Social Impact Financing’s First Flop

Way back in 2012, in one of our first Youth Services Insider columns, we discussed the potential “Dawn of the American Social Impact Bond” as interest in the concept spread from across the pond.

The way social impact bond and pay-for-success projects work is simple at the core: private money fuels a public provision of services instead of taxpayer money. If certain agreed-upon benchmarks are achieved, the taxpayers repay investors with interest. If they are not achieved, the financial loss is on the investors.

Up first for America: Adolescent Behavioral Learning Experience (ABLE), a project aimed at lowering the recidivism rates of juveniles returning home from Rikers Island, the central jail system for adults and teenagers in New York City. The primary investor, Goldman Sachs, would be rewarded with city funds if ABLE yielded a recidivism rate 10 percent better than that of a control group.

Earlier this month, the final judgment on that venture was sealed with an evaluation by the Vera Institute of Justice. The program did “not lead to a reduction in recidivism for program participants,” Vera said in its impact evaluation.

A few quick thoughts on the demise of ABLE:

Project Failure = Government Success?

This project was always, in our mind, a very tepid first step into the world of social impact bonds. Most of the “risk” being taken by financier Goldman Sachs was being protected by a philanthropic entity started by the sitting mayor of New York City at the time, Michael Bloomberg.

So as far as risk goes — and make no mistake, the risk element is the defining characteristic of social impact financing — this was pretty much a test case, a practice run.

You could say that the test worked in at least one sense: the program did not work, and New York City taxpayers will not be any worse for finding that out. The fiscal loss is all on Goldman Sachs and Bloomberg Philanthropies, and the young men in the program did not fare worse than the control group.

Bad Setting

So the city government broke even on the deal, but what was learned? The backbone of ABLE was an intervention called moral reconation therapy (MRT), which has already shown promising results on the aggregate. So hopefully the lesson isn’t to ignore MRT as an option.

But it always seemed sort of nuts to YSI that this project focused on juvenile offenders at Rikers Island, an adult jail facility. The reason juveniles are even there in abundance is because New York and North Carolina are the only two states where the default is to treat both 16- and 17-year-olds as adults in the eyes of the law.

Through recent and excellent reporting by Trey Bundy, Daffodil Altan and others, we now have a pretty clear picture of how poorly a place like Rikers is equipped to manage a juvenile population. But it’s not like everyone just found out that the massive inmate island was challenged to offer anything resembling a juvenile incarceration setting.

It is long past time for New York City to get teens the hell out of Rikers Island. But as long as Rikers houses juveniles, the city really owes them an immense effort to seek meaningful rehabilitation and development.

It just doesn’t seem like a logical place to test innovation. In a screwy way, doing so sort of validated the setting by positing the question, “Can this proven intervention work while we house teens in this adult jail?”

The project intermediary, MDRC, saw it differently. “This program attempted to change part of the culture inside Rikers Island by introducing an intervention for a very high-needs population for whom little to no programming was previously offered,” the company said in a statement following the release of Vera’s findings.

Fair enough. But what does it say that the attempt failed with an intervention that has been proven to work?

Wrong Partners, Crazy Timeline

YSI heard from one inside observer of New York City juvenile justice that, even with a setting like Rikers, the right service providers could have made a difference with the resources brought to bear through this social impact bond.

Seedco, the lead provider on the project and a favorite of former Mayor Bloomberg, had lost a slew of staff during an investigation into its practices with another major initiative. “They were in total crisis and did not have leadership or bandwidth for this,” the source told us.

If poor execution drove failure here, it raises an important question for social impact leaders to think about. If a lead provider is failing, what can be done to salvage a social impact bond or pay-for-success project?

It seems to YSI that very little can be done to salvage a three-year venture like ABLE. You figure any provider has a year to prove itself; if it doesn’t after a year, you are then banking on a new entity coming in to right the ship while also getting up to speed.

Massachusetts joined New York City as an early initiator on social impact projects, launching a juvenile recidivism reduction venture in 2013. That project has a seven-year timetable. Everyone we have spoken with has high praise for the main provider, Roca, Inc. But if the opposite were true, there might be time to switch up.

Print Friendly, PDF & Email

John Kelly, Editor in Chief, The Chronicle of Social Change
About John Kelly, Editor in Chief, The Chronicle of Social Change 1213 Articles
John Kelly is editor-in-chief of The Chronicle of Social Change. Reach him at