Cracks in Obama's Coalition or Conventional Wisdom?

There have been countless articles this week interpreting the electoral tea leaves. One common interpretation of the results is that there may be some cracks in the coalition that ushered Obama into office. The stories generally then conclude that these cracks may serve as a warning sign for Democrats in the future.

An example of this type of article appeared the day after the election in the Washington Post in an article by Dan Balz. In assessing the results of the New Jersey and Virginia gubernatorial races, he argued: "Neither gubernatorial election amounted to a referendum on the president, but the changing shape of the electorate in both states and the shifts among key constituencies revealed cracks in the Obama 2008 faction."

In particular, McDonnell's win over Democratic challenger Deeds ended eight years of Democratic control of the governorship in Virginia, while Republican Chris Christie unseated Democratic governor Jon Corzine in New Jersey.

As evidence of a fracturing coalition, Balz pointed to the fact that Independents, who overwhelmingly broke for Barack Obama in the 2008 election were key to the Republican wins in Virginia and New Jersey. Exit polls in Virginia showed that McDonnell led Deeds by a 2 to 1 ratio among Independents. Poll results leading up to the election also revealed that Independents were becoming disaffected with some of Obama's policies.

The conventional wisdom among political scientists is that voters punish incumbents during poor economic times, and this is especially the case among those who are not strongly committed to a political party. Balz's account of Independents could be correct or at least correct for some Independents. Some may have punished the Democratic candidates for discontent with Obama's handling of the national economy. This would be an indicator of cracks in Obama's coalition, at least among Independents in these two states.

There is another way to interpret the behavior of Independents. Some independents may have only been focused on state economic conditions and the race in that state. In an article in Politico, Jonathan Martin revealed information that Creigh Deeds' pollster, David Petts, was actually advising Deeds to keep a distance from Obama, which would have separated national politics from the state level race. If some Independents were only looking at who the incumbent party happened to be in their state, then they would punish the incumbent party, which happened to be the Democratic Party in each state. If the incumbent party was instead Republican, then they may have supported the Democratic candidates. According to this interpretation, Independents have not necessarily shifted their behavior from 2008 to now since they are consistently voting against incumbent parties in the face of poor economic performance.

It is likely that both types of interpretations were present among Independents voting in Virginia and New Jersey, especially since these are state level races. If we look ahead to the 2010 congressional elections, it is probable that Balz's interpretation may end up carrying more weight since congressional races are national level affairs.

As another indicator of cracks in Obama's coalition, Balz discussed the fact that turnout was much lower among voters under 30 compared to their presence in the 2008 race. In Virginia, they only accounted for 10 percent of the electorate compared to the 20 percent they represented in the presidential election. If the youth continue to stay home, then it may pose a problem for Democrats in the future.

Conventional wisdom would have predicted that turnout among the youth would be lower than it was in the 2008 race. Turnout is typically much lower when there is not a presidential race on the ballot and the voters who show up at the polls in these contexts tend to be older, more partisan, and higher in socio-economic status. It is likely that voters under 30 will again show lower turnout in 2010 compared to 2008, but that does not mean that they will necessarily stay home in 2012. If they were mobilized into politics in part by Obama's presence in the race, then they may turnout just as strong in 2012. In short, Obama's coalition may not necessarily translate into the Democratic coalition.

While some have made a fairly convincing case that there may be warning signals for the Democratic Party, I would caution against reading too deeply into the electoral tea leaves. First, there are not really enough cases to make broad generalizations. Second, there are several alternative explanations we can use to explain some of the outcomes.