News

 

A Reply to Paul Krugman

In an October 13 blog post Professor Krugman questioned ECRI's real-time forecasting track record and the validity of our leading indexes.

Here is our reply:

Professor Krugman,

Thanks for taking the time to call attention to ECRI’s work. However, many of the things that you write about ECRI are untrue. First, the record – ECRI’s public recession and recovery calls associated with the 2001 and 2007-2009 recessions are available online for review.

2001 recession (Please look to the right-hand column)

2001 recovery

2008 recession

2009 recovery

We recently highlighted these reports in a column showing how those recession and recovery calls would have helped people avoid much of the losses associated with a buy-and-hold strategy for stocks. On the contrary, annual returns would have averaged 8% more than the S&P since our September 2000 recession forecast. Perhaps that is what prompted the recent attack on our leading indexes.

Some point out that by the time we said recession was a done deal this time around, it was already March 2008. But please note ECRI’s January 2008 public statement that “a self-reinforcing downturn has already begun. If allowed to continue, it will amount to the vicious cycle known as a business cycle recession,” underscoring the extreme importance of timely stimulus that did not materialize.

We wish to address the issue raised by some economists about apparent false alarms issued by our Weekly Leading Index. Please note that since those declines were not pronounced, pervasive and persistent enough to qualify as cyclical downturns, ECRI did not make recession calls on any of those occasions. A case in point was after the 1987 stock market crash, when many did expect a recession, but The Wall Street Journal ran a front page story discussing why our group did not.

In 2005, The Economist, after having done far more thorough fact-checking than most bloggers, declared that “ECRI is perhaps the only organisation to give advance warning of each of the past three recessions; just as impressive, it has never issued a false alarm.”

Your blog post shares some fundamental misconceptions about ECRI’s leading indexes. Perhaps that is why you and other bloggers presume to know what must be driving ECRI’s leading indexes. Thus, you make the point that in the current economic situation, conventional indicators like the monetary base and yield spread don’t mean what they usually mean. But since those are not the indicators driving ECRI’s leading indexes, that critique has no validity. If you wish to understand what drives our approach, please take a look at this transcript of a seminar we gave at the IMF.

A related line of attack on ECRI is through guilt by association: dismissing the importance of the components of the Conference Board’s LEI, and then using rhetorical sleight of hand to claim that ECRI’s leading indexes must therefore be wrong. In fact, just to show how little ECRI’s indexes and the LEI have in common, please note that the LEI release in April 2009, when ECRI correctly predicted that the recession would end this summer, showed that the LEI was still falling, not having experienced a single monthly uptick since June 2008.

Of course, in May 2009, following ECRI’s recovery forecast, you yourself acknowledged that the recession “could” end by this summer.

You state that “historical correlations can’t be counted on to prevail.” Let’s be very clear. The validity of ECRI’s leading indexes is not based on such “historical correlations.” In fact, we don’t even consider correlations to be valid statistical measures where leading indexes are concerned. The inapplicability of standard statistical techniques like regression and correlation was highlighted long ago by Granger and Newbold, who wrote that leading indexes “are intended only to forecast the timing of turning points and not the size of the forthcoming downswing or upswing, nor to be a general indicator of the economy at times other than near turning points.”

Please note that, contrary to what most bloggers assume, ECRI’s leading indexes are not based on back-fitting of data to “garden-variety” postwar recessions. Rather, they go back over a century and, with no recourse to data fitting, can be shown to correctly anticipate the “jungle-variety” depressions, panics and crises of the early 20th century, as well as the recoveries from those extraordinary events. Knowing the theoretical underpinnings of ECRI’s approach, as well as the broad support from reliable leading indicators of recession and recovery that is the basis of the upturns in ECRI’s leading indexes, we fully expect the current economic recovery to prove to be stronger than the last two, at least through mid-2010.

In the decades that we’ve been monitoring business cycles, our views have always diverged from the consensus around peaks and troughs in the business cycle. Once again, last April when we predicted the end of recession, ours were very lonely voices. Yet that forecast has already been vindicated. Perhaps the intensity of the attacks today have more to do with the fact that so many (including hedge funds and investment advisors like Michael Shedlock) have missed the nearly 60% market rally since the spring, which makes it so hard to accept that ECRI’s leading indexes were correct once again.

While we don’t necessarily expect our clarifications to change your views about the near-term course of the business cycle, we would hope that if, a year from now, ECRI’s leading indexes are proven to have been correct, you would publicly acknowledge the same. After all, the proof is in the pudding.

Lakshman Achuthan and Anirvan Banerji
Economic Cycle Research Institute (ECRI)