INTRODUCTION*

^* The author would like to thank Yusuf Bangura, Solon Barraclough, Barbara Harriss-White, Carol Miller, Ruhi Saith and S. Sudha for their useful comments on an earlier draft of this paper. Responsibility for the content of the paper, however, lies with the author.

Although there are now relatively wide-ranging bodies of literature on gender and on poverty, it is arguable that the interlinkages between the two have not been adequately analysed.¹ At one level, the relationship between gender disadvantage and poverty appears to be quite straightforward, as in the tendency to equate women, or female-headed households, with the vulnerable or the poor. Such a perception underpins development agency arguments about the "feminization of poverty", and the frequent references to women as "the poorest of the poor". In these policy discourses three strands of thinking stand out: first, the equation of female headship with poverty (Jazairy et al., 1992); second, assumptions about anti-female discrimination within the household and consequent female disadvantage in well-being (UNDP, 1995); and third, synergistic arguments positing a positive relationship between "investing in women" and meeting developmental objectives such as poverty reduction or fertility decline (World Bank, 1994). But at a deeper level, the gender analysis of poverty is not so much about whether women suffer more from poverty than men (in numbers and/or in intensity), but rather about how gender differentiates the social processes leading to poverty. Critics argue that in the process of assimilating gender into their policy discourse, development institutions have screened out the differentiated and highly contextualized understanding of poverty which has been central to gender analysis. As Jackson puts it, "... 'feminization of poverty' has come to mean not (as gender analysis would suggest) that poverty is a gendered experience, but that the poor are mostly women" (1996:491).

¹ There is, however, an emerging analytical literature on the gender/poverty interface. See, for example, Jackson, 1996; Kabeer, 1996; IDS, 1997.

The question that arises with some urgency is: Why does gender appear in aid agency discussions of poverty in such generalized, and problematic, forms? One of the arguments of this paper is that there are both methodological and political reasons for this which need to be clearly understood because of their implications for future poverty measurement and analysis, as well as for policy formulation aimed at gender-sensitive poverty reduction. At the same time some new ways of problematizing the gender/poverty interface need to be elaborated which can illuminate the ways in which both the trajectories leading to poverty, and the escape from destitution, are gendered phenomena. A gendered understanding of poverty also raises some difficult questions about whether it can be assumed, as is often done, that the kinds of asset interventions that can strengthen the position of poor men are going to have much the same impact on poor women.

To take this project forward, the present paper - which draws on the existing gender and development literature as well as a number of papers commissioned by UNRISD² - makes contributions at two inter-related levels. First, it provides a critical and selective assessment of the attempts to measure gender disadvantage through a wide range of indicators of well-being. It considers how measurable, affordable and reliable these indicators are in identifying gender bias, and suggests that some indicators may be more sensitive in picking up gender bias than others. It also asks questions about what these indicators mean in different social contexts, what kinds of contextual information would be needed to facilitate their interpretation, and whether they can reveal the causal processes leading to poverty and gender bias.³

² Full references for the papers commissioned by UNRISD (Gonzalez de la Rocha, 1998; Jackson and Palmer-Jones, 1998; Kabeer, 1998; Kasente, 1998; Kynch, 1998; Lockwood and Whitehead, 1998; Razavi, 1998; Saith and Harriss-White, 1998; Sen, 1998; Sudha and Rajan, 1998; Walker, 1998) can be found in the bibliography. Some of these papers will be published in the UNRISD Discussion Paper series.

³ The issues relating to indicators are touched upon very briefly here. A comprehensive and in-depth analysis of conventional well-being indicators is available in Saith and Harriss-White, 1998, and of indicators of autonomy/empowerment in Kabeer, 1998.

An enquiry into these causal mechanisms, in turn, leads to a second set of questions about how women and men in particular social contexts relate differently to important assets such as land and labour, given the significant ways in which their livelihood strategies are distinct. To explore these issues, the social institutions within which production and distribution take place, and their "unruly"⁴ practices, rules and norms, are scrutinized. In order to get away from the determinism that has tended to pervade some structuralist analyses of social institutions, particular attention is paid to how women (and men), in specific settings, understand these rules and norms, and how their perceptions in turn feed into the positions they adopt in implicit and explicit instances of contestation and negotiation in the conjugal arena, as well as vis-is employers, community leaders and local government officials.

⁴ The term "unruly practice" is taken from Fraser, 1989, and is used here to highlight the ways in which rules, norms and practices that characterize different institutional arenas can be subverted, ignored or bypassed in explicit and implicit instances of resistance by the less powerful social actors. For a similar usage see Gore, 1993 and Kabeer, 1997.

One of the main themes emerging from this review, and the commissioned papers on which it is based, is that gender analysis can illuminate the diverse processes leading to poverty, and thereby enrich its analysis. But at the same time, it is also true that without a properly contextualized understanding of how poverty is created and reproduced, it will not be possible to comprehend the ways in which gender shapes, and differentiates, those causal processes (Lockwood and Whitehead, 1998).