Secular job creation (with caveat)
Remember “secular stagnation”? It was in vogue until not long ago. It said that advanced economies have entered a permanent state of slow economic growth and weak employment for structural causes: aging populations and terminally low productivity growth, reflected in an excess of savings over investment and low interest rates.
The U.S. seems to have instead entered “secular job creation”: sustained robust employment growth supported by a structural decline in the natural rate of unemployment, reflected in moderate wage and price growth. Consider labor market performance after the latest jobs report:
To date, 79 consecutive months of jobs gains — that is six and a half years, starting in October 2010. A total of 15.7 million jobs created. The three-month moving average of non-farm payrolls (NFP) is running at about 175k, compared to 120k needed to keep the unemployment rate constant, according to the Atlanta Fed model.
The unemployment rate is down to 4.4%, a rate last seen during the 2006–07 financial bubble. To find a better rate in recent history we need to go back to the dot-com bubble in 1999–2001. Or back to the 1960s…
The broader definition of unemployment rate, U6, is down to 8.6% compared to a peak of 17.1% in 2009. (It includes discouraged workers[1] and those working part-time for economic reasons (i.e. not by choice)). It is still a bit above the 2006 low (7.9%), so there might be room to attract some more discouraged workers back into the labor force, but not much. Earlier in the recovery people would often say, “ah, the true unemployment rate is not 5% (U3), it is 10% (U6)”.[2] They would implicitly compare this to the peak of U3 during the recession — 10% in late 2009 — making it sound like there had been little or no improvement in the labor market. A sleight of hand that used to drive me crazy… The chart below to the left shows how U6 has improved broadly in line with U3; the chart on the right shows the difference between U3 and U6, now almost back to previous lows.
The participation rate is a more complicated story. It is 3 pp lower than before the crisis, at 63%. The share of people of working age who have a job or are looking for one has fallen. The decline started well before the great recession, and there is a debate on whether it is largely structural or cyclical. I have summarized the debate in an earlier blog, where I argued that there might be room to boost participation in the next few years, but in the longer-term demographic factors will dominate and keep participation down. The fact that participation today is at the same rate as November 2013, while the unemployment rate has declined by 2.5 pp, makes me more pessimistic on the possibility of even a short-term improvement.
The participation rate of prime working age (25–54) men is the most concerning issue: it’s been falling for 60 years, by about 10pp. This is the second largest decline in the OECD and leaves the US with the third-lowest rate in the OECD — though it rose 1pp from the 2014 bottom. The decline is concentrated in lower-education (high school or less) and has affected all races and ethnicities, though especially African-American men. Part of the reason, in my view, is that while the economy has shifted job opportunities to higher skills, education and training have not kept pace.
The good news: the US economy is a job-creating machine; the recovery has proved so resilient that we seem to have entered a “secular job creation”.
The bad news: weak productivity growth and a prolonged decline in the participation rate of prime working-age men suggest serious structural problems, including a skills mis-match that will be exacerbated by technological advances.
Cyclically, we’re doing great; structurally, we’ve got a lot of work to do.
What do you think? I would love to discuss this with you in the responses below or on Twitter.
Footnotes:
[1] “Marginally attached to the labor force”, defined as those who have not looked for a job in the last 4 weeks.
[2] As in September 2015.
