Beats | December 23, 2014

Rebuilding Lives, Reducing Costs: Social Impact Bond Finances Jobs Over Jail

The team at


The snow was blowing and it was in the 20s on Wall Street on the day after New Year’s but dozens of mostly young, mostly black, and mostly unemployed men showed up for job training and placements on the first working day of 2014.

The men, all recently released from prison, were making an investment in their own future. They were eager to enroll at the nonprofit Center for Employment Opportunities, better known as CEO, which operates out of the 18th floor of a building in the heart of New York City’s financial district.

Other, more familiar, fixtures on Wall Street — including former Treasury Secretary Larry Summers — are making an investment in the young men as well.

The ex-offenders lining up for employment help were among the first of 2,000 CEO clients in New York City and Rochester whose job training costs are covered under a “pay-for-success” contract financed by private investors. Bank of America Merrill Lynch offered the investment to its private banking clients, Between Thanksgiving and New Year’s Eve last year, more than 40 high net worth investors committed $13.5 million.

If enough of the formerly incarcerated men stay out of prison, the investors stand to recoup their principal plus a return that can range between 5 and 12.5 percent. If CEO’s program fails to significantly reduce recidivism (with at least an 8 percent reduction in jail and prison days), investors will lose up to 90 percent of their money.

[blockquote author=”Tracy Palandjian, CEO, Social Finance” pull=”pullleft”]Risks are privatized and gains are socialized. That’s a new model, one harnessing private capital to serve the public good.[/blockquote]

Pay-for-success contracts, colloquially known as “social impact bonds,” are attractive to cash-strapped states and cities because they are obligated to pay only when the results are proven and the savings are realized. For investors, the investment proposition might more accurately be called “repaid-for-success.” Private investors provide the upfront risk capital to finance the preventive services. They get their capital back, plus a financial return, out of the government’s avoided costs from a successful intervention.

The contract, issued by the state of New York, is not the first pay-for-success contract, but it is the first to be offered directly to individual qualified investors. In earlier deals, institutional investors, like Goldman Sachs, backed social impact bonds with their own capital; the New York State contract is the first test of private investor interest in financing this new way to deliver preventive social services. With a minimum investment of $100,000 and a five-and-a-half-year lockup, the private investors committed an average of $300,000 each. The whole deal was brokered by an innovative nonprofit called Social Finance, which has helped bring the pay-for-success model from the U.K. to the U.S.

“The idea that there may be a different way to attract new capital, coupled with ways to improve the actual results, was naturally attractive,” says Paul Bernstein, who invested as executive director of the Pershing Square Foundation, Karen and Bill Ackman’s philanthropic vehicle. Bernstein says Bill Ackman, known as an activist investor who makes big bets, took a personal interest in the innovative structure as a way around government’s seeming inability to adequately fund even proven prevention techniques.

“If you really want this thing to scale and create a new funding model, you had to build a commercially viable approach, and they did that by bringing in BofA,” Bernstein says. As for the investment, he says, “It’s clearly not going to offer the best return you could get on any investment, but it’s a viable part of a diversified portfolio.”

[blockquote author=”Surya Kolluri, Bank of America Merrill Lynch” pull=”pullright”]Velocity is an ‘aha’ moment. It’s very different than a grant in which the money is gone.[/blockquote]

Other investors include the Utah philanthropist James Sorenson’s Sorenson Impact Foundation, the Robin Hood Foundation and the Laura and John Arnold Foundation. “Pay-for-success is a funding arrangement that allows governments to make risk-free investments in an effort to improve citizens’ lives and ensure that taxpayer dollars are allocated in the smartest, most efficient way,” said Leila Walsh, a spokeswoman for the Arnold Foundation, who added that any returns would be reinvested in future projects to scale up those that prove to have impact.

Performance-based contracting is common in areas such as energy efficiency, in which predictable savings allow energy service companies to guarantee their results. But they’re new for social services, where conventional budgeting processes generally pay for services, not outcomes. To government bureaucrats, a reduced number of prison bed-days is at least as appealing as a lower electricity bill.

“What’s most exciting intellectually is that the investment alpha is directly and explicitly linked to the social impact achieved,” says Tracy Palandjian, the chief executive of Social Finance. “It’s the very betterment of lives — the person getting a job, keeping a job, staying out of trouble — that is the source of the investment returns via government savings.”

Life Skills

For social service providers, social impact bonds represent a sea change not only in the amount, but in the kind of available capital. Payment in advance eliminates the challenge of meeting expenses while waiting for government reimbursement. Since investors are repaid based on outcomes, not inputs, unrestricted funding is not tied to specific program components and can be spent on what works best. With costs covered in full, providers can focus on services, not fundraising. All of that is intended to help high-performing nonprofits with proven interventions thrive, not merely survive.

At CEO on a recent day, dozens of men cluster around tall bistro tables with bright green chairs in the glass-enclosed reception area, waiting for their next work assignments. The agency runs its own social enterprise and contracts with city agencies and other companies to provide transitional employment that builds basic work skills and habits.

Before they go out to work, however, CEO helps the men identify their own motivations in a one week Life Skills Education class. In two classrooms around the corner from the reception area, two Life Skills sessions are underway: one for younger participants ages 18-25, the other for older participants, some of whom have served sentences for more serious offenses, including murder and armed robbery.

In the first classroom, students are reading from an essay by basketball star Michael Jordan. “Everyone had a different agenda for me, but I had my own,” reads one young man. Heads nod around the room. One student jokes his mother wanted him to be a basketball player. He wants to start a clothing line.

“I just don’t want to go back to jail,” says another.

In the other classroom, an instructor named Mary is leading about 20 men through a set of short, direct questions. “What is your goal when you leave this class?” Some of the responses, hesitant and mumbled sound like lines the men may have heard from others: “To better myself.” “To take care of myself and my family.” “To be a positive member of my community.” Mary keeps pushing.

One man, wearing a collared shirt and glasses, lifts his head. “To get a job,” he answers. Bingo.

“Today is graduation day,” Mary says, as she distributes certificates and hugs. “Today marks something you started and something you finished.”

The men will receive a badge, work boots, and their first assignment as official employees of CEO. Each employee is responsible for signing up for transitional job placements and can work at those sites for up to 75 days before moving into a permanent job placement. CEO seeks to place graduates in full-time jobs, ranging from the retail sector to food service to the construction trades. While challenges will remain, these are important steps on the ladder toward economic security and self-determination.

Measuring Outcomes

Prison recidivism has been an easy and obvious target for the first social impact bonds in both the U.S. and the U.K. But the model has promise in any case in which a proverbial ounce of prevention is worth a pound of cure. Measurement methods are improving, and social impact bonds are being developed for early childhood education, foster care, asthma, diabetes, and many other challenges.

In the case of recidivism, fewer returning prisoners means dramatic savings in prison occupancy, victim assistance, and other social costs. The average number of days of incarceration per person is easily measured, as are the state’s financial savings. Determining whether an individual is or isn’t in prison is binary, rather than the shades of gray that can color program results in other areas.

CEO estimates intensive job support for people coming out of incarceration saves $60,000 per individual per year. New York state was willing to pay about half of that, or $85 for each bed-day saved. High levels of incarceration, particularly of young black men, is an increasingly charged issue in local and national politics, but pay-for-success financing transforms it into a rational calculation.

The state of New York can repay the investors’ capital, with a modest premium, and still save millions of dollars in the long run. (It doesn’t hurt that the U.S. Department of Labor will cover the repayment for service delivery taking place in the first two years, under a pilot program to test these kinds of financing arrangements.) That doesn’t account for the improved prospects of the target population and the community at large.

Investors will start to receive repayments if the project reduces the number of nights the clients in CEO’s target group spend back in prison by at least 36.8 bed-days per person, or 8 percent, compared to a similar group that does not receive CEO’s services. If performance exceeds those thresholds, investors can earn up to 12.5 percent after five and a half years. Once the minimum is met, investors get 100 percent of the state’s savings until their capital is repaid, then split additional savings 50-50 up to the cap. If reductions are even more dramatic, the state keeps the additional savings. Most observers expect returns in the mid-single digits.

The program must also show a 5 percentage point increase in employment, perhaps the key determinant in staying out of prison. In New York state, an estimated 44 percent of formerly incarcerated individuals on parole who are unemployed return to prison within two years. For those with part-time unemployment it’s 29 percent, and for those with full-time employment, it’s 23 percent.

Pay-for-success contracts are not appropriate for bleeding-edge innovation; they typically work best to scale up proven, battle-tested interventions. CEO has honed its four-step process of life-skills training, transitional job training, full-time placement, and job retention over 35 years. A random-assignment trial in 2004 found that CEO’s program achieved a 16 to 22 percent reduction in recidivism for recently released participants; some high-risk groups showed even better results. Employment results were less conclusive in the original evaluation, but CEO’s internal data shows that additional post-placement counseling consistently boosted job retention over the last 10 years. With pay-for-success funding, CEO offers such follow-up help.

“The pay-for-success contract looks at, ‘What is the benefit? What is the cost?” says Marta Nelson, who previously headed CEO’s New York City office. “The benefits outweigh the costs, so let’s pay what it actually costs.’” This shift is significant, as organizations like CEO can be paid adequately to deliver the comprehensive suite of services clients actually need.

Perhaps even more important, the contract is driving increased cooperation between the Department of Corrections and CEO. The data shows that CEO is particularly effective with high-risk clients that it can reach as soon as possible after release. In the new program, the participant meets jointly with a parole officer and a CEO outreach worker in the very first weeks after release. That “match candidate” meeting is intended to convey that the candidate has been selected for a program specially tailored to his needs.

“We message it in a very positive way,” Nelson says. “And because it’s parole, there’s an element of a ‘special condition’ that conveys that you are required to go to the program. That combination gets people to walk in the door.”

More broadly, the shared incentives mean state officials are eager to see the program work. CEO and state officials zip spreadsheets back and forth monthly, or even weekly, tracking enrollment rates to assess if the project is attracting the desired participation.

“Under an old contract, government is buying a service. They are worrying about whether you are sending in the forms, that you are not over-spending the budget. They are looking at expenses and services and not the bigger picture,” Nelson says. “Once this is put in the frame of a benefit to the state, it opens it up to a much more collaborative way of working with the state.”

The shift from measuring activities to measuring outcomes should be welcomed by top-tier social service providers that have evidence-based, rigorously evaluated models for long-term positive behavior change in high-risk, high-cost populations. But accountability and measurement should also shake out non-performers. Agencies that deliver mixed results or low repayment rates for investors are not likely to be selected for follow-on pay-for-success contracts.

CEO is confident it can replicate the results from its earlier random-assignment evaluations. “There’s a risk we won’t, so we could suffer,” Nelson says. “If we don’t succeed, it’s going to be on the front page.”

Road Show

If it all sounds complicated, it is. In the summer of 2012, New York asked for proposals to take advantage of the U.S. Department of Labor money to test social impact bond financing for job training programs. Selected to design and manage the project, Social Finance identified CEO as the provider of choice. In April 2013, the state legislature agreed to double the length of the program from two years to four years, supplementing the federal funding with state money.

Brace Young, a former Goldman Sachs partner and chair of Social Finance’s board of directors, helped bring Bank of America on board. After several meetings, Andy Sieg, BofA’s head of global wealth and retirement solutions, told Young, “This is fascinating. I don’t know what it is, but I’m willing to dedicate a couple people from my team.”

Rockefeller Foundation, a leader in promoting the social impact bond model, agreed to provide a 10 percent first-loss guarantee for all but the philanthropic investors. That provided modest reassurance for investors but is far lower than the similar guarantees extended in other social impact bond offerings. “We didn’t want heavy-duty training wheels, but the market wasn’t ready for a completely naked vehicle,” says Palandjian of Social Finance.

Some of the negotiations were tough. CEO sought assurances its current donor list wouldn’t be cannibalized for the new investment vehicle. Investors wanted assurances that fickle future legislators wouldn’t renege on commitments. State officials sought a higher bar before payments are triggered. “You have to have a lot of room to make sure that even at the bar the investor is comfortable with, the state is saving a lot of money,” Palandjian adds.

Measurement is key. The New York State Department of Corrections and Community Supervision will evaluate a randomized control trial that compares the employment and recidivism outcomes of the 2,000 participants served by CEO with a control group referred by parole officers to traditional service providers. An independent validator, Chesapeake Research Associates, provides an additional layer of review.  (One complication: Members of the control group may enroll in CEO’s services through other channels.) Investors will get quarterly updates on the project’s performance.

In all, it took 15 months to bring the many parties and moving parts together and finalize a 130-page contract between the state and Social Finance, with CEO as a subcontractor.

Through the fall of 2013, Bank of America arranged a series of meetings between clients, their financial advisors, and the Social Finance team to drum up interest in the private placement. The complicated financial vehicle was unfamiliar to most investors. The first question asked of Palandjian in almost all the meetings: “Can you tell me again how this works?” The second question was often, “How do you measure it?”

For Bank of America, devoting hundreds of hours to a tiny deal only made sense as a way to dip a toe in the water of impact investing, an emerging must-have practice area for all major wealth managers. BofA offered the social impact bond specifically to its private banking clients with more than $10 million in investable assets.

“Our clients want it,” concludes Surya Kolluri, Bank of America Merrill Lynch’s managing director of policy and market planning. The 2014 U.S. Trust Wealth and Worth Survey found that half of high-net-worth investors consider environmental, social and governance (or ESG) issues to be an important part of investment decision-making. Kolluri says that over time, social impact bonds could become a key element of the “S” in ESG. That provides plenty of room for growth: BofA’s ESG investment platform represents approximately $8 billion in client assets.

“Was the investor investing because it was comparable to other private equity investments or because it was a better way to do social impact?” Kolluri says. “I would conclude that it was because it’s a better way to do social impact.”

The key selling point, he says, was “velocity,” the fact that the social impact bond could repay investors, who could then recirculate their money into additional social, or other, investments. “Velocity is an aha moment,” says Kolluri. “It’s very different than a grant in which the money is gone.”

Simply having that conversation helped BofA strengthen its relationships with clients. Financial advisors are eager for anything that enables them to differentiate themselves from the competition and get closer to clients and their families

“It’s not about share of wallet, it’s about share of mindset,” says Jackie VanderBrug, the U.S. Trust executive responsible for developing the firm’s sustainable investing strategy. “It’s going to be a very small percentage of their portfolio. But it’s going to be a big percentage of what they talk about around the Thanksgiving table with their grandkids.”

As investor interest grows, the pay-for-success model has the potential to scale up much more dramatically than either government spending or traditional charity. Already, more than $50 million in private capital in the U.S. has been mobilized through pay-for-success contracts targeting early childhood education in Utah and Chicago, as well as recidivism in Massachusetts and New York City, in addition to the New York state contract.

Since the first deal closed, New York has announced four finalists to its request for proposals for partners on additional pay-for-success initiatives in early childhood and child welfare, health care, and juvenile justice. CEO, Social Finance, and another California partner are pursuing an additional pay-for-success project in San Diego. “Any place where you can invest a dollar of prevention today to save more dollars downstream is an appropriate allocation,” Palandjian says.

More broadly, if the New York state deal signals a wave of private investment in social impact bonds, it could usher in something like a new social contract, aligning private capital and the common good. In an earlier era, proven approaches, often developed by nonprofits, could be taken to scale by government agencies that would implement them more broadly. With public budgets under severe constraint, private funding needs to fill the gap. Once the savings are proven with private investment at risk, government can incorporate the solutions into normal budget processes.

“In the global financial crisis, taxpayer funds bailed out some large financial institutions,” Palandjian says. Social impact bonds flip that paradigm on its head. “Here, risks are privatized and gains are socialized. That’s a new model, one harnessing private capital to serve the public good.”

[seperator style=”style1″]Disclosure[/seperator]

ImpactAlpha writer Jessica Pothering contributed reporting to this article.

This article was produced with support from Ascend at the Aspen Institute, and is published in the report, “The Bottom Line: Investing for Impact on Economic Mobility in the U.S.”