As most of you know, John McCain and Barak Obama will meet this Friday in a much anticipated and sure to be much watched presidential debate, the first of a scheduled three such debates. Media pundits undoubtedly will resurrect past debates as evidence that these events can significantly influence the outcome of the presidential election. They will cite the first Kennedy-Nixon debate in 1960, where Nixon – lacking makeup and coming off an illness – looked ill at ease while Kennedy appeared relaxed. In 1976, Gerald Ford inadvertently “freed” Poland from Soviet domination – a gaffe that some say cost him the debate and the election. In 1984, Ronald Reagan’s shaky first debate with Walter Mondale resurrected questions regarding his age. In 1988, Michael Dukakis’ bloodless reply to a hypothetical question about how he would react to the rape and murder of his wife supposedly cost him a good deal of support. Similarly, pundits claimed thta George H. W. Bush spent too much time looking at his watch and appearing disinterested in his debate with Clinton and Perot in 1992. In 2000, Al Gore’s sighing and excessive eye-rolling cost him crucial support, according to many experts. I could go on, but these anecdotal snippets suggest that presidential debates are frequently game-changing events. But are they?
The answer is: not usually, at least historically. As evidence, I refer you to a study by Tom Holbrook (see Holbrook data) who concludes that there is very little change in the underlying support for presidential candidates after a debate. Holbrook calculates that across the 13 televised presidential debates he researched in the period 1988-2004, the average swing in support for the incumbent party candidate was about 1%. There are exceptions of course – George H. W. Bush lost 2% in the second of the three presidential debates in 1992, and his son George W. Bush lost 2.2% in support after the first debate with John Kerry in 2004. Looking more broadly at changes in support starting one week before the first debate to one week after the final debate, Holbrook finds a bit more change. The most extreme change occurred as a result of the three 2000 debates, when Al Gore lost a cumulative 3.5% across this time period. However, it is hard to know just how much of this drop in support was attributable to his debate performance, and how much was caused by other factors.
Part of the reason that pundits think debates matter is because the media acts as if they do. When Ford liberated Poland in the 1976 debates, viewers understood what he was trying to say – that the Polish people, despite Soviet domination, retained their sense of independence. Right after the debate, most viewers surveyed thought Ford had won. After the media began weighing in, however, Ford’s remarks were cast in a different light, and surveys now began to show that he lost the debate. So one needs to be careful in separating out the influence of the debate from that of the media coverage. In this regard, expect both the McCain and the Obama camps to engage in an ongoing effort to spin the debate, beginning by lowering expectations for their own candidate, and raising them for their opponent. Then, during the debate, they will send out comments to underscore or clarify what the candidate just said. Finally, immediately upon concluding the debate, members of both camps will hustle to the “spin room” and explain why the opposition candidate just lost the election while failing to lay a glove on their own guy.
If history is a guide, however, the spin emanating from both camps will largely cancel one another, and there will be very little change in the underlying support for either McCain or Obama after Friday’s debate. But is the past history of debates relevant for Friday’s event? McCain, for the most part, is a known quantity to most voters, but Obama is not. For this reason, I think Obama’s potential upside in this debate is greater than McCain’s. The fundamentals in this campaign favor the Democrats, but there is a good deal of uncertainty re: Obama in terms of both his experience and his readiness to be president. He needs to calm those fears by coming across as “presidential.” If he does so, that may convince voters that despite his relative inexperience, he is ready to be president. The closest historical analogy in terms of debates that I can recall would be 1980, when the fundamentals favored the Republicans, but many voters were unsure of Ronald Reagan’s readiness to be president. For the most part, he calmed voters’ fears with a much understated, reassuring debate performance. Obama needs to do the same. He “wins” then, just by appearing on stage with McCain and not making any huge mistakes. For McCain, the key – as it was for Reagan in 1984 – is to show that he is still physically and mentally sharp, and to counteract the whispering campaign about his temperament. But, as the trailing candidate in terms of fundamentals, he also needs to be more aggressive in going after Obama on the issues, particularly since this first debate is supposed to focus on foreign policy, which plays into McCain’s policy strength.
For both candidates, this is one of the few remaining instances where they are likely to have a national audience. Just how many people will be watching? In part II of this post, I will present data on the size of the viewing audience for past televised debates.