Aid , Redistricting and Elections : Evidence from Uganda

This paper analyzes the potential impact of foreign aid on electoral outcomes of presidential elections in Uganda. We use geocoded aid data and electoral results at the county-level, as well as redistricting data, also at the county-level. The methodology used is difference-in-differences and we evaluate the impact of foreign aid in counties that were reallocated to a newly created district in the same period in which they received aid. A simultaneous regressions model is used as a robustness check and to understand the mechanisms. We find a positive impact of the treatment on the share of votes received by the incumbent. Our results suggest that administrative unit proliferation might function as a mechanism through which aid is used by the incumbent to gain electoral support.