Testimony of LaDonna Pavetti, Ph.D. Vice President, Family Income Support Policy before the House Ways and Means Committee, Subcommittee on Human Resources
Thank you for the opportunity to testify today.
In its recent report, "Opportunities to Reduce Potential Duplication in Government Programs, Save Tax Dollars, and Enhance Revenue," the U.S. General Accountability Office (GAO) examined programs throughout the government with an eye towards identifying opportunities for eliminating unnecessary duplication and reducing administrative inefficiencies that often occur when duplicative programs exist. The GAO focused on a variety of areas where government programs could be improved, including procurement within the Department of Defense and tax enforcement, both of which are likely to yield significantly more cost savings than changes in other areas the GAO examined, including programs that assist unemployed and low-income Americans.
But, we applaud the Subcommittee's interest in taking a careful look at overlap and potential duplication in programs within its jurisdiction. A re-examination of programs with a goal of maximizing the government's investments has the potential to improve program performance and the delivery of services to those who can most benefit from them.
My testimony will focus on four key points:
(1) In order to reduce program overlap and improve program effectiveness, we need to start with a common understanding of what we mean by "welfare and related programs."
(2) Program duplication is not always bad or unnecessary. In some cases, what appears as duplication is an explicit effort to ensure access to individuals who are underserved and/or whose needs differ from a larger targeted population.
(3) There are positive examples of state and local efforts to coordinate similar services and eliminate program overlap, although, there is no hard evidence that these efforts will save a significant amount of money in the short or the long term.
(4) Despite the limits of what can be accomplished through program coordination, there are actions that Congress can take to encourage greater program coordination and improve program performance.
Defining "Welfare and Related Programs"
Historically, "welfare" has referred to programs that provide economic assistance to persons in need to help them meet their basic needs. The recently introduced, "Welfare Reform Act of 2011," for example, would significantly expand this definition, creating a new budget category that would consist of all domestic programs targeted on people with low incomes. The legislation singles out low-income programs for cuts of unprecedented depth by requiring that total expenditures in this category not exceed the total amount spent on these programs in 2007, adjusted for inflation (i.e., increases in the Consumer Price Index). The majority of the expenditures for the programs in this category go for Medicaid and other health care programs like the Children's Health Insurance Program, but the category also includes many programs that are not designed to help people meet their basic needs and that extend far beyond the usual reach of the definition of welfare programs. Examples include: (1) Pell Grants, which help lower-income individuals attend college; (2) Title I grants, which provide additional resources to schools whose students disproportionately come from poor families; (3) Community Health Centers; (4) Foster Care Assistance for children from low-income families and Adoption Payments for children with special needs; (5) Child Care and Development Block Grants, which provide child care assistance to working families, including in many cases, child care that is necessary if a mother is to work rather than rely on public assistance; and (6) Community Development Block Grants, which help improve the economic infrastructure of local communities. The Center on Budget and Policy Priorities is preparing an analysis of what the proposed legislation will mean for low-income individuals and families and will provide it as an addendum to our testimony when it is complete.
A better approach for evaluating government investments would be an evidence-based strategy that uses reliable data to determine whether a program should be ended, modified, continued as is, or expanded. This would allow all programs to be considered on their merits and would not single out low-income programs for harsher treatment than programs targeted to other individuals or corporations. Evidence could be used to decide which programs are worthy of continuation as is, which show promise but need improvement, and which have failed to live up to their promise and should be eliminated.
Toward a More Detailed Understanding of Program Duplication
One program area that GAO points out in its report on reducing harmful duplication is employment and training, particularly the overlap between Temporary Assistance for Needy Families (TANF) and other employment and training programs. We will use it to illustrate what is behind some of the duplication that GAO found and what it would take to remedy it.
An important point regarding duplication that is critically important to keep in mind is that the existence of multiple programs that appear similar does not mean that the same people are receiving services from multiple similar programs. In fact, that is rarely the case. Instead similar programs are serving different segments of the population. Many programs designed to help individuals become self-sufficient are vastly underfunded, making it possible for them to serve only a fraction of the eligible population. There are tens of millions of people in need of job training to access better jobs, but only a fraction ever have the opportunity to access training. Similarly, the Child Care and Development Block Grant, which provides child care that makes it possible for many parents with children to work, serves only a small fraction of eligible families. Some states respond to excess demand by maintaining waiting lists and others simply turn families away when all their funds have been allocated.
GAO reported that in fiscal year 2009, the federal government spent $18 billion supporting 47 programs providing employment and training. Many of these programs are quite small and/or serve narrowly defined groups that face more significant employment barriers than the general population, such as individuals with disabilities, Native Americans, veterans, and disadvantaged youth. Programs with narrowly-defined target populations were created to ensure that populations with special needs receive employment services appropriate for them. Because funding for employment assistance is quite limited and only a small share of eligible individuals can be served, many of the individuals served in these more targeted programs would be far less likely to receive services if these programs did not exist -- and there is no guarantee that they would receive services that would address their special needs.
Three of the 47 programs included in the GAO analysis -- TANF, the Workforce Investment Act (WIA), and the Employment Service -- account for majority of the people served by the workforce development system. On the surface, these three programs share similar goals and provide similar services: they focus on helping unemployed and underemployed individuals find and maintain employment, primarily by providing job search assistance and working with employers to identify job opportunities. However, the similarities end there. Different performance standards, participation requirements, and target populations all discourage program coordination, and even when programs are coordinated, encourage different approaches to providing employment and training assistance.
TANF agencies' performance is judged on their ability to meet a work participation rate (WPR) of 50 percent for all families and 90 percent for two-parent families. WIA's performance for adults is judged by entirely different criteria, including the number of individuals served who enter unsubsidized employment, job retention after six months, earnings, and the attainment of recognized credentials. Because the performance measures are not adjusted based on the characteristics of the individuals served, WIA programs have been reluctant to serve the most disadvantaged job seekers, including TANF recipients (who tend to have significant employment barriers). The Employment Service's performance is judged, in part, on measures comparable to those used in WIA. Because job search assistance is a small component of the Employment Service's function and the program has limited funding for services, it provides only minimal individualized job search assistance; still, there is evidence that these services, which tend to be used by unemployed individuals with a strong labor force attachment, help to reduce the time individuals remain unemployed.
The extent to which the programs provide services beyond job search also differs quite dramatically. The employment assistance provided by the Employment Service is generally limited to job search assistance, much of it provided online or through self-service resource rooms. WIA provides three levels of services: (1) core services, that focus on job search assistance and are provided to all job-seekers requesting assistance; (2) intensive services, which are provided to those deemed in need of more personalized assistance; and (3) training services, that are reserved for those who experience the most difficulty finding unsubsidized employment. The services provided by TANF agencies differ from state to state (in some states, from county to county); many focus on job search assistance but a few encourage participation in education and training and a few operate extensive community service or subsidized employment programs.
Of the three programs, state TANF agencies impose the most constraints on the activities in which individuals can participate, primarily because of the pressure to meet the federally-established WPR. In addition, federal TANF requirements on tracking and reporting work participation also are far stricter. TANF recipients participating in job search longer than six or 12 weeks (depending on the state of the economy) or vocational training longer than 12 months do not count as "engaged in work" for purposes of the state meeting its WPR. Additionally, every hour of participation must be verified in order to count toward the state meeting its WPR. This imposes greater burdens upon other programs (such as WIA) when TANF recipients participate in them. TANF funds can be used for work-related activities beyond those that count towards the WPR, but many states shape their permissible work activities to match what can count towards the work participation rate.
Finally, the programs serve different segments of the population. The limited employment assistance that the Employment Services provides is available to all job seekers. Eligibility for WIA is significantly more complicated and varies substantially among local providers. WIA requires states and local areas to provide core services to all job seekers, but to implement a priority of service for intensive and training services. When funding is limited, public assistance recipients and other low-income adults are supposed to receive priority. Still, states and local areas have considerable flexibility to define the policy and determine whether and how to implement it. In fact, low-income individuals represent less than half of those receiving intensive or training services.
The employment services provided by TANF agencies or their contractors are generally targeted to individuals who receive TANF cash assistance, although states can use federal and state TANF funds to serve needy families who are not receiving cash benefits, and some have done so. As TANF caseloads have declined, a substantial portion of the remaining caseload faces significant barriers to employment, including significant mental health and physical health issues. Limited funding and performance measures that discourage serving the most disadvantaged have discouraged WIA-funded programs from serving TANF recipients as a part of their regular programs. However, some WIA programs serve them under special arrangements using TANF funds.
Positive Examples of Creating Integrated Service Systems
This analysis suggests that states and/or localities that want to coordinate their programs must overcome a number of barriers. Some states have done so, however, and there is much that can be learned from their experiences. While some states have focused on creating fully integrated systems that aim to integrate and coordinate employment and training as well as other public benefit programs, others have focused primarily on improving access to benefits and services. Below, I provide two examples. Utah has long been recognized as a leader in fully integrating its TANF and workforce systems while Arizona provides an example of a state that has undertaken a major initiative to reengineer its work processes to ensure that individuals and families, including working families, are able to access public benefits when they are in need.
Example: The Utah Department of Workforce Services
When Utah implemented welfare reform, it not only changed its policies to better assist low-income families in entering the paid labor market, but it also changed its administrative structure to provide all employment-related and safety net services through one Department, the Utah Department of Workforce Services (DWS). The Department's primary goal is to help individuals and families improve their situation by providing the assistance that individuals need to participate in Utah's economy. To this end, Utah provides a broad range of employment-related services including job search assistance, training and works support for individuals who are looking for work. Individuals receiving unemployment insurance, TANF, Supplemental Nutrition Assistance Program (SNAP) benefits and child care assistance are all served by the same agency. The agency aims to provide whatever assistance is necessary to help individuals and families address any immediate crisis they may be facing and to help them find employment as quickly as possible. Utah has a long history of being an innovator in addressing the diverse needs of individuals and families who come to the agency for assistance, including those who face significant and personal family challenges that prevent them from working. They may be the only TANF program in the country that places a highly trained social worker or mental health counselor in every local employment service/welfare office in the state.
An important feature of Utah's approach is that it has identified ways in which it can leverage funds from multiple programs to provide more effective services to all who seek assistance. For example, following the program model set forth by the Workforce Investment Act (WIA), Utah provides core, intensive and training services to individuals who meet the TANF eligibility requirements, using TANF dollars to do so. Similarly, it provides the same set of services using Food Stamp Employment and Training dollars for those who meet that program's eligibility requirements. This makes it possible for Utah to use its WIA dollars for those who do not qualify for other programs. It also means that it does not have to meet the WIA performance standards for these other groups of individuals. DWS has found that the use of multiple funding streams has made it possible for it to provide more services to assist job seekers to develop the necessary skills to secure employment in demand occupations - the vast majority of Utah's WIA funds are dedicated to training. Workers benefit because the training qualifies them for better jobs, and employers also benefit as they have access to more skilled workers.
Although as noted in the GAO report, DWS officials believe that the integration of program services has reduced their administrative costs, this does not mean that consolidation of existing program services will save money. In fact, the opposite is possible. In designing its integrated system Utah decided to provide a comparable set of services to all individuals in need of assistance, regardless of which program eligibility rules make them eligible for services. As a result, it has devoted a significant portion of its federal and state TANF dollars to funding work activities. In 2009, Utah spent $32.6 million (about one-quarter of its federal and state TANF funds) on work activities. That amounts to an average of $830 for each poor family with children. (The average amount spent per family served is substantially higher as only a minority of all poor families with children receives services.) To get to that same average level of spending per poor family, other states would need to more than double their current total investment of $2.4 billion of federal and state TANF funds on work activities.
It is also important to note that while Utah aims to provide similar services to all regardless of the program for which they are eligible, in order to meet the required TANF work participation rate, it may not place TANF recipients in some activities staff feel they need and the state must provide a much higher level of monitoring of TANF recipients than for individuals who are not receiving TANF cash assistance. For TANF recipients, every hour of participation must be documented and verified. If all program participants outside of TANF were subject to this same level of documentation and verification, DWS would almost certainly need to spend substantially more than it currently does to administer its programs.
Example: Simplifying and coordinating enrollment in work support programs
For more than 15 years, federal and state governments have been working together in earnest to simplify enrollment in public benefit programs. Their work has been driven in part by the need to create more effective and efficient government services and also by the need to find better ways to deliver services to working families who make up a significant --and growing-- share of most public benefit programs. New technology also has played a role. For example, states that now scan all of clients' paper documents have more options as to how to support a clients' case. Caseworkers who have access to a client's records electronically can be located in a different part of the state than the client. In addition, on-line applications, call centers and the increasing ability to link human services computer systems to information in other government systems, such as the Department of Motor Vehicles and Vital Statistics can create efficiencies in state administration. These efforts have been successful in many ways. In some form or another, most states have embraced increasing access to benefits that help individuals find and maintain employment and simplifying policies. They have streamlined processes, made procedures more client-friendly, reduced paperwork, and sought to increase outreach to potentially eligible people. As a result, millions of low-income individuals who might not have obtained work supports now do. This achievement is no small feat.
Arizona provides an example of how these efforts are playing out in many areas. After experiencing a 60 percent increase in the number of public benefit cases in recent years because of the recession and a 30 percent reduction in the number of staff to process the work, Arizonarecentlyembarked on an effort to redesign its eligibility processes for multiple public benefit programs. After discovering that a typical family had to make three to five in-person visits to the local office to secure benefits, the state sought to restructure its process to take care of as much business in as few visits as possible. Thus far, it has implemented the new process in two offices and has found that for 65 percent of its customers, it can complete the application in a single visit. In other words, in these two offices the state is saving two to four future visits per customer 65 percent of the time.
Recognizing that there is much to be learned about what is happening in this area, the Ford Foundation has embarked on a $15 million effort, "Work Support Strategies: Streamlining Access, Strengthening Families," to work with nine states to test innovative ways to streamline access and improve administrative efficiencies to work supports for low-income families. The states, Colorado; Idaho; Illinois; Kentucky; New Mexico; North Carolina; Oregon; Rhode Island; and South Carolina, were selected from 27 states that applied. Republican and Democratic governors alike have signed on to the initiative - states applying for grants had to secure the approval of their governors and the participation of key agency heads.
States participating in the initiative will design and--if chosen next year for an implementation grant-- test new integrated approaches to delivering work supports to low-income families, including health coverage, nutrition benefits, and child care subsidies. Without such supports, low-income workers often drop out of the workforce to manage family, health, and other crises, leaving them stuck in poverty and unable to climb up the economic ladder. As the initiative unfolds, it will provide detailed information on ways in which other states can change their policies and businesses practices to create more efficient eligibility and benefit delivery systems.
Actions Congress Can Take to Encourage Greater Coordination and Reduce Duplication
A number of programs that serve low-income individuals are due for reauthorization. The current authorization for TANF ends on September 30, 2011. WIA, although not within the Committee's jurisdiction, is also due to be reauthorized (and has been due for reauthorization for a number of years). Congress should use these opportunities to consider programs in relationship to other programs that serve similar purposes, rather than considering each program in a vacuum. During the reauthorization deliberations, Congress should use these opportunities to:
- Develop a common set of goals for programs that provide similar services. Programs that provide similar services that appear duplicative are not necessarily aiming to achieve the same goals. For example, the primary goals of WIA are focused on employment, earnings, job retention, and the attainment of credentials while TANF is focused primarily on placing individuals in work activities without regard to whether participation in those activities leads to employment. If similar programs were all aiming for the same or related goals, it would be easier for them to coordinate and avoid duplication.
- Reduce cross-program barriers in both TANF and WIA and other related employment and training programs to allow for better alignment between programs. States would be able to align their TANF and WIA employment services better if the program structures were more in sync with one another. While WIA provides local agencies with considerable flexibility, TANF is far more prescriptive. The mismatch between the two programs has discouraged agencies from reducing program duplication and creating fully integrated systems. If conflicting rules related to the types of employment services that can be provided were eliminated, states could more easily operate a virtually seamless employment services program in which participants are matched to appropriate work-related programs. While there are many ways to reduce cross-program barriers, one option is to allow states to apply the same set of program requirements to all program participants if they agree to fully integrate program services.
- Increase capacity and reward programs for serving individuals with the greatest barriers to employment. When programs have broad flexibility to decide how to target services, individuals with the greatest needs often get left out. New programs (which often are very small with very limited funding) are then created to meet their needs. Programs should be rewarded for serving individuals who are likely to derive the greatest benefits. Random assignment studies have demonstrated that the programs with the greatest impact on employment and earnings are often not the ones that look best under simple outcome measures. While such evaluations are not appropriate for ongoing program monitoring, performance measures should be adjusted to ensure that states and localities are not penalized for serving individuals with barriers to employment.
- Revamp program performance measures to allow for greater comparability across programs. As history has shown, it is very complicated to develop performance measures that do not discourage programs from serving individuals who are most in need even if they may be the most likely to benefit from the services. It is difficult to get agreement on how program performance across one program should be measured, let alone across multiple programs. However, as long as programs are held to different standards, it will be difficult to reduce any duplication that may exist and to encourage better integration of programs. One strategy for moving forward would be to provide some states with demonstration authority to experiment with developing and using alternative measures that can be used across multiple programs to measure performance.
These are difficult economic times and there are real fiscal constraints facing our nation. However, we should not think we can solve our fiscal problems by decimating the safety net. Cutting funding deeply for programs to help low-income Americans work, achieve self sufficiency, and secure jobs that can support their families would not represent a step forward. In this vein, I am concerned both about the cuts in a number of these programs in H.R.1, and about the severe cuts that likely would have to be made under the Welfare Reform Act of 2011 in programs to help people achieve self sufficiency, including Pell Grants to help low-income students afford a college education, job training programs, and the Title I "education for the disadvantaged" program, among others. Under that legislation, steep cuts also would seem inevitable in programs that provide child care assistance that enables many low-income mothers to work. The bill consequently would make it harder for many low-income mothers to hold jobs, while also making it less likely that they and other low-income people would receive education and training they need to obtain adequate-paying jobs.
The legislation would also pose risks for the economy because it would effectively prevent many programs that economists call "automatic stabilizers" from functioning appropriately when the economy weakens. Today, when the economy weakens and the unemployment rate rises, the number of people in poverty swells, and programs such as food stamps and Medicaid respond automatically. They serve the additional people who qualify, and their costs rise accordingly. While the official poverty measure does not account for the important poverty-reducing role these benefits play, the new alternative poverty measures do - and they show how important these programs were in keeping poverty lower than it would otherwise have been during the current recession.
Reducing program duplication and increasing program coordination are important and laudable goals. They should be part of any effort to improve the effectiveness of government services. What can be accomplished in terms of budgetary savings, however, from reducing program duplication and increasing program coordination should not be overstated. And we should be careful not to apply a meat cleaver to low-income assistance programs and employment and training and other programs serving low-income families. As discussed in this testimony, various programs can be better coordinated, and efficiency and effectiveness can and should be improved. But programs should not be starved of resources they need to function effectively.
 This target rate can be adjusted downward if a state's TANF caseload in the prior year was less than its TANF caseload in 2005.