Dean -

This question has nothing to do with what you wrote, but is useful and interesting I believe. Today we are seeing reports of "10.2% unemployment, the worst since 1983". However, my understanding is that somewhere during late Reagan/early Clinton, the official computation of unemployment was changed to count less of the underemployed and long-term employed.

How does this 10.2% figure today *really* compare to the 1983 figure, normalizing or otherwise accounting for differences in computation? Also, if that number is substantially different, why do journalists (in the vein of "Beat the Press") not report what the normalized/adjusted numbers are for comparison?

A perspective of some current economic data reporting methods compared with historical data reporting methods may be found at shadowstats.com published by John Williams.

I will leave the second part of your question to those in the economic profession.

sesli sohbet sesli chat

Posted by: MarkJ | November 6, 2009 11:41 AM

Thanks!

I am genuinely curious about what relative "real" unemployment is in the sense of "adjusted for computation differences".

(Dean Baker may be amused to note that the two letter "captcha" text I must now type to post under his writing is "alarming hero". Don't let it go to your head, Dean -)

Guy in Seattle.
To the extent that methods were changed, the historical data series maintained by the BLS and other Govt depts. would be adjusted to conform (as best they can).

So when media reports that the rate is the worst since 198x, they should be comparing apples to apples. if not be sure that Dean will nail their misleading rear-ends.

Posted by: Joe k | November 6, 2009 6:11 PM

Ah, I see, the government agencies actually employ professionals at the levels where actual work gets done, and the reporters are kept away from the pitfalls of reporting unadjusted numbers.

Thanks -

In terms of the overall economy you are correct. In terms of how many retail workers keep their jobs (who haven't lost them already), it is significant.

Here's something on the U-3 and U-6--http://blogs.wsj.com/economics/2009/11/06/broader-unemployment-rate-hits-175/

Apparently the present U-6 only goes back to 1994, but the old equivalent computed the rate at 14.3%. The equivalent for October 2009 is 14%, so it's close, but not quite there.

See the employment/population ratio - the recent drop is obviously worse than any time since WW II.

http://research.stlouisfed.org/fred2/series/EMRATIO?cid=12

If you take the prison population into account, the unemployment rate would have to be adjusted up about 1% or so based on what I've heard. The US per capita incarceration rate has more then doubled, maybe triple. I can't find the statistics I'm looking for but it's somewhere in that ballpark. NJFAC said this in an article earlier in the year: "NJFAC has always pointed to the prison population as part of the unemployment problem. If these people were added to our monthly data, the rate would rise by roughly 1-1/2 percentage points. For example, for February 2009, adding the prison population would increase unemployment from 8.1% to about 9.6% that month. Those in jail are among the most likely to have had employment problems and to continue to have them when released. One in 31 U.S. Adults are Behind Bars, on Parole or Probation."

Also, CEPR had a report earlier in the year pointing out that after demographic adjustments are made to put 1982 and 2009 on an equal footing, you would have to adjust today's unemployment rate up by 1.4%. http://www.cepr.net/index.php/press-releases/press-releases/us-unemployment-understated-in-recent-data/

When you add the adjustment for the prison population, it leads to an enormous adjustment of 2.5 or 3 percentage points.

Jesus I left some atrocious typos in that message. Please ignore them and don't assume that I'm retarded.

Nice for your sharing! Welcome @ : Nike Dunk