I keep hearing how the current recession will end in 2010 because the average United States recession from 1854 to 2001 has been 17 months. This is silly for a variety of reasons.
One reason is that there is no average recession. Post-World War II recessions have lasted from a minimum of six months to a maximum of only 16 months. If we were to apply the “average recession” logic to post World War II recessions, the current recession, which the NBER — the National Bureau of Economic Research — says started December 2007, would have ended 10 months later, last October.
Another reason is that few previous recessions have been accompanied by the financial sector collapse that we witnessed in September. Worldwide experience indicates that recessions associated with financial sector panics tend to be longer than those without panics.
Since 1854, five United States recessions have been accompanied by financial panics. These are the recessions of 1857, 1873, 1893, 1907, and 1929. The average duration of these recessions was 31 months. The 1907 recession was the shortest, at only 13 months. The 1873 recession was the longest, at 65 months. For comparison, the 1929 recession was 43 months. Interestingly enough, J.P. Morgan was instrumental in ending the financial panics of the two shortest recessions, 1893 and 1907.
If we were to engage in the same sort of fuzzy thinking as the “average recession” analysis applied to “financial Panic” recessions, and assuming we use the NBER recession start date of December ‘07, the current recession could be expected to end 31 months later in July 2010. Is that too long for you? You could use the average 20th Century recession accompanied by a financial panic length. That is 28 months, so maybe the recession will end in April 2010.
Maybe we should look at foreign data? The point is that if you play this game long enough, you can find a date you like.
Finally, the method of dating recessions changed with the 2001 recession. The new method is much more likely to declare an economy in recession. If the old method had been used — if previous criteria were applied to the current situation — I believe the recession would have commenced no sooner than July 2008. Recent data revisions increase my confidence that the NBER was wrong when they said the recession commenced in December 2007. If you have the wrong start date, any “average recession” method will be wrong.
The facts are that we have a serious recession accompanied by a financial panic and continuing massive job losses. The correct way to analyze the current recession is to recognize that it was accompanied by financial panic, and that means we had a regime shift from a good equilibrium to a bad equilibrium.
Game theory tells us that we can have multiple Nash Equilibria to certain games. A Nash Equilibrium is one where knowing your opponent’s decision you would not change your decision.
Bank runs provide an excellent example. Suppose you have a bank that does not have deposit insurance. Most of the time things plug along. People make deposits, borrow, and the like. Everybody is happy with their decisions. Call this the good equilibrium. However, in the event of a bank run, everyone wants to participate in the run, because those who do not end up loosing. Call this the bad equilibrium. Furthermore, nothing real has to change. We can switch from the good equilibrium to the bad equilibrium on unfounded rumors.
The financial panic we witnessed last September was exactly like a bank run. In an amazingly short time, we switched from a good equilibrium to a bad equilibrium. The bad news is that we have no idea how to switch from a bad equilibrium to a good equilibrium. It will surely happen, but we don’t know how to cause it. We don’t know what will cause it. We can’t predict when it will happen.
We do know that a lot of assets need to change hands. These include financial assets, auto factories, and homes. Recessions are periods when assets are reallocated to better uses.
Current policy, with its obsessive pursuit of bailouts, seems to be focused on delaying those reallocations. That will delay the recovery. So-proposed government efforts to limit the impact will be ineffective, if not counterproductive. That is why I don’t see any reason to expect a recovery in 2009.
Bill Watkins, Ph.D. is the Executive Director of the Economic Forecast Project at the University of California, Santa Barbara. He is also a former economist at the Board of Governors of the Federal Reserve System in Washington D.C. in the Monetary Affairs Division. All recession dating data in this article is from the NBER website.