Other people's business

RSS

Goldman Sachs gets into social impact bonds - but what are they?

Ryan Shorthouse of the Social Market Foundation explains the reasoning behind social impact bonds.

A guard at the entrance of Rikers Island in 1955. Photograph: Getty Images

Fiscal retrenchment is catalysing radical thinking among policymakers about how to get better public services for less money. Social impact bonds (SIBs) are currently fashionable in policy debates as a possible means of financing interventions. With SIBs, social investors fund a particular service, and only get a return if the intervention improves outcomes which will lead to reduced government expenditure in the long-term. In the current environment, Government wants to pay investors only out of identifiable savings. And an idea that started here in the UK has now gone global. Just last week Goldman Sachs announced that it was spending $9.6 million on a 4-year programme aimed at reducing recidivism of offenders at Rikers Island prison in New York.

SIBs are potentially an ingenious way of getting more bang for taxpayer bucks at a time when public money is short. They are a vehicle for encouraging innovation in public service delivery because they devolve the financial risk to investors and organisations who can affect outcomes on the ground. At a recent SMF conference, Iain Duncan Smith MP said:

It could mean a change to the whole way that Government and private sector work together to solve social problems.

The first ever SIB launched in 2010 and funds work to reduce re-offending among offenders released from Peterborough Prison. Philanthropic investors will receive a return on their investment if the interventions funded achieve at least a 10 per cent reduction in reoffending each year, or at least 7.5 per cent across all three years. Other schemes are now emerging: in Manchester, for example, the Council is sourcing funds from social investors to provide intensive support for eight young people with challenging circumstances to live in foster care rather than in residential care.

SIBs are an important part of the funding jigsaw. But they are not the magic bullet for all public services. Social investment – where investors invest in the work of charities and expect a return – is still small: in 2010, £190m was sourced for social investment compared to £3.6bn in philanthropic grant funding and £55.3bn in wider bank lending. And SIBs only constitute a small part of all social investment. The small scale is mainly down to a lack of decent investable propositions. There are at least three big reasons for this.

First, because SIBs are embryonic market information about the likely risk and reward in different service areas is poor. Investors are jumping into the unknown. Little is known about how effective new interventions could be at, say, cutting re-offending levels, so investors don’t have much to go on in assessing the investment proposition. This uncertainty is exacerbated by the length of time it may take for outcomes to be observed, especially for early intervention programmes. The Government has helped set up Big Society Capital which it hopes will co-invest with private investors to send a signal to them and mitigate their risks by accepting lower interest rates or taking on the junior part of a debt. It is also hoped that Big Society Capital will fund new products that support impact measurement.

Second, there is a risk that investors are not paid appropriately. In most public services it is difficult for government to identify whether outcomes have improved, let alone to attribute those improvements to the work of the provider. If re-conviction rates fall after an intervention how can government distinguish between its being the result of the intervention or perhaps a change in the local policing strategy? An up-tick in re-offending could be the consequence of high local unemployment, or a statistical blip, rather than ineffective interventions. Correctly attributing outcomes to their cause is notoriously difficult. But without resolving that challenge both government and investors will remain reluctant to embark on large scale SIBs.

Third, even where outcomes are measurable, quantifying the financial benefits for taxpayers is tough. Improved employment outcomes for unemployed people or better GCSE results for children in care may be good in themselves, but quantifying the public savings is no simple task. All the more so if those savings are spread across a number of government departments, making coordination difficult.

The potential for SIBs and other payment by results schemes to revolutionise public service delivery lies in the incentives they create for providers to innovate. But there are many hurdles for government to overcome if this approach is to enter the mainstream. Improving measurement and data collection, working across departmental silos, and simply taking a punt on financially risky ventures to find out what works may all be necessary steps. In time SIBs could save government money. But the first steps on the road will be costly. And right now that’s not something that government wants to hear.