Social Impact Bonds, Policy Debutante

The idea behind Social Impact Bonds is fairly simple. Instead of paying for effort, governments pay only for measureable results.
by | February 28, 2011 AT 9:00 AM

The latest social innovation tool has hit the big time. President Obama's decision to include $100 million for a pilot program on Social Impact Bonds indicates that policymakers in Washington are taking the concept seriously.

The case for piloting Social Impact Bonds is made at length by Professor Jeffrey Liebman of the Harvard Kennedy School in a new white paper from the Center for American Progress.

The idea behind Social Impact Bonds (also know as Pay for Success bonds) is fairly simple. Instead of paying for effort, governments pay only for measureable results. As The New York Times Economix blogger David Leonhardt describes them: "[P]rivate investors -- typically foundations -- pay the costs of a new program in its early years, and the government later repays the investors, often with a bonus, as long as the program meets its goals. If it fails, taxpayers pay nothing."

That last part -- government pays nothing if the results aren't delivered -- is a key lure of Social Impact Bonds. The public landscape is littered with programs that have attempted to improve the social condition, but that have delivered disappointing results at great cost.

For example, imagine a group of philanthropists "investing" $5 million to reduce recidivism among a certain prison population. Under the terms of a Social Impact Bond, a government would agree to pay back the $5 million only if the reduction in recidivism were achieved. This not only provides incentives for success, it also enables the philanthropy to "reinvest" the funds in a subsequent project.

Liebman's white paper stresses the ability of these bonds to allow and encourage private entities to innovate. "Current approaches to government funding of social services create significant barriers to innovation. Funding streams tend to emphasize inputs rather than program objectives and are often overly prescriptive, requiring grantees to use a particular delivery model. In many cases, program outcomes are not rigorously assessed, allowing unsuccessful initiatives to persist for years."

Because Social Impact Bonds are seeded with private funds, the possibility for innovation is much greater. At the same time, programs that achieve success will receive public dollars that will enable them to replicate their efforts. The approach is already being tested in the United Kingdom, and Liebman is finding that the concept is attracting interest across the political spectrum.

Many observers have noted the lack of innovation within public social service programs. Too often, political pressure from existing service providers keep ineffective programs alive indefinitely. There are a number of existing programs that currently consume resources that public officials know aren't producing results, and both conservatives and liberals should appreciate that Social Impact Bonds can be a tool to try something new. "These bonds present a way to shift resources away from ineffective programs while still remaining committed to their underlying purpose," says Liebman.

The idea is simple, but the design and execution of such programs may be another matter, however. Governments may lack the time and expertise to structure these performance-based bonds in a manner that works for all parties. Choosing realistic measures and ensuring that partner providers and their financial backers aren't bankrupted for their efforts is a big challenge. Finding innovative providers who have the management wherewithal to scale up won't be easy either.

In addition, tightly defining the treatment population is important. In order to boost results, some groups will be tempted to provide services to the most capable constituents, rather than those most in need.

Professor Liebman acknowledges that the idea has several real-world hurdles. In dealing with some of the most challenging social problems "evidence suggests that even when successful results have been demonstrated at a single site, replication and scaling up is very challenging, and it can take a significant number of false starts before a successful scalable model is discovered."

The societal ills these Social Impact Bonds are intended to ameliorate are daunting to say the least. It will be some time before we learn the extent to which they can foster the spread of effective programs. As Stephen Goldsmith noted in his book, The Power of Social Innovation, governments are "often stuck with entrenched underperforming social safety net systems.... Government's ability to collaborate has not kept pace with the growing complexity of these social production systems."

Pending funding, cities and states would be eligible for funds to design these bond programs and start pilot efforts. They would have a chance to work with philanthropic social investors to use Social Impact Bonds to try some new approaches. Ultimately, the only thing that matters is results.