The opinions expressed herein are those of the author and are not necessarily those of the Urban Institute, its trustees, or its sponsors.
Good morning Chairman Shaw and members of the Subcommittee on Human Resources. I am pleased to have this opportunity to talk with you today about time on welfare and welfare dependency. I plan to address four aspects of this issue. First, the total time families spend on the welfare rolls. Second, the percentage of the AFDC caseload which may be affected by time limits. Third, the factors associated with long-term welfare receipt. Finally, I will discuss some implications of this information for reforming the welfare system.
My testimony will amplify and support the following points:
- A complete picture of time on welfare requires an understanding of two seemingly contradictory facts: the majority of families who ever use welfare do so for relatively short periods of time, but the majority of the current caseload will eventually receive welfare for relatively long periods of time.
- A substantial fraction of the AFDC caseload potentially could be affected by policies to time-limit AFDC benefits. On average, at a given point in time, about 70 percent of current AFDC recipients have received AFDC for more than 24 months and 48 percent have received assistance for more than 60 months.
- The strongest predictors of whether a welfare recipient will leave welfare for work are recent work experience and educational attainment, including mastery of basic skills.
- Some families who spend long periods of time on the welfare rolls are likely to need more assistance than is provided by traditional welfare-to-work programs to make a successful transition from welfare to work. Providing this additional assistance will require flexibility, a broad range of "allowable" welfare-to-work activities and additional or redirected staff resources.
Estimates of Total Time on Welfare
The estimates of time on welfare that I will present today are based on the results of a simulation model that uses monthly data on welfare receipt from the National Longitudinal Survey of Youth to estimate movement on and off the welfare rolls. These estimates assume that the AFDC caseload is in "steady state," neither increasing or decreasing. During times of transition these estimates would need to be adjusted to reflect the changing nature of the AFDC caseload. In general, when the caseload is increasing there is an increasing percentage of shorter-term recipients. When the caseload is declining, longer-term recipients generally account for an increasing fraction of families receiving assistance.
The welfare system is an extremely dynamic system. In an "average" year, about one-half of the AFDC caseload leaves the welfare rolls. The best available estimates indicate that between one-half and two-thirds of those who leave do so because they have found paid employment. A small percentage (less than 15 percent) leave for marriage and the remainder leave for a variety of other reasons. Those who leave are replaced by new applicants who have never received assistance before and by families who have received assistance previously and are returning to receive assistance again.
The majority of families who leave the welfare system do so after a relatively short period of time -- about half leave within a year; 70 percent within two years and almost 90 percent within five years. But many return almost as quickly as they left -- about 45 percent return within a year and 70 percent return by the end of five years.
When one takes into account all of this movement on and off the welfare rolls, only a moderate fraction of recipients who ever turn to the welfare system for support end up spending relatively long periods of time on the welfare rolls. Over the course of their lifetimes, about one-third of women who ever use welfare will spend longer than five years on the welfare rolls and 60 percent will spend 24 months or longer receiving assistance.
It is important to note that the figures I just presented are for women who ever turn to the welfare system for support, regardless of whether they are currently receiving assistance or not. A very different picture of time on welfare emerges if one examines the total time families currently receiving welfare will spend on the welfare rolls over the course of their lifetimes. About 90 percent of those currently on the rolls will eventually spend more than 24 months on the welfare rolls and 76 percent will receive welfare for longer than five years.
An obvious question is, "Why are these two pictures of time on welfare so different?" The answer to this question is complex, but it primarily lies in the fact that we are estimating total time on welfare for two very different groups of recipients. In the first case, we are estimating total time on welfare for all families who ever turn to the welfare system for support, regardless of whether they stayed for a month or for 25 years. In the second case, we are asking the question of a much smaller and very specific group of recipients -- those who are currently receiving assistance. Over time, long-term recipients accumulate in the system so they are over-represented among the current caseload. By definition, the only short-term recipients represented among current recipients are those who have applied for assistance recently while the long-term recipients include recipients who first received welfare up to 25 years ago.
In sum, the majority of families who ever turn to the welfare system for support will use it for relatively short periods of time, but the majority of families receiving assistance at any given point in time (i.e., the current caseload) will eventually receive welfare for relatively long periods of time. While these statements often seem contradictory, both are accurate and both are necessary to present a complete picture of time on welfare.
Behind these total time estimates, are very different patterns of welfare use. Some recipients use welfare for a short period of time, leave and never return; others use welfare intermittently, returning for short-term assistance when a job ends or when a family crisis occurs. Still others spend long periods of time continuously receiving welfare. Because of these different patterns of welfare use, it is difficult to talk about an "average" welfare recipient. My research shows that, on average, women who ever use welfare will receive assistance for about six years and current recipients will receive assistance for about thirteen years. While accurate, taken by themselves, these figures are misleading because they give undo weight to the experiences of the extremely small number of recipients who spend very long periods of time (as much as 25 years) receiving welfare. These very long-term recipients do, in fact, exist, but they are the exception, not the rule. Thus, it is not accurate to describe the "typical" length of stay on welfare as 13 years.
Families Affected by Time Limits
Thus far, I have talked about the total time recipients will spend over the course of their lifetimes. For purposes of estimating the percentage of families who may be affected by policies to time-limit AFDC benefits at any given point in time, only the time welfare recipients have spent on the welfare rolls to date is of interest. On average, about 70 percent of families receiving assistance at a given point in time have already received assistance for at least 24 months and 48 percent have received assistance for more than 60 months. Assuming no change in behavior, these are the percentages of families currently receiving assistance who would be eligible for and actually receiving benefits under the current system, but would not be eligible to receive assistance with a two-year or five-year time limit in place. An alternative interpretation of these numbers is the percentage of families who would have to leave welfare sooner than they would under the current system in order to not be adversely affected by a time limit on benefits.
Characteristics Associated with Long-Term Receipt
Long-term welfare receipt is not an entirely random event. Recipients who first receive welfare when they are young, have never married, have low levels of education and have no recent work experience are all over-represented among recipients with longer stays on welfare. Minority recipients are also over-represented. When these factors are all considered simultaneously, the strongest predictors of whether a recipient will leave welfare for work in a given month are recent work experience and educational attainment, including mastery of basic skills. Thus, recipients who spend long periods of time on the welfare rolls are primarily women with limited job prospects. In fact, the employability of those who are most likely to reach a five-year time limit does not look especially promising. Half of those who spend longer than five years on the welfare rolls enter AFDC with no labor market experience and 63 percent of these women have less than a high school education. Also, 42 percent first received welfare when they were under age 25, the time when the vast majority of workers make investments in education and gain experience in the labor market that prepares them for stable future employment.
Implications for Reform: Lessons from Utah's Single Parent Employment Program
To provide some insight into the implications of this information for reforming the welfare system, I would like to briefly describe Utah's experience of reforming welfare through its Single Parent Employment Program. Utah's reform is the only reform implemented to date that does not exempt any AFDC recipients from participation in work-related activities, making it an important laboratory for examining the issues states are likely to face as an increasing percentage of the AFDC caseload, including those who have received assistance for long periods of time, are required to look for work. Utah's reform emphasizes quick entry into the labor market and provides recipients who leave welfare for work with a variety of transitional services to ease the transition into the labor market. Families who do not participate in agreed upon activities face the loss of all cash assistance after an extensive conciliation process.
According to workers, the requirement that they work with all recipients, regardless of their needs or current ability to find and sustain employment, has changed their work quite dramatically. This change has been more pronounced over time as more and more of the job ready recipients find employment and leave the welfare rolls. In hindsight, administrators and staff all acknowledge that they were unprepared for how difficult it would be to work with some families, especially those who have received welfare for extended periods of time. Over time, they have found that the vast majority of these families experience multiple barriers to employment, ranging from serious mental health problems to learning disabilities to substance abuse to children's mental health and medical problems to domestic violence.
To address the needs of these families and help move them into the labor market, Utah has started to hire more skilled workers, generally professionals with a Master's Degree in Social Work or Counseling. These workers carry smaller than average caseloads and serve as a resource for other staff who are having difficulty moving particular recipients into the labor market. In some offices, substance abuse and mental health professionals are co-located in the welfare office. The goal for families who experience a broad range of barriers to employment continues to be employment. However, Utah's experience has taught them that while it is feasible to require every family to participate in activities that will eventually allow them to find unsubsidized employment, not every family can immediately sustain full-time or even part-time employment.
Utah's experience highlights several important points. Most importantly, it suggests that reforming the welfare system in many areas is likely to involve more than simply requiring families to work. When mandated to find work, some families are likely to find employment on their own and to do so more quickly than they would had they not been faced with such a mandate. Other families will find work with limited assistance. But some families are likely to need more assistance than welfare-to-work programs traditionally provide. Providing this assistance requires flexibility, a broad range of "allowable" activities and additional or redirected staff resources. Utah's experience suggests that welfare reform can be used as an opportunity to help families with a variety of needs and circumstances take the necessary steps to become self-sufficient and it is possible to incorporate the services these families need to leave the welfare rolls in a program that emphasizes employment.
For additional information:
Pavetti, LaDonna A. "Who is Affected by Time Limits?" in Welfare Reform: An Analysis of the Issues. Edited by Isabel V. Sawhill. The Urban Institute, 1995.
Pavetti, LaDonna A. "How Long Do Families Stay on AFDC?, Who Spends Longer Periods of Time Receiving Welfare? Why Do People Who Leave Welfare for Jobs Return to the Welfare Rolls?" in Looking Before We Leap: Social Science and Welfare Reform. Edited by R. Kent Weaver and William T. Dickens. The Brookings Institution, 1995.
Pavetti, LaDonna A. "Questions and Answers on Welfare Dynamics" Urban Institute Mimeo, September 11, 1995.
Pavetti, LaDonna A. The Dynamics of Welfare and Work: Exploring the Process by Which Women Work Their Way Off Welfare. Dissertation Series #D-93-1, Malcolm Wiener Center for Social Policy, The Kennedy School of Government, Harvard University.
1. Dr. Pavetti joined the Urban Institute after completing her Ph.D. in public policy at Harvard University. She recently conducted a study on the early implementation of five state welfare reform demonstration projects designed to increase participation in work and work-related activities. Dr. Pavetti is also a noted expert on welfare dynamics. Her most recent research combines qualitative and quantitative data analysis focused on analyzing the dynamics of welfare and work.