Description and Assessment of State Approaches to Diversion Programs and Activities under Welfare Reform


Introduction - Significance of Diversion Programs

The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (P.L. 104-193, PRWORA) ended the individual entitlement to welfare benefits under Aid to Families with Dependent Children (AFDC), established by the Social Security Act of 1935. Under the new law, states are provided with a block grant to establish a new program, Temporary Assistance for Needy Families (TANF). Although states have considerable flexibility to design their TANF programs, PRWORA requires states to orient their TANF programs toward employment and to provide assistance to families for a maximum of sixty months. In addition, within two years of the enactment of PRWORA, states must require that TANF recipients participate in work and work-related activities to be eligible for benefits.
The new law also "de-links" Medicaid and welfare benefits. Eligibility for Medicaid is no longer automatic for TANF recipients and individuals can be eligible for Medicaid without having to be eligible for TANF. Although states are required to establish separate eligibility criteria for Medicaid using pre-TANF criteria, there are concerns that the complexities of TANF implementation, especially the shift to a time-limited, work-based assistance system, could result in the failure of many otherwise-eligible children and families to apply, or be found eligible, for Medicaid. Recent dramatic declines in Medicaid coverage serve to strengthen these concerns.
Even before the passage of PRWORA, many states were experimenting with ways to increase participation in work by reforming their welfare programs. One way in which states sought to encourage work was to divert individuals who were either job-ready or had other sources of income from becoming welfare recipients by offering them a one-time financial payment and/or job placement assistance as an alternative to enrollment in welfare. While these and other efforts to divert families from the welfare system appear to be an increasingly common component of states' welfare to work efforts under PRWORA, these programs are not well-understood both in terms of their actual operation or potential effects on Medicaid eligibility.

Research Goals and Methods

In an effort to better understand state efforts to divert families from welfare and the potential interactions with Medicaid eligibility, this research, funded by the Administration for Children and Families, and the Office of the Assistant Secretary for Planning and Evaluation, Department of Health and Human Services, has three broad goals: 1) to describe the range of state diversion programs and how these programs are being implemented, 2) to determine whether and how states are ensuring that families diverted from cash assistance will nonetheless apply for Medicaid and 3) to determine how diversion-related changes might affect families, particularly with respect to their eligibility for Medicaid, and the community-based entities that serve them.

This report presents the findings from the first phase of this research that involved the collection and analysis of descriptive data about diversion programs and activities in all 50 states and the District of Columbia. Data collection was accomplished primarily through a review of state documents and structured conversations with state officials.

Conceptual Approach to Examining Diversion - Three Types of Programs

Diversion programs are broadly understood as formal efforts to address the immediate needs of families seeking cash assistance in ways that avoid enrolling these families in TANF. The diversion programs implemented by states can be characterized as three distinct types.
Lump sum payment programs are designed to keep families with a short-term financial need from ever entering the welfare system with the provision of one time lump sum payment. Mandatory applicant job search programs require job searches as a condition of TANF eligibility and are designed to encourage recipients to find employment quickly to eliminate their need for cash assistance. The exploration of alternative resources is intended to discourage families from applying for cash assistance if other sources of support in their communities and from their families are available to them.
Diversion from cash assistance takes place both formally--as described above--and informally--as the result of program rules and requirements that discourage families who might otherwise be eligible for cash assistance from applying. Although the primary focus of this research is on formal diversion, wherever possible informal diversion activities are discussed as these harder-to-document activities may have just as great an impact as formal diversion.

Prevalence of Diversion Programs Nationwide

Thirty-one states have implemented at least one diversion program. As Table I-1 illustrates, 20 states are operating lump sum payment programs with three additional states planning to implement such programs in 1998. Sixteen states require TANF applicants to engage in active job searches before their application for assistance is approved. Seven states are using an aggressive approach to help TANF applicants identify alternative resources.
As Table I-2 illustrates, most states operating formal diversion programs have implemented only one program. Twelve states have implemented just lump sum payment programs and 10 states have implemented just mandatory applicant job searches. Only three states have implemented all three formal diversion programs.
Diversion programs are not equally distributed across the country. As illustrated by Table I-3, the Northeast has the fewest with 10 percent of the states having diversion programs. In the West and South, 60 percent of the states have lump sum payment programs and 40 percent have mandatory applicant job search. In the Midwest, 50 percent of the states have lump sum payment programs and 50 percent have mandatory applicant job search.

Lump Sum Payment

Under lump sum payment programs in effect in 20 states, caseworkers screen TANF applicants to determine if a lump sum payment can address effectively the immediate reason for the TANF application. Caseworkers must either find potential recipients of a lump sum payment eligible for TANF, or have gathered enough information to presume TANF eligibility. In almost one-half of the states, eligible applicants must have work-related needs directly affecting their ability to obtain or maintain employment such as loss of transportation due to needed car repairs. In the remaining states, eligible applicants can have a range of short-term or emergency needs such as overdue rent or utilities bills and child care problems. Accepting the lump sum payment in lieu of TANF benefits is not mandatory.
Lump sum diversion may affect the recipient's Medicaid application. All 20 states report using a joint application for TANF and other benefits such as Medicaid and Food Stamps, but only eight states require that the TANF application be completed for lump sum diversion. In the other states, caseworkers only collect enough information to presume the likelihood of TANF eligibility - the TANF application may not be completed or may be shown as withdrawn. This raises the possibility that the Medicaid application may likewise remain incomplete.
As illustrated by Table II-1, 23 state lump sum payment programs (three additional states planning to implement their program during 1998 are included) are analyzed on a range of characteristics. These characteristics vary across the states and include: how often one can apply for lump sum payment, amount of payment, how payment can be used, how payment is made, and the potential costs associated with receipt of lump sum payment. Most programs are relatively new, having been developed since 1996, and are operated on a statewide basis.
For example, most states prefer to make cash payments to recipients, although vouchers and third party payments are the preferred method in a few states. Maximum amounts are generally multiples of the applicable monthly TANF benefit and range from a low of $606 in Florida to a high of $3052 in Minnesota. Most states stipulate a period of ineligibility for TANF benefits after receipt of a lump sum payment during which reapplication triggers penalties such as a requirement to repay the lump sum or a reduction in the lifetime limit of TANF benefits. A few states impose these penalties or "costs" automatically on the lump sum recipient.
States can be characterized as making it easy for TANF applicants to be diverted where the lump sum payment program policies use relatively broad eligibility criteria, allow lump sum payments to be used for more than just work-related needs, and do not impose stringent repayment requirements or other penalties on lump sum payment recipients. States can also be seen as deliberately limiting the number of participants in their lump sum payment programs when program policies use very specific eligibility criteria, limit the use of lump sum payments to work-related needs, and impose onerous repayment requirements and automatic penalties.
The components of lump sum payment programs may indicate that the states view such programs primarily as supports for obtaining or maintaining employment, or suggest that the states view their program as means to expand emergency assistance for short-terms needs of families without reducing these families' lifetime TANF limits. States can opt for more or less oversight on how recipients use the lump sum payments or on how programs are administered at the county level.

Alternative Resources

Seven states are making a concerted effort to explore alternative resources with applicants before proceeding with an application for TANF. These efforts are seen as a natural outgrowth of the new concept of welfare benefits as a temporary, last resort option rather than the only option for families in need. Rather than being guided by specific policies, as in other types of diversion programs, this type of diversion is implemented through changes in the interaction between caseworkers and applicants. With the exception of Wisconsin, states generally offer applicants the option of pursuing alternative resources rather than requiring them to do so before applying for TANF. Applicants for whom these efforts are made are likely to be those with relatively minor, short-term needs.
Instead of merely documenting need, welfare caseworkers attempting to link TANF applicants with alternative resources must probe deeper to uncover the underlying circumstances. The successful implementation of efforts to link TANF applicants with alternative resources requires more sophisticated interviewing skills and a broader understanding of available community resources than has traditionally been required of caseworkers.

Mandatory Applicant Job Search

As illustrated by Table IV-1, 16 states require TANF applicants to look for work as a condition of eligibility for benefits. An additional 13 states require TANF applicants to participate in other employment-related activities such as a work-related orientation or registering with the job service for employment as a condition of eligibility. These requirements reinforce the new emphasis on employment instead of the provision of cash assistance. In this study, only mandatory applicant job search is considered a formal diversion program because the immediate and intensive search for a job is designed to eliminate families' need for assistance. Although states had experimented with work-related requirements prior to PRWORA, the emphasis of mandatory applicant job search on finding a job as quickly as possible is unique.
As illustrated by Tables IV-2 and IV-3, mandatory applicant job search programs are analyzed on a range of characteristics. These characteristics include: the target populations, the exempt populations, good cause exception for failure to job search, specific job search requirements, the amount of assistance for the job search, and the level of documentation required.
For example, most states with mandatory applicant job search requirements require all adult applicants to meet this requirement. Only one state does not require one-parent families and only two states do not require caretakers or guardians to look for work as a condition of eligibility. Most states allow exemptions from the work requirements, although there is considerable variation in how broadly or narrowly exemptions are defined. Common exemptions include families with children under age one and those who are disabled or not Ajob ready.@ Disability is defined various ways by states as is job readiness, which is often determined by caseworkers as part of their evaluation rather than based on objective criteria. In 10 of the 16 states, caseworkers have the discretion to make exceptions in addition to the formal exemptions. This discretion allows them to take into account individual circumstances that may not be explicitly addressed through formal policy.
Job search requirements differ in structure and in the expected level of effort. Most requirements are defined as a time period during which job search must take place, a specified number of employer contacts, or a combination of the two. Requirements range from two contacts in 40 to 45 days in Alabama to as many as 10 per week for four weeks in Missouri, Indiana, and Nevada. In most states, the time period coincides with the time allowed to process the TANF application. Nearly half the states have the same requirements statewide, while the rest give local offices discretion to set requirements within a broad range. Ten states also provide Agood cause@ exceptions for applicants who are required to look for work but are unable to comply with job search requirements by unforeseen circumstances.
Eleven states provide applicants with assistance to fulfill job search requirements. Types of assistance include job contacts and leads, access to a resource room where applicants can prepare resumes and conduct a job search, classes on job search skills such as resume writing and job interviewing, and auxiliary services such as transportation and child care. Eight states report that they require specific documentation to verify that applicants are meeting job search requirements.
Mandatory applicant job search programs are characterized by considerable devolution of decision making to local offices and by substantial worker discretion. These programs will probably become more prevalent as states' compliance with PRWORA requirements proceeds and are likely to divert more potential TANF recipients both formally and informally. In well-designed programs providing the assistance that applicants need to look for work, applicants will be diverted formally because they find employment. On the other hand, if programs provide minimal support and require relatively stringent job search activities that are unrealistic in terms of the abilities of most potential TANF applicants, mandatory applicant job search programs may result in substantial, and possibly undesirable, informal diversion.

Potential Impact of Diversion on Medicaid Eligibility

PRWORA not only provided states with considerable flexibility in designing their welfare programs but also "delinked eligibility for Medicaid from cash assistance." A family must be found eligible for Medicaid independent of their eligibility for cash assistance. PWRORA created a new Medicaid eligibility group of low-income families with children by adding Section 1931 to the Social Security Act. Section 1931 generally provides that all families who meet the basic requirements of a state's AFDC program in effect on July 16, 1996 are to be considered eligible for Medicaid.
Section 1931 gives states three options to modify Medicaid eligibility criteria. First, states can increase income and resource standards by the inflation index or lower these standards to May 1988 levels. Second, states can continue 1115 AFDC waiver provisions that modify income and resource standards. Third, state can establish less restrictive methodologies to liberalize income and resource standards. The third option is very broad and essentially gives states carte blanche to liberalize Medicaid eligibility criteria. Section 1931 also allows states to continue existing Title XIX "waivers" expanding transitional Medicaid assistance benefits.
Table V-1 summarizes state options to affect Medicaid eligibility post-PROWRA and the conditions for exercising these options. Additional options include Section 1115 Medicaid waivers and poverty-level-related expansions of coverage for pregnant woman and children. It is beyond the scope of this study to document states choices regarding Medicaid eligibility criteria. However, while most states report that eligibility for TANF means automatic eligibility for Medicaid, and all but two states report using a joint Medicaid/TANF application, the issue of Medicaid eligibility for nonTANF recipients raises important questions for diversion.
Diverted families are, by definition, not TANF recipients; their eligibility for Medicaid is determined by the state's Medicaid criteria. For lump sum payment recipients, this payment is counted as income in the month of receipt and will likely render recipients ineligible for Medicaid since most states' income standards are relatively low. To prevent this loss of benefits, a state can opt to modify income standards to disregard the lump sum payment as income.
Mandatory applicant job search means that many families will get jobs quickly and never receive cash assistance. This result also means that these families will be Medicaid-eligible for one or two months at most and will immediately lose their Medicaid coverage due to earned income. These families will also lose the opportunity for transitional Medicaid assistance as they will not have received Medicaid for three months (this will also be true for lump sum recipients). States can ameliorate these results by opting to disregard the first three months of earned income.
Table V-2 summarizes the state options for assuring Medicaid eligibility for diverted families. These options primarily affect coverage for adults and adolescents who are not eligible for Medicaid coverage outside of Section 1931 (i.e., poverty-level pathways). If diversion programs are successful, then a growing number of adults who previously would have been covered under Medicaid will be uninsured. While states can easily opt to change their income and resources standards under Section 1931, such changes would have to apply to all families and could represent serious financial commitments for state budgets in terms of expanded Medicaid eligibility.
A few states report having addressed the issues involved in assuring Medicaid eligibility for diverted families. There continue to be uncertainties, however, about the actual availability and consequences of these strategies. Additional guidance from HCFA is needed on this subject.
The implementation of diversion programs could also affect access to Medicaid. Because the operation of these programs typically diverts families before a joint TANF/Medicaid application has been completed, there is potential for the Medicaid application to Afall through the cracks,@ (i.e., not be processed). Informally diverted families may not be aware of their ability to apply for Medicaid without applying for TANF assistance.

Nearly all of the states report that linking diverted TANF applicants with Medicaid is not perceived to be a major issue. State have probably not considered the substantive implications for Medicaid eligibility posed by diversion programs, and believe that using the joint application form suffices to address the delinking of Medicaid and cash assistance.
Assuring Medicaid eligibility for diverted families will involve addressing the following implementation and policy questions: 1) how well diversion programs are implemented in terms of Medicaid applications "falling through cracks; 2) how aggressively states are promoting the availability of Medicaid eligibility independent of cash assistance; 3) how well-informed state policymakers are about the relationship between TANF policy and Medicaid policy and about their 1931 options for easing Medicaid eligibility requirements; and 4) how willing states are to expand Medicaid coverage to include participants in diversion programs.

Implications of Findings for Further Research

The study findings reveal that the three recognized formal diversion programs are widespread and implemented in a variety of ways. States know little about the effects of their diversion programs, however, both because most programs are so new, and because data collection efforts are lacking or in early stage of development. Important questions remain about the effectiveness of these diversion programs as well as the implications of their operation for Medicaid eligibility.
For example, questions about lump sum payment programs include: 1) which lump sum payment programs are particularly effective for what types of TANF applicants, 2) how well can caseworkers manage the implementation of these relatively discretionary and multifaceted programs, and 3) what types of lump sum payment programs are best suited for the goals of formal diversion.
Questions about mandatory applicant job search include: 1) how do workers determine who is and is not required to look for work as a condition of eligibility, 2) what fraction of applicants find employment before their application for assistance is processed, 3) how do the non-cash benefits available to families differ depending on whether an applicant finds employment before or after his or her application for assistance is approved, 4) what is the relationship between the level of job search activity required and/or the amount of job search assistance provided and the proportion of applicants who find employment, and 5) what happens to applicants who get discouraged and do not complete the application process.
A common set of data elements about the diversion program and the state=s Medicaid program is needed to fully document and analyze the effects. However, most states do not collect common data. In the case of applicant job search, for example, diverted cases generally would be coded as incomplete or denied with no specific reference to participation in and the outcome of participation in an applicant job search program. Most states report that they would measure the success of the diversion programs in terms of the number or percentage of total applicants diverted and if or for how long they then stay off the TANF rolls, but little data of this type are available.

Phase two of this study will involve 1) Case studies in five states to collect information on how the diversion programs work in practice and their effects on families from the point of view of administrators, field staff, participating families, and safety net providers, as well as assess the information available on informal diversion, and 2) Follow-up telephone conversations with Medicaid policymakers and program administrators in the 31 states with lump sum payment and/or mandatory applicant job search programs to obtain more information on how states have addressed issues of Medicaid eligibility in designing and implementing these diversion programs.
This research, building on the findings of phase one, will address four major areas of inquiry: 1) describe and examine the actual implementation of diversion programs and identify potential consequences of diversion for low-income families, particularly with respect to entry into the job market and access to Medicaid; 2) describe state policies regarding Medicaid eligibility for applicants diverted through lump sum payments and mandatory applicant job search, 3) examine whether and how the potential changes in Medicaid enrollment rates associated with diversion efforts might affect traditional health care safety net providers, and 4) examine potential strategies for monitoring changes in Medicaid enrollment rates as well as the effects of these changes over time.
Formal efforts to divert potential TANF recipients from receiving ongoing assistance represent one of many approaches states have implemented to shift to a more work-oriented, transitional system. These programs are clearly in their infancy but have the potential to affect large numbers of families. This research is creating a knowledge base that can provide a foundation for additional evaluation and monitoring of these program and activities.