after July 2008 there were many other indicators that money was far too tight. About this time the dollar began rising sharply against the euro; and commodity prices began a sharp decline. Real interest rates rose from less than 1 percent in July to more than 4 percent by November. In the first ten days of October the stock market crashed 23 percent, signaling much more bearish expectations for the economy. The recession spread beyond housing, depressing manufacturing and other sectors. Both real and nominal GDP fell at annual rates of 5 percent to 6 percent in the fourth quarter of 2008 and the first quarter of 2009.

Sumner continues,

most economists missed a fundamental change in the nature of the crisis during the late summer of 2008. Whereas the initial downturn was mostly caused by “real shocks,” such as banking turmoil and high energy prices, the far more severe second stage of the recession was triggered by a sharp decline in NGDP, which represents a failure of monetary policy…most economists simply assumed that policy was expansionary because the Fed had cut rates to relatively low levels, and the “monetary base” (which is the money produced by the Fed) rose very dramatically in the fourth quarter of 2008. Neither nominal interest rates nor the money supply, however, is a reliable indicator of the stance of monetary policy.

Read the whole thing. It is the clearest and most persuasive exposition of Sumner’s views.

Suppose that the Fed had decided to make an all-out effort to use monetary expansion to prevent a recession. What could have gone wrong?

1. We could have been in a liquidity trap. My view is that we are in a liquidity trap more often than not. The fact that the Fed can reduce the Fed Funds rate does not impress me. If the risk premium is soaring on mortgage securities, stocks, and other assets, the Fed is not going to accomplish much. On the other hand, if the Fed just keeps buying stuff until interest rates that matter start to come down, you’d think it would eventually accomplish something.

2. The shock was primarily real. In that case, if the Fed expands like crazy, we get inflation with little or no improvement in real growth.

I tend to lean on (2) more than (1) as my reason for doubts about Sumner’s views. However, I assign a probability of at least 33 percent that Sumner is right and I am wrong.

One issue is that the downturn was broadbased. This makes it seem as though a lot of the unemployment was “unnecessary,” or due to a fall in aggregate demand. What I need to demonstrate, both in theory and empirically, is that this was really a broad-based supply shock, in which weak firms in many industries were exposed and sent contractionary market signals.

I am thinking a lot about this issue these days, and I do not have definitive answers. But one way to think of it is that shakeouts that would have been stretched out over 8 to 10 years under normal circumstances were compressed into about 18 months during the recession.