Chait argues that elections are mostly about fundamentals. Andrew Gelman counters that the party in power has some control over said fundamentals:

I think it's perfectly reasonable for liberals to interpret some of the current state of the economy, and thus the predicted election outcome, to "timid policies by the Democrats." That's what Krugman is saying every day. Similarly, why shouldn't conservatives think that the current economic doldrums are partly explained by the Democrats' policies, from regulation to stimulus to health care? Republicans have been making this argument for awhile, that these activist government policies are counterproductive.

I fail to see how a stimulus-free 2009 would have helped lower unemployment by now. It seems to me that even the purists concede it would have meant a free-fall, but argue it would have been worth it in the long run. Seems like a huge human cost for the admittedly important principle of moral hazard to me, and one no actual president of either party would have considered. Kevin Drum is worth reading on the electoral repercussions if Krugman had had his way and Fox News was actually accurate about the "radical leftism" of the current administration:

  • The actual stimulus bill that was passed in Feburary 2009 amounted to about $800 billion. The biggest bill anyone was talking about at the time was $1.2 trillion. That's 50% bigger, and presumably would have been about 50% more effective.
  • For calendar 2010, CBO estimates that the stimulus bill reduced unemployment by something between 0.7 and 1.8 points. Split the difference and the consensus average is about 1.2 points. A stimulus bill that was 50% bigger would therefore probably have reduced unemployment by 0.6 points more than the actual bill.

If this is in the ballpark, it means that with a bigger stimulus bill unemployment today would be 9%, not 9.6%. That would have been well worth the price, but just because it was worth doing doesn't mean it would have made a big electoral difference.