September 02, 2003

Labor Market Forecast

A labor market forecast from -- Reports from The labor market remains in transition as the last six months of payroll declines are expected to turn into small gains over the coming months.  The dismal news from manufacturing continues with a 36th consecutive month of decline summing to 2.8 mln fewer workers.  Private service sector payrolls have been bouncing on either side of flat growth.  The monthly movement is volatile due to corporate cost cutting, strong labor productivity and an economy that's not generating a strong lift in aggregate demand.  The lagging unemployment rate reached a new 6.4% high in June and will continue to follow a path higher even as payrolls return to growth.  Hourly earnings are running at a 3% pace as compensation costs are of lesser concern given weak unit labor costs (compensation offset by productivity gains).  The workweek is a key indicator of labor demand and a leading indicator of payroll growth but shows no evidence of an impending upturn.  The labor market still shows no signs of tightening up but it lags the overall economy.

Posted by DeLong at September 2, 2003 09:21 PM | TrackBack


I would think one big issue for the labor market is whether the decline in hours, which mentions in passing, is due largely to secular or cyclical factors. It sure looks like the workweek of those already employed can be expanded to meet demand for labor for some time before hiring really picks up. The average is down about an hour from the mid-1990s.

That said, hours have been falling for decades. The 1990s saw a steadying in work hours, but the 1990s also saw an unexpected drop in the jobless rate. I don't think we know yet how much of that was the result of a special set of labor market circumstances in the 1990s. It is entirely possible the average workweek won't get back up to the 34.5 hours or so of the mid-190s. If you eye-ball a line through the data from 1964 to 1990, it suggests hours are actually quite quite high. I don't believe they are high, against what workers really want, but I also don't believe the workweek is as short as official data suggest.

Posted by: K Harris on September 3, 2003 04:23 AM

K Harris

We would argue that white collar employees work far longer hours than officially recorded, though this only by anecdotal evidence. Care to trade French or German or Dutch vacation time for American vacation time. Ah, yes.

Posted by: lise on September 3, 2003 10:54 AM
Post a comment