Brad de Long is Still Bemused
by the fact that there were 3.3 percent fewer hours worked in the nonfarm business sector in 2003:I (the first quarter of 2003) than in 2000:I. Such a prolonged shrinkage in hours worked–given America’s robust demographics–is an extraordinary labor market experience for the United States.
I’ve been thinking about this too, and, although I don’t have a fully worked-out story my view is that 2000 was exceptional, while 2003 is, in most respects, a return to normality. Looking at the decline in hours worked, between one-third and one-half is due to a decline in average hours per worker. This is a reversal of the historically unprecedented increase in average hours per worker that took place during the 1980s and 1990s.
At the same time, demand for labour was exceptionally strong. Employers were scouring remote rural areas, clamoring for access to prison labour and (Mickey Kaus’ favorite story) eager to take on former welfare recipients. The unemployment rate, at 4 per cent, was the lowest in recent times, but if anything that understates the tightness of the labour market. As things have slackened, there has been a big drop in participation
A final feature of the late 1990s was an incredible investment boom (the other side of the current account deficit). The result was that whereas labour productivity grew strongly in the late 1990s, multifactor productivity was relatively weak, a point that has attracted insufficient attention in assessments of the boom.
Not surprisingly, demand for labour was exceptionally strong. Everything fell in a heap when the bubble burst in 2001. As the investments in dotcoms and telecoms proved worthless, multifactor productivity actually fell. I would guess that MFP recovered a bit in 2002, but not much. Here’s some data I got from the Bureau of Labor Statistics MFP Home Page
Series Id: MPU750023 (K)
Measure: Multifactor Productivity (Index, 1996 = 100)
Sector: Private Nonfarm Business