« knzn Fiscal Policy | Main | Paul Krugman: Health Care Terror »

Monday, July 09, 2007

Tim Duy: On The Employment to Population Ratio

Tim Duy with thoughts on the employment to population ratio:

On The Employment to Population Ratio, by Tim Duy: Dean Baker watches the employment to population ratio for signals of job market weakness, but I would warn against reading too much into the short run variation in these statistics. For instance, Baker notes:

This drop is being driven primarily by workers ages 35-44. Their EPOP has fallen from 81.5 percent in January to 80.6 percent in June. This is 2.1 pp below the peak of 82.7 percent in Jan-Feb of 2000 and 1.2 pp above the trough of 79.4 percent in July of 2003.

It is very difficult to think of any reason why hundreds of thousands of prime age workers (both men and women, the declines are roughly equal) would suddenly drop out of the labor market, other than limited job opportunities.

Time to go to charts on this one. For men 35-44 (note, all charts are for the 35-44 demographic Baker refers to):

Employment to Population ratio: Men aged 35-44
1997:01 to 2007:06

Duy17807

Note the sharp drops in the employment to population ratios in 1999 and 2000, both of which occurred during a period of strong job growth (weakness in the labor market was not manifest until the second half of 2000). So clearly there may be reasons for variation other than limited job opportunities. For women in the same age group:

Employment to Population ratio: Women aged 35-44
1997:01 to 2007:06

Duy27807

Again, note the variation in the 1997-2000 period. One can read the recent retreat as a signal of impending doom, but then you need to explain why the job market was so strong in late 2006 (when the economy was slowing) that it dragged thousands of women into the workforce.  Alternatively, we are seeing pretty typical shifts of an indicator that exhibits a significant amount of variation.

Also, Baker claims that this is not a demographic story.  While for short term variation I agree, I think it is very misleading to examine this data without regard to the long term demographic story (and I think this applies to efforts to measure these numbers from the 2000 peak). For men:

Employment to Population ratio: Men aged 35-44
1948:01 to 2007:06

Duy37807

Since the 1970s, the employment to population ratio of these prime age male workers has ratcheted downward with each recession. For decades this has been offset by increasing female participation:

Female labor Force Participation Rate
1948:01 to 2007:06

Duy47807

For whatever reason, female employment (participation really) reached a plateau beginning in 1990 or so. My view is that we have pretty much tapped out labor force participation among females, and if the downtrend for males holds, employment to population trends in the long term have nowhere to go but down.  Consequently, it is misleading to assume that declines in these measures (similarly, labor force participation) automatically indicate cyclical weakness. [Comments welcome, or email Tim at timduy@msn.com]

Update: PGL has comments on the same topic.

    Posted by on Monday, July 9, 2007 at 12:06 AM in Economics, Monetary Policy, Unemployment | Permalink  TrackBack (1)  Comments (16)

    TrackBack

    TrackBack URL for this entry:
    http://www.typepad.com/services/trackback/6a00d83451b33869e200e008d4cafa8834

    Listed below are links to weblogs that reference Tim Duy: On The Employment to Population Ratio:

    » Employment-Population, by Age from Wcw

    Employment-to-population ratio by age groups, 1987-2007 What's up with us 35-to-44 year olds? Update: It's Tim Duy to the rescue. For whatever reason, female employment (participation really) reached a plateau beginning in 1990 or so. My vie [Read More]

    Tracked on Sunday, July 08, 2007 at 09:32 PM


    Comments

    Feed You can follow this conversation by subscribing to the comment feed for this post.

    -->