Social Impact Bonds in Criminal Justice: A Deal We Can’t Refuse?

by Jennifer R. Zelnick, MSW, ScD

Over the past several years there has been a lot of buzz about Social Impact Bonds (SIBs). More recently criticisms have emerged, along with early results of SIB projects. This article will examine existing SIBs in the criminal justice context, and raise some concerns.

What are SIBs? SIBs – known also as “pay for success” – are not bonds but complex, multi-stakeholder loans between government agencies, investors, intermediaries, service providers and evaluators in which an investor provides up-front funding for a specific project chosen to produce a specific, cost-saving outcome. If certain benchmarks are met, up-front loan money is re-paid with interest; if outcomes exceed pre-determined levels of success, pay for success payments are triggered that increase investor rewards. The “cashable” cost savings are key, because they are presumably the source of revenue used to pay back the investors. SIBs have been touted as useful in areas where government funds are tied up in remediation to the exclusion of funding prevention programs. (This describes the general model of SIBs in the United States, though there are variations).

Advocates of SIBs speak of their ability to transform government services, but their goals aren’t very idealistic. Lyndon B. Johnson introduced the War on Poverty in 1964, citing its chief weapons of attack as “better schools, and better health, and better homes, and better training, and better job opportunities to help more Americans, especially young Americans.” This was primary prevention through government social welfare policy, propelled by the success of the civil rights movement, and explicitly designed to increase opportunity and prevent the need for policies and programs to mitigate poverty and neglect.

However, times changed. Despite the fact that it was never fully funded or enacted, the backlash against Johnson’s War on Poverty and Great Society policy programs is in part what propelled the ascendancy of Ronald Reagan and the attack on social welfare spending, which has persisted ever since. Under Reagan in the 1980s, the “War on Poverty” became the “War on Drugs,” which along with mandatory minimum sentencing laid the groundwork for mass incarceration and racial disparities in our criminal justice system. These days, structural inequalities play a huge role in determining who is more likely to be imprisoned, as well as individual and community opportunities for success.

But structural inequalities and improving opportunities are not the focus of SIBs. SIBs in criminal justice focus on what might be called “secondary prevention” – reducing recidivism, preventing incarceration for those already at-risk and increasing employment in the same populations. Those are useful goals and individuals may benefit from these services, but such projects are not going to address the social issues that disadvantage huge swaths of American communities. However, the model of SIBs, with a narrow focus on pay for success outcomes and cashable savings, precludes investment in primary prevention. The cashable savings of criminal justice SIBs include the cost of prison stays and in some cases the cost of public work that would be replaced by “work crews” of formerly incarcerated people. It does not seem likely that fixing underlying structural inequalities results in short-term cashable savings.

Another important legacy of the Reagan era and the years of attacks on public social welfare expenditures is the expanded role of the private sector in social welfare policy. Experiences with prison privatization, and more recently of probation and parole services, are rife with reasons why injecting profit motives into public services runs counter to the public’s interests. In addition to outright privatization of social welfare services, government contracting has also been influenced by business practices. Performance-based contracts, which pay service providers based on outcomes rather than service delivery outputs, have been widely adopted. SIBs create a new avenue for privatization of social welfare through financing; John Roman, a senior fellow at the Justice Policy Center at the Urban Institute and an SIB advocate, has called SIBs “performance-based contracting on steroids.”

SIBs have support at the highest level of government. In 2009 the Obama administration introduced a social innovation fund to spur private investment in social problems and make up for “limited taxpayer dollars.” At a time when Republicans and Democrats can’t seem to agree on anything, a federal SIB bill, H.R.4885 (the Social Impact Bond Act), introduced in June 2014, enjoyed bi-partisan support. The Social Impact Bond Act promoted SIBs by directing the Treasury Department to seek SIB proposals from state and local governments, though the bill failed to pass. A Harvard Kennedy School Social Impact Bond incubator is one of the most popular programs among students, and provides “technical assistance” to states considering SIBs.

There are at least four criminal/juvenile justice SIBs in the United States and UK:

• The first-ever SIB at HMP Peterborough (a private prison run by Sodexo Justice Services) launched in the UK in 2010. The project intermediary, Social Finance, Inc., went on to become a prominent SIB consultant. The goal of the SIB (dubbed “One Service”) was to reduce recidivism at Peterborough by 10% in 3 cohorts over a 3-4 year period. Seventeen charitable foundation investors loaned £5 million to the project, with a potential rate of return of 7.5-13%. The intervention method included an intensive case-management and peer support model to link offenders with services/supports to ease community reentry after serving short terms (12 months or less) in prison.

• The first U.S.-based SIB launched in 2012 and aimed to reduce recidivism among adolescent offenders at New York City’s Rikers Island jail by 10%. The one-year recidivism rate for juveniles was almost 50%. The project, the Adolescent Behavioral Learning Experience, is funded by a $9.6 million investment from Goldman Sachs and backed by a $7.2 million guarantee from Bloomberg Philanthropies. The intermediary (MDRC), service provider (Osborne Group) and evaluator (Vera Institute of Justice) are all experienced in prison-related research and services. If successful, the Rikers project will pay out as much as a $2.1 million return to investors. The program at Rikers is based on Moral Reconation Therapy (MRT), a form of cognitive behavioral therapy.

• In December 2013, New York State launched an SIB targeted to reduce recidivism, increase employment and maximize participation in transitional employment among 2,000 formally incarcerated individuals in New York City and Rochester. Social Finance, Inc. is the intermediary, and project funds include $13.5 million from Bank of America, which raised money from over 40 investors. The project also received funds from the U.S. Department of Labor’s Workforce Innovation Fund. The intervention – comprehensive employment services including training, transitional employment and job placement – is provided by the Center for Employment Opportunities.

•In January 2014, Massachusetts launched an SIB to reduce incarceration by 40% among highly at-risk youth. The intermediary in this project is Third Sector Capital Partners along with New Profit, Inc., with a $27 million investment from Goldman Sachs and several private foundations. The intervention, consisting of training in job readiness, educational readiness and life skills among at-risk youth, will be delivered by ROCA (a long-standing, highly respected service provider). Like New York State, Massachusetts also received funding from the Department of Labor’s Workforce Innovation Fund.

Thus far there have been few actual results. In August 2014, the HMP Peterborough SIB failed to meet the benchmark for its first pay for success payment to investors (they achieved an 8.8% decrease in reconvictions, not the 10% required to trigger the payment). Earlier in the year, the UK Ministry of Justice announced a country-wide roll-out of a reentry program modeled on the Peterborough intervention, forcing the cancellation of the third SIB cohort because the national roll-out eliminated the control group needed to evaluate the Peterborough program.

While SIB advocates have spun both developments as a success (the government adopting the model “proves” that SIBs can spur innovation and policy change, and the project is still geared to meet a lower target of a 7.5% decrease in reconvictions for a pay for success payout when it ends in 2016), it hardly seems like strong evidence for the SIB model. In fact, it appears that providing the intervention without investor payouts makes more sense than the SIB approach. Despite the absence of results, the emergence of specific SIBs allows for a grounded critique of such projects, which includes the following.

• Cost: Because of the layers of contracting and complexity, SIBs entail extra costs (legal, evaluation, project management) beyond service delivery. If a program shows results, isn’t it cheaper for the government to just borrow the money itself, especially given the high rates of return to private investors?
• Risk: One of the key arguments for undertaking an SIB is that governments cannot afford to risk funding prevention services. Yet investors seem risk adverse too, only wanting to invest in SIB interventions that have been proven to work based on extensive study, and demanding foundation funds to back their investments.
• Performance-based Contracting: SIBs codify rigid metrics that link outcomes and pay for success financial returns, but metrics have led to problems in other performance-based contracts, such as creaming/cherry-picking (serving those more likely to succeed), perverse incentives to “cook the books” and narrowing the scope of services to the identified outcomes despite the complex needs of service users.
• Accountability: Who has a say in choosing SIB projects? Once a project is outsourced to the SIB model, what determines public accountability? As private contracts, SIBs are not transparent or accountable to the public.
• Ethics: What are the ethical implications of creating a profit incentive based on saving money on services for needy, vulnerable and disenfranchised populations?
• Public Interest: The public should be concerned about SIBs as taxpayers and citizens. Do we want our tax dollars to go to private investors in the form of profit, or would we prefer that any savings stay in the public system to reinvest in the public good?
• Human Rights: SIBs are developed for fragile clients with limited rights who have no voice in their development, yet lack oversight for the needs of these populations. Service users and frontline providers should have a say. For example, former recipients of Moral Reconation Therapy (the intervention used in the Rikers Island SIB) and frontline service providers have been critical of MRT as an adequate and appropriate intervention, but there is no place for such criticisms in SIB development.
• Scope: SIBs do not enhance the ability of government agencies to address the core issues behind social problems; instead, they constrain possible solutions to secondary prevention measures that can potentially generate short-term, high rates of return in a performance management framework. There are viable alternative approaches to reduce recidivism (e.g., rolling back policies that increase arrests for low level crimes) and to free up money tied up in costly remediation for other purposes (e.g., the Maryland Opportunity Compacts, which shift savings in remedial public service areas to prevention services).

Currently, given the desperate need for funding among service providers, the lack of taxpayer dollars for social welfare needs, bi-partisan government support and support among liberal think-tanks and foundations, SIBs seem like the proverbial “deal we can’t refuse.” Voices closer to the ground hint that exuberance may not be universally shared. At a 2013 forum on SIBs convened by the Prisoner Reentry Institute at the John Jay College of Criminal Justice, representatives from the Rikers Island SIB project countered advocates with the reality from the frontlines, where implementation of the program was messy and slow, and providers doubted the efficacy of “one small intervention” – however evidenced-based it was. A New York City leader in Mayor Michael Bloomberg’s administration noted, “if we had the ability to invest in new programs within the city budget, then why would we do this?” These voices, largely absent from the larger discussion of SIBs, indicate lukewarm support at the implementation levels.

“I’m not saying that the market is evil,” said Mark Rosenman, professor emeritus at the Union Institute and University in Cincinnati, who was quoted by the New York Times. “But I am saying when we get into a situation where we are encouraging investment in order to generate private profit as a substitute for government responsibility, we’re making a big mistake.”

One of the best things that might develop out of the SIB phenomena would be to focus attention on the contracting process and augment it with a real, community-based dialogue among service providers, advocates, current and formerly incarcerated people and their families about publicly-funded social welfare services. In order to prevent abuse of taxpayer funds and perverse incentives, and to promote human rights and investments in public policy interventions, we need to educate ourselves and make demands of performance-based contracts.

Recently the American Federation of State, County and Municipal Employees (AFSCME), which opposes Social Impact Bonds, issued a set of “guardrails” to guide decision-making around SIBs – which is a useful model for other groups that want to develop standards for their specific populations, services and public policy interests.

Maybe we can’t refuse the deal, but we can at least get a seat at the negotiations.

Jennifer R. Zelnick is Associate Professor and Social Welfare Policy Chair at the Touro College Graduate School of Social Work in New York City. She has an MSW in community organizing and planning, and a doctorate in work environment policy. Her research focuses on the impacts of privatization and workplace restructuring on the work, health and well-being of health and human service workers. She provided this article exclusively for PLN.

Ed. Note: According to recent news reports, the 2012 SIB to reduce recidivism rates for juvenile offenders at the Rikers Island jail complex was terminated by the Vera Institute of Justice in June 2015 after it failed to meet the project’s goal of reducing repeat offenses by 10%, resulting in a $1.2 million loss for investor Goldman Sachs. Another partner in the failed SIB, Bloomberg Philanthropies, lost $6 million in grant funds.

Subscribe today

Already a subscriber? Login