Although they don’t quite say so explicitly, the paper is to an important extent an answer to Robert Barro’s claim that the World War II experience shows that multipliers are low, because private spending actually fell during the war; as I and others have tried to point out, this was because it was, you know, wartime, with rationing of consumer goods and sharp restrictions on private investment.
What Gordon and Krenn point out is that we actually have more information than a simple comparison between the depressed peacetime economy and the war economy after Pearl Harbor: there was a period of more than two years when the United States was gearing up for war but not yet engaged in combat “” the Arsenal of Democracy era. Rationing was not yet in effect, and for at least part of this period the economy still had excess capacity despite a very large rise in government spending.
What they find is that when there was still excess capacity, there was a quite large multiplier on government spending; that is, fiscal policy worked.
So far so good. And Krugman overlooks another advantage that he has discussed elsewhere, the short term interest rate was close to zero. This is important because when rates are positive, an inflation targeting central bank will tend to neutralize the impact of fiscal stimulus. Krugman concludes:
But in the prewar buildup you had a clear-cut expansion of federal spending on the order of 14 percent of GDP. That’s a real experiment with the economy. And the results were clearly Keynesian.
I agree with that, but for completely different reasons. What’s most important is that monetary policy was very passive during those years. The Fed was quite content to let the wholesale price index trend downward from 1937 to mid-1940. The monetary base was determined by gold inflows. Under this sort of passive monetary policy regime fiscal stimulus can boost NGDP, and it seems to have done so in 1939-41.
But . . . this natural experiment has no implications for the efficacy of the recent fiscal stimulus. Unlike in 1939, our Fed is not content to let the US economy slide into a Great Depression. Any doubts on that score were erased by the recent Fed meeting, which strongly implied they would not allow inflation to fall below about 1%, and were seriously considering policies such as aggressive QE, in order to prevent that from happening. I won’t repeat all the evidence; my previous post discusses this in a bit more detail.
And in a sense we’ve known this all along. As far back as March 2009 the Fed responded to the sharply falling NGDP growth expectations with a substantial bond purchase program (albeit nowhere near as much as was needed.) We don’t know exactly where the “Bernanke put” is, but we can be pretty sure that we were close to it in March 2009.
Many pro-Keynesian commenters will defend the fiscal stimulus, despite its apparent failure, with two arguments:
1. It was too small.
2. At least it prevented the economy from sliding into a depression.
I agree that from the perspective of the standard Keynesian model it was too small. Nonetheless, it did far less than its proponents expected. The other excuse is much more doubtful. There is no evidence that the Fed would have allowed things to get worse than they did. Is it possible? Sure, anything’s possible. It’s also possible that in the absence of aggressive fiscal stimulus the Fed would have done much more, and the recovery actually would have been faster.
I’m even willing to concede that it’s more likely the stimulus boosted NGDP in 2009, than reduced it. But I hope everyone understands that with modern inflation targeting central banks, there is no such things as a “natural experiment” for fiscal stimulus, there is no scientific evidence pointing to how much fiscal stimulus boosted output, as compared to the counter-factual where there was no fiscal stimulus. In grad school I recall they’d say “it’s a game theory problem,” which translates as “who the hell knows.”
This appears in the Gordon and Krenn abstract:
Only the 1.80 multiplier is relevant to situations like 2009-10 when capacity constraints are absent across the economy.
I love economics, but I’m frustrated that many economists have an excessively narrow or linear way of thinking about policy issues. Their study has no relevance for 2009-2010.