Abstract

I leverage novel experimental designs and 2, 160 months of Members of Parliaments’ (MPs’) Constituency Development Fund spending to test whether fair elections promote democratic responsiveness. I nd that MPs elected in constituencies that were randomly assigned to high levels of election monitoring dur- ing Ghana’s 2012 polls spend 19 percentage points more of their CDFs, on average, compared to those who were elected from districts that had fewer monitors. These legislators were equally absent from parliamentary meetings as their low-monitoring counterparts elected during their terms in o ce, which suggests that fair elections do not motivate politicians to substitute constituency service for parliamen- tary work. The results imply that higher-integrity elections incentivize incumbents to exert more e fort to satisfy citizens’ demand for constituency service. Regarding mechanisms, I provide tentative causal evidence that politicians substitute e fort for fraud when they expect that similar intense future election monitoring will limit their ability to rig their reelection and enable voters to sanction poor performance. The paper demonstrates the downstream e fect of election monitoring on elite behavior and provides causal evidence of the impact of election quality on democratic accountability.