Essays in Development Economics

Abstract

This dissertation consists of three self-contained chapters on development economics. The dissertation is focused primarily on one of the largest input subsidy program in the world in terms of how beneficiaries are whether households hide income. As a separate project, I also look at the impact of star students on their siblings test scores in the context of a Peruvian high-achievers national boarding school. In the first chapter, entitled "Measuring Sensitive Questions: Income Hiding and Subsidy Allocation ", I document the extent to which villagers hide income from local leaders and other villagers as a strategic behavior in the context of a large scale agricultural subsidy in Malawi (FISP). My main contribution is methodological. I use three different measures of income hiding to asses the extent of this practice: direct questions, list randomization, and, willingness to pay to hide income. I find that income hiding prevalence is between 17 to 27 percent depending on the measure employed. Also, I find that villagers hide income from different people and the three most common categories are village headmen (16%), neighbors (16%), and, friends (15%). The second chapter, entitled "Decentralization and Efficiency of Subsidy Targeting: Evidence from Chiefs in Rural Malawi", is joint work with Pascaline Dupas (Stanford) and, Jonathan Robinson (UCSC). We study the trade off between centralized and decentralized subsidy targeting in the context of two large-scale subsidy programs in Malawi (for agricultural inputs and food). Decentralized targeting is carried by traditional leaders ("chiefs") who are asked to target the needy. Using high-frequency household panel data on neediness and shocks, we find that nepotism exists but has only limited mistargeting consequences. Importantly, we find that chiefs target households with higher returns to farm inputs, generating an allocation that is more productively efficient than what could be achieved through a a proxy means test used for centralized targeting. This could be welfare improving, since within-village redistribution is common in the study setting.The third chapter, entitled "On the Peer Effects of Star Students", is joint work with Manuel Barron (assistant professor at Universidad del Pacifico), and, Gabriela Cuadra (Ph.D. Student, UCSC). We estimate the effect that star students have on their siblings' learning outcomes, measured by their high school grade point average (GPA) and their math grades. To this end, we couple administrative school data on grades with an unusual natural experiment in Peru that generates exogenous variation in the presence of star students at home. We find that star students increase their siblings' GPA by 0.33 standard deviations and their math grades by 0.22 standard deviations. The effect size is inversely related to number of siblings, suggesting that the remaining siblings act as substitutes for the star student.