Temporary help service firms' use of employer tax credits: implications for disadvantaged workers' labor market outcomes.

The WOTC and WtW subsidy programs are the most recent in a long line of wage subsidy programs designed to improve work outcomes among the disadvantaged. They are most similar to the Targeted Jobs Tax Credit (TJTC), which operated from 1978 until 1994. The TJTC allowed employers to claim subsidies for workers in several target groups, including certain welfare recipients, food stamp recipients, disabled individuals, and disadvantaged youth. The WOTC program, like the TJTC, is a one-year subsidy involving several target groups; we focus on welfare recipients and food stamp youth (ages 18 24). (8) Under the WOTC, the tax credit is 40% of wages, up to a maximum credit of $2400, as long as the employee works for at least 400 total hours at the firm. The credit is 25% of wages if the employee works at least 120 but fewer than 400 hours and is zero otherwise. The WtW program applies specifically to workers with at least 18 months of welfare receipt who work 400 hours per year or more. It is administered exactly like the WOTC but involves a subsidy of 35% in the first year of employment (maximum credit $3500) and 50% in the second year (maximum credit $5000).

Employers can participate in these programs by applying for subsidy certification for any new worker they hire. Participating employers typically assess eligibility by including a short WOTC prescreening form in workers' application or new-hire materials. This prescreening form, along with a one-page subsidy application, must be completed and signed on or before the date of hire. The employer must then submit them to the state (along with fairly minimal documentation) within 21 days of hire. On approval from the state, the employer is authorized to claim the WOTC or WtW tax credits on its federal tax return. (9) Note that in the case of THS employment, the legal employer is the THS firm; it is subject to the same time limits and subsidy rates as other firms regardless of whether the worker has multiple job assignments during an employment spell with the firm.

The limited research on the WOTC and WtW suggests a low level of participation among firms that is not well understood. Hamersma (2003) estimated that less than one-third of eligible welfare recipients and fewer than 18% of food stamp youth were claimed by their employers for the credits. The U.S. General Accounting Office (2002) reported that the employers most likely to claim the subsidies are large firms that qualify for over $100,000 per year in WOTC/WtW tax credits.

In a study of the Wisconsin welfare population, Hamersma (2008) used matching estimation to assess the effects of the WOTC/WtW on employment, wages, and job tenure. The results indicate that employment probabilities do not measurably improve when workers become eligible for the programs. (10) However, within the WOTC/WtW-eligible population, those whose employers participate have about 10% higher earnings per quarter than similar unsubsidized workers (though average job tenure is unaffected). These small to nonexistent effects are consistent with the previous literature on targeted employer tax credits. (11)

Research investigating the influence of the THS industry on the employment patterns of the disadvantaged is also relatively new, reflecting the recent growth of employment in the industry. A primary concern for this study is the overrepresentation of public assistance recipients among THS employees. In recent work, Heinrich, Mueser, and Troske (2005) studied the populations of welfare recipients in Missouri and North Carolina and found that the proportion with jobs in the temporary help services sector more than doubled between 1993 and 1997. (12) Among Autor and Houseman's (2005) sample of welfare-to-work clients in Michigan, 21% who found jobs were working for temporary help agencies. In fact, THS firms are registered providers of welfare-to-work services in a number of states. (13)

There is considerable evidence that the increasing participation of the disadvantaged in THS employment may be a cause for concern. Temporary workers are paid lower wages than permanent employees in the same positions, and they are more likely to be underemployed and to work fewer hours (Nollen 1996; Segal and Sullivan 1997; Blank 1998; Houseman and Polivka 2000). THS jobs seem to lack some of the attributes that other research suggests are critical to welfare recipients' successful transition from welfare to stable employment--secure income, training opportunities, health insurance benefits, and paid time off (Blank 1998; Cohany 1998; Jorgenson and Riemer 2000; Morris and Vekker 2001). Some have also criticized THS firm arrangements specifically because they sever the traditional employer-employee relationship between workers and client firms on whose premises they work. Studies by Miles (2000) and Autor (2003), for example, link the growth of temporary jobs to THS firms' exemption from legal protections against unjust dismissal of workers. (14) THS firms can end job placements without firing workers, and they have no obligation to provide a job assignment at any time.

Of course, as we also noted earlier, there is an alternative understanding of THS employment that suggests that disadvantaged workers are more likely to take these jobs because they best match their preferences or skills. Segal and Sullivan (1997) and Carre (1992) show that a large portion of the temporary/permanent wage gap can be explained by worker characteristics and by the higher concentration of low-skilled workers in occupations mediated by THS firms (i.e., relatively low-wage administrative, clerical, and laborer jobs). Heinrich, Mueser, and Troske (2005) suggest that for these workers with less desirable characteristics, the ability to enter into a contract where the employer has no long-term obligation may facilitate their access to the labor market and possibly lead to permanent jobs if workers demonstrate high productivity. If this is, in fact, a beneficial service that THS firms provide, any additional encouragement that WOTC/WtW subsidies might give to THS firms could produce net benefits for participants and society in the form of reduced public assistance obligations.

Economic theory provides only limited help in modeling the interaction of THS employment with the WOTC subsidies and predicting the anticipated effects of each of them. For example, a typical model of subsidies predicts increased earnings (due to subsidy pass-through) and perhaps longer job tenure (in cases where the subsidy increases with time on the job). (15) Both effects are likely to improve labor force attachment in general. However, in the case of THS workers, these predictions are less clear. There is a wedge between the worker and his or her employer (the THS firm) both in terms of pay since a portion of the wages paid by the client firm goes to the THS firm rather than the worker--and in terms of productivity, which may be hard for the THS firm to directly observe. This muddies the usual prediction of increased earnings for the worker. The tenure incentives are also more unwieldy in the context of THS work. There is substantial uncertainty as to whether workers will reach the minimum hours requirement to qualify for a subsidy since workers may decline assignments and client firms may cancel contracts.

The typical expected effects of a THS job are likewise potentially different in the context of subsidized employment. While we would still expect the usual shorter tenure (due to the nature of the work), firms with very short average tenure may be less likely to participate in subsidy programs; this means the effect may be smaller in our sample of WOTC-participating firms. Predicting an earnings effect is difficult (even in the absence of a subsidy) because the compensating differential for a THS job may be positive or negative, depending on whether the worker suffers or benefits from the temporary arrangement. In effect, this becomes an empirical question that we address for the population of subsidized workers.

Moreover, this population of subsidized workers provides a convenient, more homogeneous sample than a typical study comparing THS and non-THS workers. The firms that participate in the WOTC tend to be large, national firms that provide support for processing the WOTC paperwork, and the workers qualified for the WOTC must have particular public assistance records, so both firms and workers are relatively homogeneous. This lessens econometric problems associated with controlling for differences among workers selecting into THS employment.

3. Data

Research on the effects of federal employer tax credit programs, including the WOTC and WtW tax credits, has been limited by data access problems. The state labor agencies that are responsible for certifying workers' eligibility do not keep records of hours or earnings for these workers since the responsibility for auditing the amount of subsidy claimed belongs to the Internal Revenue Service (IRS). (16) Thus, there is no single source of data that includes both certification information and labor market outcomes.