Does Automatic Voter Registration Work?

• In Event: Election Law in the States

Abstract

Scholars of American turnout concluded long ago that our system of voluntary registration was one of the main factors behind the comparatively low turnout in American national elections (Mitchell and Wlezien 1995; Powell 1986; Wolfinger and Rosenstone 1980). These scholars contended that if US voting and registration laws were fully liberalized, turnout in American national elections would approach that of other industrialized democracies. Powell (1986) estimated that if automatic voter registration (AVR) were implemented at the national level in the United States turnout in presidential elections would increase by approximately 10 percentage points. Since the 1980s, the restrictiveness of voting and registration laws has loosened somewhat with the implementation of the National Voter Registration Act in 1995, the implementation of Election/Same Day Registration in several states, and the expansion of early and mail-in voting. In 2015, Oregon became the first state to adopt automatic voter registration, and California followed suit that very same year. In both states, AVR is an “opt out” process: when a citizen applies for, renews or replaces a driver’s license or identification card his/her name is placed on the state’s registration rolls unless that citizen chooses to decline or opt out. Since 2015, another 13 states and the District of Columbia have adopted similar processes.

Using the 2018 Current Population Voter and Registration Data File, the proposed paper will be an examination of whether or not the implementation of AVR had an effect on 1) the individual probability of registering to vote and 2) the individual probability of casting a ballot in the 2018 national elections. Only seven states and the District of Columbia actually implemented AVR prior to the 2018 election (Alaska, California, Colorado, Georgia, Oregon, Rhode Island, Vermont), and thus these are the only states in which the reform could have an effect. Because AVR is newly implemented in these states any effects in the 2018 election should be muted. However, we would expect the longer a state has had the reform in place, the greater its effect on the probability of registering to vote and the probability of casting a ballot in the 2018 elections. When completed, the proposed paper will provide an initial assessment of the ability of this reform to increase registration and turnout in American elections.

Author

• David Hill, Stetson University