Deidre Schmidt, Executive Director, Affordable Housing Institute (AHI)
Social Impact Bonds and Human Capital Performance Bonds are the hottest new things in public finance. The concept is simple: the government agrees to fund preventative programs that reduce long-term public expenditures or create new tax revenues by financing against those expected cost savings or revenues. The government pays for those preventative measures, only if the desired social outcomes and/or financial savings are actually realized. The execution however, is complicated. It involves creating clear baseline and progress measures, tracking causality of social programming efforts, creating risk sharing mechanisms that make the most out of private sector participation and coordinating action across different agencies in government. Significant devils live in those details.
The first Social Impact Bond was issued in Peterborough, England in September 2010. It was structured to pay for nonprofit programs that are expected to reduce convict recidivism. If the measures are successful at reducing recidivism rates by at least 7.5% (when compared to a comparison group), then the public agency will repay investors, with the opportunity to earn a return up to 13%. If not, the government owes nothing.
This pioneering effort has caught the imagination of many of us on this side of the pond. The US Federal government and many states and are looking carefully at this concept, including my home state of Minnesota and my adopted home of Massachusetts. So far, each has taken a different approach, with Minnesota exploring HUCAP and Massachusetts examining SIBs.
There are many commonalities and differences between these two nascent efforts. Here are a few that I think are most important to understand:
Economic Measures and Benefits – US versions may vary from the Peterbourgh model in a significant way; they are likely to measure and pay for performance based on economic benefits to the public sector (as a result of future cost savings or increased tax generation). In Peterborough, the measure of success is reduction in recidivism, regardless of whether that desirable social outcome saves the public sector actual dollars. This isn’t just a philosophical shift, it has practical consequences; requiring much more refined tracking capacity.
Nature of the Investment- In the HUCAP model, the state would issue annual appropriation bonds that offer a relatively low rate of return based on investors’ assessments that the state will honor these obligations from year to year. These “moral obligation bonds” have been used for years for other purposes and run at rates similar to AA bonds. The SIB model would be structured more like equity, where the investor takes on 100% of the financial risk and won’t be paid if the service provider fails to achieve the target benchmarks. Taking on these risks means that investors will require higher rates of return than the HUCAP model; perhaps 3 fold.
Administration / Coordination – Both the HUCAP and SIBs proposals require coordination. In the SIBs model, this is envisioned to take the shape of an intermediary, a private entity which would manage virtually all aspects: negotiating terms with government, private investors and service providers; engaging an evaluation firm to measure and track outcomes; and managing any “course corrections” that would be necessary in order to achieve the desired outcomes – including the prerogative to replace non-performing providers. In the HUCAP model, the state itself manages the financial mechanism, offering the bonds and creating a “performance pool” of funds available to pay for services that are successful. An oversight Board and independent economists would then administer the performance monitoring aspects, qualifying service providers that want to participate and verifying that the economic benefit has been achieved through their efforts. They would not be able to replace service providers mid-stream (but arguably don’t need to, as their payments don’t occur until outcomes are achieved).
Interim Funding for Service Providers – In the HUCAP model, service providers are responsible for finding the resources to fund their programs in the period before they prove their success. This means that service providers with access to working capital (nonprofits who are very effective fundraisers, or for- and non-profits with existing reserves) will be able to participate. A working capital fund may be developed in conjunction with the HUCAP mechanism, which could help to solve the interim cash-flow challenge for service providers, but would add cost and complication for those service providers who don’t have the financial wherewithal to fund themselves in the interim. Under the SIB model, the intermediary will make payments to the service providers to cover the cost of the program each year. This allows service providers to participate based on their competency in the service arena and not the fundraising arena, but this adds a cost to the overall program.
Social Impact Bonds and Human Capital Performance Bonds are appealing concepts that require thoughtful structuring and prudent application. They aren’t a panacea, but they are an inspiring and fresh way of thinking about some of our most vexing societal cost burdens like premature birth, homelessness and recidivism. As a houser, I see potential in funding supportive services in housing, which are notoriously difficult to fund, but so essential to tenant stability and success. Now, comes the critical stage of figuring out those devilish details.
When Ben Franklin said, “An ounce of prevention, is worth a pound of cure”, I wonder how he arrived at his discount rate?