Sanctions and Welfare Reform

Dan Bloom and
Dan Bloom Director, Health and Barriers to Employment Policy Area, MDRC
Don Winstead
Don Winstead Principal and Founder - Don Winstead Consulting, LLC

January 1, 2002

Sanctions are financial penalties for failing to comply with work or other requirements of state welfare programs. They have been a central feature of the welfare reforms of the 1990s. Although time limits may receive more attention in the media, many more families have been directly affected by sanctions, and sanctions have arguably played a greater role in reshaping welfare recipients’ day-to-day experiences.

Sanctions will be an important topic of discussion when Congress considers the reauthorization of the Temporary Assistance for Needy Families (TANF) block grant in 2002. Some participants in the reauthorization debate will argue that Congress should require all states to use “full-family sanctions” in which a family’s entire cash welfare grant is terminated rather than the partial sanctions in place in some states today. Others will contend that sanctions disproportionately affect the most vulnerable families and that Congress should restrict the use of full-family sanctions and require states to reach out to families before and after reducing or terminating benefits to try to resolve the problems that lead to noncompliance.

The Evolution of Sanction Policies

Financial sanctions have long been used to enforce work-related requirements for welfare recipients. What changed in the 1990s was the severity of the penalties and the frequency of their use.

Until the early 1990s, sanctions did not involve terminating a family’s entire Aid to Families with Dependent Children (AFDC) grant. Rather, the individual who failed to comply (usually the parent) was removed from the grant calculation, resulting in a lower grant and reflecting the view that children should not be punished for their parent’s noncompliance. At the same time, however, noncompliance with eligibility-related requirements (for example, failure to appear for a redetermination interview at the welfare office) did result in closing the case and terminating benefits.

Some welfare staff and administrators complained that sanctions under AFDC were too small to effectively induce recipients to comply with work requirements. In addition, when a family’s AFDC grant was reduced owing to a sanction, the family’s food stamp benefits were generally increased, partly offsetting the cash sanction. Critics also contended that recipients often abused the conciliation process, a federally required procedure intended to resolve participation problems before sanctions were imposed.

In the early 1990s, the federal government began granting waivers of AFDC rules, including waivers that allowed states to impose full-family sanctions. By mid-1996, nearly half the states had received such a waiver. Then the 1996 welfare reforms required states to terminate or reduce benefits “pro rata” when recipients failed to comply with work requirements, but the amount and duration of sanctions were not otherwise specified. The act also changed the food stamp rules so that benefits are no longer increased when the cash grant is cut, and required states to reduce (or, at state option, eliminate) the food stamp grant when a TANF sanction is imposed. Finally, the law eliminated the requirement for states to have a conciliation process.

What States Are Doing

There is great variation in state sanction policies today, although most states have policies that are more stringent than required by federal law. According to the State Policy Documentation Project operated by the Center for Law and Social Policy in Washington, D.C., 36 states use full-family sanctions and 18 of these impose full-family sanctions on the first instance of noncompliance. In the other 18 states, partial sanctions can escalate to full-family sanctions with repeated or continued noncompliance. In most states, repeated noncompliance triggers a sanction that remains in place for at least a minimum period-typically three or six months-even if the individual agrees to comply earlier. In seven states, repeated or continued noncompliance can result in lifetime ineligibility for benefits. It is important to note, however, that several states with large TANF caseloads, such as California, New York, and Texas, do not use full-family sanctions; thus, a substantial proportion of TANF recipients nationwide are not subject to such sanctions.

Most states have procedures to resolve disputes before sanctions are imposed but, in many states, the process is less extensive than the AFDC conciliation process. Almost all states have specific criteria that constitute “good cause” for failing to comply with work mandates. Federal law prohibits states from sanctioning single custodial parents with preschool children if they cannot find child care. Most states also grant good cause exemptions when a recipient is ill or incapacitated, is caring for an incapacitated family member, or lacks transportation.

Although some information on state sanction policies is available, there is little systematic data on how, and how often, sanctions are imposed. One point is clear, however; during the 1990s most states extended work requirements to a greater share of the welfare caseload and began to enforce the requirements more aggressively. These trends were bound to generate more sanctioning.

The most comprehensive national study of sanctioning, conducted by the U.S. General Accounting Office (GAO) in 2000, estimated that 136,000 families (5 percent of the national TANF caseload at the time) received reduced benefits or no benefits due to sanctions in a typical month in 1998. However, GAO counted recipients as being subject to a full-family sanction only in the month when the sanction was imposed, even though sanctions cause many families to remain off assistance for more than one month. Thus, GAO’s 5 percent estimate is low.

If one simply projects GAO’s monthly estimates over several years, it is easy to conclude that well over half a million families have had their cases closed due to full-family sanctions (although there is no way to know how many families are counted more than once in this total). This compares to perhaps 85,000 families who have had their cases closed because of time limits.

There are anecdotal reports that sanctions are often imposed on clients who do not understand the program rules or who have good cause for their failure to comply, but it is impossible to determine how often this happens. Data from a Connecticut program that assists certain families who are terminated from welfare due to noncompliance show that a significant proportion are allowed to return to welfare, often because the individual qualifies for a medical exemption. A recent study by Wisconsin’s Legislative Audit Bureau found evidence that some new mothers were sanctioned in error. A detailed study of eight states conducted by the Inspector General of the Department of Health and Human Services (DHHS) found that, while the states usually explained sanctions clearly, most of the TANF clients who were interviewed did not understand the sanction rules. In addition, the study found that the sanction notices mailed to clients who failed to comply were often confusing or inaccurate.

Other studies have found that caseworkers often have substantial discretion in imposing sanctions and interpret good cause criteria differently; thus, clients who engage in the same behavior are not equally likely to be sanctioned.

A number of states and localities have developed special pre-sanction review procedures to evaluate the circumstances of non-compliant families before sanctions are imposed. These reviews are sometimes conducted by contracted service providers, and may involve home visits. A review process that precedes case closures in Tennessee results in a substantial fraction of proposed sanctions being rescinded, either because the recipient comes into compliance or because the sanction was erroneous. State officials report that the fraction of closures overturned in this manner has dropped over time, in part because the review process has helped staff better understand program policies. Other states and counties have developed post-sanction outreach programs to check on the well-being of sanctioned families and their children, identify obstacles to participation, and try to reengage clients in work activities.

Who is Being Sanctioned and How are They Faring?

A number of states have examined the characteristics of recipients who are sanctioned. Sanctioned clients are a diverse group, but the studies have consistently found that, on average, sanctioned clients have lower levels of education and are more likely than other recipients to face barriers to employment such as physical and mental health problems. This finding may be attributable to the fact that such families tend to remain on assistance longer, increasing the odds that they will be sanctioned.

States that have used administrative records or surveys to follow sanctioned families after they left welfare have also found some consistent results; notably, that sanctioned welfare leavers have lower employment rates and earnings than individuals who left welfare for other reasons.

Fewer studies have examined the material well-being of sanctioned families that are off welfare, and it is difficult to decide on the most appropriate benchmark for assessing their circumstances. One can compare families’ circumstances before and after leaving welfare, but it is not possible to definitively attribute changes to the sanctions. Several studies have compared sanctioned welfare leavers to individuals who left welfare “voluntarily,” but the meaning of this comparison is not always clear.

One of the most comprehensive studies, conducted by Thomas Fraker and his colleagues at Mathematica and published in 1997, surveyed families in Iowa that were in the midst of a 6-month period of ineligibility for benefits. This study confirms that sanctioned families are a diverse group: about 40 percent had higher income than while they were on welfare, while 49 percent had lower income. Even among the latter group, however, there was little evidence of extreme deprivation such as homelessness.

Similarly, studies that have compared the circumstances of sanctioned leavers with those of other leavers have found that sanctioned families report lower income, but not necessarily higher levels of material hardship (e.g., housing problems, food insufficiency). Such hardships are common among all categories of leavers.

Do Sanctions Work?

Most people would agree that sanctions are not designed to reduce welfare caseloads, but rather are intended to persuade recipients to comply with work requirements and to find jobs. Do they achieve this goal?

It is inherently difficult to answer this question because many people respond to the threat of sanctions and never actually experience them. Findings from the National Evaluation of Welfare-to-Work Strategies conducted by the Manpower Demonstration Research Corporation and published in 2001 suggest that programs need to enforce work-related mandates in order to obtain high rates of participation in employment activities. In programs that did not closely monitor attendance and rarely imposed sanctions, participation rates were only slightly higher for adults in the group subject to work mandates than for adults in the control group that faced no mandates. Among the programs that enforced mandates, however, higher sanction rates were not associated with higher participation rates. Some observers contend that programs that do a good job of communicating expectations to recipients do not need to sanction as often.

Another way to examine the utility of sanctions is to see how recipients respond after they are sanctioned. Staff from the Florida Department of Children and Families used administrative data to examine more than 3,000 cases that were closed due to sanctions in June 2000. About 45 percent received cash assistance in the subsequent 6 months, probably by coming into compliance with program rules (many of these clients also worked during the period). Another 32 percent worked but did not receive any cash assistance (in fact, most of these people also worked in the quarter before the sanction was imposed). The remaining 23 percent neither worked nor received cash assistance during the 6 months after they were sanctioned, although a majority of them received food stamps (figure 1).

These results provide further evidence that there are distinct subgroups within the sanctioned population. Some people respond to sanctions by coming into compliance. Others respond by finding jobs-or continue to work in jobs they had before they were sanctioned. In such cases, the sanction may have allowed for early identification of individuals who had already found jobs and stopped communicating with the welfare office. Ten years ago, when there were fewer work-related requirements, recipients’ first scheduled appointment after accepting a job would more likely have been a redetermination interview and the welfare exit might have been classified as eligibility-related. But simply taking a job and leaving welfare without notifying the welfare office now often results in losing food stamps, Medicaid, and other transitional benefits.

The people who neither worked nor returned to welfare may have had other income not captured in the administrative data. Even so, they are a cause for concern. Among those who worked and did not receive welfare in the second quarter after exit, nearly 40 percent earned less than $1,500, suggesting that they were working part-time or were unsteadily employed. The two categories of non-workers and part-time workers account for a substantial minority of the sanctioned clients, and add to the evidence cited above which suggests that some portion of sanctioned clients may face barriers to steady employment. Caseworkers interviewed by the Office of the Inspector General reported that sanctions are very effective for certain groups of recipients (e.g., those who are working off the books), but usually do not provide motivation for clients facing multiple barriers to employment.

Do Full-Family Sanctions Work Better Than Partial Sanctions?

It is not clear whether full-family sanctions generate larger changes in recipients’ behavior than partial sanctions. It may be that most of the recipients who are able to follow the rules are induced to do so by partial sanctions. If that were true, full-family sanctions could end up imposing greater penalties on people who are unable to comply.

Some analysts have argued that full-family sanctions are particularly important in the context of time limits: when clients do not respond to partial sanctions, they often end up exhausting their months of eligibility without obtaining needed employment services. Others argue that full-family sanctions increase the likelihood that families with the most serious employment barriers will simply exit from welfare without receiving needed services.

Unfortunately, there is very little direct evidence to inform this debate. Many welfare-to-work programs using partial sanctions have generated high rates of participation in employment activities-and substantial increases in employment and reductions in welfare use-but there have been very few comparable evaluations of programs using full-family sanctions.

One study by Robert Rector and Sara Youssef of the Heritage Foundation in Washington, D.C. found that states with strong work requirements and full-family sanctions have experienced much larger welfare caseload reductions than other states. This seems plausible, since full-family sanctions result in more case closures, but the study does not address whether the sanctions induced more people to work. On the other hand, a study by Sandra Hofferth, Stephen Stanhope, and Kathleen Harris of the University of Maryland did not find an association between stricter sanction policies implemented under waivers and work-related welfare exits (stronger work requirements were associated with work exits).

Despite strong views on both sides, at this point there is not enough solid evidence to draw firm conclusions about the relative effectiveness of full-family and partial sanctions.

Implications for TANF Reauthorization

Advocates will likely push for two kinds of changes in federal sanction policy when Congress debates the reauthorization of TANF. Some on the right will argue that all states should be required to use full-family sanctions. From the left will come a push for new restrictions on the use of full-family sanctions and new federal rules designed to reduce erroneous sanctions and increase outreach and assistance both before and after sanctions are imposed. Our view is that Congress should not impose new restrictions on state flexibility unless there is reasonably strong evidence of problems.

Although many administrators believe that full-family sanctions play a critical role in creating a work-focused welfare system, some states believe that such sanctions are not necessary to achieve this goal. Given the potential risks associated with full-family sanctions and the lack of definitive evidence on this issue, it seems reasonable to allow states to proceed with partial sanctions if they can achieve the outcomes required under the law (e.g., high work participation rates).

By the same token, there does not seem to be sufficient evidence to restrict the use of full-family sanctions. Because state programs include a mix of policies related to payment amounts, earnings disregards, employment strategies, and so forth, states should continue to have flexibility to set the sanction policy that works best in their program.

The question of new federal requirements for pre- and post-sanction outreach efforts is more difficult. The evidence suggests that there is some cause for concern; namely, that families with serious barriers to employment may constitute a significant minority of those sanctioned, but the knowledge base is still very thin. Moreover, extensive procedural requirements from the federal government can make it difficult for states to enforce work requirements.

One of the most effective ways to reduce inappropriate sanctions is to expand the set of work-related activities that count toward a state’s participation rate, particularly for recipients who have impairments that limit their ability to work or who need to combine work activities with treatment or rehabilitation. With more options available, states would be better able to devise employment plans that fit the needs of particularly disadvantaged clients and thereby avoid the need for sanctions.

Congress should also consider requiring states to describe, in their TANF state plan, what safeguards they will implement to assure that individuals who are subject to sanctions have information on potential exemptions and on what they must do to have the sanction lifted. This policy would send a signal that Congress is concerned about inappropriate sanctions without significantly restricting state flexibility. DHHS could study various review and outreach procedures and provide states with guidance on which appear to be most efficient and effective.

One of the most effective approaches would be to target outreach efforts on situations in which full-family sanctions have a minimum duration. Full-family sanctions that can be lifted as soon as recipients come into compliance are not substantially different from the case closures for eligibility reasons that always existed in the old AFDC program. Sanctions with a minimum duration, on the other hand, can cause considerable harm if they are imposed on families that have serious problems that limit employability, but these problems are not known to the welfare agency. In this case, the family would remain off welfare for a mandatory period without being offered services designed to eliminate or reduce their work barriers.

Finally, additional research on sanctions should be encouraged. For example, a random assignment study could compare the effects of full-family and partial sanctions (or, perhaps, other strategies for increasing engagement in work activities), examining participation rates, employment and earnings, duration and amount of welfare payments, family income, and child well-being. This type of study, along with other research on the implementation of sanctions, would provide a firmer base of evidence to inform the ongoing policy refinements of states as well as the next reauthorization debate.

Additional Reading

Fraker, Thomas, and others. 1997. Iowa’s Limited Benefit Plan: Summary Report. Washington, D.C.: Mathematica Policy Research.

Hofferth, Sandra L., and others. 2001. “Exiting Welfare in the 1990s: Did Public Policy Influence Recipients’ Behavior?” (Unpublished manuscript). College Park: Department of Family Studies, University of Maryland.

Office of the Inspector General, U.S. Department of Health and Human Services. 1999. Temporary Assistance for Needy Families: Improving Client Sanction Notices.

Office of the Inspector General, U.S. Department of Health and Human Services. 1999. Temporary Assistance for Needy Families: Educating Clients About Sanctions.

Pavetti, LaDonna, and Dan Bloom. 2001. “State Sanctions and Time Limits.” In The New World of Welfare, edited by Rebecca M. Blank and Ron Haskins. Washington, D.C.: Brookings.

Rector, Robert, and Sarah Youssef. 1999. The Determinants of Welfare Caseload Decline. Washington, D.C.: Heritage Foundation.

Wisconsin Legislature Audit Bureau. 2001. Audit Summary: Wisconsin Works (W-2) Program (Report 01-7).

Goldberg, Heidi, and Liz Schott. 2000. A Compliance-Oriented Approach to Sanctions in State and County TANF Programs. Washington, D.C.: Center on Budget and Policy Priorities.