Home / The Empty Calculi of Electability

The Empty Calculi of Electability


Drezner writes

This process meant that the Democrats ran Walter Mondale, Michael Dukakis, Al Gore, and John Kerry in November. There’s no way that any politico can justify a process that delivers that set of outcomes.

…which reminds me of this excellent post by Lance Mannion on the ghost of John Kerry:

The ghost of John Kerry haunts many of the various discussions about “electability” I’ve read on the web. There seems to be a consensus that Kerry was the nominee in 2004 because Democrats thought he was the most “electable.” There’s also general agreement that the Democrats were out of their minds on that one because Kerry turned out to be highly unelectable… In 2004 Iowa and New Hampshire decided the nomination for the rest of us by handing Howard Dean his hat. Maybe Iowans based their decision on Kerry’s supposed greater “electability.” The rest of us were just along for the ride, hoping that Kerry was in fact electable.

But the proof offered that he wasn’t actually electable is simply that…he wasn’t elected. There’s a sense out there that Kerry should have won. This idea seems to have two meanings in one. Kerry should have won because all the advantages were his and he should have won because in a fair and just universe George W. Bush would have been thrown out of the White House on his ear and by losing to that jamoke Kerry committed a sin and a crime against nature and the nation.

Behind both senses is the belief that, no matter how electable Kerry was, George W. Bush was indisputably not re-electable. Which brings us back to this: The fact that Kerry could not get elected over an obvious loser like Bush is proof that Kerry was unelectable.

Another way to think about this is to try out a counterfactual. What if Gary Hart had prevailed over Walter Mondale in the 1984 Democratic primary, then had gone on to win, say, eight states against Reagan in the general election? Is there a shadow of a doubt that Hart would have become a national joke (four years early, as it turns out), and would have joined the pantheon of utter Democratic failures that Drezner recounts? Moreover, is there even any doubt that supporters of Walter Mondale would be in the first rank of those making such claims?

Political science has some tools for differentiating between a genuinely bad candidate and a candidate in a genuinely difficult campaign, but those tools have failed to trickle into either elite media discourse or the popular conversation. Consequently, we got comments like Drezner’s, which assert that because a group of candidates lost, they must all share the dispositional characteristic of being poor candidates. This is particularly irritating from Drezner, who really should know better, but it’s a distressingly common theme in conversations about past elections.

Lance successfully demolishes much of the myth about Kerry in 2004, particularly that part that might be characterized as “If I and my immediate circle of friends really hated George Bush and thought he should be impeached, then it’s obvious he was a weak candidate, and consequently Kerry was even worse for not beating him”. I think that the other part of this claim, which runs something like “If only we had nominated Dean/Edwards, we would have won this thing,” is just as weak. Maybe Howard Dean could have beaten George Bush by being aggressive and “ballsy”, but then again when he ran assertively to the left in the subset of the population most likely to respond to an assertive leftish campaign he lost 48 states. For the electability of John Edwards I have even less regard; he’s always been attractive because of his perceived ability to appeal to moderate white men, but he’s now demonstrated in two different campaigns that he lacks the ability to appeal to very many people who aren’t moderate white men. Similarly, it’s hardly an endorsement of his electability that he felt vulnerable as an incumbent Senator in his home state, or that he was so badly beaten this time around by a guy who was running for the Illinois state Senate when Edwards was running for Vice President.

I’d also like to think that this cycle has brought the “momentum” theory of primary elections into deep question. The most common narrative of John Kerry’s nomination victory in 2004 is that by winning Iowa he acquired sufficient momentum to roll over the rest of the candidates, irrespective of his or their actual merits. It’s worth thinking, however, about how rarely such a thing actually happens. Of the major party nominees since, say, 1980, how many can we actually say won because of momentum, rather than because they had larger, better organized campaigns from the start? Really, are there ANY other extant examples of what we think happened in 2004 (the nominee rolling over candidates of similar capability because of victories in Iowa and New Hampshire)? And doesn’t this year demonstrate that momentum is rather secondary to the establishment of a good campaign around a candidate that people like, such as Hillary Clinton or Barack Obama? Another way of putting this is that Kerry may have won because of the Iowa momentum in 2004, but it would be a rather unique event in the history of modern primary campaigns, and even if so it says much more about the weakness of his opponents than the contingency of his candidacy.

I guess my point is that the problem with electability issue is even more intractable than it seems. It’s commonly observed now that electability is hard to determine in advance, which is why we should be at least a bit careful about claims that we should prefer one candidate to another because of their chances in November. It’s even worse than that, because we can’t even say all that much (using the tools we’ve been using, anyway) about electability in an election that has already happened. Winning or losing an election really isn’t much better of a determinant of candidate quality than winning or losing a baseball game is determinant of pitcher quality. And it’s doubly unfortunate that we fall into such traps because the language we use to discuss these questions tends to be more dispositional than situational. To return to the counterfactual above, Gary Hart would have lost 42-8 to Ronald Reagan because he was a terrible candidate (dispositional) rather than because he was in a situation where Democrats were unlikely to win (situational).

But then again, that might have saved him some later difficulty.

  • Facebook
  • Twitter
  • Google+
  • Linkedin
  • Pinterest
It is main inner container footer text