by LaDonna A. Pavetti and Amy-Ellen Duke with contributions from
Clemencia Cosentino de Cohen, Pamela Holcomb, Sharon K. Long, and Kimberly Rogers.
The Urban Institute
September, 1995
This study uses data from discussions with state and local officials, case record reviews and management reports to examine key implementation and operational issues in five states that are currently conducting welfare reform demonstration projects: Colorado, Iowa, Michigan, Utah and Vermont. These states were selected from the 24 states that had been granted permission to operate welfare demonstration projects by the U.S. Department of Health and Human Services (DHHS) as of September 1994. These states were chosen because their reforms emphasized strategies designed to increase participation in work and work-related activities and they were far enough along in the implementation of their reforms to provide information that would be useful to policy makers and program administrators.
The overall goal of this study is to provide an in-depth assessment of key implementation and operational issues in these five states in order to inform the policy debate over how to reform the welfare system at the national, state and local level. The specific objectives of the study are to:
Identify the strategies states have used to increase participation in the Job Opportunities and Basic Skills (JOBS) training program and/or other work-related activities and to document the problems they have encountered while trying to achieve higher levels of participation in these work-related activities;
Examine the role child care plays in a state's ability to increase participation in JOBS, to enforce social contracts or time limits, or to increase the number of recipients who are employed in unsubsidized jobs either part or full-time;
Determine the extent to which the culture of the welfare offices have changed from one of check issuance to one of finding employment for recipients as a result of program changes; and
Identify the early lessons welfare and employment program administrators and policy makers can learn from these experiences.
A. Strategies for Increasing Participation in Work-Related Activities
The states have used a variety of strategies to increase the fraction of the caseload participating in work and work-related activities including:
Expanding participation mandates to a larger fraction of the AFDC caseload by reducing exemptions and serving a larger fraction of mandatory participants than is required under current law;
Redefining participation to include unsubsidized employment and activities such as substance abuse and mental health counseling and parenting classes;
Increasing the penalty for not participating in work or work-related activities by implementing more stringent sanctions or time-limiting AFDC benefits not conditioned on work; and
Liberalizing the work expense and earned income disregards.
Since the implementation of their demonstration projects, all five states report increases in the percentage of their caseload that is engaged in employment or employment-related activities. However, it is too soon to tell whether these increases can be attributed specifically to the reforms. It is especially important to note that the current economic recovery has provided these states with a nearly ideal set of economic conditions for implementing employment-related reforms. Between 1992 and 1995, Michigan's unemployment rate declined from 9.3 to 6.0 percent. Over this same period, the unemployment rates in Colorado, Iowa, Utah and Vermont declined to 3.5, 3.3, 3.1 and 4.2 percent respectively, substantially lower than the national average of 5.8 percent.
B. Implementation Issues Associated with Increased Participation
All of these states budgeted for additional expenditures to implement their projects, primarily to hire additional staff to handle the increased number of participants in employment- related activities and to pay for additional child care expenses. To contain program costs, the states have continued to rely on education and training resources outside of the JOBS program, primarily community colleges and programs funded through the Job Training and Partnership Act (JTPA). Three of the five states (Iowa, Michigan and Utah) also shifted the focus of their JOBS program from education and training to more direct job placement.
As states have moved to engage a larger fraction of their caseload in work or work-related activities, they have also had to address several obstacles to increasing participation. In particular, staff in all the states report that it is difficult to get recipients to initially attend required program activities. Some states also report that problems such as serious mental health problems or substance abuse problems are difficult to overcome. Utah, the state that has had its welfare reform project in place the longest, has devoted considerable time and energy to understand and develop strategies for addressing the problems of harder-to-serve recipients. These strategies include: (1) hiring specialized "treatment workers" who are more skilled than regular case managers and carry smaller caseloads; (2) reviewing all cases that have been on assistance for longer than six months through formal case staffings; and (3) implementing a short-term treatment program for persons with less severe mental health problems.
C. The Role of Child Care
Each of the states in this study budgeted for and experienced an increase in the use of child care assistance, although in most states the use of child care increased more rapidly than state administrators anticipated. States were affected by and handled these increased costs in very different ways. In Utah, the increase in child care utilization was offset by a substantial decrease in their AFDC caseload, resulting in lower combined AFDC and child care costs than at the start of the demonstration project. Iowa first shifted its JOBS program from an education and training focus to more direct job placement, then restricted eligibility for its income-tested child care program to families working 30 hours or more per week with incomes below the poverty line.
In Vermont, more families are choosing to use unregulated care providers than anticipated. Thus, even though Vermont is seeing a substantial increase in the utilization of child care, they have been able to pay for the unexpectedly sharp increase in unregulated care through the additional funding originally budgeted for more costly regulated care. Colorado covered part of the initial increase in child care expenditures by shifting child care funding previously allocated for low-income working families. Shortly after the start of the demonstration project the state experienced a shortage of funds for low-income working families and was forced to temporarily establish a waiting list for low-income child care. Michigan has allocated substantial additional funding for child care, but has had difficulty processing applications in a timely manner for all of the additional recipients who are participating in required program activities. The state planned to hire new staff to address this problem. In addition, staff who previously worked as case managers for participants in education and training programs are now spending some of their time processing child care applications.
Staff report recipients required to participate in program activities have generally been able to find child care, although it takes some families longer to find care than others. Consequently, staff's overall assessment is that child care has not been a substantial barrier to participation. At the same time, staff do report that some types of care --infant care, evening and weekend care, temporary care needed for schoolage children during school breaks and during the summer-- are harder to find than other types of care. Finding care is also quite difficult for families living in areas with limited public transportation because they need providers located either close to home or work so that the two-part commute is manageable. Finally, child care centers are often filled to capacity, leaving parents to rely on informal care arrangements even though they may prefer not to do so.
D. Changing the Culture of the Welfare Office
These five states report that it is extremely difficult to change the culture of the welfare office, at least in part, because it not only involves changing program rules, but also changing attitudes and expectations among staff and recipients. In most of the states, administrators envisioned income maintenance workers and JOBS case managers both being involved in changing the culture of the welfare office. While this has happened in Vermont and Utah, in general, the JOBS program has played a much more prominent role in changing the culture of the welfare system than the welfare office itself. Vermont and Utah specifically set out to involve all staff in the planning and implementation of their projects and to instill in them the understanding that business was no longer being conducted as usual. The other states primarily focused on training staff in the technical aspects of their reforms.
In general, administrators report that is difficult to change attitudes and expectations of the income maintenance staff. They were hired primarily because they were good at paying attention to details, managing deadlines and completing technical tasks. While some are interested in talking more with recipients about their plans for the future, not all of are. In addition, because income maintenance workers have so many deadlines to meet and carry large caseloads, it is often difficult for them to focus on activities other than those absolutely necessary. But Iowa's experience suggests that it is possible to dramatically change the culture of the welfare system by increasing the prominence of the JOBS program rather than by changing the overall structure of the welfare office. Because the majority of recipients in the state are required to participate in the JOBS program and are sanctioned if they fail to participate in required program activities, uninterrupted cash assistance is only available to Iowa families who meet their obligation to participate in self-sufficiency activities, an approach that dramatically alters the social contract between recipients and the government.
E. Lessons Learned
The five states included in this study have made numerous changes in their welfare systems. While some changes represent incremental reforms to the current welfare system, others represent a significant attempt to transform the welfare system to a system that is much more orientated toward employment. Although these states are all in the early stages of implementation of their reforms, there is still much that can be learned from their experiences. The major lessons include:
There is more than one way to reform the welfare system.
It is possible to substantially increase participation in work or work-related activities in a relatively short period of time, especially within the context of a healthy economy, but, at least in the short-term, it costs money to do so.
If large numbers of recipients are placed in unsubsidized employment and caseloads decline substantially, those recipients left behind are likely to be recipients with multiple barriers to employment.
Even though sanctions have received very little attention in the national welfare reform debate, several of these states' experiences suggest they are an important part of an overall strategy for increasing participation in work and work-related activities.
Child Care plays an important role in transforming the welfare system into a more work-oriented system.
Changing the culture of the welfare office is complex and involves bringing about change on many different levels.
The up-front planning process plays an important role in bringing about change in an orderly fashion.