STANFORD GRADUATE SCHOOL OF BUSINESS—People who vote by mail may be casting their ballots prematurely and wasting their votes. An interesting new study coming out of Stanford Graduate School of Business says this phenomenon may have important implications for ballot initiatives, as well as future elections, particularly primaries.
The study found that in the 2008 California presidential primary, Democratic hopeful John Edwards and Republican contenders Rudy Giuliani and Fred Thompson got a significantly higher percentage of votes by mail than they did at polling places. Researchers say this is probably because many people mailed in their ballots before these candidates dropped out of the race. “If they had waited, many of them probably would have voted for other candidates,” says assistant professor Neil Malhotra, of the Graduate School of Business who conducted the study with Marc Meredith, PhD ’08, a visiting lecturer at MIT.
Given that mail-ins now comprise more than 30 percent of ballots in states that allow early voting, the issue of how voting by mail may affect election outcomes has becoming increasingly important.
“Absentee ballots tend to come back in two big chunks,” says Malhotra. “One chunk is returned just a few days after the ballots are mailed out, and one comes back at the last minute. The people who mail in their ballots right away do not have all the information that people at the voting booths have later.”
In their study, Malhotra and Meredith compared California precincts of less than 250 voters, where all are required to vote by mail, against slightly larger precincts where voters have the option to cast ballots live on election day. They estimate that more than 5 percent of Californians wasted their votes in the primary by choosing candidates that eventually dropped out.
Moreover, they discerned that early mail-in votes were more influenced by events taking place just prior to the ballots being sent out. “For example, Mike Huckabee, who didn’t drop out of the race, did better among vote-by-mail voters than precinct voters,” says Malhotra. “This is because the ballots were mailed out just after his huge victory in the Iowa caucuses, when he was getting a lot of media attention. By the time the California primary rolled around, however, he had lost a lot of other primaries, and his campaign had lost momentum. But it was too late for those who had already voted for him by mail.”
In close elections, uninformed mail-in votes could tip the scales in favor of one candidate over another. In the 2004 presidential election, for example, many voters in Ohio and Iowa sent in their ballots even before the first debate between George Bush and John Kerry had taken place. Although Kerry was considered to have won the debate, Bush won these states––but only by about 1 percent. “It could well be that if people who voted by mail had waited until after the debates, they would have voted differently,” says Malhotra.
In the November 4, 2008 general election,voting by mail may have influenced the outcome of ballot initiatives in particular, the researchers said. “A lot of the educating and campaigning about propositions happens at the very end,” says Malhotra. “Someone who voted in early October may have voted differently if he’d had more information at his disposal.”
With more of the public now voting early, major campaign and media developments in the last month of the election––including “October surprises”––are mattering less and less. “It’s affecting candidates’ campaign strategies, which used to have more of a ‘we’re in it to the end’ approach,” Malhotra observes.
While absentee ballots have made voting easier and more accessible to millions of Americans, the unintended effects they can have on election outcomes should have more policy makers thinking, say the researchers. They suggest that to level the playing field between mail-in and drop-in voters, the government may want to distribute ballots later, not disqualify slightly late ballots, and consider seriously whether voting by mail should ever be made mandatory––as it has been in Oregon. “What we’ve brought to light is a vote-by-mail ‘con’ that no one has considered before,” says Malhotra.
Marc Meredith, PhD ‘08
Massachusetts Institute of Technology
Assistant Professor of Political Economy
Stanford Graduate School of Business
“Can October Surprise? A Natural Experiment Assessing Late Campaign Affects,” by Marc Meredith and Neil Malhotra, Stanford Research Paper No. 2002, October 2008
“Convenience Voting,” by Paul Gronke, Eva Galanes-Rosenbaum, Peter A. Miller, and Daniel Toffey, Annual Review of Political Science, 2008
Also on Stanford Knowledgebase: