How can we more easily replicate and scale social impact bonds (SIBs) and similar contracts? This was the core question that newly published research by Ecorys and ATQ set out to answer. In this blog, the main authors of the report, Rachel Wooldridge, Senior Research Manager at Ecorys and Neil Stanworth, Director at ATQ summarise the report’s key findings.
The enablers and barriers for commissioners
Central to understanding how to replicate or scale SIBs, writes Rachel Wooldridge, is identifying the enablers and barriers that those who would lead their continued development – i.e. commissioners – face in developing them. Our study began from this premise, and explored firstly what the challenges and benefits of the SIB commissioning process are for commissioners, before considering the wider implications for further developing the market. We based this part of the research on detailed discussion with a sample of commissioners who had been involved in the development of SIBs, supplemented by a review of existing evidence from previous research.
Overwhelmingly we found that the greatest challenge commissioners faced was engaging with the various stakeholders needed to get the SIB developed, both internal to their organisation (key decision-makers, finance and legal staff, and service leads) and external (investors and providers). Sometimes these challenges were insurmountable: for example we found that lack of understanding about SIBs or misunderstandings about terminology prompted decision-makers to veto the development of the SIB. But other times, the process of overcoming these challenges, through embedding a collaborative approach from the outset, benefitted commissioning organisations through fostering new and strengthened relationships.
Our report also found commissioners have struggled with the more technical elements of developing the SIB business case, particularly the financial modelling, accessing data and navigating tools and guidance. Some commissioners have been able to engage advisors (using central programme development funding) to overcome these challenges, but the process of doing so led to up-skilled commissioning teams, that were more confident about considering outcomes-based approaches in the future.
Wider structural and cultural factors, like inflexible commissioning structures or organisational risk-aversion, also make outcomes-based commissioning difficult for commissioners. However, plenty of procurement guidance is available and commissioners told us it is just a case of knowing about, and preparing for, these processes early on in the commissioning process to ensure success. Investing time and resource into a high quality and flexible procurement process helps to establish a better contract.
Once we understood fully what commissioners had struggled with, and valued, about developing SIBs we then cast the net somewhat wider, as Neil explains further below.
The key routes to replication and scale
Having more clearly established the barriers faced by commissioners, writes Neil Stanworth, we moved on to the wider question of whether and how we can replicate existing contracts – or implement similar contracts at greater scale. In my view this question has been around almost as long as we have had impact bonds, and certainly since we have had several previous examples to consider.
Our research into this question drew on the research with commissioners mentioned above, but we cast the net somewhat wider and beyond commissioners themselves, to capture the views of stakeholders with wider experience of multiple projects, such as investors, intermediaries and other advisors. We spoke directly to many key stakeholders, and also issued a call for evidence inviting anyone with an interest in this question to let us have their views.
A key finding was that there are broadly two ways to replicate contracts, which might loosely be termed ‘bottom up’ and ‘top down’. Bottom up replication happens most obviously when a commissioner copies what another commissioner has done, implementing a SIB in the same policy area with a similar (but rarely identical) approach and intervention; it also happens when the same commissioner implements further contracts in different or adjacent policy areas, either at the same time (by implementing what some are calling a ‘multi-SIB’) or later. Top down happens when a specific type of contract is funded centrally (such as the Entrenched Rough Sleeping SIBs funded by the Department for Communities and Local Government); or when those delivering a contract on a specific model (usually a service provider or intermediary) engage other commissioners with a view to them adopting the same model. Examples of the latter include HCT Travel Training and the Mental Health and Employment Partnership, both of which started with one commissioner or one contract and now have several.
What enables and what inhibits replication?
Whatever approach is adopted, our research found that there is scope to replicate or scale some, most or nearly all the features of a previous contract; our report gives examples of such features (everything from the intervention to the contract governance structure) and highlights where and to what extent we think they can be replicated. Stakeholders told us that that there is much we can do to make this process easier and quicker, by making more information publicly available on previous contracts, and deploying tools such as ‘plug and play’ financial models and template business cases for contracts in key policy areas. Indeed the Government Outcomes Lab is already responding to some of the report’s key recommendations around this. We believe that these kinds of tools will help successor commissioners get to grips with technical issues more quickly, avoid reinventing the wheel (or worse, the flat tyre) and reduce their need to rely on external advice and support, the cost of which itself inhibits growth.
But it is equally important to note the limits to replication – notably because the experts we spoke to all pointed out that any model, however well designed and ready-made, will need adaptation to local circumstances to reflect such issues as differences in the characteristics of the target group at which the programme is aimed. Most importantly of all, any contract depends on a range of local stakeholders – from senior leaders to technical experts and front-line staff – being willing and able to engage positively in what can be a challenging process. It is an obvious point, but one worth repeating, that just because a solution comes ‘out of the box’ does not mean that everyone will want to implement it, for a wide range of reasons. Indeed we heard examples of models that had been enthusiastically and quickly adopted by some commissioners, and flatly and even more rapidly rejected by others.
Which is why there is considerable overlap between the barriers to commissioning in general, as outlined by Rachel above, and the challenges of replication. Both depend crucially on positive stakeholder engagement and therefore on the narrative around outcomes contracts being easy to understand, and dispelling unwarranted suspicion and mistrust. Some commissioners will have very good reasons not to copy or adopt what others have done, but it helps no one if those reasons are based on misconceptions or poor understanding. For this reason, we have made a number of recommendations on how the narrative around SIBs, and the language used to describe this type of contract, might be changed. This is not easy to do, but we believe that it will make the type of contract that we call a SIB easier to understand, and potentially reduce some of the barriers to their acceptance.
I would also tentatively suggest that the ultimate goal of impact bond contracts becoming ‘mainstream’ will be best served in the long term by the ‘bottom up’ approaches outlined above. Our research found that ‘top down’ approaches can make things happen more quickly, but they may not change attitudes to outcomes-based contracting as profoundly as multiple commissioners deciding for themselves that, in the right circumstances, this is the way to go.
