New Keynesiansm vs. the Retro Keynesianism

There was a not-very-enlightening exchange on the blogs recently about “New” vs. “Old” Keynesianism that I think may be better expressed as a disagreement between New and Retro Keynesianism. Here’s how I would put it. New Keynesianism has the main “Keynesian” points that expansionary fiscal and monetary policy can help reduce unemployment in a recession, but otherwise has basically all the normal old-fashioned economist-y views about what’s good and what’s bad. Something that happened during the recession is that with interest rates at zero and unemployment still sky-high, we saw the development of a number of mathematical models that brought back into style some of the wilder claims associated with Old Keynesianism. That includes things like Gautti Eggertsson’s “paradox of toil” which suggested that under certain conditions if everyone decided to increase their desire to work (or decrease their desire for leisure time or their aversion to unpleasant jobs) employment would actually fall.

I thought of that when I read Calculated Risk on labor force participation and the unemployment rate:

If the January pace of payroll employment growth continues (around 250 thousand jobs per month), and the participation rate stays at 63.7 percent, then the unemployment rate could fall to 7.3 percent in December 2012. But even at a slower pace of payroll growth, the unemployment rate could be at or below 8 percent by the end of the year—unless the participation rate rises or the economy slows sharply.

I think this is only true in a misleadingly literal way. As Karl Smith says, an increase in desired labor supply should increase demand for durables and capital goods and thus raise the natural rate of interest. In other words, if everyone suddenly became one standard deviation more “workaholic” then the rate of job creation would accelerate and there’s no reason to think the unemployment rate would go up. That Smith/Yglesias conclusion is perfectly consistent with the New Keynesian ideas I’m familiar with, but runs contrary to Eggertsson’s Retro Keynesian model.

It’s been suggested that recent events have refuted the Eggertson analysis, but I think that if you look more closely you’ll see that all this work on Retro Keynesianism was predicated on the economy being in an actual state of deflation that the central bank is unable or unwilling to reverse. What actually happened in the United States, however, is that we had a brief deflationary episode followed by an extended period of sub-trend NGDP growth. That’s produced some of the ills you might associate with inflation, but it means that the conditions under which Retro Keynesian ideas were allegedly supposed to apply don’t currently exist and haven’t existed for some time. Under the circumstances, the theoretical questions are perhaps interesting to model-builders and academics but we (thankfully) aren’t living in the kind of universe that would put them to the test.