Are Public Funds Used to Maintain Ruling Coalitions? Evidence from India

Abstract

Political considerations can distort public policies in developing countries, making them less effective. While past literature has focused on incumbent politicians manipulating policies for reelection purposes in two-party settings, around 52% of the world’s democracies have coalition governments. In such settings, a national incumbent can not only allocate funds to woo voters pre-election, but also to buy legislative support from smaller parties post-election. My paper presents a simple framework formalizing these two mechanisms and tests it using a newly constructed dataset of district-level disbursements from India’s Total Sanitation Campaign. These two mechanisms give rise to two distinct patterns in the disbursement of funds: the “Buying Votes Cycle” and the “Patronage Cycle.” The former is the traditional cycle with a pre-election increase (60%) in the release of funds in swing districts for the national parties. The latter is a previously undocumented cycle with a post-election increase (55%) in the release of funds that occurs, as predicted by the model, only in safe districts for smaller parties. Both cycles are large enough to affect the timing of household consumption and hence their welfare.