Implementing Welfare Reform Requirements for Teenage Parents: Lessons from Experience in Four States . Data Sources


This study relies on both qualitative and quantitative data. MPR and DHHS staff members conducted site visits to each of the four study states. In addition, each state provided data covering school attendance sanctions and other information for a sample of teenage parents.

Site Visits

Site visits to each state consisted of approximately one day of interviews with state-level welfare officials and an additional day or two in at least one local welfare agency. At the state level, we spoke with staff members knowledgeable about the development, implementation, and evolution of the state's policies toward teenage parents. At the local level, we spoke with agency directors, supervisors, and eligibility and intake workers about their specific procedures for (1) identifying teenage parents, (2) monitoring and enforcing their compliance with attendance and living arrangement requirements, and (3) providing teenage parents with support services.

In some instances, we conducted additional interviews with people who work closely with teenage parents receiving cash assistance but who function outside the welfare department. For example, in Arizona and Massachusetts, we spoke with the directors of GED programs serving large numbers of teenage parents about, among other topics: (1) barriers to school attendance facing teenage parents, (2) the involvement of theirprograms with attendance-monitoring procedures, and (3) their interaction with the welfare department. In Massachusetts, we spoke with a director of a teenage parent group home about the day-to-day operation of this type of facility and the advantages of this living arrangement for teenage parents and their children. In California, we spoke with Cal Learn case management staff about their program's policies and operations.


  Arizona California Massachusetts Virginia United States
1995 Population (in Millions) 4.2 31.6 6.1 6.6 262.8
1995 Per Capita Income (in Dollars) 18,987 22,035 25,099 21,940 21,188
Percentage of the Population Below the Poverty Level in 1994 15.9 17.9 9.7 10.7 14.5
Percentage of the Population Receiving AFDC or SSI in June 1994 6.5 11.7 7.5 4.8 7.7
Maximum AFDC Grant for Family of Three in January 1996 (in Dollars) 347 607 565 354 389a
Births per 1,000 Unmarried Women in 1990:          
Ages 15 to 17 41.2 34.4 22.2 26.2 29.6
Ages 18 to 19 81.5 71.4 38.2 53.2 60.7

Source: Birthrates from "Births to Unmarried Mothers: United States, 1980-92." U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Center for Health Statistics, June 1995. All other figures from "Statistical Abstract of the United States, 1996." U.S. Department of Commerce, Economics and Statistics Administration, Bureau of the Census, 1996.

a Maximum AFDC grant for the median state.

AFDC = Aid to Families with Dependent Children.

SSI = Supplemental Security Income.

Quantitative Data

In each of the four study states, we collected and analyzed quantitative data to supplement the qualitative interview data.(5) Each state provided us with longitudinal data for a sample of teenage parents who received cash assistance in the early months of 1996. The longitudinal data cover sanction data (and, in California, bonus data) associated with the school attendance requirement for a 12- to 15-month period.

Each state provided additional, descriptive, statewide data covering such items as the number of teenage parents receiving cash assistance and changes in this number over time, as well as the number of teenage parents participating in certain programs and receiving particular services. Virginia also provided statewide information on the number of case closings and application denials for violation of the living arrangement requirement during the first 18 months of the policy.


Implementing policies and programs that target teenage parents requires, first of all, the ability to identify them. The new federal TANF regulations require states to deny cash assistance to unmarried minor custodial parents who do not live in an adult-supervised setting and who do not attend school regularly if they do not have a diploma or its equivalent. Therefore, compliance with TANF may require some state welfare agencies, which previously had no operational need to identify minor parents, to develop the capability to identify this population. Accurate and efficient identification and referral of teenage parents receiving cash assistance is essential because it enables welfare agencies to enforce the relevant requirements specific to teenage parents.

Moreover, states may want to target supportive services to this high-risk population. Many teenage parents face substantial obstacles to future success: unsupportive home environments, social isolation, long-standing poor school performance, lack of role models from whom they can learn parenting and other life skills, and struggles over child care. If state welfare agencies want to provide meaningful assistance to help teenage parents meet the new requirements, they must be able to identify teenage parents and refer them to the necessary support services.

Identifying all eligible teenage parents can be a major challenge.

Welfare policies that apply specifically to teenage parents pose a special challenge, because many young parents do not head their own cash assistance case. These policies require that welfare agencies collect information on the parent status of all case members, not just the case head. Procedures and computer data systems often are not designed to collect and maintain detailed information about case members other than the case head. Therefore, successful implementation of policies focusing on teenage parents frequently requires modifications of procedures and supporting information systems.

Teenage parents on someone else's case can be particularly difficult to identify.

Research on programs for teenage parents has consistently found that identifying young parents who receive cash assistance on someone else's case is difficult (Hershey 1991; Bloom et al. 1991; and Bloom et al. 1993). For example, the evaluations of the Learning, Earning, and Parenting (LEAP) program in Ohio and the Teenage Parent Demonstration in Illinois and New Jersey found that welfare agencies in these states had considerable difficulty identifying teenage parents who did not head their own cash assistance case. In contrast, it was relatively easy for all three programs to identify teenage parents who headed their own cash assistance cases.(6) The states in our study reported similar difficulties identifying teenage parents receiving cash assistance on someone else's case.


A computer data system that tracks the parental status of all case members, not just the case head, greatly enhances a welfare agency's ability to identify teenage parents receiving cash assistance. For example, Arizona's automated data system for tracking cash assistance cases links teenage parents to their children on the grant, even if the teenager does not head the case. This capability makes it possible for the state welfare agency to identify, in an automated fashion, all teenage parents receiving cash assistance.

Identifying information may be inaccurate or incomplete.

Even if a state's data system is able to identify teenage parents, however, this information may be inaccurate or incomplete. Identifying information is particularly likely to be incomplete if local welfare agencies have no operational reason to identify this population. Virginia's computer data system illustrates this point. The state's central computer system for tracking cash assistance cases has for many years contained a field for flagging teenage parents. However, an analysis conducted by the state welfare agency as part of this study revealed that only about one in four teenage parents receiving cash assistance on someone else's case was actually flagged in the state's computer system. Because Virginia has no policies or programs that require identifying teenage parents, it appears that local welfare staff do not use the teenage parent field consistently.

Agencies may need to modify data systems to facilitate identification.

If a welfare agency's data system does not track the necessary information for identifying teenage parents, the agency may choose to modify its system. For example, when Massachusetts first implemented a school attendance requirement for teenage parents, the state welfare agency added to the data system a field for flagging teenage parents. The state welfare agency instructed eligibility workers to use the flag each time they encountered a teenage parent on one of their cases. However, even when states modify their data systems in this way, staff in local welfare offices may not use the new fields consistently. For example, welfare officials in Massachusetts reported that local staff often did not use the new fields for flagging teenage parent cases.


Identification of eligible teenage parents depends not only on data systems, but also on local welfare staff. Upon application for cash assistance, intake workers must ask about the relationships among all case members. In addition, when an eligibility worker adds a new baby to an existing case, that worker must determine whether the mother of the child is a teenager on the case who is now subject to one of the new teenage parent provisions. If the data system has the capability, the eligibility worker must enter the necessary relation and education information.

Staff must not only identify teenage parents, they must also refer the parents to the proper programs and services.

Properly identifying teenage parents is essential, because it provides the basis for referring those eligible to the appropriate programs and services. Even if the state's computer system identifies all (or almost all) teenage parents, eligibility workers may not always refer them properly. For example, in November 1995, Arizona began requiring 13- to 15-year-old custodial parents who were on cash assistance to participate in the state's JOBS program if they had dropped out of school.(7) The state's data system identifies teenage parents and links them to their children on the grant, even when the teenager is not the head of the case. Moreover, eligibility workers are trained to ask about relationships among all members of the case. In spite of these computer capabilities and procedures, statewide data provided by the state's welfare agency reveal that, during the first several months under the new policy, most parents under 16 who had dropped out of school had not been referred to JOBS. In early 1997, state welfare staff reported that eligibility workers continued to have difficulty remembering to refer these cases to JOBS.

The small number of eligible teenage parents may make it difficult to remember to refer them.

One important reason why local welfare staff may not refer eligible teenage parents to the proper program may be that eligible teenage parents represent such a small fraction of the caseload. For example, in Arizona, only 0.2 percent of cash assistance cases contain a teenage parent under age 16 (the group that may require a special referral to JOBS under the state's waiver). It may be difficult for eligibility workers to remember special rules and procedures that affect so few cases and that they so rarely need to use.


The experiences of the study states suggest some useful strategies for improving the process for identifying and referring eligible teenage parents. These strategies are summarized next.

Data systems that require welfare staff to record the mother of all minors on the case greatly facilitate identification.

A data system that includes as a required field a "pointer" that identifies the mother of each minor on a cash assistance case greatly improves the ease with which welfare agencies can identify teenage parents. This type of system links all minor parents on the case to their children, which allows welfare agencies to identify young parents in an automated fashion. Moreover, requiring local welfare staff to record the mother of a minor in order to move on to the next field in the data system will ensure that staff record this information completely. If, instead, the system uses a teenage parent flag that is not a required field, it will be unclear whether a blank field indicates that the minor is not a parent or that welfare staff simply forgot to flag the teenage parent in the system.

Training and regularly reminding staff about new procedures is crucial for complete identification.

Welfare agencies implementing new teenage parent provisions need to train local welfare staff to follow the new procedures for identification and referral. Agencies also should send regular reminders to local staff to use the new procedures during the first months under the policy. In addition, state welfare agencies may need to conduct periodic audits to determine the frequency with which local staff are referring eligible teenage parents and using the new data fields, as well as the level of accuracy of the recorded information.

Positive incentives to refer teenage parents may facilitate the referral process.

Establishing positive incentives for eligibility workers to identify and refer teenage parents may facilitate the referral process. For example, procedures for handling teenage parent cases in Massachusetts enable regular eligibility workers to reduce their workload by referring teenage parents to the appropriate services.

Teenage parent cases are handled by specialized eligibility workers known as "teen specialists" who, in most cases, work exclusively with cash assistance cases containing a teenage parent. Local staff in Massachusetts reported that regulareligibility workers (those not assigned to teenage parents) work diligently to find and refer teenage parents, because these potentially complicated cases are moved from their caseloads to those of the teen specialists. Staff believe that the desire of regular eligibility workers to make their workloads more manageable will soon cause almost all teenage parents to be identified and properly handled by local offices.

Referrals from other agencies can improve identification.

Given the potential challenges to identifying and referring all those subject to the new teenage parent provisions, it may be a useful strategy for welfare agencies to rely on referrals from other agencies that serve teenage parents. For example, Cal Learn staff considered referrals from schools and health care providers to be an extremely useful and important method of improving their coverage of eligible teenage parents and the speed with which they are brought into the program. Program staff from both California counties visited for the study reported that they were actively working to strengthen ties with schools and health care providers. These closer ties will allow Cal Learn staff to rely on these agencies for referrals of eligible teenage parents, so that the program can more quickly begin serving teenagers when they first become eligible for Cal Learn.

Local welfare staff in Massachusetts reported that the Division of Medical Assistance (which administers the state's Medicaid program) notifies the local welfare office of all births to families receiving cash assistance, so that the new baby is added to Medicaid. Staff described this process as very useful for identifying teenagers on existing cases who become parents.


The new federal TANF regulations may require some states to develop the capacity to identify minor parents. A state that imposed the most narrow mandatory attendance policy allowed under TANF, one that applied only to unmarried minor custodial parents, would need the capability to identify minor parents (including those on someone else's case) from among its cash assistance caseload.

Attendance policies covering all school-age minors allow TANF compliance without identifying minor parents.

A substantial number of states, however, have chosen attendance policies for young cash assistance recipients that are consistent with this TANF requirement and that do not require identification of minor parents on someone else's case. For example, Virginia's state welfare reform initiative, implemented in 1995, meets the TANF requirement by mandating that all school-age minors (not just minor parents) attend school. Those who do not attend regularly can have their needs removed from the cash grant. By requiring all school-ageminors (including those who are parents) to attend school regularly as a condition for cash assistance receipt, the state can deny cash assistance to minor parents who do not attend school (as required under TANF) without being able to distinguish minor parents from other school-age minors on their caseload.

The new federal TANF regulations also require that states deny cash assistance to minor parents who do not live with an adult guardian unless certain limited exceptions apply. However, states do not need to identify minor parents on someone else's case to enforce this requirement, since, presumably, these minors are already living with an adult guardian (the case head). Therefore, states such as Virginia, which impose attendance requirements covering all school-age minors, will not need to be able to identify minor parents on someone else's case to comply with TANF.

However, a state that cannot identify teenage parents cannot target services to this population.

Teenage parents are very likely to become long-term welfare recipients. For this reason, welfare agencies may want to target limited resources for special programs and services to this high-risk population. If the agency does not have the capability to identify teenage parents, however, it will be unable to target this population for special services.


1. Title I, Section 101, of PL 104-193 describes the reasons that Congress enacted sweeping changes to the welfare system.

2. See PL 104-193, Section 103, Part A, Section 408 a (4).

3. See PL 104-193, Section 103, Part A, Section 408 a (5).

4. For a more detailed discussion of the selection of states for the study, see Burghardt and Wood (1996).

5. Analyses of these quantitative data are presented primarily in Volume II.

6. The identification of teenage parents who head their own cases is typically done by simply searching the data system for case heads who are under age 20.

7. Both before and after this policy change in Arizona, 16- to 19-year-old dropouts on cash assistance were required to participate in JOBS, regardless of their parent status.