Job Retention and Advancement in Welfare Reform

With implementation of welfare waivers and the subsequent enactment of federal welfare reform legislation in 1996, state and local governments began to transform the welfare system through systematic implementation of mandatory programs to help welfare recipients get jobs. These efforts built on fifteen years of innovative efforts and rigorous evaluations at the state and local level. The resulting growth in employment for low-income mothers has surpassed all expectations. Analysis of Census Bureau data by Richard Bavier of the Federal Office of Management and Budget (OMB), for example, shows that real annual earnings (in 2000 dollars) of females heading families with children who were in the next-to-bottom fifth of the income distribution (with family incomes between $13,000 and $21,000 in 2000) increased from $2,254 in 1993 to $9,555 in 2000, a more than four-fold increase over this seven-year period.

Programs that help low-income mothers get jobs are now being augmented by programs that help them keep jobs and move up the employment ladder. Typically the term “retention” means not necessarily keeping a particular job (since changing jobs could be a step up the job ladder) but rather staying steadily employed in any job. Although “advancement” can encompass many things, including increased benefits or the quality of work itself, this brief will focus on wage or earnings growth. The major purposes of this policy brief are to examine why government has an interest in job retention and advancement, to analyze the policies that may promote retention and advancement among current and former welfare recipients, and to describe the current state and local programs aimed at promoting retention and advancement.

Research Findings on Patterns of Retention and Advancement

Research shows that the employment pattern for many welfare recipients has been getting a job, losing or quitting the job, and then experiencing long periods between jobs. Anu Rangarajan and her colleagues at Mathematica Policy Research in Princeton, New Jersey used information from a national survey to study the employment experiences of welfare recipients who found jobs. The survey showed that a substantial majority of adults leaving welfare lost their jobs fairly quickly. Few became steadily employed over five years: only 30 percent were employed for more than three-fourths of the weeks over the five-year period. Clearly, sporadic employment contributes to the low average earnings of these mothers. But it is also likely that it has contributed to poor job advancement, since working intermittently at low-wage jobs is unlikely to lead to better jobs over time.

Not only are welfare leavers likely to lose their jobs, but even when they stay employed, their wages grow slowly if at all. Rangarajan found little wage rate growth over time, around 10 percent over five years on average. However, Rangarajan found substantial increases in earnings due to increased hours of work. Based on data from a national survey, Gary Burtless of the Brookings Institution found that young women who received cash welfare between 1979 and 1981, but then left welfare for employment, experienced annual hourly wage gains of less than one percent through 1990. Despite this lack of wage growth, these young women nearly tripled their earnings over the period because they greatly increased their hours of work. Still, even after this substantial increase, average annual earnings were less than $7,000. Thus, even when wage growth is positive, it is on base earnings that are so small that the wage increases still result in extended periods of low earnings. Of course, these estimates constitute averages, made up both of individuals whose wages and earnings progress well, and of others whose wages are stagnant.

Although these descriptive statistics provide good evidence for why states should be paying attention to job retention and advancement issues, they may be somewhat outdated. This is especially the case because there were positive changes in the economic and policy environments during the 1980s and the 1990s as well as a substantial increase in the level of work among welfare recipients during the 1990s. These changes could have altered prior patterns of wages and earnings for the better. Certainly, both the rewards for working and the costs of not working are greater now.

Despite these improvements in the economy and in employment rates, Pamela Holcomb of the Urban Institute in Washington, D.C., using data from the National Survey of American Families conducted by the Urban Institute, finds that although welfare leavers in 1999 had greater stability in a particular job than did leavers in 1997, there were still substantial levels of job instability. Richard Bavier of OMB, using data from the Survey of Income and Program Participation for a somewhat different period, finds that individuals who left welfare for employment in recent years were more likely to lose their jobs than those who left earlier.

In sum, research strongly supports governments’ continued interest in enhancing the ability of low-income mothers to work steadily and advance in the workforce. If the goal of welfare reform is to substitute a system of work supports for a system of welfare supports, then investments to help low-income parents work steadily and move into jobs that can more fully support their families is integral to that agenda.

Circumstances Associated with Retention and Advancement

Anu Rangarajan and her colleagues found that the characteristics of both the individual and the job were important factors in sustained employment. Welfare recipients who became parents in their teens, had no high school diploma or GED, or had low skills worked less than those without these characteristics. Job characteristics, such as wage level and health insurance availability, were also strongly associated with employment stability. Other research has found that low-skilled workers in some industries and occupations keep their jobs longer and make higher wages than low-skilled workers in other industries and occupations. For example, those who work in health and professional/ educational services are relatively better off on these measures than those who work in other occupations.

Other research has found no link between having a high school diploma and wage growth, other things equal. Post-high school education or training, however, is strongly linked to subsequently higher wages. Some studies have found that changing jobs can be associated with higher wages, but not involuntary changes or too many changes.

Evidence on What Causes Better Retention and Advancement

While these analyses provide useful insights, their causal implications are limited because they do not account for certain intangible factors, such as motivation, that may lead to greater job retention and higher wages. In addition, the findings are often inconsistent. Better evidence comes from evaluations that randomly assign individuals to different treatments. These random-assignment evaluations almost always show that welfare-to-work programs have only small effects on employment stability and little or no impact on wage rates, even in programs that substantially increase earnings as people work more hours. These findings hold true both for programs that emphasize job search and those that emphasize education and skill training (see the policy brief by Gueron and Hamilton in this series). One exception to this pattern is a government program in Portland, Oregon that led to large gains in stable employment and earnings growth. Unlike other programs, the Portland program, while maintaining a strong emphasis on the goal of rapid employment, allowed initial participation in short-term training programs and encouraged participants to seek not just any job, but good jobs with salaries above minimum wage and with benefits. This result is consistent with the studies described above, but provides stronger evidence that the quality of the first job may play a causal role in subsequent job stability and advancement.

The findings discussed thus far are from programs providing pre-employment services such as job clubs, job search assistance, education, and training. There is a more limited amount of research on post-employment programs. A study by Rangarajan and Tim Novak of Mathematica evaluated case management-based retention programs in four states. This study found that only one program produced even small effects on employment stability. Consistent with earlier research, the evaluations found that many recipients lost their jobs within the first few months, with only about 40 percent continuously employed in the same job for 12 months. Although the impact findings were disappointing, there are some lessons that can be drawn from this evaluation. Case managers focused their efforts on helping remove employment barriers, but the evaluation suggests that this by itself did not affect employment outcomes. Also, the programs attempted to provide services to all experimental group members. Perhaps more intensive services targeted to those most at risk of poor outcomes would provide for more efficient use of resources and produce larger impacts. Finally, there is evidence from several experimental evaluations that providing wage subsidies increases job stability, especially for more disadvantaged workers.

State and Local Efforts to Increase Retention and Advancement

Current research does not provide strong evidence for specific programs and policies that could be adopted to help mothers leaving welfare increase their wages. Nonetheless, state and local policymakers and program administrators need to make informed decisions about how to invest resources to boost wages and income. The Administration for Children and Families (ACF) of the U.S. Department of Health and Human Services is currently working with fourteen intervention programs in nine states to test several interventions that may help parents acquire the skills they need to get better jobs with higher wages. The Manpower Demonstration Research Corporation is evaluating the effectiveness of these programs.

A primary assumption of the state programs is that a strong employment focus will continue to characterize welfare-to-work efforts and that enhancements to improve retention and advancement should not result in lengthy absences of individuals from the workforce. Thus, states have attempted to enhance the work first approach, which emphasizes rapid placement in a job, with additional services that may help mothers get better jobs. Within this general context, a program’s design can be examined along five dimensions:

  • Is the primary focus of the program retention or advancement?
  • Are services embedded in work, or do they occur outside it?
  • Are services embedded in pre-employment or post-employment activities?
  • What are the incentives to participate, including whether the program is mandatory or voluntary?
  • Are parents on welfare, former welfare recipients, or all low-income parents eligible for services?

Keeping these five dimensions in mind, we turn to a description of program models that illustrates the range of choices states in the current ACF study are making to promote better jobs and more permanent employment. In Texas, the program’s centerpiece is an individual plan for each parent that specifies both work supports and a career plan for how the parent can move ahead. Intensive case management is provided to assist individuals in complying with the plan. The focus is on both elements related to retention, such as planning for back-up child care, and elements related to advancement, such as options for further training. Recipients in work preparation activities are required to develop a plan and are subject to sanction for non-compliance. Because of fairly low benefit levels in Texas, however, individuals quickly lose eligibility for cash assistance when they become employed. To improve continued engagement and compliance, the state provides a $200 per month expense allowance for one year to working individuals who continue to comply with their plan. Thus, the Texas program addresses both retention and advancement, contains both pre- and post-employment services, and includes incentives-both mandatory requirements with potential sanctions and financial incentives-to participate.

The program in California faces different issues than the program in Texas in that, due to California’s high welfare benefit, a family of three remains eligible for cash assistance even if a parent is working full-time at over $8 an hour. As a result, much of the program’s focus is on job advancement in order to help families achieve a high enough level of earnings so they will no longer depend on welfare. One program being tested in Riverside involves the modification of a 32-hour work requirement to allow individuals established in the workforce to replace some hours of work with education or training. The relaxation of the work requirement allows individuals more time to obtain additional education and training primarily through post-employment, non-worksite training and education programs such as community college classes.

Illinois faces a similar challenge to California in that a substantial earnings disregard has created a very large pool of working recipients on the TANF caseload. Since Illinois does not count months of cash assistance receipt against the federal time limit for working parents, state officials are concerned about the significant number of individuals who appear to have “settled into” low-paying jobs. Illinois is now targeting working recipients who have remained on TANF after six months of employment (suggesting at best low earnings growth) and requiring them to participate in job advancement activities. These activities include substituting training and education for work (as long as a minimum 20 hours per week are worked), establishing a job advancement plan with their employer, and learning how to use past and current work experience to move into better jobs. Part of the underlying assumption in Illinois is that inexperienced workers may not know how to parlay their work experience into promotions or better jobs.

While the programs in Texas, California, and Illinois target current cash assistance recipients, other state programs target former recipients. South Carolina’s program involves aggressive outreach to individuals who have been off cash assistance for at least nine months. Through this outreach and a variety of small financial incentives that reward achievements, the program aims to helpindividuals who are not working get a job, and help current workers move up the job ladder. The South Carolina study is important not only because it will measure the effectiveness of job services, but also because it will test the ability of state agencies to locate and work with individuals who have been off cash assistance for some time.

Programs in New York, Minnesota, and Oregon are focused almost exclusively on retention for hard-to-employ individuals who have serious barriers to employment. Although goals may change in a weaker economy, most welfare agencies have come to believe that almost all individuals are capable of some kind of employment, and that retention is the primary employment problem for more disadvantaged recipients. The Portland model is targeting “recyclers,” individuals who come back on assistance after working for some period of time. New York is using vocational rehabilitation services, work-based education, and specialized work experience programs to improve employment performance by those with medical problems. And Minnesota is developing a program to provide work supports for those with mental health disorders.

Los Angeles is building on the Portland program that helped individuals locate better jobs with higher wage rates and better benefits. The Los Angeles model includes enhanced job clubs which initially assist recipients in looking for better jobs with higher wages and benefits, then reduce job quality expectations in stages over a two-week period for those who are not successful. This approach will be compared with the county’s regular job clubs, which have been shown in a rigorous evaluation to have substantial and broad positive effects on employment and earnings.

Policy Implications

State and local officials have developed and are now implementing a range of strategies for employment retention and advancement. As Congress considers the reauthorization of TANF, lawmakers should keep in mind that flexibility within the current system has allowed state and local officials to design and test programs to address the retention and advancement needs of current or former TANF recipients. The evaluation partnerships that have been formed among local, state, and federal governments, businesses, evaluation companies, and local education and training entities have also supported this effort. States will need resources to continue to test these new approaches to job retention and advancement. The history of research and state demonstration programs shows the need for multiple efforts over time in order to ensure reliable information upon which successful policies can be constructed. As we have shown, the current research on job retention and advancement is not sufficient to recommend sweeping policy changes. Rather, the federal government should continue to provide funds and flexibility to states so that better methods for increasing job retention and advancement can be created.

Additional Reading

Bloom, Dan, and others. 2002. New Strategies to Promote Stable Employment and Career Progression. New York: Manpower Demonstration Research Corporation.

Burtless, Gary. 1995. “Employment Prospects of Welfare Recipients.” In The Work Alternative: Welfare Reform and the Realities of the Job Market, edited by Demetra S. Nightengale and Robert H. Haveman. Washington, D.C.: Urban Institute.

Cancian, Maria, and others. 1999. Work, Earnings and Well-Being after Welfare: What Do We Know? Madison: Institute for Poverty Research.

Martinson, Karin. 2000. The Experiences of Welfare Recipients Who Find Jobs. New York: Manpower Demonstration Research Corporation.

Rangarajan, Anu, and others. 1998. Employment Experiences of Welfare Recipients Who Find Jobs: Is Targeting Possible? Princeton: Mathematica.

Rangarajan, Anu, and Tim Novak. 1999. The Struggle to Sustain Employment: The Effectiveness of the Postemployment Services Demonstration. Princeton: Mathematica.Demonstration Research Corporation.

Strawn, Julie, and Karin Martinson. 2000. Steady Work and Better Jobs. New York: Manpower Demonstration Research Corporation.

The views expressed in this article are those of the authors and do not represent the policy of DHHS.