Debunking the 100% Recession Chart

Note to readers -- It was either this topic or the fiscal cliff.  Since I have explained for many weeks that we will not be doing any cliff diving, I am choosing to take up the recession chart.  So much misleading commentary, and so little time!

This chart is making the rounds, confirmation bias for those who are convinced that below-trend growth equals a recession.

There are several leading sources pointing to the chart and suggesting that it implies a 100% chance of a recession.  There are so many things wrong with this that I hardly no where to begin.  Here is the commentary from non-economist talk-show guy Lance Roberts:

"As stated above, each time this indicator has signaled the probability of a recession at 20%, or higher, the economy has either been in, or was about to be in, a recession. However, while the indicator is currently at a level that is indicative of a recession, along with a host of other economic indicators confirming the same (LEI Coincident to Lagging Ratio, ADS, GDI, Final Sales, STA Economic Composite, etc.), the NBER will wait to date the recession until the data revisions are in. Historically, as shown in the table below, the amount of time between the recession probability indicator hitting 20% and the NBER confirming the start of the recession has been on average about eight months. However, since the turn of the century that lag has moved to roughly 11.5 months."

Try this.  Suppose that your favorite football team has not lost a game in the last three years in which it has scored 24 points or more.  Then they lose 27-24.   You simply revise the proposition to say that they have never lost when scoring 27 or more points.  This is what many amateurs do in interpreting data like the chart above.

Look at the chart above and pretend that we were going back to the future in 2005.   You could have drawn the line at 15% and said everything that Lance says now.  You would have been completely wrong!

This "never before" style commentary keeps coming up.  It has no solid basis in research methods, but it seems convincing to anyone who has not studied research methods.

A Novel Idea -- Read the Research!  D'oh

Here is a novel idea for the economic bloggers -- count to 10!  Maybe you should actually read the research and consult with the authors before rushing to publish.

If you actually read the cited paper you would see the following points:

1. The authors were trying to improve slightly on the NBER  recession dating process -- getting closer to real time.
2. The authors recommended waiting until a reading of 80% was reached for three consecutive months to avoid excessive false positives.  (P. 9 of the paper).

At a minimum, if you are going to use your bully pulpit on the Internet to scare the daylights out of people, wouldn't it be a good idea to confirm with the authors first?   This is the downside of a world where no confirmation is required before publishing.

Why the Average Investor is Getting Bamboozled -- Again!!

This is interesting research.  My strong recommendation is that interpreting a complex economic paper should involve the opinion of the authors.  I strongly doubt that any of the pundits writing about this could speak coherently for thirty seconds about Markov models.  So why should we listen to them?

I called Prof. Piger this afternoon (voicemail), and I'll try again.  Meanwhile, I am accepting his work at face value.  It is a tiny and preliminary recession warning.  I will be quite surprised if he agrees with the Lance interpretation of his work.

About eighteen months ago I initiated a challenge to the ECRI as the top recession forecaster.  At the time, they were not forecasting a recession, but I was concerned about the methodology.  I invited nominations from everyone, and regularly highlight those with the best records.  (Anyone who has followed my approach has done well in the market over the last eighteen months).

My highly visible search has helped me to discover an expert community  in recession forecasting.  We have regular discussions.  I frequently highlight the excellent articles by these colleagues.

I want to preserve the privacy of these emails, while still giving credit.  To this end let me say that Dwaine Van Vuuren of RecessionAlert, one of my regular featured sources, suggested to me that the published chart does not reflect original data.

That got me thinking, so I checked out Prof. Chauvet's site.  Sure enough, the data from last year shows a spike which apparently was later revised away if you believe Prof Piger's site.

Note the serious discrepancy with Piger.

Here is the supporting data:

Dwaine tells me that this is common with Markov models and revisions.  I hope that he will write on this topic, and that he will gain the wide visibility he deserves.

Investment Conclusion

Today's market was a combination of the following:

1. Investors confusing their politics with their investing;
2. Excessive fear about the fiscal cliff;
3. Excessive fear about Europe; and
4. Those mesmerized by "the chart" discussed here.

When there is excessive fear, it is an opportunity for long-term investors.  If you disagree with the recession forecast, the key choices are cyclical companies and technology -- all on sale via your favorite broker.  Regular readers know that I like CAT, AAPL, and ORCL, and I am buying them all.

I'll try to cover the fiscal cliff issues soon, but you already know my conclusion.

UPDATE -- 11/8/2012 10:30 CST

Both authors of the cited paper have confirmed (via email) the interpretation I explained in this post.  With permission of Prof. Chauvet, I am quoting her email to me:

"Real time probabilities are very noisy, and a little bump of 15%, 20% or even 30% does not mean much in terms of signaling recessions.

Please check the graph on real time probabilities of recession on my site:

https://sites.google.com/site/crefcus/probabilities-of-recession/real-time-probabilities-of-recession

These are the probabilities we get on a month-to-month basis, without data revisions, and without smoothing. As you can see, the probabilities can even be above 50% and a recession does not follow. This is why in my paper with Jeremy we set the rule that the probability would have to be above 80% and for a couple of months before one could call a recession."

Please also read a fine article on this topic by Dwaine Van Vuuren.

 

Comments

[Chris]

From an investor's perspective, I see two games taking place with respect to the business cycle. First, there is the actual business cycle. Investors should define the cycle in a way that best suits their needs. This doesn't mean we all have to create our own models; but it does mean there isn't one "true" model we all must use.

Second is the public perception of the business cycle. This is where work like that of Chauvet and Hamilton comes into play. At first glance, my guesstimation is that the lag between the business cycle dates and when the NBER announces them serves to flatten the cycle. A lag between the start of recession and the announcement of the start date can serve to mute the rate of downturn, as some people maintain the economy is slowing but not in recession. Likewise coming out of recession those who still believe the economy to be in recession will dampen the rate of recovery. Shortening the time between actual date and the announcement of that date could steepen the cycle on both sides of the peak/trough.

[Matthew]

The recession probability indicator doesn't look all that useful. From the charts it certainly doesn't look predictive at all, and while it is highly correlated with the NBER dates, I'm skeptical that they really provide any extra information than just looking at the quarterly GDP estimates and monthly employment. In fact, it really just confirms my pre-existing conclusion that the only way to improve our real-time ability to monitor the economy is by producing monthly instead of quarterly GDP estimates.

[Meteor Blades]

A thoughtful presentation and a great reply to an unthoughtful comment on your presentation.

[jacko]

If it is the case that the margin between the reliability of prediction 18 months out when compared to the actual situation that eventuates is large, with known methodology, isn't is therefore the case that new methodology is more liklely to be able to develop tighter predictions than current methodolgy than would otherwise be the case?

And therefore doesn't that diminish the importance of a proven track record?

As long as you diversified your risk you would in fact be better off giving a larger weighting to the predictions of the new approach as a result - wouldn't you?

I mean, that the future is identical to the past, and thus what has been done before, as long as it has been done a lot, is the best we can get would seem to not be consistent with an acceptance that technology drives the capacity to produce a particular good - wouldn't it?

[jacko]

If you score more points you are likely to win more games.

Therefore if you raise the cutoff point the likelihood of your prediction being right is logically more likely.

In the graph the higher the redline rises the greater is the association with a recession.

The case is proven - isn't it?

[dryfly]

It would be interesting to have an estimate of alpha and/or beta risks - that is rejecting a true test or accepting a false positive. It is really rare in any hypothesis testing process to see both low alpha risk and low beta risk - especially so in 'social sciences' like Econ. Knowing an estimate of that would be helpful in knowing what to make of the various recession calls.

Now having a high alpha or beta isn't bad - sometimes you are willing to accept a few false positives (say throw out good parts) if one missed bad result is a disaster (say a plane falls from the sky). The point though is you should know the relative alpha beta and report it and if you don't know them report that too.

[oldprof]

Steven -- It is on FRED -- the fine data source maintained by the St. Louis Fed. They provide 61,000 different data series, an excellent resource for serious researchers. There is also an excellent charting facility and references to check out the data source. They clearly provide a link to the original study, which I used as the basis for this post. It was available to anyone, as was the data on the authors' web sites.

Complaining that "the Fed" has it on their site is like objecting about some book in the library. The presence on the site is a resource, not an endorsement.

This is really an important lesson, and your comments are helping to emphasize that.

Thanks,

Jeff

[Steven Kopits]

And if so, what's it doing on the Fed's website?

[Steven Kopits]

OK. So there are three criticisms.

One is that this tool is subject to false positives, and therefore is not reliable. Jeff notes: "Look at the chart above and pretend that we were going back to the future in 2005. You could have drawn the line at 15% and said everything that Lance says now. You would have been completely wrong!"

This, of course, is incorrect. You would have been wrong 50% of the time--and right 50% of the time.

Next, you need 80% for three months, but apparently that's not visible on the chart.

And finally, data appears to have been smoothed away (and this appears to be some on-going thing) which means the graph omits certain data points which would be important in our interpretation.

Now, if you have to make this many apologies for a simple visual tool, then it's just not a good tool. Is that what you're trying to say?

[Don Ake]

You need to have a significant recovery before you can have a significant recession! - http://modeltstocktrends.blogspot.com/2012/10/will-economy-ever-get-back-in-sync.html

[Colorado Investment Manager]

Great post and Ritholz approved as I was directed here from his Big Picture blog.

[speakeasy20052000]

Im not an economist, just a network engineer who is trying to teach himself a lot about markets and money.Yes, I plan for you all to belittle and trash me but while I still have the right to free speech ill say what Im feeling. My gut tells me the country will see a DEEP recession and we are on the cusp of going into one. Look at dividend payments on a whole over the last year, the trend is NOT good and very steep on the downside as well because companies have been hoarding cash so they can survive the next downturn. Company earnings have been really bad, just take a look at UPS’s last earning report, even McDonalds took a hit. That right there is cause for concern. The tanking markets are not a knee jerk reaction to the re-election of an anti-business big government socialist, although his re-election hasn’t helped people feel confident in the business world, unless you run an abortion clinic or union based shop. The chatter all over Wall St. is about the next recession…and its on its way regardless..

[Scott Rothbort]

Jeff - as always excellent work. I would like to refresh people's attention to the Hindenburg Omen scare in 2010. That was also faulty data mining research. Unfortunately since the tech bust a decade ago, negative commentary gets more attention without having to pass the legitimacy test than good old fashioned quality research - Scott

[oldprof]

Swiss Maven -- There are many roads to investment success. The Warren Buffett school basically ignores smarty pants guys like you and takes advantage of stocks that represent good values. If you subtract Apple's cash of $125 per share, it is trading at a single-digit P/E multiple. For most of us, that is attractive.

Most people who have tried to trade Apple from the technicals find themselves chasing rallies from the sidelines.

When you have more experience, you will learn to have respect for other approaches and be less dogmatic in your opinions.

Check back with me in a year or so:)

And especially if you want to drop the cloak of anonymity and reveal some long-term performance!

Jeff

[Swiss Maven]

This is a "dash of insight?" Holy Crow. AAPL blew its top at 700, is already down 23% to 540 in just 6 weeks, and you are BUYING it? Can't you see that the macro-trend support has been broken? Any rallies from here out will no longer be “bull market rallies” but rallies in what is now a new technical bear market for AAPL. It will not again see 700 for many a year. My insight is that your insight is worthless.

[oldprof]

scm0330 -- I have often noted that the stock market is not a futures contract on GDP. People form unsound opinions about the economy all of the time.

I can't predict the zaniness of others, but I can do something about understanding the economy. Stocks respond (ultimately) to corporate earnings. Earnings get whacked during recessions. I therefore look at the trend in earnings and then adjust for financial and recession risk.

If you correctly view the recession forecasting methods as identifying where we are in the business cycle, it is helpful in identifying the right stocks and sectors. This may require accepting a little pain until the data finally convinces the skeptics.

Great question, which I should discuss more extensively.

Jeff

[oldprof]

RB - Here is a good way to think about it. I am not a recession forecaster. I do not have a method or any independent research, and therefore I have no track record.

Because of my background, I am very good at reviewing the research methods and conclusions of experts in several different fields. Think of me as a finder of experts.

When someone offers an opinion about recessions (or any similar topic) they should either have a proven record using their own research, or good evidence that they have found others who have strong evidence.

So Josh and Barry, who both are willing to change their opinions with the evidence, are free to draw conclusions. I don't know of any methods that have proven effectiveness for an 18-month prediction!

Jeff

[paul t]

Thank you very much indeed for this and similar posts. You and Krugman have the delightful habit of naming names!

[scm0330]

Jeff,

Thanks for your perspective here. I had a related question.

You spend a lot of time discussing whether or not a recession is in the offing, based on the best forecasting methods you're able to identify. How, if at all, do you relate recession calls to owning stocks? There are periods in time that are not recessionary, but are not rewarding to equity investors. Conversely, some great times to own/increase equity exposure is in the teeth of a recession. (Obviously, identifying inflection points in the business cycle is very important to success here.)

I guess what I'm trying to understand is how to relate your extensive discussions surrounding recession forecasting with the decision to own stocks.

[RB]

Scratch the last line !