There was a time when social impact bonds (SIBs) were all the rage, the shiny new policy wonk instrument. This instrument is a contract where payment is hard-wired into specified outcomes being achieved. Independent investors put up the working and risk capital and only get paid back if it all works, it was said these instruments would help us all deliver better services, would encourage innovation, guide government and philanthropists to a better way of commissioning, and would mean we only paid for the things we wanted.
But then they sort of faded away from the front-line of interesting ideas: I’ve not seen a think tank or politician talk about them in a long time, even though some keep going in the background. So, has the social sector been wise to put them on a bit of a back burner or have we made a mistake here?
Rates of return
I have followed the career of SIBs quite closely, trying to identify their strengths, how they have developed over the years and their weaknesses. But now a new attempt to rate them has come out from Big Society Capital—drawing heavily on a report written for them by consultants (ATQ) who appear to mainly work for people doing social impact bonds. In truth it is less purely about social impact bonds than their close cousin ‘social outcome contracts’ (SOCs). In other words, contracts where certain outcomes are specified (rather than activities or outputs) and payments depend to at least some degree on achieving those outcomes. With SIBs, investors put up the money in the first place—helping with working capital and taking some of the performance risk away from the provider—with the government (or other payer) paying up when outcomes are delivered.
The report highlights some of the (very big) claimed rates of return on these SOCs, in areas like homelessness, children’s services, and health, claiming that the 72 projects they looked at—of the 90 projects overall in the UK—created value of over £1.4bn, with over £10 achieved for every £1 invested. And while you may wonder why, in that case, the government has not put almost all its money into such things, they go on to say that this is in fact a conservative estimate as future benefits are not counted. Even if one is a sceptic about such very high numbers (and as an ex-Treasury hand, I certainly am) it at least makes one want to think seriously about whether SIBs and SOCs really are the answer.
The report from Big Society Capital
We know that outcome based contracts can, in theory at least, be important in allowing providers to have flexibility on how to achieve the outcomes and can make a group of organisations get together as needed to help the life of a family or individual—as they have to provide the outcomes to get the payments; and that the intensity for needing to hit the outcome means there is usually rigorous data collection relative to what is often collected by charities in real time, not least so that mid-term corrections can be made if things are not going well. SIBs tend to also create this desire to have lots of data in real time as the investors want to get their money back if they can. (See Social Impact Bonds: The Role of Private Capital in Outcome-Based Commissioning for more discussion of the benefits).
The Big Society Capital report is less strong on all the downsides of outcomes-based contracts, the perverse incentives and the tendency to go for the low hanging fruit (cream skimming), the high administrative costs and the burden on the providers. The report also does not analyse whether some of the same sort of activities (e.g. the collection of real time data) could be achieved in other ways, for instance the joining up of services through something like the Total Place approach which seeks to bring public service funding streams in a geographical area together. Or making real time data and direction-correction mid-term more normalised (as they are in agile ways of working).
The report lauds various government outcome funds—like the Life Chances Fund, the Youth Engagement Fund and the Social Outcomes Fund—and many have been very welcome in allowing new models to be tried but this is without consideration around whether the money might have created more impact if used in other ways, including in the non-profit sector. It is also telling that some sort of ‘subsidy’ or philanthropy or funder, willing to put up some funding that doesn’t require a return usually (i.e., not just joined up funds), is very often needed to make these things happen.
The report does show to some extent that such contracts seem to be most attractive in certain fields. The original SIB was in criminal justice, where in Peterborough payments were received only if the re-offending rate for a cohort of ex-prisoners fell. There has been some success in rough sleeping, and bits and pieces elsewhere.
A useful tool in the tool kit?
Some who have studied them feel that the real problem is to do with the way that contracts with outcome payments set up a conflict between commissioners and suppliers, and that what we really need is to build networks and trust and use ‘relational contracts’, where ‘rather than dictating terms, government would invest time up front to agree on a set of outcomes with providers, and negotiate how they would be measured and paid for’. (See Spotlighting Shared Outcomes for Social Impact Programs That Work for more on this argument). Others argue that in the end, you would do better to simply use grants for purpose driven organisations.
So, we continue to learn from SIBs and outcome-based contracts. So far, they have yet to really taken off despite the pleas of their fans. Are they the future or the last dregs of the New Public Management and the marketisation of everything? Once upon a time I wrote about how private finance initiative should be seen as a useful tool in the tool kit, but used with extreme care and only when a lot of conditions applied (see my chapter in Private-Public Partnerships: Policy and Experience). This feels the same for me with SIBs. Only time will tell.