October 31, 2003

How Inept Are National Review's Economics Writers?

How inept are National Review's economics writers? They are unbelievably inept. They are so inept as to be completely off the scale of ineptness.

For example, consider Victor Canto of La Jolla Economics. He asks the question:

Victor A. Canto on GDP Forecasting and Supply-Siders on NRO Financial: [W]ho had the closest GDP forecast? I went back and looked at the tables in the January 2003 Wall Street Journal Forecasting Survey, as well as the revisions to these forecasts.... I computed the annual real GDP forecast [of growth in calendar 2003] of participating economists from the quarterly figures and ranked the economists in descending order. The results were pleasing.... Leading the way was John Mueller, a long-time supply-sider, with a 5.32 percent forecast for [real GDP growth in] 2003. Our own Larry Kudlow, at 4.40 percent, was third...

But... But... But... These are forecasts for growth over calendar 2003--forecasts made last January of what the average of the first quarter, second, quarter, third quarter, and fourth quarter growth rates would be. The first quarter annual growth rate was 1.4%, the second quarter was 3.3%, the just released initial third quarter was 7.2%, and the current consensus forecast of the fourth quarter is 3.7%. These four quarterly growth rates average to 3.9%.

Thus, John Mueller's 5.3% forecast doesn't "lead the way"--it's way, way high of what 2003 economic performance is turning out to be. Larry Kudlow's 4.4% forecast is not "third" in anything but it's degree of overoptimism--it is further from what we now expect the real number to be, 3.9%, than was last January's average forecast of 3.7%. To overestimate GDP growth is a very different thing from making an accurate forecast. (Bear-Sterns, however, does look like it is about to hit it on the nose--I would argue on the stopped-clock principle.)

Does Victor Canto understand that a calendar-year growth rate is an average of the growth rates for the four quarters that make up that year? Does Victor Canto understand that a third-quarter growth rate of 7.2% doesn't wipe out the fact that the first-quarter growth rate was only 1.4%? It really seems like he does not--like he does not know what annual GDP growth is.

Here at DeLong World Headquarters, we are rolling in the aisles...

Posted by DeLong at October 31, 2003 03:50 PM | TrackBack

Comments

Brad, I think it is more likely that they have little interest in telling the truth. Fortunately their not running the country, oh wait...

Posted by: Jeff Smith on October 31, 2003 04:33 PM

____

Hm. So maybe I should train to become a conservative economics writer. Apparently I have a good enough technical foundation already. Maybe Tom Friedman will let me borrow some of his slogans and retrofit them for economics.

Posted by: Zizka on October 31, 2003 06:12 PM

____

La Jolla economics?

Posted by: oneangryslav on November 1, 2003 12:22 AM

____

What's this 'let', Ziska? Just borrow them. If he complains, say that you got them from (insert name of obscure ethnic group here) traders, while traveling through (insert random contry name here). Heck, choose a city which isn't in that country. See if 'world traveler' Friedman spots it. I'd bet he wouldn't.

Posted by: Barry on November 1, 2003 05:19 AM

____

Art Laffer wrote a similar piece 20 years ago. His model had a lower forecast error for a single quarter. No comparison of forecast errors for an extended period at all. You can't even define RMSE using one observation.

Posted by: Harold McClure on November 1, 2003 06:51 AM

____

the its/it's error in the post is really jarring.
Could you fix it?
I'll shut up now.

Posted by: Anatoly on November 1, 2003 08:03 AM

____

Speaking of rolling in the aisles at economic coverage, I'm here in Chicago now and I catch the front page of the Chicago Tribune all the time. Yesterday's headline proclaimed that the economoy is now "Red Hot!" and surging. What's the reason? I didn't bother to pay for the paper and find out, but the mystery was solved by the front page of this morning's Red Eye (the Tribune's tabloid spin-off).

The headline was "Bling Bling!" and it showed young women lining up to buy new shoes. According to the article, we all have so much money lying around now thanks to the Bush tax cuts that everyone decided to go shopping with the cash last quarter. That, according to the Red Eye, explains why the economoy is now "red hot" and "surging." ;0)

Posted by: Hannibal on November 1, 2003 10:56 AM

____

Somewhat off topic: During the big spike of oil prices in August, I believe, I had an argument on Brad's web site with a couple of guys named Joe about oil companies and profits.

The Joes, one far more than the other, seemed to think the oil companies' line about broken refineries was the reason for the spike--good old supply and demand.

I on the other hand pointed out that we should watch the results of the quarterly earnings. Well, here we come the first offerings from the Four Sisters: ChevronTexaco announced a big spike in profits despite all their talk of repairing refineries.

It's not as if there is a lot of competition in the oil industry in terms of refining. It's not as if they raised prices to pass off costs of repair. These structural issues play a role that show why a fealty to 'free market' theory is naive.

We also learned once again that the oil companies use of "just in time" inventory allows them to play this game since they know there is a largely captive audience of consumers.

When one cuts past the bombast, the oil companies are simply gouging, all within the technical protection of the law, I might add.

The loudest mouth Joe tried to change the subject at the time by talking about stock prices, as if anyone can predict why people buy or sell particular stocks. Even now ChevronTexaco stock is not seen as a great buy per some Wall St. analysts. Why? beats me...

The issue, however, remains the same and it's about oligopolies and the power to manipulate the market more often than not.

As Robert Klein put it 30 or so years ago, when talking of the first oil crisis, "The oil companies believe in supply and demand. Sure they do. 'We have all the supply, and we can demand whatever the fuck we want!'"

There are ways to grab that excess profit for the commonweal, and spend it on renewable energy for example. But that takes a political will that is sorely lacking.

Now back to Scott Peterson and Kobe Bryant and fires, oh my!

Posted by: mitchell freedman on November 1, 2003 01:07 PM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 1, 2003 10:26 PM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 1, 2003 10:30 PM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 1, 2003 10:34 PM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 1, 2003 10:41 PM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 1, 2003 10:52 PM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 1, 2003 11:00 PM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 1, 2003 11:30 PM

____

I am not inclined to defend supply-side salesmen, but it is possible that the author made an honest mistake.

If you asked me how quickly the economy was growing this year, my first instinct would be to cmpare the level of output (expected) in Q403 with the level in Q402. And if I were not aware of the frequency at which GDP is measured, I might go with a December vs December comparison or even New Year's Eve vs New Year's Eve. I would not instinctively compare the full year of 2003 with the full year of 2002, as that would seem to throw some of the late 2002 growth into the 2003 figure.

I know full-year growth rates are the conventional measure in some circles and I am not saying that they are misleading. I am saying only that it is reasonable -- and not necessarily evidence of deep incompetence -- to assume the year's growth rate refers to growth achieved during the year.

So I would not be so harsh. I would say that the author failed to distinguish between a full-year growth rate and a Q4/Q4 growth rate when evaluation the economists' accuracy. Minor error. It doesn't mean he is a fool, although an uncritical faith in supply-side economics might....

Posted by: Gerard MacDonell on November 2, 2003 04:42 AM

____

Jeez, I should just drop out of school, or at least switch my major from economics to art history or sociology or something.

Posted by: saranwarp on November 2, 2003 10:31 AM

____

Mitchell: The evidence that you cite is exactly the evidence that the refining industry does NOT operate monopolistically. Let me explain in as much English as possible.

Say that it were--that the firms had colluded to produce as if they were a monopoly. Then an unanticipated supply shock in either direction would have reduced profits. (Otherwise they're not acting like a monopoly--monopolies maximize profits for the entire industry at the expense of consumers.). But if a negative supply shock increased profits, it means that the industry had produced more than the monopolistic quantity before. So we can rule out monopoly.

This does NOT mean that they produced the competitive quantity--some kind of Cournot-like outcome is possible. My guess is that refining would look more competitive than monopolistic but I would have to develop a model and go through some data to know this for sure. Knowing what happens when we try to put price controls on energy (long lines, not increased supply) we know that this is a really, really bad idea. So this is my best hunch for intuition, and anything more is a data question.

Regarding Canto, what is it about nice places that draws hacks to them? I mean, there's real economics being done just a short ride up Torrey Pines Road. And I'm still puzzled why NR doesn't get good economists to write for them--it isn't like good conservative economists are that scarce. I know that the NR folks are a bunch of English majors but they're not unintelligent. Aaaaargh!

OK, back to shivering down here in La Jolla.

Posted by: Chris on November 2, 2003 04:53 PM

____

Okay, I admit I know little about the BEA's methodologies. However, my question is can you literally average annualized rates like this and get anything meaningful? The 7.2% is an annualized rate, not a growth rate for the third quarter itself. The growth rate from the second quarter to the third is 1.74% (GDP Q2 = 9629.4, GDP Q3 = 9797.2). The annualized rate is not just 1.74% * 4 = 6.97% != 7.2%.

Posted by: Steve on November 2, 2003 08:37 PM

____

A slight misstatement by Brad. Annual growth is not the average of quarterly growth. That instead gives you the percentage change in Q4 GDP over Q4 GDP a year earlier. Annual or y/y growth depends on your "jumping off" point. Suppose GDP this year, in each quarter, is 97, 98, 99, 100, for a total of 394 and an average of 98.5. Then GDP next year is 100, 100, 100, 100, for a total of 400 and a quarterly average of 100. Annual growth is 1.5%, despite the fact that growth in each quarter was zero.

Posted by: Matt on November 3, 2003 07:31 AM

____

Hannibal

You are right to distrust anybody who thinks "bling bling" includes shoes. "Bling bling" is mostly jewelry - flashy stuff that makes noise when it knocks together. A far narrower category than fashion accessories in general and definitely not - repeat not - an onomatopoetic reference to cash register noise.

Posted by: K Harris on November 3, 2003 10:16 AM

____

Victor Canto does know what annual GDP growth is. The point he was making is that most economists recently revised their forecasts upwards. Looking back at the January forecasts it is interesting to note which economists predicted the strongest growth for 2003. They are listed in a table in his article.

Brad is correct that if the consensus forecast comes in at 3.7% (and last quarter’s growth holds at 7.2%) then growth for the year will be 3.9%. What the table then tells me is that of the 55 economists surveyed, 48 expected the economy to grow at a slower rate.

And what if the 4th quarter grows faster than the consensus forecast? Again assuming 7.2% holds for the 3rd quarter, a 4% rate will give us 4% growth for the year, a 5% 4th quarter will give us 4.2% growth, and a 6% 4th quarter will give us 4.5% growth. We’ll have to wait and see what the final numbers turn out to be. We’ll also have to wait and see how many economists revise their forecasts in the months to come.

One last point: What is not mentioned in DeLong’s critique is who had the closest forecast for the 3rd quarter at the beginning of the year. As Dr. Canto points out, when ranking the 3rd quarter forecasts from highest to lowest Mueller was closest (6.7%), Wesbury third (4.8%), Kudlow fourth (4.5%), and Malpass/Ryding eleventh (4.1%) out of the 55 economists.

Interesting food for thought if nothing else.

Posted by: Peter Mork on November 4, 2003 09:25 PM

____

Post a comment
















__