Monday, March 28, 2005

Recession Predictions: A Mug's Game?

by Calculated Risk on 3/28/2005 11:59:00 AM

"It's hard to make predictions - especially about the future."
Allan Lamport, former Toronto Mayor.

UPDATE: Two people have commented that the above quote was from Yogi Berra. It definitely sounds like Berra, and it is usually attributed to Berra, but I'm pretty sure it is from Mayor Lamport.

I’m going to take a look at the last consumer recession (July 1990 to March 1991) for clues about what to look for in the current situation. I believe the next recession will be consumer driven, led by a slow down in real estate, perhaps triggered by high energy prices and rising interest rates.

First, I would like to point out that forecasters have a terrible record at predicting downturns. Victor Zarnowitz wrote that major "...failures of forecasting are related to the incidence of slowdowns and contractions in general economic activity. Forecasts...go seriously wrong when such setbacks occur." The reason for this predictive failure is primarily due to the forecasters' incentives. Zarnowitz wrote: "predicting a general downturn is always unpopular and predicting it prematurely—ahead of others—may prove quite costly to the forecaster and his customers".

Incentives motivate economic forecasters to always be optimistic about the future (just like stock analysts). Luckily I have no customers (my thoughts are free and worth every penny), no financial incentives and no reputation! That said, I’m not predicting a recession (yet), only suggesting tools that might help identify the next recession.

The 1990/1991 Recession

As we look back at the ’90 recession, here are a few quotes from Fed Chairman Alan Greenspan (bear in mind that the recession started in July, 1990):

“In the very near term there’s little evidence that I can see to suggest the economy is tilting over [into recession].” Greenspan, July 1990
“...those who argue that we are already in a recession I think are reasonably certain to be wrong.” Greenspan, August 1990
“... the economy has not yet slipped into recession.” Greenspan, October 1990
Source (pdf): "Booms, Busts, and the Role of the Federal Reserve" by David Altig (See macroblog)


A common belief is that the stock market predicts economic activity. Looking back at 1990, here is a graph of the S&P 500. The graph shows that the market rallied right into the recession and only started selling off after the recession started.

Click on graph for larger image.

Another coincident indicator in 1990 was employment.


This graph shows monthly job creation in 1990. Although job creation was spotty for a couple of months before the recession, it didn’t turn negative until after the recession started.

A potentially predictive tool, often cited by investors, is an inverted yield curve. An inverted yield curve exists when the rates on shorter duration instruments are higher than rates on longer duration instruments. Here are the yields for the 13 week treasury note, and 5 year and 10 year treasury bills.


The curve inverted in mid-1989, a full year before the recession started. A popular joke on Wall Street at the time was that the yield curve has predicted eleven of the last 7 recessions! The yield between the 5 year and 10 year did stay narrow right up to the recession. I'm not sure why the bond market would have more information than the stock market, but this indicator is frequently cited.


The last chart is my favorite leading indicator: New Home Sales. As the chart indicates, New Home Sales were declining for 12 months prior to the start of the recession. This has been observed (usually 8 to 12 months) for other consumer recessions (see New Home Sales as Leading Indicator and Update).

It makes sense that consumers, sensing economic weakness, would pull back on large purchases first (like homes). Then, since the Real Estate market is a large percentage of the U.S. economy, a slow down in Real Estate acts as a drag on the general economy - perhaps a self-fulfilling prophecy.

The Current Situation

So what does this mean for today?

First, here is the current chart for the 13 week treasury note, and 5 year and 10 year treasury bills. Clearly interest rates are rising, but the chart also shows that the yield curve is narrowing. The spread between the five and down year bills is close to 30 bps.
And finally, here is the current chart for New Home Sales. It is possible that New Home Sales peaked last October, however February's data was still very strong.

So, for now, I do not believe a recession is imminent. I'm just starting to watch for the early signs. Predicting recessions may be a mug's game, but I'm probably going to play.