Department or Program


Primary Wellesley Thesis Advisor

Patrick McEwan


The choices that voters make are often influenced by economic circumstances. On a macroeconomic level, better economic indicators are often correlated with better outcomes for incumbents in elections. On a smaller scale, programs that provide fiscal support to poorer households have been shown to increase support for incumbent governments. There are a number of reasons why this might be the case – voters may reward incumbents for benefits received, or they may regard such programs as signals of the incumbent government’s investment in their well-being. Conditional cash transfer (CCT) programs are one such type of fiscal intervention. In the past decade, an increasing number of Latin American governments have implemented such programs, which distribute sums of money to targeted households that meet certain conditions such as school enrolment and attendance for school-aged children, and regular health checkups for younger children. In this paper, we consider the effect of one such program, PRAF, on voting outcomes in Honduras. The administration of the program is such that it allows for two empirical strategies: the randomization of treatment among eligible municipalities sets up an experimental estimation strategy, and the threshold of eligibility for the program provides scope for a regression discontinuity estimation at the threshold. Discontinuity estimates indicate that eligibility had a positive effect for the incumbent at the threshold, as expected, but experimental estimates indicate that among the eligible group, those who receive the program are more likely to support the opposition party, perhaps due to administrative difficulties in the distribution of the cash transfers.