|Trends and Challenges for Work in the 21st Century|
Flexible Staffing Arrangements
A Report on Temporary Help, On-Call, Direct-Hire Temporary, Leased, Contract Company, and Independent Contractor Employment in the United States
Susan N. Houseman
The stereotype that workers in flexible staffing arrangements earn less than comparable workers in regular positions is only partly true. Using data from the February 1997 Supplement to the CPS, Table 5 shows the percentage of workers in each work arrangement earning "low" wages, which are defined as between $4.25 and $5.15 per hour. In February 1997 the federal minimum wage had recently risen from $4.25 to $4.75 per hour and was scheduled to be increased again to $5.15 later that year. Many individuals in the February 1997 reported hourly wages of $4.25 per hour, perhaps because they were reporting their old wage, their employer had not complied with the new minimum, or they were exempted from minimum wage laws. The $4.25-$5.15 range is designed to capture the fraction of workers in each category who were working at or near the minimum wage.(12) Compared with regular full-time workers, a larger fraction of agency temporaries, on-call workers, and direct-hire temporaries earn low wages.
Low wages are strongly associated with part-time employment and the relatively large fraction of agency temporaries, on-call workers, and direct-hire temporaries earning lower wages stems partly from the fact that a large fraction of these workers work part-time hours. (See Table 1.) However, even among those working full-time hours, a much higher fraction of workers in these three flexible staffing arrangements earn low wages compared to regular full-time employees. While just 3.6 percent of regular full-time employees earn low wages, 7.3 percent of full-time agency temporaries, 6.3 percent of full-time on-call workers, and 8.6 percent of full- time direct-hire temporaries earn at or near the minimum wage.
(a) Source: Authors tabulations from the February 1997 CPS Supplement on Contingent and Alternative Work Arrangement.
(b) Source: Authors tabulations from matched data from the February 1995 and March 1995 CPS.
Although a higher fraction of temporary and on-call jobs are low-wage, this result may stem from the fact that workers in these jobs have less human capital, on average, than those in regular, full-time positions and not from the jobs themselves. Using data on earnings from the February 1995 CPS Supplement, I showed that men and women working as agency temporaries, on-call workers, and direct-hire temporaries earn significantly less than regular full-time workers, even controlling for key worker and job characteristics.(13) For instance, among men, agencies temporaries earn about 18.8 percent less, on-call workers earn 5.0 percent less, and direct-hire temporaries earn 10.3 percent less than regular full-time men. Among women, agency temporaries earn 15.8 percent less, on-call workers earn 9.9 percent less, and direct-hire temporaries earn 15.7 percent less. Regular part-time workers also earn significantly less than regular full-time workers after controlling for worker and job characteristics.
Arguably, there are worker characteristics affecting productivity that are not measured in the CPS and that are correlated with the work arrangements. For instance, actual job experience is not measured in the CPS and those with little experience are likely to be disproportionately represented in certain flexible staffing arrangements. Similarly, those with poor work habits and social skills may be disproportionately represented in certain flexible staffing arrangements and this fact may help account for the lower average wages of workers in these arrangements. Segal and Sullivan (1997b, 1998) attempted to address this issue using two longitudinal data sets on workers in the temporary help industry. In the first paper, Segal and Sullivan (1997b) constructed a longitudinal data set from the CPS; workers reporting the temporary help industry as their industry of employment were classified as agency temporaries. Controlling for individual fixed effects and part-time status, along with worker characteristics and occupation, they found that the agency temporaries earn only 3.1 percent less, on average, than other workers.(14) One problem with these estimates is that they depend on identifying changes into and out of temporary help employment. Classification of workers in the temporary help industry comes from responses to questions concerning the worker's employer in the basic CPS. Yet, the recent February Supplements to the CPS reveal that over half of all those classified as agency temporaries in the Supplements listed the client firm as their employer in the basic CPS question. Thus, the data on which Segal and Sullivan's analysis are based contain a gross amount of reporting error.
In an alternative approach, Segal and Sullivan (1998) constructed a longitudinal file using administrative data from the State of Washington to study the magnitude of the wage differential between workers in the temporary help industry and others. In this paper, they estimated the wage differential between agency temporaries and other workers to be between 10 and 20 percent.
Concern over the low wages that workers in temporary and on-call arrangements receive arises primarily to the extent that these wages contribute to poverty. Low wages in these jobs would be of less concern if the workers primarily came from middle income and wealthy families. Table 5 shows the percentage of workers by work arrangement who were living in families below the poverty line and near the poverty line (100-125 percent of the poverty line).(15) This table shows that poverty is a problem for workers in flexible staffing arrangements, particularly agency temporaries, on-call workers, and direct-hire temporaries. Whereas just 6 percent of regular full-time workers come from poor or near-poor families, 22 percent of agency temporaries, 16 percent of on-call workers, and 17 percent of direct-hire temporaries are living below or near the poverty line.
(12) I excluded those who reported hourly earnings less than $4.25. In some cases they are valid wages, but they probably reflect inaccurate reporting in many instances. The basic pattern of wages by employment arrangement is not changed when I include those earning below $4.25.
(14)The estimates cited above from Houseman (1997), which did not control for individual fixed effects, revealed that while agency temporaries earned much less than regular full-time workers, on average, the wage differential between agency temporaries and regular part-time workers was small. Given that the estimates in Segal and Sullivan (1997) control for working part-time, and that a high proportion agency temporaries work part-time, Segal and Sullivan's estimates may be broadly consistent with the estimates presented in Houseman (1997)
(15)Data on poverty status come from the March 1995 Supplement to the CPS and refer to poverty status in 1994. Observations from the February 1995 CPS were matched to the March 1995 CPS to construct figures on poverty status by work arrangement.