electionate

As discussed earlier this week, there’s evidence of an emerging pro-Obama bias in IVR polling, mainly among white voters. While this pattern seems relatively clear in PPP and SurveyUSA, it doesn’t show up in Rasmussen’s polling, which generally provides relatively Romney-friendly numbers. However, Rasmussen’s unique methodology leaves open the possibility that there might be an IVR-related bias among white voters, since Ramussen’s likely voter model, party weights, and different state v. national methodologies, might conspire to conceal or reverse the effect.

Today, Rasmussen published a relatively innocuous poll of Nebraska, showing Obama at 39% of the vote. That’s quite reasonable without context, but it connected some disparate mental dots into an emerging pattern. Recent Rasmussen polls have shown Obama doing relatively well in white states, including Montana and now Nebraska, but also in Ohio and Missouri. In contrast, Rasmussen’s recent polling in North Carolina and Virginia looks great for Romney.

Unfortunately, there haven’t been too many Rasmussen polls over the last month, but the pattern is relatively persistent other than Nevada. When there are so few polls, one outlier can make a huge difference – r^2 of just .32 with Nevada, but .79 without.

Of course, Nevada differs from North Carolina and Virginia in an important way. While there are a similar number of white voters in North Carolina, Virginia, and Nevada, the first two states have a far larger African American population than the Silver State. Although the correlation between Obama’s drop-off and the number of African American voters in 2008 is not quite as strong as the relationship between white voters and Obama’s drop-off excluding Nevada, it does a much better job of reconciling Nevada.

This piece only considered Rasmussen polling over the last month. Ultimately, the pattern almost entirely rests on two recent polls in North Carolina and Virginia. More polling is needed before pinning down this pattern.

That said, this pattern seems relatively consistent with other observations, including Obama’s tendency to do well among white voters in IVR polls and the tendency of most polling firms to show Obama underperforming among non-white voters. The possibility that likely voter models project low African American turnout could also be the issue.

As noted many times before, while the President’s approval rating is extremely low among white voters without a college degree, it is still at 2008-levels among white voters with a college degree. From a certain perspective, this might seem counter-intuitive. The Obama Administration has pursued a relatively populist policy agenda and many of those same efforts, as well as the decision to table the debt commission proposal, would seem likely to alienate ideologically moderate but fiscally conservative suburban voters. Even if Obama’s policy decisions were consistent with the predilections of well educated voters, one might still expect Obama to be faring worse than in 2008, when a highly advantageous political climate bolstered Obama’s standing.

I can think of many hypothetical explanations for Obama’s resilience among college educated whites, but one seems powerful and sensible: adverse economic conditions have not been felt equally, and as a result, economic confidence is higher among affluent, educated, professionals than high school educated, working class voters.

While I’m not aware of any poll that disaggregated economic confidence by educational levels, it fits the available data quite well. Geographically, economic confidence is strongly correlated with educational attainment and so-called creative class workers, even more so than income. That shouldn’t be incredibly surprising, since the recession has disproportionately hurt less educated workers. The unemployment rate among high school graduates is twice that of college graduates, and many professional workers enjoy high levels of job security. The relationship between the President’s approval rating and economic confidence requires no elaboration.

If Obama’s relative strength among college educated white voters and attendant weakness among non-college white voters is due, at least in part, to different perceptions of the economy, there are important consequences for the campaign. Suppose, for instance, that voters who lack confidence in the economy might not only be less likely to vote for Obama, but more likely to focus on economic issues. Similarly, what if those with confidence in the economy might not only be more likely to maintain their support for the President, but more likely to consider cultural questions?

That might help explain campaign behavior surrounding two of the most recent foci of the campaign: gay marriage and Bain Capital. The Obama Campaign’s two messages seem aimed at different segments of the electorate. Same sex marriage is appealing to educated, secular independents, but it is also deeply unpopular with religious or less-educated voters. Conversely, attacks on Bain Capital risk alienating affluent voters, but appeal to the concerns of disaffected workers who resent growing economic dislocation and the new management techniques that increase efficiency at the expense of security.

Although those strategies might seem conflicting, differing levels of economic confidence could allow the Obama campaign to have their cake and eat it too. Deep dissatisfaction has already cost Obama considerable support among white working class voters, depleting the pool of Obama/anti-gay marriage voters who the Obama campaign might wish to avoid alienating. At the same time, the unrelenting focus of economically disaffected voters on economic questions might reduce the risk that anti-gay marriage swing voters cast their ballots on cultural issues. Attacks on Romney’s tenure might prove far more salient to their decision.

If the reverse is true for well educated voters, and relative satisfaction with the state of the economy leaves more open to casting their ballots for Obama generally and on social issues more specifically, then gay marriage should reenforce, if not improve, Obama’s standing. While attacks on Bain Capital might seem to pose risks with this critical constituency, Obama has long adopted populist rhetoric and policies, and it is hard to imagine that Obama’s strategy risks further losses, even if it might prevent socially moderate but economically conservative voters, say in the Charlotte suburbs, from switching sides.

In this light, Romney’s attempt to shift the debate back to economic issues also make sense. While greater economic confidence might relate to Obama’s strength among more educated voters, Romney can compensate with big gains among voters without a college degree, and their greater sense of dissatisfaction is consistent with the huge opportunities evident in most polls. Even if those economic dissatisfaction suggests that social issues might not be an especially effective tactic in persuading voters with unwavering focus on economic issues, deep economic dissatisfaction means that the Romney campaign doesn’t need social issues to win non-college voters. The critical task for Romney is to exploit economic dissatisfaction by demonstrating his ability to turn around the economy. For these same reasons, Obama’s shift to Bain Capital makes sense as well.

Uneven perceptions of the economy correlate with uneven perceptions of Obama’s job performance. If economic confidence is related to both Obama’s current levels of support and the willingness of voters to consider cultural issues, then Obama’s socially liberal and populist campaign strategy makes sense. Low economic confidence has already cost the Obama campaign millions of less educated, anti-gay marriage supporters. Although gay marriage poses risks with less educated whites, the same economic forces responsible for their departure from Obama’s coalition might insulate Obama against further losses if they remain focused on pocketbook issues. Well-educated voters have largely stayed in their 2008 camps, and higher levels of economic confidence might permit them to more seriously consider social issues, even if the vast majority still consider the economy the most important issue. While it is impossible to prove any of these points without more comprehensive data, all are consistent with Obama and Romney’s campaign strategies.