>I'm not sure it's attributing "evil motives" for Crudele to think these
>particular adjustments for firm startups/deaths are inaccurate.
>
>He's made his criticism, noted that the job creation numbers last July were
>unusually low because of these adjustments, then confidently predicted that
>the job numbers today would be significantly below estimates by other
>economists.
So did I.
The BLS put out a research note this morning that in part comments on those adjustments. It found they were tracking reality quite well. So he's wrong on the accuracy issue, whatever the motives.
But he's got a long history of this sort of thing. A year ago he was making claims about the CPI that were seriously wrong - and which proved he either didn't know what he was talking about or didn't care. And here's an egregious bit of Crudele from earlier this week:
>HOW COMMERCE DEPT. MADE THE RECESSION DISAPPEAR
>
>By JOHN CRUDELE
>
>August 3, 2004 -- THE U.S. Commerce Depart ment said "abracadabra"
>and made the 2001 recession disappear.
>
>While everyone last Friday was focusing on the freshly released
>estimate of the nation's gross domestic product for the second
>quarter of 2004, the real action was taking place in the revisions.
>
>As I told you in last Thursday's column, the department's Bureau of
>Economic Analysis was not only scheduled to give the most recent
>snapshot of the current economy but was also going to revise the
>entire GDP picture for the Bush administration.
>
>And guess what?
>
>With a little snip here and a tuck there, the department made the
>recession - at least as it is traditionally defined - disappear with
>the skill of a plastic surgeon getting rid of a patient's annoying
>worry lines.
>
>The magic occurred in the numbers for the second quarter of 2001.
>
>Up until last Friday, the BEA had been reporting that the nation's
>gross domestic product fell 0.6 percent during that quarter.
>
>That was a problem because the first quarter of 2001 had been down
>0.2 percent and the third quarter of that year had been lower by 1.3
>percent.
>
>Economists traditionally consider the economy to be in a recession
>if there are two consecutive quarters of contracting GDP. And until
>last Friday's revisions, the U.S. met the qualification during 2001.
>
>Not anymore.
>
>The government now says the GDP didn't really fall 0.6 percent in
>the second quarter of 2001 but actually rose 1.2 percent. So, the
>string of three losing quarters was suddenly broken up by a winner
>in the middle.
>
>How lucky!
>
>I asked economists at the BEA how a GDP number could change so
>dramatically so many years later.
>
>One of the big reasons, I was told, is that the Commerce Department
>suddenly decided to use some seasonal adjustments for car sales that
>were already being employed by the Federal Reserve.
>
>"The Fed includes special holidays" in the adjustments, a BEA
>economist told me. "We liked their (seasonal adjustments) better."
>
>How fortunate!
>
>The National Bureau of Economic Research, a private group, has the
>unofficial role of declaring recessions. It's unlikely that the
>group will change the historic record because of Friday's changes.
>
>But the move could play well on the campaign trail.
He's basically accusing the BEA of doing the Bush campaign a favor, even though he doesn't come out and say it. But he's wrong on several accounts. There were many reasons for the upward revision to 01Q2 aside from the changes to seasonal adjustment for auto sales for the upward revision - that was one of several factors. (And the Fed's technique may well have been better - the stat-heads talk to each other all the time, and learn from each other.) At the same time, they revised down the estimates of nondurables and services consumption. Also, the growth numbers out of the recession were revised downward. The net effect was mostly a wash. The personal savings rate was revised down significantly. And so were corporate profits. But Crudele just wants to argue by innuendo that the BEA made the recession disappear.
Oh, and the two consecutive months of decline definition is in wide use, but it's explicitly rejected by the NBER.
Doug