Saturday, January 19, 2008

odds of a recession

Last month I had a number of people ask me if there was going to be a recession this year. I answered that I thought there was a 40% chance. I’m no economic forecaster, but I later found out this was not too bad a guess. (And a guess it was.) The Wall Street Journal has a monthly survey of 54 real economic forecasters. As to the odds of a recession in the next 12 months, the December average was a 38% chance and the average in the first week of January was 42%.

That puts me right in the middle of the pack. Why is that good? Because the track record of this survey shows that the average forecast is a pretty good predictor, much better than any individual forecast.

I have revised my odds of a recession in 2008 to 50%. However, unlike the 54 pros surveyed, I was able to cheat by looking at their results. In December only about 10 of 54 thought the odds were 50% or more. In January, there was a large shift, actually a skew, in the distribution of forecasts with 20 of 54 at 50% or more. So I’m going with 50%.

If we have a recession it is likely to be a relatively mild one with no sharp downturn in economic activity. In fact the last two recessions were also mild (1990-91, 2001). We haven’t had a serious recession in over a quarter of a century. The consensus forecast is an unemployment rate for 2008 of a little over 5%. What is really strange that only five of the 54 forecasters predict any fall in GDP during any quarter (three-month period) in 2008. Recessions don’t happen unless GDP falls.

Heard anyone claim they “know” we’re already in a recession? If so, automatically dismiss him or her. We don’t know about recessions until after the fact because even the most current economic data is dated, collected at some point in the past. It is history. Also, don’t go by one month’s worth of data. Monthly data is notoriously volatile with frequent ups and downs. It is dumb to determine a trend from a one month change.

That said, there are some indicators that may hint of a recession. When the monthly unemployment statistics are released, I ignore the unemployment rate and look at the whether “payroll employment” went up or down. If this happens for two, or better yet, three months in a row, then we’re probably in a recession.

The unemployment rate is problematic. It is the number unemployed divided by the number in the labor force. The problem is that both the numerator and denominator in this ratio can change from month to month and yield funny results.

The stock market is considered a leading indicator. The track record? It has “predicted” 16 of the past 10 recessions. (Yes, I meant to write 16 of 10.) This is not the best of indicators. Since the end of WWII we have had 10 recessions and the stock market failed to decline prior to only one (1953). However, declines in the stock market also indicated seven recessions that never occurred. The most notorious example was on October 19, 1987, the biggest one-day absolute drop in the stock market. Overall stock values fell 22.6%. Outside of BMW and luxury condo sales in the south end of Manhattan, it didn’t even cause a blip in any important economic statistics.

1 comment:

nymrsb said...

People always want a definitive answer from those they believe are "in the know".
Know what? You've been tagged. See my site.