This is the accessible text file for GAO report number GAO-03-210 
entitled 'Welfare Reform: Former TANF Recipients with Impairments Less 
Likely to Be Employed and More Likely to Receive Federal Supports' 
which was released on December 06, 2002.



This text file was formatted by the U.S. General Accounting Office 

(GAO) to be accessible to users with visual impairments, as part of a 

longer term project to improve GAO products’ accessibility. Every 

attempt has been made to maintain the structural and data integrity of 

the original printed product. Accessibility features, such as text 

descriptions of tables, consecutively numbered footnotes placed at the 

end of the file, and the text of agency comment letters, are provided 

but may not exactly duplicate the presentation or format of the printed 

version. The portable document format (PDF) file is an exact electronic 

replica of the printed version. We welcome your feedback. Please E-mail 

your comments regarding the contents or accessibility features of this 

document to Webmaster@gao.gov.



Report to Congressional Requesters:



United States General Accounting Office:



GAO:



December 2002:



Welfare Reform:



Former TANF Recipients with Impairments Less Likely to Be Employed and 

More Likely to Receive Federal Supports:



GAO-03-210:



GAO Highlights:



Highlights of GAO-03-210, a report to the Ranking Minority Member, 

Subcommittee on Human Resources, Committee on Ways and Means, House of 

Representatives and another representative:



Why GAO Did This Study:



Debates surrounding the reauthorization of welfare reform legislation 

have involved some discussion regarding outcomes for TANF recipients 

with physical or mental impairments. To inform this discussion, GAO was 

asked to report on (1) whether recipients with impairments were as 

likely to exit TANF as their counterparts without impairments and (2) 

the sources of income reported by leavers with and without impairments.  

To obtain this information, GAO analyzed self-reported data for the 
most 

recent years available from the Census Bureau’s Survey of Income and 

Program Participation (SIPP) –- a national survey of households that 

includes questions about TANF status and functional impairments.



What GAO Found:



Recipients of Temporary Assistance for Needy Families (TANF) who had 

impairments were found to be half as likely to exit TANF as recipients 

without impairments, and recipients caring for children with 
impairments 

were found to be less than half as likely to exit TANF as recipients 
not 

caring for children with impairments, after controlling for demographic 

differences such as age, race, and marital status.  Although 
impairments 

affect exits, other factors, including family support and personal 

motivation, as well as local TANF policies, may also affect whether 

recipients exit TANF.



After leaving TANF, people with impairments were one-third as likely as 

people without impairments to be employed, according to a statistical 

model that controlled for demographic differences, and they were more 

likely to receive federal supports.  Forty percent of leavers with 

impairments reported receiving cash assistance from Supplemental 

Security Income (SSI), a federal program designed to assist low-income 

individuals who are aged, blind, or disabled.  Leavers with impairments 

were also more likely to receive non cash support in the form of Food 

Stamps and Medicaid than their counterparts without impairments.



These findings underscore the challenge states face in ensuring that 

recipients with impairments and those caring for children with 

impairments receive the supports they need to meet the work-focused 

goals and requirements of TANF.



Highlights Figure:  Employment and Receipt of SSI among Leavers with 

and without impairments, July 1997 through July 1999:



[See PDF for image]



[End of figure]



Contents:



Letter:



Results in Brief:



Background:



Impairments Were Relatively Common Among TANF Recipients:



Recipients with Impairments Were Less Likely to Exit TANF Than 

Recipients without Impairments:



After Leaving TANF, People with Impairments Were Less Likely to Be 

Employed and Were More Likely to Receive Federal Supports Than Were 

People without Impairments:



Concluding Observations:



Agency Comments:



Appendix I: Scope and Methodology:



Data Source: Census Bureau’s SIPP Data:



Logistic Regression Analyses:



Appendix II: Comments from the Department of Health and 

Human Services:



Appendix III: GAO Contacts and Staff Acknowledgments:



GAO Contacts:



Staff Acknowledgments:



Bibliography:



Related GAO Products:



Tables:



Table 1: Results of Logistic Regression Model of TANF Exits:



Table 2: Results of Logistic Regression Model of Post-TANF Employment:



Figures:



Figure 1: Prevalence of Impairments among TANF Recipients and Non-TANF 

Population, July 1997 through July 1999:



Figure 2: Demographic Differences between Adult TANF Recipients with 

and without Impairments:



Figure 3: Employment and Receipt of SSI among Leavers with and without 

Impairments, July 1997 through July 1999:



Figure 4: Proportion of Leavers with and without Impairments Reporting 

Receipt of Income from Personal Earnings, Household Earnings, or SSI in 

First Month after Leaving TANF:



Figure 5: Proportion of Leavers Receiving Food Stamps and Medicaid, 

July 1997 through July 1999:



Abbreviations:



ADA: Americans with Disabilities Act of 1990:



AFDC: Aid to Families with Dependent Children:



DI: Disability Insurance:



HHS: Department of Health and Human Services:



MDRC: Manpower Demonstration Research Corporation:



MOE: maintenance of effort:



PRWORA: Personal Responsibility and Work Opportunity Reconciliation Act 

of 1996:



SIPP: Survey of Income and Program Participation:



SSA: Social Security Administration:



SSI: Supplemental Security Income:



TANF: Temporary Assistance for Needy Families:



December 6, 2002:



The Honorable Benjamin L. Cardin

Ranking Minority Member

Subcommittee on Human Resources

Committee on Ways and Means

House of Representatives:



The Honorable Pete Stark

House of Representatives:



With the enactment of the Personal Responsibility and Work Opportunity 

Reconciliation Act of 1996 (PRWORA), the Congress made sweeping changes 

to federal welfare policy for needy families. PRWORA created the 

Temporary Assistance for Needy Families (TANF) block grant to states, 

which emphasizes work and responsibility over dependence on government 

benefits. Under TANF, states are to impose work requirements on most 

adults receiving TANF cash assistance and place a lifetime limit of 60 

months on the receipt of federal cash assistance. Some policymakers 

have expressed concerns about the ability of TANF recipients with 

physical or mental impairments to comply with program requirements, 

which has contributed to a heightened interest in what happens to 

people with impairments after they exit TANF. Although the TANF block 

grant program was due to expire on September 30, 2002, the Congress 

provided for an extension of the program until January 11, 2003. By 

that time, the Congress must either reauthorize the program or provide 

for an additional extension. As of September 30, 2002, the House of 

Representatives had passed reauthorization legislation and the Senate 

Committee on Finance had passed a reauthorization bill, although the 

full Senate had not yet voted on this bill.



To provide you with information on how people with impairments are 

faring in the current welfare environment and what issues may need 

attention as welfare reform evolves, you asked us to determine (1) how 

prevalent impairments are among TANF recipients; (2) whether recipients 

with impairments are as likely to exit TANF as recipients without 

impairments; and (3) what sources of income people with impairments 

have after leaving TANF, compared with people without impairments.



To address all three questions, we analyzed self-reported data for the 

most recent years available from the Census Bureau’s Survey of Income 

and Program Participation (SIPP)--a survey of households nationwide 

that asks respondents questions about their TANF status and functional 

impairments and uses categories of impairments comparable to those 

covered by the Americans with Disabilities Act of 1990 (ADA).[Footnote 

1] Because these data are self-reported, they may not accurately 

reflect the size of the population with impairments. We used a cross 

section of responses given between July 1997 and July 1999 and relied 

on a definition of impairments developed by Census that includes both 

severe and nonsevere physical and mental impairments. Our analyses 

included both descriptive statistics and multivariate analyses. We used 

appropriate techniques to weight the data to make population estimates. 

The sampling error for these estimates varied but did not exceed plus 

or minus 8 percentage points. For more information, see appendix I. We 

also reviewed findings of other studies to supplement the SIPP data. We 

conducted our work from March to October 2002 in accordance with 

generally accepted government auditing standards.



Results in Brief:



Impairments were relatively common among TANF recipients, with 

44 percent reporting having at least one impairment, caring for a child 

with at least one impairment, or both, compared with only 15 percent of 

the non-TANF population. The term impairments encompasses both mental 

and physical conditions. For instance, impairments could be physical 

conditions that hinder movement or require a cane or other mobility 

device, cognitive impairments, or mental conditions such as chronic 

depression. Adult recipients who had impairments were more likely to be 

over age 35 and white than adult recipients without impairments.



Recipients with impairments were less likely to exit TANF than 

recipients without impairments. Specifically, controlling for certain 

demographic characteristics, such as age, race, and marital status, 

adult recipients with impairments were half as likely to exit TANF as 

were adult recipients without impairments. Likewise, recipients caring 

for children with impairments were less than half as likely to exit 

TANF as all other recipients. Although impairments affect exits, other 

factors, such as family support, personal motivation, and local TANF 

policies, may also affect whether recipients exit TANF. For example, in 

a previous study, we found that local TANF agencies sometimes exempted 

recipients with impairments from requirements to participate in work or 

work-related activities. While in some cases this may be appropriate 

because of the level of severity of an impairment, in other cases it 

may mean that recipients may not get the encouragement or opportunity 

to acquire work skills that could help them exit TANF.



After leaving TANF, people with impairments were less likely to be 

employed and more likely to receive federal supports than were people 

without impairments. Controlling for demographic characteristics and 

other factors, we found that leavers with impairments were one-third as 

likely to be employed as leavers without impairments. Although they 

were less likely to be employed, many leavers with impairments (40 

percent) received income support from Supplemental Security Income 

(SSI), a federal program providing cash assistance to low-income 

individuals who are aged, blind, or disabled. Among those who did have 

earnings from employment or other sources, the amount of these earnings 

was similar for leavers with and without impairments and averaged 

around $1,000 per month. Leavers with impairments were more likely than 

leavers without impairments to report having no income--from personal 

earnings, household earnings, or SSI benefits--in their first month 

after exiting TANF. Specifically, 36 percent of leavers with 

impairments reported having no income from these sources compared with 

23 percent of leavers without impairments. However, leavers with 

impairments also were more likely to receive non-cash supports from 

Food Stamps (77 percent versus 62 percent) and Medicaid (89 percent 

versus 71 percent) than their counterparts without impairments.



In commenting on a draft of this report, the Department of Health and 

Human Services (HHS) noted our analysis, while possibly the best 

available approach, does not provide information on the extent to which 

outcomes differ for recipients with different types or severity of 

impairments, nor does it provide information on how local employment 

services may affect these outcomes. We acknowledge that while our 

analysis provides important descriptive information on outcomes for 

TANF recipients with impairments as a whole, much remains unknown about 

how best to help people with different types of impairments to become 

self-sufficient.



Background:



PRWORA built upon and expanded state-level welfare reforms to transform 

federal welfare policy for needy families with children. PRWORA 

replaced the individual entitlement to benefits under the 

61-year-old Aid to Families with Dependent Children (AFDC) program with 

the TANF block grant, which provides family assistance grants to the 

states, and emphasizes the transitional nature of assistance and the 

importance of reducing welfare dependence through employment, among 

other goals. HHS administers the TANF block grant program, which 

provided grants to states totaling up to $16.5 billion each year 

through September 2002. To receive its grant, each state must also 

spend at least a specified amount of its own funds, referred to as 

state maintenance of effort (MOE) funds.



State Flexibility on TANF Work Requirements and Time Limits:



While states have had flexibility to design programs that meet their 

own goals and needs, they also have been required to implement federal 

work requirements and time limits designed to promote employment among 

those able to work. First, TANF established stronger work requirements 

for those receiving aid than did the AFDC program. Specifically, to 

avoid financial penalties, states had to meet federal participation 

rate requirements, under which states were to ensure that an increasing 

percentage of adult recipients were participating in federally defined 

activities each year through fiscal year 2002. Second, states have been 

required to reduce the cash assistance benefit of an adult who did not 

participate as required by the state, referred to as a sanction, and 

could opt to terminate cash aid for the entire family. Third, states 

also have had to enforce a 60-month limit (or less at state option) on 

the length of time a family may receive federal TANF assistance.



However, the law also provided states considerable flexibility in how 

they implemented work requirements and time limits, and some states and 

localities have used this flexibility to exempt recipients with 

disabilities from these requirements. For example, in our 2002 report 

on states’ implementation of work requirements and time 

limits,[Footnote 2] we noted that states have generally faced greatly 

reduced federal participation rate requirements. This resulted from the 

law’s “caseload reduction credit” which adjusted downward the federally 

required rate if a state’s caseload declined, which is exactly what 

occurred in most states--dramatic caseload declines from 1996 through 

at least mid-2001. In fiscal year 2000, these caseload reduction 

credits reduced required rates from 40 percent (the required rate) to 0 

in 31 states. These lower participation rate requirements gave states 

more flexibility in exempting TANF recipients considered hard to employ 

from meeting work requirements. We found that while almost all states 

met or exceeded their adjusted required rate in that year, the federal 

participation rates that states actually achieved before adjustment 

ranged from about 6 percent to more than 70 percent.



Regarding time limits, we found that states generally excluded from 

time limits families with a parent or caretaker with a disability or 

caring for a family member with a disability. States could do this by 

using the 20-percent federal time limit extension established in the 

law or by using state maintenance of effort funds, as also allowed by 

the law.[Footnote 3] Our work also showed that most families had not 
yet 

reached their federal or state-imposed cash assistance time limit as of 

fall 2001.



While recipients with impairments may sometimes be exempted from work 

requirements and time limits, they may be at risk of having their 

benefits reduced or terminated through sanctions. A study in four urban 

areas conducted by the Manpower Demonstration Research Corporation 

(MDRC) found that recipients with a greater number of health problems 

were more likely to be sanctioned for noncompliance with program 

requirements than their healthier counterparts.[Footnote 4] Over 50 

percent of former recipients with at least one health problem left 

welfare due to sanctions compared with 39 percent of recipients without 

health problems. Our earlier report on sanctions under the TANF program 

found that families who left welfare due to sanctions relied on support 

from family and friends after TANF payments stopped, rather than on 

income from employment, to a greater extent than families who left the 

program for other reasons.[Footnote 5]



The Relationship between TANF and SSI:



TANF often serves, as did AFDC, as a temporary stopping point for low-

income individuals with physical or mental impairments that may be 

considered severe enough to make them eligible for the federal SSI 

program. SSI, administered by the Social Security Administration (SSA), 

provides cash assistance to low-income individuals who are aged or who 

are unable to work because of a severe long-term impairment and who do 

not have sufficient work history to qualify for SSA’s Disability 

Insurance (DI) program.[Footnote 6] To qualify for SSI, an applicant’s 

impairment must be of such severity that the person is not only unable 

to do the kind of work that he or she engaged in previously, but is 

also unable to do any other kind of substantial gainful activity that 

exists in the national economy.[Footnote 7] In most states, SSI 

eligibility also entitles individuals to Medicaid benefits. As distinct 

from TANF, SSI for adults has federally established eligibility 

requirements and benefit levels and a nationwide disability 

determination process.[Footnote 8]



Some individuals who apply for TANF may have impairments severe enough 

to make them eligible to receive SSI. Even before welfare reform, 

states had been actively identifying and referring potential SSI-

eligible welfare recipients to SSI. In these cases, individuals may be 

on TANF while they are waiting for their SSI eligibility to be 

determined.[Footnote 9] In recent years, receiving an initial 

disability determination took an average of about 4 months from the 

date of SSI application. For claims that are denied and appealed, it 

may take over a year to reach a final decision. Generally, except for 

more temporary conditions, TANF recipients who have impairments but are 

not eligible for SSI or DI may be expected to work, as their 

impairments have been deemed not severe enough to preclude substantial 

employment.



Title I of the ADA prohibits discrimination against such persons who 

have impairments but who are nonetheless able to perform the essential 

functions of the job they seek or hold.[Footnote 10] Under Title II of 

the ADA, no qualified individual with a disability shall be excluded 

from participation or be denied the benefits of the services, programs, 

or activities of a public entity, or be subject to discrimination by 

such entity. TANF, as a federal program, is subject to this 

requirement.[Footnote 11]



Identifying and Measuring Impairments:



Identifying and measuring impairments or disabilities is a complex 

undertaking, and no single survey instrument has been accepted or 

generally agreed upon as the preferred method for identifying 

impairments within a population. Census believes the extensive set of 

disability questions contained in the SIPP make it a preferred source 

to examine most impairment-related issues.[Footnote 12] Nevertheless, 

SIPP data should be interpreted with care. For instance, the SIPP 

relies on self-reports of impairments and, therefore, may not 

accurately reflect the size of the general or TANF population with 

impairments. This can result in the overreporting or underreporting of 

impairments. For example, although some impairments, such as the 

inability to walk, missing or impaired limbs, or severely impaired 

vision, are easy to identify, many impairments are not. Individuals may 

not report less obvious impairments because of certain stigmas 

surrounding them or because they may not know of their existence. Some 

examples of these impairments include learning disabilities, 

depression, and mental illness. Other surveys use different approaches 

to measure impairments. The National Household Survey of Drug Abuse and 

the University of Michigan’s Women’s Employment Survey, for example, 

use nonclinical in-depth diagnostic questioning to identify certain 

psychiatric disorders that may be overlooked by other survey 

techniques.[Footnote 13]



Impairments Were Relatively Common Among TANF Recipients:



Physical and mental impairments were reported to be relatively common 

among TANF recipients and, to a lesser degree, their children, compared 

with their prevalence among the non-TANF population.[Footnote 14] 

National survey data from the SIPP show that a total of 44 percent of 

TANF recipients reported in both 1997 and 1999 that they either had one 

or more physical or mental impairments as defined by Census or that 

they were caring for a child with such impairments.[Footnote 15] 

Specifically, in 29 percent of the TANF cases, only the adult recipient 

was reported to have impairments; in 

7 percent of the cases, only the child was reported to have 

impairments; and in 8 percent of the cases both the adult and child 

were reported to have impairments.[Footnote 16] The prevalence of 

impairments among TANF recipients is greater than among the U.S. non-

TANF population, among whom a total of 15 percent of individuals 

reported that they or their children had impairments. (See fig. 1.) 

Appendix I lists the specific criteria developed by Census that 

individuals must meet to be considered impaired as applied in the SIPP. 

We considered individuals to be impaired if they met the Census 

criteria in both 1997 and 1999.



Figure 1: Prevalence of Impairments among TANF Recipients and Non-TANF 

Population, July 1997 through July 1999:



[See PDf for image]



[End of figure]



As shown in figure 2, SIPP data show some demographic differences 

between TANF recipients aged 18 to 62 who have impairments and those 

who do not have impairments. Two-thirds of adult recipients with 

impairments were over 35 years old, while fewer than a quarter of adult 

recipients without impairments were older than 35. Age differences 

between individuals with and without impairments exist not only among 

TANF recipients, but among the non-TANF population as well. Among the 

non-TANF population with impairments, 81 percent were aged 36 to 

62 compared with 54 percent of those without impairments. Figure 2 also 

shows that TANF recipients with and without impairments differed by 

race. Forty-three percent of adult recipients with impairments were 

white compared with 28 percent of adult recipients without impairments. 

Among the non-TANF population, roughly equal percentages of people with 

and without impairments were white. Finally, as shown in figure 2, we 

found that SIPP data indicated no significant differences between 

recipients with and without impairments in the percentage who were 

married or the percentage who had no more than a high school education. 

Regardless of impairment status, about one-quarter of adult recipients 

were married and two-thirds to three-quarters had no more than a high 

school education.



Figure 2: Demographic Differences between Adult TANF Recipients with 

and without Impairments:



[See PDf for image]



[A]Differences between recipients with and without impairments are 

statistically significant.



[End of figure]



Recipients with Impairments Were Less Likely to Exit TANF Than 

Recipients without Impairments:



Impairments, whether they affected either adults or children, were 

associated with a decreased likelihood that a family would exit TANF. 

In particular, adult recipients with impairments were half as likely to 

exit TANF as adult recipients without impairments, after controlling 

for demographic differences, such as age, race, and marital status. 

Recipients caring for children with impairments were less than half as 

likely to exit TANF as others, after controlling for demographic 

differences. Different types of impairments or impairments of differing 

severity could have different effects on TANF exits, although we were 

not able to measure these effects. Furthermore, factors other than 

impairments may also affect whether recipients exit TANF.



Adult Recipients with Impairments Were Half as Likely to Exit TANF as 

Adult Recipients without Impairments:



Using a statistical model to control for basic demographic factors 

(gender, race, age, marital status, and education) and state-level 

differences,[Footnote 17] we found that adult recipients with 

impairments were half as likely to exit TANF as recipients without 

impairments. That is, an individual with an impairment who received 

TANF at some point between July 1997 and July 1999 was less likely than 

an individual without an impairment to have exited TANF by July 1999, 

all else being equal. For example, among whites, those with impairments 

were less likely to exit TANF than were whites without impairments. 

Likewise, among nonwhites, those with impairments were less likely to 

exit TANF than were nonwhites without impairments. If demographic 

factors are not taken into account, approximately equal proportions 

(about 3 out of 4) of recipients with and without impairments exited 

TANF.[Footnote 18]



Among those recipients who did exit TANF, a number of them returned to 

the TANF rolls at some point. SIPP data show that among individuals who 

received TANF and subsequently exited TANF between July 1997 and July 

1999, about 1 in 4 had returned to TANF before the end of that period. 

This was true both of individuals with impairments and those without 

impairments. Other studies of TANF leavers that have included various 

time periods, populations, and methodologies have found similar 

results.[Footnote 19] For example, a recent study using data from the 

National Survey of America’s Families found that 21.9 percent of 

families leaving welfare in 1997 returned within 2 years. They also 

found that almost half of those who returned originally left welfare to 

work and that return rates were higher for former recipients with 

little education, limited work experience, and poor health.[Footnote 

20]



Recipients Caring for Children with Impairments Were Less Than Half as 

Likely to Exit TANF as Others:



After using a statistical model to control for demographic factors, we 

found that recipients caring for children with impairments were less 

than half as likely to exit TANF as their counterparts not caring for 

children with impairments.[Footnote 21] A variety of complicating 

factors related to their children’s impairments may contribute to the 

decreased likelihood that this population of TANF recipients will 

pursue and maintain employment. For instance, parents of children with 

impairments may face demands on their time related to their children’s 

impairments in the form of special therapies, the administering of 

medications, regular medical appointments, and hospitalizations. 

Furthermore, the chronic and unpredictable nature of many impairments, 

such as severe asthma and seizures, may cause parents to be absent from 

work frequently and with little or no advance notice to their 

employers. This may be particularly problematic for TANF leavers, many 

of whom enter into low-or unskilled entry-level jobs that offer limited 

flexibility and benefits, such as vacation time, sick leave, and health 

insurance.



Finding child care and maintaining adequate health insurance coverage 

can be particularly challenging for parents caring for children with 

impairments. Children with impairments may need child care providers 

with the specialized training and equipment to accommodate their needs. 

In earlier work, we found that child care providers for children with 

special needs are sometimes in limited supply, especially in low-income 

neighborhoods.[Footnote 22] In addition to the difficulty in obtaining 

child care, families may be less likely to leave TANF if they are 

concerned about losing health care coverage. While the Congress 

established provisions to ensure that adults and children would 

continue to be eligible for Medicaid after leaving TANF, in our 1999 

report we found some evidence to suggest that the reforms of 1996 

initially contributed to confusion on the part of both beneficiaries 

and caseworkers about the criteria for maintaining Medicaid coverage 

after TANF benefits have been discontinued. Increased awareness of the 

need to ensure continued Medicaid enrollment for families exiting 

welfare has given rise to outreach efforts designed to promote 

awareness and maximize enrollment among eligible families.



Factors Other Than Impairments May Also Affect Whether Recipients Exit 

TANF:



Although recipients with impairments were less likely to exit TANF than 

recipients without impairments, SIPP data did not provide reliable data 

on several other factors that may also affect whether recipients exit 

TANF. For example, there were insufficient data to differentiate among 

individuals based on the severity, type, or number of their 

impairments. However, it is possible that these factors might affect 

whether individuals exit TANF, as evidenced in a study of SIPP data 

from the early 1990’s that suggested that respondents with more severe 

disabilities were less likely to exit welfare than respondents with 

less severe limitations.[Footnote 23] Furthermore, intangible factors 

such as family support and personal motivation might also lead to very 

different experiences with TANF for otherwise similar individuals. In 

our 1997 survey of individuals receiving Social Security Disability 

Insurance, encouragement from family and friends and high self-

motivation were identified to be among a range of factors that enabled 

these individuals with impairments to return to work.[Footnote 24]



In addition, local TANF policies, which are not measured by the SIPP 

questionnaire, may affect whether recipients with impairments exit 

TANF. For example, local TANF policies regarding screening, assessment, 

and work requirements may affect whether recipients with impairments 

receive assistance that could help them move toward employment. In a 

national survey of county TANF agencies conducted for our October 2001 

report, almost all the counties reported that they screened and 

assessed TANF recipients for impairments, but many used methods that 

may not accurately identify all impairments.[Footnote 25] In some 

cases, this may not be a problem because recipients may find jobs and 

leave welfare without special assistance. In other cases, recipients 

may need assistance targeted to their special needs to help them take 

steps toward employment or to transition to SSI.



We also found that many counties reported exempting from state work 

requirements TANF recipients who had impairments or were caring for a 

child with an impairment. While exemptions from work requirements may 

be appropriate in some cases, in other cases it may mean recipients may 

not be getting the help, direction, or encouragement they need to take 

steps toward employment and increase their chances of exiting TANF. 

Exemptions from work requirements could also leave them more at risk of 

reaching a time limit without getting the assistance they need to find 

employment or alternative means of support such as SSI. Our previous 

work and other research makes clear that recipients exit TANF for a 

variety of reasons--increased income, time limits, sanctions, and 

voluntary exits--and that the reason that a family exits TANF could 

have an effect on the family’s outcomes or circumstances. However, SIPP 

data did not provide reliable data on the reasons families exited 

TANF.[Footnote 26]



After Leaving TANF, People with Impairments Were Less Likely to Be 

Employed and Were More Likely to Receive Federal Supports Than Were 

People without Impairments:



TANF leavers with impairments were less likely to be employed and more 

likely to receive federal supports than were leavers without 

impairments. Although we found, after controlling for certain factors, 

that leavers with impairments were less likely to be employed than 

leavers without impairments, many of the leavers with impairments 

received income from SSI. Leavers with impairments also were more 

likely to receive Food Stamps and Medicaid.



Leavers with Impairments Were Less Likely to Be Employed, but Many 

Received SSI:



Leavers with impairments were one-third as likely to be employed as 

leavers without impairments, after controlling for basic demographic 

factors, state-level differences, and receipt of SSI.[Footnote 27] In 

other words, for those not receiving SSI, leavers with impairments were 

one-third as likely to be employed as leavers without impairments, all 

else being equal. Leavers caring for children with impairments were 

equally likely to be employed as others, after controlling for 

demographics and other factors.



In addition to estimating the probability of employment, we determined 

the actual percentages of adults who reported being employed at some 

point after leaving TANF between July 1997 and July 1999. Thirty-nine 

percent of adult leavers with impairments were employed at some point 

after leaving TANF, including 6 percent who also received SSI at some 

point after leaving TANF.[Footnote 28] (See fig. 3.) In contrast, 82 

percent of leavers without impairments reported being employed at least 

at some point after leaving TANF between July 1997 and July 1999. In 

addition to the 

6 percent of adult leavers with impairments who reported both 

employment and receipt of SSI, 34 percent reported receipt of SSI but 

not employment, indicating that a number of TANF recipients had 

impairments severe enough to qualify them for SSI and presumably also 

severe enough to limit their ability to sustain regular employment. 

Figure 3 shows that the proportion of leavers with impairments who 

reported either employment or SSI receipt, or both, is about the same 

as the proportion of leavers without impairments who reported 

employment.[Footnote 29]



The fact that many recipients with impairments seem to have impairments 

severe enough to qualify them for SSI suggests that many recipients are 

relying on TANF while awaiting determination of their eligibility for 

SSI. Again, it may take over a year from the time that an individual 

applies for SSI to the time that a final eligibility decision is made. 

During this time, individuals on TANF may or may not be exempted from 

work requirements.



Figure 3: Employment and Receipt of SSI among Leavers with and without 

Impairments, July 1997 through July 1999:



[See PDf for image]



Notes: For comparison, among the non-TANF populations, 52 percent of 

people with impairments were employed, while 93 percent of people 

without impairments were employed during the same time period.



“Employed” and “Receiving SSI” include people who reported being 

employed or receiving SSI, respectively, in any month after leaving 

TANF and before the end of July 1999. “Not employed” and “no SSI” 

include people who reported not being employed or not receiving SSI, 

respectively, the entire time after leaving TANF and before the end of 

July 1999.



[A] Employment and receipt of SSI were not necessarily concurrent.



[End of figure]



Leavers with impairments were not only less likely than those without 

impairments to be employed at any time after leaving TANF, but not 

surprisingly, they were also less likely to report having personal 

earnings from employment or other sources in any single month. In each 

of the first 6 months after exiting TANF, about 20 percent of leavers 

with impairments reported having personal earnings, compared with about 

60 percent of leavers without impairments. For those who did report 

personal earnings, though, the average amount of earnings for members 

of both groups was essentially equal, at about $1,000 per 

month.[Footnote 30] About 35 percent of leavers in both groups also 

reported household earnings. Regardless of their impairment status, 

their household earnings amounted on average to about $2,000 per month 

in addition to any personal earnings they may have had.



Leavers with impairments were more likely than those without 

impairments to report having no income--from personal or household 

earnings or SSI--in any single month, although they may have received 

Food Stamps or Medicaid.[Footnote 31] In their first month after 

leaving TANF, 36 percent of leavers with impairments reported having no 

personal or household earnings, or SSI, compared with 23 percent of 

leavers without impairments. (See fig. 4.) These proportions remained 

relatively constant in each of the first 6 months after leaving TANF.

 Over the course of the entire 24-month observation period, 10 percent 

of all individuals who left during that period reported never having 

income from personal or household earnings or SSI at any point after 

leaving TANF. This means that 90 percent of leavers had income from at 

least one of these sources at some point after leaving TANF. There were 

insufficient data to examine whether there were any differences between 

people with and without impairments on this measure.



Figure 4: Proportion of Leavers with and without Impairments Reporting 

Receipt of Income from Personal Earnings, Household Earnings, or SSI in 

First Month after Leaving TANF:



[See PDf for image]



[A] Differences between recipients with and without impairments are 

statistically significant.



[End of figure]



Leavers with Impairments Were More Likely to Receive Food Stamps and 

Medicaid:



A greater proportion of leavers with impairments reported receiving 

Food Stamps and Medicaid than did leavers without impairments. 

Specifically, 77 percent of leavers with impairments received Food 

Stamps compared with 62 percent of leavers without impairments. 

Similarly, 89 percent of leavers with impairments reported receiving 

Medicaid in contrast to 71 percent of leavers without impairments. 

(See fig. 5.):



Figure 5: Proportion of Leavers Receiving Food Stamps and Medicaid, 

July 1997 through July 1999:



[See PDf for image]



[A] Differences between leavers with and without impairments are 

statistically significant.



[End of figure]



Concluding Observations:



The 1996 welfare reform legislation enacted by the Congress clearly 

emphasizes the importance of welfare recipients taking steps toward 

employment and self-support. At the same time, the legislation provides 

states some flexibility to design programs that meet the needs of 

families affected by serious physical and mental impairments who may 

need special attention to facilitate the transition to work or to SSI. 

As states move beyond the first 5 years of the TANF program, a key 

challenge will be to ensure that recipients with impairments and those 

caring for children with impairments receive the supports they need to 

meet the work-focused goals and requirements of TANF.



Our findings underscore the magnitude and complexity of this challenge. 

Our findings that both adult recipients with impairments and recipients 

caring for children with impairments are less likely to exit TANF, and 

that adult leavers with impairments are less likely to be employed, 

suggest that in the early years of welfare reform at least, these 

families were not as successful as those without impairments at leaving 

welfare through work. Our finding that 40 percent of families with 

impairments who did leave welfare received SSI after leaving TANF shows 

that SSI is an important source of support for many of these families. 

This finding raises the difficult question of how best to use their 

time on TANF while awaiting SSI eligibility determination, such as what 

work expectations to have for these recipients. These findings also 

raise the more general question for policymakers about how best to 

promote work and personal responsibility--through work requirements and 

time limits--while at the same time taking into consideration the 

particular needs of recipients with impairments and those caring for 

children with impairments. While our analysis provides descriptive 

information on outcomes for TANF recipients with impairments, much 

remains unknown about how best to help people with different types of 

impairments become self-sufficient.



Agency Comments:



In commenting on a draft of this report, HHS said that the topic of 

TANF recipients with impairments is an important one. HHS also noted 

that our analysis, while possibly the best available approach, has 

limited application in providing information on the extent to which 

different types of impairments, impairments of varying severity, or 

local employment services may affect outcomes for individuals with 

impairments. We acknowledge that our analysis focuses on describing 

outcomes rather than identifying explanations for these outcomes, in 

part because information is not readily available to look at the more 

complex picture of each individual’s needs and the particular services 

received. However, our analysis provides important information on what 

is happening in the early years of welfare reform with regard to 

recipients with impairments as a whole. We added language to our 

concluding observations to state that much remains unknown about how 

best to help people with different types of impairments to become self-

sufficient. HHS also noted that an analysis that excluded recipients 

who moved onto SSI would be useful. We added language to the report to 

clarify that our finding that recipients with impairments are one-third 

as likely to be employed as recipients without impairments refers to 

recipients who did not receive SSI. HHS’s written comments are included 

in appendix II. HHS and two welfare experts also provided technical 

comments, which we have incorporated where appropriate.



We are sending copies of this report to the Secretary of Health and 

Human Services, relevant congressional committees, and other interested 

parties. We will also make copies available to others upon request. In 

addition, the report is available at no charge on GAO’s Web site at 

http://www.gao.gov.



If you have any questions about this report, please contact me at 

(202) 512-7215 or Gale Harris at (202) 512-7235. Other contacts and 

acknowledgments are listed in appendix III.



Signed by Cynthia M. Fagnoni:



Cynthia M. Fagnoni

Managing Director, Education, Workforce

 and Income Security:



[End of section]



Appendix I: Scope and Methodology:



Data Source: Census Bureau’s SIPP Data:



To describe the role of physical and mental impairments in the lives of 

families leaving Temporary Assistance for Needy Families (TANF), we 

developed estimates of the number of TANF recipients with impairments 

and investigated the differences between TANF recipients and leavers 

with and without impairments, using a 2-year cross section of data from 

the Census Bureau’s Survey of Income and Program Participation (SIPP). 

The SIPP is a national household survey conducted by the U. S. Census 

Bureau in which panels of individuals representative of the nation, 

including those receiving TANF, are interviewed over a period of 2 

years or more. At 4-month intervals, panel participants are asked a set 

of “core” questions involving such subjects as their labor force 

activity, welfare program participation, and demographic 

characteristics. Periodically, the survey also asks a detailed set of 

questions called “topical modules” on a variety of topics not covered 

in the core section, such as disabilities. For our purposes, we 

selected panels starting in 1996 and sampled TANF and non-TANF adults 

between the ages of 18 and 62. Data from the topical modules on 

disability that we analyzed were from interviews conducted from August 

1997 to November 1997, and August 1999 to November 1999, in which 

respondents were asked about their status in recent months, including 

July of that year. We included respondents who were in the sample in 

both July 1997 and July 1999 and analyzed their responses during this 

time period.



During these interviews, panel members were asked an extensive set of 

questions about their physical or mental impairments, including 

questions on a range of functional or other activity limitations. To be 

identified as having a disability or impairment in the SIPP, 

individuals must meet specific disability criteria developed by the U. 

S. Census Bureau.[Footnote 32] That is, they must meet any of the 

following criteria:



1. Had difficulty performing one or more functional activities, 

including seeing, hearing, speaking, lifting, and carrying, using 

stairs, and walking.



2. Had difficulty with one or more activities of daily living, such as 

getting around inside the home, getting in or out of a bed or chair, 

bathing, dressing, and eating.



3. Had difficulty with one or more instrumental activities of daily 

living, including going outside the home, keeping track of money or 

bills, preparing meals, doing light housework, and using the telephone.



4. Had one or more specific conditions, including a learning 

disability, mental retardation or another developmental disability, 

Alzheimer’s disease, or some other type of mental or emotional 

condition.



5. Had an other mental or emotional condition that seriously interfered 

with everyday activities, including frequently depressed or anxious, 

trouble getting along with others, trouble concentrating, or trouble 

coping with day-to-day stress.



6. Had a condition that limited the ability to work, including around 

the house.



7. Had a condition that made it difficult to work at a job or business.



8. Received federal benefits based on inability to work.



9. Used a wheelchair, a cane, crutches, or a walker.



For our purposes, we considered individuals to have impairments if 

their survey responses indicated they had impairments at both times 

that the disability topical module was administered (i.e., in both 1997 

and 1999). We considered individuals to not have impairments if their 

survey responses indicated they did not have impairments at both times 

that the disability topical module was administered. Individuals whose 

impairment status differed between the first and second modules were 

excluded from the analyses. (We excluded 12.5 percent of respondents 

for this reason). We used appropriate techniques to weight the data to 

make population estimates for 1999 as well as to take into account the 

complex sampling design when estimating variances. Because the 

estimates we reported from the SIPP were based on samples, they are 

subject to sampling error, which varied but did not exceed plus or 

minus 8 percentage points at the 95-percent confidence interval. 

Therefore, the chances are 95 out of 100 that the actual population

percentages are within no more than plus or minus 8 percentage points 

of our estimates.



Logistic Regression Analyses:



In addition to descriptive statistics, we used logistic regression 

models to examine the effects of recipients’ having impairments, and of 

recipients’ caring for children with impairments on the likelihood of 

leaving TANF and of being employed after leaving TANF, after 

controlling for age, gender, marital status, race, and educational 

attainment.[Footnote 33] Recognizing that TANF policies may vary across 

states, we controlled for state in the models as well. The models of 

post-TANF employment also controlled for receipt of Supplemental 

Security Income (SSI).[Footnote 34]



The results from the models we used are odds ratios that estimate, in 

table 1, the relative likelihood of leaving TANF for each factor and, 

in table 2, the effect of each factor on the likelihood of being 

employed after leaving TANF. If there were no significant differences 

between two groups, their odds would be equal, and the ratio of their 

odds would be 1.00. The more the odds ratio differs from 1.00 in either 

direction, the larger the effect it represents.



The odds ratios in each table were computed in relation to a defined 

reference group. In table 1 an odds ratio that is greater than 1.00 

indicates a greater likelihood of leaving TANF than the reference group 

while a ratio under 1.00 indicates a lesser likelihood of leaving than 

the reference group. In table 2 an odds ratio that is greater than 1.00 

indicates a greater likelihood of being employed after leaving TANF 

than the reference group while a ratio under 1.00 indicates a lesser 

likelihood of being employed after leaving TANF than the reference 

group. Both tables also show the

95-percent confidence intervals around the odds ratios. If these 

intervals contain 1.00, the difference is not statistically 

significant.



Table 1: Results of Logistic Regression Model of TANF Exits:



Variable: Have a disability?.



Variable: Disabled; Model of adults with impairments: Odds ratio: Have 

a disability?: 0.49[A]; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: (0.29-0.85); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: b; Model of adults caring 

for children with impairments: 95-percent 

confidence interval: Have a disability?: b.



Variable: Not disabled; Model of adults with impairments: Odds ratio: 

Have a disability?: 1.00; (reference group); Model of adults with 

impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: b; Model of adults caring for children with 

impairments: 95-percent 

confidence interval: Have a disability?: b.



Variable: Have a child with a disability?.



Variable: Child with disability; Model of adults with impairments: Odds 

ratio: Have a disability?: b; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: b; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 0.44[A]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: (0.28-0.69).



Variable: No child with disability; Model of adults with impairments: 

Odds ratio: Have a disability?: b; Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: b; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Age.



Variable: 18-35 as of 7/97; Model of adults with impairments: Odds 

ratio: Have a disability?: 0.85; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: (0.53-1.35); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 1.09; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.72-1.65).



Variable: 36-62 as of 7/97; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Gender.



Variable: Male; Model of adults with impairments: Odds ratio: Have a 

disability?: 2.21; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: (0.88-5.59); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 1.89; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.76-4.68).



Variable: Female; Model of adults with impairments: Odds ratio: Have a 

disability?: 1.00; (reference group); Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Marital status.



Variable: Currently married; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.24; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: (0.78-1.99); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 1.40; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.86-2.28).



Variable: Not currently married; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Minority status.



Variable: Minority; Model of adults with impairments: Odds ratio: Have 

a disability?: 0.40[A]; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: (0.23-0.70); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 0.44[A]; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.27-0.73).



Variable: White; Model of adults with impairments: Odds ratio: Have a 

disability?: 1.00; (reference group); Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Education.



Variable: More than high school; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.01; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: (0.59-1.71); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 1.06; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.63-1.81).



Variable: High school or less; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Note: The models also included a set of dummy variables to allow for 

effects of unmeasured state characteristics. We have omitted the 

coefficients associated with the dummy variables to simplify 

presentation.



[A] Odds ratio is statistically significant at p<=0.05.



[B] Not applicable.



Source: GAO’s analysis of SIPP data on disability.



[End of table]





Table 2: Results of Logistic Regression Model of Post-TANF Employment:



Variable: Have a disability?.



Variable: Disabled; Model of adults with impairments: Odds ratio: Have 

a disability?: 0.28[A]; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: (0.13-0.60); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: b; Model of adults caring 

for children with impairments: 95-percent 

confidence interval: Have a disability?: b.



Variable: Not disabled; Model of adults with impairments: Odds ratio: 

Have a disability?: 1.00; (reference group); Model of adults with 

impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: b; Model of adults caring for children with 

impairments: 95-percent 

confidence interval: Have a disability?: b.



Variable: Have a child with a disability?.



Variable: Child with disability; Model of adults with impairments: Odds 

ratio: Have a disability?: b; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: b; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 0.72; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: (0.33-1.58).



Variable: No child with disability; Model of adults with impairments: 

Odds ratio: Have a disability?: b; Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: b; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Age; Model of adults with impairments: Odds ratio: Have a 

disability?: [Empty]; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: 18-35 as of 7/97; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.45; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: (0.72-2.92); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 2.27; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (1.37-3.73).



Variable: 36-62 as of 7/97; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Gender; Model of adults with impairments: Odds ratio: Have a 

disability?: [Empty]; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Male; Model of adults with impairments: Odds ratio: Have a 

disability?: 4.90[A]; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: (1.45-16.51); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 1.41; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.68-2.95).



Variable: Female; Model of adults with impairments: Odds ratio: Have a 

disability?: 1.00; (reference group); Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Marital status; Model of adults with impairments: Odds ratio: 

Have a disability?: [Empty]; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Currently married; Model of adults with impairments: Odds 

ratio: Have a disability?: 0.31[A]; Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: (0.16-0.60); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 0.51[A]; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.30-0.86).



Variable: Not currently married; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Minority status; Model of adults with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Minority; Model of adults with impairments: Odds ratio: Have 

a disability?: 0.70; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: (0.33-1.50); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 1.00; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (0.58-1.73).



Variable: White; Model of adults with impairments: Odds ratio: Have a 

disability?: 1.00; (reference group); Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Education; Model of adults with impairments: Odds ratio: Have 

a disability?: [Empty]; Model of adults with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: More than high school; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.93; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: (0.89-4.18); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: 2.26; Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: (1.29-3.95).



Variable: High school or less; Model of adults with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: 1.00; (reference group); Model of adults 

caring for children with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Receipt of SSI; Model of adults with impairments: Odds ratio: 

Have a disability?: [Empty]; Model of adults with impairments: 95-

percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: [Empty]; Model of adults caring for children 

with impairments: 95-percent 

confidence interval: Have a disability?: [Empty].



Variable: Not receiving SSI; Model of adults with impairments: Odds 

ratio: Have a disability?: 28.18[A]; Model of adults with impairments: 

95-percent 

confidence interval: Have a disability?: (5.66-150.37); Have a 

disability?: [Empty]; Model of adults caring for children with 

impairments: Odds ratio: Have a disability?: b; Model of adults caring 

for children with impairments: 95-percent 

confidence interval: Have a disability?: b.



Variable: Receiving SSI; Model of adults with impairments: Odds ratio: 

Have a disability?: 1.00; (reference group); Model of adults with 

impairments: 95-percent 

confidence interval: Have a disability?: [Empty]; Have a disability?: 

[Empty]; Model of adults caring for children with impairments: Odds 

ratio: Have a disability?: b; Model of adults caring for children with 

impairments: 95-percent 

confidence interval: Have a disability?: b.



Note: The models also included a set of dummy variables to allow for 

effects of unmeasured state characteristics. We have omitted the 

coefficients associated with the dummy variables to simplify 

presentation.



[A] Odds ratio is statistically significant at p<=0.05.



[B] Not applicable.



Source: GAO’s analysis of SIPP data on disability.



[End of table]



Definitions of Other Variables:



TANF recipient: Respondents who reported receiving TANF in any month 

during the period (July 1997 through July 1999).



TANF leaver: Respondents who reported receiving TANF in some month 

during the period and subsequently not receiving TANF at some point for 

at least 2 consecutive months.



Non-TANF population: Respondents who did not receive TANF benefits in 

any month during the time period.



Employed (leavers): Respondents who reported employment in any month 

after leaving TANF during the time period.



Age: Categorized as 18-35 and 36-62 and defined as the respondent’s 

reported age in July 1997.



Education: Categorized as either having more than a high school 

education or not. For models of TANF exits, education is defined as the 

reported level of education in July 1997; for models predicting 

employment among leavers, education is defined as the reported level of 

education in the month the respondent reported leaving TANF.



Marital status: Categorized as either married or not. For models of 

TANF exits, marital status is defined as reported status in July 1997; 

for models predicting employment among leavers, marital status is 

defined as reported status in the month the respondent reported leaving 

TANF.



Received Food Stamps/Medicaid (leavers): Respondents who reported 

receiving Food Stamps/Medicaid in any month after leaving TANF during 

the time period.



Received SSI (leavers): Respondents who reported receiving SSI in any 

month after leaving TANF during the time period.



[End of section]



Appendix II: Comments from the Department of Health and Human Services:



DEPARTMENT OF HEALTH & HUMAN SERVICES:



Office of Inspector General:



Washington, D.C. 20201:



Ms. Cynthia M. Fagnoni

Managing Director, Education, Workforce, 

and Income Security Issues

United States General 

Accounting Office 

Washington, D.C. 20548:



Dear Ms. Fagnoni:



Enclosed are the department’s comments on your draft report entitled, 

“Welfare Reform: Former TANF Recipients with Impairments Less Likely to 

Be Employed and More Likely to Receive Federal Supports.” The comments 

represent the tentative position of the department and are subject to 

reevaluation when the final version of this report is received.



The department also provided several technical comments directly to 

your staff.



The department appreciates the opportunity to comment on this draft 

report before its publication.



Sincerely,



Signed by Janet Rehnquist:



Janet Rehnquist 

Inspector General:



Enclosure:



The Office of Inspector General (OIG) is transmitting’ the department’s 

response to this draft report in our capacity as the department’s 

designated focal point and coordinator for General Accounting Office 

reports. The OIG has not conducted an independent assessment of these 

comments and therefore expresses no opinion on them.



Comments of the Department of Health and Human Services on the General 

Accounting Office’s Draft Report, “Welfare Reform: Former TANF 

Recipients with Impairments Less Likely to Be Employed and More Likely 

to Receive Federal Supports” (GAO-03-210):



General Comments:



The Department of Health and Human Services appreciates the opportunity 

to comment on this draft report, which addresses the important topic of 

Temporary Assistance for Needy Families (TANF) and impairments.



Helping families move to self-sufficiency is the overarching goal of 

the TANF program. While there is general recognition that many TANF 

families face barriers to achieving the goal of self-sufficiency, 

policymakers and program managers recognize the importance of putting 

the necessary supports in place to allow recipients to successfully 

obtain and maintain jobs. Impairments are among the barriers that TANF 

families face. These impairments may be physical or mental and can 

affect both adults and children. The degree of severity varies widely. 

The occurrence of impairments may be alone or in conjunction with other 

barriers. As the report describes, some TANF recipients with 

impairments have successfully moved into the workforce and some with 

more serious impairments have used TANF during the transition into 

Supplemental Security Income (SSI) or other disability benefits. As 

states move beyond the first 5 years of the TANF program, a key 

challenge will be to ensure that all TANF clients have an opportunity 

to earn a living, including those with special circumstances such as 

impairments and those caring for children with impairments.



The report should take some opportunity to explore the limitations of 

its conclusions in employing an analysis based on a dichotomous 

comparison of outcomes. Theoretically, the degree of impairment may be 

distributed across a population more in the form of a spectrum than two 

nodes. Thus, this type of analysis, while possibly the best available 

approach, has limited application. Quite possibly, the people with 

impairments who found employment were less severely impaired than those 

who did not. It was noted that some of the impaired population would 

eventually receive SSI. However, knowing that this subpopulation is 

included also means that there should be an acknowledgement that the 

results could be different if the analysis could find a way to cull out 

those who might ultimately exit into SSI. Some additional caveats or 

discussion would be useful in exploring the limitations of the analysis 

in understanding the research problem. In addition, most employment 

services are provided at the community level. The inability to control 

for such local differences in employment services also makes drawing 

conclusions from the correlations more suspect.



Finally, no information is available in the Census Bureau’s Survey of 

Income and Program Participation or the report about whether clients 

received services, the type and extent of services and whether such 

services affect the reported results. Such information would 

dramatically affect the implications that the findings “underscore the 

challenge states face in reconciling the work-focused goals and 

requirements of TANF....”:



[End of section]



Appendix III: GAO Contacts and Staff Acknowledgments:



GAO Contacts:



Sigurd Nilsen (202) 512-7003:



Gale Harris (202) 512-7235:



Heather McCallum (202) 512-2890:



Staff Acknowledgments:



In addition to those named above, Tiffany Boiman, Wendy Ahmed, and 

Grant Mallie made important contributions to this report.



[End of section]



Bibliography:



Acs, Gregory, and Pamela Loprest. “Do Disabilities Inhibit Exits from 

AFDC?” Washington, D.C.: The Urban Institute, 1994.



Brandon, Peter D., and Dennis P. Hogan. “The Effects of Children with 

Disabilities on Mothers’ Exit from Welfare.” Paper presented at the 

Joint Center for Poverty Research, Research Conference. Washington, 

D.C.: February 2002.



Collier-Bolkus, Winifred. “The Impact of the Welfare Reform Law on 

Families with Disabled Children That Need Child Care.” Ph.D. 

diss.,Widener University, 2000.



Danziger, S.K., and others. “Barriers to the Employment of Welfare 

Recipients.” In Prosperity for All? The Economic Boom and African 

Americans. Eds. R. Cherry and W. Rodgers. New York: Russell Sage 

Foundation Press, 2000.



Lee, Sunhwa, Melissa Sills and Gi-Taik Oh. “Disabilities Among Children 

and Mothers in Low Income Families.” Washington, D.C.: Institute for 

Women’s Policy Research, June 20, 2002.



Meyers, Marcia K., Anna Lukemeyer and Timothy Smeeding. “Work, Welfare 

and the Burden of Disability: Caring for Special Needs of Children in 

Poor Families.” Syracuse University: Center for Policy Research, 

Maxwell School of Citizenship and Public Affairs. 1996.



Polit, Denise F., Andrew S. London, and John M. Martinez. “The Health 

of Poor Urban Women: Findings from the Project on Devolution and Urban 

Change.” New York, NY: Manpower Demonstration Research Corporation. 

2001.



Smith, Lauren A., MD et al. “Employment Barriers Among Welfare 

Recipients and Applicants with Chronically Ill Children.” American 

Journal of Public Health, 92, no. 9 (September 2002): 1453-1457.



Wise, Paul H., MD et al. “Chronic Illness Among Poor Children Enrolled 

in the Temporary Assistance for Needy Families Program.” American 

Journal of Public Health, 92, no. 9 (September 2002): 1458-1461.



Wood, Pamela R., MD et. al. “Relationships Between Welfare Status, 

Health Insurance Status, and Health and Medical Care Among Children 

with Asthma.” American Journal of Public Health, 92, no. 9 (September 

2002): 1446-1452.



[End of section]



Related GAO Products:



Welfare Reform: Tribes Are Using TANF Flexibility to Establish Their 

Own Programs. GAO-02-695T. Washington, D.C.: May 10, 2002.



Welfare Reform: Federal Oversight of State and Local Contracting Can Be 

Strengthened. GAO-02-661. Washington, D.C.: June 11, 2002.



Welfare Reform: States Are Using TANF Flexibility to Adapt Work 

Requirements and Time Limits to Meet State and Local Needs. 

GAO-02-501T. Washington, D. C.: March 7, 2002.



Welfare Reform: More Coordinated Federal Efforts Could Help States and 

Localities Move TANF Recipients with Impairments Toward Employment. 

GAO-02-37. Washington, D. C.: October 31, 2001.



Welfare Reform: Moving Hard-to-Employ Recipients Into the Workforce.

GAO-01-368 Washington, D. C.: March 15, 2001.



Welfare Reform: Work-Site-Based Activities Can Play an Important Role 

in TANF Programs. GAO/HEHS-00-122. Washington, D. C.: July 28, 2000.



Welfare Reform: Means-Tested Programs: Determining Financial 

Eligibility is Cumbersome and Can Be Simplified. GAO-02-58. Washington, 

D.C.: November 2, 2001.



Welfare Reform: Improving State Automated Systems Requires Coordinated 

Federal Effort. GAO/HEHS-00-48. Washington, D. C.: April 27, 2000.



Welfare Reform: State Sanction Policies and Number of Families 

Affected. GAO/HEHS-00-44 Washington, D. C.: March 31, 2000.



Welfare Reform: Assessing the Effectiveness of Various Welfare-to-Work 

Approaches. GAO/HEHS-99-179. Washington, D. C.: September 7, 1999.



Welfare Reform: Information on Former Recipientsí Status.

GAO/HEHS-99-48. Washington, D. C.: April 28, 1999.



Welfare Reform: Statesí Experiencs in Providing Employment Assistance 

to TANF Clients. GAO/HEHS-99-22. Washington, D. C.: February 26, 1999.



Welfare Reform: Status of Awards and Selected Statesí Use of Welfare-

to-Work Grants. GAO/HEHS-99-40. Washington, D. C.: February 5, 1999.



FOOTNOTES



[1] The ADA defines persons with disabilities as those who have a 

physical or mental impairment that substantially limits one or more 

major life activities, such as walking, hearing, etc.; those who have a 

record of such impairment; or those who are regarded as having such an 

impairment.



[2] U.S. General Accounting Office, Welfare Reform: With TANF 

Flexibility, States Vary in How They Implement Work Requirements and 

Time Limits, GAO-02-770 (Washington D.C.: July 5, 2002). 



[3] A state may exempt up to 20 percent of its average monthly caseload 

for hardship or having been subjected to domestic violence.



[4] Denise F. Polit, Andrew S. London, and John M. Martinez, The Health 

of Poor Urban Women: Findings from the Project on Devolution and Urban 

Change, (New York: Manpower Demonstration Research Corporation, 2001). 



[5] U.S. General Accounting Office, Welfare Reform: State Sanction 

Policies and Number of Families Affected, GAO/HEHS-00-44 (Washington, 

D. C.: Mar. 31, 2000).



[6] Cash assistance and services for persons with disabilities who have 

worked long enough and recently enough are also available from the DI 

program. Other programs, which may be available, include private 

disability insurance or pensions and state workers’ compensation 

programs.



[7] Work activity is generally considered substantial and gainful if 

the person’s earnings exceed a particular level established by statute 

and regulations.



[8] Some states provide supplemental payments to the federal benefit 

level. 



[9] Federal rules for TANF and SSI do not explicitly prohibit 

individuals from receiving cash assistance from both TANF and SSI 

simultaneously. However, in effect, states’ income eligibility rules 

for TANF generally preclude individuals from doing so.



[10] The ADA was enacted, in part, to remove barriers to employment and 

receipt of public services for people with disabilities by prohibiting 

discrimination. In the area of employment, the ADA requires employers 

to make reasonable accommodations to persons with disabilities (e.g., 

by providing a magnified computer screen for a vision-impaired person), 

unless such accommodations would impose undue hardship on employers.



[11] On January 19, 2001, HHS’s Office of Civil Rights issued “Summary 

of Policy Guidance Prohibition Against Discrimination on the Basis of 

Disability in the Administration of TANF” to all entities involved in 

the administration and operation of TANF programs.



[12] The SIPP data reported do not include impairments related to 

substance abuse.



[13] Rukmalie Jayakody, Sheldon Danziger, and Harold Pollack, “Welfare 

Reform, Substance Use, and Mental Health,” Journal of Health Politics, 

Policy and Law (Aug. 2000).



[14] “TANF recipients” are defined as those SIPP respondents who 

reported receiving TANF in any month during the period of July 1997 

through July 1999.



[15] In an earlier study (GAO-02-37), we reported that SIPP data showed 

that in 1999, a total of 44 percent of TANF adults aged 18 to 64 

reported having one or more physical or mental impairments as defined 

by Census. This figure differs from our current finding that SIPP data 

show a total of 44 percent of TANF adults either had impairments 

themselves or were caring for a child with impairments in both 1997 and 

1999. The fact that both numbers are “44 percent” is purely 

coincidental.



[16] These data capture individuals who reported functional or other 

activity limitations generally covered by the ADA.



[17] A variable is included in the model to control for any differences 

among states, although the model does not evaluate the specific effects 

of different state policies. See appendix I for model results.



[18] This helps to explain why the proportion of the caseload that had 

impairments did not increase between 1997 and 1999, as we found in our 

October 2001 report, even though one might expect the proportion of 

recipients with impairments to increase if recipients with impairments 

were less likely to exit TANF than those without impairments.



[19] See U.S. General Accounting Office, Welfare Reform: Progress In 

Meeting Work-Focused TANF Goals, GAO-01-522T (Washington, D.C.: Mar. 

15, 2001).



[20] Pamela Loprest. Who Returns to Welfare? Policy Brief B-49 

(Washington, D.C.: The Urban Institute, Sept. 2002).



[21] See appendix I for model results.



[22] See U.S. General Accounting Office, Child Care: States Increased 

Spending on Low-Income Families, GAO-01-293 (Washington, D.C.: Feb. 2, 

2001); and Welfare Reform: States’ Efforts to Expand Child Care 

Programs, GAO/HEHS-98-27 (Washington, D.C.: Jan. 13, 1998).



[23] Gregory Acs and Pamela Loprest, Do Disabilities Inhibit Exits from 

AFDC? (Washington, D.C.: The Urban Institute, Aug. 1994).



[24] U.S. General Accounting Office, Social Security Disability 

Insurance: Multiple Factors Affect Beneficiaries’ Ability to Return to 

Work, GAO/HEHS-98-39 (Washington, D.C.: Jan. 12, 1998).



[25] U.S. General Accounting Office, Welfare Reform: More Coordinated 

Federal Effort Could Help States and Localities Move TANF Recipients 

With Impairments Toward Employment, GAO-02-37 (Washington, D.C.: Oct. 

31, 2001). The terms screening and assessment are often loosely applied 

and can have different meanings in various treatment and service 

communities. We defined screening as “any means of gaining information 

about an individual that can be used to detect warning signs that 

suggest that some form of impairment might exist.” If there is an 

indication that an impairment may exist, the next step is to perform an 

assessment. We defined assessment as “a comprehensive examination of an 

individual that is used to identify the specific impairment(s) he or 

she has.”



[26] The survey did ask respondents about their reasons for exiting 

TANF, but 81 percent did not answer the question.



[27] The model analyzed employment status for the period spanning July 

1997 through July 1999 and controlled for gender, race, age, marital 

status, education, and receipt of SSI. A variable was included in the 

model to control for any differences among states, although the model 

did not evaluate the specific effects of different state policies. See 

appendix I for model results.



[28] Employment and receipt of SSI were not necessarily concurrent.



[29] There is no statistically significant difference between the 

proportions of leavers with impairments and without impairments who are 

neither employed nor receiving SSI.



[30] Estimates are based on small numbers of respondents, so 

differences too subtle to measure could exist.



[31] There were insufficient data to determine how many recipients who 

reported no income also did not receive Food Stamps or Medicaid.



[32] Children were identified as having an impairment or not based on 

the questions in the SIPP disability module related to children’s 

disabilities. These questions differed somewhat from the questions 

related to adults’ disabilities.



[33] The purpose of our model is only to examine the effects of 

recipients or their children having impairments on the likelihood of 

leaving TANF and of being employed after leaving TANF. We are not 

attempting to explain TANF exits or post-TANF employment generally.



[34] In developing our model, we examined the effects of all of the 

reasonable variables available to us in the SIPP. In addition to those 

included in the final model, we tested the effect of the number of 

children in a household and found that this variable had no significant 

effect on the results. 



GAO’s Mission:



The General Accounting Office, the investigative arm of Congress, 

exists to support Congress in meeting its constitutional 

responsibilities and to help improve the performance and accountability 

of the federal government for the American people. GAO examines the use 

of public funds; evaluates federal programs and policies; and provides 

analyses, recommendations, and other assistance to help Congress make 

informed oversight, policy, and funding decisions. GAO’s commitment to 

good government is reflected in its core values of accountability, 

integrity, and reliability.



Obtaining Copies of GAO Reports and Testimony:



The fastest and easiest way to obtain copies of GAO documents at no 

cost is through the Internet. GAO’s Web site ( www.gao.gov ) contains 

abstracts and full-text files of current reports and testimony and an 

expanding archive of older products. The Web site features a search 

engine to help you locate documents using key words and phrases. You 

can print these documents in their entirety, including charts and other 

graphics.



Each day, GAO issues a list of newly released reports, testimony, and 

correspondence. GAO posts this list, known as “Today’s Reports,” on its 

Web site daily. The list contains links to the full-text document 

files. To have GAO e-mail this list to you every afternoon, go to 

www.gao.gov and select “Subscribe to daily E-mail alert for newly 

released products” under the GAO Reports heading.



Order by Mail or Phone:



The first copy of each printed report is free. Additional copies are $2 

each. A check or money order should be made out to the Superintendent 

of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 

more copies mailed to a single address are discounted 25 percent. 

Orders should be sent to:



U.S. General Accounting Office



441 G Street NW,



Room LM Washington,



D.C. 20548:



To order by Phone: 	



	Voice: (202) 512-6000:



	TDD: (202) 512-2537:



	Fax: (202) 512-6061:



To Report Fraud, Waste, and Abuse in Federal Programs:



Contact:



Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov



Automated answering system: (800) 424-5454 or (202) 512-7470:



Public Affairs:



Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800 U.S.



General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C.



20548: