Why do contractors contract? The experience of highly skilled technical professionals in a contingent labor market.(Statistical Data Included)

GIDEON KUNDA, STEPHEN R. BARLEY, and JAMES EVANS (*)

This study examines 52 highly skilled technical contractors' explanations, in 1998, of why they entered the contingent labor force and how their subsequent experiences altered their viewpoint. The authors report three general implications of their examination of the little-studied high-skill side of contingent labor. First, current depictions of contingent work are inaccurate. For example, contrary to the pessimistic "employment relations" perspective, most of these interviewees found contracting better-paying than permanent employment; and contrary to optimistic "free agent" views, many reported feeling anxiety and estrangement. Second, occupational networks arose to satisfy needs (such as training and wage-setting) that employing organizations satisfy for non-contingent workers. Third, regarding their place in the labor market, high-skilled and well-paid technical contractors cannot be called--as contingent workers usually are--"secondary sector" workers; and their market is not dyadic, with individuals sel ling labor and firms buying it, but triadic, involving intermediaries such as staffing firms.

For most of the twentieth century, institutions of bureaucratic employment dominated the social organization of work. Bureaucratic employment was built on a simple bargain: as long as firms remained profitable, they would provide employees with secure jobs in return for effort and loyalty. Although there were notable exceptions on the economy's periphery (Doeringer and Piore 1971; Baron and Bielby 1984; Piore 1979; Friedland 1975), until the 1980s most middle-class Americans believed that working for a reputable company would guarantee life-long employment, as long as they performed conscientiously and the economy remained strong.

In recent years, three developments have shaken this belief. First, in the name of efficiency, firms in the economy's core have repeatedly laid off large numbers of employees independent of economic cycles. For the first time in history, layoffs have targeted significant numbers of managers and professionals (Heckscher 1994; Osterman 1996; Cappelli 1999). Second, despite inconclusive data (Diebold, Neumark, and Polsky 1997; Bureau of Labor Statistics 1997), many people believe that job tenure has become shorter and labor markets more volatile. Finally, perhaps the most radical break with the system of bureaucratic employment has been the expansion of the so-called "contingent labor force" (Barker and Christensen 1998).

The term contingent labor has been applied to a wide range of short-term employment arrangements, including part-time work, temporary employment, self-employment, contracting, outsourcing, and homebased work. Estimates of the size of the contingent labor force vary widely. The most conservative data come from the Bureau of Labor Statistics. Under the most liberal of its restricted definitions, the BLS estimates that 13.3% of Americans were contingently employed in 1995 (Polivka 1996a, 1996b; Cohany 1996; Cohany et al. 1998). (1) Estimates for 1997 were nearly identical. (2) More liberal estimates suggest that the number may be as high as 30% (Dillon 1987; Belous 1989; Kalleberg et al. 1997).

Estimates of the spread of contingent work are hampered by the absence of longitudinal data. However, data on the temporary service industry suggest two significant trends (U.S. Department of Commerce 1997). First, between 1986 and 1996 there was spectacular growth in the relative size of the temporary service industry: in those years, employment in temporary services grew 10.3% while total employment in the United States grew by only 1.7%. Second, there has been a change in the distribution of contingent jobs. Between 1991 and 1996, the percentage of the temporary service industry's payroll represented by office, clerical, and medical work declined, while the industrial, technical, and professional segments (which include managers) became more important (Staffing Industry Report 1997).

Thus, three conclusions seem reasonable on the basis of available data: a significant proportion of Americans are contingently employed; this proportion has increased over the past decade; and technicians, professionals, and managers represent a larger portion of the contingent labor force than in the past.