created, one can conclude that this is so only after looking at a range of indicators and trying to assess the levels of well being that emerge, as the introduction to this book has argued. However, the argument relating to work participation itself needs to be confirmed.

All of the country studies included here suggest that female employment in EPZs or export oriented production is indeed high. They also show that the quality of this employment is generally poor. There is strong evidence that women’s workloads have

substantially gone up and very little evidence of any change in the allocation of household duties and responsibilities between men and women. Although widespread harassment has not been reported (as found by other researchers, see for example Swaminathan, 2002) this may be due to reluctance to speak on the part of the respondent, or it may reflect the situation in the sample selected. Field insights confirm that employers’ ‘preference’ for women has much to do with their docility and inability to protest against discomfort (see also Thorat, 1995). The interesting and somewhat counter-intuitive finding that emerges here is that women are positive about their new roles outside the household, despite the additional burdens of work and the poor quality of the employment itself. This cannot be attributed only to the increase in household economic security; women also report feeling positive about new social networks, new experiences and increased self-confidence.

The sample selected in each of the countries is purposive, with each household having at least one woman employed in export oriented production. In comparison with average female work participation rates in the country as a whole, the participation rates are thus higher in the sample. In addition, it appears that these new export oriented jobs have gone to women who were previously not working. However the men employed in these jobs had generally been previously employed elsewhere.

Salma Chaudhuri Zohir in her paper on Bangladesh presents the findings of a survey conducted in the Dacca EPZ among 112 households and 246 members of households. It was found that 84 per cent of the women workers were migrants, and a third of the women who had migrated to the area had come alone: notable because they are from Muslim households, where the level of seclusion imposed on women is generally high. The average size of a household in the EPZ was found to be smaller than among the general population (3.3 as against 5.6), educational levels are higher, and more women than men, 61 per cent as against 49 per cent, were found to be regular earners. Of the regular workers, 98 per cent of women (and 79 per cent of men) were salaried workers. Many women previously not in the labour force were found to be working in the EPZs. In the case of men, however, EPZ jobs had replaced others. Women seem to have gained more, workwise, than men: new jobs, more regular work, more benefits. In many ways the situation of these households is therefore very different from

elsewhere in the country. This has had its effect on household arrangements. It was found that, in general, men and women ate together, and men were willing to share in household chores. As the authors put it, ‘They also washed clothes, provided nobody was around while they were doing so’. And the attitude towards education of children has become gender neutral, that is, the desire to educate boys and girls is similar. Overall, there are both positive and negative impacts of female employment. On the positive side, there is greater employment security for women, higher standing in the family and greater influence over household decision-making. Against this have to be put the negative impacts, which were perceived more strongly by women than by men, and included conflict over domestic chores, neglect of children and a perceived threat to the joint family. In addition the total workload of women has gone up, and they face a reduction in mobility, since apart from going to work and coming home there is neither time nor approval for any other kind of mobility. Moreover, the environment outside the house has not changed in respect of other matters apart from work: women had less access to formal banking than men, were less aware of any trade union activity, and had low participation in decision-making outside the household, in the public/community sphere. Despite the evidence of some change in gender roles within the household, few men wanted to see women’s influence in decision-making increase in public and community matters.

In her paper on India, Swapna Mukhopadhyay cautions that any one study cannot lend itself easily to generalizations, especially in a country as large and as diverse as India. Macro data on the Indian economy show a similar direction of change in men and women’s employment over the last two decades, although rates of change differ. The share of the primary sector has fallen, and that of the tertiary sector has gone up, with some closing of the gap between males and females in urban areas. One notable change is the increase in the incidence of regular employment for women in urban areas. As far as levels of unemployment are concerned, educated women seem to be generally worse off than educated men; wages earned by women are systematically lower than those of men. On the whole, liberalization has created new opportunities for a small group of educated, urban women, with an increase in low paying, low productivity jobs for others: a picture that is broadly confirmed in the other countries too. The systematic decline in the

sex ratio in the 0-6 age group in India underlines the persistence of gender bias. The focus of the field surveys was to understand the manner in which an increase in female labour force participation may influence labour use within households. For this, a survey was carried out of 114 households and 272 individuals from these households, drawn from two locations in the states of Haryana and Uttar Pradesh. One sample was drawn from workers in an EPZ, the other from workers in EOUs. This was done so as to be able to contrast the situation of an EPZ, which is a created environment, with a more ‘normal’ one. The selected households all had at least one woman working in either an export processing zone or an export oriented unit. The nature of the work in the EPZs was such that women had better benefits and greater job security than men (as also seen in the Bangladesh sample). In the EPZ area, the women workers are largely migrants, living in a nuclear family context. This has necessitated some work sharing within the household, with a marginal change in men’s roles within the home, but the net effect is that women’s workloads have substantially gone up. Working outside the home is not accompanied by an increase in women’s control over their earnings. In the second sample, with women working in EOUs, the household context is different, since the families here are locally settled agricultural families. Most households have one or more non-working women. As some women go out to work, there has been a redistribution of household work between the women in the household, resulting in some conflicts emerging between women in the family; men seemed largely unaware of these conflicts and certainly had not changed their contributions to household activities. Despite the increased workload, increased conflicts over housework, and harassment at work, women were positive about their work. While men were positive largely because of the increased economic security that resulted from women’s work, women also seemed to value highly the new social networks they gained access to, as well as their increased confidence and experience. Men either did not notice these changes, or did not see their value. It could be argued that this has resulted in sowing the seeds of change, in a ‘changing mindset about women’s agency among women’.

Household surveys were not conducted in Nepal. A review of data suggests that over the last few years, the rate of male migration from the hills to the terai regions, always high, may have gone up

even further. This has left women with ever higher work burdens, including all agricultural activities. Despite having to live and work alone, the level of dependency upon males continues to be very high, a result of socio-cultural norms, high gender gaps in health and education, and limited access to earning opportunities. There may be differences in the experience of women from different ethnic groups, but overall, it is reported that the levels of male violence, and depression among women, are both high and increasing. Both trafficking of women and young girls, and commercial sex work, appear to be on the increase. It is possible that in Nepal, as in other South Asian countries, there has been some increase in employment for women in export oriented production, but clearly the scale is so small that an observer of the scene may not remark upon it. On the whole, there is little evidence of changing gender roles and in fact some worsening of the status quo because of high levels of domestic violence.

Rehana Siddiqui, Shahnaz Hamid, Rizwana Siddiqui and Naeem Akhtar, find that there has been some marginal diversification of women’s employment in the post-SAP period. The context in Pakistan is one where the share of women in manufacturing is very low to start with, and women workers are made more vulnerable by their concentration in certain occupations, informal contracts, and a lack of benefits. Gender disparities in education and access to health care are still glaring. For this study, a survey was conducted in three cities, among workers from 31 formal and informal establishments in the textile and garments sector, sports goods, surgical goods and fisheries, in other words, the main exportable products of the country. Over the period 1987–97, fisheries show a decline in their share in total exports, while the others show an increase. In all, around 298 workers from 250 households have been interviewed. A high concentration of women workers was found in stitching and related activities. Gender biases are also evident in earnings. Even when wage rates are similar, earnings were reported to be differentiated by gender because of differences in the rate of overtime payment. It was found that a much higher proportion of women workers than men reported seeking treatment for ailments from private doctors. The authors suggest that this may be because women do not want to lose more than the minimum number of working days, and are willing to spend more on their treatment by private doctors in the belief that

this treatment is superior. Economic hardship was an important explanation behind the incidence of children dropping out from school, with greater reluctance being evident in withdrawing boys from school. A wider range of jobs was acceptable for boys than for girls. The survey found that young women had a low level of control over assets or income, and it was older women within the household who exercised effective control. Men faced fewer controls on their spending. Women’s mobility was restricted to work/schooling; and the only sphere in which they could exercise effective decision-making related to household provision or child care. Employment outside the home has clearly added to household income and led to higher self esteem and an increased role in decision-making within the household, but both men and women were found to be reluctant to see more decision-making power with women.

The paper on Sri Lanka by Swarna Jayaweera has examined the impact of economic policy changes on women’s lives by looking at the experience of those working in the garment and textiles industries. The increased export orientation of industry has been the most visible result of Sri Lankan macro economic reform, and in this scenario garments have been a leading growth sector, while textiles have seen a mixed impact. Both are major employers of women. With garments, for example, 80 per cent of the labour employed in EPZs, and 90 per cent of that in the rural garment factories, consists of women between the ages of 18-30. As a spinoff from these garment factories, there has been an expansion in home based (sewing) work. In addition, large private textile mills have been opened. At the same time, withdrawal of subsidies has led to a collapse of village handloom industries, and the closure of public sector textile centres. The impact even within this one group of garment and textile workers has therefore been fairly heterogeneous. The survey conducted for this study has tried to capture this range of impacts and has not looked exclusively at those areas in which employment has clearly gone up. Garment factories in EPZs as also in rural areas have meant the creation of new work opportunities, mainly for women. However, there is no prospect of upward mobility for women in these factories, while this is not the case for men. New work has also become available to home based women workers, both on a self employed and sub-contracted basis. These new workers in garments are now part of a global market. Garment

work was not seen to confer higher social status on workers. However, workers in a privatized textile factory felt that their status in the community had improved. This factory provided more spaces for women, offering access to middle level employment. Handloom workers have stayed in this low return work, despite the general collapse of the sector, partly because state subsidy allows them to do so, and partly because of a lack of alternative employment. Textile workers retrenched from the public sector have been unable to find alternative work. For the latter two groups, having a reasonable level of education has not helped in finding other employment. Overall, there has been an extension of women’s productive roles for those employed in garments and in the private textile factories, but the quality of employment is poor, including heavy work loads, no job security, and no opportunities for upward mobility. The study also tried to examine the extent to which there has been any change in gender roles, through an in-depth survey of a sub-sample of the households selected above.

On the whole, there is little evidence of any change in gender roles, in for example a sharing of housework, or care of children, resulting in some role conflict. But there is a wide measure of gender equality in access to education and health care, and marital relations are not excessively patriarchal. Most men and women perceived positive changes resulting from women’s work outside the home, including greater economic security, self confidence and higher standing in the family. But at the same time, gender inequality is entrenched in many norms and practices that continue to be widely observed. For example, a majority of the women interviewed had undergone the ‘virginity test’ on marriage, and did not resent this, not even working women. Similarly son preference is present, although the situation of the girl child is probably better than elsewhere in South Asia.

Across South Asia, then, it seems that female work participation is high in export oriented production. In the EPZ areas, where most workers are migrants, there have been greater changes within the household. But women working in EPZs did not always disclose this to their families back in the village, suggesting that the attitudes to women’s work outside the home are still somewhat ambivalent. In the new households that have been created in the vicinity of EPZs, gender roles do appear to have changed somewhat with men making some contribution to household chores. Clearly, this is

born out of necessity, and it is not clear that men (or women for that matter) regard the new roles as anything other than context dependent. It is unlikely, given the unchanged perceptions in regard to other matters such as women’s decision-making role that these new roles would survive if the couple were, for instance, to return to the village.

Levels of well-being and stress

Women are positive about their new roles as workers outside the household, but are they in fact better off? The households in which women have found work are better off in the sense of having more economic security; is this reflected in higher levels of well-being and lower levels of mental stress?

To answer this difficult and complex question, an attempt has been made by four of the country studies — India, Bangladesh, Sri Lanka and Pakistan — to estimate the GHQ and SUBI, briefly mentioned above and discussed in detail in the paper by Shobhna Sonpar and Ravi Kapoor, and to correlate this measure with other characteristics such as levels of income, age, marital status and of course, sex. Levels of stress are found to be higher among women, but this is reportedly a worldwide phenomenon. More intriguingly, all four studies report lower levels of stress among those who are working in comparison to the non-working group.

In other respects, the findings from the four countries are in all cases credible, but not necessarily identical. For example, in India, an increase in age, marriage and education are all associated with higher levels of stress. This could be because of the physical hardships inflicted by age; the expectations and responsibilities that come with marriage; and the difficulty of meeting the expectations generated with education. However, a larger household size reduced stress, perhaps because it makes possible a greater sharing of responsibilities. In Bangladesh, an increase in household size, age, and migration increased stress, while marriage and higher incomes brought it down. Here, it seems that the greater work generated by larger households might be offsetting the work sharing possibilities to increase stress. Migrants probably have higher stress because their situation is more precarious with fewer fallback options. The effect of marriage in reducing stress may be linked to a sense of security, while that of income could be linked to greater coping mechanisms.

In Pakistan, age seemed to reduce stress for women, but increase it for men; an increase in income reduced stress levels more significantly for men than for women. The effect of age is probably linked to the marked increase in autonomy and control within the household that older women have in this culture. The gender difference in the effects of income suggests that economic security for women could be set off by non-economic considerations more than in the case of men. In Sri Lanka, education and marriage reduced stress and increased well-being, while age tended to increase stress and reduce well being.

In abstract terms, one can argue several possible ways in which any of these factors might work. For example, education can be expected to increase coping skills, perhaps increase opportunities and hence contribute to lower stress; on the other hand it could also generate new expectations and an inability to fulfill them might result in higher stress. Similarly larger households generate more work but also a stronger support system. The differences observed across countries therefore probably reflect the particular combinations most evident in that sample. But the fact that being a worker — despite the poor quality of the work, and all the other possible sources of higher stress — is associated with lower stress and higher well-being is important, and more than justifies the attention that is being given to this variable.

The Nepal study did not attempt this kind of analysis, but it has identified depression among women, and violence against women as a special concern, based on other data. It has been suggested that the reasons include low levels of social development as well as very limited economic opportunities for women, both of which have made women highly dependent on males within a strongly patriarchal society.

Within South Asia, Kerala (and Sri Lanka) are held out as examples of the high levels of gender equality and social development that can be attained even at low levels of economic achievement. The case of Kerala is specially interesting because in many ways it shares a common tradition with other parts of India. More recently, some people have argued that the position of women is not as good as may appear from a review of ‘conventional’ indicators in Kerala, and that in fact women face high levels of male violence, and display other manifestations of mental distress. Moreover domestic violence and dowry deaths have increased even

as women’s educational levels have increased. Mridul Eapen and Praveena Kodoth present an analysis of why, and how, the position of women has possibly been on the decline in this high achieving state. They suggest that social development in this area has gone hand in hand with changing social practices that have led to a strengthening of patriarchy and lower status for women. Family practices and structures have changed in such a way that even within the traditionally matrilineal groups women’s access to inherited resources has gone down. The patrifocal or male centred family has been encouraged through legislation and the adoption of practices associated with patriarchal family structures. This has eroded the traditional strengths that women had within a matriarchal tradition, which include less stringent control over sexual choice, an uncontested right to live in the natal home, a more positive attitude towards girls, and lineage through women. At the same time education and employment have not been able to compensate through greater access to independent earnings. This is because the emerging norms of masculinity and femininity direct women towards lower productivity and lower skill occupations. The net effect may have been that women are left with lower control over resources and a weaker position within the household. The high rate of out migration from Kerala to the Gulf countries has led to high remittances, which have been channeled largely into consumption, including large expenditures on dowries. The net effect has been an increasing observance of customs that are linked to the emerging ‘domestic’ and subservient role of women within the home. Overall, the Kerala experience has been that ‘education and employment have not played the transformative role so generally expected of them’. On the contrary, education and employment are being used to mediate new norms of ‘masculinity’ and ‘femininity’, strengthening the notions of a male ‘worker’ and of ‘domestic’ women.

It seems then that we have a somewhat paradoxical situation that women’s increased participation in paid work outside the home has clearly added to the total burden of their responsibilities, and probably contributed to role conflicts within the home and higher levels of stress. And yet, the general levels of stress among those who are working are lower than among those who are not working, and women value their new roles for reasons other than the fact of greater economic security. It can perhaps be hypothesized that while

the ‘static context’ shows little sign of change — in fact the case of Kerala suggests a retrograde movement as far as women’s position is concerned — it is still possible that women’s work may lead over time to a stronger women’s agency. However, these studies also suggest that education has so far played a weak role in any social transformation of gender roles. The link between education and employment also seems to be weak. The fact of being educated may be used, as Swapna Mukhopadhyay suggests, as a screening device in offering employment, and not as a route to any occupational mobility. Although the levels of education among the women workers surveyed were higher than in the general population, this is not generally associated with a higher status in the community. And among those who have been retrenched, as shown by the Sri Lankan study, being educated has not helped in finding new work. The strongest finding about education comes from Kerala, where other forces have proved much stronger in influencing the position and status of women, and neither education nor work has been able to lead to any transformation.

Intra household matters and the world outside

This study has been motivated by a desire to see how far women’s work may have affected roles and responsibilities within the home. Women’s work outside the home could be an important conduit for change. The household surveys attempted to document the opinions of respondents regarding women’s decision-making roles. In all the countries, there is little evidence that women’s control over income or assets has increased as a result of their work outside the home. Where this control is high, as in Sri Lanka, it is the result of a more equitable context that is already in existence. Women’s mobility is also restricted, and although there is a high degree of toleration in so far as work timings are concerned, women are expected to and indeed need to be at home when not at work. The reporting of violence is low, but it exists, and a strong sense of the need to conform, to take care lest violence be provoked was evident everywhere. So that, despite the fact that women have placed a high value on the experience, confidence and social networks that come with work, there is little evidence that any observable change has actually occurred in their position or status. Women do participate in decision-making within the home; but should their role in this

increase? In Bangladesh, women were found to have more autonomy in daily provisioning, and in respect of jewellery. While two-thirds of the women felt they should have more decision-making power in the community, only a third of the men thought so. Those who were opposed cited women’s lower levels of experience and knowledge, but also felt it would affect men’s dignity. A similar finding comes from Pakistan, with few men being in favour of increased decision-making by women either within or outside the household. It should also be noted that a large proportion of women felt that they were not equipped to have a greater say in decision-making. In India, women felt their role as decision-makers had gone up, although there is little objective evidence of this. In Sri Lanka, a much higher proportion of both men and women were in favour of increased participation by women in the community, than were in favour of increasing their participation within the home. This could reflect the low participation in community matters, or as Swarna Jayaweera puts it, it could mean that ‘increase in decision-making outside the family was less problematic than an increase within the family’.

There is, on the whole, no evidence that women’s role as decision makers has greatly changed, and in fact given the demands on their time, even if attitudes had changed, active participation in community matters may have been difficult. But this has not happened so far.

The Macro Perspective

The country papers are based on the findings of household surveys and case studies mainly conducted among households in which women have found new work outside the home. It is difficult to generalize from these studies however valuable they are as micro perspectives, given the large size of the region and the heterogeneity of experience of different groups within each country. The third section of the book tries to present a macro perspective as well to examine the tools that are currently available for macro analysis of the gender-differentiated experience of economic reform.

The paper by Manju Senapaty briefly reviews available evidence on the impact of trade liberalization on employment. Her reading of the literature suggests that export oriented manufacturing has

benefited women through the creation of jobs, and regular earnings may have increased their influence on household decisions. However, gender wage gaps persist and the quality of employment may be poor. All attempts to link gender and trade face the absence of reliable gender disaggregated data. Detailed data on employment of men and women in production sectors are not available. Using whatever data are available on the manufacturing sector for India, she finds that the female intensity of employment in this sector showed an increase from 30.16 per cent in 1987-88 to 30.33 per cent in 1993-94, which is an extremely small change. Further, in major export sectors in which female intensity has increased, the proportion of employment going to women was small. This makes the evidence of feminization and new work opportunities somewhat less convincing than from the country studies. Data inadequacies are a major reason for this, and must remain an important area for future research efforts. Apart from the absence of gender disaggregated data, the under counting and the relative invisibility of informal and home based workers is a further deficiency in official statistics at present. However, the development of the international trading regime is likely to have a major impact on some important female intensive sectors. For example, textiles and garments, are among the most important export sectors for South Asian countries, with a contribution to total export earnings ranging from 25 per cent in India, to 52 per cent in Sri Lanka, 60 per cent in Pakistan, and over 73 per cent in Bangladesh, in 1998. A review of the Agreement on Textiles and Clothing (ATC) and its impact on South Asia suggests that effective market access for developing countries is still low.

Accepting the limitations posed by data, what are the tools that are available for gender sensitive macro analysis? The paper by Anushree Sinha is an attempt to use a computable general equilibrium (CGE) model structure to examine the gendered impact of trade liberalization, using data on India. The disaggregation of data using a simple social accounting matrix framework is a necessary and interesting prelude to such modelling. This confirms the usual observations of women being concentrated in certain sectors, in low status labour categories, and at lower levels of earning than men. It is also seen that in all types of households the share of female unpaid workers is much higher than that of female paid workers, while the reverse is true of men. This brings to the fore the need

to include the reproductive and non-market sector in any attempt to understand the allocation and re-allocation of women’s labour in changing economic environments, an area for future research. The preliminary exercises included here suggest that liberalization would bring welfare gains for all, albeit more for men. It is difficult however to read much value into the policy prescriptions of this model, simply because the assumptions are very restrictive and do not allow for simulation of situations closer to reality.

Both these macro economic perspectives recognize the specific nature of the constraints on women and the interaction between housework and work outside the home. But using available data, and existing methodology, clearly constrains the findings. It is simply not possible to introduce the kind of nuanced analysis that emerges from the country papers. Quite simply, there is a large research agenda here.

From Research to Policy

Ultimately, we want not just to understand the world better, but also to change it. In discussions of globalization and its possible adverse impacts, there is a ‘pro’ and an ‘anti’ position. The picture emerging from this book, however, shows up the shades of grey that lie between these positions.

One clear finding of the country studies is the poor quality of employment that has been generated for women. Although this new employment may have brought women into the global market, it is on terms unequal with men and in conditions of work that have not, so far, created any change in their situation within the home or outside it. Those who have found work have been able to contribute to improving the economic situation of the household and reducing the intensity of poverty, but they have not been ‘empowered’.1

It should also be remembered that a large number of people are excluded even from these minimal gains. The studies presented here

1Economic empowerment has been defined as ‘economic change/material gain plus increased bargaining power and/or structural change which enables women to secure economic gains on an on-going and sustained basis.’ See Carr, Chen, and Jhabvala, p. 11.

have focused on the situation of those households where women have found some work. The Sri Lanka study gives us a little insight into the difficult situation in which those households that have not benefited in this way, find themselves.

In countries in which only a small absolute number of people have been drawn into the global economy, and a large majority are excluded — as seems to be true of South Asia — there can be no assessment of globalization that does not take a careful look at what is happening outside this limited sphere. This is an area for future research. However, South Asians who argue the need to look carefully at what would best serve the interests of the excluded are often accused of ‘localism’ and of an uncritical rejection of an economic policy approach that actually holds out the most potential of maximum gain for the majority of people. From a South Asian perspective, however, it is difficult to understand the tremendous rejection of what is loosely described as ‘localism’ by those who are strong proponents of globalization. For example in our present context, of trying to understand the changes that have taken place and those that have not taken place in women’s lives, there seems to be little gain in focusing exclusively on the ‘dynamic’ context of change, and ignoring the ‘static’ context. The static context, which is the social reality, shapes our experience of economic change and can, as the experience of Kerala shows, result in unexpected outcomes. If concern with the static social reality is a part of localism, it is a crucially important concern.

Limiting ourselves to the question of gender, change in gender relations can be initiated or led ‘from above’, through policy advocacy, legislation, international opinion and conventions, and so on. Or women themselves can demand it, ‘from below’. In South Asia, the former is far more visible. The basic reason has been well put by Shobhna Sonpar and Ravi Kapur in this volume: ‘conflicts of interest may be suppressed not only from the decision-making agenda, but also from the consciousness of the parties involved. Here both the dominant and the subordinate parties subscribe to accounts of social reality that deny that any inequality exists.’

If women’s agency is strengthened, it will increase their bargaining power, and influence the outcome for those issues that are on the decision-making agenda. What, however, of those matters that are suppressed even from consciousness? To what extent do the processes of liberalization and globalization contain

the possibility of changing people’s perceptions of social reality, and hence of generating a demand ‘from below’?

In suggesting that increased participation by women may have sown the seeds of change, we are suggesting that liberalization may indeed contain the possibility of changing such perceptions. But the evidence so far is inadequate to suggest that such change has actually occurred up to now. The only situation in which it may have happened to some limited observable extent is those in which women have been drawn into global connections at a high end and therefore enabled to change the manner of their participation in social and economic spheres in substantive ways.

But for the large majority who have gained a global connection at the low end, for home based workers or women in EPZs with no prospect of any job mobility, it is business as usual.

From a policy perspective, the question is: what kinds of interventions could alter the external environment of the household in such a way that an impact is made both on the consciousness of men and women, as well as in their relative bargaining power, such that a movement towards more equitable and mutually supportive gender roles becomes possible.

Most of the papers in this volume have not attempted to formulate clear policy recommendations, with some exceptions. For example, the paper on Pakistan recommends schooling, child care and training as effective tools to empower women. It also recommends the need for shelter and effective legal cover to help women who are victims of gender-based violence. The role of education however, can be uncertain as the study on Kerala shows. The macro studies suggest that available tools are not yet sensitive enough to be used for gender sensitive policy analysis.

But even if no single, clear message is available of the kind ‘do this, and economic reform can be accompanied by greater gender equality’ the research presented here clearly demonstrates the need for economists to incorporate a gender dimension in their analysis. It is true that women’s work may hold the potential for strengthening women’s agency. It is also true that this will not happen unless the quality of the employment improves substantially, and unless responsibilities for household work are more equitably distributed. Indeed, if such changes do not happen soon enough, even the present levels of women’s work participation may become unsustainable. These concerns need to be made central

to an assessment of economic performance. It is equally clear that such an assessment cannot be exclusively based on quantifiable indicators, and that it has to be embedded in social reality.


Carr, Marilyn, Martha Chen, Renana Jhabvala (ed.) (1996) Speaking Out: Women’s Economic Empowerment in South Asia, Delhi: Vistaar

Economic and Political Weekly (1996) Special Issue on GDI and GEM for Indian states, 31(43) 1996

Greif, Avner (1994) ‘Cultural Beliefs and the Organization of Society: A Historical and Theoretical Reflection on Collectivist and Individualist Societies’, Journal of Political Economy, Vol. 102, No. 5, October 1994. pp. 912–50

Mukhopadhyay, S. and Ratna Sudarshan (1997) ‘Gender Discrimination in an Age of Liberalization: Some Issues’, paper presented at Workshop on Gender Discrimination under Structural Reforms, ISST, New Delhi

Palaskas, Theodore B. and Barbara Harris-White (1993) ‘Testing market integration: new approaches with case material from the West Bengal food economy’, Journal of Development Studies, Vol. 30, No. 1, October 1993, pp. 1–57

Saith, Ruhi and Barbara Harris-White (1998) ‘Gender sensitivity of well-being indicators’, UNRISD Discussion Paper No. 95

Swaminathan, Padmini (2002) ‘Labour-Intensive Industries but units without ‘workers’: where will ILO’s social dialogue begin?’ Madras Institute of Development Studies, Working Paper No. 168

Thorat, Minakshi (1995) ‘Labour Administration, Enforcement and Gender’ in Cherian Joseph and K.V. Eswara Prasad (ed.) Women, Work and Inequity: The Reality of Gender, National Labour Institute, New Delhi, pp. 169–229

UNDP (1995) Human Development Report, New York: Oxford University Press



Mridul Eapen’s areas of professional interest include: Industry (has worked on Textile and Automobile Industry); Rural Transformation focussing on Rural Industrialization, and, (more recently) Gender Studies. She has edited a volume (together with others) on Indian Industrialisation: Structure and Policy Issues, 1992; is author of the book, Rural Industrialisation in Kerala: Its Dynamics and Local Linkages, 2000.

Swarna Jayaweera is co-ordinator in the Centre for Women’s Research in Colombo, Sri Lanka and is an Emeritus Professor of Education at the University of Colombo. She has published widely in national and international journals, and has served as a consultant to several international agencies.

R.L. Kapur is an internationally known expert in Clinical Psychiatry. He has been associated with the University of Birmingham as a Professor, the National Institute of Mental Health and Neuro Sciences in Bangalore and is currently Professor at the National Institute of Advanced Sciences, Bangalore. He has been associated with a number of research projects dealing with mental health.

Praveena Kodoth is a Research Associate at the Centre for Development Studies, Thiruvananthapuram. Her research interests very broadly are on the concerns of ‘Development’, Gender and Property Relations.

Swapna Mukhopadhyay is overall Project Director, Gender Network Project and Director, Institute of Social Studies Trust New Delhi. She has taught in several institutions in India and abroad and has been a consultant with many international organizations. Her research interests range from theoretical economics and econometrics to labour market issues and women’s studies. She has designed and supervised the Gender Planning Network project from its inception.

N. Sangeeta has an M.Phil (Statistics) from the University of Delhi.

Manju Senapaty is an economist with the Government of India (Indian Economic Service) and has served in the Ministries of Finance, Human Resource Development and in the Ministry of Commerce. Currently she is working as an Economic Adviser with the Delhi office of Department for International Development (DFID), UK.

Rehana Siddiqui is currently working as Senior Research Economist at the Pakistan Institute of Development Economics. She completed her PhD from Columbia University, New York, and has worked on the issues of economic growth, gender, exchange rate misalignment, and human resource development.

Shahnaz Hamid is currently working as Research Economist at the Pakistan Institute of Development Economics. Her area of specialization includes human resource development, including gender.

Rizwana Siddiqui is currently working as Research Economist at the Pakistan Institute of Development Economics. She has worked extensively on trade issues. Using CGE models and has also worked on issues related to poverty and structural adjustment.

Naeem Akhter is currently working as Research Economist at the Pakistan Institute of Development Economics. She has worked on issues related to trade, industry, and exchange rate misalignment.

Anushree Sinha is Principal Economist in the Macroeconomic Monitoring and Forecasting Division at the NCAER. She has wide experience in macroeconomic analysis with a focus on issues of development and policy assessment. During the last few years she has been extensively involved in research on macro issues of the Indian informal sector and gender.

Shobna Sonpar is a clinical psychologist with special expertise in the field of counselling and psychotherapy. She had worked as a student counsellor at IIT Delhi for 11 years and is now in private practice. She is a trustee of Psychological Foundation, an organisation that seeks to facilitate the work of psychologists and other mental health professionals in clinical practice and has been the editor of its
journal. Her current interests are in the area of gender and mental health, and in collective violence.

Ratna M. Sudarshan is currently Principal Economist, Human Development Programme Area, National Council of Applied Economic Research, New Delhi. She directs a programme of research on gender and informal economy concerns. Other research interests include education, the policy process, and researcher-activist linkages.

Salma Chaudhuri Zohir is a Research Fellow in the Bangladesh Institute of Development Studies at Dhaka. She has done extensive research on women working in the textile sector of Bangladesh. Her research interests are mainly in gender and development issues.