POLITICS MARCH 1, 2012
Judging from recent headlines, things are looking up for President Obama. The Republican presidential nominees have been serially embarrassing themselves; the White House communications department has successfully focused its messaging on jobs and economic fairness; and consumers are feeling ever more confident about the economy.
But it's not just anecdotal evidence that suggests Obama’s re-election chances have improved—most of the polling data suggests the same. Obama has been running consistently ahead of his most likely challenger, Mitt Romney, in national polls—by an average of 4 points according to the Pollster.com website. Indeed, the closer you look at the numbers, the more reassuring the news: Obama, it seems, is well on his way to reconstructing the very coalition that elected him in 2008.
Consider these results from a recent Pew Center poll. In this poll, Obama is 8 points ahead of Romney, close to his victory margin in 2008 (7 points). But what is especially fascinating in this poll is its internals—how Obama is faring with key subgroups of voters. Start with minorities. Obama gets 93 percent of the black vote (he got 95 percent in 2008) and 79 percent of minorities overall (he got 80 percent in 2008). (The poll does not provide data on Hispanics, but the two most recent national polls of Hispanics give him 67 percent of these voters, identical to his 2008 performance.)
He also gets 44 percent of the white vote, compared to 43 percent in 2008. Moreover, if you break the white vote down by working class and college-educated, his performance is even more impressive. Among white college-educated voters he ties Romney 49-49, compared to the 4 point deficit he ran against John McCain, and loses white working voters by only 41-55, compared to his 18 point deficit against McCain.
These demographics are excellent news for Obama. But what of the longer term? Here we must depart the realm of demographics and current polls and look to election forecasting models. The best way to sort them out is not to focus on one particular model—there are so many to choose from!—but rather to look at the factors driving these models and what they have to tell us. The one factor that stands above the rest is economic performance in the election year. Moreover, there is general agreement that what matters most is change in economic performance, not the absolute level of performance. Thus, change in the unemployment rate in the election year is more important than the level of unemployment, GDP growth is more important than the size of the economy, and so on.
Recent economic performance by these standard measures has been fairly good: The unemployment rate fell from 9 percent in September of last year to 8.3 percent this January; job growth accelerated steadily from 112,000 last October to 243,000 in January; and the overall economy grew by 3 percent in the last quarter of 2011, up from 1.8 percent the previous quarter. Current projections of economic growth for 2012 are in the 2.5 to 3 percent range. Based on these projections, models that incorporate GDP growth tend to see an Obama victory this November. Nate Silver’s model, for example, gives Obama a 60 percent chance of winning, given a 2.5 percent growth rate this year and current job approval levels.
There are a couple of other big factors that I think are particularly worth considering. One is that Obama is the incumbent and that his party has only occupied the White House for one term. This is generally viewed as a positive for the incumbent party, since it damps down sentiment that “it’s time for a change.” Alan Abramowitz, who incorporates this factor into his model, predicts that, based on 2.5 percent growth and current job approval levels, Obama will get around the same share of the vote in 2012 as he did in 2008.
Another factor, especially relevant for Obama, is that voters may consider the economic performance from an incumbent’s first year in office when they think about performance in an election year. This possibility has been raised by Larry Bartels, who finds that bad first year economic performance—and Obama’s was very bad—has a positive effect on an incumbent’s election chances, presumably because voters cut the incumbent some slack for digging the economy out of hole even if election year economic performance is not stellar. If this effect is real, Obama, according to Bartels’ model, will receive a 7 percentage point increase in his 2012 margin due to the poor economy in his first year, considerably boosting his chances of being re-elected.
None of these forecasts indicate that Obama is a lock to win in November. Plenty can go wrong between now and then, particularly with the economy. But the factors underlying these forecasts suggest that the recent re-emergence of Obama’s coalition is no fluke. If current trends continue, there's a good chance that election day 2012 will look a lot like the one from 2008.
Ruy Teixera is a Senior Fellow at the Center for American Progress Action Fund and editor of America's New Swing Region: Politics and Demographics in the Mountain West, just out from Brookings.