If there is any intellectual framework that should have been discredited over the past decade it is old-style Keynesianism. Unfortunately, just the opposite has happened. Marcus Nunes directed me to a Noah Smith post that discusses the revival of old Keynesian ideas:
Another way of putting this is that Paul Krugman was right. Krugman has long advocated that macroeconomists learn to once again think in terms of simple simple Keynesian theory. And when more fully developed, complex models are needed, Krugman uses the kind of models that Christiano endorses.
As Christiano mentioned, the New Keynesian revolution isn’t so new. Even in the 1990s, economists like Greg Mankiw and Olivier Blanchard were arguing that monetary policy had real effects on demand. And at the same time, international macroeconomists were realizing that Japan’s post-bubble experience of slow growth, low interest rates and low inflation implied that demand shortages could last for a very long time unless the government rode to the rescue. Krugman, Adam Posen, Lars Svensson, and others were already referring to a Japan-type stagnation as a liquidity trap in the late 1990s, and warning that standard monetary policy of cutting interest rates wouldn’t work in that sort of situation. . . .If economists gravitated toward anti-Keynesian theories, it was at least in part because evidence wasn’t strong enough to push them in the right direction. It’s just very hard to assess the impacts of fiscal stimulus. For example, Japan’s tremendous government spending binge in the 1990s looks to a casual observer like it had no effect, since the economy didn’t recover until years later — but government spending might have been the only thing saving the country from a deeper recession.
I certainly agree that Japan tells us a lot about the validity of old Keynesian thinking. Here are some things it tells us:
1. Depreciating the yen is a foolproof way of creating inflation. Thus Keynes was wrong about monetary policy being ineffective at the zero bound.
2. From 1993 to 2013 Japan ran up by far the largest peacetime fiscal deficits ever seen by a major economy. And all that “stimulus” led to by far the worst growth in AD over 20 years ever seen by a major economy. Roughly zero growth in NGDP over two decades. And the Keynesian takeaway is that this was a great success, as it prevented an even more record-breaking fall in NGDP. This is like a religious person who believes in the efficacy of prayer, prays for peace in 1939, and then later argues that his prayers prevented an even bigger war and Holocaust. Okaaaay . . .
3. Then in 2013 Abe takes office and raises consumption taxes. This fiscal tightening causes the debt to GDP ratio to level off at 250%. Instead Abe relies on monetary stimulus, raising the inflation target. And both inflation and NGDP growth actually increase, the opposite of the prediction of the old Keynesian model.
Of course I could go on and on. There’s the letter signed by 350 Keynesians warning that the fiscal austerity of 2013 risked recession (growth actually sped up.) Or the fact that Keynesians don’t even no how to estimate the multiplier (as documented recently by Ryan Murphy.)
Smith points out that Paul Krugman realized in the late 1990s that the standard policy of cutting interest rates would no longer work at zero. But he doesn’t tell you that everyone already knew that, even Milton Friedman. AFAIK, not one economist in the entire world in the late 1990s thought that cutting interest rates when they were already zero would work. Perhaps you think I’m being too picky; what Smith really meant is that Krugman discovered that monetary stimulus no longer worked in Japan, and that fiscal stimulus was needed. Except that’s not true, in the late 1990s and early 2000s Krugman ridiculed the idea of using fiscal stimulus in Japan, and suggested that monetary stimulus was the obvious solution. Whatever “new facts” caused Krugman to revert to old Keynesianism more recently; it certainly wasn’t his famous 1998 study of Japan’s liquidity trap. So what caused Krugman to change? I’m not sure, but Smith hints at one possibility:
When evidence is sparse or inconclusive, things like sociology and politics often fill the gap.