Do Elections Encourage Public Actors to be More Responsive?

Abstract

In the U.S. many public services are provided by individuals who are selected in local elections. The incentives created by the election mechanism are relatively understudied. One of these public actors is the prosecutor. The U.S. is unique in the world in its use of popular elections to select and retain local prosecutors. They exercise an enormous amount of discretion, and the election mechanism serves as the primary accountability tool. Critics point to voters' poor information, weak incentives to become informed, and insufficient knowledge to properly evaluate as arguments against the ability of elections to properly incentivize prosecutors to do their job well. Nevertheless, empirical research has documented important distortions in prosecutorial decision making close to election periods. This has called into question further the institution's appropriateness. Here, we ask whether elections at least encourage local prosecutors to be responsive to citizens. We design a novel field experiment where we send an information request to a randomly selected sample of offices in the country. Whether the office responds to the request is our measurement of responsiveness. We show that offices whose head is up for re-election in 2020 are more likely to respond than offices led by prosecutors not up for re-election this year. We also show that offices in states that appoint their local prosecutors are substantially less likely to respond than a matched, similar set of offices in states that use popular elections. Thus, we provide evidence that the election mechanism encourages public actors to be more responsive.

Keywords: election, email, field experiment, prosecutor, responsiveness, survey

JEL Classification: K4, H1, C93