How bad is 10% unemployment?

Nothing substantial to debate, but this very short commentary makes an interesting point that gives some context to the high unemployment rates we face in the US.

http://www.brookings.edu/opinions/2009/1207_unemployment_gale_harris.aspx

Basically, if you adjust for the differences between the age distribution of the labor force today and that in 1982 (when we had very high unemp), the comparable unemployment rate today would be over 11%.

(crickets chirping)

But do they apply the same criteria to being unemployed now as they did in 1982?

I don't know how useful this thought experiment is. It assumes that unemployment rates within each class would not change along with the age distributions.

If we believe that markets are segmented by age group, then an increase in the proportion of people in an age group would lead to an increase in that group's unemployment rate (tightness would fall).

So plugging an experimental age distribution into the current unemployment rates by age group doesn't make much sense to me.

Originally posted by KazetNagorra
But do they apply the same criteria to being unemployed now as they did in 1982?

Originally posted by Palynka
I don't know how useful this thought experiment is. It assumes that unemployment rates within each class would not change along with the age distributions.

If we believe that markets are segmented by age group, then an increase in the proportion of people in an age group would lead to an increase in that group's unemployment rate (tightness would fall).
...[text shortened]... distribution into the current unemployment rates by age group doesn't make much sense to me.

Originally posted by telerion
According to the author, William Gale, (who I admire as an economist), the relationship between unemp and age has remained roughly the same over the years. All else equal, the rise in the proportion of older workers would tend to dampen unemployment rates.

Originally posted by telerion
According to the author, William Gale, (who I admire as an economist), the relationship between unemp and age has remained roughly the same over the years. All else equal, the rise in the proportion of older workers would tend to dampen unemployment rates.

Originally posted by zeeblebot
(crickets chirping)

does the age data only include legal employment, and the unemployment survey data also include illegal employment?

Originally posted by Palynka
Mmm... If you plug the age distribution of year X and get a different unemployment rate of year X, then all the changes come from changes in unemployment rates of age groups.

I'm sure Gale knows more about this than me (not being a labour economist) but IF we admit there is labour market segmentation by age, then why on earth would we expect changes in tightness to not affect unemployment rates?

Originally posted by telerion
If you change the population weights in the 2009 unemployment calculation, the statistic can change even if the likelihood of being unemployed conditional on age remains the same across years. The change does not come from unemployment in each age group changing, it comes from the changes in each group's relative weight in the calculation.

Originally posted by zeeblebot
does the age data only include legal employment, and the unemployment survey data also include illegal employment?

Here's the somewhat longer testimony before congress about the changing age distributions impact on measuring unemployment. Gale has this paper in mind when he makes his argument in the OP link.

http://jec.senate.gov/archive/Hearings/03.07.08%20BLS/Blank%20Testimony.pdf

Originally posted by Palynka
I meant the difference between unemployment in 82 (year X) and the thought-experiment for last week's figures.

That difference must come from changes in unemployment rates in each age group.