Social Impact Bonds Matter a Lot in Public Health (Governing)

By Michael Ollove

South Carolina is turning to an unusual source to finance a new program intended to reduce the state's high rate of premature births: private investors.

If the program succeeds in reducing that rate among Medicaid beneficiaries, and reduces state spending as a result, the state will repay the principal plus a yet-to-be determined rate of return. If the program doesn't meet expectations, the state will owe the investors less or even nothing at all.

The state sees the plan as a way to launch a promising health program without having to bear the costs for years to come, if ever.

"If we achieve the results we hope for, the state wins by getting an innovative program that works and the investors get a payback," said John Supra, a deputy director of the state Department of Health and Human Services.

The financial instrument powering the project is called a social impact bond (SIB), also referred to as a pay-for-success contract. In the past several years, a number of states and municipalities have used SIBs for multiyear projects pertaining to education, homelessness and prisoner recidivism. In most cases, the money goes to a nonprofit that provides the services.

Lately, as in South Carolina, projects related to health care have begun to turn to the same mechanism. In Fresno, Calif., for example, where more than 20 percent of children have been diagnosed with asthma (compared with 8 percent nationally), the city is expected to use an SIB in a project to reduce the number of emergency room visits and hospitalizations resulting from childhood asthma.

Connecticut, New Jersey, North Carolina, Utah and Oregon also have or are considering using SIBs to fund a variety of health-related programs for such things as substance abuse treatment and early childhood health interventions. The District of Columbia is turning to an SIB for a teenage pregnancy prevention program.

Many of these health projects are geared toward prevention and save money if they help states avoid paying for expensive services, such as hospitalizations and surgeries. Because SIBs extend for many years, there is time for those savings to emerge and be measured.

Enthusiasm among some of those involved in SIBs is high. "When I first heard about this concept, I was not sold, but over time, I have become a believer," said Kevin Hamilton, chief program officer at Clinica Sierra Vista, a health-care organization providing services to the poor in central California that is a partner in the Fresno asthma project.

Hamilton says he now believes that SIBs could be used to create prevention programs for all types of chronic diseases, including diabetes and heart failure.

Others are skeptical. SIBs relieve governments of having to pay money upfront for these projects. But if the projects succeed, they have to repay the money along with what might be a sizeable return on investment, not to mention additional administrative expenses.

Critics are also quick to point out that SIBs have been around only since 2010_not long enough to have had to pay out on their investments. So while more jurisdictions are pursuing SIBs, they are unproven.

The United Kingdom is usually credited with launching the first SIB, a six-year project started in 2010 to prevent recidivism among prisoners released after serving short sentences. Although the project showed some success in the early stages (though below its goals), it was aborted earlier this year in favor of another approach. So far, it has not paid out to investors, but still could if the first-phase goals are reached.

In 2012, New York City became the first American jurisdiction to use an SIB, for a prisoner rehabilitation program that received $9.6 million in private investment money.

With funding from the Rockefeller Foundation, the Harvard Kennedy School created the Social Impact Bond Technical Assistance Lab, which is helping some states and municipalities. Applicants identify a social service project they would like to launch, often one that has shown success in a pilot program, and the lab helps them put financing together and establish measurable benchmarks.

Recently, South Carolina health officials began to think that an SIB might be the best way to attack the state's high rate of premature births. The state ranked fourth highest in those births in 2013 and more than 11 percent of Medicaid births in the state were premature.

The state had been involved in a small but effective program that sent visiting nurses or other trained health-care workers into the homes of poor, pregnant women before births and for 18 to 40 months afterward to provide support and education. The result was healthier births and healthier children in the first three years of life. With SIB funding (the amount has yet to be determined), the state plans to increase the number of beneficiaries from a few hundred to 4,000.

John Supra, deputy director of the South Carolina Department of Health and Human Services, said the hope is that the seven-year program will lead to reduced state costs in several areas, from less use of neonatal intensive care, to less spending on child abuse and special education services_all of which are linked to premature births. Supra expects the full program to be up and running in April.

As with all SIBs, in order for investors to recoup their money and turn a profit, the projects must achieve measurable success in improving health outcomes and saving money. Goldman Sachs and Bank of America are among the commercial institutional investors that have invested in SIBs.

The reason, says David Juppe, an analyst with the Maryland General Assembly who worked with McKay on his report, is that the SIBs promise high rates of return. "I haven't seen anything less than double-digit returns," Juppe said. Often, though, McKay has testified, the SIB deals are so complicated and involve so many entities, that determining exactly what results deliver what rates of return is almost impossible.

For many commercial investors, the financial risk isn't as great as it first appears to be. Many of the SIBs are designed with the involvement of foundations, which often guarantee at least some payout to investors if the program doesn't achieve the established goals. In the New York City prison project, for example, former Mayor Michael Bloomberg's foundation, Bloomberg Philanthropies, has guaranteed to pay back investors up to $7.2 million from their $9.6 million investment if the program falls short.

Cohen also questions whether SIBs actually fulfill the goal of spawning innovations. Because of the investment risk, he says the projects that are backed by SIBs tend to be models that have already proven that they do work.

"If you're investing the SIBs in things with a high probability of success," he said, "why doesn't government just do it themselves in the first place?"