Essays on Development Economics

This dissertation is a collection of three independent papers in empirical development economics. The first chapter studies the effect of a family planning program in Bangladesh, which successfully reduced fertility, on households' asset accumulation. In developing countries parents expect their children to take care of them when they are old. Children also help parents to smooth consumption over their life cycle. They send remittances when parents are old and have relatively low income. The chapter presents a model where asset accumulation and children are substitutes, and finds that households exposed to a family planning program have lower fertility and more assets than those who were not exposed to the program. Chapter 2 examines effect of the same program on female autonomy. Policy makers and planners often view family planning programs (FPP) as being conducive to female autonomy. They argue that when women have fewer children they can earn more income and enjoy more property rights, higher mobility and greater decision making power inside and outside the household. But this may not be true in all situations. Using household data from 142 villages in Bangladesh this paper shows that although a family planning program reduces women's fertility and thereby allows them to enjoy higher levels of private consumption through expanding their outside opportunities, it significantly reduces their decision making power within the household. A simple analytical model is presented to reconcile this empirical evidence. Finally chapter 3 explores the macroeconomic usage of aid using panel data for a broad sample of aid-recipients. An increase in aid must go toward a reduction in the current account balance (in which case there is a real transfer of resources from donor to recipient and aid is said to be absorbed), an increase in capital outflows, or into international reserves. We find that short-run absorption is typically very low. While absorption increases in the long-run, it is still significantly less than complete and only a tiny fraction of the absorbed aid dollars go towards investment. It is likely that the remaining aid is lost through the capital account. Moreover, aid spending, defined in terms of the increase in government fiscal expenditures as a result of aid, is significantly greater than aid absorption, implying that aid systematically leads to an injection of domestic liquidity in recipient economies, with possible adverse consequences for macroeconomic management. The evidence marshaled here may help illuminate the rather weak link between aid and growth found in the literature. Thesis Supervisor: Esther Duflo Title: Abdul Latif Jameel Professor of Economics Thesis Supervisor: Abhijit Banerjee Title: Ford International Professor of Economics To my parents and Ashiq Acknowledgement I am deeply indebted to my advisors, Esther Duflo and Abhijit Banerjee. They have been a source of inspiration for the last five years. And it is their valuable guidance and support that helped me to complete this dissertation. I am also obliged to the faculty and students of the Economics Departments for their valuable comments and feedback. The Japan-IMF Scholarship Program provided financial support for the first two years. The third chapter would not have been possible without the cooperation of the Development Issues Division of the IMF. It has been a pleasure working with them. I am also grateful to RAND, the library of ICDDR,B for providing data for the first two chapters of this thesis. The last couple of years would have more difficult without the constant support of Tuli Banerjee. She made my stay at MIT a wonderful experience. I will always be in debt to her. I am profoundly grateful to my parents for making me the person I am today. They are the best. Finally I would like to thank my husband, S. M. Ashiquzzaman, who inspired me to come to MIT. Without his affection, encouragement and help I could not have come this far.