The theory behind social impact bonds is pretty straight forward: Investors provide upfront money for development/ social programs, based on an informed assumption/ risk that these programs will achieve certain outcomes/ milestones. If these outcomes are met, an underwriter (traditional donor or government) pays back the investor with a profit.
This model is a win-win: Donors only pay for successful projects, private investors profit from development programs, and eventually, most development money goes to organizations that actually achieve an impact. This is very different from the way things work today in traditional development.
However, when this model gets put in practice, there are two huge problems:
1. It is hard to objectively determine a causality between certain activities (that benefited from the funding in question) and impact – which means that investors need to do a leap of faith that their investment will lead to impact; and
2. Even when a causality is determined, there is a significant time gap between investment and the determination of causality, which means that once invested, investors are not in control of their investment;
Both of these barriers are inherent in the way development funding works. Here is a handy chart:
Basically, an investment is made. Implemented activities are decided and rolled out. Then a bunch of years pass. Eventually, there is an evaluation, towards the end of the project (3–5 years after the investment). The evaluation has several distinct phases, spread over another year or so. Eventually, years after the original investment, the evaluation results are published. These are usually complex, stuffy academic works.
Even if the investor can draw clear causality lines between investment/ activities and specific results, at this point there is not much they can do about it either way. This doesn’t make any sense to a typical investor, used to track the performance of their investment daily and continuously re-asses whether they will keep a specific investment or sell it.
There is another way.
Make it possible for people to earn reward every time they complete an impact behaviour – vaccinate a child, attend a consultation, refill a prescription etc. These rewards are triggered in real time, through a simple validation logic (i.e. we verify that someone actually completed that behaviour).
This allows us to define highly precise behaviours and track them in real time. For example, we can track every vaccination that happens in a certain district. Or every consultation. We can even track behavior by audience – say, every girl attending school in a certain community. This means we can also price every one of these custom behaviours very precisely. (We call them “RET” – short for Rafiki Earning Tiko, where Rafiki is our archetypal client and Tiko are the rewards they earn through impact behaviour).
Basically, every time a desired behaviour occurs, there is a blip on the investor’s screen. Which means that the investor can really manage their investments – not happy with progress? Take their money out. Are they happy? Put more money in.
Now for the link between the verified behaviour and the impact: we only focus on behaviours where the link is presumed. There is a presumed impact to a child actually receiving a vaccination or going to school or someone refilling their prescription regularly. The only thing that needs to be verified is the accuracy of our validation – that can be done through regular audits – fast and a lot cheaper than a population-level evaluation report. In fact we include such audit in our regular operation.
Once the behaviour itself is verified, it can be converted into any impact indicator required.
This brings an unprecedented level of rigor in impact investment. Indeed, we hope it will also eliminate the prevalent emotional heuristics behind funding decisions currently and that it will usher in a new era of data-driven investment.