Political economy of third-party interventions

Abstract

Public policy interventions by both national and international (non-governmental) organizations are common in developing countries. In this paper we show that such interventions by third party organizations may have unintended political consequences which may, in turn, undermine the effects of the intervention. We study a World Bank capacity building program for local governments in an Indian state which provided incentivized training to local politicians on various governance practices in a selected set of village governments and allocated discretionary grants to the better performing ones. The selection criteria for the villages into the program allows us to use a regression discontinuity design to look at the causal effects of the program. We first show that even though the program villages received training and were allocated larger discretionary funds, there was no effect on the incumbents’ rerunning and reelection behavior. We then show that the state government responded to the program by allocating higher resources towards the program villages where the incumbents switched their party affiliations in favor of the ruling party. Consequently, party switching went up in program villages. The switchers also got reelected with higher probability. The party loyalists, on the other hand, received reduced state grant and consequently, their reelection probability went down. We argue that the ruling party at the state reallocated discretionary grants in response to the program in order to maximize its presence in local governments. The paper provides a cautionary tale for evaluating third party public policy interventions in future.