Staircases or Treadmills?: Labor Market Intermediaries and Economic Opportunity in a Changing Economy

Chapter Five

Chapter Five The Impact of Intermediaries on Job Outcomes In the previous chapter, we reviewed data from the Survey of Labor Market Intermediary Use to examine the nature of LMI use by workers in two regions, and we considered the differential experience of workers who are disadvantaged by education, income, or race. Here we give further consideration to the impact of intermediaries. We look at job outcomes—occupation, industry, wages, hours, and benefits—with an eye to whether workers obtained their job through an intermediary agency, the kind of agency used, and their level of disadvantage. Job outcomes, of course, are jointly determined by complex factors on both the supply and demand sides of the market. Raw outcomes may reflect not only the efficacy of the agency making the job placement but also differences in the mix of workers who do or do not use each type of agency. Thus, although a simple comparison of wages and other outcomes across intermediary types is an important descriptive exercise, these comparisons cannot be taken as indicative of causality. In the analysis that follows, we first evaluate raw outcomes across LMI users and non-users, disadvantaged and otherwise. We then turn to a multivariate analysis of outcomes, measuring the impact of intermediary use in the presence of other controls for worker and job characteristics. Even though this latter analysis still cannot definitively identify causality, it is more suggestive of potential causal factors. 126 STAIRCASES OR TREADMILLS? In particular, we find a strong negative relationship between temporary agency use and wage and benefits outcomes. We compare these findings with those of other authors, reconciling differences in datasets and reproducing a variety of specifications. We pay particular attention to a study by Fredrik Andersson and his colleagues (2005), who found a positive association between previous temp use and earnings for low-income workers; we rework our data to mimic their results and illustrate how the positive impact is likely to be due to longer hours rather than higher wages, with the net welfare benefits up for evaluation by observers and policymakers. In general, we find that the negative relationship between temporary agency use and hourly wages is robust and holds up in any comparison of jobs obtained through temp agencies and jobs obtained without LMI intervention. Occupations and Industries The nature of a job can in part be inferred from its occupational and industrial location. In figures 5.1 to 5.4, we show the distribution of jobs according to industry and occupation and according to whether the job was obtained through a temp agency (narrowly defined), some other type of LMI, or no LMI. It is clear (if not surprising) that both the occupational and industrial distribution of jobs vary according to how the jobs were obtained. In both regions, temp users were largely found in administrative support occupations, machine operator and assembler occupations, and transportation and material moving occupations. They were also more likely to be located in the manufacturing sector than were other workers, perhaps belying a common notion that temp workers are mostly found in office occupations.1 Differences between those who used other LMIs and those who used no LMI were much less stark. A separate question is whether job placements made by LMIs are more highly concentrated in a narrower slice of the economy. We measure this by calculating the cumulative percentage employed in occupation or industry categories , sorted by concentration of employment. We conduct this analysis separately by region and for temp agency users, other LMI users, and non-LMI users. We also repeat the analysis separately for those with low education levels only. We find that across all education levels, LMI users are not more or less concentrated than non-users in their industry or occupation locations. However, among those with low levels of education (high school diploma or less), the industry/occupation locations of those who obtained jobs through LMIs were (text continues on page 131) THE IMPACT OF INTERMEDIARIES ON JOB OUTCOMES 127 Figure 5.1 Distribution of Employment by Occupation and LMI Status, Milwaukee 0 5 10 15 Percentage 20 25 30 Executive, Administrative, and Managerial Occupations Professional Specialty Occupations Technicians and Related Support Occupations Sales Occupations Administrative Support Occupations, Including Clerical Private Household Occupations Protective Service Occupations Service Occupations, Except Protective and Household Farming, Forestry, and Fishing Occupations Precision Production, Craft, and...