Center for Policy Alternatives
CPA - Header Photo

Election Day Registration

Hundreds of thousands of Americans could not exercise their right to vote in the 2004 elections due to inefficient or discriminatory voter registration systems.
Reports indicate that registration-related problems were widespread during the 2004 election. The Election Protection Coalition’s Election Incident Reporting System tallied over 10,000 registration-related incidents on Election Day 2004, including voters left off the rolls and voters who never received voter cards or polling place information in the mail. Often, these voters were not given provisional ballots, and in many cases provisional ballots cast were not counted.1
Voter registration deadlines limit voter participation.
Many voters do not take an interest in elections until a few weeks before Election Day, when political campaigns do most of their advertising and races inevitably tighten. Yet 36 states cut off registration opportunities 20 to 30 days before Election Day. A series of Gallup polls in 2004 found that the proportion of Americans giving “quite a lot” of thought to the election rose from 77 percent in mid-September—shortly before voter registration usually closes—to 91 percent by mid-October.2 The 14 percent who became more interested during the final month of the campaign generally could not vote unless they were already registered.
Seven states have enacted legislation that allows voters to register on Election Day.
Idaho, Maine, Minnesota, New Hampshire, Wisconsin and Wyoming have allowed eligible citizens to register to vote and cast a ballot on Election Day for several years.3 In 2005, Montana adopted a law that permits Election Day registration and voting at county election administrators’ offices starting in 2006.
States with Election Day registration have voter turnout significantly higher than the national average.
In 2004, when nationwide voter turnout totaled slightly more than 60 percent, the six Election Day registration states had a combined turnout of almost 74 percent.4 Researchers estimate that elimination of voter registration deadlines and implementation of Election Day registration would result in an average seven percent increase in voter turnout.5 According to a May 2001 poll, 64 percent of nonvoters said that the option to register on Election Day would make them more likely to vote.6
States with Election Day registration report few problems with fraud or administrative 
complexity.
Officials in the six Election Day registration states report minimal incidence of fraud and no unusual administrative problems. Indeed, Election Day registration can help address one of the most frustrating administrative problems exposed during the 2004 elections: incomplete or inaccurate registration lists that bar people from voting. In the states that use Election Day registration, the work of adding new voters has proven manageable. Election officials in these states educate registration clerks on how to make reasonable estimates of voter turnout, ensuring that polling places are adequately staffed and have enough materials.
States that implement Election Day registration do not face substantially higher costs.
The major cost associated with Election Day registration is increasing the number of polling place workers and training them to handle new registrations on Election Day. But, Election Day registration lowers other costs of processing provisional ballots and handling the onslaught of forms turned in just before the voter registration deadline.
Election Day registration reduces the need for cumbersome provisional ballots.
The Help America Vote Act (HAVA), enacted by Congress in 2002, requires states to offer provisional ballots to voters who claim to be registered but who are not listed on the voter rolls. Election Day registration would greatly reduce the need for provisional ballots. Most importantly, while provisional ballots often go uncounted, Election Day registration provides certainty to citizens that their votes will count.
Research supports the use of Election Day registration to increase turnout of traditionally underrepresented groups.
Underrepresented groups—youth, people of color and those with lower educational attainment—would gain the most from the implementation of Election Day registration.7 Research has found that Election Day registration could increase youth turnout in presidential elections by as much as 14 percent.8
States are moving toward implementation of Election Day registration.
Montana adopted a form of Election Day registration in 2005. In recent years, Connecticut, Massachusetts, New Mexico and Vermont have seriously considered legislation to adopt Election Day registration.

This policy summary relies in large part on information from Dēmos.

Endnotes
  1. Election Incident Reporting System, “Nationwide Election Incidents,” 2004.
  2. CNN/USA Today/Gallup polls conducted September 11-14, 2004 and October 15-18, 2004.
  3. An additional state, North Dakota, does not require voter registration.
  4. Dēmos, “High 2004 Turnout for States with Election Day Registration,” January 10, 2005.
  5. Craig Leonard Brians and Bernard Grofman, “Election Day Registration’s Effect on U.S. Voter Turnout,” Social Science Quarterly, March 2001.
  6. Medill School of Journalism and Medill News Service, “America’s No-Shows,” Northwestern University, May 2001.
  7. R. Michael Alvarez, Stephen Ansolabehere, and Catherine H. Wilson, “Election Day Voter Registration in the United States: How One-Step Voting Can Change the Composition of the American Electorate,” Caltech-MIT Voting Technology Project Working Paper, June 2002.
  8. Mary Fitzgerald, “Easier Voting Methods Boost Youth Turnout,” CIRCLE Working Paper, February 2003.
Updates