BUFFALO, N.Y. -- A new study by researchers at the University at Buffalo modifies the longstanding theory of "retrospective voting," which holds that in presidential elections voters happy with the in-party's performance will support its candidate and when unhappy are more likely to support the opposition party's candidate.

The study supports the theory of conditional retrospective voting, which holds that in such elections, voters are more likely to give consideration to national conditions when evaluating an incumbent seeking re-election than when considering a successor candidate from the incumbent's party.

The research, "The Theory of Conditional Retrospective Voting: Does the Presidential Record Matter Less in Open-Seat Elections," is published in the latest issue of the Journal of Politics (Cambridge University Press).

The authors include noted researcher James E. Campbell, PhD, professor and chair of the UB Department of Political Science.

Click here to see a video interview with Campbell about his findings.

The theory of retrospective voting holds that (as well-known political commentator and public intellectual Walter Lippman wrote) the essence of popular government is "to support the Ins when things are going well (and) to support the Outs when they seem to be going badly.…"

Campbell notes, however, that things are not quite as simple as Lippman described.

His study found that voters do not respond to a successor candidate in the same way as they respond to an incumbent. Voters unhappy with an incumbent do not "take it out" on a successor candidate of the same party to the same degree as they would on the incumbent he were running.

For example, in 2008 President George W. Bush would have been treated more roughly by the voters than they treated Sen. John McCain. McCain was in a tight race with Obama until the Wall Street meltdown hit in mid-September of 2008. With his low approval ratings, Bush would already have been far behind Obama even before the economic crisis hit, according to Campbell's theory of conditional retrospective voting.

By the same token, when times are good, the successor candidate only gets part of the credit that would have gone to the incumbent. For example, Al Gore received only part of the credit for what most voters saw as good times in the 2000 election. As the incumbent, President Clinton would have received the full credit for the conditions that most voters regarded as being quite positive in the 2000 election.

Campbell explains, "From the end of World War II right up until today, voters assigned only partial credit or blame for national conditions to in-party successor candidates. Perhaps this is because, unlike incumbents, successors are not considered to have had personal power over policies that might have affected the national conditions leading up to the election."

The article is co-authored by Bryan J. Dettrey, PhD, visiting assistant professor of political science, Oklahoma State University, and Hongxing Yin, a PhD candidate in political science at UB.

"Our research examined what we call the theory of 'conditional retrospective voting' at both the aggregate level on elections since 1948 and with individual-level survey data since 1972," Campbell says.

"Our analysis consistently found, as we expected, that for most voters, the record of the past administration in evaluating would-be successors does matter," he says, "but not as much as it would if the incumbent president was running. Voters don't seem to assume the same degree of continuity as when they are voting for or against an incumbent."

He says the theory explains why would-be successor candidates have done better than expected when the incumbent president is unpopular -- as McCain did in 2008 before the financial crisis hit -- and worse than expected when the incumbent president was popular -- as was the case with Al Gore in 2000.

Campbell says political scientists care about the issue of retrospective voting because it helps to explain how leadership is determined in this country: why candidates are elected.

"Voting is central to the democratic process," he says, "and this research helps us identify how it actually works as opposed to how it was intended to work or how it was assumed to work. It's hard to imagine any question in the field that is more important.

"Why someone is elected has great implications for how they are likely to govern and huge consequences for virtually every aspect of American life, from war and peace to prosperity and economic hard times and everything in between," he says.

Campbell says that in our polarized electorate, conditional retrospective voting among those in the middle is what has determined the nation's leadership time and time again.

"The country is more polarized now than at any time since the Civil War," he says.

"We have a group on the left and the right whose ideas of governance are based on what they consider to be 'core values.' But that leaves a huge group of people in the middle, and that is the group that really holds the balance of power," Campbell says.

"Moderates see some things in liberal principles that they like and some things in conservative principles that they like, so they are torn. They often decide their votes not so much on the principles or ideologies of either side, but on the performance of the parties. This study sheds light on how this crucial part of the electorate -- the center -- bases its voting decisions on the past performance of political parties and political leaders," he says.

"We saw this born out in the 2006 midterm election, and again in the 2008 election, which turned on performance issues like whether or how Republicans had handled power," Campbell says.

"And in 2008, voters were concerned with what was happening in the Iraq and Afghanistan Wars and how the Wall Street meltdown had been handled by the Republicans. Then in the 2010 midterms, with the Democrats in power, the elections turned largely on how political moderates thought the Democrats had dealt with high unemployment levels and the recovery from the recession.

"Understanding how people respond to performance issues and their retrospective judgments about the record," says Campbell, "is crucial to understanding what is likely to happen in future elections."

The University at Buffalo is a premier research-intensive public university, a flagship institution in the State University of New York system and its largest and most comprehensive campus. UB's more than 28,000 students pursue their academic interests through more than 300 undergraduate, graduate and professional degree programs. Founded in 1846, the University at Buffalo is a member of the Association of American Universities.