Friedrich von Blowhard writes:

Dear Blowhards,

I noticed a current (October 10, 2007) story posted on the Bloomberg.com website by Joe Richter and Alex Tanzi, "U.S. Economy to Avoid Recession as Housing Sinks, Survey Shows."

As the story leads off:

The U.S. economy will skirt recession even as the housing slump takes a bigger bite from growth, according to a survey of economists.

The economy will grow at an annual rate of 1.8 percent in the fourth quarter, 0.4 percentage point less than forecast last month, according to the median of 71 analysts participating in Bloomberg News's monthly survey. Estimates for the first six months of next year were also reduced.

Well, that's not fabulous, exactly, but it sounds like we're going to dodge the bullet, right?

Unfortunately, maybe not. My recent reading of economics bloggers has tipped me off to a painful truth: economists are notoriously bad predictors of recessions. Indeed, they seem to have a systemic bias for what might be termed rose-colored glasses.

Nouriel Roubini, a Serious Economist (see his Wikipedia profile), brought this up in a post (which you can read here) from a year ago in which he defended his call that the U.S. would see a recession in 2007:

These days I get asked daily in interviews and talks: "How do you explain that the market consensus is still so far from your recession call for 2007? Why does almost everyone on Wall Street believe that there will be no recession? What do you know that they do not?"

Dr. Roubini also linked to a IMF Working Paper by Prakash Loungani, , "How Accurate are Private Sector Forecasts? Cross-Country Evidence From Consensus Forecasts of Output Growth" from April 2000. This study systematically examined forecasts for 63 countries for the period from 1989-2000. One of its conclusions:

How well do forecasters predict recessions?

The simple answer is: "Not very well." Only two of the 60 recessions that occurred over the sample were predicted a year in advance, two-thirds remained undetected by the April of the year in which the recession occurred...This predictive failure is well-known in the case of U.S. recessions; the contribution of the work here is to show that it is a ubiquitous feature of growth forecasts. This predictive failure could arise either because forecasters lack the requisite information (in terms of reliable real-time data or reliable models) or because they lack the incentives to predict recessions; further work would be needed to discriminate between these two classes of theories. [emphases original]

Granted, there seem to be fairly good theoretical arguments that cast doubt on our ability to ever accurately predict anything as complex as a national economy; still, I read Mr. Loungani's remarks as a strong hint that there are pressures exerted on economic forecasters to not call 'em as they see 'em.

(Perhaps economists, being highly trained professionals, are unable to bring themselves to follow the three simple rules of intelligent forecasting: (1) Don't label your axes, (2) Don't date the forecast, and (3) Don't sign your name).

Anybody got any insight into what those pressures might be?

Cheers,

Friedrich von Blowhard

Friedrich is right. The investment analyst/economics profession always likes to paint the bright picture of the near-term economic future. This is because their income depends on it.

In 1989, the economists talked about the end of the business cycle. They also talked about a "soft landing". Job-wise, the economy tanked in 1990 and stayed bad until 1993 or so. The job market did not pick up until around 1995 or so.

In 2000, the economists once again talked about a "soft landing". The following 4 years made it clear that they had their heads up their arses.

In 2001, the telecoms industry crashed. An investment analyst I talked to a lot at the time described the situation as there being "no visibility". I said that there is perfectly good visibility, but that it was all bad. The start-up I was in died the following month.

Right now, lack of job creation suggests that we are already in a recession.

Posted by: Kurt9 on October 10, 2007 12:31 PM

Hi Friedrich, I can't speak to the pressures exerted on forecasters, except for the ONE pressure they are free of: being right.

This reminds me of a story about the 1992 Election. Newsweek or Time ran a story saying that the consensus of the forecasters said that 1.) Clinton would win and 2.) he would get about 48-51% of the vote. (This was done in October). Well, they were right about #1 but way off on #2 (I think he got like 43%).

Well, they also mentioned that a Betting Market had opened on the Election and their they predicted that 1.) Clinton would win and 2.) he would get about 43% of the vote. Actually, this was the "line" for the bookies.

In other words, the people who have to put their money where their mouth is were much better at predicting (on the aggregate).

This is something that Tim Harmon talks about a lot over at MarginalRevolution. The importance of Prediction Markets (i.e. Gamblers and Investors).

Actually, Steve Sailer talks about this as well. Many people who make political/sports/economic predictions are never held accountable to those predictions. If you see that some major economist has made 20 predictions in the last 10 years and he has been close 4 times, well, you would probably start ignoring him.

Posted by: Ian Lewis on October 10, 2007 12:31 PM

My guess is that it's no so much a matter of not calling 'em as they see 'em, as much as it's not seeing 'em at all. People can draw greatly differing conclusions from the same data sets, depending on their own biases, which is something we struggle with a lot in the sciences.

Posted by: Derek Lowe on October 10, 2007 12:45 PM

I agree with Mr Lowe about 'not seeing them'. Investment firm types who's firms deal with the public might always have a bullish bias that might be exarcerbated in individual stocks for the sake of investment banking/underwriting business, and since the public generally doesn't sell short. But taking away all that, what's the best predictor of tomorrows high temperature if you don't know anything? Today's high temperature. What's the best predictor of next quarter's GDP growth? Last quarter's.

Lastly, being right when everyone else is wrong is a good thing for a forecaster's reputation, but it's not as good as being wrong when everyone else is right is bad for a forecaster's reputation. I suppose that goes for polling too.

Posted by: j mct on October 10, 2007 2:56 PM

A safer alternative is to ignore individual predictions and look at the collective predictions of millions and millions of people, as reflected in stock market prices.

Did you know that the stock market has predicted ten of the last five recessions?

Posted by: Peter on October 10, 2007 3:53 PM

What people should really take away from this is that the economics profession isn't much of a science at all, since it isn't predictive. All sciences seek strong correlations between cause and effect--the ability to replicate and predict results. Despite the mathematical masking, economics is really more of a sociological and psychological field than a science. Like sports, forecasts are only made for financial gambling, which is yet another indication that it isn't a science.

Why anybody gives such "learned" guessers much creedence is beyond me. I guess people want some way to deal with the future, especially with regard to money. The simplest way to deal with that is to spend less than you make, save, and buy a variety of assets. Of course, this becomes harder all the time.

Economists to me are no better than sociologists--the same crew always giving sob stories and excuses for one favorite group or another. In economist's cases, the "free market" or government is the poor, pitiable group always given the benefit of the doubt--in other words, they make excuses for the elites who move the markets and determine money creation.

Big surprise there, eh?

Posted by: BIOH on October 10, 2007 4:10 PM

"Did you know that the stock market has predicted ten of the last five recessions?"

I am not sure I know what that sentence means, but I love it.

Posted by: Ian Lewis on October 10, 2007 4:11 PM






Post a comment

Name:


Email Address:


URL:


Comments:


Remember your info?