When does redistricting matter? Changing conditions and their effects on voter turnout

Abstract

Redistricting has been the subject of a great deal of scholarship, conducted under the suspicion that redistricting has negative effects on voting behavior via mechanisms like competitiveness, incumbency, and partisan balance of the district. Most of this scholarship, however, falls short in measurement of the voter-level effects of redistricting and its different mechanisms. In this study, I test the effects of these mechanisms, and thus of redistricting as a whole, at the highly precise individual level and among different but significant subsets of voters. By predicting turnout with individual-level changes in competitiveness and candidate and district characteristics, this project specifies the impact of redistricting on voting behavior with a focus on the individual voter that has been missing from the literature. Redistricting in the state of Florida, which was subject to national legislative redistricting and reapportionment prior to the 2012 election, offers a valuable exogenous test by which changes in competitiveness, incumbency, and partisanship resulting from redistricting can be definitively assessed as drivers or depressors of turnout. Results indicate that while the overall effects of redistricting on individual turnout appear to be substantively low, the effects of changes in partisan composition in a voter's district that result from redistricting can decrease turnout likelihood by more than 4%. As such, these results call into question what it really means for a voter to be “redistricted.” This project precisely defines what aspects of the redistricting process are most consequential to an individual's voting behavior in a way that the previous literature has been unable to do.