Three Essays on Empowerment: Issues of Subjective Well-Being, Identity, and Agricultural Productivity in Rural Bangladesh

Women's empowerment is universally recognized as an important development goal and intrinsically valuable for women's well-being. Each of the three essays comprising this dissertation examines a different aspect of women's empowerment.The first essay investigates the relationship between women's empowerment and subjective well-being (SWB). Understanding the full implications of women's empowerment for improvements in women's lives is an especially salient concern for development researchers and policymakers in Bangladesh and around the globe. This study argues that a complete understanding of the potential for women's empowerment to positively impact women's well-being in rural Bangladesh requires consideration of women's SWB. However, existing studies of the relationship between women's empowerment and SWB are scarce and tend to rely on indirect measures that capture only certain aspects of empowerment and may not perform well in rural settings.Using data from the 2011-2012 Bangladesh Integrated Household Survey (BIHS), this study hopes to answer the following questions. First, what is the relationship between empowerment and SWB among women in rural Bangladesh? Second, is the relationship between women's empowerment and SWB the same for women across consumption quintiles? The latter question addresses an emerging debate within that literature that the association between women's empowerment and SWB may depend on women's level of income--in particular, that the association between women's empowerment and SWB is weaker among poorer households. Women's empowerment is operationalized in terms of two indicators derived from the Women's Empowerment in Agriculture Index (WEAI) developed by Alkire et al. (2013): the uncensored Five Domains of Empowerment (5DE) index, which captures women's overall level of empowerment in agriculture, and the Relative Autonomy Index (RAI), which reflects women's ability to act on behalf of their own personal values as opposed to external socially imposed values. Women's SWB is measured in terms of their overall life satisfaction, based on the question: "How would you rate your satisfaction with your life overall?" To answer the questions posed in this study, a series of multivariate regressions are estimated to assess the relationship between women's empowerment, using alternative measures, and SWB Because empowerment itself may be endogenous, instrumental variables (IV) regression is used as opposed to ordinary least squares (OLS). The relationship between women's empowerment and SWB is estimated, first, for the entire sample of women and, second, for women within each consumption quintile, measured in terms of annual per capita consumption.The main results of the analysis indicate an unambiguously positive and significant correlation between women's empowerment and SWB among women in rural Bangladesh, regardless of women's level of household wealth. Moreover, the results remain unchanged when empowerment is measured in terms of women's empowerment score (uncensored 5DE) or women's autonomy (RAI score).Building on insights gained in the first essay, the second essay probes more deeply into the relationship between women's empowerment and SWB. Specifically, this essay examines the relationship between the time women spend on unpaid work and SWB. Addressing gender inequalities in the time men and women spend on unpaid work is an important development objective with the potential to directly impact women's well-being. Indeed, the freedom to allocate one's time in a manner of one's own choosing represents an important aspect of women's empowerment. This is a particularly important goal in rural Bangladesh where the competing claims on women's time of extreme poverty and traditional gender division of labor often force women into difficult sacrifices that can have debilitating effects on women's physical and mental health. However, while existing time use research provides a great deal of insight into the costs imposed on women's livelihoods by traditional gender norms, little direct empirical evidence exists on the impact of social norms on women's time use.This essay extends both the time use and SWB literatures by developing a conceptual framework based on Akerlof and Kranton (2000) that links women's identities and the time women spend on unpaid work. Specifically, this essay empirically tests the hypothesis that women's personal attitudes about gender roles condition the relationship between SWB and unpaid work. This hypothesis is largely confirmed using multivariate (OLS) regression analysis and primary data collected in early 2014 for a small sample of women from 10 villages in rural Bangladesh. Among women who tend to strongly disagree with patriarchal notions of gender roles, higher levels of unpaid work are associated with lower levels of SWB, and among women who tend to strongly agree with patriarchal notions of gender roles, higher levels of unpaid work are associated with higher levels of SWB. However, among women who tend to neither disagree nor agree with patriarchal notions of gender roles, unpaid work is found to be unrelated to SWB. These results suggest that in certain circumstances--when women strongly identify with patriarchal or egalitarian notions of gender roles--identity may alter the payoffs associated with the amount of time women choose to spend on unpaid work. Identity may, thus, influence women's preferences for how much time they spend in unpaid work.The third essay examines the relationship between women's empowerment and agricultural productivity. Although a great deal of research exists on the sources and consequences of gender differences in agricultural productivity, few studies investigate linkages between women's empowerment and agricultural productivity, and the question of whether female farmers fail to achieve the same level of agricultural productivity as male farmers due to their inability to exercise agency within the agricultural domain remains unanswered.Using data from the 2011-2012 BIHS, this essay poses two questions about the gender and agriculture in rural Bangladesh. First, do gender gaps in agricultural productivity exist in rural Bangladesh? Second, what is the relationship between women's empowerment and agricultural productivity in rural Bangladesh? Gender gaps are estimated based on women's participation in decision-making and ownership status for a particular plot of land. Agricultural productivity is measured in terms of technical efficiency (i.e., the ratio of actual output to the maximum technologically feasible level of output given a set of inputs). Women's empowerment is operationalized in terms of two indicators derived from the WEAI: an aggregate measure of women's empowerment (the uncensored 5DE) and a measure of women's group membership. All estimation procedures are carried out using stochastic frontier analysis to address econometric problems related to the simultaneity of productivity and input choice in the estimation of production functions.The main findings of this essay emphasize the importance of women's empowerment, particularly in terms of group membership, in any initiative for promoting agricultural productivity. First, both the uncensored 5DE score and group membership are found to be positively associated with higher levels of agricultural productivity for all plots operated by women's households. Thus, positive spillover effects may exist, in terms of access to social capital or credit, that extend the benefits of women's empowerment outward to all household members. Second, no evidence of gender gaps in agricultural productivity is found in terms of women's actual participation in decision-making or women's ownership status for a particular plot of land, nor when based on female headship.