“It is better to be vaguely right than precisely wrong”, so said the famous economist, John Maynard Keynes. Is this the right attitude to take when measuring the non-marketed social benefits from social services programs? Or should all possible effort be expended to precisely quantify the benefits and costs of a social services program? I would argue that it isn’t possible to precisely quantify social benefits because of the difficulty of collecting data. However, that doesn’t mean that we should accept that a social benefit is material without some transparency. And this is where the Social Returns on Investment (SROI) methodology can help improve the rigour of investment decision-making in social programs.
SROI has some superficial resemblance to benefit-cost analysis (BCA). Like BCA, the net benefits of the program (the social return on investment) can be expressed as a ratio such as ‘$x social benefits to $1 of costs’. However, there are some substantive differences with the methodology. First, is the focus on stakeholders’ views in collecting data. Stakeholders can include beneficiaries, investors or program staff. Stakeholders are the centre of SROI analysis. They help define how a program’s activities generate social change and therefore impact. They are the key source of data used to calculate the social value generated by a program. This differs from the BCA because it is possible to undertake one without speaking to a single stakeholder.
Second, non-marketed and therefore unquantified social benefits are more important than quantified economic and financial data. This is obvious, since the aim of SROI is to estimate the value of social benefits created from a program. It would be pretty pointless only including quantified economic and financial data if there was no relationship to the social benefits being created. This is where it is important to include the stakeholders in understanding the magnitude of the impact. For example, expressing values in terms of items that the stakeholder use can be used to implicitly quantify social benefit. An example, is where stakeholders valued increased happiness as equivalent to going on an overseas holiday (around $5000). This is an example of choice modelling. Other non-market valuation technique such as revealed preferences and stated preferences can also be used. Unlike BCA, the SROI must include financial valuation of social benefits. In many BCA reports, social benefits and costs are often relegated to the ‘qualitative discussion’ section of the report and are effectively omitted from the analysis. SROI forces the analyst to explicitly includes social benefits and justify the valuation.
Third, SROI can comfortably incorporate subjective data. Indeed, the valuation from stakeholders may be completely subjective and based on their personal frames of references and experiences. As mentioned before, SROI places stakeholders in an almost sacrosanct position of the analysis. The SROI analyst has a responsibility to eliminate their own and the stakeholders’ bias but this may not be always possible. In the end, if the objective is to value the impact on stakeholders, wouldn’t the stakeholders’ perspective be the most important? Arguably, BCA is also not free of subjectivity. Choice of assumptions, time period and variables can lead to wildly variable results on the same program.
Finally, SROI analysis explicitly accounts for deadweight, attribution, displacement and drop off. Deadweight is whether or not the value would have been created anyway. Attribution (or contribution) is how much can be directly attributed to the program in question. Displacement is whether the program didn’t actually change behaviours but instead forced the targeted behaviours to move to another location. And drop-off is how quickly the benefits dissipate. This places the onus on SROI analysts to justify that the social program is the key catalyst for the social value creation. This also has the useful function of helping identify potential collaborators (in terms of attribution) or potential design flows. Deadweight, attribution, displacement and drop-off are expressed as percentages which are used to weight the social benefit valuation. Thus, it is possible for social value to be reduced to zero if it is found that the social value would have happened anyway, or there are many other duplicate programs, or if the targeted behaviour moved or if there is rapid drop-off in benefits. In BCAs, there isn’t a convention to include such explicit weightings. Instead, these effects are taken account within the modelling of the benefits which may make it difficult to understand the assumptions used.
The great strength of SROI is that it requires rigorous analysis of how social value is created from a specific program. Compared to BCAs, SROIs place less emphasis on technical modelling but places higher priority on stakeholders’ perspectives for calculating social value. The engagement of stakeholders also strengthens the analysis in my view because it gains the insights from the ‘customers’ and stakeholders of the service. After all, the customer is always right, right?