This piece was written by James Ronicle, an Associate Director at Ecorys UK and a former Fellow of Practice at Oxford University’s Government Outcomes Lab.
An impact bond has the potential to employ scarce resources more effectively by increasing the focus on outcomes, encouraging adaptation when things are not working and ensuring governments or donors only pay when things are a success.
But they are not a panacea, and they are not appropriate in all circumstances. There are many cases where impact bonds have been explored but not pursued.
For example, by December 2016 the National Lottery Community Fund, a grant giving organisation in the UK, had provided 62 organisations with grants to test the feasibility of launching a social impact bond (SIB); over half of these (37) decided not to launch one. These organisations concluded that the necessary conditions weren’t in place to launch a SIB.
• Want to write for us? Take a look at Apolitical’s guide for contributors
So what are the necessary conditions? Ecorys, a research and evaluation organisation, has undertaken multiple evaluations into impact bonds in high, middle and low-income countries, looking at over 40 impact bonds that have launched and 20 that have not.
Based on this research, it is our view that any organisation considering developing an impact bond needs to ask themselves five key questions:
- Is there a clear rationale for pursuing an impact bond?
- Is there sufficient support for an impact bond within all the organisations involved?
- Are there clear and measurable outcomes?
- Is there sufficient data to build a solid business case?
- Is there a strong service provider market?
We cover each of these questions in further detail below.
1. Is there a clear rationale for pursuing an impact bond?
It is essential an organisation starts with the problem first, then considers the possible solutions to this problem and how they could be commissioned, of which an impact bond could be one option.
Many impact bonds fail because organisations want to do an impact bond and go in search for an area to apply it. Research by the Government Outcomes Lab (GO Lab) in the UK suggests that impact bonds are most appropriate when you face the following problems:
- Fragmented delivery and “siloed” budgets: Impact bonds can be good at breaking down silos between different funders and service providers by creating a shared focus on outcomes.
- Inability to prevent issues from escalating: Outcomes payers can struggle to fund preventative services because all of their funding is tied up in responding to the problem. By only paying if and after the outcomes are achieved (and if those outcomes lead to reduced demand for more acute services) the outcomes funder can use the savings from the reduced demand for the acute services to fund for the preventative ones — known as “double running”.
- Limited current progress and a need for innovation: Impact bonds can support innovation by:
- Enabling more service providers to compete in Payment by Results (PbR) / Results Based Finance (RBF) contracts by transferring some financial risk to external investors, widening the range of interventions available
- Encouraging innovation in delivery by enabling service providers to focus on outcomes and not outputs
- Ensuring projects adapt when things are not working through increased performance management from external investors.
2. Is there sufficient support for an impact bond within all the organisations involved?
Strong leadership is necessary because impact bonds are novel for most stakeholders; they can be complex to develop, and require establishing new relationships and trust between individuals and organisations that often share little prior collective experience. Some stakeholders can be concerned about the risks associated with an impact bond, and so strong leadership is necessary to drive the process and bring other stakeholders on board.
3. Are there clear and measurable outcomes?
Defining a cohort of target beneficiaries, and choosing outcomes that are clear and attributable to the intervention, is an important part of the development process of any PbR or RBF contract, including impact bonds. Ultimately, the organisation paying for outcomes needs to be satisfied that the outcomes are worth paying for, and the service providers (and investors) need to believe that the outcomes are achievable.
This can be difficult when:
- It is not possible to calculate the added value of the SIB — i.e. the outcomes that would be achieved over and above those that would have happened without the intervention
- It is not possible to identify outcomes that can be defined clearly and measured objectively
- It is hard to attribute the outcomes to the intervention. One advisor on impact bonds reflected that they work best in more “closed” systems — i.e. when there is a very clear link between the intervention and the desired outcomes
- It is not possible to find outcomes that would be achieved in a relatively short time period; generally if it is going to take more than five years between the intervention and the outcomes occurring an impact bond starts to become unattractive for outcomes payers (because they then span electoral cycles) and/or investors (as it can be too long for them to wait for returns).
4. Is there sufficient data to build a solid business case?
A substantial amount of data is required to develop an impact bond. This is primarily because an impact bond contract requires an estimate of what outcomes are likely to be achieved, on what scale and at what point. This is so the business case can be developed, and the outcome payer can calculate what the potential outcomes are worth to them, and the investor can calculate the levels of risk and rates of return.
In order to build the impact bond business case, it is necessary to have data on:
- The eligible cohort, including the size of the cohort, the outcomes they are currently achieving and the current cost of supporting them
- The cost of delivering the new intervention
- The outcomes likely to be achieved by the intervention, including when these would take place and for how long they are likely to be sustained.
5. Is there a strong service provider market?
Impact bonds require service providers to have a credible track record, a strong performance management infrastructure and be willing to adapt when things are not going to plan.
We would strongly encourage any organisation considering an impact bond to ask themselves these five questions. If the answer to all five questions is “yes” then, based on past experience, there is a strong likelihood that the impact bond is feasible. But if the answer to any of these is “no”, then this needs to be addressed before your progress. — James Ronicle
(Picture credit: Rawpixel)