James Hamilton on the Decline in the Employment-Population Ratio
My updates of the labor force participation rate (LF) and the employment to population rate (EP) will have the starting date being January 1997 in part due to my desire to say thank you to James Hamilton.
James nicely summarizes my view nicely:
The primary concern seems to be that, although the unemployment rate has come down nicely from its peak during the last recession, the fraction of the population that is employed has failed to return to its values of 2000.
The news on Friday was that EP did not rise and LF fell – with latter being why the unemployment rate fell from 5.1% to 5.0%. James questions whether a fall in EP is necessarily a bad thing in a spirit similar to what Andrew Samwick wrote. As one follows the logic of Andrew and James, note we can discuss developments in the labor market entirely in reference to EP and LF without addressing their difference (the unemployment rate). While economists like Paul Krugman and myself see part of the reason for the decline in LF (and an implied fall in EP for any given unemployment rate) coming from a discouraged worker effect – Andrew noted that at least some of the decline in LF may have been due to people wishing to leave the labor force.
James has taken Andrew’s suggestion to its logical conclusion – if less people wish to work in 2005 than they did in 2000, we should expect at least some decline in EP. While this is a fair point, I have two problems with this apparent “don’t worry, be happy” view. One comes from reading the paper from the BLS on teenage LF, which James kindly links to. Its very first sentence notes that teenage LF is highly cyclical – something several AB readers have noted. It is true that there has been a long-term downward trend in teenage LF, which the BLS explains in part by teenagers decided to get more education and training after high school. During the dotcom boom of the late 1990’s, having strong technical skills was especially important. So some teenagers delayed entering the strong job market in order to obtain these skills, which is why LF did not rise. After the dotcom boom ended, LF fell dramatically. Is the dramatic decline due to even more benefits from training as James suggest or is it due to a discouraged worker effect as Paul Krugman suggests?
The second problem is my excuse for providing a second graph – the EP ratio for 25-year old and over. It is not surprising that this ratio is higher for 25-year old and over than the overall EP ratio. Notice that this ratio fell after 2000 and has only partially recovered. Also note that this ratio fell from 64.4% to 64.3% in June 2005.
James also kindly links to an analysis from Jared Bernstein who notes some good news about long-term unemployment, but also writes:
Still, it is unusual for long-term unemployment to be this high with unemployment so low. Historically, 5.0% unemployment rates have been accompanied by an average long-term jobless rate of 10.7%, well below June’s level of 17.8% … the rate of job growth is far below that of past recoveries.
Some of our pro-Bush AB readers have suggested I put a negative spin on the employment news. James in his kindly and gentler fashion claims I was angry and bearish. While one could argue (as Andrew and James do) that my previous hopes for a 64% EP may set the bar too high, setting the bar for full employment at EP = 62.7% strikes me as economic malaise. I find it odd that the allegedly pro-growth supply-side crowd would rather excuse lackluster economic performance during a Republican Administration than set the bar a bit higher. But then I’m a pro-growth liberal – who is not bearish on where the economy could be.