Social impact bonds: an agnostic stance

Social impact bonds elicit strong opinions – from great enthusiasm about what they could achieve to passionate opposition to their complexity. Mara Airoldi cuts through the noise to take a hard look at the evidence we have about their effectiveness to date. 

The debate around social impact bonds (SIBs) ranges from extraordinary expectations to passionate opposition. Such a polarised debate can risk us failing to learn useful lessons about what has and hasn’t worked in the 45 SIBs that have been launched in the UK so far (with over 100 worldwide). At the Government Outcomes (GO) Lab, a centre of academic research and practice based at the Blavatnik School of Government at the University of Oxford, we take an agnostic stance.

Our first major study on SIBs, published today, moves beyond ideological debate to provide a constructive response based on a robust analysis of the available evidence. Later today, we will host the UK premiere of The Invisible Heart, the first feature-length documentary film about social impact bonds, after which there will be a panel discussion about the future of social services in the 21st century.

We know that the first ever SIB in Peterborough reduced reoffending rates in prisons, but was this because of the SIB?

The first question we asked in our research was not “do SIBs work”, but “what are SIBs for?”

SIBs have enjoyed much enthusiasm from social investors, successive governments, and certain charities and social enterprises. Despite this, uptake has not been as high as expected. It seems that it has not always been clear to government departments and local authorities what a SIB offers that is more than conventional ways of contracting services.


Overcoming public sector challenges

In order to understand whether SIBs may be beneficial, we looked at the justifications and how they may help overcome perennial challenges in government: fragmented and siloed agencies and budgets; a short-term political and financial focus; and risk-aversion and difficulty creating change. SIBs seem to have the potential to:

  • Overcome fragmentation through collaboration: multiple commissioners and service providers work together so that services can ‘wrap around’ citizens and respond to their needs more holistically.
  • Intervene earlier to prevent crisis: many SIBs have a preventative focus and enhance individual life chances before issues become entrenched. This supports citizens before crises occur as well as saving money in the longer term.  
  • Overcome difficulty creating change through innovation: SIBs can create room for commissioners to try new interventions as financial risk is transferred to the investor. There may also be enhanced performance management and systematic learning.


Four key dimensions of a SIB

We then asked, what is it about a SIB approach that might unlock collaboration, prevention and innovation?

What we found here will come as no surprise to those familiar with the space: there was a huge amount of variation between the way SIBs worked. We boiled these differences down to four dimensions.

The first was the nature and amount of payment by results – was payment made only on the end outcomes, or were some conventional service payments or payments for short-term outputs included? 

The second was the nature of working capital or ‘investment’ – was it tied to the achievement of outcomes, or did have to be repaid no matter what? 

The third was the level of social intent of the provider organisation – was the delivery being done by a charity/social enterprise, or a company without explicit social values? 

And the final was the performance management approach – how hands-on were project stakeholders in managing the performance of the project throughout delivery? 

We don’t really know yet how these different factors interact in different circumstances to produce different results. Developing this understanding is a core objective of the GO Lab.


Do SIBs work?

That brought us to the question we are most often asked: do SIBs work? 

Unfortunately, this is hard to answer based on evidence available at the moment.The current evidence-base around UK SIBs focuses on whether the intervention achieved results, but few studies look at the impact of the SIB commissioning approach itself. For example, we know that the first ever SIB in Peterborough reduced reoffending rates in prisons, but was this because of the SIB? Could they have used another commissioning approach that produced the same or better outcomes? 

We don’t yet understand how to identify the most appropriate commissioning tool or the ways in which the commissioning arrangement interacts with service delivery. The major aim of the GO Lab is to develop this evidence-base through further evaluations in the coming years.

Looking at existing evidence, it seems that SIBs might alleviate perennial public service challenges like fragmentation, reactive spending and difficulty innovating, but we’re not seizing the opportunity to learn from where they work well and where they don’t. The GO Lab seeks to create an environment where policy makers, practitioners and researchers are open and frank about SIB failures as well as successes, and evaluations are published promptly. Being transparent about weaknesses is not a sign of failure, it is an opportunity to get better. The real failure would be not to learn.

Building the Tools for Public Services to Secure Better Outcomes: Collaboration, Prevention, Innovation, the Go Lab's Evidence Report on UK social impact bonds is available here.

The Go Lab is hosting a webinar, Is there any Magic Dust in Social Impact Bonds?, to discuss the report with the authors on 18 September 2018. To register your interest, see here.

Header image: A still from The Invisible Heart showing children at Chicago's Melody Elementary School whose education is supported by a social impact bond funded by Goldman Sachs and billionaire venture capitalist JB Pritzker. The funders will make a profit if the children succeed in their tests.