January 14, 2007

Backing and Filling: Forecasts for 4th Quarter GDP Growth Are Revised Upwards

Lordy, Lordy (HT Instapundit; Update — Welcome Instapundit readers!), the economic “surprises” don’t stop. The report is from AFP:

Economists are hastily upgrading their forecasts for the US economy after a series of surprisingly strong reports suggesting the so-called “soft landing” may be over and growth is accelerating.

Over the past week, surprises have come in stronger-than-expected reports on US job creation, the trade balance and retail sales — all key contributors to economic activity.

Lehman Brothers chief US economist Ethan Harris on Friday boosted his forecast for fourth quarter 2006 growth to an annualized rate of 3.3 percent, a leap from the firm’s prior call for just 2.0 percent growth.

“After slowing in November, the economy seems to have regained its stride,” Harris said.

….. The latest data defy predictions that the slump in real estate would filter into other areas of the economy, notably consumer spending.

The latest data showed US employers added a healthy 167,000 new jobs in December (196,000 with revisions to prior months — Ed.), with unemployment holding at a low 4.5 percent. Average wages were up 4.2 percent annually.

That’s what happens when economists like Ethan Harris read too much Paul Krugman and Rex Nutting, and not enough of yours truly (see cautious “3% or more” prediction about halfway through the post).

Even the above report doesn’t get it right, because the “slowing” in November cited by Mr. Harris isn’t supported by facts he and other analysts should have known. While the 14% of the economy represented by the manufacturing sector barely went into contraction mode in November, the other 86% representing services and everything else was in serious expansion mode (that was also a reported “surprise” at the time). That, and November’s addition of 174,000 jobs (132,000 plus revisions to prior months), render the idea that there was any kind of “slowdown” in November quite silly.

Whether these upward revisions, or for that matter the late-January preliminary result, get wide coverage beyond the business pages remains to be seen. Prognosis: doubtful.

As to 2007, Mr. Harris provides the formerly Mainstream Media plenty of cover with his skeptical optimism:

Harris at Lehman Brothers said “a natural question is whether the solid fourth-quarter growth is a fluke.”

He said data could be distorted by things such as the introduction of the new Windows Vista operating system, shifts in holiday shopping patterns, problems in the auto sector, the plunge in energy prices and unusually warm winter weather.

“However, looking closely at these factors on net they argue for more, not less strength in the first quarter of 2007,” he said.

Of course, I fully realize that there’s no victory dance until the official announcement of preliminary GDP, which won’t be until (zheesh) January 31, followed by two more revisions in late February and March.

Cross-posted at NewsBusters.org.


UPDATE: In a surprising development, Dr. Sanity beats Weapons of Mass Discussion to the inevitable punch line — “Personally, I think we should blame Bush.”

UPDATE 2: Before someone accuses me of it, I’ll plead guilty — to inadvertent Memeo-bombing. This post and the cross-post at NewsBusters are both listed as blog links at the Memeo entry for the AFP article that led to this post; in fact, right now, they’re the only ones, which constitutes a Memeo-post monopoly.

UPDATE 3: Anchoress links back, with keen insight into my R-Rated mind as it attempts to run a PG-13 blog.



  1. Some of these so-called forecasters are so bad at predicting economic results that you really have to wonder why their calls aren’t more often taken as a clear indication that the opposite going to happen.

    More seriously, you really have to question why these analysts are really paid much at all when they say things like:

    “… data could be distorted by things such as the introduction of the new Windows Vista operating system, shifts in holiday shopping patterns, problems in the auto sector, the plunge in energy prices and unusually warm winter weather.”

    Translation: “we really have no idea at all what will happen, so let’s throw up a bunch of garbage and see if anything sticks!”

    The truth is the economy may be driven a lot more by inertia than anyone really appreciates and it really takes some kind of government action, or some other, more natural, disaster to derail the economy’s growth prospects.

    Comment by Ironman — January 15, 2007 @ 9:08 am

  2. #1, I like and agree with your “translation.” Went to your links; hope you’re wrong about 2Q07, though you may have updated since.

    Comment by TBlumer — January 15, 2007 @ 9:17 am

  3. The more the median and mean diverge, the more meaningless is a total. As bad as a good economy is for us, I see no future in predicting anything but.

    Comment by Tracy Coyle — January 15, 2007 @ 9:44 am

  4. Predicting economic activity is futile. Makets are the collective activity of the individual purchasing decisions of hundreds of millions of people. Given, that nobody knows for sure what and how much they will be buying or investing in 3 months, let alone what 300 million other folks will be doing, you might as well throw darts at numbers to come up with the estitmates.

    Comment by Harry Calamari — January 15, 2007 @ 11:47 am

  5. #4, Points well taken. The frustration here is that some of the data I’ve cited has been available for weeks, yet the econs weren’t changing their outlook. They also got sucked into the media “slowing” meme, which has been suspect for some time.

    Comment by TBlumer — January 15, 2007 @ 1:06 pm

  6. [...] Surprising to who? We’ve been in a recovery since, oh about 2003 and it’s now 2006. See BizzyBlog, who asks the appropriate questions: [...]

    Pingback by NixGuy.com » The Economy is Good… — January 15, 2007 @ 2:54 pm

  7. #2, The “Climbing Limo” forecasting method only draws on official GDP data – only revisions to the reported data can change the forecast. For what it’s worth, the method tends to forecast a bumpier path forward than what Real GDP actually takes in relatively stable economic conditions, such as we’ve enjoyed recently.

    My best guess is that the forecast for Real GDP in 4Q06 is overshooting what the actual results for the quarter will be. This will make it much more likely that if the 2Q07 data comes in as forecast, it will not be as “catastrophic” as it appears compared to the other forecast values.

    In any case, it’ll be two-three months before the 4Q06 data is finalized and we’ll know for sure. The first preliminary GDP estimates should be out later this month.

    #5, Bad news sells better and provides more screen time.

    Comment by Ironman — January 16, 2007 @ 9:05 am

  8. #7, my theory is that they’re underestimating a bit even now — but to channel Berman, that’s why they release real data.

    Climbing Limo would always seem to forecast trouble 2-3 quarters out, if I understand your explanation correctly, because it needs new data to justify any projected increase.

    Comment by TBlumer — January 16, 2007 @ 10:49 am

  9. #8, The method doesn’t necessarily forecast trouble 2-3 quarters out (remember, that high prediction for 4Q06 Real GDP was predicted 3 quarters in advance!) It only points where the data says to point!

    As a result, the method is very data dependent. One of the things I’ve had to learn to resist doing is to plug the preliminary GDP figures into the calculations as they’re released. There is a lot of volatility in the forecast values and until the source data is finalized, the forecast values can vary wildly.

    On the other hand, The Skeptical Optimist, who created the method, reports that its track record over the past 40 years is such that it provides a better prediction of economic performance 55% of the time (compared to the Bureau of Economic Analysis’ method), and since 1993, more than 63% of the time.

    Comment by Ironman — January 17, 2007 @ 1:03 am

  10. #9 thanks, I would guess that the GDP’s general track record of usually getting revised upward by more than at least I would expect on rev 2 and rev 3 would drive you nuts if you didn’t wait until the finalized number to update Climbing Limo.

    Comment by TBlumer — January 17, 2007 @ 1:15 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.