The Temporary Assistance for Needy Families (TANF) program was designed to replace indefinite cash assistance that substituted for employment with temporary assistance that helps recipients obtain employment. Employment outcomes for TANF recipients are the result of an interaction between labor supply issues — the skills and characteristics of the individuals seeking employment — and labor demand issues — the needs and practices of employers in the entry-level, low-skill labor market in which many TANF recipients seek employment.
The research and policy community has given substantial attention to labor supply issues, including extensive study of how individual liabilities can serve as barriers to employment, and the means by which TANF recipients overcome such challenges and obtain employment. Far less attention has been given to labor demand issues. A number of demand-side studies have examined particular companies, job sectors, industries or locales, but we lack a comprehensive picture of the range of employment policies, practices, and opportunities encountered by TANF recipients as they enter the labor market.
A recent ASPE report, Private Employers and TANF Recipients,” completed in collaboration with Abt Associates, Inc. and the Upjohn Institute for Employment Research, provides a comprehensive review of relevant literature and data sources, and discusses a range of opportunities for advancing knowledge in this area.
The study began with the following questions:
- What types of employers are most likely to hire TANF recipients?
- Why do employers hire TANF recipients?
- Why do employers not hire TANF recipients?
- How do employers employ TANF recipients successfully?
- Which segments of the TANF population present the greatest challenges?
- What would enable employers to increase and improve their efforts?
ASPE contracted with Abt Associates and the Upjohn Institute to study the degree to which these questions have been answered in the research literature, to assess the value of existing datasets for answering the remaining unanswered questions, and to identify the options for future study that would be most valuable to researchers, policymakers, program administrators, workforce intermediaries, and employers.
The study team reviewed three fields of research:
- Employer-focused research on the entry-level, low-skill labor market in which most TANF recipients gain employment, although little of this literature focuses on recipients;
- Research on the experiences of TANF recipients and other low-skill workers in the labor market that includes attention to the practices of employers; and
- Research on workforce intermediaries, the public and private organizations that connect TANF recipients and other low-skill workers to employers.
The first question — what types of employers are most likely to hire TANF recipients — can be answered reasonably well. These employers are heavily concentrated in the service sector, are large companies, in urban areas, and offer irregular schedules or alternative arrangements (e.g. on-call, temporary employment) and relatively low wages.
The second question — why do employers hire TANF recipients — is also reasonably well answered. Although some companies have made extensive efforts to employ TANF recipients, based on a sense of social responsibility or in response to tax credits and other incentives, survey and focus group evidence shows that most employers are indifferent to whether or not a job applicant is a TANF recipient or recent recipient. Rather, employers hire TANF recipients for the same reason they hire other employees — for business reasons, if the applicant is able to do the job the employer needs to fill. For this reason, effective efforts to place recipients have made a business case to employers as to why the recipients should be hired. However, given the relative disadvantage of most TANF recipients, in comparison to the jobseekers with whom they compete, tight labor markets are more conducive to their being hired.
The third question — why do employers not hire recipients — is partially answered. There is evidence that employers are skeptical that TANF recipients possess adequate “soft skills,” such as work ethic and social skills. But there is also evidence that employer concerns are often allayed after they have hired recipients and found their performance to be comparable to their peers. The limited “hard skills” of many recipients, such as low education levels and reading abilities, are the same skill deficits that can prevent workers in general from finding employment. Employers also report that TANF recipients often experience other challenges, such as unreliable child care or transportation, that may interfere with their job performance.
The fourth question — how do employers successfully employ recipients — cannot be confidently answered. Answering this question would ultimately require a large-scale evaluation of the sort that has not been done. Moreover, “success” has been defined inconsistently in the smaller studies that have sought to answer this question by identifying “best practices. Much more is known about recruitment, screening and initial training than about later management and human resources practices. There is little comprehensive information on work supports, such as child care or employee assistance programs, offered by employers of TANF recipients, though there are case studies of employers that have made extensive efforts in this area. And there is little comprehensive information on the role of workforce intermediaries in these areas.
The fifth question — which segments of the TANF population present the greatest challenges — can be partially answered, on the basis of employer testimony but also on the basis of studies of the TANF caseload. Several surveys, including recent ASPE surveys covering six states, have shown that low education levels, limited work experience, physical health problems, and child care responsibilities are correlated with limited employment among TANF recipients. Additional challenges, such as substance abuse and domestic violence, while experienced by only a small minority of TANF recipients, may also present challenges to employers. Employers report difficulty providing the full range of work supports necessary for employees who experience these challenges.
The final question — what would enable employers to increase and improve their efforts — also cannot be confidently answered, given the near absence of systematic evaluation of various efforts. Evidence indicates that most employers are unaware of the range of government incentives and services that might encourage them to hire TANF recipients, suggesting that better publicizing these might increase interest on the part of employers. However, studies have also shown that employer incentives, such as the Work Opportunity Tax Credit, may discourage employers from hiring recipients, as the incentives may imply that the potential employees are deficient and could prove costly to the employer, if the employee requires extensive remedial training or is not able to perform. There is also evidence that training is most effective when it is targeted to an employer's specific needs, leading to recent pilot projects in which employers are actively involved in the selection, screening, and training of potential employees.
The literature search confirmed that research on TANF has given far less attention to the demand side of the labor market than to the supply side. The TANF research that does focus on demand side issues has been mainly descriptive rather than evaluative, qualitative rather than quantitative, local rather than nationally representative, and has tended to neglect the important role of workforce intermediaries. Little research examines outcomes — either employee outcomes, such as job retention and advancement, as a function of employer practices, or employer outcomes, such as utilization of publicly provided work supports, as a function of the efforts of TANF administrators or workforce intermediaries.
The most informative extant data source in this area is economist Harry Holzer's survey of a random sample of 3,000 employers in Cleveland, Milwaukee, Chicago and Los Angeles in 1998-1999. However, this dataset has a number of limitations — it covers only four cities, three of which were in the Midwest; it was fielded at the peak of the economic expansion, when labor markets were unusually tight; it focused on recruitment and hiring practices, much more then post-hiring management practices; and it gave limited attention to workforce intermediaries.
Another valuable data source, which has not been fully utilized, is a survey conducted by Abt Associates for the Welfare-to-Work Partnership, which surveyed over 200 workforce intermediaries in Atlanta and New Orleans. This survey contains extensive information on the activities of intermediaries, and the types of jobseekers and employers they served. However, the survey was administered during the economic conditions of the late 1990s, in only two cities, and contains limited information on the employers with which the intermediaries worked.
Administrative data collected by Workforce Investment Act (WIA) one-stops — Workforce Investment Act Standardized Records Data (WIASRD) — help to answer some of the above questions, but only partially, in part because they contain no information on employer-provided training, and no information on non-WIA intermediaries. And the Unemployment Insurance wage records relied on in WIASRD contain no information on the subtleties of compensation practices, or management practices in general.
The Longitudinal Employer-Household Dynamics (LEHD) dataset promises to be very useful in this area. Sponsored by the Census Bureau and ASPE, LEHD consists of linked data from individual surveys such as the Current Population Survey, the Survey of Income and Program Participation, and the decennial census; employer surveys providing information such as size, industry, job classifications, and location; and Unemployment Insurance wage records linking the two. While this dataset is not national, the 10 participating states amount to roughly half of the U.S. population. However, this dataset includes limited information on specific employer hiring, training, and retention practices.
All of these sources could further answer the questions set out above, but none provide the comprehensive descriptive evidence or rigorous evaluation evidence needed to answer the questions more completely. Rather, extant data provides descriptive evidence that either is comprehensive, in the sense of nationwide, but with too little detail, or is detailed but far from comprehensive, and also becoming dated. There is no rigorous evaluation evidence on outcomes and effectiveness that covers employers and workforce intermediaries in sufficient detail, in terms of examining both employer and employee outcomes.
The literature review revealed that several critical questions remain unanswered. These include:
- What is the range and prevalence of employer and workforce intermediary practices?
- What practices are most helpful in enabling individuals leaving TANF to succeed in employment?
- What leads employers and workforce intermediaries to adopt these practices?
- Are these practices as effective as labor-supply interventions, such as skills training?
Questions on effectiveness are difficult to answer in the absence of a random assignment evaluation. Significant challenges that would arise in attempting to conduct such studies include the difficulty of inserting random assignment into hiring practices, sample size problems, and the fact that many intermediaries are in a state of evolution, such that it would be premature to conduct controlled experiments.
Moreover, without knowing the range and prevalence of employer and workforce intermediary practices, it would be difficult to focus the evaluation and to formulate hypotheses. Consequently, the report suggests that the next key step is to obtain a detailed, comprehensive picture of the range of employer and workforce intermediary policies, practices and attitudes. This would require a national survey of employers and the intermediaries with which they partner, examining business characteristics, workforce characteristics, business conditions, interest in and perceptions of TANF recipients, knowledge of government incentives and supports, and practices in terms of recruitment and hiring, compensation, employee assistance, training, and performance assessment. It is especially important to capture the interactions of employers and workforce intermediaries.
Such a survey could range in terms of scope and complexity, and the report discusses a range of possibilities in detail. For example, a core survey project might consist of a mail survey of private employers, but could be broadened by including public and nonprofit employers, and/or surveying the intermediary organizations that partner with the employers surveyed, and/or conducting a follow-up survey or site visits for a subsample of employers, in order to examine priority areas, such as child care, in greater depth. The sample could be stratified geographically, by industry, or by type of company, in order to capture the companies most likely to hire TANF recipients
Such a survey would encounter a number of difficulties, especially those that are characteristic of attempts to survey private companies. One challenge is in sample design, in terms of selecting the sample universe and, even more, the appropriate sampling unit (e.g., enterprises comprised of more than one company, a single company, individual establishments within a company). Getting up-to-date contact information for companies is difficult, as is identifying and gaining access to the right person to interview (e.g. the person responsible for entry-level hiring and management), gaining complete and accurate information, given company policy regarding release of information (especially for written surveys), and the difficulty of achieving high response rates and of creating appropriate incentives for participation.
Nevertheless, such a survey could be effectively designed and conducted, and would provide extremely valuable information for researchers, policymakers, program administrators, caseworkers, workforce intermediaries and employers.
There is much work to be done in this area, and this report clarifies the issues at stake and means by which to improve our understanding. Hopefully this will be useful for the research community in beginning such work.
Prepared by: John Tambornino
Office of Human Services Policy
Office of the Assistant Secretary for Planning and Evaluation
Department of Health and Human Services