The first Social Impact Bond (SIB) in the United States ultimately failed in its mission to reduce youth recidivism at Rikers Island by 10 percent, generating a host of questions as to the scientific and analytical usefulness of the data gleaned from the experiment, as well as to the efficacy and future use of SIBs.
As Key Elements Group LLC has previously discussed in our ongoing series, SIBs are a unique intersection of public and private sectors working to ameliorate pressing social issues. A private investor – in the case of Rikers Island, Goldman Sachs – fronts money for a program that employs new or innovative strategies to address an intractable problem. Following a third-party evaluation at contractually predefined benchmarks, the initial investors can potentially profit from these projects if they succeed in saving the government money.
The key selling point is the minimal risk to the tax-payer, who is theoretically off the hook if a project goes south. If a project succeeds, proponents argue that everyone wins: the government secures future savings, individuals assisted by the program lead better lives, the investors receive a return on investment, and the tax-payer benefits from an improved social state and lower social spending costs that require fewer tax dollars.
Popular on both sides of the Atlantic, SIBs have made waves in elite British and U.S. philanthropy circles. The Obama Administration has displayed its interest, earmarking millions of dollars in federal funding for potential SIB projects.
The Rikers Island project – however – failed to make a measurable impact. A therapy program designed to decrease recidivism among interned youth, the project sought to reduce return-to-prison rates by 10 percent. A third-party assessment by the Vera Institute concluded that recidivism among the youth population at the prison did not decrease.
Looking at the program’s trajectory, the failure is not surprising. A number of essential components fell apart, setting the experiment up with a number of sizable hurdles. Wardens were organizationally incapable of maintaining the project’s control group, thus complicating scientific comparison and accurate assessment; the New York City Education Department withdrew its operational support; and a contractual complication reduced the budget of the nonprofit organization slated to actually implement the program. Some project insiders also point out that incidental costs accrued by in-kind support from public officials and staff still cost public dollars.
Even outside of the contours of this one project, critics point to a number of issues concerning SIBs. Some fear that they divert philanthropy dollars from other causes. If the liability of SIBs proves greater than proponents originally thought, philanthropic underwriters may step in to mitigate the potential loses for investors. For the Rikers Island project, Bloomberg Philanthropies guaranteed over $7 million, meaning that Goldman Sachs could only lose $2 million max on the investment. This not only contradicts the claim to cost-cutting and private risk, but could also saps valuable philanthropy dollars from other causes and projects.
A single failure – especially in the first-ever project of this type in the United States – is not sufficient enough to spurn SIBs. In fact, as a scientific and quantitative system, SIBs stand to benefit from the hard data of underachieving projects. New methodologies for enacting social policy are virtually never perfect when first instituted. Only a number of similiar projects, faithfully followed through to their completion, can begin to construct a clearer idea as to how well and to what extant SIBs will fit into U.S. philanthropy.