APPENDIX L. MONITORING THE EFFECTS OF WELFARE REFORM CONTENTS Evaluations and Reports Required by Law Major Research Programs Monitoring or Evaluating Welfare Reform National Surveys State Policy Monitoring Development of Program and Social Indicators Implementation Studies Topical Studies State Welfare Demonstration Programs Other Evaluations EVALUATIONS AND REPORTS REQUIRED BY LAW Numerous governmental and other organizations plan to monitor the effects of the 1996 welfare reform law. The law itself requires information and data reporting by States; calls for several reports by States and Federal agencies; and authorizes evaluations of programs and policies it created or modified. Table L-1 summarizes the various reporting requirements and evaluations related to welfare reform specified in law. MAJOR RESEARCH PROGRAMS MONITORING OR EVALUATING WELFARE REFORM A variety of research activities are planned or underway to monitor and evaluate welfare reform. These activities include the fielding of new national surveys of families likely to be affected by welfare reform. Most notable among these survey efforts are the Census Bureau's Survey of Program Dynamics (SPD) and the Urban Institute's Survey of American Families. Several organizations are monitoring State welfare policies under the devolved welfare system. Other organizations are developing and compiling program and social indicators that will aid in evaluating the new welfare program. Some groups are undertaking studies to examine how States go about implementing policy changes under the new welfare system. A number of researchers are investigating special topics relating to welfare reform, such as the effects of benefit termination and grant diversion, others are studying the role of child care under the new system, while still others are focusing on certain groups such as the urban poor and immigrants. TABLE L-1.--WELFARE REFORM: SURVEY AND DATA COLLECTION, REPORTS TO CONGRESS, AND EVALUATIONS REQUIRED OR AUTHORIZED BY LAW ------------------------------------------------------------------------ Section \1\ Description Date due ------------------------------------------------------------------------ National surveys ------------------------------------------------------------------------ 414 Census Bureau household Ongoing sample survey to provide data on TANF recipients and other low-income families. Funding $10 million each year, fiscal years 1996-2002 \1\. 105 (PRWORA) Census Bureau required to November 22, 1996 expand data collection to enable analysis of grandparents as primary care givers. 429A Department of Health and Ongoing Human Services (DHHS) national random sample of child welfare containing longitudinal data reliable at the State level for children who have been subject to, or are at risk of, abuse and neglect. Funding $6 million each year, fiscal years 1996- 2002 \2\. 1110 The fiscal year 1998 Interim report appropriation law for DHHS submitted to the includes $5 million for a Appropriations DHHS study of ``welfare Committees by April outcomes.'' The study 1998 should involve State- specific surveys and data sets, survey data on the impact of State waiver programs, and administrative data. ------------------------------------------------------------------------ Evaluations ------------------------------------------------------------------------ 413 DHHS research and Ongoing evaluation of TANF, welfare-to-work programs, and abstinence education programs. Also funds for State-initiated evaluations. Funding: TANF research and evaluations: $15 million per year (fiscal years 1996-2002) \1\; welfare-to-work evaluations: $9 million per year (fiscal years 1998-1999); and abstinence education evaluations: $3 million per year (fiscal years 1998-1999). ------------------------------------------------------------------------ Periodic reports to Congress ------------------------------------------------------------------------ 411(b) TANF annual report by DHHS April 1, 1998 and includes information on annually thereafter whether States are meeting TANF work requirements and program objectives; demographic and financial characteristics of families receiving TANF; and characteristics of TANF Programs. 452(a)(10) Child Support Enforcement 3 months after the annual report by DHHS end of the fiscal includes program costs and year collections; number of cases in which an obligation has been established and in which collections have been made; information on collections for families who receive, or used to received, public assistance. 1637(a) Supplemental Security May 30 of each year Income (SSI) annual report by the Social Security Administration includes current and historical data on applications and allowances, characteristics of recipients, program costs, and 25-year projections of program enrollment. 658K (Child Care and Child Care and Development July 31, 1998 and Development Block Block Grant (CCDBG) annual every 2 years Grant Act of 1990) report based on data thereafter supplied by States, includes number of families served, demographic and financial characteristics of families receiving care, monthly cost of child care services by type of care, number of providers that gave subsidized care. ------------------------------------------------------------------------ Other reports to Congress ------------------------------------------------------------------------ 106 (PRWORA) DHHS report to Congress on Feb. 1997 the status of automated systems to manage TANF Programs. 107 (PRWORA) DHHS report on outcome September 30, 1998 measures as alternatives to TANF work participation requirements. 5001 (Balanced GAO study of the effect of August 5, 1998 Budget Act of 1997) family violence on the need for public assistance. 232 (PRWORA) GAO study of the effects January 1, 1999 of: (1) changes in SSI eligibility for children; (2) extra expenses incurred by families with children; (3) bar on benefits for those who have committed welfare fraud, fugitive felons, and parole violators. ------------------------------------------------------------------------ Information and data reported by the States ------------------------------------------------------------------------ 402(a) TANF State plans must July 1, 1997, and outline State program to every 2 years serve low-income families thereafter with children. 411(a) Quarterly reports by the No later than 45 days States to DHHS that after the end of the include TANF disaggregated fiscal quarter case record information, beginning July 1, aggregated caseload 1997 and every information, and financial quarter thereafter information. 411A Report by the States to the At least 4 times Immigration and annually and at the Naturalization Service request of the INS (INS) furnishing information on illegal aliens in the State. 413 Chief executive officer of May 31, 1998 and each State must submit a annually thereafter statement about the child poverty rate in each State. 469 DHHS must collect and Ongoing maintain up-to-date statistics by State of the child support enforcement caseload and services provided for paternity determination, location of absent parents, and establishment of child support obligations. 453A States must establish Directory must be in directory of new hires place October 1, under the Child Support 1997 (no deadline Enforcement Program. DHHS for when data may make this directory becomes available available for research for research purposes likely to purposes) contributing to meeting the goals of the TANF Program. 658K (Child Care and States must submit Aggregate data due Development Block information on the from the States Grant Act of 1990) demographic and financial December 31, 1997 characteristics of and annually families receiving child thereafter; case- care assistance and level information information about the cost first due from the of care provided to States August 31, families. Information is 1998 and quarterly collected on a monthly thereafter basis, but submitted to DHHS quarterly. ------------------------------------------------------------------------ \1\ Unless otherwise noted, section references are to the Social Security Act. \2\ The appropriation law for the Departments of Labor, Health and Human Services (HHS), and Education (Public Law 105-78) rescinded the fiscal year 1998 funding of $15 million for welfare reform evaluations (section 413) and $6 million for the child welfare sample (429A) and included funding for these activities in a $26 million line item for Social Services and Income Maintenance Research. Note.--PRWORA = Personal Responsibility and Work Opportunity Reconciliation Act of 1996; TANF = Temporary Assistance for Needy Families Program (title IV-A of the Social Security Act). Source: Congressional Research Service. A number of evaluations employing experimental or quasi- experimental methods are also being conducted on various components of States' welfare programs. These include studies of work requirements, time limits, and increased earnings disregards. For example, the Department of Health and Human Services (DHHS) has awarded funds to States to continue evaluations of demonstration programs that were begun under the Aid to Families with Dependent Children (AFDC) Program. These demonstrations were authorized under section 1115 of the Social Security Act.\1\ A national evaluation of welfare-to-work strategies, the former AFDC Job Opportunities and Basic Skills (JOBS) evaluation, is also being undertaken. --------------------------------------------------------------------------- \1\ Under section 1115 of the Social Security Act, States were required to support an evaluation of the impact of program changes granted under waivers. The costs of section 1115 waiver evaluations were matchable by the Federal Government. While States are allowed to continue their waivers under the Temporary Assistance for Needy Families (TANF) Block Grant, they receive no automatic additional funding beyond the block grant for conducting evaluations. However, DHHS has allocated discretionary research funds to protect its investment in these evaluations. In fiscal year 1997, DHHS invited States with waivers to apply for additional funding for up to 95 percent of the costs of continuing the evaluations under TANF. Nine States were awarded funding to continue their evaluations under their existing designs. Most of the waiver evaluations used an experimental approach in which participants were randomly assigned to experimental or control groups. In addition, DHHS has thus far funded grants to 10 States to implement modified evaluations for their new TANF Programs. Under these evaluations, DHHS is supporting projects that will include different evaluation methodologies, such as pre- and posttreatment comparisons of the same groups and quasi-experimental designs. --------------------------------------------------------------------------- In addition, the DHHS 1998 appropriation (Public Law 105- 78) includes $14 million for policy research, of which $5 million has been earmarked for a study on the outcomes of welfare reform. The House and Senate conferees on the bill recommended that the study involve State-specific surveys and data sets, survey data on the impacts of State waiver programs, and administrative data such as food stamp, Social Security, and Internal Revenue Service records. The study should measure outcomes in both low and high economic growth areas of the country. The conferees urged DHHS to submit its research plan to the National Academy of Sciences to provide guidance on research design and recommend further research. DHHS is expected to provide an interim report to the Appropriations Committees by May 13, 1998 (within 6 months of the bill being signed into law). In addition to these efforts, researchers will be monitoring, reviewing, and assessing the evaluations. Some groups are forming to set up information clearinghouses relating to welfare reform. Others are setting up review panels to assess the various research and evaluation efforts that are being undertaken. Some groups are attempting to synthesize the research and evaluations with the purpose of identifying ``best practice'' models of welfare programs; i.e., State or local programs whose design, administration, or implementation are promising or judged to be effective and which other jurisdictions may wish to adapt. The following summary provides an indication of the types of research activities that are now being planned and undertaken. Our intent is not to be exhaustive but to provide summary information about several of the major studies. National Surveys Survey of Program Dynamics (SPD) The welfare reform law of 1996 provides the Census Bureau with $10 million annually, from fiscal year 1996 to fiscal year 2002, to continue to collect data on families in the 1992 and 1993 panels of the Survey of Income and Program Participation (SIPP). Called the Survey of Program Dynamics (SPD), the Census Bureau continuation of SIPP will provide a nationally representative sample of the population to evaluate the impact of TANF. The 1992 and 1993 SIPP panels provide 3 years of longitudinal data which can serve as a baseline for examining TANF impacts. The Census Bureau interviewed members of these panels again in the spring of 1997 and will collect annual information from this sample on household characteristics, income, and program participation. The legislation also requires Census to obtain additional information on the status of children. Assessing the New Federalism--National Survey of America's Families Assessing the New Federalism, being conducted by the Urban Institute in Washington, DC, is a major foundation-funded, multiyear comprehensive assessment of how State's income support systems change as a result of the New Federalism. The project includes a National Survey of America's Families which surveys nearly 50,000 people to provide comprehensive information on the well-being of adults and children as welfare reform is being implemented in the various States. About half the households in the sample have incomes below 200 percent of the poverty line. The survey collects information on: employment, economic hardship and income; child care and child support; health insurance coverage, access to care, and use of care; and the behavior and emotional well-being of children, families, and individuals. The survey is designed to allow State-specific estimates in the 13 States (Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin) selected for intensive study, as well as to allow national estimates. Two waves of interviews are being conducted, one in 1997 and the other in 1999. Child Welfare Longitudinal Study The welfare reform law required DHHS to conduct a longitudinal study to describe the outcomes for children and families who come in contact with the child protection system. The study, which is being conducted under contract by the Research Triangle Institute in North Carolina, will examine system-level and service factors, as well as child and family characteristics, that contribute to child and family outcomes. The study will select a nationally representative sample of 6,000 children upon entry into the child protection system. One hundred agencies will be included in the sample. Information will be collected at baseline and at three annual followup interviews from the children and their care givers, caseworkers, and other agency personnel and service providers. The contract will cover a 6-year period, with field work expected to begin in the spring of 1999. State Policy Monitoring Monitoring State welfare reform The General Accounting Office is conducting a multiyear study that monitors and reports on the progress States are making in achieving the stated objectives of TANF as they reform their public assistance programs for needy families. The study's major questions are: (1) How are States using the design and funding flexibility provided by the law to restructure their assistance programs for needy families?; (2) What approaches are States taking in promoting work opportunities to reduce welfare dependence and what are the major challenges they are encountering?; (3) What approaches and polices are States using to promote personal responsibility to reduce out-of-wedlock pregnancies and births, encourage formation and maintenance of two-parent families, and promote marriage? The study will include site visits to seven States. Information for all 50 States will be drawn from State welfare plans and compilations of State programs from other sources. Survey on status of States' implementation of welfare reform The American Public Welfare Association is conducting a comprehensive baseline survey on the implementation of welfare reform in the States in 14 topical areas relating to welfare reform. Future surveys will focus on specific welfare reform issues such as management, service delivery, and policy decisions; eligibility, work, and benefits; children's issues; and food stamps. Database on State legislative activity The National Council of State Legislatures is developing a database on welfare legislation enacted in each State. Topics to be covered include: time limits, eligibility, impacts on immigrants, benefit levels, responsibility contracts, grant diversion, minor parent provisions, school attendance requirements, work and participation requirements, job creation strategies, transitional services, and administration and oversight. State Policy Documentation Project This project, which is being run by the Center for Law and Social Policy and the Center on Budget and Policy Priorities in Washington, DC, monitors and reports on program changes (administration, eligibility, services/benefits provided) that States legislate and implement. The project covers changes in food stamps, general assistance, TANF, and Medicaid Programs. State child care policies The Urban Institute of Washington, DC, is conducting a project that will summarize the key features of State child care assistance programs for low-income families, including detailed descriptions of reimbursement rates, eligibility rules, and benefit levels, in all 50 States. State policies in effect in 1997 will be compared to those in effect prior to the new child care and Federal welfare reform legislation. State Medicaid Eligibility Decisions and Practices Report The American Public Welfare Association is monitoring State's choices in establishing Medicaid eligibility. The welfare reform law changed State options about Medicaid eligibility. TANF recipients are not automatically eligible for Medicaid as they were under AFDC. Rather, Medicaid eligibility now incorporates the States' standards used for determining AFDC eligibility which were in effect on July 16, 1996. States may also decline Medicaid services for most populations of resident aliens. This study will examine States' choices in these areas. Development of Program and Social Indicators Urban Institute's State database The Urban Institute is developing a State database as part of its ``Assessing the New Federalism'' project. The database will incorporate State-specific data in several broad areas, including: income security, health, well-being, State fiscal and political conditions, demographic characteristics, and social services. The database will contain aggregate measures of budget growth or decline, tradeoffs among major spending categories, discretionary tax increases and decreases, and certain indicators of how programs are changing (e.g., welfare programs' benefit levels, eligibility rules, time limits, and behavioral incentives). Welfare Reform Information and Support Project The Census Bureau has funded a project, being carried out by the Institute for Research on Poverty at the University of Wisconsin, that will devise a framework for classifying welfare reform programs along important program dimensions. The classification scheme will then be applied to welfare reform programs in counties that constitute the primary sampling units (PSUs) in the Census Bureau's Survey of Program Dynamics (SPD) described above. The project will allow linkage of county program parameters with the SPD data. Supporting State efforts to link administrative data systems These DHHS grants fund State efforts to link administrative program data from a variety of low-income assistance programs in order to analyze the collateral impacts welfare reform has on recipients and on other State and Federal assistance programs. While States may pursue any type of data linkages which examine potential impacts of welfare reform on other programs, DHHS has identified several topical areas of interest, including: foster care/kinship care, child abuse and neglect, mental health and substance abuse, teen pregnancy, out-of-wedlock childbearing, domestic violence, and work and work-related support services. South Carolina, Maryland, Wisconsin, Massachusetts, and Mecklenburg County, North Carolina, have received funding for this project. Implementation Studies Assessing the New Federalism--case studies This project is part of the Urban Institute's comprehensive assessment of how State's income support systems change as a result of welfare reform (see above). In addition to TANF, the project will track general assistance, unemployment insurance, employment and training programs, child support policies, child care subsidies, and State tax policies. Case studies will be conducted in 13 States selected for intensive study; the case studies focus on the development and implementation of reform policies. The 13 States are: Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin. The first round of intensive case studies started in late 1996 and were completed in early 1997. A second round in the same locations is planned for 1999. Study of State capacity Being conducted by the Rockefeller School of Government at the State University of New York in Albany, this 4-year study focuses on State governments' capacity to implement social programs, including family and children's services, welfare, work force development programs, and Medicaid. The study focuses on such issues as: planning and management; information systems; integration and coordination of services and facilitation of access and ease of use; and interaction between welfare agencies and other entities including local governments, nonprofits, businesses, and service providers. The study will conduct indepth case studies of up to 18 States. Particular management topics will be addressed in all 50 States. A small number of States that appear to be innovative, or whose implementation appears to be especially interesting, will be chosen, along with the case study States, for analysis of particular topics in greater detail. DHHS has funded a topical study that will examine culture change at the frontline worker level in local offices. Topical Studies States' early experiences with benefit termination The General Accounting Office (GAO) is examining the early experiences of three States in benefit termination (mostly as a result of work requirement sanctions). The study examines implementation problems and effects on families that were terminated. An examination of State diversion programs and impacts on Medicaid This DHHS-funded study being conducted by George Washington University in Washington, DC, will focus on State welfare diversion programs and their effects on Medicaid enrollment. The link between AFDC and Medicaid that existed before welfare reform was one of the primary means of entry into Medicaid for poor families. Formal and informal diversion programs, because they limit the contact that poor families have with welfare offices, may potentially limit their access to Medicaid coverage if steps are not taken to formally assess them for Medicaid eligibility. This study proposes four broad areas of inquiry: (1) How are State diversion programs being conceived and structured?; (2) How are these diversion programs actually being implemented and what are the effects on participants, particularly with respect to Medicaid enrollment?; (3) How might local communities and institutions be affected by diversion programs, in particular by changes in Medicaid enrollment rates?; (4) Can potential strategies for monitoring changes in Medicaid enrollment rates and the effects of such changes over time be developed based on existing databases? The project is expected to be completed by November 1998. A study of diverted welfare applicants in Milwaukee Mathematica Policy Research of Princeton, New Jersey and the Hudson Institute of Indianapolis are conducting a study of applicants who are diverted from receiving welfare. About 250 Wisconsin families in each of 3 sample groups will be interviewed: families that left Wisconsin's welfare rolls during the sample interview; families that began but did not complete the application process (diverted applicants); and families that began and completed the application process. The study seeks to determine what precipitated the applicants' decision to apply for welfare, what was their experience during the application process, when and why they decided to stop the application process, and how they are doing relative to when they decided to apply for assistance. Wisconsin administrative data project on former AFDC recipients This DHHS-funded study by the Institute for Research on Poverty is intended to get early results on the economic and employment outcomes of women who left the AFDC rolls in Wisconsin--a State that has experienced a rapid decline in caseloads over the past 10 years. The study will use linked administrative data, including: (1) AFDC and food stamps data; (2) earnings data from unemployment insurance records; and (3) income and family status data from State tax records. This project will not explain caseload decline; rather it will describe the characteristics and outcomes of two groups of women who have left the AFDC Program in Wisconsin: one group of recipients in January 1995, and another group of recipients in January 1990, who left AFDC over the course of the year. Evaluation of community-based job retention programs This DHHS-funded project, being carried out by the Pittsburgh Foundation, includes two phases. In phase 1, the project will provide a detailed implementation analysis and short-term outcome analysis for participants receiving various mixes of job retention and postemployment services. The services will be provided to about 700 employed TANF recipients through 5 community-based neighborhood service organizations in Pittsburgh. A common set of services will be provided but within the context of somewhat different existing service delivery systems among the community-based neighborhood service organizations. Phase 2, in project years 2 and 3, will include State-funded job retention programs. Assessing effective welfare-to-work strategies for domestic violence victims and survivors in the Options/Opciones Project This DHHS-funded research project by the Taylor Institute of Chicago, Illinois will study effective strategies for addressing the needs of abused women as they try to enter the labor force. This project will document the needs of battered girls and women on welfare and will identify successful strategies employed to eliminate violence and exit welfare. The research will focus on services provided through Options/ Opciones in partnership with the Illinois Department of Public Aid in one major welfare department office on the west side of Chicago. Options/Opciones will provide the following services: case management; preemployment training groups; and weekly peer support groups for those who have completed the preemployment training. Child care under welfare reform The General Accounting Office is undertaking a study that will focus on several areas relating to child care and welfare reform. Among other issues, the study examines how States are allocating resources to provide child care to welfare families, to families transitioning off welfare, and to working poor families. The study also examines State efforts to increase the supply of child care given the growing demand for child care under welfare reform and whether States are changing standards for child care providers in response to welfare reform. The study will rely on site visits, telephone interviews, State plans filed with DHHS, and results from other research. Child care research partnerships DHHS has awarded five Child Care Research Partnership projects (California Child Care Resource and Referral Network, Columbia University, Harvard University, Linn Benton College, and Wellesley College) to examine three broad questions: (1) How do child care policies and market dynamics influence child care demand, supply, and outcomes for low-income families, particularly welfare clients, those moving from welfare to work, and the working poor?; (2) How do child care opportunities and constraints affect the lives of low-income families and children?; and (3) What factors affect the cost, quality, and delivery of subsidized child care services? Collectively, these partnerships involve many local, State and national partners, including university research teams; agencies which administer child care, welfare, education, and employment programs; major corporations; national research and professional organizations; many community agencies and organizations; child care providers; and families from a wide spectrum of American society. Participating States include: Alabama, Connecticut, Florida, Illinois, Maryland, Massachusetts, New Jersey, New York, and Oregon. National study of low-income child care This DHHS-funded project being carried out by Abt Associates of Cambridge, Massachusetts will study the low- income child care market in 25 communities in several States. A substudy will examine the license-exempt family care market in neighborhoods drawn from these communities. The studies will examine how significant shifts in welfare policy and programs affect the child care market for welfare recipients and the working poor at the community level. They will examine the impact that subsidies have on the types, amounts and quality of child care available, on children's child care placements, and on family decisionmaking. Additionally, the studies will report on the largely unknown license-exempt family care market for both preschool and afterschool cohorts. Improving the States' capability to evaluate child care policy options as a component of their welfare-to-work strategies The Urban Institute and Mathematica Policy Research are conducting a DHHS-funded project to develop an expanded simulation model that would enable State welfare administrators to consider the interactions between child care assistance and welfare reform policies. Currently, Mathematica Policy Research is nearing completion in the development of a model to assess alternative welfare reform policies--STEWARD II. This project would add a component to STEWARD II that allows States to assess child care funding policies as they intersect with welfare reform. The role of child care in low-income families' labor force participation The Urban Institute and Mathematica Policy Research are also conducting a DHHS-funded project that will develop research designs to address the relationship between child care and labor force attachment. The major work under this contract will consist of a series of working papers that critically evaluate research related to child care and labor force attachment, and that develop the rationale for the factors included in possible research designs. The devolution of welfare: assessing the effects on communities, families and young children The University of California-Berkeley and Yale University Child Study Center are collaborating on a study that will focus on the effects of welfare reform on early child development, child care, and early education in eight sites in four States. In years 1 and 3 of the study, investigators will inventory local welfare agencies at each site to determine changes in Federal and State welfare polices and services. A sample of 60 low-income families with a birth in the last year will be drawn from each community and interviewed annually. Data will include economic and demographic information as well as information on child care (including quality of care). The study will use various tests to assess the physical, social, and mental development of children in the sample families. Study of the effect of family violence on the need for public assistance Legislation passed in 1997 (Public Law 105-33) requires the General Accounting Office to conduct a study on the effect of family violence on the use of public assistance programs, particularly on how family violence prolongs the use of or increases the need for assistance. Devolution and urban change The Manpower Demonstration Research Corporation is conducting a foundation-funded, 5-year study of the impact of welfare reform on urban areas. Four urban areas are included in the study: Los Angeles, Miami, Cleveland, and Philadelphia. The study will focus on three broad questions: (1) How are new State policies implemented by the local welfare office, and how do they affect the structure of the safety net and the mission, message, benefits, eligibility, and service mix offered by public agencies?; (2) How do the new policies affect the role and activities of community organizations?; and (3) How do the new policies affect the dynamics of program participation, employment, and the overall well-being of low-income families with children? The study will rely primarily on administrative data from TANF, unemployment insurance, and food stamps, going back over time as well as current data. Surveys will be administered to at least two cohorts of people who received food stamps in 1995 and 1997. Between 750 and 1,000 people per cohort per city will be selected for the survey. The project will include four additional separate studies in each of the areas: an implementation study of how the mission, message, benefits and service mix offered by welfare agencies change; a study of neighborhood indicators; an ethnographic study of approximately 30-50 families in each city; and an indepth study of nonprofit organizations and community institutions (e.g., churches, schools) and their responses to welfare reform. Information for action: understanding the challenges of welfare reform from a neighborhood perspective The Urban Institute's National Neighborhood Indicators Project will collect neighborhood-level data about characteristics of welfare recipients, the jobs that are potentially available to them, and the impact of welfare reform in six cities (Atlanta, Boston, Cleveland, Denver, Providence, and Washington). The data will be incorporated into geographic information systems, which can be searched by zip code or census tract, for use by communities and local decisionmakers. The database will help researchers address such questions as: Where do the people live who are at greatest risk of becoming welfare recipients?; How does welfare reform affect people and neighborhoods over time?; Are recipients able to get and keep jobs, obtain health coverage, and improve their overall family well-being? Multicity study of the effects of welfare reform on children A consortium of universities including Johns Hopkins, Harvard, Pennsylvania State, and Chicago are conducting an interdisciplinary study of the effects of welfare reform on children and youth living in three cities (San Antonio, Boston, and Chicago). The study will collect information on child and youth well-being which includes measures on health and disability, cognitive development, emotional development, school achievement, and fertility in order to address such issues as: the effects on children of being in a family that is receiving welfare; the effects on children of living in a single-parent working family; the implications of welfare reform for persons with disabilities; the effects of poverty on children; the importance of income per se; and factors that mitigate or exacerbate the impacts of having low income. Interviews, direct assessments of the well-being of children, and administrative records provide the data on which the study is based. The study will follow families over time, and new families may be added. Detailed information on the rules of the local welfare system in the three sites will be collected to help determine whether and how impacts relate to welfare policies. Neighbors, service providers, and welfare reform in Los Angeles County This Rand project will examine neighborhood variation in the availability of public and private social services throughout Los Angeles County early in the implementation of welfare reform. The study will also investigate how agencies are adapting to the current and anticipated changes in demand for their services as a result of welfare reform. The focus of the inquiry will be on services directed toward children and families. Welfare reform studies and analyses (rural TANF) The purpose of this DHHS-funded project by Eastern Washington University in Cheney, Washington, is to conduct an indepth process evaluation of family response to the implementation of welfare reform in three remote rural counties of northeastern Washington State: Stevens, Ferry and Pend Oreille. The study will examine family strategies to attain economic self-sufficiency, family support from social networks, and assistance provided by community agencies, including transportation and child care. Restricting welfare eligibility for legal immigrants The General Accounting Office is preparing a report that will describe the impact of changes in the welfare law on immigrants, particularly on their use of the Supplemental Security Income, Medicaid, TANF and Food Stamp Programs. GAO will examine: (1) how many and what types of legal immigrant households have their Federal welfare benefits terminated; (2) how States implement the limits on legal immigrants' eligibility for Federal welfare benefits (both optional and mandatory); (3) what plans States have to provide assistance to legal immigrants; and (4) what major implementation issues and challenges Federal agencies and States face in administering the provisions restricting welfare assistance to legal immigrants. Study on the economic and health status of immigrants, their communities, and the organizations which serve them The Urban Institute in collaboration with the University of California at Los Angeles is conducting a study, funded by DHHS, the Immigration and Naturalization Service, and the U.S. Department of Agriculture, which profiles the status of immigrants in Los Angeles and New York City with regard to health, employment, and participation in programs administered by public and private agencies. The study will pay special attention to distinguishing different categories of immigrants and drawing comparisons between them and the native population. To the extent possible, the project will explore the impacts of welfare reform on immigrants, communities, and the organizations which serve them, with attention to both individual and institutional adaptations. The impact of the loss of food stamps among current legal immigrants will be a particular focus of the study. Welfare to work: monitoring the impact of welfare reform on American- Indian families with children This DHHS-funded project at the Washington University School of Social Work in St. Louis is addressed primarily to monitoring the implementation and assessing the impact of welfare reform on Indian reservations. The study will examine the prospects for economic independence of parents currently on welfare and the potential changes in the service structure and socioeconomic conditions on reservations brought about by State and tribal responses to TANF. Descriptive information on TANF Program implementation and, if available, evaluative data on short-term program impacts will also be generated by the study. Arizona was chosen as the study site because it has the largest reservation-based American Indian population, with 21 reservations of varying characteristics. Partner and father involvement in the lives of low-income first-time mothers and their children This DHHS-funded project at Children's Hospital in Denver will investigate the role that fathers play in improving the material, emotional, and developmental well-being of low-income women and children. The project consists of a set of intensive secondary analyses using data from three longitudinal randomized experiments of a program of prenatal and infancy home visitation serving first time mothers from various ethnic and racial groups. All three experiments used home visitation by nurses or health paraprofessionals to assist the first time mothers; the projects were conducted in Elmira, New York; Memphis, Tennessee; and Denver, Colorado. Most of the mothers were low income, and previous findings from this evaluation demonstrate many positive impacts of the program on women's maternal life course, caregiving, and child outcomes. Data has been collected by these studies on the role of fathers and their possible relation to program effects. This project is designed to analyze those data. State Welfare Demonstration Programs The following 20 projects are being funded by DHHS as components of the State Welfare Reform Evaluation Project, under which States continue, possibly with modifications, the evaluations of their section 1115 AFDC waiver demonstrations. Arizona's EMPOWER Program Abt Associates is evaluating Arizona's EMPOWER (Employing and Moving People Off Welfare and Encouraging Responsibility) Program, a continuation of the State's waiver demonstration program. EMPOWER's main provisions are: (1) time-limiting cash assistance for adult recipients; (2) withholding cash assistance for additional children born while the family receives assistance; (3) extending transitional medical and child care benefits from 12 to 24 months; (4) eliminating the 100 hour per month work restriction for two-parent family eligibility; (5) allowing individual development (savings) accounts; (6) requiring unwed minor parents to live with a responsible adult; (7) requiring 13- to 16-year-old parents (including pregnant girls) to participate in JOBS/TANF; (8) imposing automatic sanctions for initial noncompliance with JOBS/TANF Program requirements; and (9) implementing a limited subsidized employment pilot called JOBSTART. The study randomly assigned 5,829 applicants and recipients to control and experimental groups. The evaluation will study a variety of impacts, including: income and employment; program participation; program duration; family structure (including fertility) and stability; child well-being; and employer health benefit provision. The study will also include process studies on how the demonstration was implemented and a cost-benefit analysis. The evaluation is scheduled to be completed by June 2002. Connecticut's Jobs First Program The Manpower Demonstration Research Corporation is evaluating Connecticut's Jobs First Program, a continuation of the State's waiver demonstration program. Jobs First includes: (1) a 21-month time limit for nonexempt recipients; (2) 24 months of transitional Medicaid; (3) liberalized earnings and asset disregards; (4) limited eligibility for children born while the family receives assistance; (5) a strong emphasis on job search as a labor market screen and use of employability assessment only if a job has not resulted from extensive search; and, (6) progressive sanctions for noncompliance. Jobs First was one of the first programs to impose a time limit in a major urban area (New Haven and Hartford). The study randomly assigned 6,000 recipients to Jobs First or traditional AFDC. The study examines implementation and staff and recipient attitudes toward the changes in the welfare system under welfare reform. It will examine the impacts of Jobs First on: work, earnings and poverty status; welfare receipt and costs; job-retention and recidivism; family composition and stability; the well-being of children; child support and other sources of income; the use and perceived quality of program services and child care; attitudes toward work, welfare, and Jobs First; and, material hardship especially as hardship may be affected by time limit policies. The evaluation's scheduled completion date is April 2002. Florida's Family Transition Program The Manpower Development Research Corporation is evaluating Florida's Family Transition Program (FTP) in Escambia County, a continuation of the State's waiver demonstration program. Provisions of the FTP demonstration include: a 24-month time limit on assistance (with some exceptions); increased earned income disregards; requiring parents of children age 6 months or older to participate in employment and training programs; a higher motor vehicle asset limit; increased eligibility for two-parent families; and school attendance requirements for teenagers. A total of 5,430 AFDC applicants who met FTP eligibility criteria were randomly assigned to either a control group or experimental group. The study examines such questions as: Is the FTP Program being implemented as intended?; How do program staff across all components of FTP attempt to achieve the multiple goals implicit in the program model?; How do welfare recipients respond to the new package of rules and services?; Who reaches the time limits?; What happens prior to and at the point when time limits are reached, and after benefits are terminated?; Does FTP increase the rate at which welfare recipients participate in and complete various education and employment activities?; Does FTP lead to increases in employment and earnings?; Does FTP reduce welfare receipt?; Is FTP cost effective? The FTP evaluation, which began in 1994, is scheduled to be completed by April 2000. Illinois' Youth Employment and Training Initiative Illinois State University is completing an evaluation of the State's Youth Employment and Training Initiative (YETI) operated as a welfare reform demonstration from November 1993 until July 1997 when TANF was implemented in the State. YETI focused on inner-city youth in welfare families and provided counseling and classes designed to: help participants stay in high school and graduate; increase life skills and self-esteem and reduce the incidence of pregnancies and substance abuse; increase job-readiness and vocational skills; and facilitate the transition from school to work. The overall objective was to reduce the likelihood of welfare dependency among inner-city youth. The evaluation is a random assignment experiment in which 1,000 students in 3 high schools who volunteer to participate were randomly assigned to an experimental or control group. The major impact questions center on whether the interventions will help participants complete high school, obtain and hold regular jobs, avoid the pitfalls of substance abuse or early parenthood, and achieve economic self-sufficiency. The process study centers on whether the program can be effectively implemented and operated in inner-city schools. The cost/ benefit analysis determines the cost effectiveness of the program for government, participants and the general public. The evaluation is expected to be completed by July 2000. Indiana's IMPACT Program Abt Associates is evaluating Indiana's IMPACT (Indiana Manpower Placement and Comprehensive Training) Program, a continuation of the State's waiver demonstration program. IMPACT includes the following major policy provisions: grant diversion; a 24 month benefit time limit for adults determined to be JOBS eligible; a family benefit cap; child immunization and school attendance requirements; and a higher resource limit for AFDC eligibility. Under TANF, effective April 1997, two major policy changes were added under IMPACT: the 24 month welfare benefit time limit was expanded to apply to all mandatory employment or training cases, and adults with children over the age of 12 weeks and under the age of 3 years are no longer exempt from participating in employment or training activities. Applicants and recipients were randomly assigned to control and experimental groups. Central research questions to be explored include: Were the reforms successfully implemented?; What organizational characteristics, service features, and aspects of the demonstration environment determine its impacts?; How do these reforms affect benefit payments, income, self-sufficiency, children's well-being, and family stability?; What are the consequences of the changes on welfare entry and the size and composition of the welfare caseload under TANF?; What trends should the State expect in levels of benefit payments and services needed?; What happens to individuals after they leave welfare, especially as a result of reaching the time limit? The evaluation is scheduled to be completed by April 2002. Iowa's Family Investment Program Mathematica Policy Research is evaluating Iowa's Family Investment Program (FIP), a continuation of the State's welfare waiver demonstration. The FIP combines program changes designed to ease a family's transition from welfare to work with strict requirements that recipients participate in the development and execution of a social contract, the family investment agreement (FIA). The FIA details the steps a family will take to become self-sufficient and establishes a timeframe for doing so. Families which opt not to develop an FIA or fail to follow through with the self-sufficiency plan outlined in the agreement are placed on a 6-month limited benefit plan (LBP) which leads to the complete loss of cash assistance for a following 6-month period. Applicants and recipients were randomly assigned to control and experimental groups. The study seeks to determine the long-term impacts of the FIP initiatives on: cash assistance, food stamp and Medicaid recipiency and benefit amounts; participation in employment and training activities; employment, earnings and total family income (including the FIP cash grant, food stamps, earnings, and child support); the incidence of poverty; and the number and duration of welfare spells, frequency of recidivism, and the number and duration of employment spells. In addition to an impact analysis, the study will also include process studies on how the demonstration was implemented and a cost-benefit analysis. The scheduled completion date is April 2000. Iowa's limited benefit plan Mathematica Policy Research is conducting a study, using funds from DHHS and foundations, that is one of two add-on studies to Iowa's AFDC waiver evaluation. This study will examine families, sanctioned under Iowa's Family Investment Program, that are placed in the limited benefit plan (LBP), which leads to complete loss of cash assistance for a following 6-month period. The study will describe the experiences and outcomes of welfare cases that have been assigned to the LBP more than once. The study will be based on administrative and survey data. Major research questions for the study of repeat LBP assignments are: To what extent do recipients assigned to the LBP for a second time understand the family investment agreement, a social contract required under Iowa's regular TANF Program?; After experience with LBP, what efforts do clients make to meet program requirements?; What events, barriers and circumstances contribute to their failure to meet these requirements following their first LBP?; What are the nature of changes in family income, employment, housing and stability following benefit termination under a second LBP?; What are the demographic characteristics of cases assigned to the LBP for a second time? The study is scheduled to be completed by September 2000. Iowa study of postemployment services Mathematica Policy Research is also conducting the other add-on study to Iowa's AFDC waiver evaluation. This study examines postemployment services by describing and comparing standard and enhanced postemployment services and assessing their contribution to clients' progress toward self- sufficiency. The study, which will be based on administrative and survey data, scheduled for completion by September 2000, documents the processes for developing, implementing, and delivering postemployment services. Maryland's Family Investment Program The University of Maryland School of Social Work is conducting a statewide process study to examine and document frontline assessment and allocation practices under TANF. The objective of this analysis is to help clarify the relationship between agency factors and county-level characteristics by describing county-level variation in frontline procedures. The evaluation is scheduled to be completed by September 2000. Minnesota's Family Investment Program The Manpower Demonstration Research Corporation is evaluating Minnesota's Family Investment Program (MFIP). Major features of MFIP include: making work pay for public assistance families, primarily by using increased disregards to decrease grant reduction for earned income until an income threshold is achieved; requiring long-term public assistance recipients to participate in intensive employment/training services, including mandatory case management; and consolidating benefits and simplifying rules and procedures for TANF, general assistance and food stamps. Impact analyses will be based on random assignment methodology comparing experiences of control and experimental group members with respect to service participation, welfare receipt, employment, earnings, and income and poverty. The evaluation is scheduled to be completed by December 2000. Minnesota's WorkFIRST Program Maximus, Inc. of McLean, Virginia, is evaluating Minnesota's WorkFIRST and the State's TANF Program. The TANF Program is a continuation of the Minnesota Family Investment Program (MFIP) welfare reform demonstration. WorkFIRST uses a mandatory labor force attachment strategy which includes providing benefits through vendor payments. MFIP represents what might be characterized as a progressive labor force attachment model in which work is rewarded through a more liberalized structure of income disregards for TANF recipients. The study will consist of both a process and impact evaluation, the latter employing a quasi-experimental, pretest/ posttest comparison group design. The major research questions for the study are: Will the WorkFIRST Program have a greater impact on moving first-time applicants into employment and promoting long-term self sufficiency, or will MFIP's ``make work pay'' strategy prove more effective in moving people off welfare?; Is WorkFIRST more effective in preventing first-time applicants from becoming dependent on public assistance?; and Do tough sanctions reinforce the WorkFIRST message of personal responsibility? The evaluation is scheduled to be completed by September 2002. Nebraska's Employment First Program This project evaluates State welfare policies initiated under a welfare reform demonstration begun on November 1, 1995 and now incorporated statewide under the Employment First (TANF) Program. These policies include: intensive case management based on assessment and a self-sufficiency contract/ employability plan; time-limiting cash assistance to 2 out of every 4 years; strict full-family sanctions imposed only as a last resort and after careful review and mediation; temporary or permanent good cause exemptions from participation and time limits; liberalized earnings and assets disregards; and 24- month child care and Medicaid transitional benefits. The evaluation will primarily focus on implementation and operation of the program and will include a special study to assess successful case management. A separate impact evaluation will be carried out comparing an intensive up front job search strategy with an up front assessment approach. The evaluation is scheduled to be completed by September 2002. New Hampshire's Employment and Training Program An implementation study and a separate impact study will be fielded to evaluate New Hampshire's employment focused welfare program. New Hampshire's TANF Program includes up front job search followed by a client assessment and placement into one of two types of employment program groups if employment is not found within 4 weeks of job search. The groups are divided according to employability of clients, and services are structured appropriately. Participants who do not find employment after a 26-week job search component enter a work benefits program, which includes activities such as on-the-job training and subsidized employment. Special needs payments are available to address employment obstacles, and individuals are exempt from employment requirements if personally providing care for a child under age 3. The impact evaluation will rely on administrative data to compare cohorts of persons who were on or entered AFDC between January 1992 and December 1993 and would have been designated State TANF-mandatory participants with cohorts of persons who were in TANF or entered TANF between October 1997 and September 1999. The evaluation will seek to answer such questions as: Does New Hampshire's welfare reform program impact State welfare case dynamics, including exit rates, length of benefit receipt and recidivism?; Does the program have an impact on welfare caseloads compared to forecasted caseload values such as new welfare case openings, case closings, denials and withdrawals, ``child only'' cases, earned income cases, and benefit payments?; and Does the program affect welfare-related measures such as foster care placements, child abuse and neglect, homelessness, paternity establishments, and child support collections? The impact evaluation is scheduled to be completed by September 2002. North Carolina's Work First Program Maximus will evaluate North Carolina's Work First Program which requires TANF families to work to support themselves and their families. Through Work First, parents can receive short- term training and families can get child care and other services to assist them in becoming self-sufficient. Work First emphasizes three strategies: (1) diverting families from the welfare system by helping them cope with unexpected emergencies; (2) shortening the length of time that families receive assistance by making work mandatory and time-limiting assistance; and (3) helping families who leave TANF to stay off by encouraging them to save and providing supportive services. The evaluation is scheduled to be completed by September 2000. North Dakota's Training, Education, Employment and Management Project Berkeley Planning Associates will produce a process evaluation of North Dakota's Training, Education, Employment, and Management (TEEM) Project. TEEM is a welfare reform demonstration initiated prior to implementation of the State's TANF Program. The TEEM project consolidates TANF and home energy assistance into a single cash assistance program. TEEM includes the following major provisions: a social contract, increased work incentives, sanctions, raised asset limits and incentives for family stability and marriage. The process study is scheduled to be completed by September 2000. Achieving change for Texans The University of Texas will continue the evaluation of the State's original welfare reform demonstration. The demonstration consists of three major components: (1) a number of policies implemented statewide addressing such recipient responsibilities as immunization for children, school attendance and adhering to a personal responsibility agreement; (2) a number of policies implemented in counties operating JOBS, the primary feature of which is differential benefit time limits based on consideration of work experience and need for education; and (3) several policy options implemented in four counties providing for individual development accounts and fill-the-gap budgeting. The evaluation will produce implementation and impact studies. Major research questions include: How does the differential time limit function taking into consideration work experience and education for setting the duration of the limit?; Will personal responsibility measures help TANF recipients become self-sufficient faster?; and Can policies designed to promote and reward work help the process of self- sufficiency? For all policies tested there are three general questions to be researched: Did welfare use decline?; Did clients become self-sufficient?; and Did clients who went off cash assistance stay off longer than clients subject to prior law policies? The State will also conduct a survey of TANF recipients who have left the rolls or been diverted at application through the offering of a one-time payment of $1,000 in lieu of regular TANF with no reapplication for benefits for 1 year. The evaluation is scheduled to be completed by March 2002. Vermont's Welfare Restructuring Project The Manpower Demonstration Research Corporation is continuing the evaluation of Vermont's Welfare Restructuring Project which was begun as a waiver demonstration project. One of the key features of Vermont's welfare demonstration is a time limit that requires some adult recipients to participate in community work experience after 15 (two-parent families) or 30 months (single-parent families) receiving cash assistance. Other policies of the program include asset and disregard changes, requiring that minor parents live at home or in a supervised living situation, and a requirement that parents with temporary disabilities participate in rehabilitation and training programs. The evaluation will include an implementation/process study, an impact study, and a cost-benefit study. Applicants and recipients were randomly assigned to experimental or control groups for the impact study. Major questions to be addressed by the impact study include: To what extent does the interaction of financial incentives with time limits for work and related activities increase unsubsidized employment and earnings, increase job retention, and affect the types of jobs and wage levels?; What are the effects attributable to changing financial incentives for AFDC single parents and two-parent cases?; To what extent does the program reduce or increase welfare benefits?; and What is the impact of WRP on various subgroups of applicants and recipients? The evaluation is scheduled to be completed by April 2002. Virginia's Independence Program (VIP) The Virginia Polytechnic Institute will examine the State's TANF implementation which continues policies begun under the State's welfare reform demonstration. Among the policies to be studied under the State's VIP Program are: diversionary assistance, a family cap, time-limited assistance, personal responsibility agreements, expanded earned income disregards, school attendance requirements, paternity establishment rules, requiring minor parents to live in adult settings, child immunization requirements, incentives promoting individual savings accounts, and transitional Medicaid and child care benefits. The evaluation will include a description of cases reaching the time limit, and an assessment of the early impacts based on information from a discontinued random assignment experiment. Major questions to be addressed include: What happens to families after they reach the time limit?; What are the effects of post employment and job retention services offered by the State?; What are the implications of the growing number of child-only cases the State is finding on its rolls?; What are the issues specific to localities where unique innovations have been introduced in the State?; How does diversionary assistance work in a limited number of localities?; and What are the dynamics of the Virginia caseload over time under the TANF regime? The evaluation is scheduled to be completed by September 2002. Wisconsin's Pay for Performance/Self-Sufficiency First Maximus is evaluating the Pay for Performance/Self- Sufficiency First Programs in Wisconsin. Under the Self- Sufficiency First Program, all nonexempt welfare recipients must complete 60 hours of work and training during a 30-day application period or face denial of benefits. Under the Pay for Performance Program, individuals not exempt from work and training requirements and who are not working 30 hours or more per week must participate in up to 40 hours of employment and training activities per week. Cash benefits are reduced for each hour of nonparticipation in employment and training without good cause by the Federal minimum hourly wage with no corresponding adjustment in the food stamp allotment. In any month when the hours of participation in employment and training activities fall below 25 percent of assigned hours, the assistance grant is reduced to $0 and the food stamp allotment to $10. Applicants and recipients were randomly assigned to control and experimental groups. Major questions to be addressed by the evaluation include: Do the program policies improve employment rates, length of employment, amount of earned income, hours worked per month, child support collections, total family income, and accumulated savings?; Do these policies affect public assistance participation and program costs?; Does the program affect participation in employment and training activities?; Do these program rules and policies affect marriage and separation rates and use of foster care?; and Does the demonstration affect the incidence of reported child abuse and neglect and health insurance status of children? The evaluation is scheduled to be completed by April 1999. Federal studies of impact of welfare reform on children DHHS is funding child impact studies that will augment the welfare reform demonstration evaluations in five States (Connecticut, Iowa, Minnesota, Indiana, and Florida). The studies will assess the effects of different welfare reform approaches on child well-being. The ongoing demonstration evaluations focus primarily on adult behaviors and outcomes, such as changes in earnings and welfare dependency. This project adds detailed data on children to these evaluations. Impacts on children will be measured through random assignment of families in each site. Treatment groups subject to welfare reform policies will be compared with control groups subject to former AFDC policies. Child outcomes data, focusing on children ages 5-12, will be collected through surveys and administrative records. Major research questions addressed by the studies include: What are the effects of alternative approaches to welfare reform on child well-being, including school achievement, behavioral problems, and health status? What intervening mechanisms, such as the quality and regularity of the home environment, child care arrangements, and parental employment and income, affect these outcomes? The project periods vary from 2 to 4 years. Other Evaluations National evaluation of welfare-to-work strategies (formerly the JOBS evaluation) and child outcomes substudy The Manpower Demonstration Research Corporation and Child Trends of Washington, DC, continue to conduct this major study, funded by DHHS and foundations (with support from the Department of Education). Under the JOBS Program, random assignment of recipients to experimental and control groups began in March 1991 and ended in December 1994. Approximately 55,000 sample members are being followed for 5 years from the time they entered the study using administrative records (unemployment insurance earnings and AFDC payment data). A subsample of about 10,000 individuals from all research groups in all sites was interviewed 2 years after they entered the study; some of these individuals will be reinterviewed after 5 years. In three sites, Child Trends, Inc. will be measuring outcomes for children who were between the ages of 3 and 5 when their mothers entered the study. Major questions the study is designed to answer include: How much can welfare-to-work programs contribute to reducing welfare dependency and increasing employment? Do certain approaches work better than others--for example, human capital development (i.e., education and training) versus labor force attachment (e.g., job search and other employment-focused activities)?; How do the children of welfare recipients fare under welfare-to-work programs?; and What do welfare-to-work programs cost? Bridges to work Public/Private Ventures of Philadelphia has undertaken a $17 million demonstration program and evaluation of a transportation-based antipoverty strategy. The project is funded primarily by the Department of Housing and Urban Development (HUD) and major foundations. The demonstration includes three program elements: a metropolitan placement mechanism that gives participants information about the suburban labor market and links them to existing private sector jobs; a targeted commute component that connects inner-city workers to previously inaccessible job destinations through increased public transportation and/or van pooling; and support services to mitigate demands created or exacerbated by the long commute. A random assignment evaluation and implementation study will be conducted. The demonstration is being conducted in five sites: Baltimore, Chicago, Denver, Milwaukee, and St. Louis.