Poverty is not only about 'poor' people but also about the social and economic inequalities that compound and reproduce poverty. Email: poverty@ids. ac. uk Web: www. ids. ac. uk/research-teams/vulnerability-and-poverty-reduction-team/researchthemes/ poverty-inequality-and-wellbeing PI WP001 The Vulnerability and Poverty Reduction (VPR) Team aims to construct dynamic and multidimensional perspectives on vulnerability and poverty in order to transform thinking, policy and practice. The VPR team produces working papers on social protection; conflict, violence and development; and poverty and inequality.
Follow this link to view a full list of publications: www. ids. ac. uk/go/research-teams/vulnerability-and-poverty-reduction-team Women’s Empowerment Revisited: From Individual to Collective Power among the Export Sector Workers of Bangladesh Naomi Hossain IDS Working Paper 389 First published by the Institute of Development Studies in March 2012 © Institute of Development Studies 2012 ISSN: 2040-0209 ISBN: 978-1-78118-046-4 A catalogue record for this publication is available from the British Library. All rights reserved.
Reproduction, copy, transmission, or translation of any part of this publication may be made only under the following conditions: • with the prior permission of the publisher; or • with a licence from the Copyright Licensing Agency Ltd. , 90 Tottenham Court Road, London W1P 9HE, UK, or from another national licensing agency; or • under the terms set out below. This publication is copyright, but may be reproduced by any method without fee for teaching or nonprofit purposes, but not for resale. Formal permission is required for all such uses, but normally will be granted immediately.
For copying in any other circumstances, or for re-use in other publications, or for translation or adaptation, prior written permission must be obtained from the publisher and a fee may be payable. Available from: Communications Unit, Institute of Development Studies, Brighton BN1 9RE, UK Tel: +44 (0) 1273 915637 Fax: +44 (0) 1273 621202 E-mail: bookshop@ids. ac. uk Web: www. ids. ac. uk/ids/bookshop IDS is a charitable company limited by guarantee and registered in England (No. 877338) 3 Women’s Empowerment Revisited: From Individual to Collective Power among the Export Sector Workers of Bangladesh
Naomi Hossain1 Summary Bangladesh has become known as something of a success in advancing gender equality since the 1990s. There have been rapid gains in a number of social and economic domains, yet by most objective standards the current condition and status of women and girls within Bangladeshi society remain low. Rapid progress has come about under conditions of mass poverty and interlocking forms of social disadvantage, political instability and underdevelopment, overlain with persistent ‘classic’ forms of patriarchy.
Mass employment of women and girls in the country’s flagship export sector – the readymade garments (RMG) sector – has been one of the more visible and prominent changes in women’s lives since its late 1970s’ introduction. Whether and the extent to which RMG or garments employment has changed the lives of women workers for the better has been the subject of much debate, and the research and analysis it has generated offers valuable insights into the processes of economic and social empowerment for poor women in low income developing countries.
Yet as this paper notes, close observers of social change in Bangladesh have become dissatisfied with the limits of a focus on individual economic empowerment. Paid work may enable some women to negotiate the ‘structures of constraint’ that shape their lives and relationships, but what of the structures of constraint themselves? In the Bangladesh context the experience of mass RMG employment has given rise to questions about whether women have gained greater recognition as citizens with rights and roles as carers in the private and political actors within the public spheres.
Revisiting the question of women’s empowerment in this context means interrogating whether paid employment has contributed to investments in the education and skills of women and girls, improvements in their public safety and rights to occupy public space. Given labour militancy in the sector and its partial successes in raising the minimum wage, what has the experience of labour politics meant for women’s political empowerment? Drawing mainly on the rich literature available on women’s RMG employment, this paper explores the wider and less well-documented effects of such employment on public policy relating to gender equality in these areas.
It concludes that the overall direction of change in the industry points plainly to the need for investments in worker productivity, with a host of implications for women’s work and gender equality more broadly. Factory owners have to date shown few signs of recognising their interests in supporting better state health, education and public safety for women and girls, or changing management practices to retain and raise productivity of skilled women workers.
Yet with downward pressure on wages increasingly effectively resisted by workers at a time of global economic volatility and rising living costs, the tide may now be turning for the RMG workers of Bangladesh. Productivity gains require the state and the industry to treat women workers as full citizens with public policies that promote their skills and safety and respect, and which guarantee the representation of their rights and demands.
RMG employment continues to be a source of empowerment for women in Bangladesh, but social and economic change means that that power now depends less on the individual economic effects of paid work on household decision-making than it once did. RMG employment is increasingly a source of power for women because of its more collective effects on women’s citizenship and political agency. This matters all the more because of how this group is exposed to the volatilities of the global economy. 1 Research Fellow, Participation, Power and Social Change Team, Institute of Development Studies. 4 Contents
Summary 3 Acknowledgements and Disclaimer 5 1 Introduction: the state of women’s empowerment in Bangladesh 6 1. 1 The idea of success 6 1. 2 Social indicators 7 1. 3 Economic participation 8 1. 4 Political participation and security 10 1. 5 Changing society in the post-conflict era 11 1. 6 Empowerment revisited 13 2 The RMG sector 14 2. 1 The economic significance of the RMG sector 15 2. 2 The political significance of the RMG sector 15 3 Women as RMG workers 17 3. 1 Hard work 18 3. 2 Workers’ experiences of economic and social empowerment through garments work 21 3. 3 Women’s wages and male power 23 The wider effects of RMG employment 24 4. 1 The impact on men’s attitudes 24 4. 2 RMG employment and the expansion of basic education for girls 25 4. 3 Women’s mobility and the feminisation of public space 27 4. 4 Citizenship and political empowerment 28 5 Conclusions 30 Annex 31 References 33 Tables Table 1. 1 Selected changes in Bangladeshi women’s health and education 7 Table 1. 2 Years of education in the labour force by quintile 2000–5 9 Table 3. 1 Gender division of RMG jobs (% of workers) 19 Table 3. 2 Workers’ reports of incidents that contribute to insecurity (% of workers) 20
Figures Figure 1. 1 Labour force participation rates for men and women, 1999–2000 and 2005–6 8 Figure 1. 2 Proportion of seats held by women in national parliament, 1990–2009 10 5 Acknowledgements The author is grateful to Carolina Sanchez-Paramo and Sudhir Shetty of the World Development Report Team for comments on the first draft, Samer Al-Samarrai for insights into the issue of education and skills, and colleagues at BRAC Development Institute, in particular Ferdous Jahan, Sohela Nazneen and Mamunur Rashid for work done jointly on women’s empowerment (Nazneen et al. 010; 2011) and on the effects of the food, fuel and financial crisis (Hossain et al. 2009; Hossain et al. 2010). The work with BDI colleagues has generated many of the insights into the wider context of women’s empowerment, and the recent global economic shocks experienced by garments workers discussed in this paper. Useful comments were also received from colleagues at the Dhaka World Bank office. The author is particularly grateful to Sohela Nazneen for her insights and knowledge of relevant recent literature and public perceptions of garments workers. Disclaimer
While a version of this paper was originally prepared as a background paper for the World Development Report 2012 on Gender Equality and Development, it does not reflect the views of the World Bank. All errors and omissions are those of the author alone. An earlier version has been published as: ‘Exports, Equity, and Empowerment: The Effects Of Readymade Garments Manufacturing Employment On Gender Equality In Bangladesh’, http://siteresources. worldbank. org/INTWDR2012/Resources/7778105- 1299699968583/7786210-1322671773271/Hossain-Export-Equity-employment. df 6 1 Introduction: the state of women’s empowerment in Bangladesh This paper is going to try to assess the effects of RMG employment on women’s empowerment in Bangladesh. To do so, it will start by setting out evidence of change in women’s lives in Bangladesh, as well as of areas of persistent disadvantage and domination. It will present some perspectives on what these changes mean, and in so doing set out some markers for the future direction of analysis of women’s empowerment in this setting. 1. 1 The idea of success
Bangladesh has come to be seen as a success in promoting gender equality in recent years. A recent study summarised it as follows: Bangladesh stands out as the shining new example in South Asia of a poor country achieving impressive gains in gender equality ... a country that had been famously written off by Henry Kissinger as a ‘basket case,’ which now dwarfs India and Pakistan in many areas. Between 1971 and 2004, Bangladesh halved its fertility rates. In much of the country today, girls’ secondary school attendance exceeds that of boys. The gender gap in infant mortality has been closed.
The micro-credit revolution continues to boost women’s solidarity groups and earning potential, and vast numbers of young women are leaving their villages to work in garment factories where, in earlier generations, young women were rarely seen outside their homes. (World Bank 2008a: 3) Other donor documents similarly note ‘first generation’ achievements towards gender equality, typically with reference to gender parity in basic education, Bangladesh’s ‘pioneering’ role in micro-credit, and formal sector employment of two million young women in the readymade garments (RMG) industry (Nazneen et al. 011). The official view is a moderate version of the same. The Government’s recent National Strategy for Poverty Reduction document noted that ‘[w]omen in Bangladesh have made important gains along with changes in social attitudes towards women’s economic participation’ (Government of Bangladesh 2009a: 3): [M]easurable progress in women’s advancement and rights in a number of areas including education, participation in labour force, health and nutrition, and participation in public services.
In the area of women’s advancement and rights, the government has made strong commitments and undertaken various initiatives to reduce the gap between men and women. (Government of Bangladesh 2009a: 62) The idea of successful progress towards gender equality could be read as a comment on the unpromising context in which these rapid changes have taken place, characterised by chronic mass, severe and interlocking forms of poverty and disadvantage (Sen and Hulme 7 006); political instability and weak governance (IGS 2006; 2008), and persistent ‘classic’ forms of patriarchy. 2 The idea of success chimes to some extent with perspectives from the women’s movement. For the women’s movement, progress in the present era is slower compared to the 1990s’ ‘golden age’ of democratic transition (Nazneen and Sultan 2010: 70). There are also signs that advances can be reversed, and the growing influence of conservative versions of Islam on national politics and society are cited (ibid. . The mood in the women’s movement is more cautious; having itself emerged out of the two-part national struggles for independence, the women’s movement has a healthy respect for the possibilities of change through mobilisation and collective action, and extensive experience of negotiating tricky political terrain. So the feminist struggle continues. 1. 2 Social indicators What is the substance of this success, and what are the remaining challenges?
In 2010, Bangladesh ranked 116th in the Gender Inequality Index of the UNDP’s Human Development Index, below Pakistan’s 112, yet above India’s 122 (UNDP 2010); even within South Asia, then, it is not apparent that Bangladesh has much to celebrate in terms of the levels of gender equality (nor, indeed, Pakistan or India). Yet the pace of change merits attention: gains for Bangladeshi women were made from a lower starting point, and caught up fast given the modest pace of poverty reduction, particularly in the 1990s.
Table 1 summarises key changes in health and education indicators over the last three decades. Table 1. 1 Selected changes in Bangladeshi women’s health and education Indicator Outcome Fertility Total fertility rate declined from 7. 3 (1974) to 2. 7 (2007) 1 Maternal mortality Maternal mortality ratio (maternal deaths per 100,000 live births) more than halved from 648 (1986) to 315 (2001) 2 Basic education Gender parity in primary and secondary enrolment achieved by early 2000s; girls now outnumber boys3 Sources: NIPORT (2009) 2 World Bank (2007) 3 Chowdhury et al. (2002) The most striking changes have been seen with respect to girls’ education. Bangladesh closed the gender gap in enrolment at primary by the end of the 1990s, ahead of the MDG targets and many comparator countries (Chowdhury et al. 2002). More girls than boys now enrol in secondary school, drawn in substantial part, it is believed, by the availability of cash stipends for all unmarried girls who attend and perform to a minimum level (Al-Samarrai 2009).
Yet while Bangladesh has done well compared to other countries in widening girls’ access to school, quality and attainment for boys and girls are absolutely low and the closure of the gender gap in basic education may reflect stagnating educational access among the ‘boys left behind’ (World Bank 2008a; Shafiq 2009; Tariquzzaman and Hossain 2009). After the rapid progress of the 1990s, Bangladesh’s more recent MDG performance has been disappointingly average. Kandiyoti’s (1988) term refers in the Bangladesh context to a society characterised by ‘patrilineal principles of descent, patriarchal structures of family organization, the practice of female seclusion, and a marked preference for sons over daughters’ (Kabeer 2004: 14). 8 1. 3 Economic participation Patterns of women’s economic participation have changed fast. Women’s labour force participation rate grew faster than that of men in the 1984–2000 period (Rahman 2005), yet remained comparatively low at 22. 8 per cent in 2000 (25. in urban areas). However, women’s wage employment increased considerably after that, growing at 4. 3 per cent each year between 2000 and 2005 (World Bank 2008b; see also Figure 1). Figure1. 1 Labour force participation rates for men and women, 1999–2000 and 2005–6 Source: Labour Force Survey 1999–2000; 2005–06 (BBS 2002; 2008) In relation to girls’ increased access to education it is worth noting that official figures suggest that labour force participation rates for younger women (aged 15 to early 20s) actually declined during the early 2000s.
This suggests that more women in this age group were entering education in preference to early entry into the low-skill end of the workforce. This would be consistent with the finding that the gender wage gap narrowed at the upper end in the first half of the 2000s, as higher educated women gained access to employment (particularly in the public sector; Al-Samarrai 2007). The narrowing of the gender wage gap reversed the pattern of change over the second half of the 1990s, in which the share of women’s formal sector employment shrank from 20. (1995–96) to 8. 9 (1999–2000) per cent (Rahman 2005). The result of this shift has been that women’s wages increased much faster than men’s over the first half of the 2000s, and the gender gap in income and wages narrowed considerably, particularly at the upper end of the occupation hierarchy (Al-Samarrai 2007). This was substantially to do with the changing composition at the top end of the female labour force: of the richest 20 per cent of women, 58 per cent had been in salaried work (e. . , skilled public sector employment) and 26 per cent self-employed in agriculture in 2000; these proportions changed to 76 per cent in salaried work and only 3 per cent selfemployed in agriculture in 2005 (Al-Samarrai 2007). In short, there are not only more women in paid work in Bangladesh than in the past, they are also increasingly concentrated in formal sector employment. 9 A critical factor in the narrowing of the gender wage gap appears to have been gains in education (see table 2).
However, as recent real wage and income gains for women have been concentrated among the more educated and typically more affluent population, there will have been more limited impacts for poor women. And despite these gains for women who have succeeded in catching up with men educationally, on average, women continue to earn 21 per cent less per hour than men (Kapsos 2008). A considerable proportion of the gender wage gap remains explained by simple discrimination, as well as by labour market segmentation that excludes women from better-paid occupations (Ahmed and Maitra 2010; Kapsos 2008).
Table 1. 2 Years of education in the labour force by quintile 2000–5 quintile 2000 2005 male Female total male female total 1 1. 6 0. 6 1. 4 2. 0 1. 4 1. 9 2 2. 2 1. 2 2. 1 2. 7 1. 5 2. 6 3 3. 0 1. 7 2. 9 3. 7 3. 1 3. 7 4 4. 5 3. 4 4. 4 5. 3 4. 7 5. 3 5 7. 7 6. 8 7. 6 8. 6 8. 4 8. 5 Total 3. 9 2. 7 3. 8 4. 6 4. 0 4. 5 Source: Al-Samarrai (2007) based on HIES data Some 60 per cent of the increase in women’s paid work during the 2000s was concentrated in urban areas, half overall in manufacturing sectors.
Over two million women are estimated to be employed in the RMG industry, which dominates the Bangladesh manufacturing export sector. Given that garment factory careers may be short, many more than the current two million women are likely to have experienced factory work over their lifetimes, so that the social effects of RMG employment are likely to have been more widely shared across the population than the two million figure suggests.
There has also been significant recent growth in public sector employment as teachers or health workers, and in self-employment and household enterprises, often in response to public policy measures to recruit women in order to better advance gender equality goals in the social sectors (health, education, fertility control and poverty reduction). An older, much-debated pathway of women’s economic empowerment has been microcredit, in which the scale of Bangladeshi women’s collective participation has been unprecedented.
Micro-finance programmes for women expanded fast from the second half of the 1990s; by 2006, it was estimated that there were some 16. 4 million micro-credit borrowers in Bangladesh, the overwhelming majority of them women; the Government of Bangladesh is itself a major source of wholesale financing for micro-credit and of credit programmes (World Bank, 2006). Debates about the impacts of micro-credit on women’s empowerment have rehearsed themes around control over resources and the effects on women’s domestic bargaining power (Goetz and Sen Gupta 1996; Kabeer 1999).
Despite enduring scholarly scepticism about the benefits of micro-finance, it features prominently in domestic development debates as a strategy for addressing vulnerability and poverty. Evidence is available to show a range of empowering effects, including that participation in micro-finance programmes enabled discussion of family planning with spouses, an expanded role in household decision making, more access to ‘financial, economic, and social resources’ and greater mobility (World Bank 2006: 24).
In relation to intra-household relations and women’s economic empowerment, the 2007 Demographic and Health Survey found that while fewer women reported making decisions about the use of their income alone 10 than in 2004 (from 39 to 31 per cent), the proportion of women claiming involvement in joint decision making had risen, from 47 to 56 per cent (NIPORT, 2009). There are some good reasons to believe that the economic roles within Bangladeshi gender relations have experienced a significant shift, and that micro-credit has played a part in that.
More recent concerns about possible market saturation and serial indebtedness have shifted attention to the problem of too much, as opposed to too little, access to finance. 1. 4 Political participation and security In contrast to their striking gains in human development and new economic opportunities, and despite the two top political leaders both being women, Bangladeshi women have fared far less well with respect to their political participation at the centre than women in comparator countries (see figure 2).
Figure 1. 2 Proportion of seats held by women in national parliament, 1990–2009 Source: UNSTATS UN Millennium Development Goals Indicators (http://mdgs. un. org/unsd/mdg/Default. aspx) The provision of reserved seats has ensured 30 per cent representation of women in local and national government, but few women have won party endorsement to contest general seats.
In the last parliamentary election, 17 women won general seats – by far the highest figure in all parliamentary elections to date (the previous record being 8 in 2001, suggesting an upward trend from an abysmally low starting point (Pandey 2008). Many women candidates are, however, believed to be proxies contesting on behalf of disqualified or jailed male family members. All political parties have delayed introducing direct elections to women’s reserved seats in Parliament, since these seats are valuable patronage resources, particularly during coalition negotiations.
In terms of women’s access to justice and security, the most persistent concern is about the prevalence of violence against women. The most recent DHS data indicate that 49 per cent of ever-married women had experienced some spousal physical violence in their most recent marriage; 18 per cent had experienced rape within marriage; 53 per cent had experienced some form of physical and/or sexual violence, while 13 per cent had experienced both (NIPORT 2009: 201).
While the evidence does not indicate violence against women has increased over time,3 evidence from the WHO (2005) multi-country study and other research 3 Focus group discussions about domestic violence as part of research for the World Bank’s gender survey did not indicate a perceived rise in domestic violence (World Bank 2008a: 92). It seems likely that the efforts of the women’s 11 indicates that Bangladeshi society suffers from a comparatively high prevalence of violence against women (Naved et al. 006). A series of different sources have arrived at similar estimates of the prevalence of domestic violence, including that between 40 per cent and half of all women experience violence from husbands in their life time, with the proportion rising to two-thirds among poorer groups; in addition, just under one-fifth of married women of reproductive age (16–19 per cent) experienced current or recent violence from husbands (cited in Naved and Persson 2010). 1. Changing society in the post-conflict era: dowry, violence and the feminisation of public space The larger social backdrop to the above changes in women’s lives was that of the dislocation to gender norms associated with the war in 1971 and the post-conflict period. This period was catastrophic for many hundreds of thousands of families, so much so that the destruction of old certainties and customs may well have primed gender relations for the rapid social and economic change that followed. This was a period in which the assurances of male protection under the ‘patriarchal bargain’ became shakier. The account of one early entrant into the RMG industry illustrates the situation in which many women found themselves in the post war period: During the [1971] troubles, my husband and father were murdered by the razakars… My son was just over a year old then. We never found their bodies … My mother and sister died of illness.
I had two brothers, one drowned and the other died of illness. All of them are dead … I have no one left. (Kabeer 2000: 102) An unknown number of women were raped during the war, possibly as many as several hundred thousand;5 many more experienced trauma, displacement, and the loss of fathers, brothers, husbands and sons (Kabeer 2000). Women had themselves been prominent within the nationalist struggle – a formative experience for important sections of the women’s movement.
Yet the most significant state response was to rewrite the patriarchal bargain as one between (victimised) women and the state through the ‘birangona’ programme intended to socially rehabilitate the women raped during the war by declaring them war heroines and arranging marriages for them (Kabeer 1991a). As Mookherjee notes ‘the kinship norms of purity and honour were articulated in a public discourse that made [the women who had been raped] the concerns not merely of the family or the community but also of the new nation’ (2008: 40).
Arguably we see in this first action an appropriation of the responsibility to act on behalf of women whose victimhood brought them directly within the jurisdiction of the state. It was soon after that the state began to experiment with social protection for women through the Vulnerable Group Development (VGD) programme; it seems likely that these shifts in the way the state ‘saw’ poor women at this time are related. The VGD was an innovation in its era, for among other reasons it targeted destitute women directly, rather than operating through male household heads (Hossain 2007).
This could be reasonably interpreted as marking official recognition that the patriarchal bargain had broken down en masse for its most vulnerable citizens, and that it was, therefore, mandated to act. movement to draw attention to this as an issue have borne fruit to the extent that it receives more attention and is spoken of as a problem, in contrast to the past. 4 War frequently involves sharp shifts in gender relations and related public policy (see Moghadam 2003). The number of women raped during the Bangladesh liberation war has been the subject of recent controversy. Susan Brownmiller used the 300,000-400,000 estimate in her definitive account, Against our Will. Yet Germaine Greer, herself active in securing abortions for Bangladeshi wartime rape survivors in 1972, stated on television in 2011 that the wartime rapes in Bangladesh were ‘an urban myth’ (she may have been referring to the idea that rape was a Pakistani army policy).
This bizarre volte-face is of course easily discounted by numerous credible sources which suggest that around 25,000 abortions and adoptions are known to have taken place; as these were generally concealed or undertaken in secret, the number of rape survivors is likely to have been far greater, in line with the official lists that were collected and later destroyed. Bina D’Costa and Nayanika Mookherjee have published recent analyses of these events. 12 The rupture of the patriarchal bargain meant many women were in the position of having to support themselves.
The famine and wider economic crisis that followed the war in the mid- 1970s produced growing evidence that women were seeking paid work outside the home ‘in the face of considerable resistance from family and community’ which still assumed a norm of female seclusion (Kabeer 2000: 65). The combination of women without male protection, acute poverty and deprivation and high fertility rates meant the country was ‘a ripe field’ for positioning women at the centre of development as beneficiaries and ‘targets’ (Azim 2001: 392).
Predating the war, longer-term processes of landlessness, impoverishment and new employment opportunities for educated men were also affecting the economic resources of women. These processes drove a reversal of marriage payments from the gifts to the bride’s family (customary till the 1950s) towards payments to the groom’s family – dowry or as is more commonly used, dabi (literally, ‘demand’) (Kabeer 2000). The persistent strength of the norm of universal and early marriage, particularly for women, means that dowry has become a significant burden across sections of society.
This seems to be particularly true among poor households with many daughters. Recent research found that around 53 per cent of rural households and 14 per cent of urban households reported facing dowry demands (Naved and Persson 2010, based on 2001 data). It also seems that dowry demands are a relatively more serious problem among poorer people, with 61 per cent of rural women in the poorest quintile facing demands compared to 37 per cent in the top quintile; comparable urban figures were 20 per cent for the lowest and 7 per cent in the top quintile (ibid. p. 841). This may not mean that richer households do not give and take dowry, but that they do so as ‘gifts’ and can generally afford to do so without serious damage to the household economy. A nationally representative survey by the new National Human Rights Commission of Bangladesh placed dowry at the top of the list of issues reported to be affecting women in 2011 (NHRCB 2011). Dowry or demand is a strikingly common source of discontent and insecurity for the society as a whole, which has to date eluded effective regulation or social change.
Amin argues that the necessity of marriage amidst a ‘powerful perception of insecurity and risk of sexual violation of girls living in households without a male guardian’ (1997: 230) coupled with the demands of dowry leave the poorest women vulnerable to marriages in which there is ‘no obligation on the part of the man to support his wife’ (ibid. ) ‘Dowry violence’, including murders, immolations, suicide and other forms of violence has become categorised as a specific policy problem in the popular media and within official discourse (see references in Government of Bangladesh 2009a: 10–11, 62–3).
Public opinion and public safety surveys now routinely return ‘dowry’ among the most pressing social issues of the day. The 1999 Bangladesh study for the World Bank’s Voices of the Poor study consistently identified dowry as among men and women’s key concerns in relation to poverty and gender relations (Nabi et al. 1999). A 2007 survey of human security found that some 56 per cent of respondents listed ‘dowry’ as among the crimes and insecurities they worried about most, second only to personal property crimes at 77 per cent (Saferworld 2008).
A final aspect of the wider social backdrop against which women’s lives are being lived in 21st century Bangladesh is the feminisation of public space, a much-commented on change exemplified by the visible urban fact of young women garments workers en masse en route to work. Other symbols of this feminisation of public space include the presence of large numbers of young girls going to school and college in rural areas and upazilas and small towns, as well as the by-now familiar figure of women NGO staff, some on motorcycles and bicycles, others walking or using public transport.
Many wear head coverings of various kinds and styles, but by no means all do at all times, highlighting that this remains a matter of choice for women in Bangladesh. BRAC women staff have innovated an outfit which resembles a cross between a lab coat and burkah; this denotes their official and professional 13 status without requiring that they cover their heads. A mobile phone billboard in 2011 features a confident smiling young woman cycling down a sun-dappled country lane – a gulf away from earlier imageries of women.
We will look more at what this claiming of the right to move in public space has meant for gender relations in Bangladesh in relation to garments work more below. It should be noted that the immense cultural significance of women having gained access to the public sphere will be lost on anyone who cannot recall the Bangladesh of the 1980s or earlier. 1. 6 Empowerment revisited It is against this backdrop that the contribution of women’s RMG employment to women’s empowerment is to be assessed.
A recent analysis of the meaning of women’s empowerment among key actors in Bangladesh concluded that the focus was predominantly on individual and economic empowerment, and that this was at the cost of more collective and more political forms of power. In common with other recent analyses of the meaning of women’s empowerment within development globally, the paper concluded that the understanding of women’s empowerment was a neutered idea of power as ‘choice’. In particular, it excluded recognition of the central importance of women’s political and collective empowerment in achieving fuller forms of gender equality. Apart from the Bangladeshi women’s movement, the key actors studied – NGOs, donors, political parties – all to greater or lesser degrees featured a ‘residual instrumentalism’ in their view of the pathways to gender equality in Bangladesh (Nazneen et al. 2011). The perspective from the women’s movement, by contrast, was of a more multi-dimensional understanding of the kind of power women need to gain in order to secure their rights on a lasting basis, and not merely to the extent that this serves the interests of the nation more generally.
This included, in addition to individual women’s agency and economic empowerment, recognition of the need for collective and political empowerment, so as to jointly shift the ‘structures of constraint’ (Folbre 1994; Kabeer 1997) within which all women operate. The paper concluded that inadequate and potentially reversible progress towards gender equality in Bangladesh required tackling the structural political, institutional and cultural changes that mark the ‘second generation’ challenges to Bangladeshi women’s progress.
These have not often come into view as critical determinants of empowerment to date. In an effort to contribute to the widening of the debate about the empowerment of women in Bangladesh, the analysis in this paper proposes to explore the effects of the RMG across the domains of women’s empowerment, looking at not only the individual and economic forms of power that women have gained, but also the more explicitly feminist concerns of collective and political power.
So to a substantial degree, the present paper provides the empirical analysis of the theoretical arguments set out by the author and colleagues in Nazneen et al. 2011). The paper will proceed with its empirical analysis by discussing the effects of RMG employment on women’s citizenship, access to basic education, rights to public space, and political empowerment. 6 For example Batliwala (2007); Cornwall et al. (2008); Eyben and Napier-Moore (2009). 14 2 The RMG sector Some background to the RMG sector is necessary.
A history of the emergence and growth of the sector is available from other sources,7 so we focus here on the contemporary economic, political and social significance of the RMG sector. The history of the RMG sector in Bangladesh resembles that of most low income country expansions into export manufacturing with the global restructuring of the garments trade, enabled by trade liberalisation and low labour costs, with some local variants. Several features of the growth and development of the RMG sector are worth noting, however.
It emerged initially in response to the Multifibre Arrangement (MFA) in 1974, which set quotas on garments exports from the newly industrialising countries of Asia, and encouraged ‘quota-hopping’ as entrepreneurs sought to establish manufacturing sites in quota-free sites. The Korean firm Daewoo was an early entrant in Bangladesh, when it came to an agreement with a Bangladeshi firm, Desh Garments, to which it trained and provided equipment (Kabeer and Mahmud 2004).
Nationally, economic reform processes were also in train from the early 1980s to reverse the socialist and state-directed thrust of economic policy, abandoning import substitution industrial policies in favour of promoting export-led growth. This was most marked under the National Industrial Policy in 1982, which set up incentives and support for foreign direct investment, including through export processing zones (EPZs) (Kabeer and Mahmud 2004; Kochanek 1993). The 1990s saw more incentives for RMG investment, encouraging the growth of more locally-owned firms (Bhattacharya and Rahman 2000).
Early on, domestic entrepreneurs typically drew on their client networks in their home areas to mobilise their initial workforce. There was a paternalistic aspect to some of this, as the factory owner essentially undertook the protection of women workers from these areas (Kabeer 2000). Early on, the strategy of employing women appears to have been aimed in part at circumventing the possibility of labour organisation, as the chief lesson from the state-owned enterprises was that this made it impossible to run a profitable business.
Women were seen as likely to be docile, and this, with the extreme cheapness of their labour, made an emphasis on women’s employment more likely (Kabeer 2000). An idea that tailoring was in some sense customarily or at least appropriately women’s work may have contributed to the social acceptability of what was, in effect, tailoring. But as an explanation of why more conservative rural households accepted girls’ garments work, this has not generally been accepted by leading scholars in the field, who note that tailoring is a traditionally male occupation.
RMG employment had no important opponents and many potential beneficiaries, so that whether or not public policy was overtly supportive, it did not deter women’s employment, even during shifts towards more Islamic versions of the state in the 1980s. Despite employers’ concerns that male workers may be more militant and organised, the more recent emergence of the knitwear industry has been on the basis of a considerably more male workforce (as knitting machines are seen as too heavy and/or too complex for women).
The implications of the changing gender and skills of the export labour force have not been very thoroughly considered to date. It is surprising, for instance, that the changing gender composition of the workforce did not make developing more constructive workerowner- state relationships a higher priority. 7 On the origins and development of the RMG sector, see Bhattacharya and Rahman (1999); Kabeer (2000: 69–76); Khundker (2002); Rock (2003); Siddique (2003); Khatun et al. (2007); Rahman et al. (2009); Morshed (2007). 15 2. 1 The economic significance of the RMG sector
It is difficult to overstate the contemporary significance of the RMG sector in Bangladesh’s economic development. With around 1. 9 million workers directly employed in the sector (Ahmed 2009), it employs only 4 per cent of the total labour force of 51. 8 million, over onethird of the 5. 3 million in manufacturing employment (Rahman, Moazzem and Hossain 2009). Yet some 76 per cent of all export earnings were from apparel in 2008–9 (MoF 2009, chapter 6), and a 2002 estimate was that the RMG sector contributed some 10 per cent of GDP (Bhattacharya, Rahman and Raihan 2002).
Its central significance for the national economy means its continued growth has been a key public policy concern. The industry has faced several challenges, but to date has proven to be remarkably adaptive to its global market environment. The episodes include the US Harkin Bill to prevent use of child labour in the early 1990s, the shock to global trade which hit exports to the US (a major market for Bangladeshi garments) after 9/11 (see Siddique 2003; Ward et al. 2004); and the end of the favourable Multi Fibre Arrangement in 2005, which exposed Bangladesh’s sector to more competition, including from China (Ahmed 2009).
More recently, the sector has recovered well from the global downturn of 2008–10. It did not go untouched: garments exports contracted substantially in the first half of the 2009–10, but this was compared to unprecedentedly rapid growth the previous year. The industry was hit somewhat less and later than export sectors elsewhere, very likely because the Bangladeshi industry benefited (in order volumes) from downward pressure on prices – the so-called ‘Wal- Mart effect’ (Rahman, Bhattacharya et al. 009; CPD 2011). But the idea that the sector’s continued success depends solely on ‘a race to the bottom’, and rests on Bangladesh’s continued capacity to compete amidst downward wage pressures now lacks credibility. One reason is that a process of restructuring is already happening, and there is evidence of improvements in compliance and management practices to raise worker productivity in the modern sections of the industry (Rahman et al. 2007; MoF 2009; CPD 2011).
There are signs that these improved managerial and compliance practices are enabling investments in worker productivity, helping factories cope with the uncertainties of export production in Bangladesh (political and labour unrest, transport, energy and other infrastructural bottlenecks). Other changes afoot include new Rules of Origin that may be opening up new opportunities in the European market and the expansion of the Asian market for Bangladeshi garments.
Changes to the Indian trade regime may also favour RMG exports from Bangladesh, as should the global economic recovery more generally (CPD 2011). 2. 2 The political significance of the RMG sector Economic significance is accompanied by its growing political significance as a wellorganised and resourced and relatively unified interest group. The first generation RMG owners were mainly from the professional middle classes – engineers and civil servants, factory managers and technicians (see Kabeer and Mahmud 2004).
The growth of the RMG and the establishment of the Bangladesh Garments Manufacturers and Exporters Association (the BGMEA) and later the Bangladesh Knitwear Manufacturers and Exporters Association (BKMEA) marked the first real entry of organised business interests (as distinct from more particularistic interests) within the Bangladeshi polity. Business in the region had historically been dominated by non-Bengali Muslims prior to Independence, so that the RMG owners collectively represented the emergence of a distinct indigenous entrepreneurial class (see Sobhan and Sen 1988; Kochanek 1993).
Most notably through the BGMEA, business interests began to organise for political influence through the 1990s (see Kochanek 1996 and 2000). It is also significant that a high proportion of national politicians are drawn from business and industry (57 and 56 per cent in the previous and present parliaments 16 respectively),8 replacing the mainly urban professional middle class politicians from the independence era (see Khan et al. 1996; IGS 2006).
While a figure for the number of RMG owners among the business people in parliament is not available, the prominence of garments manufacturing within the business sector as a whole in Bangladesh suggests that it is likely that a significant proportion has personal interests or close connections with the industry. Persistent infrastructural bottlenecks (problems of electricity, gas, transport, the Chittagong Port) underpinned by problems of governance (grand and petty corruption, slow procurement and investment, labour unrest and political violence) plague the lead times of garments exports. There is potential for BGMEA to act as a lobbying power for the greater good of manufacturing interests and the general public, particularly in relation to energy supply and infrastructural investment. Yet optimism that the relatively powerful BGMEA, with its high degree of insulation from national politics afforded by its position as lead exporter (IGS 2006), might exert positive pressures on economic governance has not to date been realised.
Instead, where the BGMEA has weighed in on governance concerns, it has mainly attempted to insulate the industry from them, rather than forcing a political settlement on improved governance of wider public benefit. An example is the political parties’ concession to exempt garments factories from hartal (the all-out strikes that are the main weapon of the political opposition within Bangladeshi politics) in the 1990s, which in theory should have permitted export production to continue unhindered. 10 The BGMEA has also been successful in attracting special concessions and incentives.
This was illustrated most clearly in the fiscal stimulus response to the global financial crisis as it began to unfold in Bangladesh in 2009. Compared to the primary exporters, the early phase of the crisis saw the RMG sector affected through dropping garment prices globally, but saw order volumes rise with the ‘Wal- Mart effect’. The sector felt excluded from the first fiscal stimulus package of $488 million in April 2009, and lobbied successfully for more support as garments export growth declined in the first half of the 2009–2010 financial year (Rahman et al. 010). But it was not until the third stimulus package that its lobbying bore fruit with a wide-ranging set of incentives to help the industry through the crisis, focused on cash incentives and easing the costs of business (MoF 2009). However, efforts to extract government cash to pay wages and Eid bonuses in 2009 were embarrassingly shot down, with the Government dismissing industry warnings that without help with paying wages and bonuses labour unrest would result, citing statistics showing the sector’s continued profitability.
The BGMEA denied trying to blackmail the government, and as the leading English language newspaper put it, was forced to eat ‘humble stimulus pie’. 11 The limits to the political clout of the BGMEA are becoming increasingly clear, for two reasons. One is that it is widely believed that since the 2000s, the leadership of the sector has become more closely aligned to party interests. This may ensure greater access to the seats of power, but it also compromises its autonomy with respect to public policy. 2 The second may be equally important, and this is the limits on the political influence of the BGMEA from what appears to be widespread popular sympathy with the RMG workers’ 8 Personal communication from Inge Amundsen of Chr. Michelsen Institute. 9 Bhattacharya and Rahman (2000) note that partly because of the need to import raw materials, lead times in RMG manufacture in Bangladesh are some 120–150 days, compared to only 19-45 days in Sri Lanka, and 12 in India. More recent estimates were that lead times for woven garments were in the 90-130 day range (Nuruzzaman and Haque 2009; Haider 2007; World Bank 2005). 0 Although factory owners said that transporting goods remained an issue, and so the exemption was ultimately toothless. But it explains why so many vehicles on Dhaka streets at this time carried signs explaining that they were engaged in ‘urgent export work’, to signal their exclusion from potential political violence. 11 ‘BGMEA eats humble stimulus pie’, 8 September 2009, The Daily Star. 12 This insight arises from interviews conducted with garments factory owners including some active within the BGMEA, as part of a multi-donor country governance assessment in 2010. 7 longstanding struggle for decent (and regular) wages. This is a complicated issue that merits far more sustained attention than it has received to date. Several aspects of the popular perception of this struggle merit attention. The first is that for many within the business elite, wage struggles and rights of labour to organise spell the doom of the industry, as organised labour is seen to have captured key (mainly public sector) industries in the past.
There is a marked tendency among the RMG owner classes to detect the outside hand of economically powerful regional neighbours seeking to sabotage this nationally important sector, with much dark reference to ‘vested quarters’ and outside actors manipulating otherwise compliant workers. Such views are dismissive of the workers’ struggle. In the popular and middle class perspective, however, the gap between the conspicuous consumption of the garments-owning classes and the pitiful pay packets of these young workers has become a glaring inequality.
Sympathetic literary, journalistic and other depictions of Bangladeshi garment workers may have heightened this awareness, although this appears to be more the preserve of elite and English-language culture than mass or popular culture (see also Kabeer and Mahmud 2004). 13 That RMG sector employment has meant a reduction in the supply of women’s and girls’ domestic labour has caused considerable hardship to middle class households, and in particular to working women; this is very likely to moderate their sympathies and support. 4 Yet while there is no middle class support for worker violence, there is certainly widespread awareness within this group of the rising cost of living since the mid-2000s and so some recognition of the legitimacy of their struggle (if not their methods). As food prices spiked in 2008 and 2011 the issue of low wages has remained a concern with which all urban people can sympathise (see Hossain et al. 2009; Hossain et al. 2010). Governments have responded in a series of ways since 2006. Efforts by the state to put down protests have achieved little.
The introduction of the Industrial Police is part of the effort to prevent unrest, although workers generally report that labour organisation more generally is repressed. The priorities of the state with respect to garment workers is highlighted by this innovation, which was not intended to police worker behaviour, rather than to protect workers against the harassment and abuse they routinely face in and en route to work (Siddiqi 2003). The present Awami League Government has to date been more responsive to the garment workers’ demands than some previous regimes, and the minimum wage was fixed at Tk 3000 in 2009.
Yet the struggle continues in some sections of the growing RMG labour movement, with claims for a Tk 5000 minimum wage now being argued for. While the present campaign dates at least to the May 2006 incidents which involved violence, arson and criminal damage, violent actions continue within the sector, reinforcing the need for more regular channels of labour-industry dialogue and policy-making processes. 3 Women as RMG workers We turn now to women’s experiences as RMG workers. This section will explore two apparently contradictory issues about women’s garment work experience.
The first is the tough and exploitative nature of garments work; such work has always been tough – physically demanding and featuring unsafe conditions of fire risks, sexual harassment and physical and verbal abuse. By virtue of their gender, women typically enter the industry on terms of comparative disadvantage in terms of pay, conditions and promotion prospects. Irregular wage, overtime and bonus payments have long been and remain one of the most 13 For example, Monica Ali’s somewhat controversial Brick Lane, itself inspired by Naila Kabeer’s The Power to Choose (2000).
Sohela Nazneen reports that a search for popular representations of women garments workers in songs, films etc returned relatively few examples, suggesting this is a glaring gap in popular cultural representations of the heroines of the country’s export industry! 14 Many thanks to Sohela Nazneen for this point. 18 significant problems workers face in the industry. All of this has meant that until recently, few women garments workers stayed longer than five years, with their garments career usually overlapping with their pre- and early marriage lives.
Most women find it difficult to balance care work with garments employment, which involves long working days and unpredictable over-time; facilities for childcare are rare, as is return to work within the same factory after maternity. Yet the hard nature of garments work appears to be contradicted by the second issue about women’s RMG employment which is that despite its hardship, women and girls who have worked in the sector have experienced a range of forms of social and economic empowerment as a result of that work.
The important issue here is that these findings highlight the relative advantages afforded by RMG employment within the context of women’s limited labour market opportunities in Bangladesh. 3. 1 Hard work Women make up the majority of workers in the RMG sector as a whole, with most recent estimates suggesting that up to 1. 7 million women are employed in the sector at present (Ahmed 2009). 15 The gender composition of the apparels sector workforce has changed in recent years, with the growth of the knitwear industry which employs more men.
That knitwear manufacturing is seen as ‘men’s work’ has been linked to the belief that women are less capable of the physical and skill demands of the machinery used in knitwear and sweater production. Women and girl workers are concentrated in the woven RMG sector, in which the gendered pattern of the occupation structure is that senior management, supervisors and occupations seen as skilled technicians (e. g. , cutters, finishers) tend to be men; most machine operators and helpers are women and girls.
Siddiqi summarises the gendered hierarchy of production: Production is divided into three main tasks – cutting, sewing and finishing. A production manager (PM), almost always male, is in charge of the entire production process from cutting to shipment, and reports only to the managing director. The cutting section is supervised by a male master cutter (cutting master) under whom other cutters and helpers, all male, work. The responsibility of overseeing the entire sewing section is in the hands of the floor-in-charge, as he or she is called.
The floor-in-charge of sewing is preferably a woman, ‘so that the workers will feel comfortable bringing their problems to her’. Almost three quarters of the line supervisors in the sewing section are men … The sewing machine operators and helpers are predominantly female, around 80 per cent. In the finishing section, the floor-in-charge is always male, as are line supervisors and those who do the ironing. Folding is done by men and women. Male workers do the packaging and cartoning (2000: L-13).
The force of gender segmentation within the RMG industry means that women who do jobs that are seen as more skilled – and therefore as ‘men’s’ jobs – can be ashamed of their ‘unfeminine work’ (Kabeer 1991b). Most recent survey data suggests that occupation segregation is less acute in the more ‘modern’ EPZs than in the non-EPZ factories (see Table 3). 15 It may be a feature of the informality of the industry that reliable up-to-date worker data is hard to come by for the RMG sector. Estimates of the proportion of women workers in the sector vary widely, from 66 (Paul-Majumder and Begum 2000) to 90 percent (Ahmed 2009).
Given the growing importance of the knitwear industry and its predominantly male workforce within the RMG sector as a whole, the lower figure is likely to be closer to the truth in the present day. 19 Table 3. 1 Gender division of RMG jobs (% of workers) Job EPZ Non-EPZ Men Women Men Women Operator 70 54. 5 54. 5 77. 3 Helper 20 27. 3 - 4. 5 Packer - - - 6. 8 Other 10 18. 2 45. 5 11. 4 All 100 100 100 100 Source: CPD/GATE survey, in Khatun et al. (2007). Gender segregation within the RMG occupation hierarchy is reflected in gender wage differentials.
The CPD/GATE survey found that wage differentials remained significant (indeed, they may have increased over time) in the RMG industry, with women machinists or operators earning only 71 per cent of the earnings of men operators; female helpers earned a mere 53 per cent of male helpers’ earnings in 2005. Only in ‘other’ categories of work – ironing, folding, cleaning and packing – were wage differentials slightly in favour of women (103 per cent) (Khatun et al. 2007: 42). Yet promotion prospects were generally dim for most women workers (Chaudhuri Zohir and Paul-Majumder 1996).
Analysis indicates that simple discrimination explained part of the wage differential and that the wage gap may have increased over time (see also Paul-Majumder and Begum 2000; 2006; Paul-Majumder 2003). This is despite the fact that skill differentials tend to explain more of the gender wage differentials in the RMG than in other sectors (Ahmed and Maitra 2010; Kapsos 2008; Uddin 2008). However, despite strong evidence that simple discrimination affects wage differentials, business leaders, policymakers and trade unionists persist in the view that skill differentials entirely explain women’s lower pay (see CPD 2007).
Garments work has not been a life-time career for most women who have worked in it. Women tend to be young when they work in the garments, and in the 1990s lasted an average of four years (Paul-Majumder 1996). However, there are some signs of change. The average age of women garment workers appears to have risen since the 1990s, from 19 in 1990 to 25 in 2006 (Khatun et al. 2007). The proportion of married women also increased, from 38 per cent in 1990 and 1997, to 59 per cent in the 2006 CPD/GATE survey (Khatun et al. 2007).
This suggests more enduring factory careers. Yet the overall pattern remains of a very young female workforce (60 per cent under 25 in the CPD/GATE survey (Khatun et al. 2007; see also CPD 2007). It is also the case that factory work has not provided stable, respectable industrial jobs, although EPZ employment comes closer to that ideal than employment in other kinds of factories. Export sector factory work under the New International Division of Labour needs to be situated on the continuum between formal and informal work (Kabeer and Mahmud 2004b).
The labour contract is certainly an informalised and fluid arrangement between factories and workers, enabled in part by the absence of recruitment costs in a context of no training and a steady supply of adequately schooled and ‘unencumbered’ labour. Demands for identification cards and contract letters and other formalising aspects commonly feature among the demands of garment workers, who appear to see these as tools with which to formalise their unevenly honoured claims to wages, overtime and benefits.
Other reasons these are not good working class jobs include issues of worker safety. Factory fires occur regularly, despite the provision of fire escapes and safety equipment having been made mandatory and among buyers’ compliance requirements. Many young workers have died because of being locked inside to prevent theft of materials or to meet deadlines. In a 20 particularly tragic recent event, a large, established factory caught fire in December 2010, and 26 people died – 23 from jumping out of the 11th floor windows – and around 100 people were injured. 6 In one estimate, some 60 per cent of garment factories continue to lack fire safety equipment,17 despite extensive and growing pressure for factories to comply with social and labour requirements. Garments work is also so physically demanding that few people can last more than a few years. Chaudhuri Zohir and Paul-Majumder found a higher incidence of illness among women workers than among men. These ranged from eye and head pains, to respiratory and gastric conditions and urine infections (1996). The high turnover rate of workers was attributed partly to ‘sustained exhaustion from work’ (pp. 01), a finding that other research has corroborated. An important but typically neglected aspect of the lack of safety at work is that factories and routes to work can be significant sites of sexual harassment and other forms of abuse, consistent with the fact that paid work outside the house remains an anomaly albeit increasingly common. Verbal and other forms of abuse within factories were found to negatively affect the productivity of almost half of surveyed workers, and practices appear to be widespread (see Siddiqi 2003).
Yet as table 4 suggests, sexual harassment and others sources of threats to personal security are actually more common during travel to and from than in the workplace, a
