Friday, November 7, 2008

How Bad is the Labor Market?

BLS reports:

Nonfarm payroll employment fell by 240,000 in October, and the unemployment rate rose from 6.1 to 6.5 percent

While a 240 thousand drop in just one month for the payroll survey measure of employment is quite significant, how on earth did the unemployment rate increase this much? Only a small part of the story comes from the labor force participation rate, which rose from 66.0% to 66.1%. The household survey shows that measured employment dropped by 297 thousand, which lowered the employment-population ratio from 62.0% to 61.8%. This compares to an employment-population ratio of 63.4% as of December 2006 and an employment-population ratio of 64.4% as of December 2000. The need for aggregate demand stimulus is clear and it is good to see that President-elect Obama is quickly assembling his economic team.


Sandwichman said...

The August and September unemployment totals were also revised upward by substantial amounts. 54,000 for August and 125,000 for September. My suspicion is that the October number will also be revised upward by somewhere in the order of 100,000 next month or the month after. The BLS is using a model that clearly can't handle the magnitude of the job decline.

Anonymous said...

Tell me: are these unemployment numbers seasonally adjusted in some fashion? Because I imagine the absolute level of employment should be much higher in October due to the approaching holidays bonanza.

If the absolute numbers got so much worse in October, it must be a real disaster.

Daro said...

Aren't people dropped off the unemployment tally if they've been out of work 6 months or longer under the risible theory they're no longer looking for work? From the wailing I hear in blog commentors from country towns it seems the unemployment number is much, much worse..

Anonymous said...

No, in the household survey thing, I understand, they actually ask people if they are looking for a job or not.

ProGrowthLiberal said...

abb1 - the reason I focus on the employment to population ratio rather than the unemployment rate is precisely the problem you note. The unemployment rate can fall even if the employment to population rate falls if the labor force participation rate drops even more.

Anonymous said...

According to BLS' "Alternative measures of labor underutilization" (, if you count "Total unemployed, plus all marginally attached workers, plus total employed part time for economic reasons" then the unemployment rate (seasonally adjusted) is 11.8%.

But I seem to remember reading somewhere that the Clinton admin in the 90s redefined "marginally attached" to exclude those who gave up looking for a job more than a year ago. So, in reality it's even higher than 11.8%.