Tuesday, November 22, 2005Was the election of 2004 winnable? John Kerry has come in for a lot of criticism for his performance in the Presidential race. As usual, the party lost and the candidate took the blame. The question remains whether this attribution is fair.
David Gopoian's answer is no. Using extensive poll data, he argues that John Kerry performed as well or better than any other candidate could have - including John Edwards, Hillary Clinton and even the Big Dog, Bill Clinton himself. Kerry successfully united the Democratic base, but Bush's support among Republicans was so fanatic that it was impossible to unseat him. No Democrat could have done better than Kerry.
I concur with Gopoian that Kerry was the best of the candidates running in 2004 and that his campaign was better than it has been given credit for. I do have some real objections to Gopoian's assertions, however. First, if the Democratic base was as united as the Republican base (as Gopoian's piece implies), then party unity is a nullified factor - independents would have decided the race. In other words, Kerry could have won if he persuaded independent voters, unless one supposes that there are more Democrats than Republicans (which there aren't).
Which brings me to the second point: that Gopoian's method is characteristic of empirical political science practices - practices I'm suspicious of. His study examines voter preferences as they were in November 2004, after the Presidential race was over. All of the moves made by both parties and the frames surrounding them had been set. Now partisan identity is notoriously tied to perceptions of the party's standard-bearer, perceptions that had already been firmly established by a year of campaigning. What Gopoian's analysis cannot capture is the counterfactual - what if a different nominee had been selected, or Kerry had run a different campaign? Asking people what they feel about a party or a candidate after a campaign tells you nothing about what they would have felt had the candidates behaved differently.
Campaigns matter. Yes there are powerful structural features that shape elections. If the underlying political dynamic is overwhelmingly in favor of one party or another, then the campaign itself is pretty irrelevant. But in a close election, campaigning is hugely important. Voters, particularly independent voters, are open to persuasion. In 2004 I believe that there were constituencies willing to support a Democrat but who supported Bush. I still think Kerry was the best candidate last year - I just think he failed to make the most persuasive case to swing voters.
So no, we weren't doomed. We just failed.
P.S. More Dynasty Watch. Via Daily Kos, the son of former Senator John Culver is the leading Democratic candidate for governor of Iowa. Boo!