The impact of geography and institutions on Indian economic development

This thesis contributes to the debate on the fundamental determinants of economic growth and development. It tests the importance of geography and institutions in the long run development of India. It consists of three chapters, the first two of which are self-contained, while the third builds on ideas emerging in the preceding chapters. The first chapter investigates the role of rainfall variability in determining development in late colonial India. An original dataset on rainfall shocks is employed to show that rainfall variability and specifically negative shocks are significantly associated with lower district level literacy. Both ex-ante and ex-post coping strategies that could explain such an effect are explained and evidence of the impact of rainfall shocks on agricultural incomes and school enrolment is provided alongside. The second chapter estimates the causal impact of colonial land revenue institutions on long run rural development in India. A novel village-level dataset that links colonial era villages to a modern village census is used in a geographical Regression Discontinuity Design (RDD) framework. Results indicate that villages assigned to landlords in the colonial era have lower literacy, irrigation and public good provision at the start of the 21st century. The channels proposed include migration, public good provision and the differential influence of the railways. The third chapter attempts to explain how institutions have had persistent impacts in India. To do so, it traces the influence of historical institutions in explaining patterns of success in the implementation of the land reforms and the uptake of Green Revolution technologies across the different states of independent India. The key determinants in both cases are identified and their relationship with historical land tenures is examined. Quantitative evidence for the significance of institutions in the take-up of High Yielding Varieties (HYVs) is presented, alongside evidence of a more restricted role for rainfall variability.