Moved out, Moved On : Assessing the Effectiveness of Voter Registration List Maintenance

The quality of voter registration lists has been at the center of debates about election administration in the United States for over a decade. Lists with excessive numbers of ineligible registrants, whether due to death, mobility, or other reasons, complicate the logistical task of running elections and an easy target for those concerned with the electoral integrity or voter fraud. This paper seeks to answer two questions regarding the accuracy of voter lists. First, how well do the registration cancellation rates line up with the rates we would expect to see given demographic trends? Second, when in an electoral should we expect that voter lists will be the most reflective of the population of registered and eligible voters. Using administrative and demographic data over four two-year election cycles, we find very strong evidence that voter lists are almost entirely absent of deceased registrants. On the other hand, the data suggest that election officials have a much more difficult time removing registration records of those who have moved out of the jurisdiction. We supplement these findings with data from Florida and Virginia which allow us to explore temporal variation in removal rates. We find that the states employ two different paradigms for registration list cleaning