News and Comment

Social impact bonds in Japan – it’s not always about saving money

Thursday 16 June 2016


In the previous blog in this series, I shared some key messages from a recent speech I delivered at the 2016 Social Investing and Corporate Social Responsibility Forum, held at Meiji University in Tokyo. In this blog, I summarise some of the thinking shared with Japanese colleagues on a second topic they had asked me to touch on:

Can you structure outcome metrics in a way that is not motivated directly by budgetary savings but by social wellbeing?

Level of outcomes

We can look at outcomes at the individual level, the group level and the system level. At the individual level, outcomes are obviously about achieving and improving individual wellbeing. At the group level, we may aim for improving social wellbeing. At the system level, there can be policy or political priorities that focus on social wellbeing. There can also be economic benefits that are being aimed for, and also budgetary ‘cashable’ savings.

Meeting the priorities for outcomes at one level may not always lead to (or support) achievement of priorities at another level. Outcomes for individuals, for instance, may not always lead to savings for the system. An intervention aimed at reducing the social isolation of a socially excluded group may uncover previously unmet needs, resulting in these people being put in touch with other services. While this is a good thing, it does mean there is a direct service use cost, at least in the immediate short term, that was not previously there.

Wellbeing, rather than savings, as a driver

If we are to take an approach that prioritises the achievement of individual/social wellbeing as opposed to the achievement of budgetary savings, there are different ways for structuring outcome metrics in support of this goal.

(a) Taking a wider definition of outcomes and accounting for their value: This recognises that there are different perspectives of the value of the same outcome. For example, when we talk about the ‘cost of crime’, members of the public do not really think of the cost of policing, the cost of incarceration, etc. Instead, they are likely to think about the economic, social and emotional harm inflicted on the victim and communities experiencing crime. A wellbeing perspective will therefore need to take these into account, over and above the system costs.

(b) Acknowledging public ‘willingness to pay’: The public can value things even when the financial implications of those things may not be clear. This ‘willingness to pay’ is underpinned by what we as a society think of as ‘right’ and ‘of value’. It can be influenced by cultural, ethical or moral norms. For example, there can be a strong moral case for supporting war veterans.

Public ‘willingness to pay’ is subjective, contextual, and can change. For example, prior to the UK Government announcing its Comprehensive Spending Review (CSR) at the end of 2015, there was widespread expectation that the government would cut policing budgets significantly. In fact, the government had hinted this will happen. The CSR took everyone by surprise when it was announced that policing budgets would be protected. Why had this happened? In a nutshell, the global terrorism threat had become increasingly significant. Public expectations around policing and security have gone up the political agenda, and it would have been untenable politically to cut policing budgets in such a context.

How may this look like in terms of structuring outcomes

First, there can be situations where the primary outcome metric that triggers payment is savings related, but there are also a range of other wellbeing outcomes being monitored and tracked. For example, in the Essex County Council SIB, the primary outcome metric that triggers payment is ‘care days avoided’. There are a range of secondary outcomes, such as strength of family functioning, emotional wellbeing, educational outcomes, etc. Data on all of these are analysed and discussed by the Programme Board on a regular basis. The seriousness to which the various players consider outcomes holistically is demonstrated by the fact that the social investors have committed to re-contacting the families who have received Multi-Systemic Therapy some time after the intervention has ended to check whether a range of outcomes are sustained over and above the duration for formal outcomes monitoring required by the SIB.

Second, there can be outcome payers who are prepared to pay for outcomes even when they are not linked directly to savings. These outcomes may be paid for as stand-alone outcomes. One of the outcome metrics that triggers payment in the Lisbon Junior Code Academy SIB, for example, is ‘logical thinking’. Logical thinking, obviously, is not something that leads directly to budgetary savings, but is a life skill that will serve an individual well throughout his and her life and can operate in very different situations.

Third, rather than to pay for stand-alone wellbeing outcomes, these may be blended in with other types of outcomes that may be more savings-related. The New South Wales, Australia, Benevolent Society Social Benefit Bond shows how this may be done. Payment is triggered by the Performance Percentage. Performance Percentage is a weighted average of three separate measures. This provides a model for how wellbeing outcomes can be woven in with others to create a composite outcome that triggers payment.


As many of the initial SIBs are underpinned by a savings logic, it is easy to think that SIBs can only be structured with this objective in mind. The outcomes focus of SIBs should challenge us to be clear about ‘outcomes for whom?’ Shifting the gaze away from system-level outcomes onto outcomes for others requires us to think of different ways to identify and structure outcome metrics to support the achievement of individual and social wellbeing.