The Auto Bailout’s Electoral Impact

by Dan Hopkins on December 9, 2012 · 2 comments

in Presidency,Public opinion

That’s the topic that I take on in a new “Poli-Sci Perspective” post over at Wonkblog.  The results (or the weakness thereof) will be surprising to those who followed the pre-election reporting on Ohio, but not to those who have followed the decades-old research in political science on self-interest.


MTNance December 10, 2012 at 5:13 pm

I think that work captures the dynamic pretty well. But I think a more complete analysis would have to look at the supplier chain, too, which stretches from Milwaukee down to Mexico, and I think (without double-checking) along the purple-ish route in the map you provide. Those firms profited as much from the bailout as the car companies. That could strengthen your argument or moderate it–I’m not sure which. Of course, much of that is in red states, so it would be more about explaining individual level behavior, not aggregate outcomes, but that makes it more interesting in a way. In terms of explaining the relative unimportance of the auto bailout, beyond a more general “self interest doesn’t determine everything” notion, there’s also the fact that the automobile industry may not be as important to the US economy as I think it is imagined to be in the national self-image.

dirkmjk December 11, 2012 at 6:21 am

Interesting analysis. In a less sophisticated manner, I also found that the bailout doesn’t seem to have had much of an impact in Ohio ( What surprised me is that there didn’t appear to be much of a correlation between changes in unemployment and changes in election outcomes. Maybe a more detailed analysis would have yielded better results. Or maybe, as you suggest, voters simply aren’t as self-interested as I implicitly assumed..

Comments on this entry are closed.

Previous post:

Next post: