Yesterday, President Obama spoke at a large rally at the University of Wisconsin that was intended to help rally the Democratic base for the midterm election. But will he and his party be able to narrow the enthusiasm gap with Republicans? The indicators aren't encouraging.
One possible obstacle was suggested recently by The New Republic's Jon Chait, who suggested that Democrats can't sustain enthusiasm when their party holds the presidency like Republicans:
The Democratic base tends to lose interest in the threat of right-wing politics when their party holds power. Republicans, I'm guessing offhand, have had more success energizing their base during Republican rule. (Anybody want to quantify this?) Specifically I'm thinking of the 2002 and 2004 elections, which featured revved-up Republican bases despite total GOP control of government.
My seat of the pants analysis is that this reflects a psychological difference between the left and the right. The liberal coalition is more ideologically diffuse and attracted to individualism. Sometimes you see left-wing splintering at the end of periods of Democratic control -- 1948, 1968, 2000 -- but more often it's simply harder to make liberals understand the urgency of preserving their party's control of power against a hypothetical threat. Conservatives, by contrast, may find the idea of rallying behind a leader more attractive. Liberals were obviously very enthusiastic about the historical nature of Obama's election, but the enthusiasm has waned since. The conservative cult of personality around George W. Bush actually seemed to peak in 2004.
Is this claim supported by the data? Gallup has asked survey respondents whether they are more or less enthusiastic about voting than usual in every election since 1994. In previous years, I use the last available poll before the general election. However, Gallup changed their question wording this election cycle for the enthusiasm question so I rely on the June 11-13, 2010, survey (the last using the old wording) to make sure the results are comparable with previous years (the current estimates of enthusiasm using the new wording are very similar).
Using this measure, I calculate net enthusiasm by party (% more enthusiastic - % less enthusiastic) and then take the difference between parties, constructing a measure of the net enthusiasm advantage for the president's party.* (This abstracts away from features of the election that may increase or decrease enthusiasm in both parties.) The results are more ambiguous than Chait's claim:
Democrats have been less enthusiastic relative to the other party in the first midterm under both Clinton and Obama than Republicans were under Bush, but it's important to keep in mind that the 2002 election is an outlier due to 9/11. By comparison, 1994 and 2010 were extremely unfavorable electoral environments. In more favorable conditions (principally, a booming economy), we see that Democrats were relatively more enthusiastic for Clinton in the 1996-2000 elections than Republicans were for Bush in 2004-2008. It's unlikely that Democrats will close the enthusiasm gap with Republicans in this election -- the conditions are just too unfavorable -- but the historical record doesn't indicate that they are incapable of enthusiastically supporting a Democratic president.
* I relied on Gallup's tabulation of enthusiasm by party (including leaners) when available. I calculated results myself for 1996 and 2000 using survey data archived by the Roper Center. Note: The 1996 survey includes "the same" as an option for the enthusiasm question; in other years, it was only recorded if volunteered by the respondent.