fb-pixel Skip to main content

Social impact bonds: Getting better at doing good

When you walk into the unassuming Chelsea headquarters of Roca, you see the hallmarks of a thriving, urban non-profit. Teenagers draped on couches in the lobby. Posters proudly touting participants’ achievements lining the walls. Doors leading to a modest gym, job training classes, and crowded offices.

But there is a word you hear repeatedly around Roca that sets it apart: data. Dedicated to reducing recidivism in high-risk youth and helping disengaged young people break the cycle of violence and poverty, Roca is compulsively data-driven and outcomes-oriented.

Their four-year program for juvenile offenders is measured, analyzed, and evaluated every step of the way. Each participant is closely monitored, which allows Roca to see which interventions work and which don't. An enormous database tracks progress through the program and holds the organization and its participants accountable for the results. Those results, so far, have been exceptional. Last year, 89 percent of young men in the final stages of the program had no new arrests and 69 percent were holding down jobs.


That success is worth replicating. But in a time of limited government resources and tenuous economic recovery, how can we take best practices carefully honed by local service providers like Roca and scale them, to systemically address some of our Commonwealth's and country's most wrenching social shortcomings?

That challenge brought Roca and several partners to the helm of Massachusetts's first iteration of the "social impact bond" model, an innovative financial tool rooted in the principles of "pay-for-success" financing. It's a project — and an approach — that deserves careful consideration from Beacon Hill to Capitol Hill.

It works like this: Government, investors and service providers come to the table around an area of pervasive and systemic social need, such as recidivism, homelessness, unemployment, early childhood education, and public health. The philanthropists and investors provide additional working capital. The service provider uses this capital to expand their efforts and reach more people — in Roca's case, to grow their highly successful model out from the Boston area to Lynn, Springfield, and several other communities in between.

Programs are given a set amount of time to put those funds to work before a third-party evaluator determines whether or not the initiative has met its goals.


Should the Roca project fall short, the government won't pay a dime. However, if an independent evaluation reveals that Roca has reduced days of incarceration, improved job readiness, and boosted employment outcomes compared to a control group, the Commonwealth will pay returns to the initial investors. At the same time, Massachusetts taxpayers stand to save up to $45 million from a reduction in the number of people we pay to incarcerate or rehabilitate. That's a government investment worth making.

Pay-for-success financing allows us to rigorously evaluate and improve the effectiveness of our social services. It gives us the ability to scale what works and replicate best practices. And it brings traditionally untapped resources off the sidelines to correct our nationwide underinvestment in prevention — a costly misallocation of public funds that we pay for dearly in our courts, hospitals, schools, and streets.

In Washington, social impact bonds hold the potential to shake up today's over-simplified and over-partisan approach to federal budgeting and reveal common ground. Too often the debate gets undermined by political talking points that try to fit our fiscal challenges into two sides — cutting or spending. What's missing is a conversation about investing. In the era of "big data" we can test, measure, and quantify the impact of a dollar more precisely than ever before. Government has no excuse not to use these tools to improve its capacity to help those who need it most.


With that in mind, I joined a bipartisan group of eight colleagues, led by Representatives Todd Young, a Republican from Indiana, and John Delaney, a Democrat from Maryland, in introducing the Social Impact Bond Act earlier this year. Our bill — the first major legislative proposal to bring the social-impact bond model to federal policy — would establish a $300 million fund within the Treasury Department to support state-level projects across the country. Projects made possible by the Treasury fund would be selected for their ability to demonstrate measurable outcomes, potential savings to state and federal government, and real impact on real lives.

There is undoubtedly work to do to solidify this exciting new model and ensure its effectiveness. There will be kinks to work out and improvements to be made. But by breaking down traditional barriers between the public and private sectors, social impact bonds hold the potential to help government do what organizations like Roca have been working towards for decades: Getting better at doing good. That's a mission Washington should take to heart.

Joe Kennedy III is a US representative from Massachusetts.