1. The FSP Caseload 2

B. The Medicaid Caseload 4

IV Factors Associated with the Decline in Participation 5

A. Changes to the FSP and Medicaid under PRWORA 6

1. Overview of the FSP and Recent Policy Changes 6

2. Overview of Medicaid and Recent Policy Changes 6

B. Changes in the Macroeconomy 8

1. The Economy and FSP Participation 8

2. The Economy and the Welfare Caseload 8

3. The Economy and Medicaid Enrollment 9

4. Summary of the Economy=s Effects on Caseloads 9

C. Changes in the Pool of Eligible Families 10

1. Participation among TANF Recipients 10

2. Participation among Families Who Have Left AFDC/TANF 11

D. The Design and Implementation of TANF Programs 14

1. TANF Diversion Policies 14

2. Work-First Programs 16

3. TANF Sanctions 16

4. TANF Time Limits 17

E. The Design and Implementation of the FSP and Medicaid in a

Work-Oriented Welfare Environment 19

 

CONTENTS (continued)

 

Section Page

 

1. Confusion about Potential Eligibility among Families and

Workers 19

2. Participation Costs 21

3. Use of Automated Eligibility Systems 22

4. Federal Guidance on the FSP and Medicaid under Welfare

Reform 23

5. Stigma Associated with Food Stamps and Medicaid 23

V Strategies for Increasing Participation and Increasing Our

Understanding 24

A. Increasing Participation in the FSP among TANF Applicants and Leavers 24

2. Increasing Participation in Medicaid and SCHIP for TANF

Applicants and Leavers 25

CONTENTS (continued)

Policy Changes to Medicaid. Unlike changes to the FSP, recent changes to Medicaid were primarily designed to increase rather than decrease participation in the program. A new category was created for family coverage under Section 1931, and states were given considerable flexibility to modify their program rules in order to expand enrollment (Mann 1999). The 1996 legislation also changed the rules under Section 1925, making eligibility for Transitional Medicaid Assistance (TMA) dependent on prior receipt of Medicaid, instead of AFDC. Finally, the Balanced Budget Act of 1997 provided funds to states for a new program to expand health coverage to low-income uninsured children who do not qualify for a state=s Medicaid program. Under the new State Children=s Health Insurance Program (SCHIP), states may expand Medicaid, establish a new program separate from Medicaid, or develop a combination of both.

TANF Diversion Policies. In keeping with the Awork-first@ philosophy of most state TANF programs, diversion policies are becoming increasingly common. Diversion seeks to provide the potential TANF applicant with one-time assistance that is intended to deter that person from applying for benefits. Common diversion strategies include (1) lump-sum payment programs, (2) alternative resource programs that help families access resources in the community or from family or friends, and (3) mandatory job search programs, which require potential applicants to complete a job search as a condition for processing the TANF application (Maloy et al. 1998). There is some evidence to suggest that TANF diversion policies may be unintentionally contributing to lower rates of FSP and Medicaid participation by eligible families. According to FSP and Medicaid rules, program applications should be processed as soon as they are submitted, yet because they are often part of a joint TANF application, the eligibility determination for these programs may be delayed, or the application may not be processed at all (Maloy et al. 1998, GAO 1999a and 1999b). Because many families seek Medicaid and food stamp assistance through the welfare office, certain practices associated with diversion policies may inadvertently close off a significant means of access to the other two programs.

Stigma. Several studies ranging from focus groups with human service administrators to national surveys and focus groups with low-income families suggest that there may be a stigma attached to the receipt of food stamps and Medicaid, even more so now because of the public=s increasingly negative view of welfare (Smith et al. 1998, McConnell and Ponza 1999). Recent findings from a national survey of low-income parents indicate that 4 in 10 parents of Medicaid-eligible uninsured children (42 percent) would be much more likely to enroll in Medicaid if they did not have to go to a welfare office (Perry et al. 2000). This finding supports the idea that families may be more likely to enroll if the public could be persuaded to view Medicaid not as a welfare program but as a support for working families.

Costs of Participating. There is also evidence that potential participants weigh both the costs and benefits of participation, and when the scale tips toward the former, another barrier arises. Participation costs can include complicated application processes, intrusive income verification and reporting requirements, and the need to be absent from work in order to apply or be recertified. In fact, most participants are required to return to the food stamp office four times a year to be recertified for benefits; moreover, working families, because their income is more likely to change, are required to be recertified more often than those with a fixed income. These Ahassles@ and others were reported in studies of potential applicants to both the FSP and Medicaid (Ponza et al. 1999, McConnell and Ponza 1999, Perry et al. 2000). Most potential Medicaid applicants in a national sample of low-income families (72 percent) reported difficulty obtaining the required documents as the primary reason for not completing the application process, and 66 percent gave as their primary reason the overall hassle of the enrollment process (Perry et al. 2000). In the National Survey of America=s Families, 10 percent of former welfare families and 17 percent of nonwelfare families reported leaving the FSP because of administrative problems or hassles (Zedlewski and Brauner 1999). Participation costs may be perceived to be greater among families who are working because of the decreased availability of free time.

There is broad consensus among researchers and practitioners that the rules regarding Medicaid eligibility have become extremely difficult to understand and that adding Section 1931 to the array of eligibility categories (estimated to range from 30 to 100) has further complicated the already complex program rules (GAO 1999b, Ellwood 1999). Process studies reveal that welfare staff often lack the necessary training in Medicaid rules to explain them to clients and collect the appropriate information (Ellwood 1999, Smith et al. 1998). Although the Medicaid program is well known and held in high regard by low-income families, they do not consider it to be Auser friendly@ (Perry et al. 2000).

1. Introduction

Participation in two of the nation=s most important federal safety net programs, the Food Stamp Program (FSP) and Medicaid, has declined in recent years. Of particular concern is the possibility that many eligible low-income families with children may not be participating in these programs due to the changes that have occurred as a result of welfare reform. The purpose of this review is to summarize the literature relevant to the recent decline in the Medicaid and FSP caseloads, particularly within the context of the reforms introduced by the Personal Work and Responsibility Act of 1996 (PRWORA). We discuss the existing information on both barriers and enhancements to initial and continuous enrollment, focusing on the period from 1995 to 1999. Our review includes government and privately sponsored research projects, studies of participation in the FSP and Medicaid at the national and state level, studies of families who have left welfare, reviews of research, and ongoing analysis and data collection efforts.

We begin Section II below by discussing the policy context in which the changes in FSP and Medicaid participation have taken place. Section III presents evidence of the decline in participation in each program. Section IV covers the major factors that are considered to be associated with the declines, including the strong economic environment, changes in program eligibility rules, effects of the implementation of welfare reform, and effects of the operation of the FSP and Medicaid programs within the context of a work-oriented welfare environment. In Section V, we discuss strategies that have been proposed to increase participation in Medicaid and in the FSP, and we conclude this review in Section VI, drawing together our findings by comparing the issues related to, reasons for, and proposed solutions to the problem of diminished participation in the two programs. The appendixes to this document list the recent key publications in the literature on the decline in FSP and Medicaid participation, along with selected information on study methodologies.

 

2. Policy Context

Historically, eligibility for the FSP and Medicaid has been closely connected to the system for welfare cash assistance, and as a result, most welfare recipients also received Medicaid and food stamp benefits. In 1996, PRWORA made sweeping changes to the nation=s cash assistance program, the implementation of which many believe may be related to recent declines in the Medicaid and FSP caseloads. The welfare reform legislation eliminated the Aid for Dependent Children (AFDC) program and replaced it with a block grant to states to implement the Temporary Assistance for Needy Families (TANF) program. States were given considerable flexibility to design their TANF programs, but the federal legislation requires TANF recipients to work within 24 months of receiving assistance and limits lifetime eligibility for federally funded benefits to 60 months. One of the goals of the new legislation was to more quickly move people from welfare to work, and thus toward self-sufficiency. Since the implementation of this legislation, the welfare caseload has declined in nearly every state (General Accounting Office 1999a, see Table II.3).

PRWORA also included changes to the FSP and Medicaid, but with the exception of provisions for legal immigrants, most of these changes were not designed to restrict program eligibility for families with children. Indeed, policymakers took precautions to ensure that low-income parents and their children would not lose Medicaid coverage as a result of welfare reform. Nevertheless, participation in the FSP and Medicaid overall and by low-income families with children in particular has dropped substantially. There is widespread belief that the drop in participation is much greater than what should have been expected even when taking into account Medicaid and FSP policy changes and other factors such as the economy.

Decreased caseloads have spawned concern about the food sufficiency and health care coverage of low-income families. Some recent surveys have indicated that families who leave the FSP and welfare continue to have problems affording food (see, for example, Zedlewski and Brauner 1999). Moreover, recent federal data indicate that several states with high levels of food insecurity also have low levels of participation in the FSP (Nord et al. 1999). A report by GAO (1999a) indicates that requests for food assistance directed toward resources other than the FSP have increased as participation in the FSP has decreased. For example, the report cites an increase in the number of children participating in the USDA=s National School Lunch Program from 1994 to 1997. In addition, a survey by the U.S. Conference of Mayors indicates that requests for emergency food assistance increased by about 14 percent in 21 of 30 major cities from 1997 to 1998.

The picture for low-income families is similarly uncertain when it comes to health care. The number of low-income children and adults without health insurance is on the rise. The Urban Institute has estimated that 42 percent of uninsured children are eligible for Medicaid, with another 27 percent eligible for the State Children=s Health Insurance Program (SCHIP) (Lyons 1999). Although uninsured low-income individuals who visit emergency rooms or acute care facilities are often assisted in applying for Medicaid on the spot, the lack of health insurance results in less preventive and ambulatory care to begin with (Nathan and Thompson 1999), and in less access to prenatal care for pregnant women (Lyons 1999). Low-income mothers without Medicaid coverage have about a 40 percent higher infant mortality rate than those with coverage (Lyons 1999). Moreover, an analysis by Wooldridge and Hoag (1996) concluded that greater Medicaid participation can increase the likelihood that welfare recipients making the transition to work will remain off cash assistance.

 

3. Caseload Declines in the FSP and Medicaid

1. The FSP Caseload

Recent data from the USDA=s Food and Nutrition Service (FNS) document an ongoing decline in Food Stamp Program participation. The caseload dropped by 27 percent over 3 2 years, from fiscal year 1996 to the first half of fiscal year 1999 (GAO 1999a). Prior to this period, the caseload had climbed for several years, peaking in 1994 at 27.5 million participants. The caseload steadily declined thereafter to 25.5 million in 1996, 19.7 million in 1998, and finally to 18.3 million in February 1999, the lowest level since 1979 (Office of Analysis, Nutrition, and Evaluation, FNS, USDA 1999a).

The figures above indicate that most of the decrease in participation occurred between 1996 and 1998, coinciding with the implementation of welfare reform legislation and with a period of continuing strong growth in the U.S. economy. GAO (1999a) reported that the FSP caseload declined in each of the 50 states across this period (see Appendix A). The extent of the decline varies widely by state, from 4.9 percent to 48.1 percent. Table 1 groups states by magnitude of percent decline over this period, highlighting the 10 states with the smallest declines and the 10 states with the largest.

TABLE 1

PERCENT CHANGE IN FSP ENROLLMENT FOR TOP AND BOTTOM 10 STATES
	State	Percent Change in Food Stamp Caseload 8/96-8/98	Percent Change in TANF 1/96-12/98
Lowest Decline (Top 10 States)	Nebraska Hawaii Arkansas South Dakota District of Columbia South Carolina Alaska Montana North Dakota West Virginia	-4.9 -6.9 -7.8 -8.5 -9.3 -9.9 -10.3 -12.1 -12.2 -12.2	-9.9 -31.9 -48.3 -10.7 -23.7 -41.1 -28.1 -50.4 -30.6 -72.0
Largest Decline (Bottom 10 States)	Mississippi Washington Kansas California Ohio Texas Wisconsin Arizona Rhode Island Vermont	-30.2 -30.4 -30.6 -30.7 -30.7 -33.2 -33.6 -35.0 -40.2 -48.1	-67.3 -35.4 -54.2 -30.1 -47.3 -46.8 -81.6 -43.9 -52.1 -53.7

Source: GAO. "Food Stamp Program: Various Factors Have Led to Declining Participation.@ July 1999.

Preliminary FNS data for the percent caseload change from May 1998 to May 1999 indicate that a few states may have begun to reverse the negative trend in participation (USDA 1999). Whereas the FSP caseload in all 50 states had declined in the previous year, it increased slightly in Hawaii, Missouri, North Dakota, Rhode Island, and Kansas from May 1998 to May 1999.

 

2. The Medicaid Caseload

Medicaid enrollment for children and their parents began to decline in 1996 for the first time in almost a decade (Ellwood and Ku 1998) and has steadily declined each year since (Lyons 1999; Ku and Bruen 1999). Using edited state data for average monthly participation, Ku and Bruen (1999) report that Medicaid enrollment among nonelderly, nondisabled adults and children declined by 5.3 percent nationwide between 1995 and 1997. This drop reflects substantial variation in state Medicaid enrollment, ranging from a 21 percent decline in West Virginia to an increase of almost 30 percent in Oregon (see Appendix B). The overall 5.3 percent drop resulted because the decline in participation among welfare families (21.9 percent) was not completely offset by the increase in families not on welfare (18.2 percent). The 10 states in the top of the distribution (those with Medicaid caseload increases) and the 10 states that fall into the bottom of the distribution are shown in Table 2.

A drop in participation may occur because fewer people meet eligibility criteria or because fewer people who are eligible actually participate. Three broad factors have been considered as possible influences on the decline in participation in the FSP and Medicaid: changes to eligibility provisions in the FSP and Medicaid programs under PRWORA, the ongoing strong economy and low unemployment, and effects of the implementation of PRWORA, including changes in the pool of FSP- and Medicaid-eligible families, and the operation of these programs in the context of a work-oriented welfare environment. In this section, we explore what is known about the effect of these factors.

The FSP is the nation=s largest food assistance program, helping low-income individuals and families gain adequate and nutritious diets that they might not otherwise be able to afford. The FSP serves not only parents and children, but all individuals in households that meet income criteria, even unrelated individuals. Regardless of the employment status of household members, households without a disabled or elderly member, with an income below 130 percent of the federal poverty level, and that meet asset tests and other procedural requirements of the FSP are eligible for food stamps. Different eligibility criteria apply to households with disabled and elderly members.

2. Overview of Medicaid and Recent Policy Changes

Medicaid was also retained as a federal entitlement under PRWORA. In general, Medicaid serves four main groups of low-income individuals: the disabled, the elderly, parents, and children. Almost three-quarters of the 1997 caseload was composed of parents and children (The Kaiser Commission 1999). Prior to PRWORA, parents and children usually enrolled in Medicaid via welfare. PRWORA delinked Medicaid and welfare, meaning that welfare receipt is no longer an eligibility category for Medicaid and that Medicaid eligibility requirements are now separate from TANF rules.

Unlike the changes to the FSP, recent changes to Medicaid were primarily designed to increase rather than decrease participation in the program. The one exception is PRWORA=s provisions that restrict Medicaid eligibility for new legal immigrants. Immigrants legally admitted to the U.S. prior to PRWORA continue to be eligible for full Medicaid coverage (assuming they meet income and categorical standards), but immigrants entering the country after the legislation was passed are no longer eligible for full coverage (although they can still receive emergency services) (Ellwood and Ku 1998). However recent research has indicated that Medicaid participation among eligible immigrant families has declined (Fix and Passel 1999, GAO 1998, Ellwood 1999). In particular, GAO reported that one out of three uninsured children eligible for Medicaid but not enrolled lived in immigrant families in 1996. It appears that many immigrant families erroneously believe that as a result of PRWORA, they and their native-born children are automatically ineligible for Medicaid. Ellwood (1999) also notes that immigrant families may be reluctant to apply for Medicaid out of fear that enrollment would jeopardize their immigration status.

Below we briefly review provisions and programs that have recently been created or expanded to increase Medicaid enrollment and participation. We include a discussion of Section 1931, a new eligibility category created under PRWORA; Section 1925, which provides for transitional Medicaid benefits for families no longer meeting Medicaid eligibility criteria; and the State Children=s Health Insurance Program (SCHIP), a new program for low-income uninsured children.

Section 1931. The welfare reform legislation created a new category for family coverage under Section 1931. Under this new category, states are required, at a minimum, to set standards not more limiting than their AFDC income and resource standards in effect prior to August 1996 for Medicaid eligibility. In addition, states were given considerable flexibility to go beyond this requirement and modify their old AFDC/Medicaid rules in order to expand enrollment in Medicaid. For example, states may adopt less restrictive methods of counting income and assets for AFDC-type families, may raise their income and resource standards, and may cover two-parent families as well as single-parent families (Mann 1999). A report by the American Public Welfare Association and the Health Care Financing Administration compiles information on state decisions and options with respect to Medicaid as of January 1997 (Peller and Shaner 1999), yet the utility of this information is limited because it was collected so early after the enactment of PRWORA. More recent information on this topic is available from a study of 31 states (Darnell et al. 1999), which indicates that many states have not taken advantage of the policy options available to them under Section 1931.

Section 1925. Prior to PRWORA, welfare recipients who found work and left welfare were eligible for Transitional Medicaid Assistance (TMA), which extends Medicaid benefits for six months to a year. Although PRWORA delinked TANF and Medicaid, TANF recipients can still receive Medicaid benefits under TMA after they lose Section 1931 eligibility. Section 1925 of the Social Security Act entitles families no longer meeting Medicaid criteria because of increased income to receive an additional year of Medicaid coverage. Individuals are required to have received Medicaid for three of the previous six months in order to qualify for TMA. Families receiving TMA are entitled to six months of coverage regardless of the amount of their earnings as well as an additional six months if their earned income, minus child care costs, does not exceed 185 percent of the federal poverty level (GAO 1999b). Families receiving TMA are subject to strict reporting requirements. Because many states do not report TMA as a separate Medicaid group, national data on TMA are not available.

SCHIP. The Balanced Budget Act of 1997 provided funds to states for a new program to expand health coverage to low-income uninsured children who do not qualify for a state=s current Medicaid program. Under SCHIP, states may expand Medicaid, establish a new program separate from Medicaid, or develop a combination of both. Thirty-four states have opted to expand Medicaid or combine an expansion with a separate program (Ullman et al. 1999). HCFA had approved the SCHIP plans of 48 states as of January 1999 (Nathan and Thompson 1999).

 

B. Changes in the Macroeconomy

1. The Economy and FSP Participation

Historically, FSP participation has been influenced by changes in macroeconomic factors (Beebout 1998, Castner and Anderson 1999). For example, the surge in FSP participation from 1989 through 1994 was related to a worsening economy, while the decline that began in 1994 coincided with an improving economy.

By the end of 1998, the unemployment rate was at a 30-year low, even for workers without a high-school diploma (Wallace and Blank 1999), and the U.S. had not experienced a single quarter of negative economic growth since 1992 (Gunderson, LeBlanc, and Kuhn 1999). There is some disagreement about the extent to which the favorable economy may have created jobs with wages high enough to potentially reduce the need for safety net programs such as the FSP. GAO (1999a) conducted a survey of all 50 states and the District of Columbia in an effort to understand the factors that have led to declining FSP participation. According to the survey results, most states believe that the primary factors are the strong national economy and the PRWORA provisions that tightened FSP eligibility requirements (e.g., provisions applying to immigrants and ABAWDs). Yet states also indicated that their own initiatives and those of local governments designed to reduce welfare rolls have also affected participation in the FSP. Estimates from models of the relationship between the economy and FSP participation indicate that the ongoing decline in unemployment rates can account for about 28 to 44 percent of the decline in the FSP caseload at best, suggesting that the drop in FSP participation must be due at least in part to factors other than the strong economy (Wallace and Blank 1999; Beebout 1998).

2. The Economy and the Welfare Caseload

Because the FSP and Medicaid caseloads may be affected by the number of families receiving cash assistance welfare, it is helpful to examine the degree to which the welfare caseload has declined due to economic conditions. There is no clear consensus about how much of the decline in the welfare caseload is explained by state welfare reform policies as opposed to the economy, at least during the period of welfare waivers (1993-96) (see, for example Danziger 1999, Primus et al. 1999, Martini and Wiseman 1997). Welfare researchers tend to agree that both factors have played a role, but they disagree about their relative importance. An analysis conducted by the President=s Council of Economic Advisors (CEA)(1997) attributes more than one-third of the decline between 1993 and 1996 to the unemployment rate and less than one-third to state welfare waiver policies. Figlio and Ziliak (1999) replicated the CEA model and also attribute one-third of the decline to the economy but only about one-sixth to welfare waivers; however their preferred models, which control for caseload dynamics, attribute about half to three-quarters of the change to the economy and none to welfare reform.

When post-PRWORA data are included in similar analyses, researchers have found that a much smaller part of the decline is due to the unemployment rate (Wallace and Blank 1999). The CEA (1999) recently updated its prior analyses using post-PRWORA data and found that because the unemployment rate did not decline as much in the latter period as in the earlier period of welfare waivers (1993-96), a smaller proportion of the post-PRWORA decline can be attributed to broad economic conditions. Only 8 to 10 percent of the 1996-98 drop in the TANF rolls is estimated to reflect the improved labor market, with an additional 10 percent attributed to the higher minimum wage. Roughly one-third of the decline in the welfare caseload in the post-PRWORA period is attributed to reforms implemented under TANF. The CEA (1999) estimates that 35 to 45 percent of the change in the welfare caseload from 1996 to 1998 is accounted for by factors other than changes in the economy (including the unemployment rate, minimum wage, and level of welfare cash benefits) and TANF policies.

3. The Economy and Medicaid Enrollment

GAO (1999b) reported that most state officials attribute the decline in Medicaid enrollment, as they did the decline in FSP participation, to favorable economic conditions and their welfare-to-work policies. Yet the GAO study indicates that most TANF leavers continue to be eligible for Medicaid, especially when federally required and state-selected income disregards and other options are considered. The agency also found that former welfare recipients in its study generally held low-wage jobs and worked less than full-time. Forty-three percent of adults in these families did not have health insurance, and although over half had been offered employer-sponsored health coverage, 60 percent had declined it because they were unable to afford it or for other reasons. Only 8 percent of those declining coverage were participating in Medicaid. Other studies have also indicated that families who move from welfare to work typically get jobs that pay less than $8 an hour and that do not provide benefits such as health insurance or paid sick leave (Families USA 1999, Parrot 1998).

4. Summary of the Economy=s Effects on Caseloads

The literature to date appears to indicate that a portion of the decline in the FSP and Medicaid caseloads may remain unexplained even after one accounts for changes in the economy, including lower unemployment, changes in policy that make certain groups ineligible, and increased wages that could move families out of poverty. The findings suggest that a portion of the decline in caseloads may be accounted for by low-income families who are eligible for benefits but who do not participate. This raises the question of whether the FSP and Medicaid caseloads are declining partly as a result of how welfare reform has been implemented, especially insofar as how the reforms as implemented interact with the FSP and Medicaid.

1. Changes in the Pool of Eligible Families

The decline in TANF caseloads has shifted large numbers of people from one eligibility category to another within the pool of potentially eligible people in both the FSP and Medicaid, but it has not necessarily reduced the number of families who are eligible for program benefits. A significant number of families, especially those headed by women, remain in poverty and are likely to continue to be eligible for the FSP and Medicaid (Primus et al. 1999). For example, more families are now likely to be members of categories of people who have historically had lower participation rates (e.g., those not receiving cash assistance). A recent study suggests that the pool of families eligible for food stamps has declined slightly, but not enough to account for the caseload drop (Castner and Cody 1999b, USDA 1999b). Thus, in attempts to improve program participation, it will be necessary to address not only new problems, but old ones as well.

In general, declines in program participation among eligible families can occur because fewer eligible people apply for benefits or because people participating cease to do so. For example, fewer people may apply for food stamps or Medicaid because they are unaware that they may be eligible (perhaps confusing TANF rules with those of these other programs), or because of a reluctance to approach the welfare office due to stigma. We will explore these reasons and others for nonparticipation in a later section (see Section V). Here, we examine how participation declines may occur as a result of discontinued participation by eligible families. We focus on families who currently receive TANF benefits and those who have left AFDC/TANF.

1. Participation among TANF Recipients

The proportion of all food stamp households who receive welfare decreased from 38.1 percent of the caseload in 1994 to 34.6 percent in 1997 (USDA 1999b). This decline likely reflects the drop in the welfare caseload during this period rather than a drop in participation among families receiving TANF benefits.

Data on food stamp receipt within the welfare caseload indicate a small decline in program participation since welfare reform. Recent tables of AFDC/TANF data from the USDHHS, Administration for Children and Families reveal that the proportion of the welfare caseload that receives FSP benefits has declined slightly since the implementation of PRWORA. Prior to PRWORA, 89.3 percent of AFDC families received food stamps (from October 1995 to September 1996) (USDHHS 1997). After TANF was implemented (October 1997-September 1998), participation dropped to 83.5 percent (USDHHS 1999), although this change may reflect the adjustment to new reporting requirements. In addition, greater variability in FSP participation among states occurred during the latter period compared to the former, with proportions ranging from a low of 61.2 percent in Idaho to a high of 99.4 percent in Indiana (USDHHS 1999).

Because eligibility for welfare and Medicaid was linked prior to PRWORA, 100 percent of welfare recipients were enrolled in Medicaid. USDHHS (1999) reported that Medicaid participation among TANF recipients averaged 98.1 percent after Medicaid and TANF were delinked (from October 1997 to September 1998). Thus, there has been little change in Medicaid enrollment since welfare reform among families currently receiving cash assistance.

To examine participation patterns among low-income families with children who have left welfare, it is helpful to use longitudinal data because they show what happened to the same group of people over time following their exit from welfare. Here we are interested in understanding the degree to which families discontinue their participation in the FSP and Medicaid after leaving cash assistance. In this section, we review information available from a nationally representative household survey, from the CPS, and from a variety of state- and local-level studies that document program participation.

State and Local Exit Studies. Table 3 presents findings from state and local exit studies. We restrict our discussion, and Table 3, to those studies with response rates above 50 percent (see Appendix C for a complete list of all studies reviewed, with key information about their methodologies). As shown in the table, there appears to be considerable variation in program participation across this set of state and local exit studies. For Medicaid, participation ranges from 20 percent to 80 percent at follow-up (usually 12 months after exit). For the FSP, participation ranges from about 29 to 61 percent at follow-up. Thus, on average, roughly half

We should note that the studies in Table 3 are similar in a number of respects, yet they vary significantly in method, scope, and length of time families are followed. All were conducted after the passage of PRWORA, when states were in the process of implementing welfare reform, and most examine participation at least to the 12-month point. In addition, each study (except New Jersey) defines a Aleaver@ in the same way: a case that has left cash assistance for at least two months. One should bear in mind that clients within each sample may cycle back on cash assistance once the two-month period has lapsed. Among the studies reviewed, between 20 and 35 percent of clients returned to TANF assistance within a year of case closure. This point is particularly important because clients who return to cash assistance are likely to also return to food stamps and/or Medicaid. Finally, the reader should note that families may leave welfare for reasons other than employment, such as marriage, a move, or ceasing to have an eligible child.

aData were gathered from administrative records and averaged over the quarter after cash assistance case closure, unless otherwise indicated.

bData are based on the average number of cases that exited during the first quarter of 1996.

cData are based on the average number of cases that exited during the fourth quarter of 1998.

dData are based on the average number of cases during the quarter in which clients= cash assistance cases were closed.

eData were gathered from administrative records during the quarter in which the clients= cash assistance case was closed.

fClients interviewed 6 months (rather than 12 months) after cash assistance case closure.

gBased on the portion of the study sample who left TANF at some point from July 1997 to June 1998. Medicaid use for New Jersey is Apublic health insurance@.

hEmployment rates for all South Carolina studies indicate the percentage of sample families whose cases were closed due to an increase in earnings.

iClients were interviewed between 6 and 9 months after cash assistance case closure.

leaving welfare. In addition, only about one-quarter of the employed families in these studies were receiving employer-based health coverage. We should note that a few of these exit studies have extremely low response rates (e.g., New Mexico=s was 12 percent; Louisiana reported 17.5 percent). Greenberg=s review of evaluation studies indicates that in each case, the welfare-to-work program was successful in increasing employment in the program group but unintentionally resulted in a decline in health care coverage that was greater than that in the control group.

Using administrative data, Ellwood and Lewis (1999) found that in California and Florida, about half of the children who left AFDC also left Medicaid within six months. One-half to two-thirds of the AFDC adult leavers also left Medicaid during this time. Furthermore, only 6 percent of adult leavers in California and 9 percent of Florida leavers had TMA. The rates of TMA were only slightly higher at three months after exit. Although the data for this study were collected in 1995, prior to PRWORA, both states had welfare reform underway and were experiencing declines in welfare caseloads. A draft report on Medicaid participation in another three states (Alabama, Michigan, and New Jersey) shows similar results (Ellwood and Irvin 1999). In this study, at least half of the children and parents leaving welfare also left Medicaid.

National Exit Studies. Loprest=s (1999) results from a study based on a nationally representative sample of families are similar to the state-level results. Within two years of leaving TANF/AFDC, less than one-third of former welfare recipients were receiving food stamps. This finding relies on data from the National Survey of America=s Families (NSAF), in which a sample of families were asked about their program entries and exits from 1995 to 1997. Zedlewski and Brauner (1999) also used NSAF data in their analysis and found that the earnings and income of former welfare families and nonwelfare families who left the FSP were very similar; yet former welfare families left the FSP at higher rates (62 percent) than their non-welfare counterparts (46 percent), even when they remained eligible.

The NSAF data have also been used to examine Medicaid participation for welfare leavers. Garrett and Holahan (1999) report in a working paper that 41 percent of mothers and 30 percent of children who left welfare were not enrolled in Medicaid. Furthermore, they found that rates of uninsurance for families who left welfare increased with the number of months since exit. Only about 23 percent of mothers who left welfare were able to obtain private or employer-sponsored health insurance.

Program Eligibility among Leavers. The earnings of leavers in the state/local studies in Table 3 suggest that many families who found employment would likely still be eligible for FSP and Medicaid benefits, even 12 months after leaving cash assistance. Reported average monthly earnings range from $648 to $1,187 at the point of leaving cash assistance, with higher earnings occurring a year later (between $766 and $1,504). Even though earnings may increase over time, it is unlikely that they alone would account for the decline in FSP and Medicaid participation. These findings are consistent with those reported in the NSAF study, as described above, and with Greenberg=s review of exit studies, in which only about 25 percent of former welfare families who were employed reported having employer-based health insurance.

There is evidence to suggest that Medicaid participation by families who leave welfare has historically been low. Moffitt and Slade (1996) examined the Medicaid participation of welfare leavers in the National Longitudinal Survey of Youth (NLSY) and found that 52 percent of mothers who left AFDC had Medicaid benefits one year later. Three years later, only 17 percent of mothers and 34 percent of children were receiving Medicaid benefits. The take-up rate for transitional Medicaid assistance has been observed to be even lower. Ellwood and Adams (1990) examined participation in TMA in Georgia and California from 1980 to 1986, long before welfare reform, and found the take-up to be only 5 to 6 percent in California and 12 percent in Georgia.

 

1. The Design and Implementation of TANF Programs

Although states have considerable flexibility to design their TANF programs, many TANF programs operate in very similar ways (Pavetti and Wemmerus 1999). In particular, many states have designed their programs to help families avert crises so as to eliminate the need for ongoing cash assistance. This approach is know as diversion. The states also require nearly all families to find employment as quickly as possible. Extensive use of sanctions--financial penalties for noncompliance with TANF work requirements--and time limits also distinguish current welfare reform efforts from previous efforts. Below we discuss how these features might affect participation in the Food Stamp and Medicaid programs.

 

1. TANF Diversion Policies

In an effort to keep families from ever receiving cash assistance, a number of states have implemented formal policies to divert families from the welfare system. There is some evidence to suggest that formal TANF diversion policies may be unintentionally contributing to lower rates of FSP and Medicaid applications by eligible families. This effect can occur through formal procedures that are intended to deter applicants from participation in TANF, but that may inadvertently discourage application to other programs as well. Beyond these formal means, there is also reason to suspect that potential FSP and Medicaid applicants are deterred informally.

TANF diversion programs encompass a range of state efforts designed to assist families seeking cash assistance in ways that avoid enrolling them in welfare. The first nationwide survey of state diversion programs found that these efforts most commonly include (1) lump-sum payment programs, (2) alternative resource programs that help families access resources in the community or from family or friends, and (3) mandatory applicant job search programs, which require applicants to complete a job search as a condition for processing the TANF application (Maloy et al. 1998). A follow-up study examined the implementation of diversion in five states and documented a range of problems in ensuring access to Medicaid (Maloy et al. 1999).

According to FSP and Medicaid rules, program applications should be processed as soon as they are submitted, yet diversion programs increase the likelihood that this will not occur. Because Medicaid and FSP applications are commonly part of a joint TANF application, eligibility for these programs may be determined incorrectly or not at all (Maloy et al. 1998, GAO 1999a and 1999b). For example, an eligibility worker may incorrectly believe that the FSP or Medicaid application cannot be processed until a TANF-required job search has been completed. Or a worker may improperly deny a Medicaid or FSP application along with the properly denied TANF application for a family receiving a lump sum payment. In other cases, workers may confuse the rules for TANF with other programs, for example, by automatically denying food stamps or Medicaid if the TANF seeker finds employment during the pre-application job search. Some workers have failed to inform families of their potential eligibility for FSP and Medicaid, while others appear to have delayed application processing because of what they perceive as conflicting messages with regard to TANF and the receipt of other program benefits. For example, an FNS review found that caseworkers at two New York City job centers were failing to inform TANF applicants about the availability of food stamps when applicants accepted a diversion payment or were denied TANF benefits (GAO 1999a).

Aside from issues arising from worker-level practices, administrative systems may also be responsible for delays and other problems related to the receipt of FSP and Medicaid benefits as a function of diversion program practices. For example, joint applications may delay or terminate food stamp and Medicaid requests because there is no established process for separating the paperwork.

Confusion regarding diversion rules may also exist among families applying for benefits. If not informed otherwise, some families may assume that diversion requirements apply to all programs, so these families may not apply for benefits or may even abandon their application process before it is complete. Other families may be informally diverted, failing to submit an application at all or even to come to the welfare office because they assume they are ineligible due either to their inability to meet TANF work requirements or to other reasons.

GAO (1999b) also identified state welfare diversion policies as potential obstacles to Medicaid enrollment. Eighteen of the 21 states in the agency=s sample required job search or offered a lump sum payment or both. The agency reported that in states with combined welfare and Medicaid applications, mandatory job search requirements have been found to delay Medicaid eligibility determination. For example, South Carolina officials indicated that they hold combined applications until 10 verified employer contacts have been made. If after 30 days the job search is not completed, the TANF application is denied, and the Medicaid portion is sent to a separate unit for eligibility determination.

Darnell et al. (1999) surveyed 31 states with diversion programs to determine whether these states had used their options under Section 1931 to take into account the effects of diversion on access to Medicaid. Several policy options are available to states to expand Medicaid eligibility criteria to accommodate the increased income of diverted applicants. For example, states can disregard lump sum payments when considering eligibility for Medicaid, and applicants who have been diverted through job search may still be eligible for Medicaid through expanded earnings disregards. At the time of this study, however, most states with diversion programs had not taken advantage of these options.

In an effort to support states in gathering information about families who are formally or informally diverted from TANF, the Office of the Assistant Secretary for Planning and Evaluation at the USDHHS awarded grants to Arizona, Illinois, Iowa, New York, Texas, Washington, and a group of counties in California last year. Most of these studies are currently in process.

 

 

2. Work-First Programs

Almost all TANF programs have a strong emphasis on moving recipients into employment as quickly as possible. This can inadvertently cause problems with continued receipt of FSP and Medicaid benefits for eligible families.

Regardless of how soon after application a TANF recipient goes to work, the TANF case is closed when the individual is no longer eligible for welfare cash assistance. According to FSP and Medicaid rules, the individual=s FSP or Medicaid case should not also be automatically closed at this time. Yet there is concern that this may be occurring with some frequency, probably for a variety of reasons. For example, eligibility workers may be confused about eligibility rules for the various programs, may fail to inform clients about their potential continued eligibility for certain programs, or clients may have difficulty providing the additional information necessary to retain eligibility for FSP or Medicaid.

The Health Care and Financing Administration (HCFA) has also noted that several states have cited as a barrier to transitional Medicaid the requirement that clients must have received Medicaid on the basis of AFDC-related criteria in three of the previous six months in order to be eligible for TMA. Florida officials note that because most families are on TANF for only about three months, many leave before meeting the three-month criteria for TMA (GAO 1999b).

 

3. TANF Sanctions

PRWORA allows states to sanction TANF families who fail to comply with work requirements. Sanctioned families generally lose their cash assistance for some period of time or until they comply with the requirement. As of 1997, 37 states had elected to extend the TANF sanction to all members of a household (Gallagher et al. 1998). Whether TANF sanctions also apply to food stamp benefits can be a somewhat complicated issue, at times leading to litigation against states (GAO 1999a).

According to FNS guidance in The Nutrition Safety Net: At Work for Families, states must sanction families that receive food stamp benefits in the event of a TANF sanction imposed because of noncompliance with TANF work requirements, if the person is also subject to food stamp work requirements (known as Acomparable disqualification@). However, the federal food stamp termination requirement does not automatically extend to all members of the household. PRWORA gave states the option to disqualify the entire household from food stamp benefits if the adult head-of-household refuses to comply with FSP work requirements. Twelve states chose to implement this option as of 1998 (Gabor and Botsko 1998). In the event of a nonwork-related TANF violation (such as falling behind in childhood immunizations), states are not allowed to extend the FSP sanction to the full household, but only to the person violating the requirement. According to GAO (1999a), FNS has determined that some states= implementation of the comparable disqualification provision is not supported by law, resulting in some qualifying households not receiving benefits to which they are legally entitled. For example, a federal district court directed Michigan to stop disqualifying an entire household for food stamps because of one household member=s TANF violation (which in this case was non-cooperation in obtaining child support).

States are not required by federal law to sanction Medicaid benefits of the person violating a TANF work requirement, nor are they permitted to sanction the Medicaid benefits of children or pregnant women at all. Federal law also does not permit states to sanction Medicaid benefits for nonwork-related offenses. States may opt to sanction the Medicaid benefits of adults who violate TANF work-related requirements (except in the case of pregnant women), but few states have elected to do so.

Even in states that have not exercised the option to sanction individuals receiving Medicaid (or the full family in the case of food stamps), participation in these programs could still be affected by TANF sanctions. Families who have been sanctioned under the TANF program may be reluctant to continue participating in other programs, particularly if their eligibility is determined by the same worker who imposed the sanction or if they do not understand that TANF sanctions do not affect their Medicaid or food stamp benefits in the same way.

FSP and Medicaid Participation among Sanctioned Families. Early state experiences with sanctions in Massachusetts, Iowa, and Michigan were summarized by the Kaiser Commission with respect to Medicaid participation (Greenberg 1998). Two to five months after benefit termination, the proportion of families with at least one member receiving Medicaid fell in Massachusetts from 100 percent to 58.5 percent, a drop very similar to that in Wisconsin, where the proportion of such families fell from 100 percent to 53.5 percent. In Iowa, Medicaid enrollment declined from 86.3 percent at exit to 54.4 percent. The results of an early study of welfare reform in Iowa by Fraker et al. (1997) indicate that in about one-third of sanctioned families, neither the child or the parent was receiving Medicaid benefits, and in 28 percent of sanctioned families, neither the parent or child was receiving either Medicaid or employer-based coverage.

Table 4 shows findings from other state studies of TANF sanctions. In every case except one, food stamp and Medicaid use declined with time after sanctioning. The Arizona study directly compares program participation by sanctioned versus nonsanctioned families and shows that participation in food stamps and Medicaid was lower for sanctioned families, both at exit and at 12 months after exit. The South Carolina study compares program participation for sanctioned families to cases that were closed due to increased income. In this study, sanctioned families had slightly higher rates of participation than families who left welfare because of increased income.

 

4. TANF Time Limits

Although most states have adopted TANF time limits as a part of their efforts to reform the welfare system, only a few states have had significant numbers of families reach these limits. States are prohibited from applying time limits to Medicaid and the FSP (USDHHS 1999b, USDA 1999c). Unlike other features of state and local TANF programs, time limits may not necessarily adversely affect participation in the food stamp and Medicaid programs. Families who reach even the shortest TANF time limit (21 months) and are working may continue to qualify for Medicaid under other eligibility provisions (such as Section 1931)(Schneider et al. 1998). If they are ineligible under Section 1931, they may receive TMA, assuming that they received Medicaid for three of the six months prior to reaching the time limit. Furthermore,

Table 4

FOOD STAMP AND MEDICAID USE IN TANF SANCTION AND TIME LIMIT STUDIES: SUMMARY FINDINGS^a

TANF SANCTION STUDIES
	Employment Rate		Monthly Earnings		Food Stamps Use		Medicaid Use
Study	Exit	Follow-Up	Exit	Follow-Up	Exit	Follow-Up	Exit	Follow-Up
Arizonab Sanctioned Nonsanctioned	55.1% 40.0%		$744 $550		50.6% 73.5%	42.4% 52.4%	56.5% 88.9%	45.7% 55.1%
Iowac (May 1997) Admin data Survey data d Iowa (Aug 1999)e	33.0%	53.0% 43.0%		$731 $925	60.5% 88.3% 80.0%	35.5% 63.5% 71.0%	95.6% 71.0%	66.4% 76.0%
GAOf Iowa Massachusetts Wisconsin	17.4% 14.3% 35.9%	30.4% 23.0% 31.5%	$692 $394 $551	$741 $540 $870	83.6% 95.9% 76.2%	60.8% 25.6% 55.6%	86.3% 100% 100%	54.4% 58.5% 53.5%
South Carolinag Sanctioned Closed due to income All Closures	36.0% 80.0% 50.0%	43.0% 70.0% 50.0%			73.0% 73.0% 63.0%	58.0% 54.0% 52.0%		54.0% 46.0% 46.0%
TANF TIME LIMIT STUDIES
	Employment Rate		Monthly Earnings		Food Stamps Use		Medicaid Use
Study	Exit	Follow-Up	Exit	Follow-Up	Exit	Follow-Up	Exit	Follow-Up
Connecticuth		83.0%		$1,011		50.2%
Virginiai	62.5%	71.3%	$848	$902	83.2%	70.7%

 

there is a federal expectation that states will reassess eligibility under other categories before terminating an individual=s Medicaid benefits. Thus, as families reach time limits, staff may spend considerable time reassessing the family=s situation to ensure that benefits are not terminated in error or that the termination will not cause undue harm to the children in the household. Finally, in some states, most families who reach time limits are employed. Since these families have been receiving food stamp and Medicaid benefits while employed, they may be less likely to think they are no longer eligible. On the other hand, it is possible that some families who reach time limits and lose their benefits may wrongly assume that they are no longer eligible for any government assistance.

We reviewed two studies that tracked families who reached TANF time limits (also shown in Table 4). In Connecticut, 83 percent of families were employed six months after reaching the time limit, and 50 percent were receiving food stamps. Although almost half of the sample was offered employer-sponsored health insurance, only 33 percent were enrolled in a plan. In Virginia, almost three-quarters of the sample was employed six months after reaching the time limit and had average monthly earnings of $902. Seventy-one percent of the sample was receiving food stamps. Ninety-two percent of households reported that at least one family member was covered by some type of health insurance, which in most cases was Medicaid.

 

1. The Design and Implementation of the FSP and Medicaid in a Work-Oriented Welfare Environment

Recent findings reveal that factors other than TANF policies may be contributing to the decline in FSP and Medicaid participation under PRWORA, including the added complexity of rules and procedures, inadequate automated communication and eligibility systems, and poor coordination between welfare and Medicaid at the local level (Ellwood 1999). Some of the issues we review below are related to operating entitlement programs in a work-oriented welfare environment, including the effect of the stigma associated with dependence on welfare, eligibility worker and client confusion due to new program rules that have been put in place as a result of welfare reform, and the use of automated eligibility systems that have not been brought up to date in terms of the new policies. Other more general issues we review are related to participation in public programs irrespective of the welfare context, specifically the cost-benefit ratio for families who consider participation in entitlement programs and federal guidance on the implementation of these programs.

 

1. Confusion about Potential Eligibility among Families and Workers

Confusion about Medicaid and FSP eligibility rules has been reported throughout the system, from directors of regional offices to potential beneficiaries themselves.

Medicaid. The rules regarding Medicaid eligibility have become extremely complex, making them more difficult to implement (GAO 1999b, Ellwood 1999). GAO notes that adding Section 1931 to the eligibility categories has further complicated already complex program rules. In interviews with officials in 21 states, the number of welfare-related Medicaid eligibility categories was reported to range from almost 30 to over 100. During site visit interviews in five states, workers reported that Medicaid rules were difficult to understand and explain to their clients (Ellwood 1999). In particular, welfare staff were often charged with explaining Medicaid rules to clients, informing them of eligibility in different circumstances, and collecting eligibility information, yet many times they lacked the necessary training in Medicaid eligibility to do so. The most confusing areas in the Section 1931 rules were reported to be (1) the steps for determining TMA, (2) the impact on eligibility when families fail to meet reporting requirements, and (3) income disregards that differ across eligibility categories. A study conducted by Smith et al. (1998) is consistent with these findings. Interviews with administrators of human services indicated that eligibility staff and potential Medicaid beneficiaries themselves are confused about eligibility rules, and there appears to be a general lack of information explaining the major reforms and new rules (Smith et al. 1998).

Potential Medicaid beneficiaries themselves confirm these findings. According to a recent nationwide survey conducted for the Kaiser Commission on Medicaid and the Uninsured, the Medicaid program is well known and held in high regard by low-income families but is not considered Auser-friendly@(Perry et al. 2000). Sixty-two percent of program applicants who did not receive benefits (some of whom were ineligible, but many of whom did not complete the application process) said they believe the process to be complicated and confusing. Moreover, a 1999 survey of approximately 1,100 heads of households and patients at 20 community health centers in 10 states has yielded preliminary findings indicating that 30 percent of the sample incorrectly believe that TANF work requirements also apply to Medicaid. In addition, 27 percent think that TANF time limits apply to children=s Medicaid benefits, and 17 percent believe that a person had to be on welfare in order to receive Medicaid (Stuber et al. 1999).

Welfare recipients may historically have had little awareness of their potential continued eligibility for medical benefits in the event of employment. In interviews with AFDC recipients, Shuptrine et al. (1994) found that 41 percent of AFDC recipients in Tennessee and North Carolina were unaware that they could work full time and potentially still receive benefits. Similarly, two-thirds of AFDC recipients did not realize that children in two-parent families could be eligible for Medicaid. In a later North Carolina study, Shuptrine and McKenzie (1996) found that 39 percent of parents receiving AFDC did not understand that if they left welfare for work, their children would still be able to receive Medicaid benefits. In addition, the study found that health care providers and community organizations had a poor understanding of Medicaid eligibility provisions.

The Food Stamp Program. Confusion about FSP eligibility among potential beneficiaries has also existed for some time. Studies conducted in the 1980s indicated that one-half to one-third of FSP eligible nonparticipants did not apply for benefits because they thought they were ineligible (Coe 1983, GAO 1988, Hollenbeck and Ohls 1984). Findings from focus groups conducted after welfare reform (1998) suggest similar results (McConnell and Ponza 1999). Furthermore, the National Food Stamp Survey (NFSS), a nationally representative randomized digit-dial survey of low-income households conducted between June 1996 and January 1997, indicates that 72 percent of nonparticipants estimated to be eligible for the FSP are not aware that they could be eligible, indicating that this was their primary reason for not participating (Ponza et al. 1999). Of this group, 74.5 percent had household earnings. These results suggest that there has been widespread misunderstanding about FSP eligibility among low-income families both before and after welfare reform.

In a review of FNS regional offices, four food stamp directors reported that the implementation of TANF has been an important factor in the decline of participation. They cited confusion about eligibility rules for food stamps, on the part of both eligibility workers and food stamp applicants. According to these directors, many people do not apply because they assume that if they are not eligible for TANF, they are also not eligible for food stamp benefits (GAO 1999a).

There may be differences in the degree of awareness of eligibility rules depending on whether the family has recently been on welfare. Among people who left the FSP, significantly more former welfare families than nonwelfare families reported that they were not participating in the FSP because of a new job or increased earnings (Zedlewski & Brauner 1999). The authors suggest that more welfare families than nonwelfare families may believe that earnings disqualify them for FSP benefits. Consistent with this possibility, a study of TANF recipients who reached time limits in Virginia indicated that six months later, 54.1 percent of all cases not receiving food stamps thought they were not eligible, despite the fact that 54.4 percent of the sample had an income below 130 percent of poverty (Gordon et al. forthcoming).

Awareness of potential eligibility for food stamps may be greater in some areas compared to others. Two leaver studies (in Wisconsin and South Carolina) reported that most families were aware they might qualify for Food Stamps after leaving welfare. Most also knew that their children might be eligible for Medicaid, but they were less likely to know of their potential eligibility for adult Medicaid. In South Carolina, leavers who were working were also more likely than those who were not working to be aware that they might qualify for adult Medicaid (72.2 percent compared to 33.8 percent).

 

2. Participation Costs

The Food Stamp Program. About five percent of eligible nonparticipant households in the NFSS responded that too much money, time, and hassles were involved in participating in the FSP (Ponza et al. 1999). Examples include too much paperwork (2.8 percent), a problem with transportation (1.5 percent), and a perception that the benefit was too small for the effort required (2.8 percent). In the National Survey of America=s Families, 10 percent of former welfare families and 17 percent of nonwelfare families reported leaving food stamps because of administrative problems or hassles (Zedlewski and Brauner 1999). These concerns and others were also raised in focus groups by nonparticipants (McConnell and Ponza 1999). Focus group members reported difficulty getting to the FSP office (e.g., need to arrange for dependent care, losing pay while at the FSP office, physical difficulties getting to FSP office because of poor health or disabilities). They also felt that the application and recertification questions were too personal.

It is possible that these perceived Ahassles@ may stem from the fact that most participants are required to return to the food stamp office four times a year to be recertified for benefits; moreover, working families, because their income is more likely to change, are required to be recertified more often than those with a fixed income. Most food stamp offices are open only during business hours, requiring at least some working heads of household to take time off from work to be recertified. Other office procedures that can be perceived as Ahassles@ were identified in an FNS review of food stamp program operations. For example, in certain area offices with Afirst-come, first-served@ procedures, clients who arrived before 8:30 AM waited all day without being interviewed and were told to return another day and start over again (GAO 1999a). Potential food stamp beneficiaries have reported in focus groups that applying and getting recertified is time-consuming and complicated. They cited having to wait a long time to be served in the FSP office, finding the application form to be long and complex, and having difficulty obtaining all the necessary paperwork (McConnell and Ponza 1999).

Medicaid. Major findings from a national survey of parents of low-income children sponsored by the Kaiser Commission on Medicaid and the Uninsured indicate that most (72 percent) of those who unsuccessfully applied for Medicaid benefits on behalf of their child reported having difficulty obtaining the required documents as the main reason for not completing the application process (Perry et al. 2000). Sixty-six percent indicated that the overall hassle of the enrollment process was the primary reason for not completing the application. Despite these problems, the vast majority of low-income parents in this national sample reported believing that having health coverage for their children is very important and that Medicaid is a good program. Sixty percent of parents of eligible uninsured children indicated that they would be more likely to enroll their children if the enrollment process included a phone-in option, 55 percent indicated that extended office hours would encourage their enrollment, and 56 percent said that being able to enroll immediately but complete the forms later would make them much more likely to apply.

Families who receive TMA have found the reporting requirements difficult to meet (Ellwood 1999, GAO 1999b). GAO has urged Congress to consider revising section 1925 of the Social Security Act to allow states to reduce or eliminate income reporting requirements for families receiving TMA. Currently, in all but 3 of the 21 states they contacted, clients must follow very specific reporting requirements. Income must be reported at entry and by the 21st day of the 4th month, even though benefits are guaranteed during the first six months regardless of income. During the second six months, clients must submit quarterly income reports by the 21st day of the 1st and 4th months. In HCFA=s FY 2000 budget, legislative changes were requested to remove the burdensome reporting requirements for TMA.

 

3. Use of Automated Eligibility Systems

In site visits to a number of states, Ellwood (1999) found that automated eligibility systems could not meet the requirements of Medicaid=s new rules. The systems were designed to meet welfare, rather than Medicaid, needs. Medicaid workers indicated that clients were sometimes erroneously terminated due to the automated system. In addition, Medicaid termination notices sent automatically to clients were very technical, difficult to understand, used unexplained acronyms, and often did not give reasons for terminating a case.

According to GAO=s interviews with workers and state officials, many computer systems have not kept up with welfare policy changes (GAO 1999b). Workers in Florida reported that they must either manually determine Medicaid eligibility or understand the policies well enough to verify the accuracy of the computerized system=s determination. Workers in California also reported having to manually determine Medicaid eligibility for the section 1931 category. California indicated that much of its programming expertise has been devoted to the Y2K issue.

4. Federal Guidance on the FSP and Medicaid under Welfare Reform

Inadequate federal guidance on the implementation of the FSP under welfare reform is, in part, responsible for the caseload decline, according to a report produced by the General Accounting Office (GAO 1999a). On the basis of its findings, GAO recommends and USDA agrees that FNS should promulgate regulations regarding PRWORA=s revisions to the Food Stamp Act, publicize eligibility requirements for the FSP and distinguish them from TANF requirements, and target participant access issues when reviewing food stamp office operations. GAO further concludes that mistakes that have been made by states that attempt to go farther than the law permits in limiting eligibility (e.g., requiring job searches on an applicant=s first visit; denying whole households rather than individuals who commit a TANF violation) also account for a portion of the declining caseload.

In a study on the decline in Medicaid enrollment, GAO (1999b) concludes that despite federal protections to prevent declines, it is now more complicated for eligible low-income families to gain access to and continue participation in Medicaid. The agency recommends that HCFA take several steps to provide additional guidance to states regarding this issue. HCFA has responded to the GAO report by indicating that it is sponsoring several studies to examine the factors contributing to declining enrollment, developing plans to issue additional guidance to states on TMA, providing on-site technical assistance to states, and visiting every state to ensure each is meeting the challenge of sustaining enrollment under welfare reform.

As noted, USDA and USDHHS have produced guides for states, local and regional offices in order to clarify rules for participation in the FSP and Medicaid within the context of welfare reform. The FNS guide, The Nutrition Safety Net: At Work for Families (USDA 1999c) outlines what food stamp regulations require of states in terms of eligibility, application processing, recertification and other matters, particularly within the welfare reform context. Further guidance has been provided to state or regional offices through a series of letters and notices, most of which are posted on the FNS web site.

A joint ACF/HCFA publication, Supporting Families in Transition: A Guide to Expanding Health Coverage in the Post-Welfare Reform World, (USDHHS and USDA 1999), offers a road map to the administration of Medicaid and SCHIP under welfare reform. It includes information on mandatory eligibility, application and enrollment policies, as well as optional policies that states may implement to increase participation. Guidance is also offered on administrative practices that improve potential applicants= access to the programs. In addition, in 1997, HCFA amended the state Medicaid manual to reflect changes to the law made by welfare reform and has issued further guidance in the form of fact sheets and letters.

 

5. Stigma Associated with Food Stamps and Medicaid

Approximately seven percent of the NFSS sample identified a stigma-related or psychological reason as the most important grounds for not participating in the FSP. Some respondents reported that they do not like to rely on government assistance or charity. Other specific reasons included not wanting to be seen shopping with food stamps, being too proud to ask for assistance, being asked questions of a personal nature, not wanting peers to know they need help, being treated badly as a result of receiving food stamps, and having had a bad experience with the FSP. Findings from focus groups of nonparticipants indicate that some low-income families do experience embarrassment, a sense of failure, hurt pride, or the feeling that they have lost their independence when applying for food stamps (McConnell and Ponza 1999). Yet it is unclear how important stigma itself is as a reason for nonparticipation; less than 25 percent of respondents to a survey pretest based on these focus group findings indicated stigma as a reason for not participating, and less than 5 percent said it was the most important reason.

According to Smith et al. (1998), who conducted focus groups with human service administrators, state experts on Medicaid eligibility, and welfare researchers, potential beneficiaries of Medicaid appear to still closely associate Medicaid with welfare, and the negative connotation associated with TANF receipt appears to generalize to Medicaid. Preliminary findings from a survey of clients at 20 community health centers indicate that one-quarter to one-third of patients and heads of households perceive some degree of stigma to be associated with Medicaid (Stuber et al. 1999). In addition, 25 percent of the sample believe that people are treated poorly when they apply for Medicaid, 30 percent believe that Medicaid recipients are not respected, and 31 percent believe that doctors do not treat Medicaid patients as well as they do people with private health insurance.

Recent findings from a national survey of low-income parents indicate that 4 in 10 parents of Medicaid-eligible uninsured children (42 percent) would be much more likely to enroll in Medicaid if they did not have to go to a welfare office (Perry et al. 2000). This finding supports the idea that families may be more likely to enroll if the public could be persuaded to view Medicaid not as a welfare program but as a support for working families.

 

V. Strategies for Increasing Participation and Our Understanding of the issues

An important, but challenging, part of the effort to increase program participation in the context of welfare reform involves changing public attitudes toward food stamps and Medicaid. Conducting public awareness campaigns and increasing outreach to target nonwelfare working families are examples of strategies that may begin to reshape the public=s view of these programs as important supports for working families. In addition to strategies for accomplishing this broader objective, a fair number of more specific strategies and ideas have been developing for increasing participation in Medicaid and the FSP. In this section, we review suggestions for improving access and continued participation among welfare applicants and welfare leavers and for reaching families who have no contact with the welfare system. We also discuss new and ongoing research projects that are intended to increase our understanding of issues related to participation in the FSP and Medicaid in the context of welfare reform.

 

A. Increasing Participation in the FSP among TANF Applicants and Leavers

The FNS guide (USDA 1999c) referred to above provides suggestions for ways to increase participation in the FSP, particularly by TANF applicants and welfare leavers. To improve access to the FSP for TANF applicants, the guide makes the following recommendations:

$ Train TANF agency staff to educate TANF applicants about FSP eligibility.

 

• Where TANF and FSP programs are administered separately, place food stamp eligibility workers in TANF offices. This practice may be particularly important in sites that have diversion policies.

 

• Use Afamily friendly@ scheduling procedures to allow working parents to attend appointments outside work hours or days.

 

• Use less intrusive employment and income verification procedures.

 

• Develop shortened and simplified common access application forms.

 

• Make greater use of technology to simplify program access.

 

To help eligible families retain food stamp benefits after they leaveTANF, the guide suggests the following strategies and Aexemplary state practices@:

$ Educate families about their potential continued eligibility before they become employed

 

• Simplify recertifications

 

• Review closed TANF cases

 

• Request income reporting waivers to allow 6-month rather than 3-month recertification periods

 

• Request waivers to increase the income reporting threshold from $25 to $100

 

• Follow up by calling families whose TANF cases have been closed for non-financial reasons

 

• Develop and implement a customer service plan

 

 

2. Increasing Participation in Medicaid and SCHIP for TANF Applicants and Leavers

The ACF/HCFA guide for Medicaid and SCHIP parallels the FNS guide for the Food Stamp Program in providing information on increasing enrollment. A number of other individuals and organizations have also distributed information on practices that could increase participation in Medicaid and SCHIP (e.g., Shuptrine et al. 1998, Families USA 1998, Center on Budget and Policy Priorities 1998b, Guyer and Mann 1998, Ross and Guyer 1999). Many strategies involve making greater use of the flexibility that states have been given with respect to Medicaid eligibility policies. For example, states can use less restrictive methodologies for calculating income, continue certain AFDC waivers, provide continuous eligibility for children, eliminate the extension of sanctions in certain cases, and give presumptive eligibility to children and pregnant women.

Other strategies focus on administrative practices, and several overlap with those described above for increasing participation in the FSP. Some practices that are exclusive to increasing participation in Medicaid include:

$ Acceptance of mail-in applications and redetermination forms

 

• More emphasis on Medicaid eligibility training for welfare workers

 

• Delinking Medicaid and TANF redeterminations

 

• Simplifying redeterminations

 

• Making greater use of TMA

 

• Educating TANF families about termination of benefits and TMA

 

 

In addition to these recommendations, Ellwood (1999) and other researchers (Maloy et al. 1999) have concluded that the Medicaid enrollment system needs to be more streamlined and customer-friendly, which could include simplifying rules, application and redetermination forms, making notices easier to understand, and increasing use of the phone and mail for communication. Ellwood (1999) notes that Medicaid staff in some localities are trying to make themselves more available to welfare staff. For example, one county set up a Abuddy@ system to help welfare staff understand the complexities involved in Medicaid eligibility. In another location, $20 gift certificates were offered to families who reapplied to Medicaid after losing welfare benefits.

 

2. Access to Medicaid and the FSP for Families Not in Contact with TANF: Outreach and Education

We have already reviewed evidence that FSP participation by eligible low-income families who have no contact with the welfare system has historically been low. Ellwood and Irvin (1999) have reviewed evidence from several researchers who have found this to also be the case for Medicaid. According to one study, for example, the participation rate for children qualifying under welfare-related rules was substantially greater than the rate for children eligible under the child poverty-related expansion provisions (Dubay and Kenney 1997).

We have also noted that there may be many eligible low-income families with children who do not apply for food stamp or Medicaid benefits because they incorrectly believe that they must also apply for TANF in order to be eligible for the other two programs. Alternatively, these families may be reluctant to approach a welfare office to apply for Medicaid or FSP benefits because of stigma. For these and other reasons, outreach efforts may be even more important in the context of welfare reform. Federal funds have been allocated for Medicaid outreach and enrollment activities with a more generous match rate than that for administrative activities (Mann and Guyer 1998). A recent survey of states indicates that as of early 1999, most of the federal funds available for increased outreach and enrollment activities had not been used (Darnell, Lee, and Murdock 1999), possibly due in part to a time limit that was originally scheduled to sunset on October 1, 2000. However, Congress recently removed the time limit and restored access to federal funds for states that had lost it. States are now allowed to draw on the funds until the money is fully expended (Ross and Guyer 1999), using their share to support such activities as redesigning their application forms, updating their computer systems, and conducting outreach and education programs.

Increasing participation in Medicaid may involve more than outreach efforts that facilitate enrollment. It may also be necessary to educate families about the value of preventive medical care for themselves and their children. For instance, because it is usually possible to have emergencies covered by Medicaid, families may be reluctant to bother going through the enrollment process when they believe it is not necessary. However, if these families were made aware of the importance of preventive and ongoing care, they might be more inclined to enroll in Medicaid in order to receive such care.

To increase access for families who have no contact with TANF, the FNS and HCFA guides recommend the following strategies:

$ Build FSP and Medicaid partnerships in communities

 

• Place food stamp and Medicaid eligibility workers in community-based organizations

 

• Train volunteers and employees of other government agencies to take applications at community locations

 

• Develop and disseminate information on food stamps and Medicaid

 

• Improve accessibility of application sites

 

• Integrate the health and social service systems

 

• Coordinate the FSP and Medicaid with the food donor community at the local level

 

• Coordinate the FSP and Medicaid with the School Lunch Program at the local level

 

• Adopt initiatives that require workers to screen for and refer program applicants to other programs for which they may be eligible

 

• Outstation FSP eligibility workers in clinics, health centers, and other community organizations

 

• Create full-service application sites that are independent of the welfare office

 

 

Several outreach strategies are being used to increase Medicaid/SCHIP enrollment. These strategies tend to focus on enrolling children. Community organizations are being encouraged to help link children to Medicaid by distributing information about the availability of Medicaid and SCHIP for children in low-income families, by incorporating Medicaid income eligibility screening as part of their program=s routine intake procedures, by conducting outreach at special events, and by enlisting the business community in efforts to promote Medicaid (CBPP 1998a). Other outreach strategies for community organizations include using the media, partnering with schools and school clinics, and agreeing to become a Medicaid outstation. The Children=s Defense Fund has developed a community guide for enrolling children in SCHIP and Medicaid (Children=s Defense Fund 1999).

The USDHHS and HCFA have jointly developed two guides to inform child care providers and Head Start centers of the benefits of Medicaid and SCHIP enrollment and to provide examples of linkages that encourage families= enrollment (Child Care and Medicaid: Partners for Healthy Children; and Head Start, Medicaid and CHIP: Partners for Healthy Children). An initiative supported by the Robert Wood Johnson Foundation, Covering Kids, works to help states and communities design and conduct outreach programs, simplify the enrollment process, and coordinate coverage programs. Certain community action groups such as the Children=s Partnership and Philadelphia Citizens for Children and Youth are organizing innovative outreach activities in their communities to increase the enrollment of children in SCHIP and Medicaid. APHSA has recently issued a report describing responses to a 33-state survey about information on outreach strategies for SCHIP (Micky 1999). Finally, information on the FNS internet website outlines ways it can encourage enrollment in SCHIP for children who participate in its own programs, such as the FSP, the School Lunch, School Breakfast, and other feeding programs.

Examples of existing state outreach and enrollment strategies for increasing child participation in Medicaid/CHIP appear in profiles of four states visited by staff from Health Systems Research for the Kaiser Commission on Medicaid and the Uninsured (Schwalberg et al. 1999). For outreach, the study sites used mass media, community-based outreach, and one-on-one outreach. Their eligibility and enrollment strategies included the simplification and abbreviation of application forms, the use of mail-in applications, and expanding points of access in the community. The experiences of the study sites in the implementation of these strategies are also profiled in the report. Implementation issues involved such questions as whether it is best to align SCHIP with Medicaid or to operate it separately, the relative value of media-focused versus community-based outreach approaches, the challenge of reaching low-income working families through employers, and whether to tailor messages to local needs or to send the same message across the state.

Despite these efforts, there is still a need to identify more efficient and strategic methods of reaching eligible uninsured children. A recent report by the Urban Institute indicates that most low-income uninsured children (almost 75 percent) live in families who participate in other government programs--specifically, the National School Lunch Program, WIC, the FSP, or Unemployment Compensation (Kenney, Haley and Ullman 1999). These programs, especially those sponsored by the USDA, may therefore be an important conduit to eligible children not receiving Medicaid. The report notes that states that have attempted to use enrollment in the National School Lunch Program as an outreach tool have faced some challenges, such as the need to protect the confidentiality of lunch program participants and the fact that Medicaid and SCHIP require more information than the lunch program to establish eligibility. If these challenges were met, it is likely that many more eligible children would become insured.

 

D. Ongoing Efforts to Increase Our Understanding of the Issues

A large-scale USDA-funded study of the effect of local office procedures and operations on access and participation in the FSP, which is in the design phase at this time, will be conducted by Abt Associates and its subcontractor Health Systems Research. The project will examine the likelihood that eligible individuals will obtain or continue to obtain food stamps as a function of local office procedures. Data will be collected through surveys of applicants and eligible non-participants, extensive supervisor and caseworker interviews, a telephone survey of local office operations, and case record abstractions. The study will seek to explain program entries and exits and how they are affected by household characteristics, local office practices, and local area characteristics. However, the study does not directly seek to identify promising strategies that could increase participation, nor is it designed to provide technical assistance to localities on this issue.

An additional USDA study sponsored by the Economic Research Service is being conducted by MPR. The study will use FSP administrative data to examine the changes in FSP caseload characteristics associated with different types of state welfare reform programs. A classification scheme for state welfare reform policies will be developed, and a series of tabulations will be used to compare the characteristics of FSP households before and after the implementation of welfare reform. Finally, project staff will perform an econometric analysis of the relationship between welfare reform policies and state food stamp caseload characteristics holding constant the state, year, unemployment rate, and the state=s nonwelfare policies.

The Rockefeller Institute of Government (IOG) is conducting a study of the linkages between TANF and other safety net programs, particularly Medicaid. With support from the W.K. Kellogg Foundation and other funders, IOG staff will examine 20 states in a field network evaluation of the implementation of PRWORA. In addition, the Economic Research Service of the USDA has developed a cooperative agreement with the IOG to add a component to this larger multi-site study to analyze local-level institutional changes in service delivery that could be contributing to the FSP caseload decline. The project will focus on the growing institutional separation between the food stamp and welfare cash assistance systems and the increasing complexity of the programs. In another IOG study, researchers will conduct site visits to examine the front-line management and practices of local TANF offices in Georgia, Michigan, New York City, Texas, and possibly Wisconsin. The study seeks primarily to answer how state and local policies and management systems affect the front-line practices of TANF programs (Nathan and Thompson 1999).

The Center for Health Services and Policy at the George Washington University School of Public Health and Health Services, with support from the Medi-Cal Policy Institute, is conducting a study of the diversionary effects of CalWORKS, California=s TANF program. The study, which will be completed in spring 2000, involves case studies of four counties and includes site visits to welfare offices and structured interviews with welfare staff at various levels as well as with representatives of community-based organizations. The purpose of the study is to identify the range of effective local strategies and state policies for implementing CalWORKS as well as diversion efforts that will ensure adequate and appropriate access to Medi-Cal and Transitional Medi-Cal (California=s transitional Medicaid assistance).

The Center for Health for Health Services and Policy is also conducting final analyses for a 1999 survey of patients and heads of households at 20 community health centers in 10 states in order to understand attitudes toward and barriers to Medicaid participation. A report is due to be released in spring 2000.

In a study sponsored by USDHHS, MPR and its subcontractor, the Urban Institute, are studying the following several issues involving access to and participation in Medicaid and SCHIP:

$ Low-income populations that have or have not enrolled in Medicaid before and after welfare reform; the data source is the NCHS National Health Interview Survey

 

• National trends in Medicaid enrollment for groups affected by welfare reform in all states; the data source is 1991-98 HCFA 2082 data

 

• Changes in the characteristics of Medicaid participants before and after welfare reform in selected states; the data source is 1994 and 1997 State Medicaid Person Summary Research Files (SMRF)

 

• Changes in longitudinal Medicaid enrollment patterns in selected states, including TMA, and continuous and discontinuous enrollment; the data source is 1993-94 and 1996 and 1997 SMRF

 

• Completion of the Five-State Medicaid participation study for the USDHHS Office of the Assistant Secretary for Planning and Evaluation

 

 

VI. Summary and Conclusions

In this final section, we discuss the extent to which issues related to the declines in the FSP and Medicaid caseloads may be similar. In particular, we discuss the relationship between the caseload declines in the two programs, the degree to which the reasons for the declines appear to be similar or different, and whether the potential solutions for improving participation are common to both programs.

 

1. Relationship between FSP and Medicaid Caseload Declines

Our review of the literature indicates that, since welfare reform, the drop in participation has been greater in the FSP than in Medicaid, both overall and at the state level (see Tables 1 and 3). We also note that within each state, the percent change in participation for one program often differs in magnitude and even sometimes in direction from that of the other program. That is, states with large declines in one of the programs (FSP or Medicaid) do not always have similarly large declines in the other program (the same is true for states with small declines or increases). The reasons for this lack of correspondence may be explained by several differences between the two programs, such as the wide variation among states in eligibility policies for Medicaid, and increased Medicaid and SCHIP outreach activities. States have far less flexibility in determining eligibility for food stamps than for Medicaid, and there has been less activity related to FSP outreach than to Medicaid outreach. Furthermore, families may consider Medicaid to be Aworth more@ in light of the high cost of health care and therefore more inclined to enroll.

Focusing on the caseload changes for AFDC/TANF shown in Tables 1 and 3, we further observe that although changes in participation in the FSP and Medicaid tend to mirror changes in the welfare caseload, this is not always the case. For example, the AFDC/TANF caseload in Oregon declined by almost 40 percent, but its Medicaid participation increased by 30 percent over the same period. States with a large drop in the TANF caseload but smaller declines or even increases in Medicaid or FSP participation may suggest strategies and practices that move welfare recipients into work without reducing participation in safety net programs (although an analysis of state policy options should be taken into consideration in determining whether this is the case, particularly with respect to Medicaid).

 

B. Have Declines in the Participation of Both Programs Occurred for Similar or Different Reasons?

Aside from effects of low unemployment and recent policy changes that made certain groups ineligible, there appear to be several similar reasons for the caseload declines in the Medicaid and the Food Stamp programs. The literature suggests that the declines in both programs have occurred to some extent as a result of administrative issues affecting welfare families (welfare leavers, welfare families that are sanctioned because of TANF violations, and families who have reached welfare time limits). Worker and applicant confusion about eligibility rules was observed in both programs (though confusion was more pronounced in the case of Medicaid), and there is evidence that some families may be unaware of their potential continued eligibility after leaving welfare. Factors that may act as barriers to program access for both Medicaid and the FSP involve the costs of participating (e.g., hassles, reporting requirements), and the stigma that may be carried over from TANF receipt.

Although we did not identify any studies examining the effect of increased income on program participation, we noted that changes in the average monthly income of welfare leavers did not appear to make many families ineligible for benefits in either program. However, the literature suggests that people may become ineligible through other means (i.e., as a result of lump-sum TANF diversion payments that are counted as income).

Beyond these similarities, there are reasons for diminished caseloads that are unique to each program. With respect to Medicaid, for instance, there is evidence of an urgent need for welfare workers to be given assistance in understanding complex eligibility rules, and for automated eligibility systems to be brought up to date. We also note that although the rapid transition to employment may inadvertently prevent recipients from meeting eligibility criteria for TMA (i.e., receipt for three months), this is not a factor in FSP eligibility. With respect to the FSP, it appears that promulgating rules particularly with respect to specific areas such as sanctions may be necessary to enforce the lawful implementation of the FSP under welfare reform.

 

 

C. Should Strategies for Improving Participation Be Similar or Different Across the Programs?

Our review of the literature suggests that several strategies may be effective in increasing access to and participation in the FSP and Medicaid that do not involve major changes to state policies or initiatives. The evidence further suggests that strategies that are effective for one program would be likely to work for the other as well. Although research remains to be done to explore the extent to which this may be the case, the findings to date suggest that outreach programs, outstationing eligibility workers, simplifying applications, reviewing closed cases, improving worker knowledge about eligibility rules, and decreasing the need for face-to-face meetings between workers and clients (through increased use of phone and mail, for example) would be helpful in both programs. Our review further suggests that correcting specific administrative practices related to TANF may be particularly important to ensure access and continued participation among eligible welfare applicants, recipients, and leavers. Procedures, practices, and systems involved in TANF pre-application requirements, diversion strategies, sanctions and time limits, as well as issues related to termination of TANF benefits and a clear understanding of the independence of eligibility rules for the three programs are key areas that may need attention.

Going beyond these strategies that are common to both programs, it may be especially important in the case of the FSP to simplify recertification procedures. To increase participation in Medicaid it may be important to take steps to encourage greater use of TMA, to pair welfare and Medicaid eligibility workers in a buddy system, and to focus on improving automated eligibility systems.

REFERENCES

Beebout, Harold. AFluctuations in Food Stamp Program Participation.@ Testimony before the U.S. Senate Committee on Agriculture, Nutrition, and Forestry. Washington, DC, April 23, 1998.

Castner, Laura, and Jacquelyn Anderson. ACharacteristics of Food Stamp Households: Fiscal Year 1998.@ Advance Report. Washington, DC: Mathematica Policy Research, Inc., July 1999.

Castner, Laura, and Scott Cody. ACharacteristics of Food Stamp Households: Fiscal Year 1997.@ Washington, DC: Mathematica Policy Research, Inc., February 1999a.

________. ATrends in FSP Participation Rates: Focus on September 1997.@ Washington, DC: Mathematica Policy Research, Inc., November 1999b.

Center on Budget and Policy Priorities. ATen Ways Community Organizations Can Link Children to Medicaid [and other Public Health Insurance Programs].@ Washington, DC: CBPP, 1998a.

________. ASteps States Can Take to Facilitate Medicaid Enrollment of Children.@ Washington, DC: CBPP, 1998b.

________. AAn Analysis of the AFDC-Related Medicaid Provisions in the New Welfare Law.@ Washington, DC: CBPP, 1996.

Children=s Defense Fund. ACHIP Toolkit: A Community Guide to Enrolling Children in Free and Low-Cost Health Insurance Programs.@ Washington, DC: Children=s Defense Fund, 1999.

Coe, Richard D. AParticipation in the Food Stamp Program, 1979. In Five Thousand American Families--Patterns of Economic Progress, vol. 10, edited by Greg J. Duncan and James N. Morgan. Ann Arbor: University of Michigan, Institute for Social Research, 1983.

Council of Economic Advisors. AThe Effects of Welfare Policy and the Economic Expansion on Welfare Caseloads: An Update.@ Washington, DC: CEA, 1999.

________. ATechnical Report: Explaining the Decline in Welfare Receipt, 1993-1996.@ Washington, DC: CEA, 1997.

Darnell, Julie, Hye Sun Lee, and Jonah Murdock. AMedicaid and Welfare Reform: States= Use of the $500 Million Federal Fund.@ Chicago: Northwestern University, Institute for Health Services Research and Policy Studies, October 1999.

Darnell, Julie, Kathleen Maloy, and Lea Scarpulla-Nolan. AStates= Use of Options under Section 1931 to Account for the Effects of Diversion Under Welfare Reform.@ Washington, DC: George Washington University, Center for Health Policy Research, 1999.

Darnell, Julie, Lea Nolan, and Kathleen Maloy. AThe Impact of Diversion Programs on Medicaid Eligibility: Preliminary Findings Based on Interviews with Officials in 29 States with Diversion Programs.@ Washington, DC: George Washington University, Center for Health Policy Research, 1999.

Dubay, L., and G. Kenney. AEffects of Medicaid Expansions on Insurance Coverage of Children.@ Future of Children, vol. 6, no. 1, 1996, pp. 152-61.

Ellwood, David T., and E. Kathleen Adams. AMedicaid Mysteries: Transitional Benefits, Medicaid Coverage, and Welfare Exits.@ Health Care Financing Review 1990 Annual Supplement, December, 1990.

Ellwood, Marilyn. AThe Medicaid Eligibility Maze: Coverage Expands, but Enrollment Problems Persist. Findings from a Five-State Study.@ Washington, DC: Mathematica Policy Research, Inc., September 1999.

Ellwood, Marilyn, and Carol Irvin. AWelfare Leavers and Medicaid Dynamics: Five States in 1995.@ Washington, DC: Mathematica Policy Research, September 1999.

Ellwood, Marilyn, and Leighton Ku. AWelfare and Immigration Reforms: Unintended Side Effects for Medicaid.@ Washington, DC: The Urban Institute, 1998.

Ellwood, Marilyn, and Kimball Lewis. AOn & Off Medicaid: Enrollment Patterns for California and Florida in 1995.@ Washington, DC: The Urban Institute, 1999.

Figlio, David, and Ziliak, James. AWelfare Reform, the Business Cycle, and the Decline in AFDC Caseloads.@ In Economic Conditions and Welfare Reform, edited by S. Danziger,. Kalamazoo, Michigan: W.E. Upjohn Institute for Employment Research.

Fix, M., and J. Passel. ATrends in Noncitizens= and Citizens= Use of Public Benefits Following Welfare Reform: 1994-1997.@ Washington, DC: The Urban Institute, 1999.

Fraker, Tom, Lucia Nixon, Jan Losby, et al. AIowa=s Limited Benefit Plan: Summary Report.@ Washington, DC: Mathematica Policy Research, Inc., and the Institute for Social and Economic Development, May 1997.

Garrett, Bowen, and John Holahan. AHealth Insurance Coverage and Health Status of Former Welfare Recipients.@ Draft Working Paper. Washington, DC: The Urban Institute, 1999.

Gordon, Anne, Carole Kuhns, Renee Loeffler, et al. AExperiences of Virginia Time Limit Families in the Six Months after Case Closure: Results for an Early Cohort.@ Princeton, NJ: Mathematica Policy Research, forthcoming.

Greenberg, Mark. AParticipation in Welfare and Medicaid Enrollment.@ Washington, DC: Kaiser Family Foundation, 1998.

Gundersen, Craig, Michael LeBlanc, and Betsey Kuhn. AThe Changing Food Assistance Landscape: The Food Stamp Program in a Post-Welfare Reform Environment.@ Washington, DC: U.S. Department of Agriculture, Economic Research Service, 1999.

Guyer, Jocelyn, Matthew Broaddus, and Michelle Cochran. AMissed Opportunities: Declining Medicaid Enrollment Undermines the Nation=s Progress in Insuring Low-Income Children.@ Washington, DC: Center on Budget and Policy Priorities, 1999.

Guyer, Jocelyn, and Cindy Mann. AEmployed but Not Insured: A State-by-State Analysis of the Number of Low-Income Working Parents Who Lack Health Insurance.@ Washington, DC: Center on Budget and Policy Priorities, 1998a.

Guyer, Jocelyn and Cindy Mann. ATaking the Next Steps: States Can Now Expand Health Coverage to Low-Income Working Parents through Medicaid.@ Washington, DC: Center on Budget and Policy Priorities, 1998b.

Hollenbeck, Darrell, and James C. Ohls. AParticipation among the Elderly in the Food Stamp Program.@ The Gerontologist, vol. 24, no. 6, 1984.

Kaiser Commission on Medicaid and the Uninsured. AMedicaid: A Primer.@ Washington, DC: The Henry J. Kaiser Family Foundation, 1999.

Kenney, G.M., Jennifer H., and Ullman, F. AMost Uninsured Children Are in Families Served by Government Programs.@ Washington DC: The Urban Institute, 1999.

Klein, Rachel, and Cheryl Fish-Parcham. ALosing Health Insurance, Unintended Consequences of Welfare Reform.@ Washington, DC: Families USA Foundation, 1999.

Ku, L., and B. Bruen. AThe Continuing Decline in Medicaid Coverage.@ Assessing the New Federalism, Brief A-37. Washington, DC: The Urban Institute, 1999.

Ku, Leighton, and Theresa A. Coughlin. AHow the New Welfare Reform Law Affects Medicaid.@ Assessing the New Federalism, Number A-5. Washington, DC: The Urban Institute, 1999.

Ku, Leighton, Frank Ullman, and Ruth Almeida. AWhat Counts? Determining Medicaid and CHIP Eligibility for Children.@ Washington, DC: The Urban Institute, 1999.

Loprest, Pamela. AFamilies Who Left Welfare: Who Are They and How Are They Doing?@ Discussion Paper DP 99-02. Washington, DC: The Urban Institute, 1999.

Loprest, Pamela, and Sheila Zedlewski. ACurrent and Former Welfare Recipients: How Do They Differ?@ Discussion Paper DP 99-17. Washington, DC: The Urban Institute, 1999.

Lyons, Barbara. AHealth Coverage and Access to Care: Key Issues for Low-Income Children and Their Families.@ Paper presented at the Welfare Reform and Child Outcomes Conference hosted by the Assistant Secretary for Planning and Evaluation and the David and Lucille Packard Foundation, Washington, DC, June 1999.

Maloy, K. A., J. Darnell, and S. Rosenbaum. AWork Instead of Welfare: States= Responses to the New Welfare Paradigm and the Implications for Access to Medicaid.@ Submitted to Health Affairs, August 1999.

Maloy, Kathleen, LaDonna Pavetti, Julie Darnell, et al. ADiversion as a Work-Oriented Welfare Reform Strategy and Its Effect on Access to Medicaid: An Examination of the Experiences of Five Local Communities.@ Washington, DC: George Washington University, Center for Health Policy Research, 1999.

Maloy, Kathleen, LaDonna Pavetti., Peter Shin, et al. AA Description and Assessment of State Approaches to Diversion Programs and Activities under Welfare Reform.@ Washington, DC: George Washington University, Center for Health Policy Research, 1998.

Mann, Cindy. AThe Ins and Outs of Delinking: Promoting Medicaid Enrollment of Children Who Are Moving in and out of the TANF System.@ Washington, DC: Center on Budget and Policy Priorities, 1999.

Mann, Cindy, and Jocelyn Guyer. AThe President=s Budget Includes New State Options for Reducing the Number of Uninsured Children.@ Washington, DC: Center on Budget and Policy Priorities, 1998.

Martini, A., and M. Wiseman. AExplaining the Recent Decline in Welfare Caseloads: Is the Council of Economic Advisors Right?@ Washington, DC: The Urban Institute, 1997.

McConnell, Sheena, and Michael Ponza. AThe Reaching the Working Poor and Poor Elderly Study: What We Learned and Recommendations for Future Research.@ Washington, DC: Mathematica Policy Research, Inc., 1999.

Mickey, Michelle. ACHIP Outreach and Enrollment: A View from the States.@ Washington, DC: American Public Human Services Association, 1999.

Moffitt, R. A., and E. Slade. AHealth Care Coverage for Children On and Off Welfare.@ The Future of Children, no.1, 1997, p. 87-98.

Moore, Judith D. AWelfare Reform and Its Impact on Medicaid.@ Issue Brief/No. 732. Washington, DC: National Health Policy Forum, 1999.

Nathan, Richard, and Frank Thompson. AA Preliminary Analysis of TANF-Social Safety Net Linkages: Especially for Medicaid.@ National Academy of Social Insurance 11^th Annual Conference. Washington, DC: Rockefeller Institute of Government, 1999.

Nord, M., K. Jemison, and G. Bickel. APrevalence of Food Insecurity and Hunger by State, 1996-1998.@ Washington, DC: Food and Nutrition Service, U.S. Department of Agriculture, 1999.

Parrott, Sharon. AWelfare Recipients Who Find Jobs: What Do We Know About Their Employment and Earnings?@ Washington, DC: Center on Budget and Policy Priorities, 1998.

Pavetti, LaDonna, and Nancy Wemmerus. AFrom a Welfare Check to a Paycheck: Creating a New Social Contract.@ Journal of Labor Research, vol. XX, no.4, pp. 517-537, 1999.

Peller, J., and H Shaner. AMedicaid Eligibility Standards for Low-Income Families and Children.@ Washington, DC: American Public Welfare Association, May 1998.

Perry, M., S. Kannel, R. B. Valdez, et al. AMedicaid and Children: Overcoming Barriers to Enrollment, Findings from a National Survey.@ Washington, DC: Henry J. Kaiser Family Foundation, January 2000.

Ponza, Michael, James Ohls, Lorenzo Moreno, et al. ACustomer Service in the Food Stamp Program.@ Washington, DC: Mathematica Policy Research, Inc., July 1999.

Primus, Wendell, Lynette Rawlings, Kathy Larin, et al. AThe Initial Impacts of Welfare Reform on the Incomes of Single-Mother Families.@ Washington, DC: Center on Budget and Policy Priorities, 1999.

Rangarajan, Anu, and Robert G. Wood. AWork First New Jersey Evaluation: How WFNJ Clients Are Faring under Welfare Reform: An Early Look.@ Princeton, NJ: Mathematica Policy Research, October 1999.

Ross, Donna C., and Jocelyn Guyer. ACongress Lifts the Sunset on the >$500 Million Fund=, Extends Opportunities for States to Ensure Parents and Children Do Not Lose Health Coverage.@ Washington, DC: Center on Budget and Policy Priorities, 1999.

Schneider, Andy, Kristen Fennel, and Peter Long. AMedicaid Eligibility for Families and Children.@ Washington, DC: Kaiser Commission on Medicaid and the Uninsured, 1998.

Schwalberg, R., I. Hill, H. Bellamy, and J. Gallagher. AMaking Child Health Coverage a Reality: Lessons from Case Studies of Medicaid and CHIP Outreach and Enrollment Strategies.@ Washington, DC: Health Systems Research, Inc., 1999.

Shuptrine, Sarah C., Vicki C. Grant, and Genny G. McKenzie. ASouthern Regional Initiative to Improve Access to Benefits for Low-Income Families with Children.@ Columbia, SC: Southern Institute on Children and Families, February 1998.

________. AA Study of the Relationship of Health Coverage to Welfare Dependency.@ Columbia, SC: Southern Institute on Children and Families, 1994.

Shuptrine, Sarah C., and Genny G. McKenzie. AInformation Outreach to Reduce Welfare Dependency: A North Carolina Welfare Reform Initiative.@ Columbia, SC: Southern Institute on Children and Families, May 1996.

Smith, Vernon, Robert Lovell, Karin Peterson, et al. AThe Dynamics of Current Medicaid Enrollment Changes: Insights from Focus Groups of State Human Service Administrators, Medicaid Eligibility Specialists, and Welfare Analysts.@ Washington, DC: The Kaiser Commission on Medicaid and the Uninsured, 1998.

Stuber, J., K. A. Maloy, and S. Rosenbaum. AEligible but Not Enrolled: Stigma and Other Barriers to Medicaid Participation.@ Washington, DC: George Washington University, Center for Health Policy Research, 1999 (forthcoming).

Ullman, Frank, Ian Hill, and Ruth Almeida. ACHIP: A Look at Emerging State Programs.@ Washington, DC: The Urban Institute, 1999.

U.S. General Accounting Office. Food Stamp Program: Various Factors Have Led to Declining Participation. Washington, DC: GAO, 1999a.

________. Medicaid Enrollment: Amid Declines, State Efforts to Ensure Coverage After Welfare Reform Vary. Washington, DC: GAO, 1999b.

________. Demographics of Nonenrolled Children Suggest State Outreach Strategies. Washington, DC: GAO, 1998.

U.S. Department of Agriculture. Characteristics of Food Stamp Households: Fiscal Year 1998 (Advance Report). Washington, DC: USDA, July 1999a.

________. Who is Leaving the Food Stamp Program? An Analysis of Caseload Changes from 1994 to 1997. Washington, DC: USDA, 1999b.

________. The Nutrition Safety Net: At Work for Families. Washington, DC: USDA, 1999c.

________. Household Food Security in the United States 1995-1998 (Advance Report). Washington, DC: USDA, 1999d.

U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research, and Evaluation. Characteristics and Financial Circumstances of TANF Recipients: FY 1998. Washington, DC: USDHHS, 1999.

U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research and Evaluation. Characteristics and Financial Circumstances of AFDC Recipients: FY 1996. Washington, DC: USDHHS, 1997.

U.S. Department of Health and Human Services, Administration for Children and Families, and the Health Care Financing Administration. Supporting Families in Transition: A Guide to Expanding Health Coverage in the Post-Welfare Reform World. Washington, DC: USDHHS, 1999a.

________. Head Start, Medicaid and CHIP: Partners for Healthy Children. Washington, DC:USDHHS, 1999b.

________. Child Care and Medicaid: Partners for Healthy Children. Washington, DC: USDHHS, 1999c.

Wallace, Geoffrey, and Rebecca Blank. AWhat Goes Up Must Come Down? Explaining Recent Changes in Public Assistance Caseloads.@ Chicago: Northwestern University, 1999.

Wooldridge, Judith, and Sheila Hoag. AMedicaid=s Role in Encouraging Transitions from Welfare to Work.@ Princeton, NJ: Mathematica Policy Research, Inc., October 1996.

Zedlewski, Sheila R., and Sarah Brauner. AAre the Steep Declines in Food Stamp Participation Linked to Falling Welfare Caseloads?@ Washington, DC: The Urban Institute, 1999.

________. ADeclines in Food Stamp and Welfare Participation: Is There a Connection?@ Discussion Paper DP 99-13. Washington, DC: The Urban Institute, 1999.

Ziliak, James, and David Figlio, Elizabeth Davis, and Laura Connolly. AAccounting for the Decline in AFDC Caseloads: Welfare Reform or Economic Growth?@ Discussion Paper 1151-97. University of Wisconsin-Madison, Institute for Research on Poverty, 1999.

 

APPENDIX A

PERCENT CHANGE IN FOOD STAMP AND TANF ENROLLMENT

1996-1998

State	Food Stamp Program 8/96 - 8/98 Percent Change	TANF 1/96 B12/98 Percent Change
Alabama	-17.7	-54.3
Alaska	-10.3	-28.1
Arizona	-35.0	-43.9
Arkansas	-7.8	-48.3
California	-30.7	-30.1
Colorado	-23.3	-58.2
Connecticut	-15.9	-39.7
Delaware	-27.1	-46.8
District of Columbia	-9.3	-23.7
Florida	-29.8	-60.5
Georgia	-22.0	-57.9
Hawaii	-6.9	-31.9
Idaho	-26.1	-86.7
Illinois	-19.6	-37.5
Indiana	-18.9	-22.7
Iowa	-23.0	-34.7
Kansas	-30.6	-54.2
Kentucky	-15.6	-40.7
Louisiana	-18.3	-46.5
Maine	-15.0	-34.5
Maryland	-15.7	-52.0
Massachusetts	-25.1	-37.9
Michigan	-18.0	-47.9
Minnesota	-24.2	-19.7
Mississippi	-30.2	-67.3
Missouri	-25.3	-42.0
Montana	-12.1	-50.4
Nebraska	-4.9	-9.9
Nevada	-29.4	-42.9
New Hampshire	-29.7	-35.2
New Jersey	-24.0	-38.8
New Mexico	-23.1	-21.5
New York	-22.1	-38.8
North Carolina	-16.3	-21.5
North Dakota	-12.2	-30.6
Ohio	-30.7	-47.3
Oklahoma	-16.0	-38.8
Oregon	-20.2	-42.1
Pennsylvania	-19.4	-49.7
Rhode Island	-40.2	-52.1
South Carolina	-9.9	-41.1
South Dakota	-8.5	-10.7
Tennessee	-16.0	-59.4
Texas	-33.2	-46.8
Utah	-17.0	-43.8
Vermont	-48.1	-53.7
Virginia	-28.8	-27.4
Washington	-30.4	-35.4
West Virginia	-12.2	-72.0
Wisconsin	-33.6	-81.6
Wyoming	-24.4	-85.9
US Totals	-24.1%	-41.0%

Source: Food Stamp Program Quality Control data, as cited in GAO 1999a; USDHHS

Administration for Children and Families.

APPENDIX B

PERCENT CHANGE IN MEDICAID ENROLLMENT FOR THE NONELDERLY, NONDISABLED POPULATION AND IN THE AFDC/TANF CASELOAD

1995-1997

State	Medicaid 1995-1997 Percent Change	AFDC/TANF 1995-1997 Percent Change
Alabama	-2.7	-27.1
Alaska	-7.0	-4.1
Arizona	-3.5	-22.5
Arkansas	2.9	-16.0
California	-2.1	-10.3
Colorado	-9.3	-26.9
Connecticut	0.5	-9.6
Delaware	23.9	-11.2
District of Columbia	-1.6	-0.1
Florida	-11.1	-27.4
Georgia	-2.9	-26.3
Hawaii	-3.4	8.4
Idaho	-7.6	-32.6
Illinois	-4.9	-16.6
Indiana	-15.8	-37.9
Iowa	-7.4	-22.1
Kansas	-11.7	-32.5
Kentucky	-7.5	-16.7
Louisiana	-9.2	-25.4
Maine	-5.6	-17.5
Maryland	-8.4	-27.0
Massachusetts	-1.1	-24.9
Michigan	-9.4	-24.9
Minnesota	6.8	-13.1
Mississippi	-8.7	-28.9
Missouri	-6.6	-22.4
Montana	-9.8	-14.5
Nebraska	13.7	-6.0
Nevada	-19.3	-27.9
New Hampshire	-0.8	-29.4
New Jersey	-6.9	-20.7
New Mexico	14.9	-21.4
New York	-7.1	-16.6
North Carolina	-2.4	-22.4
North Dakota	-7.4	-21.4
Ohio	-18.4	-19.3
Oklahoma	-8.4	-33.8
Oregon	29.5	-39.9
Pennsylvania	-7.0	-22.8
Rhode Island	-7.4	-11.1
South Carolina	7.8	-30.3
South Dakota	1.5	-21.6
Tennessee	-6.5	-33.2
Texas	-7.0	-23.4
Utah	-15.4	-25.8
Vermont	18.1	-15.4
Virginia	-6.8	-29.4
Washington	12.1	-11.3
West Virginia	-21.0	-21.8
Wisconsin	-18.6	-42.5
Wyoming	-12.8	-50.0
US Totals	-5.3	-20.0

Source: Health Care Financing Agency 2082 data, as edited by the Urban Institute; and AFDC/TANF data from ACF-3637, Statistical Report on Recipients Under Public Assistance, as reported in Ku and Bruen (1999).

Note: Medicaid data reflect average monthly participation levels for nonelderly, nondisabled adults and children.

 

APPENDIX C SELECTED RESEARCH ON THE FOOD STAMP PROGRAM
Author(s)	Study	Date of Report	Data Source	Time Period Covered
Bartlett, S., Hamilton, C., Burnstein, N. Abt Associates Gabor, V. & Botsko, C. Health Systems Research	A Study of Food Stamps Access and Declining Program Participation	Oct 1999 (proposal)	New telephone surveys; in-person interviews; case record abstractions	2000-2001
Beebout, H.S. Mathematica Policy Research Testimony to U.S. Senate	Fluctuations in Food Stamp Program Participation	Apr 1998	FNS data	1988-1998
Bickel, G. and Carlson, S. Food and Nutrition Service; Economic Research Service	Household Food Security in the United States: 1995-1998 (Advance Report)	Jul 1999	Food Security Supplement to the CPS 4/95; 9/96; 4/97/ 8/98	1995 - 1998
Castner, L. & Anderson, J. Mathematica Policy Research	Characteristics of Food Stamp Households: Fiscal Year 1998 (Advance Report)	Jul 1999	Food Stamp Program Quality Control data	FY 1998 1989-1999
Castner, L. & Cody, S. Mathematica Policy Research	Characteristics of Food Stamp Households: Fiscal Year 1997	Feb 1999	Food Stamp Program Quality Control data	FY 1997 1994-1997
Castner, L. & Cody, S. Mathematica Policy Research	Trends in FSP Participation Rates: Focus on September 1997	Nov 1999	March CPS; FSP Admin data, Food Stamp Program Quality Control data	1996-1997 1994-1997
Food and Nutrition Service U.S.D.A.	The Nutrition Safety Net: At Work for Families	1999	N/A	N/A
Food and Nutrition Service, U.S.D.A.	Food Stamp Program: Number of Persons Participating (Preliminary Data as of July 22, 1999)	Aug 1999	Food Stamp Program Operations data	May 1998-May 1999
Fraker, T., Nixon, L., Losby, J., Prindle, C., & Else, J. Mathematica Policy Research	Iowa=s Limited Benefit Plan: Summary Report	May 1997	Welfare Administrative Records, Survey of 137 Households, Case Studies of 12 Families	1995-1996
Genser, J., Office of Analysis, Nutrition, and Evaluation, Food and Nutrition Service U.S.D.A.	Who is Leaving the Food Stamp Program? An Analysis of Caseload Changes from 1994 to 1997	Mar 1999	Food Stamp Program Quality Control data	1994; 1997
Gordon, A., Kuhns, C., Loeffler, R., & Agodini, R. Mathematica Policy Research	Experiences of Virginia Time Limit Families in the Six Months After Case Closure: Results for an Early Cohort	Nov 1999	Administrative Data and Follow-up Surveys with Families	1998 Cohort of Families Reaching the Time Limit
Gundersen, C., LeBlanc, M., & Kuhn, B. Food and Rural Economics Division Economic Research Service, U.S.D.A.	The Changing Food Assistance Landscape: The Food Stamp Program in a Post-Welfare Reform Environment	Mar 1999	FNS data	1996-1998 1964-1997
Loprest, P.J. & Zedlewski, S.R. The Urban Institute	Current and Former Welfare Recipients: How Do They Differ?	1999	National Survey of America=s Families	1995-1997
Loprest, P., The Urban Institute	Families Who Left Welfare: Who Are They and How Are They Doing?	1999	National Survey of America=s Families	1995-1997
McConnell, Sheena, and M. Ponza Mathematica Policy Research, Inc.	The Reaching the Working Poor and Poor Elderly Study: What We Learned and Recommendations for Future Research	Dec 1999	Random digit-dialing study of FSP nonparticipants, and focus groups
Nord, M., Jemison, K. & Bickel, G. Economic Research Service Food and Nutrition Service U.S.D.A.	Prevalence of Food Insecurity and Hunger by State, 1996-1998	1999	CPS Food Security Supplement 9/96, 4/97, 8/98	1996-1998
Parrott, S. Center for Budget and Policy Priorities	Welfare Recipients Who Find Jobs: What Do We Know About Their Employment and Earnings?	Nov 1998	Review of TANF Leaver Studies	1994-1998
Ponza, M., Ohls, J.C., Moreno, L. Zambrowski, A., & Cohen, R. Mathematica Policy Research	Customer Service in the Food Stamp Program	Jul 1999	National Food Stamp Program Survey	1996-1997
Primus, W., Rawlings, L., Larin, K., and Porter, K. Center for Budget and Policy Priorities	The Initial Impacts of Welfare Reform on the Incomes of Single-Mother Families	Aug 1999	CPS	1993-1995 1995-1997
Rangarajan, A. & Wood, R. G. Mathematica Policy Research	Work First New Jersey Evaluation: How WFNJ Clients Are Faring Under Welfare Reform	Oct 1999	Household Longitudinal Surveys and Administrative Data	1997-1999
U. S. Department of Health and Human Services, Administration for Children and Families	Characteristics and Financial Circumstances of AFDC Recipients: FY 1996	1997	Administrative Data	FY 1996
U. S. Department of Health and Human Services, Administration for Children and Families	Characteristics and Financial Circumstances of AFDC Recipients: FY 1998	1999	Administrative Data	FY 1996
U.S. General Accounting Office	Food Stamp Program: Various Factors Have Led to Declining Participation	Jul 1999	GAO survey of states; survey of FNS regional offices; FNS reviews; FNS participation data; CPS data	1996-1999 1989-1998
Wallace, G., & Blank, R.M. Northwestern University and Council of Economic Advisers	What Goes Up Must Come Down? Explaining Recent Changes in Public Assistance Caseloads	Feb 1999	CPS; state unemployment data; FNS participation data	1994-1998
Yelowitz, Aaron S. University of California, Los Angeles	Did Recent Medicaid Reforms Cause the Caseload Explosion in the Food Stamp Program?	Jul 1996	Survey of Income and Program Participation	1987-1995
Zedlewski, S.R. & Brauner S. The Urban Institute	Declines in Food Stamp and Welfare Participation: Is There a Connection?	1999	National Survey of America=s Families	1995-1997

APPENDIX D SELECTED RESEARCH ON THE MEDICAID PROGRAM
Author(s)	Study	Date of Report	Data Source	Time Period Covered
Center on Budget & Policy Priorities	An Analysis of the AFDC-Related Medicaid Provisions in the New Welfare Law	Sep 1996	N/A	N/A
Center on Budget & Policy Priorities	Steps States Can Take to Facilitate Medicaid Enrollment of Children	Nov 1998	N/A	N/A
Center on Budget & Policy Priorities	Ten Ways Community Organizations Can Link Children to Medicaid [and other Public Health Insurance Programs]	Nov 1998	N/A	N/A
Darnell, J., Northwestern University, Maloy, K.A., George Washington University, Scarpulla-Nolan, L.	States= Use of Options Under Section 1931 to Account for the Effects of Diversion Under Welfare Reform	1999	Telephone survey of 29 states	1998
Darnell, J., Northwestern University, Lee, H.S., and Murdock, J.	Medicaid and Welfare Reform: States= Use of the $500 Million Federal Fund	October 1999	Telephone survey of 40 states
Dubay, Lisa and Kenney, Genevieve The Urban Institute	Effects of Medicaid Expansions on Insurance Coverage of Children. Future of Children, vol. 6, no. 1, pp. 152-61	1996	CPS-based microsimulation	1989;1994
Ellwood, M. and Lewis, K. Mathematica Policy Research	On & Off Medicaid: Enrollment Patterns for California and Florida in 1995	Jul 1999	HCFA State Medicaid Research Files	1995
Ellwood, D., and Adams, E. K. Mathematica Policy Research	Medicaid Mysteries: Transitional Benefits, Medicaid Coverage, and Welfare Exits	1990	HCFA	1980-1986
Ellwood, M., and Ku, L. Mathematica Policy Research and the Urban Institute	Welfare and Immigration Reforms: Unintended Side Effects for Medicaid	May/Jun 1998	HCFA	1995-1997
Ellwood, M., and Irvin, C. Mathematica Policy Research	Welfare Leavers and Medicaid Dynamics: Five States in 1995	Septem ber 24, 1999	State Medicaid Research Files (SMRF) from HCFA	1995
Ellwood, M. Mathematica Policy Research	The Medicaid Eligibility Maze: Coverage Expands, but Enrollment Problems Persist Findings from a Five State Study	Sep 1999	Site visits to five states	Site visits conducted 1998-1999
Fix, M. and Passel, J.	Trends in Noncitizens= and Ciritizens= Use of Public Benefits Following Welfare Reform: 1994-1997	1999	HCFA; various other sources	1994-1997
Fraker, T. and L. Nixon, J. Losby, C. Prindle, and J. Else Mathematica Policy Research, Inc., and the Institute for Social and Economic Development	Iowa=s Limited Benefit Plan	1997	Iowa DHS records	1994-1995
Garrett, Bowen, and Holahan, John The Urban Institute	Health Insurance Coverage and Health Status of Former Welfare Recipients: Draft Working Paper	1999	NSAF	1997
General Accounting Office GAO/HEHS-99-163	Medicaid Enrollment: Amid Declines, State Efforts to Ensure Coverage After Welfare Reform Vary	Sep 1999	HCFA, state surveys, and site visits to 21 states	1995-1999
Greenberg, Mark Kaiser Commission on Medicaid and the Uninsured	Participation in Welfare & Medicaid Enrollment	Sep 1998	Various state TANF leaver and sanction studies	1996-1998
Guyer, J., and Mann, C. Center on Budget & Policy Priorities	Employed But Not Insured A State-by-State Analysis of the Number of Low-Income Working Parents Who Lack Health Insurance	Mar 1999	CPS	1998
Guyer, J. and Mann, C. Center on Budget & Policy Priorities	Taking the Next Steps: States Can Now Expand Health Coverage to Low-Income Working Parents Through Medicaid	1998	N/A	N/A
Guyer, J., Broaddus, M., Cochran, M. Center on Budget & Policy Priorities	Missed Opportunities: Declining Medicaid Enrollment Undermines the Nation=s Progress in Insuring Low-Income Children	Oct 1999	CPS	1996-1998
Kaiser Commission on Medicaid and the Uninsured	Medicaid: A Primer	Aug 1999	Various sources	1997
Kenney, G.M., Haley, J.M., and Ullman, F. The Urban Institute	Most Uninsured Children Are in Families Served by Government Programs	Dec 1999	National Survey of America=s Families, 1997 data	1996-1997
Klein, R. & Fish-Parcham, C. Families USA	Losing Health Insurance, Unintended Consequences of Welfare Reform	May 1999	CPS and HCFA	1996 & 1998
Ku, L., Ullman, F., Almeida, R. Urban Institute	What Counts? Determining Medicaid and CHIP Eligibility For Children	May 1999	Variety of Sources
Ku, L., & Coughlin, T. Urban Institute	How the New Welfare Reform Law Affects Medicaid		N/A	N/A
Ku, L., and Bruen, B. The Urban Institute	The Continuing Decline in Medicaid Coverage. Assessing the New Federalism Brief A-37.	1999	HCFA 2082 data, edited using Unemployment Insurance data	1995-1998
Lazarus, W.	Letter to Donna Shalala detailing the Express Lane Eligibility	Jun 1999	N/A	N/A
Lyons, B. Kaiser Commission on Medicaid and the Uninsured	Presentation on - Health Coverage and Access to Care: Key Issues for Low-Income Children and Their Families	Jun 1999	Various	1994-1997
Maloy, K.A., George Washington University, Pavetti, L., Mathematica Policy Research, Shin, P. George Washington University & Darnell, J., Northwestern University	A Description and Assessment of State Approaches to Diversion Programs and Activities Under Welfare Reform	1998	Survey of states	1998
Maloy, K.A., George Washington University, Darnell, J., Northwestern University, Rosenbaum, S.	Work Instead of Welfare: States= Responses to the New Welfare Paradigm and the Implications for Access to Medicaid (Submitted to Health Affairs in August 1999)	1999	Survey of states	1998
Maloy, K., George Washington University, Pavetti, L., Mathematica Policy Research, Darnell, J., Northwestern University, Shin, P.	Diversion as a Work-Oriented Welfare Reform Strategy and its Effect on Access to Medicaid: An Examination of the Experiences of Five Local Communities	Mar 1999	Site visits to five areas	1996-1998
Mann, C. Center on Budget & Policy Priorities	The Ins & Outs of Delinking: Promoting Medicaid Enrollment of Children Who are Moving In & Out of the TANF System	Mar 1999	N/A	N/A
Mickey, M. American Public Human Services Association	CHIP Outreach and Enrollment: A View from the States	Sep 1999	Survey of 33 states	1999
Moffitt, R. A. & Slade, E.	Health Care Coverage for Children On and Off Welfare	1997	NLSY
Moore, J. D.	Welfare Reform and Its Impact on Medicaid	1999	Variety of Sources	1994-1998
Nathan, R. & Thompson, F. Rockefeller Institute of Government	A Preliminary Analysis of TANF-Social Safety Net Linkages: Especially for Medicaid	Jan 1999	CPS, HCFA	1995-1997
National Health Policy Forum Issue Brief No. 732	Welfare Reform and Its Impact on Medicaid: An Update	Feb 1999	HCFA, various studies	1995-1998
Perry, M. & Kannel, S. Lake Snell Perr;y & Associates R.B. Valdez, UCLA School of Public Health C. Chang, the Kaiser Commission on Medicaid and the Uninsured	Medicaid and Children: Overcoming Barriers to Enrollment, Findings from a National Survey	Jan 2000	Telephone survey and focus groups of low-income parents	198-1999
Peller, J., American Public Welfare Association, and Shaner, H., Health Care Financing Administration	Medicaid Eligibility Standards for Low-Income Families & Children	May 1998	Various sources	1997
Primus, W., Rawlings, L., Larin, K., and Porter, K. Center for Budget and Policy Priorities	The Initial Impacts of Welfare Reform on the Incomes of Single-Mother Families	Aug 1999	CPS	1993-1995 1995-1997
Rangarajan, A. & Wood, R. G. Mathematica Policy Research	Work First New Jersey Evaluation: How WFNJ Clients Are Faring Under Welfare Reform	Oct 1999	Household Longitudinal Surveys and Administrative Data	1997-1999
Ross, D.C. & Guyer, J. Center for Budget & Policy Priorities	Congress Lifts the Sunset on the >$500 Million Fund= Extends Opportunities for States to Ensure Parents and Children Do Not Lose Health Coverage.	Dec 1999	Various Sources	1996-1999
Schneider, A., Fennel, K., Long, P., Kaiser Commission on Medicaid and the Uninsured	Medicaid Eligibility for Families & Children	Sep 1998	Various Sources	1995-1997
Schott, L. & Mann, C. Center on Budget & Policy Priorities	Assuring that Eligible Families Receive Medicaid When TANF Assistance is Denied or Terminated	Nov 1998	N/A	N/A
Schwalberg, I. Hill, H. Bellamy, and J. Gallagher Health Systems Research, Inc.	Making Child Health Coverage a Reality: Lessons from Case Studies of Medicaid and CHIP Outreach and Enrollment Strategies	Sept. 1999	Case studies of 4 states= outreach strategies
Shuptrine, S. C., Grant, V.C., & McKenzie, G. G., Southern Institute on Children and Families	Information Outreach to Reduce Welfare Dependency: A North Carolina Welfare Reform Initiative	May 1996	Interviews with welfare recipients in North Carolina	1995-1996
Shuptrine, S., Grant, V., McKenzie, G. Southern Institute on Children and Families	Southern Regional Initiative to Improve Access to Benefits for Low Income Families With Children	Feb 1998	Site visits to 17 southern states and the District of Columbia	1997
Smith, V., Lovell, R., Peterson, K. Health Management Association, O=Brien, M., The Lewin Group	The Dynamics of Current Medicaid Enrollment Changes: Insights from Focus Groups of State Human Service Administrators, Medicaid Eligibility Specialists, & Welfare Analysts	Oct 1998	APHSA and Eligibility TAG focus group discussions	1998
Stuber, J., Maloy, K.A. and Rosenbaum, S., George Washington University	Eligible But Not Enrolled: Stigma and Other Barriers to Medicaid Participation	Forthcoming, 1999	A survey of community health center patients/heads of households	May-October 1999
U. S. Department of Health and Human Services, Administration for Children and Families	Characteristics and Financial Circumstances of AFDC Recipients: FY 1996	1997	Administrative data	FY 1996
U. S. Department of Health and Human Services, Administration for Children and Families, Health Care Financing Administration	Supporting Families in Transition: A Guide to Expanding Health Coverage in the Post-Welfare Reform World	1999	N/A	N/A
U. S. Department of Health and Human Services, Administration for Children and Families	Characteristics and Financial Circumstances of AFDC Recipients: FY 1998	1999	Administrative data	FY 1998
U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation	Understanding Estimates of Uninsured Children: Putting the Differences in Context	Jan 1999	NHIS, CPS, SIPP, MEPS, NHIS	N/A
U.S. Department of Agriculture, Food and Nutrition Service	Children=s Health Insurance Program: FNS Outreach Strategies		N/A	N/A
U.S. General Accounting Office	Medicaid Enrollment: Amid Declines, State Efforts to Ensure Coverage After Welfare Reform Vary	Sep 1999	HCFA data in 50 states; Review of state policies and practices in 21 states	1995-1997
Ullman, F., Hill, I., & Almeida, R., The Urban Institute	CHIP: A Look at Emerging State Programs	Sep 1999	State CHIP plans	1999
Wooldridge, J. and Hoag, S. Mathematica Policy Research	Medicaid=s Role in Encouraging Transitions from Welfare to Work	Oct 1996	Literature Review	1985-1996

APPENDIX E SELECTED STATE/LOCAL WELFARE LEAVERS STUDIES
Author(s)	Study	Date	Data Type	Sample Size	Response Rate	Time Period
Arizona Department of Economic Security	Arizona Cash Assistance Exit Study: Cases Exiting Fourth Quarter 1996 Arizona Cash Assistance Exit Study: Cases Exiting First Quarter 1998	Jul 1999   May 1999	Admin   Admin	N=9,439   N=10,646	n/a   n/a	Jul 1996 - Dec 1997 Oct 1997- Mar 1999
Berkeley Planning Associates, Rebecca London, Stephen Walsh, Courtney Smith, and Marlene Strong	Evaluation of Arkansas=s Transitional Employment Assistance (TEA) Program	July 1999	Admin	N=34,091	n/a	Jun 1996 - Oct 1998
Center for Applied Research, Millsaps College: Jessie Beeler, Bill Brister, Sharon Chambry, and Anne McDonald	Tracking of TANF Clients: First Report of a Longitudinal Study C Mississippi=s Temporary Assistance for Needy Families Program	Jan 1999	Admin Survey	N=351	87%	Jun 1998- Oct 1998
Center on Urban Poverty and Social Change, Claudia Coulton; Manpower Demonstration, Nandita Verma	Employment and Return to Public Assistance Among Single, Female-Headed Families Leaving AFDC in Third Quarter 1996, Cuyahoga County, Ohio	May 1999	Admin	N=2,794	n/a	Oct 1996-Dec 1997
Hudson Institute, Rebecca Swartz and Jay Hein; Mathematica Policy Research, Inc., Jacqueline Kauff, Lucia Nixon, Tom Fraker, and Susan Mitchell	Converting to Wisconsin Works: Where Did Families Go When AFDC Ended in Milwaukee?	1999	Survey	N=296	74%	Oct 1998-Feb 1999
Institute for Research on Poverty, University of Wisconsin: Maria Cancian, Robert Haveman, Thomas Kaplan, and Barbara Wolfe	Post-Exit Earnings and Benefit Receipt Among Those Who Left AFDC in Wisconsin	Jan 1999	Admin	N=26,047	n/a	Aug 1995-Dec 1997
Manpower Demonstration Research Corporation, JoAnna Hunter-Manns and Dan Bloom	Connecticut Post-Time Limit Tracking Study: Six-Month Survey Results	Jan 1999	Survey	N=373	82%	1/98-4/98 4/98-7/98
Mathematica Policy Research, Thomas Fraker, Lucia Nixon, Jan Losby, Carol Prindle, and John Else Mathematica Policy Research, Lucia Nixon, Jacqueline Kauff, and Jan Losby	Iowa=s Limited Benefit Plan: Summary Report     Second Assignments to Iowa=s Limited Benefit Plan	May 1997     Aug 1999	Admin Survey   Admin Survey	N=137     N=185	85%     76%	Feb 1996-Mar 1996   May 1998- Aug 1998
Mathematica Policy Research, Anne Gordon, Carole Kuhns, Renee Loeffler, and Roberto Agodini	Experiences of Virginia Time Limit Families in the Six Months After Case Closure: Results for an Early Cohort	Forth-coming	Survey	N=256	78%	Aug 1998-Feb 1999
Missouri Department of Social Services; University of Missouri	Report on the Status of Temporary Assistance to Needy Families (TANF) Recipients: Preliminary Outcomes for 1996 Fourth Quarter AFDC Leavers	Apr 1999	Admin	N=11,373	n/a	Oct 1996- Dec 1998
Rockefeller Institute; New York State Office of Temporary and Disability Assistance; New York State Department of Labor	After Welfare: A Study of Work and Benefit Use After Case Closing	Jul 1999	Admin	N=8,983	n/a	Jan 1997- Mar 1998
South Carolina Department of Social Services	Survey of Former Independence Program Clients: Cases Closed During July Through September 1997 Survey of Former Independence Program Clients: Cases Closed During October through December, 1997 Survey of Former Independence Program Clients: Cases Closed During January through March, 1998 The Post-Welfare Progress of Sanctioned Clients: A Study Using Administrative and Survey Data to Answer Three of Four Important Questions	Oct 1998   Feb 1999     Jun 1999   Nov 1999	Survey   Survey     Survey   Admin Survey	N=403   N=429     N=384   N=15,412 N=2,500	76%   80%     75%   n/a 75-80%	June 1998-Aug 1998 Sep 1998-Dec 1998   Jan 1999-Apr 1999 Oct 1996-Mar 1997
SPHERE Institute; County of San Mateo Human Services Agency (California)	Examining Circumstances of Individuals and Families Who Leave TANF: Assessing the Validity of Administrative Data (Interim Report)	May 1999	Admin	N=2,402	n/a	Jan 1996-Dec 1996
United States General Accounting Office (GAO)	Welfare Reform: States= Early Experiences with Benefit Termination	May 1997	Admin	Iowa N=408 MA N=636 WI N=651	n/a n/a n/a	Aug 1996 Sept 1996 Oct 1996
Washington State Department of Social and Health Services	A Study of Washington State TANF Departures and Welfare Reform Washington=s TANF Single Parent Families After Welfare	Apr 1999   Jan 1999	Admin   Survey	N=105,165 N=91,244 N=592	n/a n/a 52%	10/95-12/97 10/96-12/98 Oct 1998-Nov 1998
Wisconsin Department of Workforce Development	Wisconsin W-2 Works: Those Leaving AFDC or W-2 January to March 1998: Preliminary Report	Jan 1999	Survey	N=375	69%	Aug 1998-Nov 1998

 

 

ACF Home | Questions? | Site Index | Contact Us | Accessibility | Privacy Policy | Freedom of Information Act | Disclaimers
Department of Health and Human Services | First.Gov: The U.S. Government's Homepage

Administration for Children and Families • 370 L'Enfant Promenade, S.W. • Washington, D.C. 20447