It's always the economy, stupid
With merely five months before the election and the outlook grim for Democrats, we're starting to hear rumblings of a fight within the White House. The political side, we're told, wants to focus on the swelling deficit, which it believes is contributing mightily to the public's sense that the economy isn't being effectively managed. Karl Rove, who's been on the political side of a White House himself, agrees with them: "People's concern about the spending and the deficits and the debt and the out-of-control government have been growing and growing and growing," he said on Fox News. "And it's one of the key drivers in the 2010 election."
The economists disagree: They see a weak economy that still needs government support. And if the government fails, and the economy worsens, the midterms will be a massacre. Richard Trumka, president of the AFL-CIO, joins them in this view: "The jobs hole -- and the decades-long stagnation in real wages -- are the source of the anger that echoes across our political landscape."
Beneath all this sits an age-old tension: The job of governing is different than the job of getting reelected. What do you do when good politics and good governance point you in the opposite directions?
But maybe we don't have to choose. For decades now, political scientists have been building election models that attempt to predict who will win in November without making any reference to candidates or campaigns. They can generally get within two percentage points of the final vote, and they don't need to know anything about the ads, the gaffes or the ground games to do it. All they really need to know about is the economy.
"In presidential elections," Princeton political scientist Larry Bartels says, "a 1 percent boost in election-year income growth has typically increased the incumbent party's vote share by about 2 percent. So an incumbent party that won 51 percent of the vote in an average economic year like 2004 would be expected to win only 46 percent in a recession year like 2008." Which is, as you may remember, pretty much exactly what happened.
Congressional elections are a bit more difficult because they're more local, but they end up being predictable, too. Gary Jacobson, a political scientist at the University of California at San Diego, has a model that uses the number of seats the majority party holds, the approval rating of the president and the change in real disposable income, and predicts about 70 percent of the change from one election to the next.
All of this suggests political scientists have a pretty good handle on what wins elections, so I began asking them the question that some say is bedeviling the White House: Should the White House focus on polls or paychecks?
The answers were unequivocal: "A policymaker reading polls who finds that people are concerned about the deficit and says I should rein in spending and I'll get credit for that, I don't think there's evidence that'll move voters," the University of Denver's Seth Masket says. "You want to get as much money in voters' hands in the months before the election as possible."
John Sides, of George Washington University, helped me run the numbers on presidential elections: We made one graph comparing the share of the vote the incumbent party got with the change in the deficit that it had presided over. It looked as if we'd spilled a bag of dots onto a piece of paper. The next graph plotted vote share against change in real disposable income. The line showing a correlation fit perfectly -- more perfectly, in fact, than I'd anticipated.
We all agree that the economy is important, but there have been elections where the candidates mattered, right? In, say, 1964, when Barry Goldwater's extremism repulsed the electorate; and in 1972, when George McGovern's campaign of hapless hippies ran into Richard Nixon's silent, and annoyed, majority; and of course 1984, when Ronald Reagan picked up 49 states against Walter Mondale's promise to raise America's taxes.
Those three campaigns showed up all the way at the top of my graph. They were the three presidential elections with the most income growth in the months before the race. Mondale, McGovern, and Goldwater might have been bad campaigners, but they were running against impossible fundamentals. We understand elections in terms of candidates, but it seems awfully improbable that the three worst candidates happened to run in the three most impossible election years (though there is a theory that bad economic years attract bad candidates, as better candidates wait for a better shot).
Rather, this is a mistake that afflicts a lot of our election analysis: We think of campaigns in terms of people, but they're often decided by circumstances. "The media and the popular mind really think that candidates and the campaigns make a huge difference," says Michael Lewis-Beck, author of "Forecasting Elections." "But it's not as big a difference as the fundamentals operating behind the scenes every day." In some ways, that's comforting: Politicians are judged more on the condition of the country than on the elegance of their campaign.
But for the Obama administration, it's likely chilling: The economy is still weak, and there aren't 60 votes in the Senate for further stimulus spending. And even if there were, it is too late for them to make a major difference in the economy before November. Democrats needed to pass a bigger stimulus back in 2009, not in late 2010. What they do from here might help the economy, but it's not likely to affect their reelection.
If it's any comfort to the economic team, however, the political team isn't in any better shape: Focusing on the deficit won't make a difference, either. The White House may as well just govern.