The world’s first social impact bond (SIB) has reduced recidivism among offenders and repaid private investors in full, with a return of 3% interest per annum. Over five years, One Service embedded itself in the Peterborough, UK community and coordinated service providers to help male short-sentence offenders before and after their release with housing, employment, mental health support and financial advice. They worked with 2,000 prisoners and reduced recidivism by 9%, exceeding the target of 7.5%. Since then, 89 SIBs in 19 countries have mobilised over $300 million.
Results & Impact
An independent evaluation found that recidivism of 2,000 short-sentence offenders was reduced by 9% compared to a national control group. This exceeded the target of 7.5% set by the Ministry of Justice. National reoffending rates for such prisoners were at around 60% when the program was designed. Investors were repaid in full: their initial investment plus a return representing 3% interest per annum. Since then, social impact bonds have spread across the world.
Ministry of Justice, Big Lottery Fund, Social Finance, One Service, Sodexo Justice Services, St Giles Trust, Ormiston Families, Sova, MIND, TTG Training, MegaNexus, YMCA, John Laing Training
Social Finance liaised with government, designed the program and solicited investment. One Service embedded itself in the community and coordinated the many service providers. They worked with adult male short-sentence offenders, contacting them before release to assess their needs, plan their resettlement and introduce case workers and community volunteers. Some of these case workers were ex-offenders who had received training to act as points of contact. From their release date, One Service worked with the ex-offenders for up to 12 months, helping with needs such as housing, medical issues and benefits, and ultimately moving them towards employment or training. Prison leavers weren’t forced to use the service, but the great majority did.
Cost & Value
$6.6 million was committed, but as the project was terminated early not all was drawn. It costs around $45,000 a year to keep someone in prison in the UK.
Completed in 2015
Data collection and consistency was a challenge. One Service had to collect data from the courts, the prison, the police, service providers and the local authorities in a timely fashion. As a result, the data it received was often incompatible, inconsistent or inaccurate. Another challenge was the complexity of working with so many service providers. Even after five years, this was still a challenge, although some service providers adapted to the task and worked together better than others.
A change of direction in national policy meant that the program was curtailed at five rather than seven years. Informally, elements of it are maintained by the local community.
A SIB is a type of payment-by-results contract where a charity or another organisation with a social purpose agrees to deliver a given outcome. The funding typically comes from third-party investors, and if the project hits the target, the investors are paid back by the government—with interest. But if it fails, it is investors, not the government, who lose out.
Over five years, the Peterborough SIB worked with 2,000 adult male offenders with sentences of less than a year. It reduced recidivism by 9%, exceeding the target of 7.5% set by the Ministry of Justice. When the program was designed, national reoffending rates for such prisoners were at around 60%.
“I’ve got 25 years experience in the social sector, and you always bump into the same people just going round and round the system,” said Janette Powell, SIB Director at Social Finance. “When I came across this social impact bond, I immediately thought it might be a solution to some of the social issues we were looking at on the ground.”
“Social Finance came up with the idea in the first place, had the conversations with government to put numbers on the problem and then helped attract investment,” continued Powell. “The government were very much hands-off in terms of service delivery. They agreed to the cost, settled on the most important outcome [reoffending], and then they took a step back. I used to work in the Ministry of Justice and that was quite a bold move.”
The program started in 2010, with around $6.6 million raised from 17 investors to launch the SIB. This funded One Service, an umbrella organisation that coordinated many service providers to respond to the needs of offenders.
Many of the clients had reoffended before, had housing needs, lacked the right skills to find employment, had no access to income, and were in debt to moneylenders – altogether, a sure-fire formula for recidivism.
One Service reached out to them while they were in prison to assess their needs, design their resettlement activities and introduce them to their case workers.
These case workers are part of a peer support model used by the provider St Giles Trust. Some are ex-offenders themselves who have received professional training to become points of contact between each offender and the providers looking to help them.
Offenders weren’t forced to use the service, but participation levels were high: 74% for the first cohort of 1,000, and 87% for the second.
The One Service was delivered by many providers, including St Giles Trust, Ormiston Families, Sova, MIND, TTG Training, YMCA and John Laing Training. It also relied on the support of the prison, operate by Sodexo.
Each provider did something different. MIND provided mental health support. St Giles Trust trained the case workers. Sova managed a large team of community volunteers. TTG Training set up a building with a training centre and accommodation, and had links with local construction firms to get the men work.
“What went really well was having a long-term goal and being able to embed ourselves in the area,” said Powell. “We were completely embedded with the police, probation, local authorities, housing associations, local businesses, Jobcentre Plus—to the extent that we even had a job centre person working in our office.”
There were some persistent challenges, though.
One Service made extensive efforts to engage service users. However, long-term engagement remained a problem. Most users disengaged from the One Service well before the expiry of their 12 months of available support.
“Data was another big challenge,” added Powell. “One thing I asked when I started the job was: are we going to get access to all the data we need? The court data, the prison data, the police data, the local authority data. Accessing the right data, in a timely fashion, and making sure it was accurate and consistent—this continued to be a problem from start to finish.”
“Another issue was around working with multiple service providers,” said Powell. “This is always a challenge, but you would’ve hoped that over the course of a five-year project with a single goal, the competition between the providers would’ve diminished—but it never did. Working across all these providers, and with external agencies, remained a challenge.”
The program was initially intended to span three cohorts over seven years, but in the end it was curtailed after just two cohorts and five years due to a new government program that targeted all short-term prisoners, thus removing the ability to judge the SIB against a national control group.
With more time, the results might have been better still.
“We were on an upward curve,” said Powell. “We only got to do two cohorts; had we got to do the third, I think the results would have been even better. I think we really would’ve cracked it.”
(Image credit: Flickr/Alexandar C. Kafka, RAND)