One of the more remarkable phenomena of U.S. politics is the relatively low utilization rate of the findings of political scientists by practitioners and journalists in the field. Until fairly recently, as a matter of fact, there was astonishingly little interaction between the two worlds of the academic study of politics and its day-to-day analysis and operations. One reason for the rise of the ubiquitous think tank industry was this gap between the thinkers and doers, which extended to a host of public policy areas.
That’s changed a fair amount in the very recent past, and today political scientists are a robust minority in the chattering classes, some via conventional news outlets and some in blogs (prototypically The Monkey Cage, which we regularly cross-post from here).
But the gap remains pretty large. Today Nate Silver fired a shot across the gap with a strong empirical challenge to the “economic fundamentalists” of the political science community: those who believe that for all the noise, campaign activity, money, advertising, and partisanship involved, presidential elections are essentially decided by economic conditions in the country, and careful scrutiny of key economic indicators makes it possible to predict electoral outcomes with some precision.
I won’t attempt to summarize Nate’s massive post, which he appears to have generated as a byproduct of his upcoming book on electoral forecasting. But his bottom-line is pretty harsh:
The “fundamentals” models, in fact, have had almost no predictive power at all. Over this 16-year period, there has been no relationship between the vote they forecast for the incumbent candidate and how well he actually did — even though some of them claimed to explain as much as 90 percent of voting results.
He goes on to made a less damning statement about economic models that take into account some empirical campaign-dependent factors like polling data, and concedes, of course, that economic conditions are generally an important factor in presidential elections, if only as a strong headwind or tailwind for incumbent and opposition parties.
You should read the whole thing for its intrinsic merits, but I bring it up here mainly to make a point about the wildly varying assumptions that people bring to political discussion and analysis. Many, many people are election “determinists” to one degree or another, believing that factors beyond the framework of actual campaigning, candidates, issues, messages, and events largely control outcomes. Some are like the academics that Nate critiques, holding very strong views based on “models” of past elections; others believe that shadowy forces, usually financial, control politics and elections; still others believe that relatively immutable factors like demographics are all-important. Determinists of every stripe tend to be impatient or even angry at people who pay attention to “ephemeral” factors, events and people in the election process, often accusing them of deliberately lying about such factors to keep themselves employed, provide something controversial to talk about, or even to hide “fundamental factors” from the public.
At the other end of the spectrum, particularly in political journalism, we have people who pretty much ignore fundamentals, and treat the character and abilities of candidates and their staffs, the thrust-and-parry of campaigns, the salience of particular issues, and the battle for persuasion of voters, as essentially non-determined phenomena that can only be understood and assessed via close “insider” inspection. At the extremes, such observers often seem to behave as though they think all voters are swing voters, every day of a presidential campaign is potentially critical, and that candidacies are highly purposeful and relatively autonomous operations that “win” or “lose” according to “rules” derived from military conflict, sports history or game theory.
I’d guess an impressive percentage of the unproductive arguments in politics come from dialogues of the deaf between observers holding different positions along the spectrum of assumptions about the basic nature of politics, and the very different optics these assumptions encourage or even dictate.
This is all to say that regular examination of our assumptions should be a regular feature of political talk. The best way to do that is to acknowledge our biases about what matters most and to subject them whenever possible to empirical verification and adjustment. I try to do that here, but probably ought to do it more. I’d encourage readers to do the same. At some point, if you really do believe politics are essentially controlled by iron laws that rarely bend, then political discussion becomes pretty much an exercise in expressing contempt for anyone who hasn’t read the right textbooks or examined the right indicators. If, on the other hand, you think presidential campaigns are like sports playoffs or video games or battles between roughly equal military forces, you should at least be aware why the whole spectacle is a lot less engaging and fun to people who don’t look at them that way at all.
There’s certainly room for widely varying perspectives, but we’ll all get along better if we try to show our cards–or if you prefer, our paradigms.
