THE INCREASE IN CONTINGENT WORK ARRANGEMENTS--particularly the secular growth of personnel supply services (PSS)--is one of the most striking features of employment relations in the United States during the past quarter-century. The personnel supply industry has experienced large, secular growth in both absolute numbers and as a percentage of the labor force from 1972 to 2000. During this period, employment in the U.S. personnel supply industry increased from about 200,000 workers to more than 3.5 million, as compared to total nonfarm employment in the United States, which increased from 71 million to 127 million workers. As a percentage of the U.S. labor force, personnel supply employment grew from 0.27 percent in 1972 to 2.81 percent by the beginning of 2000.
Despite the dramatic growth of personnel supply employment, there is much that we don't know about why employers have adopted this type of flexible staffing arrangement. Most research on temporary work has focused on individual workers' motivations for undertaking this kind of work (e.g., Rogers 2000), cross-sectional analyses of correlates of employment in personnel supply firms (often based on data from the Current Population Survey, e.g., Kalleberg et al. 1997), a single organization's use of temporary workers (e.g., Smith 1998), how the legal context of employment and the diminution of the "employment-at-will" doctrine have affected the growth of the personnel supply industry (Autor 2003), or surveys of personnel practices within firms (Abraham 1988). The relatively few time-series analyses of the personnel supply services industry have focused on the period of the 1980s and mainly on reasons for the secular growth of temporary employment, not on volatility in this industry (e.g., Golden and Appelbaum 1992; Laird and Williams 1996).
In this article, we examine the reasons for the volatility in the U.S. personnel supply industry over the past 29 years (1972:i-2000:iv) rather than employment growth in this sector. This time period allows us to examine the effects of the last three business cycles on the demand for temporary workers. In particular, by examining volatility in employment of personnel supply workers, we are able to provide suggestive evidence as to whether employers use temps to achieve numerical flexibility and buffer their regular full-time workers from downturns in demand, rather than to obtain wage flexibility.
This article is organized as follows. In Section II we discuss the factors that have led to the growth in temporary employment and make a distinction between growth and volatility. In Section III we review theories about the effects of dual labor markets on temporary employment volatility. Section IV presents our data and empirical model, and in Section V we report our empirical results. In Section VI, we summarize our conclusions.
Growth and Volatility in the Personnel Supply Industry
THE PERSONNEL SUPPLY INDUSTRY IN THE UNITED STATES did not really develop into its modern form until after World War II, though its roots in the United States can be traced to agencies supplying calculating-machine operators on a temporary basis in Chicago during the late 1920s. Personnel supply agencies existed even earlier in Britain (Moore 1965). By 1956 there were still only about 20,000 employees in the U.S. personnel supply industry (Gannon 1984). Starting in the early 1970s, the PSS industry expanded very rapidly and grew faster than regular full-time employment, as is shown in Figure 1.
[FIGURE 1 OMITTED]
Increasing Personnel Supply Employment
The explosive growth of personnel supply services employment attracted the attention of social scientists, several of whom attempted to explain this growth using time-series data. A key question addressed by these researchers was whether this growth in temporary employment had been generated by the needs and decisions of employers or by the preferences of employees. …