This item by TDS Co-Editor William Galston is cross-posted from The New Republic.
In some presidential cycles, an incumbent’s reelection strategy doesn’t matter all that much. When the economy is very strong (1984), the incumbent wins big; when it’s very weak (1932), he loses even bigger. And when a party chooses a nominee seen as outside the mainstream (1964, 1972), it suffers a crushing defeat. It’s possible that one or more of these circumstances could prevail next year. The economy could over- or under-perform current projections; the Republicans could choose a nominee who’s too conservative or lacks credibility as a potential president. But it’s more likely that both the economy and the presidential nomination contest will yield results in the zone where strategic choices could prove decisive. In that context, two recent events are alarming, because they offer clues to what may well become President Obama’s reelection strategy.
The first was a Ron Brownstein interview with David Axelrod, who said that he saw Michael Bennet’s 2010 senatorial victory in Colorado as “particularly instructive.” As Brownstein noted, Bennet prevailed by mobilizing “enough minorities, young people, and socially liberal, well-educated white women to overcome a sharp turn toward the GOP among most of the other white voters in his state.” The second event was DNC chair Tim Kaine’s selection of educated, new economy Charlotte, North Carolina, as the site for the 2012 Democratic convention. In the process, he rejected three Midwestern finalists: St. Louis, Minneapolis, and, most notably, Cleveland.
Taken together, these clues suggest that the Obama’s 2012 campaign will focus more on the Democratic periphery–territory newly won in 2008–than on the heartland, where elections have been won and lost for the past half-century. This could turn out to be a mistake of epic proportions. Why? Because the United States looks a lot more like Ohio than like Colorado.
Here’s a snapshot from the exit polls in the 2008 election, versus the 2010 Ohio and Colorado Senate elections, broken down by education level:
4% of the 2008 national electorate had less than a high school education; that’s true of 1% of the 2010 Colorado electorate, and 3% of Ohio’s.
20% of the 2008 electorate had a high-school diploma but nothing else; the number dropped to 12% in Colorado and rose to 25% in Ohio.
31% of the 2008 electorate had “some college;” the numbers were 20% for Colorado and 32% for Ohio
28% of the 2008 electorate were college graduates; 39% of 2010 Colorado voters achieved that level, and only 24% of Ohio’s.
And the number for post-graduates is 17% for national 2008 voters, 28% for Colorado’s 2010 voters, and 15% for Ohio’s 2010 voters.
In other words, the 2010 Colorado electorate was a total outlier (67 percent with a B.A. or more), while Ohio was a near-microcosm of the national presidential electorate. Every Midwestern state for which exit polls are available looked pretty much like Ohio.
Now let’s look at the new Gallup state-by-state survey released on February 21. It shows that between 2008 and 2010, Democratic identifiers and leaners have declined as a share of the electorate in every state. The median loss was 6.1 percent. And every Midwestern state was at or above the median, raning from 6.1 percent in Minnesota to 9.2 percent in Wisconsin.
No wonder that Democrats lost 2010 Senate races in Wisconsin, Ohio, Iowa, Missouri, Indiana, light blue Pennsylvania, and even dark blue Illinois, or that they lost gubernatorial races in Pennsylvania, Ohio, Michigan, Wisconsin, and Iowa. Small consolation that Democratic gubernatorial candidates managed to scratch out victories in Illinois (by 20,000 votes out of 3.5 million cast) and Minnesota (by 9,000 out of 2 million).
The Midwest is home to large numbers of white working-class voters, who accounted for nearly 40 percent of all voters nationwide in 2008. Obama has never done very well with this group, losing them by 2 to 1 against Hillary Clinton in the primaries and by 58 percent to 40 percent against McCain in the general election. And they turned against Democratic candidates in the vast majority of 2010 House and Senate races.
As a recent Washington Post/Kaiser/Harvard survey reveals, white working-class voters are even more disaffected today. They are gloomy about the present and think it will be a long time before the economy recovers. They fear that the job market of the future will have little use for workers with their level of education and skills. Only 14 percent say that the president’s economic policies are working, and a large majority think that his administration isn’t doing enough to look after their economic interests. Remarkably, when asked which party better understands the people’s economic priorities, they give the Republicans a 14-point edge. Moreover, white working-class voters have long tended toward conservatism on social issues, nationalism in foreign policy, and hawkishness on defense.

Given all this, it’s tempting to look toward states like Colorado, where, Brownstein notes, Bennet won despite losing blue-collar white women by double digits and blue-collar white men by more than 2 to 1. His winning coalition was built on minorities, young people, and college-educated white women–precisely the groups Ruy Teixeira highlighted in an impressive analysis published in March 2009. He argued that the confluence of demographic, geographic, and attitudinal changes underway for decades heralded a “new progressive majority.” Not only was the political salience of social issues in decline, but also majorities of Americans had endorsed a stronger role for government, guaranteed health coverage, and clean energy. A principal driver of these shifts was the declining share of white working-class voters and the rising tide of minorities and highly educated professionals.
In a less noticed portion of his analysis, Teixeira offered a cautionary note. The white working class “is still an enormous group of voters–still larger than white college graduate voters–and there are good reasons to suspect that the exit polls may significantly underestimate the size of this group.” He went on to observe that “Progressives ignore that large a group at their peril … [their] already large deficit among the white working class–clearly their biggest political vulnerability–could easily become larger. If that happens, any fall-offs in support among their core and emerging constituencies could put the progressive majority at risk, despite continuing demographic trends in their favor.” This is a pretty good description of what transpired during Obama’s first two years.
The seductiveness of the Colorado model is obvious. But the consequences of succumbing to it could be dire. The last Democrat to win the presidency without prevailing in Ohio was John F. Kennedy. The electoral college math worked only because he won South Carolina, Georgia, half of Alabama’s electoral votes, and even Texas, thanks to LBJ’s presence on the ticket. None of these states is remotely within Democratic reach today. Ohio is more than a rich pool of votes; it is the closest state we have to a microcosm of the nation.
Barack Obama’s path to reelection runs through Ohio and the Midwest, not around them. And that means taking seriously the concerns of the voters throughout the region who deserted Democrats in droves last year–Americans unlikely to be moved by an agenda of high-speed rail, cleaner energy, and educational reforms that rarely seem to yield good jobs for themselves or their children. Instead, ratcheting up efforts to boost exports would work better; so would toughening our line against the excesses of China’s economic nationalism. In addition, Obama should acknowledge clearly that no region has gained more from the administration’s forward-looking restructuring of the U.S. auto industry; and none would have lost more if the administration had allowed it to collapse. Despite the continuing unpopularity of this initiative nationwide, the administration might as well take credit and tout it full-throatedly. Obama should also use his proposed Infrastructure Bank as the leading edge of a program to “rebuild America” that would create large numbers of good jobs that can’t be exported. And white working-class skepticism about the first two years of the administration’s economic agenda gives the president one more reason to participate in broader bipartisan discussions about our fiscal future.
It’s easy to claim that the administration need not choose between a Colorado strategy and an Ohio strategy. At the level of tactics, that may be true. If the Obama campaign raises as much money as, or more than it did, in 2008, it will be able to organize and advertise everywhere. But at a deeper level, that doesn’t hold. The administration can have only one agenda. If its policy choices over the next 18 months are directed principally toward mobilizing voters from its base, it will pay a price among the millions of others without whose support Obama would never have been elected.