No, swing state economies don’t matter more
posted at 9:11 am on June 12, 2012 by Karl
After a hiatus of a few months, the establishment media is again promoting the notion that improving swing state economies boost Pres. Obama’s reelection odds. Here’s Matt Negrin at ABC News:
Because the United States still holds elections under the arcane rules of the electoral college, only a handful of states will determine who the next president is. And in those states, the state of the economy is much more important, politically of course, than in the old party standard bearers.
The good news for Obama is that five key swing states have unemployment rates that are below the national average, which is 8.2 percent as of last week. New Hampshire is at 5 percent; Iowa is at 5.1 percent; Virginia is at 5.6; Ohio is 7.4; and Colorado, 7.9.
Three swing states, however, have unemployment rates higher than the national average: Florida, with 8.7 percent; North Carolina, with 9.4 percent; and Nevada, with 11.7 percent, the highest in the country.
But in the states at the high end of unemployment, a downward trend could work in Obama’s favor.
Politico’s Ben White recently sounded a similar theme:
President Barack Obama has watched a stubborn national economy dim his reelection chances.
But the real economic battle for election 2012 will take place in the states.
Some key battlegrounds are doing better than the nation as a whole. And even if the president’s policies didn’t cause the improvement, it could help his case that the economic outlook has brightened under his stewardship.
Bloomberg’s Mike Dorning is also riding this narrative:
From extra shifts at auto and steel plants in Ohio to office buildings rising in Northern Virginia, the geography of the U.S. economic rebound is providing an edge to President Barack Obama’s re-election.
The unemployment rates in a majority of the 2012 battleground states are lower than the national average as those economies improve. Coupled with the growth of adult minority populations in those states, the trends create a higher bar for presumed Republican Party presidential nominee Mitt Romney in his quest to unseat Obama.
Both White and Dorning rely on Xu Cheng, a senior economist at Moody’s Analytics, which has a forecasting model based solely on state-by-state economic data and past voting behavior. However, the model’s decent prediction of the 2008 outcome is less impressive when you consider that 2008 was not a difficult election to forecast. Furthermore, most of the media discussion focuses on state unemployment rates, when the rate itself is not predictive of election outcomes; the rate of change of unemployment is a better indicator (I doubt Cheng makes this mistake). More broadly, political scientists tend to find little evidence state economies matter, above and beyond the effect of the national economy (and voter perceptions thereof). Indeed, Andrew Gelman has found that partisan swings have become more uniform across states in recent elections.
The cheap shot here would to be to claim media bias, as establishment outlets look for good news for Obama amid a limping economy. But it’s just as easily explained by a superficially plausible hypothesis: “It’s the economy” + “the Electoral College matters in a close election” = “swing state economies matter.” And once Xu Cheng gets mentioned once, it’s just a matter of reporters sticking his phone number in their contact lists. When deadlines loom, the pressure to move with the herd is likely as important as media bias.