What Drives the Electorate? The Answer May Surprise You
For US pundits, each national election is an opportunity to discuss whether the results indicate a mandate for one party or the other. But the political scientist Larry M. Bartels, in a November 27 lecture at the Institute titled “The Elusive Mandate: Searching for Meaning in Presidential Elections,” warned that the idea of a mandate is “so elusive as to be unhelpful.”
Just three weeks after the presidential election, Bartels said his aim was to immediately dispense with the idea that the 2012 election provided anyone with an impressive—or even a slim—mandate. Rather than focusing on a particular election, he offered an “interim report on a career spent thinking about the American electoral process and what it means.” Bartels fleshed out his argument by applying recent research to influential books on electoral politics, most notably The Responsible Electorate: Rationality in Presidential Voting, 1936–1960 (Belknap Press of Harvard University Press, 1966), in which V. O. Key Jr.— whom Bartels regards as the greatest political scientist of his era—paints the electorate as ultimately moved by the relevant issues of the day.
In his much-quoted summary, Key states, “In the large, the electorate behaves about as rationally and responsibly as we should expect, given the clarity and the alternatives presented to it and the character of the information available to it.”
Bartels offered a point-by-point analysis of the major arguments in Key’s book, saying that in his opinion, the subsequent political science literature has not upheld Key’s optimistic portrait of the electorate and the electoral process.
Bartels suggested that many Americans are low-information voters who make decisions on the basis of first impressions, choosing their candidate and then adjusting their policy views according to what that candidate believes.
So what do American voters vote on? As it turns out, Key’s emphasis on governmental performance and executive personality has largely been supported by many additional studies. “Elections can be viewed as referenda on presidential performance and some version of economic performance or income growth,” said Bartels.
There’s a catch, however, which is that voters are myopically focused on current conditions. “The best predictor of a president’s reelection chances is the level of income growth in the 14th and 15th quarters of his term, which is the middle of the election year,” said Bartels, and what happens during that six-month period is more likely to drive voters’ opinions than the incumbent’s overall performance.
According to Bartels, voters seek alternatives to the incumbent party in times of economic hardship, whether that hardship is the product of the markets or of the natural world. “The incumbent party is punished at the polls when weather is too wet or too dry—or even when shark attacks go up,” he said.
“Election outcomes have very significant and systematic policy consequences,” Bartels said. “But I’m forced to conclude that those election outcomes themselves are largely random, shaped much more by short-term income growth and other idiosyncratic factors than by what Key referred to as ‘relative questions of public policy, of governmental performance, and of executive personality.’”