Favoritism and Flooding: Clientelism and Allocation of River Waters

Political favoritism is commonly documented in developing democracies, demonstrated by better outcomes in favored regions. There is little understanding of the mechanisms driving such allocation of resources. I demonstrate favoritism in the allocation of agricultural resources, which eventually affect agricultural production. I use close elections to get random variation in a region's alignment with the political party in power and find that water may be diverted to favor areas aligned with the ruling party. Water availability is in favor of upstream districts and against downstream districts when upstream districts are aligned with the ruling party. The opposite is true when downstream regions are aligned with the ruling party. As a result of this favoritism, floods (or droughts) are less likely to occur in downstream regions, and agricultural yields are better there when the ruling party has incentives to favor them. I argue that the ruling party's influence over autonomous agencies that control water allocation allow them to favor their constituents.