Hii Dunia

Written by David Parduhn

Introduction

In the year 2000, the United Nations member states identified ‘women’s empowerment’ as one of the most important development goals. Since then, ‘gender equality and women’s empowerment’ was codified as the third out of eight United Nations’ Millennium Development Goals (MDGs). However, ‘neither the World Bank nor any other major development agency has developed a rigorous method for measuring and tracking changes in levels of empowerment’ (Malhotra et al. 2002: 3); it still means different things to different people. The three
indicators which are used by the United Nations to monitor progress in achieving this goal are about women’s share in education, employment and political participation (UNDP 2010). However, cross-cultural and ethnographic research has shown that, while each of the three aspects certainly holds the potential for positive changes with regard to women’s empowerment, particular cultural circumstances and social relationships influence whether and how this potential can be realised (Kabeer 2005: 13).

The present paper will closely examine the second indicator for the third MDG, namely the idea of ‘increasing women’s share of wage employment in the non-agricultural sector [NAS]’ (UNDP 2010). According to this indicator, wage employment opportunities in the NAS are seen as an essential resource to empower women. Female economic activity in general, without the qualification ‘in the NAS’, is also one of the numbers used for the construction of the Human Development Index (HDI) through the Gender Empowerment Measure and the Gender Inequality Index, which both measure the inequalities between men and women with regard to labour force participation. This suggests that the sole act of taking up paid employment by women is seen as a step towards empowerment (Kabeer 2005: 15).

Does the presence of women in paid employment really indicate female empowerment? One link between wage employment and empowerment consists of control over resources: The assumption is that women’s economic position will be enhanced by taking up waged employment, which will in turn result in their empowerment (Horton 1996, cit. in Elson 1999: 614; cf. Kapadia 2010: 277). Beyond doubt, the presence of women in the workforce can be important for improving their material conditions, thus increasing their independence. However, the present paper will put the prevalent underlying assumption into perspective and demonstrate that ‘putting women to work’ does not necessarily empower them.

In order to do this, this paper will first examine the concept of empowerment in the context of gender relations. Thereafter, it will outline the anthropological discussion around intra-household dynamics with a special focus on the gendered division of labour. In the second part, the paper will draw on case studies from Kenya and Bangladesh in order to demonstrate in the subsequent analysis that the inclusion of women in the cash economy does not necessarily entail a transformation towards empowerment as any change is mediated through power relations, particularly those which unfold within the household. Finally, on the basis of these findings the concept of women’s empowerment itself and its usefulness will be critically reassessed.

II. Measuring and monitoring an ill-defined concept

Strikingly, the term ‘empowerment’ itself is not defined in the MDGs, which clearly shows the unquestioned nature of this concept. Notwithstanding, ‘empowerment’ is one of the buzzwords for policymakers and practitioners in development and, as mentioned above, one of the frequently cited goals of development interventions (Mosedale 2005: 243). Sarah Mosedale notes that ‘empowerment’ is frequently ‘used to add glamour (rather than value) to interventions which actually seek to achieve a variety of economic and social outcomes’ (Mosedale 2005: 252).

The absence of a clear definition complicates an analysis of the achievement of the goal itself. In order to have a basis for discussion, it is necessary to clarify what is actually meant by ‘empowerment’ in the present paper: Drawing on the extensive discussion of the term and its implications by Kabeer (1999, 2005), the
paper defines ‘empowerment’ as the ability to make choices or rather the ‘process by which those who have been denied the ability to make choices acquire such an ability’ (Kabeer 2005: 13). Thus, empowerment entails change, since people can be very powerful, but may have never been disempowered in the first place (Kabeer 2005: 14). Central to the ability to decide is the existence of ‘alternatives without punishingly high costs’ (Kabeer 1999: 460), such that there is a real possibility to have chosen otherwise (Kabeer 1999: 441). In this context, the term ‘agency’ is often used to describe one’s ‘ability to define one’s goals and act upon them’ (Mosedale 2005: 249), including acts of resistance, ‘bargaining and negotiation, deception and
manipulation’ (Kabeer 1999: 438; cf. Agarwal 1997: 25).

Gendered power relations

Several political, legal, economic and cultural circumstances influenced by social categories such as class, caste, ethnicity, race or age, amongst others, affect an individual’s ability to make choices. One of the most defining social categories which influence women’s possibilities and decisions is certainly the crosscutting category of gender. Gender-related practices and norms, rights and responsibilities define the appropriate behaviour and position of women in society as well as their access to resources and control over them (Francis 1998: 75; cf. Mosedale 2005:244).

As one is always empowered relative to others or to oneself at a previous time, empowerment is an ongoing process rather than an absolute product (Mosedale 2005: 252). Interestingly, the standard which is applied is usually a male one: Women’s level of empowerment is judged by ‘what is possible for [them] to be and to do compared to men’ (Mosedale 2005: 252).4 As a result, the indicators used to measure empowerment show that empowerment is often conflated with gender equality and reduced to quantitative parity. In this regard, Mosedale (2005: 243) criticises that fact that the term is intuitively related to women, who have to be empowered in relation to already-empowered men.

Power relations are not always noticed because, as Kabeer points out, they ‘are most effective when they re not perceived as such. Gender often operates through the unquestioned acceptance of power’ (2003: 169). Processes of identification, that is the interplay between social ascriptions and self-perceptions, can circumvent and deny the cognition of gendered power inequalities or even justify (cf. Kabeer 2005: 14), internalise and naturalise such inequalities. Gendered identities, that is roles and modes of appropriate behaviour for men and women, are developed throughout life and perceived as natural or inherent (Mosedale 2005:249). Therefore, they cannot easily be shaken off, though, as Agarwal points out, ‘social norms are not immutable, and are themselves subject to bargaining and change, even if the time horizon for changing some types of norms may be a long one’ (1997: 19).

Furthermore, even if inequalities are perceived as such, less powerful groups do challenge them only if it does not carry ‘substantial personal and social costs’ such as social exclusion or domestic violence (Kabeer 2005:14). Accordingly, one is likely to be disempowered if more powerful people can override one’s ability to access resources, to define priorities and to enforce claims (Kabeer 2005: 14). According to this, women are not empowered as long as power relations8 shape their choices, opportunities or well-being, thus preventing them from making ‘decisions on matters which are important in their lives’ (Mosedale 2005: 244).

Gendered power relations are constructed and maintained differently in different locations and vary over time (Mosedale 2005: 244). However the household is the central site where they unfold and where identities are negotiated and transmitted from one generation to the next. For this reason, the following section will outline the discussion about households and intra-household dynamics.

III. Inside the Black Box: Dynamics within the household

The definition of ‘household’ has been debated in anthropological circles for decades: It became clear that there is no cross-culturally shared concept and that the nuclear family is not always the basis for a household (Moser 1993: 15); rather, different forms of kinship, residence, reproduction, consumption and others can serve as defining commonalities (Kabeer 1994: 113). Another important characteristic was added by Caroline Moser: She argued that households are not static, but flexible with shifting boundaries (1993: 19; cf. Kabeer 1994: 113). This is especially so in times of migration and transnationalism, when members are not linked to a household through proximity but through remittances. One can argue that what constitutes a household is very dynamic and changes over time. This complexity shows that a universal definition of the household is impossible due to cross-cultural variations. However, as a category of analysis, a household is usually constructed as a close network of social relations between individuals who share and exchange material or immaterial resources and depend on each other for productive and reproductive work.

For many years, macroeconomic development discourse treated a household ‘as a single decision-making unit with a joint welfare function’ (Francis 1998: 73). It was assumed that individuals within one household pool their resources and share the same preferences (Quisumbing 2003: 1). However, since the early 1990s, researchers have started to pay attention to interactions within households, as empirical research has shown that the so called unitary model is inappropriate in many settings in both industrialised and developing countries (see Haddad et al.1997). In the same way, the advanced conceptualisation of the household as operating on the basis of generosity or altruistic decisions of a ‘benevolent dictator’ (Becker 1974), the so called common preference model or New Home Economics (Katz 1991), does not do justice to realities in many places (Quisumbing 2003: 1). However, these approaches were common amongst economists for a long time (Francis 1998: 74).

But what if the individual members within one household do have different, conflicting preferences and the power to improve their own welfare at the expense of others? What happens if one person has the power to allocate material goods such as money or food within the household without consent? These questions indicate that the above mentioned approaches, by treating households as a ‘black box’ (Kabeer 1994: 98), ignored or obscured intra-household dynamics and thus disguised power inequalities which unfold within a household, such as resource allocation (Quisumbing & Maluccio 2003: xi). Indeed, ‘many of the allocational activities [...] are organized through “intimitate” relations of marriage, parenthood and kinship’ within the household, as Kabeer (1994: 96) notes.

Sen (1990) argued that resources are allocated through ‘bargaining in a process of co-operative conflict’ (Francis 1998: 72). In several studies in all parts of the world, the economist Quisumbing (2003) found that such co-operative conflicts affect the outcome of development policies as all decisions which are taken within households or families can affect the well-being of its individual members (cf.Moser 1993: 94). Factors such as household composition, property rights and gender ideologies determine power relations (Francis 1998: 93). Thus, the assumption that members of a household have equal control over resources also neglects the particular position of women (Moser 1993: 15). Critics of the neoclassical unitary model therefore have advocated a focus on bargaining and conflicts, instead of assuming generosity or harmony (see Folbre 1986; Kabeer 1994). According to Folbre (1986: 251), such so called bargaining models or collective models are more appropriate, due to the fact that they have, in contrast to unitary models, at least ‘the potential to offer a truthful understanding of household relations’ (Francis 1998: 75). They accommodate the idea of gender asymmetry and recognise the potential for conflicts, without ruling out altruism (Kabeer 1994: 111).

The role of women within households is shaped by the gendered division of labour: In general, they are associated with biological reproduction and so called ‘reproductive work’, namely to care for and to maintain the current and the future workforce, that is male household members and children (Rosaldo 1974; Moser 1993: 29). Additionally, women often account for secondary paid work, often within the informal sector (Moser 1993: 27). Both types of work are most often invisible, neither valued as ‘real work’, nor captured by statistics (Moser 1993: 30; Kabeer 2003: 27). Conversely, men are often regarded as the primary ‘breadwinner’ by both researchers and themselves (Moser 1993: 28), being engaged in so called ‘productive work’ which is generating income in cash or kind (Moser 1993: 31).

Furthermore, women tend to occupy a subordinate social position in society in general: Despite many differences between cultures and even within one culture or locality in the social value placed on women (Quisumbing & Maluccio 2003: xi), scholars have recognised that discrimination and the subordinate position of women is a recurring norm within developing countries (Sen 1990; Kabeer 1994, 1996: 20, 2003: 81; Francis 1998: 76-77). The control over income by men on the one hand and the secondary status of women on the other hand account for women’s poorer breakdown or fallback position, that is the prospective positions in which they would be if the household did not persist (Francis 1998: 73). This position in turn results in a lower bargaining-position, that is the ability to influence intra-household negotiations and decisions in their interest.

Thus, advocates of women’s employment for empowerment argue that women’s access to income through wage employment would improve their bargaining-position within the household as increased economic independence would lead to a better breakdown position. In order to assess this claim, the next part will look closely into case studies from Kenya and Bangladesh, trying to unpack power relations which unfold within the household and shape the lived realities of women.

IV. Micro-level evidence: Working women and households

Many studies have attempted to trace the positive impacts of wage employment for women in developing countries and their families (e.g. Dolan & Sorby 2003). Research has indicated that women gained greater independence and increased their power in intra-household decision-making through becoming part of the labour force. They were furthermore able to widen their social networks and sometimes to escape abusive marriages (Kabeer 2005: 18-20). However, the link between women’s employment and their empowerment, that is greater control over their lives, is not as straightforward as it seems (Kabeer 2005: 20), as the following examples will demonstrate.

Cash Crop Production by Kenyan Smallholders

Francis (1998) evaluated seven studies about changing rural livelihoods in Kenya with regard to the impact on gender relations (see Francis 1998: 72). On the basis of Francis’ analysis of Davison (1988) and Mackenzie (1993), the following part will show how power relations within the household shaped female livelihoods in the Kenyan Central and the Rift Valley Province and answer the question whether changes in household livelihoods give rise to changes in gender relations.

The Kenyan household is characterised by a patriarchal model in which women were responsible for subsistence food production but subordinated in decision-making processes (Francis 1998: 89). Since the 1960s, smallholder production of coffee, tea, sugar and dairy products did expand rapidly in Central Kenya, not only because of its proximity to the labour and product market in Nairobi, but also due to the ‘encouragement of state agencies and transnational companies, which controlled marketing and provided infrastructure’ (Francis 1998: 78). With the spread of this agricultural development, women’s tasks shifted away from domestic food production to commercial farming (Francis 1998: 72-78).

Mackenzie (1993) found that even in regions which are characterised by a high rate of rural out-migration of men (over sixty per cent), women had only limited access to the revenues of commercial farming. Despite of the women’s greater responsibility for commercial production, crop payments by the district farmers’ cooperative union are directly made to shareholders, of whom the large majority are men (Francis 1998: 79).

These findings are also supported by Francis’ own research (1998: 84): Contractors pay the income from crops to the male head of household, regardless of the respective labour input of wife and husband (Francis 1998: 84). Though Francis (1998: 77, 91) found substantial differences in household composition and gender ideologies within one country and even within a single locality, all studies have demonstrated that women’s actual contribution to the production of cash crops does not necessarily give them access to the income derived from it (Francis 1998: 80, 91).

While commercial farming has increased men’s dependency on their wives for labour input (Francis 1998: 91), and women effectively generated a considerable income through their work, it seems that their bargaining position within the household did not improve and they did not have greater control over their lives (Francis 1998: 91-92). Their breakdown position was still weak as they were dependent on men to gain access to land and had few alternative prospects in the labour market (Francis 1998: 92).

Export-oriented garment industry in Bangladesh

In contrast to rural Kenya, in Bangladesh and more widely in Southeast Asia and Latin America, women entered the labour market not in agricultural cash crop production but in export-oriented manufacturing, particularly in Free Trade or Export Processing Zones (FTZ/EPZ) established in the context of neo-liberal economic reforms and Structural Adjustment Policies (SAP) since the late 70s (Kabeer 2005: 18). The establishment of EPZs in many developing countries led to a rapid increase in labour migration, particularly of young single women from rural to urban areas (Kabeer 2005: 18). In Bangladesh trade liberalisation policies adopted in the early 80s have also to the emergence of a number of export-oriented industries and a ‘feminisation of the labour force’ (Kabeer 2005: 18, cf. Elson 1991).

It is important to recognise that many regions in Bangladesh are characterised by strict patriarchal authority structures within the family, as well as the practice of female seclusion, leading to extremely low female labour force participation rates (Kabeer & Mahmud 2004: 94). The expansion of exportoriented industries has changed the face of female employment in the country: The export-oriented ready-made garment industry, which expanded from a handful of factories in the late 1970s to over 3500 by the mid-1990s (Kabeer & Mahmud 2004: 95), provides employment for about 1.5 million women alone (out of 1.8 million workers in total) (BGMEA, cit. in: Kabeer & Mahmud 2004: 95),most of whom had migrated from rural areas (Kabeer & Mahmud 2004: 105).

Unlike developments in Kenya, women had access to their income as they were paid directly. Furthermore, the fact that some of them were single or migrated ‘away from patriarchal controls of kinship and community’ (Kabeer 2005: 19) enabled them to maintain more control over their income.

However, most of the migrants were accompanied by their husbands or other male household members. In these cases, ‘they had no control about their earnings – they simply handed over their income to other family members’ (Elson 1999: 615); sometimes as a result of pressure or even violence by men (cf. Kabeer 1998). If their income was not appropriated, it was often spent directly on areas seen as a woman’s responsibility, such as child care (cf. Francis 1998: 75; Whitehead 1981). Furthermore, women had to face a so called double burden (see Elson 1991): Despite full-time wage employment, they continued to fulfil all domestic chores and were responsible for reproductive work. In some cases, women even had to withdraw their daughters from school as their help was needed in the household. Thus, in spite of the additional workload for women, the gendered division of labour was rarely re-negotiated (Kabeer 2005: 20).

These findings suggest that the bargaining position does not inevitably improve as a result of access to economic resources. This could be attributed to social and cultural norms related to gender which are more decisive in determining women’s fallback position than material independence: In patriarchal societies, living as a single woman is not an option. Furthermore, irrespective of their economic position, women depend on their husbands in a variety of ways.

Summary: Employment without empowerment?

These cases show that, while it could be argued that at least for young single migrant women, factory-work entails empowerment through access to income and even more so through the fact of migration, which puts them out of the reach of patriarchal structures limiting their decision-making (Kabeer 2005: 19), this relationship can not easily be transferred to all women workers. Certainly, it is not possible to generalize from the findings in Kenya and Bangladesh, ‘but they provide sufficient evidence to undermine the simple assumption that earning a wage will automatically empower women’ (Francis 1999: 615; cf. Kabeer 2005: 13).

First, due to norms and customary or even legal frameworks in society, waged work does not necessarily lead to women having access to and control over the income their work generates. The evidence suggests that this is particularly prevalent in small-scale family production, where work relationships and terms of employment are less formalised. Second, notwithstanding the access to income, it does not necessarily translate into more bargaining power, as is evident in the fact that women did not renegotiate the responsibility for domestic tasks and had to work a double shift. The reliance on other female household members such as daughters for assistance reinforces and reproduces a gendered division of labour. Under these conditions, instead of leading to women’s empowerment, gender inequalities within the household are even intensified as a consequence of female wage employment (Kabeer 2005: 20).

Furthermore, it is questionable whether the change from the position of rural wife to factory worker is indeed a form of empowerment as this form of employment can be characterised by highly exploitative and precarious working conditions (Kabeer 2005: 20): Thus, Goldman argues that

independence, emancipation and equality will continue to be illusory if the
narrowness and the lack of freedom in the home is exchanged for the
narrowness and lack of freedom in the factory, sweatshop, department store or
office. (1970: 14, cit. in Stolcke 1981: 45)

V. Re-evaluating the concept of empowerment

Embodiment of inequalities

The case studies presented above raise questions which are related to wider debates within the social sciences: Sociologists like Bourdieu (1977) have argued that in the process of growing up, human beings internalise social and cultural practices and norms, which then seem to be natural and inherent to them (cf. Mosedale 2005: 249). This embodiment of social norms leads to women internalising a gendered habitus and unconsciously behaving accordingly, that is, in ways that reproduce and reinforce their subordination. Sen described women therefore as ‘implicit accomplice’ (1990: 126; see also Sen 1987; cf. Spivak 1988; Bourdieu 2002). This naturalised self-perception can circumvent the recognition of gendered power inequalities, rendering these power relations even more effective (Kabeer 2003: 169, cf. Kabeer 2005: 14).

In the cases described above, this embodiment of social norms related to gender could explain why women did not challenge men’s dominant position: Francis found for the Central Region of Kenya that women’s ‘sense of what are legitimate distributions within the household affect the positions they adopt’ (1998:73). Although husbands did not pay their wives reliably or commensurate with their labour input, Mackenzie (1993) maintains that most of the women were satisfied with these arrangements (Francis 1998: 79). A similar argument is presented by Kabeer (1998) about South Asia, where women consider the inequitable distribution of goods and the inequitable decisions about consumption to be legitimate.

All this points to the fact that economic resources are not always sufficient in effectively changing women’s position in the household, as they see no need to challenge existing relations within the household which they perceive as legitimate. Without the individual or collective ability to recognize their subordinate status and to utilize resources in their own interests, resources cannot bring about empowerment (cf. Kabeer 2005: 16). This also calls into question the assumption that an individual’s bargaining position can be derived from his or her fallback position, as this implies that people always act in a rational way, that is by consciously and constantly weighing up the costs and benefits of a certain decision.

Empowered to choose what?

More generally, if empowerment is about the ability to make decisions in one’s own interest, it needs to be questioned how these decisions are identified from the outside perspective. It has been taken for granted that if women had more bargaining power, the inequitable allocation of work and resources within the household would have changed. But could it not be that women themselves decide consciously and without constraints to act in ways which reproduce gender norms such as confinement to the domestic sphere?

Many Western researchers tend to assume that everybody wishes to be individualistic, independent and self-determined and acts accordingly, whenever one has the chance to do so. However, this idea is Eurocentric as it takes Western values to be universal and claims a ‘cognitive authority’ over other people (Nussbaum 2000). Mosedale argues that ‘[w]hen we use universal norms as benchmarks for societies, we are telling people what is good for them and showing too little respect for people’s freedom as agents’ (2005: 246). In this context, Mohanty (1984) maintains that it is crucial to discard the assumption that Western understandings are universal, such as seeing the veil as a symbol of domination, or the gendered division of labour as necessarily entailing a devaluation of women’s work (cf. Nussbaum 2000: 40). On the contrary, Nussbaum pointed out that ‘[f]emale modesty, deference, obedience and self-sacrifice have defined women’s lives for centuries and should not be assumed to be incapable of constructing good and flourishing lives for women’ (2000: 41). Thus, as women’s empowerment is supposed to enable women to make their own decisions, this should imply the possibility to choose a ‘traditional’ life or make decisions which, seen from a Western standpoint, seem to contradict their own interest (Mosedale 2005: 246).

VI. Conclusion

In the context of development projects, it has often been assumed that women’s employment would lead, as a result of access to financial resources, to their empowerment, which is conceptualised as the ability to make choices. However, the evidence from Kenya and Bangladesh has exemplified that this link is not as straightforward as is believed. On the one hand, due to intra-household dynamics, women do not always have direct access to the wage they earn as it is appropriated by or handed over to men. On the other hand, even when they have control over their income, it is not necessarily translated into more bargaining power and a subsequent transformation of their status within the household. ‘Money does not always talk’ (Francis 1998: 75), and, as Batliwala pointed out, rich women ‘suffer domestic abuse and rape too’ (1994: 132, cit. in Mosedale 2005: 248). Furthermore, employment often entails negative consequences such as bad effects on health and exploitative working conditions (Kabeer 2005: 20), and results in a double burden (Elson 1991). The transformatory potential of employment for empowerment should therefore not be overestimated.

The thoughts elaborated in the previous part, however, point to fundamental difficulties in measuring empowerment. It has been demonstrated that often universal criteria modelled on normative Western ideas about empowerment, emancipation and modernity are applied in order to determine whether women exercise agency and make their own decisions. Instead, careful empirical analysis of the complex, highly varying and even contradictory material realities of particular women’s lives in particular locations is needed (cf. Quisumbing 2003: 2). Particularly, long-term ethnographic studies are crucial to track transformations over time.

Instead of constructing and essentialising a homogeneous Other, as the nowadays critically contested category of ‘Third World Women’ (Mosedale 2005: 245, cf. Mohanty 1984), Cornwall (2003) points out the importance of remaining agnostic over the respective value of gender and being open to diverse forms of different identities and categories of everyday life. Rather than transferring conventional gender assumptions which inform practice, researchers have to be open and flexible with regard to assumptions about women’s needs. It remains crucial to focus on local knowledge, concepts and preferences instead of taking Western values as universals, and to pay attention to women’s own presentation of their lives and to accept their priorities in order to construct locally relevant indicators for empowerment (cf. Mosedale 2005: 245; cf. Chambers 1995). However, despite the importance of the recognition of subjectivity, it is crucial not to overlook structural constrains placed on women’s decision-making capabilities and not to replace research by subjective statements.

The same holds true for policy-making and development interventions: Without appreciating the local circumstances, they are likely to fail (Mosedale 2005: 246). Empowerment must be claimed from within, as Mosedale (2005: 244; cf. Kabeer 1994: 223) argues. Therefore, while it is questionable whether outsiders should attempt to transform existing gender roles and relations at all (Moser 1993: 74), the inclusion of the concerned women themselves into the whole policy making process may do more to improve their position.

For Appendices and Bibliography please see PDF download

‘Money does not always talk’ - Reassessing the Empowerment Potential of Women’s Employment (306)

Tags: Economics, Gender

This entry was posted on Wednesday, June 1st, 2011 at 12:39 pm and is filed under Article, Economics, Featured Articles, Gender, Human Rights. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.