The growing use of temp workers by manufacturing -- about 890,000 now of a total above 19 million -- partially explains the flatness in the level of manufacturing employment in the 1990s (as recorded by the Bureau of Labor Statistics) despite the substantial increases in output of American manufacturers.
For years, official government statistics have indicated high productivity gains for manufacturing firms. But new research indicates that such gains, though still far above the average for nonfarm industries, may not be quite so spectacular. In a recent NBER Working Paper, Marcello Estevao and Saul Lach find that manufacturers in the 1990s have been hiring more and more temporary workers. These workers remain on the payroll of the temp agency, and thus are not counted when manufacturers supply numbers to the government on the size of their work force and hours worked. The result has been a modest exaggeration -- about 0.5 percentage points a year, or somewhat less, from 1991 to 1997 -- in the annual improvements in manufacturing productivity. Instead of annual productivity gains of 3.8 percent in that period, the adjusted figure amounts to 3.3 percent. That compares with 1.3 percent annual productivity gains for all nonfarm industries, a group that includes the manufacturing sector.
In Measuring Temporary Labor Outsourcing In U.S. Manufacturing (NBER Working Paper No. 7421), the authors write that the growing use of temp workers by manufacturing -- about 890,000 now of a total above 19 million -- partially explains the flatness in the level of manufacturing employment in the 1990s (as recorded by the Bureau of Labor Statistics) despite the substantial increases in output of American manufacturers. Manufacturers added only 550,000 jobs between 1992 and 1997, even though the economy was thriving. In the last 10 years or so, employment in the U.S. "temporary help supply" industry has more than tripled. Looking at a longer period, the number of temps has risen at an annual rate of 11 percent a year. Total nonfarm employment -- a far larger number than manufacturing employment alone -- expanded at only 2 percent per year.
As of 1997, however, temps accounted for only about 2 percent of total nonfarm employment, Estevao and Lach note in The Evolution of the Demand for Temporary Help Supply Employment in the United States (NBER Working Paper No. 7427). Temporary help accounted for 10 percent of the net increase in nonfarm employment between 1991 and 1997, though. In this paper, the authors estimate where these temp workers are employed, making use of input-output data (cross-industry sales and purchases information) and numbers from surveys of contingent workers in 1995 and 1997. They find that the proportion of temp employees in each of eight major American industries, except the public sector, increased from 1977 to 1997. The proportion of temp workers employed in the public sector (including federal, state, and local administration, and public enterprises) declined from a 40 percent peak in 1982 to almost zero in 1997. By 1997, close to 4 percent of the employees in manufacturing and services were temps; about 75 percent of all temporary employees worked in manufacturing or service sector firms. That compares to 40 percent in 1982.
The surge in demand for temporary blue-collar workers by manufacturers os consistent with reductions in the proportion of temporary white-collar and pink-collar (clerical) workers. Further, Estevao and Lach find that more temps now are men. The proportion of males among those working for temp agencies in 1992-7 (38 percent) was more than 10 percentage points higher than in 1977-87. The share of temps working part-time (less than 35 hours of work per week) declined in that same time span, although it remained well above the average for the whole labor force.
On average, the authors find, temps are a bit better educated: the proportion of temporary workers with at least some college increased from 52.5 percent between 1977 and 1987 to about 55 percent between 1992 and 1997. The authors also find that the rise in the proportion of temp workers in the economy is attributable to a conscious decision by employers to hire more temps, rather than to a shift in employment from temp-light to temp-heavy industries. The hiring of more temps, they write, may enable firms to implement a two-tier wage structure; they may contract with temp agencies that pay less for workers doing similar work to that of regular workers, thereby lowering overall wage costs. Employers also may get the advantage of temps specializing in specific tasks, or obtain temps who have been better screened or trained than if the firms were to hire these temporary workers directly themselves. The use of temps may facilitate rapid changes, up or down, in the number of employees the firm needs because of shifts in demand. Indeed, Estevao and Lach find a much greater variation in temporary employment in manufacturing -- perhaps six to 10 times greater -- than in regular, on-payroll workers.
-- David R. Francis