The Impact of Debates on Presidential Elections

My colleague at OTB, Steven Taylor, considers the impact of the debates in the last six presidential elections:

Now, on the one hand, the number of observations is small (6 elections and 16 debates), so no, we cannot draw ironclad inferences from the figures. On the other hand, as Holbrook notes, “Across all 16 presidential debates the average absolute change in candidate support was just less than 1 percentage point.”

I think there are a couple of problems in comparing last night’s debate with those from 1988 to 2008. First, in only three of those, 1992, 1996 and 2004 [ed. corrected], was an incumbent running for re-election. Second, the president’s re-election campaign appears to be unique in that he’s de-emphasizing his record in favor of emphasizing that he’s a better candidate than his opponent. In that respect it more closely resembles his primary campaign against Hillary Clinton or his successful presidential run against John McCain.

The prevailing wisdom is that re-election campaigns are always referenda on the incumbent. Assuming otherwise is going against the evidence of history. The conventional way of stating this going along the lines of “No president since FDR has been re-elected with an unemployment rate above 8%” or something along those lines.

The president is, in effect, running as a challenger rather than as an incumbent. Will he be allowed to do so? If not, performances in the debates by Mitt Romney that suggest that he’s a viable alternative to Barack Obama may provide a lot more than “less than 1 percentage point”. In an election as close as I suspect that this one will be that could be significant.

5 comments… add one
  • Icepick Link

    First, in only two of those, 1996 and 2004, was an incumbent running for re-election.

    Forgot 1992.

  • Icepick Link

    Also, I think looking at the “bump” isn’t really the correct metric. The debates are when a lot of undecided folks decide whether a non-incumbent is someone they can really vote for. It can also make a candidates support more solid than it was, insuring a better turnout of the faithful. (Or it can lessen the support of the faithful, obviously. I think that’s one of the things that happened to McCain in 2008.)

    Reagan needed to look like he wasn’t going to blow up the world. He succeeded. Dukakis and Kerry needed to look like they were up to the foreign policy challenges of the time. They failed.

    Last night Romney needed to look like he could handle the pressures of the economy. It looks like he succeeded. Also I’ve seen comments here and there (I’ve been having another bought of insomnia) include a fair number of Anybody But Obama (ABO) types stating that they’re now moving from voting against Obama to voting FOR Romney. That will help with turnout and with efforts to convince others to vote Romney. At the moment it seems like Romney has achieved everything he could have hoped to achieve last night.

  • TastyBits Link

    Most analysis is missing the tepid voters. They decided to vote for one candidate or the other, but they are not wedded to him. Bill Clinton swung these voters to President Obama. Romney just swung these voters to him.

    The election will be determined by Bill Clinton and Mitt Romney. These voters are solid Romney votes, and it is going to be difficult to dislodge them. Romney will need to screw-up badly, and Clinton will need to make President Obama a valid choice.

    Make no mistake this was a game changer. It was not the substance, but that was important. All the fact checking is going to be pointless, and it is going to hurt rather than help.

  • Andy Link

    The problem is Steven used the average.

  • I guess that’s what I’m saying, Andy. I’m concerned about fallacies of aggregation.

Leave a Comment