Abstract
Voter registration is widely viewed as a barrier to voter participation in general, and especially so for likely Democratic voters. A popular remedy for both turnout effects is election day registration (EDR), which eliminates the closing date by permitting registration at the polls. Following earlier research we posit a small positive effect of EDR on turnout. But contrary to conventional wisdom, we theorize that individuals most likely to take advantage of EDR are in fact Republican voters. To investigate these causal effects we make use of a natural experiment in Wisconsin. When EDR was implemented in Wisconsin in 1976, only municipalities that already required registration were affected by the change in the law. Analysis of this intervention shows that EDR did increase turnout in Wisconsin but actually decreased the Democratic share of the two-party vote for president.
Get full access to this article
View all access options for this article.
