Reflecting on the Social Outcomes Conference 2020

We went virtual for the fifth edition of the international Social Outcomes Conference. While much is naturally lost in the absence of in-person interaction, there are some upsides. We were pleased to welcome an even more international audience than usual to the conference, and we enjoyed seeing participant’s live reactions via the Zoom “chat” function while sessions were underway. There was much to digest for those involved in thinking about, establishing and managing cross-sector partnerships for social outcomes.  

A new approach to contracts?  

Contracts matter a lot in partnerships for social outcomes. The formality of a contract often offers important protections for both parties that a grant does not. A conventional contract, at its most basic, sets out what a buyer expects to get for their money. Sometimes, though, both buyer and seller might understand the end goal, but face a lot of uncertainty about the means (and exact cost) of achieving it. Certainly, that is often the reality of efforts to improve social outcomes, and it makes it hard to set out in the contract exactly what will be delivered. Much of the conference discussed outcome-based contracts (including social impact bonds or SIBs), which address this uncertainty by specifying measurable end goals and paying against those. But our opening session introduced something new: relational contracts. These address the same issue by setting out how the parties should interact and negotiate with one another as the contract goes along, particularly when things get challenging or the unexpected happens. Our keynote speaker, Professor David Van Slyke, is a world-renowned expert in the topic. He suggested relational contracts might combine the trust and informality that typifies a good relationship with the protection offered by a typical formal contract – increasing the chance of a “win” outcome for both sides. 

The relational contracting approach has its pedigree in the private sector and has been little used in the public sector – and barely at all, it seems, in relation to social issues. Might relational contracts do a better job than conventional contracts in acknowledging the complexity of social problems, and the many actors involved in solving them? Might the explicit consideration of governance and shared decision-making complement the performance focus that often typifies an outcomes contract? Unlike typical outcome-based contracts, where realistic, payable targets need to be agreed upfront, a relational contract needn’t rely on an accurate prediction of the future to set the terms of exchange. That means there is more room for uncertainties that might arise because the parties are doing something quite new and innovative, or because the delivery environment rapidly changes and evolves (as we have seen, at a particularly extreme level, over the past six months). 

Some might argue that grants can perform the same flexible function as relational contracts. But grants suffer their own limitations. They are not necessarily any less prone to reinforcing the power imbalances in a transaction. They often lack the checks and balances that govern a contract – leaving them open to abuse in circumstances of low trust or weak scrutiny. And they don’t necessarily do a very good job of aligning the goals of their parties, or of assuring quality. 

Relational contracting is certainly not a panacea. Such contracts might lack legal enforceability. Their reliance on trust means they take time (and cost money) to set up and manage. And the reputational incentives the parties rely on to keep each other in order may not be strong enough to prevent opportunistic behaviour – an even more acute problem, perhaps, in countries with weak governance, as is the case in much of the developing world.  

However, relational contracting surely deserves experimentation in the world of social outcomes. I hope governments promote this, just as many have done with outcome based contracting. 

Data as an instrument of power 

Another recurring theme throughout the conference was how social outcomes are measured, and whether such measurement might support or undermine both the quality of services and the voices of citizens. Many people crave the top-down clarity that might be achieved if measurement of outcomes could be standardised and data could be collected and shared publicly. Such insights seem to provide clear utility to the investment industry, academia, and of course governments and multilateral agencies. However, many expressed concerns that such a “data revolution” in the world of social outcomes could further consolidate the position of these powerful actors. 

The issue is not so much in the old cliché that real people get reduced to a number on a spreadsheet or a point on a graph. Rather, it reflects a legitimate anxiety that such information might create the illusion of control, and therefore be used to support top-down, chain-of-command approaches that deny agency to those doing the messy on-the-ground work of making people’s lives better. 

Such statistical abstraction can also shut out the voice of service users themselves. But there are promising ways to mitigate this risk. Besides the moderating effect of interpreting data alongside related qualitative information, we heard of several creative approaches that position service users as data users too. For example, citizens might be asked to sense-check the insights gleaned from data. They could be asked to make decisions based on what the data says – particularly when it directly affects them. 

We hope to be able to dive into the issue of power more deeply at next year’s conference, and would welcome suggestions on who we might invite. 

Another day in the life of social impact bonds 

The other thing that was clear from the conference is that social impact bonds (SIBs) continue to energise discussions on every continent of the world. As they do, it becomes ever harder to identify a common approach. We heard many contrasting, and often competing, narratives – not just regarding what a SIB is, but also why the mechanism might be useful to address a wide array of weaknesses in social service systems. SIBs are being touted as a response to diminished government funds, to a need for fresh innovation in service delivery, to a perceived allergy to risk in the public sector, or sometimes as little more than some stardust to get a new project going. Perhaps a SIB can address all these things and more – but the answers seem to be context-dependent to the extent that the term “SIB” may “obscure more than it illuminates”, as one of our session chairs, Traverse’s Chih Hoong Sin, eloquently put it.  

It seems to me that a new language that can tease apart the different use-cases for SIBs is urgently needed, to allow for a more meaningful categorisation of like projects. 

These are just a few reflections from a very rich four days of discussion that 1000 words could never do justice to. I look forward to hearing your own reflections, and to continuing to engage with you all on these questions and more over the coming 12 months.