A common elections adage is that Republican candidates should “pray for rain” on election day. The logic is that rain suppresses voter turnout among unlikely voters (who, in turn, disproportionately identify as Democrats). The other day Mike Huckabee extended this logic to the upcoming Iowa primary, telling Chris Wallace of Fox News:
if the weather is bad and it’s real tough to get out, Ron Paul would win… Ron Paul has an exceptional organization there. And it very well could be that he could end up winning because of the extraordinary devotion of his followers.
That seems plausible. But is it true? A 2007 paper by Brad Gomez, Thomas Hansford, and George Krause suggests it is (an ungated version can be found here). From the abstract of their paper examining the “weather-turnout thesis”:
We argue that much of the intuitive appeal of the [Weather-turnout] thesis results from it comporting well with both socioeconomic status and rational choice models of voter turnout. We also address the theoretical underpinnings of the partisan bias conjecture associated the weather-turnout thesis, which contends that if bad weather does affect voter turnout, the resulting suppression of voters may benefit one party over the other. We examine the effect of weather on voter turnout in the over 3,000 U.S. counties for 14 U.S. presidential elections (1948–2000)—the most exhaustive empirical test of the weather-turnout thesis to date… In the end, we find that bad weather (rain and snow) significantly decreases the level of voter turnout within a county. We also demonstrate that poor weather conditions are positively related to Republican party vote share in presidential elections.
In particular, the results of their impressive analysis reveal that Republican presidential candidates added about 2.5% to their vote share for every one inch of rainfall above normal. Now the usual caveats about generalizability apply, but if we extend this effect to ideology (rather party identification) and primary elections (rather than general elections), it seems reasonable to me that bad weather will suppress turnout among Mitt Romney voters. The logic is that ideologically extreme primary voters are more likely to cast their ballot despite inclement weather compared ideologically moderate primary voters.
Now if this logic holds then, yes, rain in Iowa may help Ron Paul’s chances as Mike Huckabee claimed (of course it may help Gingrich for the same reasons). But does this mean that rain will be the deciding factor between a Ron Paul victory and a Mitt Romney victory? I’m skeptical for two reasons. First, though the most recent polling shows Gingrich, Romney and Paul in a statistical tie, my intuition is that the effect of rainfall in primary elections (even significant rainfall) is lower than that reported by Gomez, Hansford and Krause’s analysis (2.5% in their study of general elections). Primary voters are, after all, more ideological and (I’m assuming) more motivated to turn out despite inclement weather compared to general election voters. Moreover, a 2003 simulation by Citrin, Schickler and Sides (see an version here) shows that even full turnout rarely sways election outcomes (though there is a Democratic advantage associated with greater turnout). The second problem is that rain is uncommon in Iowa in January. According to one not-so-scientific source, the average rainfall Iowa for the entire month of January is only 1.1 inches. Moreover, an early forecast from Weather.com (left) predicts a whopping 0% chance of rain on Tuesday (and only 10% on the 2nd and 4th).
So could rain affect the Iowa Caucus? That seems reasonable to me. But will it affect the outcome? Unlikely.