Unemployment, U-6, and U-turns

Ever since Fed Chairman Ben Bernanke said a few weeks ago that we may be glimpsing the first “green shoots” of recovery, it’s been a cockeyed-optimism fest among media commentators.  As always, developments in the stock market have gotten way too much attention (Paul Krugman wryly noted on TV that the stock market has predicted six of the last one recoveries), and they’ve done much to fuel the optimism.  As of yesterday’s close the S&P 500 is up more than 30% from its low of 676.53 just two months ago (March 9). Sucker’s rally or not, it’s moving in the right direction.

Commentators have also seized on the BLS’s latest unemployment release as a good tiding, which would seem like a reach given that the official unemployment rate rose from 8.5% in March to 8.9% in April.  The cliche of the day is that employment has “bottomed out.”  Let’s crunch the April numbers:

  • Employment on nonfarm payrolls fell by 539,000, and private sector employment fell by 611,000.  Total job losses since the recession began in December 2007 are nearly six million (5.7 million to be exact).
  • The number of long-term unemployed rose by 498,000, to a total of 3.7 million.
  • Job losses were substantial in every major industry group except health care and government (and the government jobs were mostly temporary 2010 Census jobs).
  • The unemployment rate for African Americans jumped from 13.3% to 15.0%, its highest since 1986.

So where’s the good news?  Well . . .

  • The BLS reports on two measures of employment:  establishment-data (employer payroll) employment, and household-data employment (which I’m guessing includes the self-employed in addition to those on a payroll).  Employment actually rose slightly using the household data, by about 120,000.  The household sample is the one they use for the unemployment rate, so the rise in the unemployment rate was from a large rise in the number of jobless who said they were actively looking for work.
  • The not-seasonally-adjusted (NSA) data (which I tend to prefer for unemployment — see previous posts) actually do show some overall improvement.  The NSA unemployment rate was 0.4% point worse than the official unemployment rate in March, but in April it’s 0.4% point better, 8.5% instead of 8.9%.  (The seasonal adjusters at the BLS evidently agree with T.S. Eliot that April is a cruel month.)
  • The U-6 unemployment rate (another hobbyhorse of mine, it includes discouraged job-seekers and involuntary part-timers as well as the regular unemployment) improved even more, dropping from 16.2% to 15.4% (NSA data).   Granted, those numbers are still ghastly  (and the seasonally adjusted U-6 rates were a couple ticks worse in April than in March).
  • But what most optimistic-minded economists and media cheerer-uppers are pointing to is the fact that the huge drop in payroll employment is actually a little less huge than the drops in the previous two months.   As usual, economists prefer to put it in terms of the incomprehensible, breaking out the calculus and saying “the second derivative of unemployment is positive.”  While it’s true that employment was falling less rapidly in April than in March or February, one data point does not a trend make.

This whole “second derivative” bit irks me not merely as another case of economists being willfully confusing, but more because it implicitly assumes that the economy must follow some well-behaved mathematical function like a sine wave or a quadratic function.  It just ain’t so.   Economists who get all misty about the second derivative of employment or GDP are imagining a curve on a blackboard, not a real-life economy whose fluctuations are unpredictable and (gasp!) not curvilinear.   When they assume, contrary to historical data, that all recessions are U-shaped,  they end up making bogus predictions on the basis of two or three data points.  Look elsewhere for your forecasts.

UPDATE, 29 May 2009:  Gillian Tett of the Financial Times offers a helpful corrective to all the talk about U-shaped and V-shaped (the optimists’ special) recessions.  (Full access is available for free if you register with the FT.)


Tags: , , , , , , ,

3 Responses to “Unemployment, U-6, and U-turns”

  1. democommie Says:

    Fortunately, I HAVE bottomed out and did not eat my stud-driver, so that’s to the good.

    I have to say that you are one of the few economists that I actually trust.

  2. Ranjit Says:

    Thanks, Terry. Here’s a “green shoot” of recovery- we finished the port! And are ready to move on to some of the wines we picked up at yesterday’s tastings. You’ll have to come over sometime.

  3. democommie Says:

    Glad you enjoyed the Port. Of course now that you’ve “Plamed” me to your vast audience of readers I will have to come over to your house and kill–a bottle or two!

    I gotta tell ya, I’m amazed, if not amused, at how many people are telling me that Obama is not who they thought he was. I thought he was a politician and he turns out to be one. I still think he was not just the best, but really the only, alternative.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: