For investors who believe that stocks eventually reflect the fundamental strength of the companies, understanding the business cycle is crucial. Since popular sentiment and stock prices are often excessively negative about economic prospects, there is great opportunity for those who can get this right.
Regular readers know that I have been conducting research on this topic since May, providing some occasional updates and background. I am now ready to describe what I have learned. This is going to require about five articles over the next couple of weeks. Today’s installment is the first. Tomorrow I plan to discuss the best single method that I found.
Many pundits, including some of the most famous and powerful, are playing fast and loose with the “R” word. John Hussman, a leader in keeping investors scared witless (TM OldProf eumphemism), now warns us that everyone else is looking at lagging or coincident indicators, and that we might all be blindsided by a recession. Here is a key argument:
Let’s examine the seemingly most “compelling” data point first – the fact that December payrolls grew by 200,000. Surely that sort of jobs number is inconsistent with an oncoming recession. Isn’t it? Well, examining the past 10 U.S. recessions, it turns out that payroll employment growth was positive in 8 of those 10 recessions in the very month that the recession began.
Let’s keep that one in mind for a moment while we consider another popular allegation. It goes like this:
Mr. X (substitute nearly any mainstream analyst) thought that the economy was fine in December of 2007, the very month the recession was starting.
These statements are accurate, but deceptive. Understanding why is crucial to an objective analysis of recession forecasting.
What is a Recession
For the purposes of my series, and also the basis for the deceptive arguments, we are looking at the official recession dating by the National Bureau of Economic Research (NBER). The NBER is not a government agency. It is a private, non-profit, non-partisan group.
I did an extensive discussion of the NBER process in April, 2008 (when the recession was still actively debated) in my article, How to Win a Recession Predicting Contest. You can go back and see who was predicting what at that time, but the more important point is the process used by the NBER. To summarize that description, the NBER does two things:
- Wait until there is a significant move lower in a group of four factors, with special emphasis on personal income less transfer payments and employment.
- Only after noting such a decline does the recession-dating committee go back to find the last cycle peak and declare that to be the start of the recession.
In the case of 2007-08, the NBER determined that the recession started in December of 2007 in December of 2008, a year later. If the economy had rebounded during the year, or maybe if the Lehman failure had been avoided, the dating and scope would have been different.
When a recession ends, the NBER does the same two steps in the opposite direction. The end of the recession, June 2009, was declared fifteen months later in September of 2010.
Here is a simple rule for understanding the NBER dating:
Forget the term “recession.” Think of their work as identifying peaks and troughs in the business cycle.
To avoid mistakes, the NBER waits for solid evidence before declaring a peak or trough.
This definition does not correspond to what most people think of as a recession. Most of us just mean that the economy is growing at a rate significantly below potential, usually reflected in the long-term trend. By this definition we never emerged from the 2008 recession. Others like to use two quarters of negative GDP growth. Most people have never heard of the NBER and mistakenly think that it is just another clueless government agency.
Feel free to use your own definition, but when you are reading about economic forecasting, the shaded areas you see on the charts reflect the official NBER dating.
Revisiting the Deceptive Statements
If you keep the NBER definition in mind, you will see why the deceptive statements are unfair and misleading. Of course most recessions started when there was positive employment growth. If you grasp that the start of a recession is defined by the NBER as an economic peak, this is almost inevitable. If the employment growth were weak, it would not be the peak; some earlier month would be selected as the start of the recession.
Similarly, citing what someone said in December of 2007 is unfair, since no one knew for sure that it would be the peak. It is like someone giving directions — “If you see the school on the left, you have gone too far.”
We need to find true leading indicators — the focal point of my mission in this series.
For those who are predicting that we will enter a recession next month, it could, of course, prove to be correct, but it is very unlikely. Meanwhile, it is safe to say that the recession did not begin before December. Why not? None of the four major NBER factors have shown a peak so far. Unless we have an instant and surprising decline in January, December will not be the peak either. Check out The Bonddad Blog for a fine review of the major NBER factors and an analysis of how close we came to a recession in 2011.
Let us revisit the Hussman “leading indicators” with this quotation from Business Insider:
A few weeks ago, I noted that our recession warning composite was on the brink of a signal that has always and only occurred during or immediately prior to U.S. recessions, the last signal being the warning I reported in the November 12, 2007 weekly comment Expecting A Recession. While the set of criteria I noted then would still require a decline in the ISM Purchasing Managers Index to 54 or less to complete a recession warning, what prompts my immediate concern is that the growth rate of the ECRI Weekly Leading Index has now declined to -6.9%. The WLI growth rate has historically demonstrated a strong correlation with the ISM Purchasing Managers Index, with the correlation being highest at a lead time of 13 weeks.
- Openness — with the potential for peer review
- Small number of input variables. Most people do not understand that “small is good.” If you have a lot of variables, it is easy to do back-fitting on a few cases. Beware.
- Real-time performance. This means that you do not go back in history doing any data-mining. You create an indicator and live with it through time. (While the ECRI predictions are a matter of record, no one knows what changes they have made in their indicators).
In Part 2 of this series I will describe the method that best meets these criteria.
This post originally appeared at A Dash of Insight and is posted with permission.