Short-term re-election strategies are widely used by governments around the world. This is problematic if governments can maximize their re-election chances by prioritizing short-term spending before an election over long-term reforms. This paper tests whether longer program exposure has a causal effect on election outcomes in the context of a large anti-poverty program in India. Using a regression-discontinuity framework, the results show that length of program exposure lowers electoral support for the government. The paper discusses a couple of potential explanations, finding that the most plausible mechanism is that voters hold the government accountable for the program's implementation quality.