Social Impact Bonds: A New Model for Investing in Social Services

Given the uncertain fiscal climate of the past several years, state and local governments are under pressure to keep budgets down. At the same time, payments for previous commitments to entitlement programs are coming due, limiting the amount of spending that can be put toward discretionary programs like those in public health, criminal justice, early childhood education, and homelessness.

Where can state and local governments find the funding needed for these social imperatives? Social Impact Bonds (SIBs), also known as Pay for Success (PFS) financing, may be the solution in certain cases. SIBs are a new form of public-private partnership that match state and local governments with non-profit service providers and private funding.

To many philanthropists and financial institutions, SIBs are the wave of the future—an opportunity to put more capital to work toward socially beneficial causes. Moreover, because these investors expect a return, supporters believe that this funding model increases accountability for service providers. Others, however, express concern that private funders who seek financial returns should not be in the business of traditional public investment and that SIBs are an inefficient solution.

This memo outlines the development, the pros, and the cons of SIBs through three case studies.

The SIB Landscape

SIBs sit under the broader umbrella of impact investing. The term impact investing has become a catch-all for nearly any type of investment made with the intention to positively impact society. This can range from organizations that provide microfinancing like Kiva, to venture capital funds that support small businesses in underserved areas, like Citi’s investment in SJF Ventures, to investment managers like Calvert and BlackRock that offer mutual funds dedicated to companies that meet environmental, social, and governance benchmarks. Today, the global market for impact investing is estimated to be about $60 billion, but supporters believe that it could grow to $2 trillion over the next 10 years.¹

The impact investing subset of SIBs and similar investment vehicles is fairly new. As of June 2016, 60 SIBs have been launched in 15 countries with a total investment over $200 million.² The majority of these early SIBs were located in the UK, where the first SIB was created in 2010. Other iterations of the SIB model include Development Impact Bonds, which are being piloted globally, and Environmental Impact Bonds, a new concept under consideration by the DC Water and Sewer Authority.

In the U.S., the first domestic SIB —the Rikers Island Adolescent Behavioral Learning Experience—was launched in 2013. Although there are as many as 50 SIBs in development, only nine have reached the implementation stage thus far.³ These nine programs are backed by a total investment of $91.9 million.⁴ Two additional SIBs recently announced by Connecticut and South Carolina have attracted $12.5 million and $17.5 million in funding, respectively.⁵ SIBs are poised to grow further with at least 20 states considering laws that would enable SIBs to expand.⁶

Thus far, SIBs have been underwritten by investment banks including Bank of America Merrill Lynch, Goldman Sachs, and Northern Trust. These banks put up some of their own money, and they seek out institutional investors like high net worth individuals and foundations. SIBs do not trade on the bond market, so these investors hold the bonds for their duration, and ordinary investors cannot yet buy these products.

The recipients of SIB funding operate at the state and local level. The federal government’s main role thus far has been to encourage the adoption of SIBs by allocating funding to states, counties, and cities to conduct feasibility studies, defray evaluation costs, and, in some cases, contribute to or guarantee repayment costs.⁷ The Obama Administration has expressed support for SIBs by directing funding to the Social Innovation Fund and similar PFS initiatives in the Departments of Education, Housing and Urban Development, Justice, and Labor. In July, the House of Representatives passed bipartisan legislation introduced by Congressman Todd Young (R-IN-09) and Congressman John Delaney (D-MD-06) to create a $100 million pool for social impact partnerships. A companion bill in the Senate sponsored by Finance Chairman Orrin Hatch (R-UT) and Senator Michael Bennet (D-CO) awaits action.

How SIBs Work

Case Study #1

The Green & Healthy Homes Initiative

 The Green & Healthy Homes Initiative (GHHI), a nonprofit organization, has discovered a way to improve health outcomes for asthma sufferers and create significant savings for the government simultaneously. GHHI inspects homes in low-income areas for common health hazards, performs remediation services, and provides health and safety education for the home’s residents. These services have been proven to reduce the number of asthma episodes, ER visits, and hospitalization, in some cases by as much as half.⁸

Here’s where the rubber meets the road. The cost of GHHI’s program is $2,500 per home. Meeting these expenses for a large number of families is prohibitive for many local governments. But the net benefits of the program could be as much as $17,500 per home after four years.⁹ That’s because GHHI frequently works in the homes of Medicaid recipients, and the federal government spends about $10,000 per person per year on Medicaid recipients with asthma. Official estimates put the return on investment at 5.3 to 14 times the initial expense.¹⁰ So, an all-in cost-benefit analysis shows that there is an excellent return on investment, provided that the upfront costs can be met.

This is where a SIB comes in. GHHI is developing a SIB contract with the Calvert Foundation to fund its work with members of Johns Hopkins’ Medicaid Managed Care Organization (MCO). The Calvert Foundation has agreed to provide a loan in three installments over three years. That funding will cover the cost of GHHI’s services. The Hilltop Institute at the University of Maryland Baltimore County will evaluate GHHI’s performance, and if its services result in Medicaid savings, then Johns Hopkins’ MCO will repay the Calvert Foundation for the loan. If not, then a loan guarantee provided by an investor to be determined would kick in to protect the Calvert Foundation’s initial investment. While there is no purely governmental entity involved in this particular case, the federal government ultimately benefits if the projected Medicaid cost savings are realized.

Upsides of SIBs

Case Study #2

The Rikers Island Adolescent Behavioral Learning Experience

In 2013, an average of 682 juveniles were in jail serving time at New York’s Rikers Island on any given day.¹¹ About half of the teenagers who pass through the infamous jail return within a year. That year, the New York City Department of Corrections launched the Adolescent Behavioral Learning Experience (ABLE), a program intended to reduce recidivism among teenagers at Rikers Island. Every individual aged 16 to 18 who stayed in jail for more than four days would receive a form of cognitive-behavioral therapy created for prison populations shown to reduce short-term recidivism by 50%.¹² Lower recidivism rates would not only improve outcomes and quality of life for these kids and their communities, but also create significant savings for the city government.¹³

Goldman Sachs underwrote a $7.2 million SIB to a nonprofit research organization called MDRC, which was selected to coordinate the ABLE program. The loan was backed up by a $6 million guarantee provided by Bloomberg Philanthropies. The funding partners, along with the city government, set a goal to reduce recidivism by at least 10% by the time of the first evaluation, as measured by nights spent in Rikers Island by returning juvenile offenders. The 10% benchmark represented the break-even point for the total investment. Any additional improvement would not only result in increased government savings, but also increased returns for the private investors. Furthermore, if the program passed evaluation, the investment would increase to a $9.6 million loan with a $7.2 million guarantee.¹⁴ Unfortunately, the evaluator found that program fell short of its 10% goal. MDRC used the $6 million guarantee from Bloomberg Philanthropies to compensate Goldman Sachs for 75% of the initial investment, and the program was discontinued.

Even though the program did not succeed, the SIB should not be considered a failure. Here’s why: The government spent nothing. Had the government taken on the initial investment, it would have been out $7.2 million. Instead, the initial investment was borne by private investors, and the risk was shared with a philanthropy. The philanthropy preferred this arrangement to actually funding the project itself because shifting its investment to the back end allows it to invest in more programs. And all parties involved learned lessons from the experiment that informed ongoing efforts to reduce recidivism.

Case Study #3

The Utah High Quality Preschool Program

As of 2013, when the Utah High Quality Preschool Initiative was launched, Utah was one of 10 states that did not fund public preschool. As a result, a high percentage of low-income children enter kindergarten ended up enrolling in special education: 17% to 18%. Serving a child with special needs costs the state $2,600 more annually than serving general education students.¹⁵ Over the course of 13 years in public school, this adds up to $33,800 per child.

Some children with special needs will be served best in special education programs, regardless of whether or not they attend preschool. However, a specially designed preschool program piloted in a Utah school district was shown to keep 95% of at-risk low-income children out of special education.¹⁶ The preschool program cost approximately $1,700 per child each year for two years.¹⁷ Yet even though it is less expensive than special education needed later in childhood, it represents an added cost until the student population turns over. Therefore, the local and state governments turned to a nonprofit, a philanthropic investor, and an investment bank to launch a SIB that would expand this program to more children.

For this program, the United Way of Salt Lake served as the project manager, so that all of the funding and expenses involved with the project flowed through this one central coordinator. Goldman Sachs and J.B. Pritzker partnered to make a $7 million loan for the upfront and operating costs of the preschool program. The parties structured the loan so that the bank’s $4.6 million loan took precedence over the foundation’s $2.4 million loan. If the program failed, the foundation would take on losses before the bank. If the program succeeded in reducing special education expenditures, the county and state government would turn over the money they saved to the project manager, who would then make an annual payment to the investors for each year of the students’ schooling: 95% of projected savings per student, plus 5% interest, subject to a limit on total payments.¹⁸

The program evaluator identified 110 at-risk four-year-olds with a high likelihood of future enrollment in special education based on their scores on the Peabody Picture Vocabulary Test (PPVT), an assessment selected by the Utah School Readiness Board.¹⁹ When the first results of the program came in, only one of the 110 children was placed in special education. This triggered a $260,000 repayment to the investors for year one.²⁰ But such overwhelming success led critics to question the evaluator’s methodology for determining at-risk students, and a public debate unfolded over whether the particular test and the score cutoff line. In particular, some noted that many low-income preschoolers who come from homes in which English is not the primary language are inappropriately identified as having special needs. Others accused the bank of planting an evaluation methodology that all but ensured a large payout by the school district ensured a large payout by the school district.²¹ However, the methodology of the program was defended by the president of the United Way of Salt Lake, which served as the lead project manager and one of the payors.²²

Despite the controversy, the preschool program continues. Utah’s experience demonstrates both the enormous potential of SIBs and the intense scrutiny given to all participants.

Conclusion

SIBs are an innovative, new idea that could become a significant source of funding for a variety of social services. The promise of SIBs stems from their ability to advance new social interventions with a new source of financing, bounded by measurable markers of success. It also may be the case that SIBs, because of their complexity and downsides, work best only in a narrow set of cases that have robust scientific evidence and quantifiable results.

It is also important to keep in mind that PFS funding cannot replace general public funding. We still need federal, state, and local governments to fund the operating costs of school systems, public health care, and safety net programs—services that do not necessarily generate tangible financial returns but are essential for daily life.

The only way to find out the ultimate role for SIBs is to push the experiment further. Doing so appears to be worthwhile, because even when SIBs fail, important knowledge can be gained about what techniques do and do not work in addressing social needs.