P.L. 114-254, the further continuing resolution for FY2017, extended funding and program authority for the Temporary Assistance for Needy Families (TANF) block grant through April 28, 2017. Though several bills that would have changed TANF policies were reported by the House Ways and Means Committee to the full House during the 114th Congress—and one bill passed the House—none were enacted.
The TANF block grant funds grants to states, tribes, and territories for providing benefits, services, and activities to broadly address both the effects and root causes of childhood economic and social disadvantage. It was created in the 1996 welfare reform law, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (P.L. 104-193), which culminated decades of debate over how to change public assistance programs for needy families with children. Most TANF policy dates back to the 1996 welfare law, as a full reauthorization of the block grant has never been enacted.
On June 21, 2016, the House passed H.R. 5170, a bill that would have established a demonstration project for “social impact partnership projects.” That bill also would have extended TANF and mandatory child care block grant funding for one year (FY2017) and revised TANF-related research funding. The bill did not pass the Senate.
H.R. 5170, as it passed the House, would have established a mechanism through which state and local governments could apply to the Secretary of the Treasury for demonstration projects, which would have been called “social impact partnership projects.” These projects would have used funds provided through philanthropic and other private-sector partnerships to finance social programs to meet specified social goals. Under this type of financing, the government would “pay off” investors only when a program evaluation demonstrates that desired outcomes are met. H.R. 5170 would have funded social impact partnership demonstrations at $100 million, financed from a set-aside from FY2017 TANF contingency funds. The bill would have required that at least 50% of funds provided for agreements awarding funding for social impact partnership projects be used for initiatives that directly benefit children. As it passed the House, H.R. 5170 would have also extended TANF and mandatory child care block grant funding through FY2017 and revised TANF research requirements and funding.
H.R. 2990, as reported by the House Ways and Means Committee, would have created a demonstration project for subsidized employment programs for TANF assistance families. Subsidized employment programs are those where public funds are used to pay all or part of the wages, benefits, and other costs of employing an individual. Subsidized employment can be a part of TANF programs under current law. However, except for a brief period when such programs were financed from special funds enacted in response to the 2007-2009 recession, subsidized employment has been little used in TANF. The subsidized employment demonstration would have been funded at $100 million, financed from an additional set-aside from FY2017 TANF contingency funds. The bill was not considered by the full House.
The House Ways and Means Committee also reported additional TANF bills: H.R. 2959 would have altered the rules for the TANF state spending requirement, H.R. 2966 would have added reducing child poverty as a statutory TANF goal, and H.R. 2952 would have established new employment outcome performance measures for TANF. These bills were not considered by the full House.
P.L. 114-254, the further continuing resolution for FY2017, extended funding and program authority for the Temporary Assistance for Needy Families (TANF) block grant through April 28, 2017. Though several bills that would have changed TANF policies were reported by the House Ways and Means Committee to the full House during the 114th Congress—and one bill passed the House—none were enacted.
The TANF block grant funds grants to states, tribes, and territories for providing benefits, services, and activities to broadly address both the effects and root causes of childhood economic and social disadvantage. It was created in the 1996 welfare reform law, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (P.L. 104-193), which culminated decades of debate over how to change public assistance programs for needy families with children. Most TANF policy dates back to the 1996 welfare law, as a full reauthorization of the block grant has never been enacted.
On June 21, 2016, the House passed H.R. 5170, a bill that would have established a demonstration project for "social impact partnership projects." That bill also would have extended TANF and mandatory child care block grant funding for one year (FY2017) and revised TANF-related research funding. The bill did not pass the Senate.
H.R. 5170, as it passed the House, would have established a mechanism through which state and local governments could apply to the Secretary of the Treasury for demonstration projects, which would have been called "social impact partnership projects." These projects would have used funds provided through philanthropic and other private-sector partnerships to finance social programs to meet specified social goals. Under this type of financing, the government would "pay off" investors only when a program evaluation demonstrates that desired outcomes are met. H.R. 5170 would have funded social impact partnership demonstrations at $100 million, financed from a set-aside from FY2017 TANF contingency funds. The bill would have required that at least 50% of funds provided for agreements awarding funding for social impact partnership projects be used for initiatives that directly benefit children. As it passed the House, H.R. 5170 would have also extended TANF and mandatory child care block grant funding through FY2017 and revised TANF research requirements and funding.
H.R. 2990, as reported by the House Ways and Means Committee, would have created a demonstration project for subsidized employment programs for TANF assistance families. Subsidized employment programs are those where public funds are used to pay all or part of the wages, benefits, and other costs of employing an individual. Subsidized employment can be a part of TANF programs under current law. However, except for a brief period when such programs were financed from special funds enacted in response to the 2007-2009 recession, subsidized employment has been little used in TANF. The subsidized employment demonstration would have been funded at $100 million, financed from an additional set-aside from FY2017 TANF contingency funds. The bill was not considered by the full House.
The House Ways and Means Committee also reported additional TANF bills: H.R. 2959 would have altered the rules for the TANF state spending requirement, H.R. 2966 would have added reducing child poverty as a statutory TANF goal, and H.R. 2952 would have established new employment outcome performance measures for TANF. These bills were not considered by the full House.
P.L. 114-254, the further continuing resolution for FY2017, extended funding and program authority for the Temporary Assistance for Needy Families (TANF) block grant through April 28, 2017. Though several bills altering TANF policies were reported by the House Ways and Means Committee to the full House during the second session of the 114th Congress—and one bill passed the House—none were enacted.
On June 21, 2016, the House passed H.R. 5170, a bill that would have established a demonstration project for "social impact partnership projects." That bill would also have extended TANF and mandatory child care block grant funding for one year (FY2017) and revised TANF-related research funding. The bill was not considered in the Senate.
In May 2016, the House Ways and Means Committee approved separate bills that would have made additional policy changes to TANF by (1) establishing a demonstration program for subsidized employment for TANF assistance recipients and noncustodial parents; (2) making revisions to TANF state spending requirements; (3) adding child poverty reduction to the statutory purpose of TANF; and (4) establishing new employment outcome measures for TANF. These bills were not considered by the full House.
The TANF block grant funds grants to states, tribes, and territories for providing benefits, services, and activities to broadly address both the effects and root causes of childhood economic and social disadvantage. It was created in the 1996 welfare reform law, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (P.L. 104-193), which culminated decades of debate over how to change public assistance programs for needy families with children. The 1996 welfare law provided funding for TANF through FY2002. Since the end of FY2002, TANF has been extended numerous times on a short-term basis, and had one long-term (five-year) extension. That long-term extension expired at the end of FY2010, and TANF has been extended on an annual basis since then. Though there have been policy changes to TANF attached to some TANF extensions, most TANF policy dates back 20 years to the 1996 welfare law.
In July 2015, the House Ways and Means Committee released a "discussion draft" proposing a five-year reauthorization of TANF with wide-ranging policy changes, including changes to TANF's work standards. (See CRS In Focus IF10315, TANF Reauthorization: House Ways and Means Committee Discussion Draft of July 10, 2015.) The President's FY2017 budget proposals also proposed a number of changes to TANF, including an increase in block grant funding. (See CRS In Focus IF10367, Temporary Assistance for Needy Families and Related Programs: The President's FY2017 Budget Proposal.)
H.R. 5170, the Social Impact Partnerships to Pay for Results Act, was introduced by Representative Todd Young, for himself and Representative John Delaney.1 As ordered reported with an amendment by the Ways and Means Committee, H.R. 5170 would have provided for the establishment of "social impact partnerships," also sometimes referred to as "social impact bonds."2
H.R. 5170 was ordered reported by the House Ways and Means Committee on May 11, 2016. Its report (H.Rept. 114-616) was filed on June 10, 2016. The version of H.R. 5170 considered (and passed) on the House floor on June 21 differed from the bill reported from the Ways and Means Committee. The version considered on the floor and passed by the House added a provision that would have required at least 50% of funds provided for agreements awarding funding for social impact partnership projects be used for initiatives that directly benefit children. The version considered and passed by the full House also added provisions that would have provided for a one-year extension of TANF funding as well as a revision of TANF's provisions on research.3 That version was considered by the House under suspension of the rules and passed by a voice vote.
There is no provision currently under TANF for social impact partnerships. Hearings in the Senate and House focused on earlier versions of the legislation from the same sponsors, however, as well as the topic in general.4 The Obama Administration pursued a few related efforts under other authorities,5 and it advocated for legislation similar to the committee bill.6 In characterizing these proposals, Obama Administration officials suggested that the use of social impact partnerships/bonds represents a "new form of grant-making."7 Other observers have viewed the approach as a form of complex contracting.8
As passed by the House, H.R. 5170 would have had several enumerated purposes, including, among others, to improve the lives of families and individuals in need in the United States by funding social programs that achieve real results, to facilitate the creation of public-private partnerships, and to incorporate outcome measurement and randomized controlled trials or "other rigorous methodologies" for assessing program impact.9
Generally speaking, the measure would have established a mechanism through which state and local governments could apply to the Secretary of the Treasury (hereinafter, the Secretary) for "social impact demonstration projects."10 If the Secretary awarded an agreement for one of these demonstrations—a "social impact partnership project"—to an applying state or local government, the applicant and several other entities could have entered into contracts that govern the operation, financing, and evaluation of the project and related interventions. The legislation specified a "social impact partnership model" that involved interactions among the federal government, a state or local government, service providers, investors (if applicable), and potentially an "intermediary" to coordinate among the non-federal actors.11 The federal government would have made a payment to the state or local government only if the project met the requirements of the agreement and achieved one or more of the outcomes specified in the agreement, as determined by an independent evaluator using methodologies specified in the legislation.
The legislation specified related procedures. Selected provisions are summarized below. To carry out these provisions, H.R. 5170, as passed by the House, would have directed the Secretary to reserve $100 million of amounts made available for the Contingency Fund for State Welfare Programs for FY2017 (see discussion in "The TANF Contingency Fund").
Not later than one year after enactment, the Secretary, in consultation with the Federal Interagency Council on Social Impact Partnerships (hereinafter Council, as established in the committee bill),12 would have been required to publish in the Federal Register a notice that seeks proposals from state or local governments for social impact partnership projects. A state or local government would have been required to submit an application describing its proposal for a project. As passed by the House, the bill would have required social impact partnership projects to produce one or more measurable outcomes that result in social benefit and federal savings through one or more of 21 enumerated categories of outcomes.13
An application for a social impact partnership project and related funding would have been required to, among other things, describe the intervention and a plan for delivering the intervention.14 In the application, the state or local government would have been required to address project outcome goals; address "rigorous evidence demonstrating that the intervention can be expected to produce the desired outcome" in the applicant's context; and include projections of the federal, state, and local government costs of the project as well as associated cost savings. The application would also have been required to include a description of the experience of the state or local government in raising private and philanthropic capital to fund social services; the experience of the service provider in delivering the proposed intervention; and certain information about the intermediary for the project. Furthermore, the application would have been required to address the project's evaluation design, metrics, and terms for payment if the program succeeds in producing specified outcomes.
A state or local government would have been able to apply for separate funding to conduct a feasibility study prior to an eventual application for a social impact partnership project. Applications for funding for feasibility studies could have included information drawn from feasibility studies funded through other sources, and applications would have been required to address multiple topics including descriptions of the outcome and intervention.15
In determining whether to enter into an agreement awarding funding for a social impact partnership project, the Secretary would have been required to consult with the Council and the head of any federal agency administering a similar intervention or serving a population similar to the one described in the project application, and to consider several factors before making a determination on whether to enter into an agreement.16 These factors would have included, among others, recommendations made by the Commission on Social Impact Partnerships (hereinafter commission, as established in the committee bill);17 the value of the expected outcomes; the likelihood, based on evidence provided in the application and "other evidence," that the state or local government will achieve the expected outcomes; the projected savings to the federal government and state and local governments; and the expected quality of the evaluation under the agreement.
The Secretary, in consultation with the Council and certain federal agencies, would have been authorized to enter an agreement for a social impact partnership project with a state or local government if the Secretary determined, in consultation with the Council, that several requirements were met.18 The requirements would have included, among other things, that the state or local government demonstrated through its application that, "based on prior rigorous experimental evaluations or rigorous quasi-experimental studies, the intervention can be expected to achieve each outcome."19
The Secretary would have been required to make a determination about whether to award funding for a feasibility study to an eligible applicant not later than six months after receiving an application for feasibility study funding. In making this determination, the Secretary would have been required to consult with the Council and the heads of certain federal agencies. When considering an award of feasibility study funding, the Secretary would have been required to consider the recommendations made by the commission, the likelihood that social impact partnership projects included in the feasibility study would achieve the desired outcomes, the value of the expected outcomes, and the potential savings to the federal government and to state and local governments if the projects identified in the feasibility study were successful.20 Federal funds could provide up to 50% of the cost of a feasibility study. The study could then be used to apply for social impact partnership project funding.
The state or local government would have been required to agree to achieve one or more outcomes specified in the agreement. The duration of a social impact partnership project could not exceed 10 years. Furthermore, the Secretary would have been authorized to transfer to the head of another federal agency the authority to administer an agreement and any necessary funds. Federal payment to the state or local government would have been made if an independent evaluator determined that the project had met the requirements of, and achieved an outcome specified in, the agreement. As it passed the House, the bill would have required that at least 50% of funds provided for agreements awarding funding for social impact partnership projects be used for initiatives that directly benefit children.
A state or local government would have been able to receive one or more payments under the terms of the agreement.21 The evaluation used to determine payments to state and local governments for project outcomes would have been required to use experimental designs using random assignment, or certain other research methodologies certified by the Council that "allow for the strongest possible causal inferences" when random assignment is not feasible. Progress reports would have been due from the evaluator within two years of the approval of the project and biannually thereafter. A report would have also been due before the scheduled time of the first outcome payment and each subsequent payment, in addition to a final report within six months of the completion of the social impact partnership project.
Various potential issues of design and implementation may be associated with provisions in H.R. 5170, in addition to the issue of capabilities and limitations of certain types of evaluation to support learning, improvement, and policymaking. These issues include federal-state relations in grant administration; comparison of the costs and benefits of using federal funds for social impact partnerships/bonds in contrast with those of traditional grant program structures; and the capacity of federal, state, and local governments and other grant recipients to engage in complex contracting and evaluation.
Proposals for social impact partnerships/bonds often have been associated with advocacy for the use of a particular method of evaluation known variously as experimental design, random assignment, or randomized controlled trial (RCT).22 An RCT is a type of impact evaluation.23 Some advocates have referred to a primary emphasis on using impact evaluations, and especially RCTs, as "evidence-based policy" and "investing in what works."24 They frequently reserve the terms "rigorous evaluation" and "rigorous evidence" to refer primarily or only to RCTs. Other observers, by contrast, have argued that a sole or predominant emphasis on RCTs is too narrow for selecting interventions to bring to wider scale in different contexts and for facilitating continuous improvement.25
P.L. 114-113 provided TANF funding through September 30, 2016. It also provided both FY2016 and FY2017 funding for the TANF contingency fund. For a discussion of the TANF contingency fund, see "The TANF Contingency Fund." Funding into FY2017 was provided through two continuing resolutions, P.L. 114-223 and P.L. 114-254.
H.R. 5170 would have provided FY2017 funds for the TANF basic block grant, mandatory child care block grants, grants to Indian tribes for work programs, and healthy marriage and responsibility grants. Except for the TANF basic block grant, FY2017 funding for TANF grants and mandatory child care block grants would have been provided at the same level as was provided for FY2016.
The bill would have made a total FY2017 appropriation for the basic TANF block grant at the same amount as was provided for in FY2016. However, a separate provision of the bill would have set aside 0.33% of funding for the basic block grant for TANF-related research. Thus, the FY2017 TANF basic block grants to each state would have been reduced by 0.33% from their FY2016 level. This reduction would also have triggered a commensurate proportional reduction in the amount of spending required of states under the TANF maintenance of effort (MOE) requirement.26
The bill would not have provided additional funding for the TANF contingency fund, which received an FY2017 appropriation in P.L. 114-113. Under the bill, contingency fund grants to states would have been reduced in FY2017 from their FY2016 levels. As discussed in "Social Impact Partnerships," the House-passed bill would have set aside $100 million from the TANF contingency fund for the social impact partnership demonstration.
The Department of Health and Human Services (HHS) is required to conduct research on the benefits, effects, and costs of state programs under the TANF block grant. The research must include studies that relate to time limits, welfare dependency, illegitimacy, teen pregnancy, employment rates, and child well-being. HHS may conduct studies on other policy issues.
The 1996 welfare law provided the Census Bureau with funding to continue following families from the mid-1990s sample in the Survey of Income and Program Participation (SIPP). That survey, known as the Survey of Program Dynamics, followed the same families from 1992/1993 through 2002. Subsequently, the funds were used for Census Bureau research related to the SIPP.
In general, funding for HHS welfare-related research has been $15 million per year. HHS used funds to continue the evaluations of pre-welfare reform "waiver" programs and large-scale, multisite experiments on Employment Retention and Advancement, and initiatives to help the "Hard-to-Employ." It currently helps fund multi-site experiments on Career Pathways programs for low-income individuals, subsidized employment programs, job search assistance programs, and other employment-related research. Funding for Census Bureau research has generally been $10 million per year. In recent years, HHS and Census Bureau welfare-related research funds were set aside from the TANF contingency fund.
The House-passed bill would have continued to require the Secretary of HHS to conduct research on the effect of TANF programs on employment, self-sufficiency, marriage, family stability, economic mobility, and poverty. HHS would also have been required to conduct research on the effects of healthy marriage and responsible fatherhood grants on child well-being. HHS would have been required to develop ways to distribute information on any research and evaluation conducted as a part of this amendment. The House-passed bill would have made states eligible for funds to evaluate their TANF and related programs.
The House-passed bill would have required research conducted by HHS and the states to use experimental designs using random assignment when feasible. If a random assignment experiment was not feasible, the research would have been conducted using other reliable evidence-based research methodologies.
The House-passed bill would have required HHS, in consultation with the Department of Labor, to develop a database named "What Works Clearinghouse of Proven and Promising Projects to Move Welfare Recipients into Work." This database would have consisted of the projects that used a promising or proven approach in delivering services to move TANF recipients into work. The database would also have included a list of projects that used a developmental approach, and a list of projects that were ineffective in moving recipients to work. The categorization of these projects as proven, promising, ineffective, or developmental would have been based on rigorous evaluation of them.
The House-passed bill would have required the Census Bureau, in consultation with the Secretary of HHS and the Bureau of Labor Statistics, to implement a new household survey and/or enhance existing household surveys to provide for the assessment of the effects of welfare reform on the economic and child well-being of low-income families. The Census Bureau, the Secretary of HHS, and BLS would have been required to consider ways to improve the surveys, and data derived from the surveys, to address underreporting of means-tested benefits; increase understanding of poverty spells, long-term poverty, and intergenerational poverty; better understand the geographical dimensions of poverty; increase understanding of the effects of means-tested benefits and tax benefits on the earnings of low-income families; and improve how poverty and economic well-being are measured, including the use of consumption measures.
These research activities would have been funded through a set-aside from the basic TANF block grant, 0.33% of the total block grant amount. This would have provided a total for welfare-related research of $54.7 million. Of this total, at least $10 million would have been required to be used for research related to Census Bureau household surveys, though the Secretary of HHS would have had the authority to increase funding for those activities above $10 million.
In addition to H.R. 5170, the House Ways and Means Committee reported additional changes to TANF in separate bills during May 2016. These bills would have established a subsidized employment demonstration project (H.R. 2990), revised the rules for TANF state spending requirements (H.R. 2959), added reducing child poverty as a statutory goal of TANF (H.R. 2966), and established new employment outcome measures for TANF (H.R. 2952). These bills were not acted upon by the full House.
H.R. 2990, the Accelerating Individuals into the Workforce Act, was introduced by Representative Dold. The bill, as amended and reported to the House by the Ways and Means Committee, would have established a subsidized employment demonstration program. Subsidized employment programs are those where public funds are used to pay all or part of the wages, benefits, and other costs of employing an individual.
Under current law, states have the authority to operate subsidized employment programs as part of their TANF programs. Additionally, subsidized employment in the private and public sectors are work activities that states may count toward meeting federal TANF work participation standards.
While allowed under current law, subsidized employment has been a relatively small part of TANF, with the exception of a brief period when extra TANF funding was provided in response to the 2007-2009 recession. From FY2001 through FY2007, TANF expenditures on wage subsidies averaged $51.5 million per year. The American Recovery and Reinvestment Act of 2009 (P.L. 111-5) created an Emergency Contingency Fund (ECF) within TANF that provided additional funding for basic assistance, emergency aid, and subsidized employment for FY2009 to FY2010. With special funding from the ECF, TANF expenditures for employment subsidies were $1.050 billion in FY2010. With the expiration of ECF funding at the end of FY2010, TANF expenditures for subsidized employment again receded (though they remained above the FY2001-FY2007 average). In FY2015, TANF expenditures on wage subsidies totaled $186 million (0.6% of total TANF expenditures).
In addition, relatively few cash assistance recipients participate in subsidized employment. From FY2001 to FY2007, an average of 6,500 TANF recipients per month was participating in subsidized private or public sector employment. This number increased to 19,500 per month during FY2010, though participation declined thereafter. In FY2015, a monthly average of 12,960 TANF recipients participated in subsidized private or public sector employment, out of a total of 1.2 million TANF "work-eligible" individuals.27 Thus in FY2015, 1.1% of TANF work-eligible individuals participated in subsidized employment.
HHS is currently fielding an evaluation of subsidized employment programs in seven locations, though few findings on program impacts are available.28 Earlier evaluations of subsidized employment initiatives for cash assistance recipients in the 1970s, 1980s, and early 1990s showed some positive employment impacts (higher employment rates and earnings) and reduced cash assistance receipt, with lasting, long-term impacts. The most recent subsidized employment demonstration (for TANF recipients in Philadelphia) showed some short-term positive impacts, but the impacts faded over the longer term.29 TANF ECF-funded subsidized employment programs were not subject to additional reporting or evaluation.
H.R. 2990, as reported by the House Ways and Means Committee, would have established a one-year (FY2017) subsidized employment demonstration program, administered by HHS. The Secretary of HHS would have awarded grants to states for the development of subsidized employment demonstration projects that would evaluate strategies for providing wage subsidies to help low-income individuals obtain and retain employment. The bill would have provided that wage subsidies extend for no more than 12 months, and that federal funds account for no more than 50% of a participant's wage. It would have required that those participating in the subsidized employment program
States in the subsidized employment demonstration would have been prohibited from displacing regular employees with subsidized job participants. The committee bill would have required HHS to conduct an evaluation of the demonstration project using an experimental design with random assignment, unless such research design was not feasible. If a random assignment experiment were not feasible, the research would have been conducted using other reliable evidence-based research methodologies. HHS would have been required to report recommendations to Congress on how to increase employment, retention, and advancement of individuals currently receiving, or who formerly received, TANF assistance.
The demonstration would have been funded using $100 million of the FY2017 contingency fund appropriation. (For a discussion of the TANF contingency fund, see "The TANF Contingency Fund.") Of that amount, 15% ($15 million) would have been reserved for the operation of career pathways training programs. Career pathways programs are a combination of education, training, and other services that align with the skill needs of industries in the state or regional economy involved. They are designed to help an individual enter or advance within a specific occupation or occupational cluster.
H.R. 2959, the TANF Accountability and Integrity Improvement Act, was introduced by Representative Noem. The bill, as amended and reported by the House Ways and Means Committee, would have addressed the rules of the TANF state spending requirement known as "maintenance of effort" (MOE).
The 1996 welfare reform law created the TANF block grant by ending and consolidating funding from several predecessor programs. The predecessor programs were matching grant programs that shared costs between the federal and state governments. The TANF block grant is based on federal funding under the predecessor programs. Under TANF, states are required to continue spending a minimum amount from their own funds under a MOE requirement. The amount is based on the state share of expenditures in TANF's predecessor programs. Under the MOE requirement, states are required to expend from their own funds at least 75% of their spending in predecessor programs in FY1994. If a state fails to meet TANF work participation standards, the state spending requirement is increased to 80% of historical state expenditures.
HHS has interpreted the TANF MOE as a cost-sharing requirement, subject to general rules of what expenditures count toward meeting cost-sharing requirements.31 These rules are not in TANF statute or regulations, but in HHS's general regulations regarding grant management.32 Under the general rules for cost sharing, states may count both cash donations by non-federal third parties as well as the value of third party, in-kind contributions.
While the MOE requirement specifies a minimum spending level, resulting in a reduced TANF block grant for noncompliance, there are two major incentives for states to spend above that minimum. States receive credit against their work participation standards for "excess MOE": the percentage of the TANF caseload a state must engage in work or activities is reduced by state spending that exceeds the minimum required. Additionally, states must spend more than the minimum under the MOE to access extra federal funding from the TANF contingency fund.
The Government Accountability Office (GAO) has twice surveyed the states on their use of third-party contributions toward the MOE. In February 2016, GAO reported that 16 states had counted third-party contributions as state expenditures toward the requirement.33
Federal law prohibits states from using federal TANF funds to provide medical services (though pre-pregnancy family planning services are excluded from this prohibition). While federal TANF funds cannot be used for medical services, there is no corresponding prohibition on states counting medical services toward their TANF MOE requirement. Medical services may be counted if they are for eligible families and otherwise meet the requirements that apply to expenditures qualifying for the MOE.
H.R. 2959, as amended and reported by the House Ways and Means Committee, would have, beginning with FY2017, frozen the amount of third-party contributions that a state may claim toward meeting its MOE requirement. It would have excluded from the definition of qualified state expenditures counted toward the MOE any third-party contributions that exceed the value of third-party contributions claimed by the state as MOE expenditures in FY2016.
Additionally, H.R. 2959, as amended and reported, would have prohibited states from counting medical expenses toward their MOE.
H.R. 2966, the Reducing Poverty through Employment Act, was introduced by Representative Smith of Missouri. The bill, as amended and reported by the House Ways and Means Committee, would have added reducing child poverty as a statutory goal of TANF.
The purpose of the TANF block grant is to increase the flexibility of states in operating a program designed to achieve four statutory goals: (1) provide assistance for needy families so that children may reside in their own homes, or with relatives; (2) promote job readiness, work, and marriage in order to decrease the dependence of parents on government programs; (3) have states establish goals to prevent and reduce the incidence of out-of-wedlock pregnancies; and (4) encourage the formation and maintenance of two-parent households.
TANF's statutory purpose is both an aspirational statement about the goals of the block grant and a basis for how states may expend TANF funds. Federal TANF funds may be spent on any activity that can be reasonably calculated to achieve TANF's purpose and its goals. States may also count toward their MOE requirement expenditures that can be reasonably calculated to achieve TANF's purpose and its goals.
H.R. 2966, as reported by the House Ways and Means Committee, would have added "to reduce child poverty by increasing employment, retention, and advancement of needy parents" as a fifth main purpose of the TANF program. By specifying that child poverty is to be reduced through employment, retention, and advancement, the bill would have limited any TANF-funded activities aimed at reducing child poverty to those that would have reduced child poverty through employment.
H.R. 2952, the Improving TANF Outcomes of Recipients Act, was introduced by Representative Boustany. As amended and reported by the House Ways and Means Committee, it would have established a new system of employment outcome performance measurement for TANF.
TANF is a broad purpose block grant, with states given flexibility in the design of their cash assistance programs for needy families with children. However, states are held accountable for meeting TANF's federally specified statutory goals related to moving families from welfare-to-work through measuring their performance.
The TANF block grant has two sets of performance measures related to work: (1) the Work Participation Rate (WPR), and (2) employment outcome measures that were used for the TANF High Performance Bonus before 2006. The WPR is computed to determine compliance with TANF federal work standards. It is the percentage of families receiving assistance in the state who are considered engaged in work during a fiscal year. A state with a WPR that fails to meet the numerical goal specified by the federal work standard is at risk of a financial penalty (reduced block grant amount).
Before FY2006, TANF paid a bonus to states that received scores indicating high performance toward achieving the block grant's statutory goals. A state's performance for the purposes of this bonus was partially determined by a set of employment outcome measures for TANF-assistance adults: job entry, job retention, and earnings gain. The Deficit Reduction Act of 2005 (P.L. 109-171) eliminated the performance bonus. However, the TANF statute requires HHS to rank state welfare-to-work outcomes. HHS continues to collect the employment outcome measures that were used for the performance bonus to conduct that ranking. The ranking is not used to determine program funding or rules, and is for informational purposes only.34
The committee bill would have required each state to collect and report information to measure the state performance levels based on a new set of employment outcome indicators beginning in FY2018. The new indicators would have been the following:
These three measures are conceptually similar to those specified as employment outcome performance measures for the nation's workforce programs under the Workforce Investment and Opportunity Act (P.L. 113-128), enacted in July 2013.35 The measures were also recently adopted for the Supplemental Nutrition Assistance Program (SNAP) through regulations implementing the 2014 Farm Bill's requirements that performance goals be set for SNAP employment and training programs.36
The Secretary of HHS would have been required to use the information collected for FY2018 as the baseline level of performance of each state for each indicator. The committee bill would have required the state and the Secretary to come to an agreement in establishing the requisite levels of performance for each indicator for FY2019 and FY2020. It would have required that the state and the Secretary consider how the performance levels for the state compare with levels established in other states. Additionally, in establishing performance levels, the state and the Secretary would have been required to use an objective statistical adjustment model to ensure such levels are adjusted to reflect the economic conditions of the state and characteristics of the participants during that fiscal year. The committee bill would have required performance levels to be set to promote continuous improvement by each state.
The committee bill would also have required the Secretary to develop a template, before October 2017, for each state to use to report on various characteristics and information related to the operation of the program in their state. After September 2020, the Secretary would have been required to annually publish every report submitted for this purpose.
Two of the bills approved by the House Ways and Means Committee on May 11, 2016—H.R. 5170 as passed by the House (social impact partnerships) and H.R. 2990 (subsidized employment)—proposed demonstration projects. Each project would have been funded at $100 million, with the funding coming from the FY2017 appropriation for the TANF contingency fund.
P.L. 114-113 made appropriations to the TANF contingency fund for both FY2016 and FY2017. Under current law, the FY2017 appropriation to the fund is $608 million. These funds are available for grants to states that qualify for them. H.R. 5170, as passed by the House, would have reduced the amount available for FY2017 grants to states from the contingency fund by $100 million. Additionally, H.R. 2990 would also have reduced FY2017 contingency fund grants to states by an additional $100 million.
The contingency fund was created in the 1996 welfare reform law to make extra grants to states during economic downturns. It was initially funded at $2 billion, to remain available as needed. The original funding was exhausted in FY2010; since then Congress has provided annual appropriations for the fund.
States have to meet criteria of economic need in order to access the contingency fund. The criteria of economic need are (1) a three-month average state unemployment rate of at least 6.5% and at least 10% higher than in the corresponding three months of either of the prior two years; or (2) a state's Supplemental Nutrition Assistance Program (SNAP) caseload is at least 10% higher than it was in FY1994 or FY1995. Additionally, states have to spend more from their own funds than they spent in FY1994 on TANF-related programs in order to access the fund.
The number of households receiving SNAP increased through the 2000s, even before the onset of the 2007-2009 recession, with additional increases during the recession. Though SNAP caseloads have begun to decline, for almost all states they remain elevated well above their FY1994 or FY1995 levels. Thus, almost all states qualify as economically needy for the purposes of the contingency fund.
In FY2016, 19 states met both the criteria as economically needy, spent sufficient state funds, and drew down TANF contingency grants. The available FY2016 funding was exhausted in April 2016. The Congressional Budget Office (CBO) estimated in March 2016 that all contingency fund dollars will be spent in each of the next 10 years—regardless of the projected economic growth forecast for that period. Thus, the TANF contingency fund is not expected to serve a countercyclical function of providing extra grants when the economy experiences a downturn. It does, however, provide extra TANF grants to states that qualify for contingency funds. (For a discussion of the TANF contingency fund in the context of overall TANF financing issues, see CRS Report R44188, Temporary Assistance for Needy Families (TANF): Financing Issues.)
Author Contact Information
1. |
In the 114th Congress, similar, though not identical, legislation was introduced in both the House and Senate: H.R. 1336, introduced by Rep. Todd Young on March 4, 2015, and referred to the House Committee on Ways and Means, and S. 1089, introduced by Sen. Hatch on April 27, 2015, and referred to the Senate Committee on Finance. |
2. |
Many observers use the terms "social impact partnerships," "social impact bonds," "pay for performance," and "pay for success" interchangeably. |
3. |
The revisions to TANF research had earlier been reported from the House Ways and Means Committee as H.R. 5169. |
4. |
See U.S. Congress, House Committee on Ways and Means, Subcommittee on Human Resources, Social Impact Bonds: Can They Help Government Achieve Better Results for Families in Need?, 113th Cong., 2nd sess., September 9, 2014, at http://waysandmeans.house.gov/event/chairman-reichert-announces-hearing-on-social-impact-bonds-can-they-help-government-achieve-better-results-for-families-in-need/; and U.S. Congress, Senate Committee on the Budget, Task Force on Government Performance, Investing in What Works: Exploring Social Impact Bonds, 113th Cong., 2nd sess., May 1, 2014, at http://www.budget.senate.gov/hearings/performance-task-force-investing-in-what-works-exploring-social-impact-bonds. |
5. |
U.S. Executive Office of the President, Office of Management and Budget (OMB), "Improving Outcomes through Pay for Success Financing," fact sheet, February 2016, at https://obamawhitehouse.archives.gov/omb/evidence. |
6. |
OMB, "Building and Using Evidence to Improve Results," fact sheet, February 2016, p. 13 of PDF file, at https://obamawhitehouse.archives.gov/omb/evidence. |
7. |
Cecilia Muñoz, Director of the Domestic Policy Council, and Robert Gordon, Executive Associate Director, OMB, "Pay for Success: A New Results-Oriented Federal Commitment for Underserved Americans," White House Blog, January 24, 2012, at https://obamawhitehouse.archives.gov/blog/2012/01/24/pay-success-new-results-oriented-federal-commitment-underserved-americans. |
8. |
See Trevor L. Brown, Matthew Potoski, and David M. Van Slyke, Complex Contracting: Government Purchasing in the Wake of the US Coast Guard's Deepwater Program (Cambridge, UK: Cambridge University Press, 2013). Authors of the book suggested at a May 2013 forum hosted by the IBM Center for the Business of Government in Washington, DC, that social impact bonds constitute a form of complex contracting. The authors use the term "complex contracting" to refer to a situation when a product is difficult to specify in a contract and requires specialized investments for a market exchange, such that the buyer and seller can find themselves locked into a mutually dependent relationship in which both win-win and lose-lose outcomes are possible. |
9. |
See Section 2 of the bill, the proposed Section 403(c)(1), H.R. 5170 as passed by the House. |
10. |
Social Impact Demonstration Projects is the title of the proposed Section 403(c), H.R. 5170 as passed by the House, containing provisions on social impact partnership projects. |
11. |
See the proposed Section 403(c)(13)(E), H.R. 5170 as passed the House. |
12. |
The Council would have been established to, among other things, coordinate the administration of funded projects, assist the Secretary in the implementation and development of projects, and provide oversight for projects. The council would have been composed of 11 members. The bill would designate the Director of the Office of Management and Budget to be chair, and would require the remaining ten members to be designated by certain federal agencies. |
13. |
See Section 2 of the bill, the proposed Section 403(c)(2)(B), H.R. 5170 as passed by the House. |
14. |
See Section 2 of the bill, the proposed Section 403(c)(2)(C), H.R. 5170 as passed by the House. |
15. |
See Section 2 of the bill, the proposed Section 403(c)(4), H.R. 5170 passed by the House. |
16. |
See Section 2 of the bill, the proposed Section 403(c)(3), H.R. 5170 passed by the House. |
17. |
The commission would have been established to assist the Secretary and the Council in reviewing applications for funding, make recommendations to the Secretary and the Council regarding funding for agreements and feasibility studies, and provide other assistance. The commission would have been composed of nine members: a chair appointed by the President and eight members selected by designated Members of Congress. |
18. |
See Section 2 of the bill, the proposed Section 403(c)(3)(C), H.R. 5170 as passed by the House. |
19. |
The term "quasi-experimental" refers to evaluations that attempt to estimate an intervention's impact, but, in contrast with RCTs, that do not have random assignment to treatment and control groups. Some quasi-experiments are controlled studies (i.e., with a control group and at least one treatment group), but others lack a control group. See CRS Report RL33301, Congress and Program Evaluation: An Overview of Randomized Controlled Trials (RCTs) and Related Issues, by Clinton T. Brass, Erin D. Williams, and Blas Nuñez-Neto. |
20. |
See Section 2 of the bill, the proposed Section 403(c)(4)(B), H.R. 5170 as passed by the House. |
21. |
The legislation does not specify in detail how a state or local government would pay a service provider or intermediary for its services. In addition, the legislation does not specify in detail how an investor or other entity would have been paid or otherwise compensated for use of its capital and the potential risk of losing the capital, in the event that the intervention does not achieve the outcomes that are specified in the agreement between Treasury and the state or local government. Instead, the legislation would require an applying state or local government to address all of these topics in the application to be submitted to Treasury, presumably with plans for multiple contracts. |
22. |
For example, see Results for America, "RFA Accomplishments and Activities," at http://results4america.org/rfa-accomplishments-activities/. For an overview of RCTs, see CRS Report RL33301, Congress and Program Evaluation: An Overview of Randomized Controlled Trials (RCTs) and Related Issues, by Clinton T. Brass, Erin D. Williams, and Blas Nuñez-Neto. |
23. |
An impact evaluation quantitatively estimates the extent to which a policy causes a change in an outcome of interest, compared to what would have happened without the policy (i.e., the difference is the "impact"). An RCT randomly assigns subjects to treatment and control groups to estimate an intervention's impact. See ibid. |
24. |
See Results for America, "RFA Accomplishments and Activities," and Ron Haskins and Greg Margolis, Show Me the Evidence: Obama's Fight for Rigor and Results in Social Policy (Washington, DC: Brookings Institution, 2015). |
25. |
Some observers have argued that impact evaluations do not provide sufficient evidence that an intervention will have an impact in another time and context, and that a "full spectrum" of evaluation methods is necessary (1) to assess the generalizability of an intervention and (2) for an organization with a system of multiple interventions to improve its performance. For example, see Friends of Evidence at the Center for the Study of Social Policy, An Evidence Framework to Improve Results, Washington, DC, November 2014, at http://www.cssp.org/policy/harold-richman-public-policy-symposium. See also Nancy Cartwright, "Knowing What We Are Talking About: Why Evidence Doesn't Always Travel," Evaluation & Policy, vol. 9, no. 1, January 2013, pp. 97-112; and American Evaluation Association, statements regarding Evidence-Based Policymaking Commission Act of 2014, at http://www.eval.org/p/cm/ld/fid=52. |
26. |
See Section 409(a)(7)(B)(iii)(II). For a discussion of the TANF maintenance of effort (MOE) requirement, see "H.R. 2959: The TANF State Spending Requirement" in this report. |
27. |
TANF work-eligible individuals are adults or teen parents in TANF families who are not exempted from being considered in the calculation of the TANF work participation rate. This excludes minors (children) as well as certain nonrecipient adult caretakers of children. The families of some work-eligible individuals (e.g., single parents caring for infants) may be disregarded when determining the TANF work participation rate. |
28. |
See discussion of the demonstration at http://www.acf.hhs.gov/programs/opre/research/project/subsidized-and-transitional-employment-demonstration-sted. |
29. |
The National Supported Work Demonstration, fielded in the 1970s, produced positive impacts for long-term assistance recipients (usually single mothers) in terms of increased employment and earnings and reduced welfare receipt for a period of 27 months after entry into the program. See Board of Directors, Manpower Demonstration Research Corporation, Summary and Findings of the National Supported Work Demonstration (Cambridge, MA: Ballinger Publishing Company, 1980). In the 1980s, the AFDC Homemaker-Home Health Aide Demonstration included subsidized jobs, and produced positive impacts in terms of increased earnings and reduced welfare receipt. See Stephen H. Bell, Nancy R. Burstein, and Larry L. Orr, Evaluation of the AFDC Homemaker-Home Health Aide Demonstrations, Overview of Evaluation Results, Abt Associates, December 1987. The Supported Work and the AFDC Homemaker-Home Health Aide programs were multisite demonstrations. In the 1990s, the New Hope project, operated in Milwaukee, WI, also produced long-term positive impacts. The project included earnings supplements, child care, and guaranteed health insurance for those employed and community services jobs for those who were not employed. See Aletha C. Huston, Cynthia Miller, and Lashawn Richburg-Hayes, et al., New Hope for Families and Children, Five-Year Results of a Program to Reduce Poverty and Reform Welfare, MDRC, June 2003. A transitional (subsidized) job program in Philadelphia for TANF recipients who received assistance for more than one year was tested in the 2000s. That program showed a positive impact in the first year of the evaluation on employment—both subsidized and unsubsidized. However, the impact faded after the first year. Thus, the subsidized job sped up employment, but did not over the long-term affect the likelihood that a recipient would be employed. See Erin Jacobs and Dan Bloom, Alternative Employment Strategies for Hard-to-Employ TANF Recipients: Final Results from a Test of Transitional Jobs and Pre-employment Services in Philadelphia, U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research, and Evaluation, OPRE Report 2011-19, December 2011. |
30. |
See the section on "H.R. 2959: The TANF State Spending Requirement" for further information about TANF MOE. |
31. |
See U.S. Department of Health and Human Services, Administration for Children and Families, Office of Family Assistance, Clarification that third party cash or in-kind may count toward a State's or Territory's TANF maintenance-of-effort (MOE) Requirement, TANF-ACF-PA-2004-01, December 1, 2004. |
32. |
See 45 C.F.R. §92.24. |
33. |
U.S. Government Accountability Office, Temporary Assistance for Needy Families, Update on States Counting Third-Party Expenditures toward Maintenance of Effort Requirements, GAO-16-315, February 2016, http://www.gao.gov/assets/680/675110.pdf. That report updates a previous GAO report published in 2012. |
34. |
H.R. 5170, as it passed the House, would have repealed the authority for HHS to collect the employment outcome measures formerly used by the High Performance Bonus. In effect, if both bills were enacted the employment outcome performance measures of H.R. 2952 would replace the current employment outcome measures associated with the High Performance Bonus. |
35. |
See CRS Report R44252, The Workforce Innovation and Opportunity Act and the One-Stop Delivery System, by David H. Bradley |
36. |
For a description of SNAP employment and training programs and the new performance goals requirement, see CRS Report R43332, SNAP and Related Nutrition Provisions of the 2014 Farm Bill (P.L. 113-79), by Randy Alison Aussenberg. For the interim rule implementing SNAP employment and training monitoring and report requirements, see U.S. Department of Agriculture, Food and Nutrition Service, "Supplemental Nutrition Assistance Program (SNAP): Employment and Training Program Monitoring, Oversight, and Reporting Measures," 81 Federal Register 15613-15623, March 24, 2016. |