Search Type Title Author/Investigator Search Advanced Search

Table of Contents | Previous | Next

I. Introduction

The 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) provided a block grant to states to create the Temporary Assistance for Needy Families (TANF) program.  In doing so, it required states to engage certain minimum percentages of their TANF caseloads in specified work and work-related activities for a specified number of hours per week.  States that do not meet these required work participation rates are financially penalized.  In reauthorizing the TANF program, the Deficit Reduction Act of 2005 (DRA) effectively increased the work participation rate required of states by changing the way the rate is calculated.  As of October 1, 2006, each state is required to engage 50 percent of all families and 90 percent of two-parent families in work activities, although these standards will be adjusted downward for any caseload decline that occurs after 2005 for reasons other than changes in federal requirements and changes in state rules since 2005 that directly affect a family’s eligibility for assistance.  The work participation rate achieved by each state will be calculated by using a base (that is, a denominator) that includes not only families receiving TANF assistance but also families receiving assistance in separate state programs that count toward maintenance of effort (MOE) requirements.  This is a substantial change from previous law, which also set standards of 50 and 90 percent but adjusted the requirements downward for caseload declines that occurred after 1995.  Moreover, the work participation rate was previously computed by using as the base only families receiving TANF assistance and not those in state-funded MOE programs.  Largely because of how the caseload reduction credit was defined, the real participation rates for active cases that states had to meet before the DRA were substantially below 50 and 90 percent.  As of fiscal year 2007, states face the challenge of achieving participation rates that are considerably higher and close to the 50 and 90 percent standards set in the law.

Sanctions, or financial penalties for noncompliance with program requirements, have long been perceived as a major tool for enforcing program requirements and for encouraging TANF recipients who might not be inclined to participate in work activities to do so.  They may be particularly useful in motivating recipients to work toward employment before exhausting their time limit on cash assistance.  The logic behind sanctions is that adverse consequences—such as a reduction in the TANF grant (a partial sanction) or gradual or immediate termination of the TANF cash grant (a full-family sanction)—can help influence the decisions that welfare recipients make.  Therefore, as states consider their options for meeting the higher work participation rates, they are likely to consider how they might make better use of sanction policies and procedures to encourage higher levels of participation in program activities.

Sanctions may influence the work participation rate in one of two ways.  First, sanctions may encourage recipients who are not inclined to participate in program activities to do so.  In this case, a state’s work participation rate will be higher than it would be in the absence of sanctions because the numerator of the rate will increase.  This effect may vary, depending on the type of sanction.  If gradual or immediate full-family sanctions induce more recipients to participate than do partial sanctions, all else equal, work participation rates would be higher in states with full-family sanctions.  Second, when gradual or immediate full-family sanctions are applied to noncompliant recipients, they eliminate those cases from the TANF caseload, thereby removing them from the denominator of the TANF work participation calculation.

States interested in modifying their sanction policies and procedures are likely to seek strategies that can help them meet the effectively higher participation requirements.  Given that families sanctioned and still on the caseload for more than 3 months in a 12-month period remain in the denominator of the participation rate, states with partial-sanction policies may consider several alternatives to raise their participation rate.  First, they can move to a gradual or immediate full-family sanction to eliminate noncompliant cases from the caseload.1 Second, they can implement new outreach and service strategies to encourage sanctioned recipients to come into compliance with program requirements, or to reduce the likelihood that a sanction is ever imposed.  Third, they can reduce the amount of time a sanction must be in place before it can be “cured” by a participant’s compliance (e.g., from 6 to 3 months) or eliminate entirely the minimum sanction durational requirements.  States with a gradual or immediate full-family sanction policy in place may aim to impose full-family sanctions more quickly.  For states with gradual full-family sanction policies, rapid imposition of full-family sanctions could mean reducing the time to move from a partial to a full-family sanction (e.g., from 6 to 3 months).  States with gradual or immediate full-family sanction policies may consider changing how they implement such policies in order to ensure that all sanctions are imposed more swiftly.

This report documents how some jurisdictions are responding, in their sanction policies and practices, to the new requirements imposed by federal law.  The study focuses on sanction policies and practices in eight sites located in seven states.  It provides information on changes these jurisdictions have made with respect to sanctions and ways in which they use the sanction process to increase engagement in program activities—information that may be useful to other states and counties considering whether adjustments in sanction policies can increase participation in program activities.  This chapter describes the study and the context in which it was conducted.  Chapter II describes the study sites and the reason for their selection.  Chapters III through V discuss the approaches the study sites have taken with respect to three major program components of the sanction process: (1) defining and communicating program requirements; (2) monitoring participation in program activities; and (3) re-engaging noncompliant clients in program activities.  Chapter VI reports on an analysis of the relationship between sanction policies and practices and work participation and related outcomes such as employment.  Finally, Chapter VII summarizes our findings, identifies unanswered research questions, and presents opportunities for future study.

A. State Policy Context

The most crucial aspect of a sanction is its effect on a family’s TANF grant.  There are four basic categories of sanction policies with respect to their effect on the TANF grant (1) partial grant sanctions; (2) gradual full-family sanctions; (3) immediate full-family sanctions; and (4) pay-for-performance.  When a partial sanction is imposed, a family’s cash grant is reduced—by a specified dollar amount, by a specified percentage, or by the noncompliant adult’s portion of the grant—but the family continues to receive some portion of its benefits.  When a gradual full-family sanction is imposed, the first instance of noncompliance results in a partial grant reduction for a specified period.  The full grant is then restored for families that come into compliance, or the case is closed for families that do not.  Depending on the state, a second or subsequent instance of noncompliance may result in immediate TANF case closure.  When an immediate full-family sanction is imposed, a family loses all of its cash assistance either immediately or soon after it is identified as noncompliant.  Under the pay-for-performance model, which only Wisconsin has implemented to date, a family receives assistance only for the hours it participates in required work activities; families that do not participate at all receive no assistance, and families participating for fewer hours than required receive proportionately reduced assistance.

In recent years, many states have shifted the basic structure of their sanction policy, sometimes as part of a larger reform of their welfare systems.  All states that have done so have moved to a more stringent model—that is, from a partial to a full-family (six states) or from a gradual full-family to an immediate full-family sanction policy (three states).  Table I.1 compares states’ sanction policies at the time of an earlier review of sanction policies Mathematica Policy Research, Inc. (MPR) conducted in March 2003 (see Pavetti et al. 2003) to their current sanction policies.  At present, six states have partial- sanction policies, 23 states have gradual full-family sanction policies, and 21 states have immediate full-family sanction policies in place.

 

Table I.1. State Sanction Policies: March 2003 and March 2007a

Sanction Policy in 2003 and Change to 2007		States
Partial-Sanction States in 2003	Partial Sanction in March 2007	CA, DC, ME, MO, NY, VT
	Gradual Full-Family Sanction in March 2007	IN, MN, NH, RI, WA
	Immediate Full-Family Sanction in March 2007	TX
Gradual Full-Family Sanction States in 2003	Gradual Full-Family Sanction in March 2007	AK, AL, AR, AZ, CO, CT, DE, GA, IL, MA, MT, NJ, NM, ND, OR, SD, UT, WV
	Immediate Full-Family Sanction in March 2007	LA, NC, NV
Immediate Full-Family Sanction States in 2003	Immediate Full-Family Sanction in March 2007	FL, HI, IA, ID, KS, KY, MD, MI, MS, NE, OH, OK, PA, SC, TN, VA, WY

a Policies reflect the most stringent policies in the state.  States in italics made changes to the basic structure of their sanction policy between 2003 and 2007.  The table does not include Wisconsin, which operates a pay-for-performance system and reduces benefits by an amount equal to the minimum wage multiplied by the number of hours the individual failed to participate in required activities.  Wisconsin also imposes “strikes” for noncompliance within each program component.  After imposition of a third strike, the case is permanently ineligible for benefits in that component.

 

In addition to the effect on the TANF grant, four other major dimensions define the structure and stringency of a state’s sanction policy: (1) the minimum duration of the sanction, (2) the steps a recipient must take to reverse or cure a sanction, (3) the agency response to multiple acts of noncompliance, and (4) the sanction’s effect on Medicaid and food stamp benefits.  The minimum duration of the sanction is the length of time a sanction must remain in place.  This period varies—among states and for repeated instances of noncompliance within a state—from 1 to 12 months, or it can simply last until program compliance.  The requirements to reverse or cure a sanction dictate what a family must do to be considered compliant again and resume receipt of full benefits.  Some states simply require the noncompliant adult to indicate a willingness to comply while others require a minimum level of participation in program activities.  Jurisdictions also vary in how they deal with multiple acts of noncompliance—that is, how they treat families that move in and out of sanction status.  Typically, states increase the stringency of the sanction with each act of noncompliance either by intensifying the effect on the TANF grant or by imposing longer minimum durations or stricter cure requirements.  Finally, in some states, sanctions result in loss of Medicaid for the adult and/or in a reduction of the family’s food stamp benefit; in other states, neither is affected.  The effect of TANF sanctions on food stamps in particular is critical to the overall effect of sanctions on disposable income.  The food stamp benefit calculation and benefit levels are established at the federal level  (with some state options) and take TANF income into account.  Thus, for states with lower TANF benefits and in which sanctions do not affect food stamp benefits, the effect of full-family sanctions on disposable income may not differ very much from the effect of full-family sanctions in states that have higher TANF benefits and in which sanctions do not affect food stamp benefits.  Since MPR’s last review in 2003, some states have made changes along these four dimensions while maintaining the basic structure of their sanction policy, though no comprehensive information is available on which states have made such changes.  Some of these changes increased and others eased the stringency of sanctions. 

B. Study Questions and Methodology

There are two primary objectives of the current study of sanction policies and practices.  The first is to shed light on how sanctions are used in efforts to increase participation in program activities.  The second is to define a possible research agenda to address outstanding questions about sanction policies and practices.  To accomplish these objectives, the study addresses five key research questions:

• What sanction policies exist, how have they changed, and what has been the role of sanctions in encouraging participation in work activities?  To understand the role of sanctions in encouraging participation in work activities, it is important to develop a picture not only of the details of states’ sanction policies—including how and when sanctions affect cash and other benefits, minimum sanction periods, cure requirements, and approaches to repeat acts of noncompliance—but also of the goals of the sanction policy and the underlying philosophy about when, why, and how sanctions should be used.  Exploring recent changes in sanction policy—the nature of the changes, the impetus for change, who was involved in making the changes—can help us understand the underlying philosophy as well as the observed changes in sanction rates and related outcomes over time.  We examine staff perceptions of the role of sanctions in encouraging participation in work activities and examine administrative data as well.  We pay particular attention to (1) how sanction rates and engagement in activities have changed with the growing emphasis on achieving higher work participation rates, and (2) the contribution that explicit sanction policy, procedural, or service delivery changes may have made to these trends.

• How are sanction policies implemented at the local level?  A review of state sanction policy can provide basic information about the consequences of noncompliance in theory but reveal little about how policies play out in practice.  Procedures for implementing sanction policy can vary across local welfare offices or even among staff within the same office.  The way in which policies are applied may depend on how well local welfare office staff understand sanction policies and procedures, how and how well local welfare staff communicate work requirements and sanctions for noncompliance to recipients, how local staff perceive the role and utility of sanctions, how much discretion local staff may exercise in implementing sanctions, and how staff make decisions regarding sanctions.  Understanding how sanction policies are implemented in practice can shed light on how existing policy can be best used, and can help explain variation in sanction rates and related outcomes between states or localities with similar policies.

• What strategies do states/local sites use to avoid imposing sanctions?  An important use of the sanction process is to encourage program participation before recourse to a sanction becomes necessary.  The extent to which efforts to avoid sanctions are successful may depend on when and how local welfare office staff identify barriers to participation, the services available to address those barriers, and how participants gain access to those services.  An exploration of the work requirement exemptions developed by sites, how sites define good cause, how sites assign and monitor work activities, and how quickly and when participants move from one activity to another can shed light on who is most likely to be at risk of sanction, how quickly noncompliance may be identified, and perhaps when noncompliance may be more likely to occur in the service delivery process.  Some state and local welfare offices have developed specific processes for conciliating acts of noncompliance in order to avoid sanctions.  The extent to which these efforts are successful may depend not only on the design of the process but also on the degree to which program staff use the process to encourage participation.

• What strategies do states/local sites use to re-engage noncompliant clients?  Once clients are sanctioned, many welfare offices have minimal contact with them and most of that contact occurs for purposes of managing ongoing monthly benefits and/or supportive services in partial-sanction states.  Other offices, however, have developed unique initiatives targeted specifically to sanctioned clients for purposes of re-engaging them in program activities.  In study sites that have developed such initiatives, we explore the extent to which local office staff set priorities for re-engaging sanctioned clients and preventing recurring sanctions, the nature of services and supports for sanctioned clients, who provides such services/supports, when and for how long, how clients access them, and how much they cost.  We also examine the extent to which sanctioned clients cure their sanctions or return to TANF.

• What future research could advance understanding of how sanctions increase engagement and participation rates?  At the completion of this study, questions about the role of sanctions in encouraging participation in work and work-related activities will undoubtedly remain.  We identify the questions and how they might be best addressed in future evaluations. 

To answer the study’s research questions, we selected eight counties in seven states and collected data from three sources: (1) in-depth case studies in all eight sites; (2) a telephone survey of frontline TANF workers in all eight sites; and (3) administrative data in three sites.2

1. Case Studies

The purpose of the case studies was to gather qualitative information from a variety of sources to create a comprehensive picture of the implementation of sanction policies and procedures.  We conducted in-person visits to one local welfare office in each county in fall 2006.  While some states had already solidified revisions to their sanction and related policies, many state and local programs were, at the time of our visits, still in the process of determining their approach to addressing the new requirements under the DRA, which took effect in October 2006.  Thus, the case studies may not reflect all the ways in which the selected sites have responded to the DRA.

For most sites, a two-person team conducted the in-person visits, which lasted two to three days each.  The visits involved one-on-one interviews and small group meetings.  Using semi-structured guided discussion techniques, we collected data from local welfare office administrators, case managers, eligibility workers, and employment service providers in each site.  Where applicable, we gathered data from specialized staff such as those responsible for conciliating or imposing sanctions and those responsible for conducting home visits or other outreach to noncompliant clients.  In each site, we asked case managers to identify a small number of clients who had most recently been sanctioned and we reviewed the case files of those clients.  The case file reviews enriched the interview data by providing concrete examples of the concepts and issues raised by interviewees.  Finally, we obtained written reports and copies of sanction notices and other relevant material in each site and conducted telephone interviews with state TANF administrators.

2. Telephone Survey of Frontline Workers

Concurrent with the case studies, we conducted a telephone survey of frontline staff in each local office to which we made a site visit in order to interview those involved in the sanction process.  For purposes of the telephone survey, “front-line staff” was defined as case managers, eligibility workers, and staff with specialized roles in the sanction process (such as staff responsible for conciliating or imposing sanctions and those responsible for conducting home visits or other outreach to noncompliant clients).  The primary goal of the survey was to shed additional light on (1) how staff perceive the role of sanctions; (2) the extent to which staff understand sanction policies and procedures; (3) the extent to which staff use discretion in the sanction process; and (4) the way in which staff use sanctions to encourage program participation.  The 25-minute telephone interview asked respondents about their role in:

• Deciding who is required to meet TANF work requirements and what an individual must do to meet the requirements

• Informing clients about the work requirements and consequences for nonparticipation

• Monitoring clients’ participation in program activities

• Deciding whether a client should be sanctioned for nonparticipation in program activities

• Carrying out procedures to impose a sanction

• Contacting clients who are in sanction status for purposes of re-engaging them in program activities 

In each site, we attempted to interview either the complete universe of staff or, in larger offices with substantially more staff, a random subset.  In total, we completed 161 interviews—ranging from 9 in one site to 30 in another—from a sample of 176.  Table I.2 presents the sample size and total number of completed interviews by site.  Throughout the report, we cite selected findings from the survey.  The survey instrument, or questionnaire, may be found in Appendix B.  A full compendium of tables presenting data from the survey may be found in Appendix C along with a more complete description of the telephone survey methodology.

 

Table I.2. Sample Size and Response Rate for Telephone Survey of Frontline Workers

	Total Sample Size	Nonrespondentsa	Completed Interviews	Response Rate (%)
Suffolk County, NY	29	1	28	96.6
Pima County, AZ	9	0	9	100.0
Duval County, FL	16	1	15	93.8
DeKalb County, GA	29	3	26	89.7
Tarrant County, TX	12	1	11	91.7
Los Angeles County, CA	24	3	21	87.5
Kern County, CA	30	5	25	83.3
Salt Lake County, UT	27	1	26	96.3
Total	176	15	161	91.5

a Nonrespondents include those who refused to participate in the survey and those who were unavailable during the survey fielding period.

 

3. Administrative Data

To explore the relationship between sanction policy and procedural changes on the one hand and engagement in work and work-related activities on the other, we obtained and analyzed administrative data from management information systems or detailed management reports in two of the states—Texas and Georgia—and one of the counties—Los Angeles—included in the study.  We focused on these sites because they had implemented major sanction-related changes, at least in part, to increase their participation rates.  Moreover, these sites introduced the changes recently enough that the changes reflected the current policy environment, but not so recently that ample data were not available to track outcomes of the changes over an extended period.  We examine participation rates and related outcomes (for instance, employment rates) before and after the changes went into effect (1) to explore whether there is any correlation between policy and programmatic changes and changes in participation rates and (2) to consider the ways that policy and programmatic changes may influence the rates.  Without data on what work participation rates and related outcomes would have been in the absence of sanction policy and procedural changes, we cannot draw rigorous conclusions about the effect of the changes on these outcomes.  This is because we cannot distinguish the effects of changes in sanction policies from the effects of other factors, such as other changes in TANF policy or practice, economic influences on the behavior of low-income families, or changes in policy or practice in other programs serving low-income families.

Our objective in analyzing data in Texas was to learn about how the number of recipients participating in work activities and the employment rate changed with the implementation of major statewide sanction policy changes.  We analyzed TANF and Unemployment Insurance data for one cohort of cases that were on TANF one year before the policy changes and one cohort of cases on TANF in the month before the changes.  The state shifted from a partial to an immediate full-family sanction in September 2003.  The first cohort consisted of the 131,556 cases on TANF in August 2002 and the second cohort of the 138,916 cases on TANF in August 2003.  We tracked the TANF, work participation, sanction, and employment status of each cohort for one year to determine the extent to which outcomes differ across clients subject to different sanction policies.

Our objective in analyzing data in Georgia was to learn about how changes in applicant job search procedures may have affected who entered the TANF caseload and thus became subject to sanctions and how the new procedures may have affected the work participation rate.  In 2004, the state implemented a new employment-focused initiative that prompted many counties to change the way in which they were implementing their TANF applicant job search processes.  In January 2005, the state implemented changes to its gradual full-family sanction policy.  We analyzed data on the TANF caseload, TANF applications, TANF case closures, and sanctions for each month from January 2001 through December 2006.  For each month, we obtained the total number of applications received statewide, the percentage of applications approved and denied, and the reasons for denial.  We also obtained the number of sanctions imposed, the number of TANF cases closed, and the reasons for case closure.  Using these data, we tracked trends over time—particularly before and after implementation of the new initiative and new sanction policies—and we attempt to identify correlations between (1) recent caseload declines and work participation rate spikes and (2) changes in application, case closure, and sanction rates.

Finally, our objective in Los Angeles County was to assess the outcomes of the county’s home visiting process, which was designed to avoid the imposition of sanctions.  We obtained program management reports on the 41,233 TANF recipients who were noncompliant and subject to the county’s home visit process in the first year of its existence.  We examined the percentage of cases successfully resolved (and resolved with and without a home visit), sanctioned, or unresolved.  We also looked at trends in the sanction rate over time.

C. Review of Literature on the Role of Sanctions in Welfare Reform

Although consensus holds that sanctions have been one of the most important policy changes implemented through state welfare reform efforts, they are among the least studied.  Most studies to date have focused on the incidence and duration of sanctions and the characteristics and circumstances of sanctioned families.  We refer to this literature where relevant throughout the body of the report.  Fewer studies have focused on those aspects that are most relevant to the current study—the relationship between sanctions and participation in work and related activities, and the implementation of sanction policies in practice.  In this section, we briefly summarize the existing literature in these two areas and describe how the current study adds to the literature.  It is important to note that all of the previous studies rely on the use of simple descriptive statistics and/or non-experimental methods (primarily regression analyses) to explore the relationship between sanctions and outcomes of interest; thus, none of the studies provide rigorous evidence of the impact of sanctions or causal relationships between sanctions and outcomes.

1. The Relationship Between Sanctions and Participation in Work-Related Activities

Sanctions are intended to change TANF recipients’ behavior by encouraging those who would not otherwise participate in work activities to do so; yet, research on the success of sanctions is scant.  A few studies from 1999 and 2000 that address the impact of sanctions suggest that more stringent sanction policies may lead to greater welfare exits and TANF caseload declines, though they offer little insight into how these changes occur (Hofferth, Stanhope, and Harris 2000; Rector and Youssef 1999; Mead 2000).  An earlier study suggests that high rates of caseload declines may result from differences in office performance (Mead 1997).  Researchers found that well-performing welfare offices make program expectations clear and threaten sanctions for nonparticipation but rarely need to impose sanctions.  Conversely, welfare offices that do a poor job of clearly stating recipient expectations and perform poorly in job placement and other performance measures frequently sanction recipients.  Large caseload declines may thus occur because agencies either do well in pressing recipients to find work and leave the caseload, or do poorly in communicating expectations such that many recipients leave the caseload without work.

More recent studies have attempted to discern whether sanctions motivate behavioral change.  For example, Lee et al. (2004) distinguish sanctions imposed from sanctions that were initiated but lifted before benefits were reduced.  They label the latter a “threat to sanction” and hypothesize that if sanctions induce behavioral changes, a threat to sanction would predict employment and welfare behavior as well as, or better than, imposed sanctions that result in actual grant reduction.  If, on the other hand, sanctions are primarily a punitive tool, we would expect greater effects on behavior from imposed sanctions.  Using longitudinal survey and administrative data to perform multivariate analyses, Lee et al. found (1) that there is no evidence that sanctions are correlated with less dependency on welfare and, in fact, are correlated with less work and lower earnings, and (2) that threats to sanction are not systematically related to formal work and welfare outcomes, although they are positively related to informal work and job training.  The authors conclude that sanctions do not appear to promote work or reduce welfare dependency and that any induced behavioral change occurs in the context of information about work and job preparation.  They caution, however, that causal relationships between welfare grant reductions and observed outcomes cannot be ascertained because the characteristics of sanctioned and non-sanctioned recipients are likely to differ in other unmeasured ways.

Hasenfeld et al. (2004) postulate that sanctions intended to change behavior assume (1) that welfare recipients can comply with work requirements but do not do so because they are unmotivated or not truly needy, and (2) that recipients understand and can calculate the costs and benefits of compliance.  Their study, however, calls both assumptions into question.  Findings from the study run counter to the first assumption in that sanctioned recipients in the study faced a greater number of barriers to the work requirements than non-sanctioned recipients despite no differences between groups in whether they looked for work in the past 30 days (a proxy measure for motivation to work).  Annual household income from all sources was lower among sanctioned families than non-sanctioned families, indicating the former group’s relatively greater financial need.  These findings suggest that many sanctioned recipients fail to comply with program requirements not because they are resistant to them but rather because they face barriers that make it difficult for them to meet such requirements.  The researchers find the second assumption equally questionable because a significant proportion of recipients, whether sanctioned or not, reported that they were poorly informed about sanction policies.  There are seven reasons why a TANF recipient may be sanctioned in California, but almost one-fifth of the sample only knew of three or fewer rules.  Almost two-thirds of the sample was unaware that they could be sanctioned for not participating in an assigned activity.  Other studies have also found that clients are unaware of program rules and the consequences of violating them (Nixon et al. 1999; U.S. Department of Health and Human Services 1999).

Other findings suggest that sanctions may have their desired effect.  Wu et al. (2004) find that many TANF recipients cure their sanctions and remain in sanction status for relatively short periods.  They interpret the results as suggesting that sanctions may change behavior toward program compliance.  However, they note that some sanctioned recipients moved off TANF altogether.  More research is needed on whether such cases leave the program because they are employed, or whether they and their families leave even though they face serious economic distress (the former suggesting that perhaps sanctions had their desired effect, and the latter suggesting that they did not).

2. The Implementation of Sanctions

In an earlier literature review, MPR highlighted uncertainties about factors that cause variation in sanction rates across local welfare offices.  Several recent studies find substantial variation in the implementation of TANF sanctions not only between different agencies within the same locality but also between staff members within the same agency.  These studies point to a variety of factors that can affect implementation of sanctions and sanction rates:

• Personal Judgment and Decision-Making. Three recent studies illustrate the extent to which the staff of local welfare offices use personal judgment in applying policy.  In a study by Los Angeles County (2005), case managers reported basing their sanctioning decisions on their perceptions of participants’ motivation, attitude, needs, and likelihood of benefiting from the program.  In a study by Berkley Planning Associates (2004), case managers in some local welfare offices in Louisiana said that their usual practice was to avoid sanctioning clients if at all possible.  Case managers in other local offices throughout the state, although given the same amount of discretion, were more likely to use sanctions whenever the circumstances justified so doing, without any personal judgment.  And a study by the Wisconsin Department of Workforce Development (2004) found that case managers within and across agencies in the state made markedly different decisions with respect to sanctioning when presented with the same case scenarios.

• Variation in Personal Perceptions of the Purpose and Effectiveness of Sanctions.  Research indicates that caseworkers differ in their philosophy about ways to ensure program compliance among TANF recipients.  Caseworkers in the Los Angeles County (2005) study fell into two camps: (1) those who use sanctions as a way of removing from the caseload those who fail to comply; and (2) those who use the sanction process as a way of working with participants to encourage their involvement.  In the BPA (2004) study, 45 percent of program managers across Louisiana agreed or strongly agreed that sanctions are an effective way to help program participants become more self-sufficient while 25 percent disagreed or strongly disagreed. 

• Variation in Personal Background Characteristics Among Staff.  The Los Angeles County (2005) study explored the relationship between caseworkers’ backgrounds and their use of sanctions (self-reported).  Case managers with more years of employment experience had a 54 percent greater probability of curing more than 30 sanctions over the past year than those with less experience.  The researchers suggest that those with more work experience may more thoroughly understand sanction policy and the implications of sanctions.

• Variation in Office Structure and Performance Measurement Practices. Fording et al. (2005) find two noteworthy variations.  First, they find that TANF cases in Florida managed by for-profit providers are more likely to be sanctioned than cases managed by government or nonprofit agencies.  They also find that regions in the state where the rate of welfare exits to employment is low appear to sanction clients more heavily than regions that perform better. 

• Variation in the Local Welfare Office Environment. The Los Angeles County (2005) study suggests that various aspects of the work environment, such as caseload size and the privatization of case management services, may affect sanction practices and rates.  For instance, researchers found that large caseloads were significantly associated with higher sanction frequencies.  In contrast to the Fording et al. (2005) study in Florida, the Los Angeles study found that contract caseworkers (from for-profit providers) had a lower probability of sanctioning than county caseworkers, although the finding may be at least partially attributable to differences in the types of participants served by the two types of caseworkers. Other aspects of the work environment—for instance, lack of coordination between the TANF and employment and training agency management information systems, lack of communication between the TANF and employment and training agency workers, or staff time and resource constraints due to a growing number new programs, new services, and increased workloads—have created challenges to implementing sanctions correctly and contributed to an increased rate of sanctioning.

• Variation in Sanction Policies and Procedures.  The BPA (2004) study revealed varied staff responses to Louisiana’s policy shift from partial to full-family sanctions in August 2003.  Staff in six local offices in the state reported using sanctions more often after the switch, staff in four offices reported using sanctions at the same rate, staff in two offices reported using sanctions less often, and staff in the other offices in the study could not come to consensus on how their use of sanctions might have changed.  Hasenfeld et al. (2004) suggest that the way in which local welfare office staff communicate with TANF recipients also has implications for sanction rates.  They found that, in a county that invested resources in counseling sanctioned recipients to help them cure their sanctions, 64 percent of recipients were aware of the rules governing sanctions.  By contrast, in a county in which staff communicate with recipients primarily through formal notifications, only 41 percent of recipients were aware of how sanctions work.  The authors conclude that interactions between workers and recipients play a significant role in the degree to which recipients understand program requirements and consequences and can respond rationally to them.

• Variation in the Social and Political Environment. Fording et al. (2004) and Keiser et al. (2004) suggest that the implementation of welfare reform and sanctions in particular is not race-neutral.  Fording et al. (2005) find that clients are less likely to be sanctioned if they live in a county with a relatively large Hispanic population and more likely to be sanctioned if they live in a county with a relatively high wage rate.  Keiser et al. (2004) find that, in any given county, nonwhites are more likely to face sanctions than whites with similar demographic characteristics, work histories, family structures, and welfare experience.  Yet, in the aggregate, white TANF recipients are more likely to face sanctions than are nonwhites due to the fact that they are more likely to live in areas with higher sanction rates.

3. Contributions of the Current Study of Sanction Policies

Despite the literature just cited, there is still little research on how states and localities have used sanctions to increase work participation rates and the extent to which they have succeeded.  In the absence of an experiment in which families are randomly assigned to groups subject to different sanction policies, it would be difficult to determine rigorously how sanctions affect work participation or related outcomes, and the extent to which sanctions lead to behavioral change among welfare recipients.  The current study, however, explores the mechanisms by which sanction policies and procedures can help achieve desirable outcomes and presents opportunities for exploring the effect of sanction-related initiatives in future evaluations.  It describes how policymakers and program administrators in selected states have either changed sanction policies or practices or improved implementation of existing policies and procedures specifically to increase participation rates.  It also presents the challenges associated with such initiatives, the costs of the initiatives, and how sanction rates, work participation rates, and related outcomes differ before and after the implementation of the initiatives in some states.  This information will be valuable to other states or counties, most of which are likely to be considering a range of options to increase their work participation rates in the wake of the DRA.  This report should not be used to deduce the impact of various sanction policies and procedures; ideally states that choose to adopt sanction-related changes as a result of information from this report would study the impact of those changes using random assignment methodologies.

---

1 Although the shift from a partial to a full sanction policy could result in a substantial decline in the TANF caseload, any decline that results from the change would not count toward a state's caseload reduction credit because it is the result of an eligibility change that a state must net out of its credit. (back)

2 Chapter II identifies and describes the study sites and the site selection process. (back)

 

Table of Contents | Previous | Next