Social Impact Bonds (SIB), also known as “pay for success” loans, are a novel form of financing social service interventions, including education initiatives. First piloted six years ago in the United Kingdom and now making their way to the United States, SIBs aim to leverage private funding to start new programs or scale proven ones. Broadly speaking, the instrument works like this: Private lenders and philanthropists deliver dollars—the bond—to a nonprofit provider that, in turn, implements the intervention. A government agency pays back the bond principal with interest, but only if the program achieves pre-specified results.
In its ideal form, an SIB has the potential to be a triple win: Governments receive risk-free funding to test or expand social programs that could help them save money; investors reap a financial return if the program works; and providers gain access to new sources of funding. To ensure that the deal will benefit all parties, due diligence occurs on the front end, including selecting a program provider, estimating government savings, and developing an evaluation method.
To date, the discussion on SIBs has been largely conceptual, engaging both supporters and skeptics alike. But a fascinating new report written by MDRC President Gordon Berlin provides a first-hand look at an early SIB experiment. In July 2012, Bloomberg Philanthropies, Goldman Sachs, MDRC, and New York City teamed up to create a four-year, $9.6 million SIB that funded a behavioral therapy program at Rikers Island jail. The objective was to reduce adolescent recidivism by at least 10 percent. The deal was structured in such a way that after three years (summer 2015), the program evaluators would determine whether to continue into a fourth year or pull the plug.
The program fell short of its goal and folded early. Using quasi-experimental methods, independent evaluators found that the intervention did not reduce recidivism for its participants. Berlin describes the implementation challenges, among them the harsh prison environment and hurdles to program completion (just 9 percent of participants completed the full intervention). The null findings resulted in Goldman taking a loss of about $1.2 million (Bloomberg’s grant protected the bank from absorbing the entire loss). Meanwhile, the New York City government was freed from payment: Under the SIB terms, it would have paid back the principal if the program met the 10 percent threshold—and more (principal plus interest) if the reduction had been greater than 10 percent.
Based on this experience, Berlin provides helpful lessons for those contemplating SIB financing. To highlight just a few: First, he notes that SIBs, as currently structured, tend to carry more risk on the lenders’ side. SIBs are commonly conceptualized as high-risk/high-return “venture capital” for the social service sector. Yet the “bond-like returns” governments may be willing to pay aren’t commensurate with the risk that SIB investors have to bear, thus limiting their viability. Second, Berlin discusses the incredible intricacy of SIB deals and how the “tyranny of inflexible loan agreements” can stifle on-the-ground flexibility. For instance, the Rikers Island program found it more difficult than expected to recruit participants; consequently, the parties had to negotiate changes in the already complex SIB contract. He concludes, “Each SIB will confront this tension between inflexible contract terms and the need to respond to unexpected operational challenges.”
This report is important reading for anyone interested in SIBs, including their potential use in financing education interventions like pre-K or scaling high-quality schools. (Utah is presently using an SIB to fund pre-K for 595 low-income children, with the aim of reducing special education placements in elementary school.) To be sure, SIBs hold the potential to unlock non-traditional financing to help grow and replicate innovative programs; they also rightly focus on results and rigorous evaluation—features not traditionally emphasized in the public sector. But as this report acknowledges, such arrangements must be done thoughtfully and with a strong dose of realism. Berlin observes, “Reality is turning out to be more nuanced than either proponents or detractors have promised.”
SOURCE: Gordon L. Berlin, “Learning from Experience: A Guide to Social Impact Bond Investing,” MDRC (March 2016).
