Wednesday, October 17, 2012

The MUSS Method and John Taylor

In a recent blog post, Economist John Taylor makes the comparison between our current recovery and previous U.S. recoveries. He argues that our current recovery, which he calls "unusually weak," is not, as economists like Ken Rogoff and Carmen Reinhart have argued, slow because it was caused by a financial crisis. He argues that Rinhart/Rogoff's original work is inapplicable to the U.S. because it largely covered financial crises in foreign countries. Instead, he asserts that poor policy-making is what holds our recovery back, and that an observation of previous U.S. financial crises and their recoveries proves him right.

As evidence of this, he uses research from Michael Bordo and Joe Haubrich to show that previous recoveries from American financial crises have been much, much faster. Specifically, he uses this chart:
Definitive proof, right? Here's the problem: at first glance, a number of these recessions listed weren't even caused by financial crises. For example, the 1973 recession was largely an oil-price supply shock. The 1981 recession was caused by the Fed. And those are just the ones I personally know about off the top of my head! In response to Taylor's arguments, Ken Rogoff and Carmen Reinhart released a new paper as a rebuttal, in which they compare actual financial crisis-induced recessions in the U.S. (not things like 1973 and 1981!) and their subsequent recoveries to our current recovery. This chart encompasses their findings:
At worst, the U.S. is par for the course by comparison, and at best, we did far better than relevant historical recessions and recoveries. John Taylor seems to have fallen into the same sort of trap Phil Gramm and Glenn Hubbard fell into in their June op-ed, except that this time, instead of falsely comparing one recession and recovery to another, Taylor drew false comparisons writ large. Last night, Noah Smith criticized Taylor's argument, saying that he was using the MUSS (Make Up Some...Stuff) method of economic debating. It sure looks that way, doesn't it?

Update: Paul Krugman picks up on something I should have:
"The first is that looking at the rate of recovery from the trough is a very peculiar criterion — especially when, as Taylor does, you look only at the first year (!) of recovery. By this standard, the New Deal was a tremendous success story, because growth was fast in 1933-4. Never mind the fact that pre-crisis per capita GDP wasn’t restored for more than a decade. As R-R say, surely the relevant comparison is with the pre-crisis peak, especially given the fact that post-crisis economies often suffer periods of relapse (as is happening in Europe now)."
Excellent point.