Recession Forecasting: News from the ECRI

Today we got the "official" recession forecast from the ECRI.  As regular readers know, I have been a long-time fan of the ECRI methods, but more recently raised questions.

Last week's WTWA column, was right on target, since I emphasized recession forecasts and methodology. A crucial idea is what we can and should believe.  Here are key questions:

•  Is it enough that there is a good track record, even though we do not know anything about methodology?

• If we can see the methods, how much do we  dismiss the results because "this time is different?"

In the case of the ECRI, I admire their record of forecasts, especially that there have been no false positives.

Meanwhile, there are concerns-- issues which I have been questioning for several months -- the lack of transparency, the secret stuff for paying clients, and the apparent irrelevance of the public data.

Since I observed this several months ago I have  been looking for indicators that meet Jeff's acid test for relevance:

• 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).

A Candidate for Review

I have been considering several candidates using the acid test criteria.  What do you think of this chart?

I am going to assume that you agree with me that this is a pretty good fit.

What if I now told you that the chart was advanced by nine months?

You should be impressed, as I was when I first saw the data.  The implication is that we are not at a cycle peak, and such a peak is more than nine months in our future.

Conclusion

How should we evaluate this?  The background is that everyone knows economic performance is below trend and something better is needed.

Despite this common-sense knowledge, people wonder about the official recession definition.  I encourage readers to consider the candidate I have nominated in this article, and feel free to offer your own suggestions.

I'll take a deeper look in the regular WTWA column.

Comments

[thiazole]

Keep in mind that the ECRI's track record hasn't necessarily been predicting recessions in "real time". In Feb 1990, they called a recession that didn't begin until July 1990. That is 5 months in advance of the recession. I think there is another earlier example as well where they called it fairly early. Not necessarily a bad thing, but something to be aware of.

[oldprof]

RB -- Thanks for the pointer. I have been trying to highlight a number of different viewpoints. Dueker's odds are similar to most -- perhaps a touch better. You are right in observing that the approach meets the tests.

Thanks,

Jeff

[Andrew]

Jeff - I can tell you are optimistic, but do you believe a recession is unavoidable within the next year? Thanks and I appreciate your hard work and time you put into this blog.

[oldprof]

John -- Thanks to you and to Bond Sales Person. I think I have fixed the problem.

Jeff

[John]

For some reason the image isn't there completely. When you right click it and press "View Image" you can see the complete image.

[Bond Sales Person]

Hi Jeff, You do great work and I appreciate you letting me in on your thought process. This is one of the few posts that I can not follow. What is the chart showing? On my view it ends in the year 2000?

[RB]

This, from Mike Dueker, is a qualified warning, but I think it meets the simplicity criterion. Still I hope that your optimistic view proves right!