Here's an excerpt from a much lengthier piece in the New York Review of Books, in which Krugman makes the not-implausible case there's reason to hope for a return to Keynesian policies before the election:
Pundits are always making confident statements about what the American electorate wants and believes, and such presumed public views are often used to wave away any suggestion of major policy changes, at least from the left. America is a “center-right country,” we’re told, and that rules out any major initiatives involving new government spending.
And to be fair, there are lines, both to the left and to the right, that policy probably can’t cross without inviting electoral disaster. George W. Bush discovered that when he tried to privatize Social Security after the 2004 election: the public hated the idea, and his attempted juggernaut on the issue quickly stalled. A comparably liberal-leaning proposal—say, a plan to introduce true “socialized medicine,” making the whole health care system a government program like the Veterans Health Administration—would presumably meet the same fate. But when it comes to the kind of policy measures I have advocated—measures that would mainly try to boost the economy rather than try to transform it—public opinion is surely less coherent and less decisive than everyday commentary would have you believe.
Pundits and, I’m sorry to say, White House political operatives like to tell elaborate tales about what is supposedly going on in voters’ minds. Back in 2011 The Washington Post’s Greg Sargent summarized the arguments Obama aides were using to justify a focus on spending cuts rather than job creation:
A big deal would reassure independents who fear the country is out of control; position Obama as the adult who made Washington work again; allow the President to tell Dems he put entitlements on sounder financial footing; and clear the decks to enact other priorities later.
Any political scientist who has actually studied electoral behavior will scoff at the idea that voters engage in anything like this sort of complicated reasoning. And political scientists in general have scorn for what Slate’s Matthew Yglesias calls the pundit’s fallacy, the belief on the part of all too many political commentators that their pet issues are, miraculously, the very same issues that matter most to the electorate.
Most real voters are busy with their jobs, their children, and their lives in general. They have neither the time nor the inclination to study policy issues closely, let alone engage in opinion-page-style parsing of political nuances. What they notice, and vote on, is whether the economy is getting better or worse; statistical analyses say that the rate of economic growth in the three quarters or so before the election is by far the most important determinant of electoral outcomes.
What this says—a lesson that the Obama team unfortunately failed to learn until very late in the game—is that the economic strategy that works best politically isn’t the strategy that finds approval with focus groups, let alone with the editorial page of The Washington Post; it’s the strategy that actually delivers results. Whoever is sitting in the White House next year will best serve his own political interests by doing the right thing from an economic point of view, which means doing whatever it takes to end the depression we’re in. If expansionary fiscal and monetary policies coupled with debt relief are the way to get this economy moving, then those policies will be politically smart as well as in the national interest.