Evolution through learning - reflecting on the Social Outcomes Conference 2019

Nigel Ball is the Executive Director of the GO Lab. After the Social Outcomes Conference 2019, he shares his thoughts about where we are up to with impact bonds. He looks at the lessons we have learned in the last decade and the ones we are still learning. We'd love to hear your feedback, email us on golab@bsg.ox.ac.uk 

Towards the end of our international Outcomes Conference, I asked our delegates to stand up and stretch their legs. It had been a deeply thoughtful couple of days. I then told everyone I would only let them sit down again if they could confidently say the conference had given them a clear sense of exactly what they needed to do differently on Monday morning. Unsurprisingly, no-one sat down. Does this mean that thinking too deeply about hard questions paralyses us into inaction? Actually, it acknowledges the unfortunate reality that there are no perfect answers to hard questions. How should we fund efforts to improve social outcomes? What is the role of governments versus the private and voluntary sectors? If something has “worked” in one place at one time, will it work again in a different place and time? And how do we work out whether our efforts have been effective? The plurality of perspectives on offer at the conference – from academics, policy makers and practitioners from all over the world – should surprise no-one.

The peculiarities of one particular financial instrument brought many of these discussions into sharp focus. Ten years ago, a new funding model in a forgotten town in the UK midlands gained unprecedented attention and sparked ten years of hype-fuelled efforts to repeat it. The era of the “social impact bond” or "SIB" was upon us. A decade on, many speakers at the conference agreed that the idealised version of this model has not survived its contact with reality. This has certainly led a few of the most committed sceptics to gleefully chant “we-told-you-all-along”. However, the impact bond model has shown remarkable adaptability, and in the words of one delegate, “morphed” into something quite different from that idealised original. That process itself is instructive. My prediction is that impact bonds are not ready to disappear quite yet – and that is something that everyone – even the sceptics – ought to welcome. Here’s why.

Firstly, the things we call “impact bonds” are so different from each other that they can’t easily be understood as a defined model. Instead, we might be better comparing how each project makes lesser or greater use of a set of independent mechanisms. The GO Lab’s 2018 report Building the Tools presents four possible dimensions of this variability. Understanding which of these mechanisms is useful, to what degree it should best be employed, and in which contexts, could be of immense practical use to policymakers.

Secondly, the benefits these mechanisms unlock are often not those that were billed when the impact bond model was first dreamt up, which chiefly revolved around more preventative spend and a transfer of financial risk to the private sector. Those coming back to the impact bond conversation after some time may have been surprised to hear themes of collaboration and relationship-based working, flexibility at the frontline, and the creation of learning and feedback loops. Almost everyone agreed that these were positive and useful practices, and they were clearly present in the many impact bond case studies we heard – Educate Girls in India, Manchester Rough Sleeping, Sheffield Futures, West London Zone and more. 

Of course, it would be a mistake to argue that an impact bond is the only route to unlock these types of helpful practices. Several alternative approaches were discussed at the conference. One is to bring services back in-house and attempt to run them through a traditional centralised, pyramid-shaped hierarchy. I suspect most reformers would consider this to be a nostalgic idea – though in the UK, it is the official policy position of the Labour Party opposition.

Another alternative approach is almost the exact inverse of this: push power downwards to front-line workers and citizens themselves, and trust the very people who are usually put at the bottom of the traditional hierarchy to work things out for themselves. While some people might deem this idea nostalgic too, it does seem to mesh with current social circumstances. Several speakers pointed out that countries across the world are facing a crisis of legitimacy in government and public institutions, and the greater involvement of communities (defined through shared places, beliefs or interests) might help to address it. Interestingly, some of the most active practitioners of this approach in the UK are Labour Party-run local governments. This is despite the idea’s most recent appearance in the national political discourse being in the mouth of a Conservative Prime Minister: David Cameron’s Big Society.

No doubt these ideas deserve more exploration – both intellectually and practically. They will fit more neatly in some national contexts than others. But if nothing else, the transformation of impact bonds into multiple different forms that diverge from the idealised original should serve as a warning against pinning our hopes on snake oil. There is no single approach that is going to address the manifest challenges of delivering excellent services to the public and addressing complex social issues. We know that shiny new ideas never survive first contact with reality. But that doesn’t make them useless: it starts a cycle of learning and evolution. There may be no perfect answers – but there is still a lot to try.