Bruce Bartlett has a column at Forbes that discusses how to apply the theory of the second best to voting. He discusses his calculations from 2008:
Surveying the political landscape, I didn’t think the Republican candidate, whoever it might be, was very likely to win against whoever the Democratic candidate might be. Therefore I concluded that it was in the interest of conservatives to support the more conservative Democratic candidate so that candidate would be more likely to get the nomination and become president than the more liberal one.
That led him to favor Hillary Clinton over Barack Obama. He thinks he made the right choice:
think the evidence suggests that Hillary Clinton could have won the Democratic nomination with just a little bit more support, and probably would be governing significantly more conservatively than Obama. For one thing, given her disastrous experience with health care reform in 1993-1994, it’s reasonable to assume that she would have stayed away from that issue at all costs.
I’m not so sure. It’s a tough counterfactual on how Clinton might have governed viz. Obama. Would she have stayed away from health care? I doubt it. It was an important issue in her campaign, and politicians tend to follow up on campaign promises. Indeed, she was already talking about health care in 2004. I also think it’s pretty certain she would have pursued a stimulus package similar to what Obama got. That strategy just seems like mainstream economics on the Democratic side of the aisle. Perhaps the open question is one of priority. Would she have pursued health care as soon as Obama or would she have subordinated it to other issues?
After reading Bartlett’s piece, I searched in the political science journals on JSTOR but turned up no applications of the theory of the second best to voting behavior. If anyone knows of any research, please leave a comment. It seems like there could be some interesting and unexplored applications of the theory here.