Jeff Liebman Testifies to the Senate Budget Committee

May 2, 2014

Testimony by Jeffrey Leibman, Malcolm Wiener Professor of Public Policy, Harvard Kennedy School of Government - May 1, 2014, Senate Budget Committee Hearing - "Investing in What Works: Exploring Social Impact Bonds."

Chairman Warner, Ranking Member Ayotte, and Members of the Government Performance Task Force, thank you for inviting me to testify this morning about Pay for Success contracts.

Despite spending hundreds of billions of dollars each year, our country is not making rapid enough progress in addressing social problems -- from recidivism to school readiness, and obesity to workforce development. And, for the vast majority of this spending, we have little to no evidence about which programs actually improve social outcomes. Instead, governments continue to fund the same services year after year, paying based on metrics like the number of people served, regardless of whether the programs make a difference in the lives of people they aim to help. We can and must do better to produce more value with each taxpayer dollar.

Starting with the GPRA Modernization Act of 2010, the Government Performance Task Force has been at the front lines of the effort to strengthen performance management and improve how our government works. I believe that Pay for Success, or PFS, contracts could be a critical next step in this Task Force's efforts.

Under the most common PFS model, the government contracts to obtain social services from a local service provider (or team of providers). The government pays entirely or almost entirely based upon the provider achieving performance targets, such as a 10 percent increase in employment or a 50 percent reduction in emergency room visits. Performance is rigorously measured by comparing the outcomes of individuals referred to the service provider relative to the outcomes of a comparison (or "control") group that is not offered the services.

If the program fails to achieve a minimum performance target, the government (and taxpayers) do not pay. Payments increase for performance exceeding this target, up to a maximum payment level. Importantly, the level of payment is often tied directly to the taxpayer savings that occur as a result of the intervention, in some cases producing net savings in excess of the total cost of the original intervention.

As an example, consider a program that works with young men exiting the juvenile justice system. We know that, without transitional support, over 60 percent of these youth will end up back in jail within five years of release. In addition to the social cost of new crimes, these high recidivism rates impose a large budgetary cost on governments--each inmate costs taxpayers thousands of dollars per year.

Under a PFS model, the government contracts with a social service provider to help support these young men as they transition out of the juvenile justice system. The government then tracks recidivism rates for young men referred to the program as well as for a comparable group of young men who were not referred. At a pre-determined time, the government compares the outcomes for these two groups, and pays an amount based on the actual reduction in prison time generated as a result of the program. In other words, by investing in preventative services up-front, governments can actually save money down the road, improving lives and reducing crime in the process.

Under this model, there is often a several year lag between when services are delivered and when results can be measured and government payments occur. Private investors bridge this gap, providing capital to fund the upfront operating expenses of the service provider. The private investors get repaid only if the provider achieves the required level of performance. This financing arrangement is often called a "social impact bond." In most pay for success projects to date, a nonprofit private sector intermediary has helped assemble the project, raising funds from investors and coordinating other key activities such as the contract and evaluation design.

Over the past two years, we have observed the PFS model improve government performance in four ways:

. Improving decision making. The PFS model brings important market discipline to government decisions about which programs to expand as investors will only put money behind programs with a strong evidence base. In addition, the rigorous measurement of project impacts in PFS projects will enable smart decisions to be made about whether or not to continue services beyond the period of the initial project.

. Shifting from remediation to prevention. The injection of private sector financing allows the government to shift resources so as to pay for preventative services rather than pay for the remedial costs associated with bad outcomes.

. Fostering multi-year collaboration. Tackling challenging social problems often requires a sustained multi-year collaboration among public, private, and non-profit actors. It is very difficult to sustain the necessary energy and partnerships with traditional annual government procurements, especially given the high frequency of turnover of government appointees. Pay for success contracts establish a framework for sustaining multi-year collaboration.

Dozens of state and local governments around the country are exploring this model. In 2012, with generous support from the Rockefeller Foundation, the Laura and John Arnold Foundation, and the Dunham Fund, I launched the Social Impact Bond Technical Assistance Lab (SIB Lab) at the Harvard Kennedy School. In the past two years, the SIB Lab has provided pro bono technical assistance to ten state and local governments that are developing PFS projects. These include the states of Colorado, Connecticut, Illinois, Massachusetts, Michigan, New York, Ohio, and South Carolina, as well as the cities of Chicago and Denver. These governments are developing PFS contracts to address a wide range of policy issues--from early childhood education to homelessness and prison recidivism to diabetes prevention.

There are four U.S. pay for success projects that are already delivering services. New York City launched the first PFS initiative in the United States in 2012, designed to help young adults incarcerated at Riker's Island transition back to society upon release. In 2013, our initial state partners, Massachusetts and New York, launched the first state-led PFS projects in the United States--the largest PFS projects anywhere in the world to date and the first to base payments on rigorous, randomized evaluations. Both of these projects are focused on delivering services to young men who have been involved in the criminal justice system and have the aim of increasing their employment and preventing their return to jail or prison. Finally, a PFS initiative to expand access to pre-school for underserved children is being piloted in Utah.

If the PFS model is going to achieve its full potential, the Federal government will need to play a larger role than it has played to date. There are three primary ways in which the federal government could play a critical role:

. Collaborating with states in projects that produce federal budgetary savings. Many of the most promising PFS projects being developed by state governments will produce federal budgetary savings along with state savings. This is particularly true of interventions that reduce future Medicaid costs. For most of these projects, the total federal and state savings exceed project costs, but the state savings alone do not. These projects are viable only if the federal government partners with the state government and enables performance payments to be based on the combined government benefits.

. Establishing PFS initiatives in areas where nearly all of the benefits accrue to the federal government. There are promising policy areas for PFS initiatives where the federal government accrues most of the benefits and where states and cities will have little impetus on their own to set up a project. For example, consider an initiative that would enable individuals with health impairments to remain in the workforce, thereby reducing federal spending on Supplemental Security Income, Disability Insurance, Medicare, and Medicaid. Unless the Federal government initiates a pay for success project in this sort of policy area, it is highly unlikely that any projects will occur.

. Using PFS to foster learning about what works. As members of this Task Force are acutely aware, we often have minimal evidence on program effectiveness. Pay for Success offers the most promising technique available to quickly gather a lot more evidence across a wide range of social spending areas so that better budget decisions can be made both by the federal government and by state and local governments. In particular, well-designed PFS projects provide rigorous, real time, assessments of program impacts. Federal support can help make viable projects that offer the greatest learning opportunities.

There are several specific mechanisms through which the Federal government can encourage the spread of the PFS approach. Different mechanism will be appropriate in different circumstances.

First, in cases in which a state-led project will produce Federal budget savings, the Federal government can make grants to states to cover a portion of the success payments. This was the approach used by the U.S. Department of Labor in its 2012 solicitation for PFS pilot projects. DOL agreed to pay for the benefits that would accrue at the federal level, injecting an additional $22M total in potential success payments to the New York State and Massachusetts PFS projects.

Second, federal agencies can give state and local governments the flexibility to use federal funding streams to make success payments.

Third, federal agencies can initiate pay for success projects in policy areas that are federal priorities, either through direct solicitations for service providers or through state and local government competitions.

Fourth, in cases in which the potential for learning is great, the Federal government can help reduce the financial risk of projects through the use of "credit enhancements." In practice, some of the most interesting and innovative PFS projects are too risky for commercial investors. Given that a successful intervention discovered in one jurisdiction has the potential to spread nationwide, there is a useful role for the federal government in absorbing a portion of the risk so that learning can occur. A credit enhancement or other guarantee from the federal government for a minimal level of repayment can make it possible for commercial investors to finance this sort of project.

The President's proposal for a Pay for Success Incentive Fund at the Treasury Department is quite promising because it envisions a range of strategies matched to the particular needs of different types of PFS projects.

Pay for Success is based on the simple premise that governments should pay for demonstrated results rather than for unverifiable promises. By focusing attention on achieving outcomes and evaluating impacts, PFS has the potential to produce better results at a lower cost to taxpayers. I look forward to working with members of this Task Force to further explore how the Federal government can best encourage the use of this promising approach.

Read this original document at:

Print print | Email email
Jeffrey Liebman, Malcolm Wiener Professor of Public Policy

Jeffrey Liebman, Malcolm Wiener Professor of Public Policy