Two essays on the women’s empowerment in agriculture: an empirical assessment of the WEAI and a theoretical proposal with relationship to time allocation

Abstract

Several authors argue that one mechanism to promote economic growth in agriculture is a sector that is more inclusive and equitable towards women. Thus, there is increasing interest in the measurement and drivers of women’s condition and efforts directed at improving their agency, status and power. One measurement strategy center on the novel concept of the Women’s Empowerment in Agriculture Index (WEAI). This dissertation is composed of two essays to empirically evaluate women’s empowerment and focuses on the WEAI. The first paper of this dissertation centers on the evaluation of the Women’s Empowerment in Agricultural Index and the Abbreviated Women’s Empowerment in Agricultural Index (A-WEAI). These tools have been used extensively to measure and track women’s empowerment in agriculture and is based on the multidimensional empowerment concept that relies on different indicators to measure the latent concept of women’s empowerment. After a quantitative assessment of the indices using the Multiple Indicators Multiple Causes approach (MIMIC), we conclude that holding all other variables constant, the probability of increasing the correlation between the variable women’s empowerment and the indicators of the indices is higher under the WEAI than under A-WEAI. Ownership of assets and workload indicators require attention, but due to uniqueness that workload indicator brings to the women’s empowerment discussion, the way this indicator is used in the calculation of the indices needs refinement. The paper concludes by proposing a theoretical revision of the links between women’s empowerment and time allocation. Motivated by the finding of the first paper, the second paper proposes a new theoretical framework that is based on Becker’s model of allocation of time, but incorporates Sen’s and Kabeer’s definitions of empowerment, with respect to time allocation. Under this framework, women’s empowerment is part of a utility maximization problem, and new relationships and explanations offered to understand some of the apparent contradictory results found in empirical studies. These studies found empowerment contributes to decreasing the allocation of time to leisure. The objective of the paper is not only to propose a different theoretical approach, but to test it empirically using the Bangladesh Integrated Household Survey and modeling time allocation using Two-Stage Least Squares to control for endogeneity of empowerment status. We conclude, after accounting for the endogeneity in the models, that the Sen model generates more consistent estimates of the relationship between time allocation and empowerment. We suggest data collection strategies to understand the actual freedom of individuals when making time allocation decisions in order to refine assessment of the model. These refinements, combined with our modeling, will assist in rigorous assessment of interventions designed to improve women’s welfare.