Essays in Development Economics and Political Economy

Abstract

This dissertation is a collection of essays on three important areas of study in development economics and political economy.

The first chapter asks whether local state legislators of a party provide decision-relevant information to voters about the party's overall quality. I study this question mostly in the context of the two main national parties in India. Using a regression discontinuity design to analyze election data, I find that a party that barely wins instead of loses an election in a state constituency has a lower vote share in that constituency in the following national election; in other words, there are negative spillover effects of local incumbency. Alongside this spillover effect, I also find a direct negative incumbency effect in state legislatures. The findings indicate that, on average, the information that local legislators reveal tends to be bad for the party.

The second chapter, co-authored with Sabyasachi Das, examines whether political alignment between local and state governments is beneficial for the local area. In particular, while local administrative units that are politically aligned with the state government may receive more financial resources, the politicians in these units may have a greater incentive to seek rent. We build a dynamic model that studies how the state government may control the rent-seeking activities of local politicians through the assignment of public servants of varying ability. The model predicts that the state government will assign lower ability police officers to aligned local units and rent-seeking in such units will be higher. Our regression analysis, which uses data from the state of Rajasthan in India, is broadly consistent with the predictions of our model.

The third chapter, co-authored with Shoumitro Chatterjee, asks whether trade liberalization, often considered to promote economic growth, is good for child health, an important development indicator. Specifically, we study the impact of tariff changes in India, over the period 1987-1997, on infant and child mortality. We demonstrate that most of the variation in infant mortality over this period can be attributed to trends at the national and district level, leaving little room for tariffs to have any explanatory power.