Pay For Success

Lessons Learned: Improving Stateside PFS Transactions Through Findings from India’s first DIB for Education

By Alexis Herschkowitsch, Innovation Fellow

In June, the Center for Universal Education at Brookings hosted a webinar to discuss year-two results of the world’s first Development Impact Bond (DIB) for education. For those unfamiliar, Development Impact Bonds work similarly to Social Impact Bonds; the key difference is that with a DIB the outcomes payor is generally a donor, as opposed to the government (as is the case with SIB’s). Webinar presenters represented the broad spectrum of perspectives of the various agents in a DIB.

This particular DIB is in Rajasthan, India, and its goal is to improve educational outcomes for girls. The service provider both aims to increase enrollment among girls in the area, and increase test scores. Instiglio has a helpful primer here:

Image from Instiglio

Now that the DIB has results from its second year, there are many learnings to extract and apply to our work on PFS transactions here in the US. Here at ICS we are excited about the findings from this particular DIB, given its focus on measuring educational outcomes. What follows are some key takeaways

There were continued big increases in children’s learning From Year 1->Year 2. Progress towards the goal nearly doubled in the DIB’s second year, indicating that there was still plenty of growth occurring in Year 2 of the program. While it’s a natural assumption that we might expect to see gains and improvement immediately, in some cases there might be a steeper learning curve. It’s a reminder to (a) build in a small buffer time to allow the project to truly take hold, and (b) monitor outcomes closely, but not become overly alarmed if goals at the first evaluation point are not met. Instead, this checkpoint should serve as a time to analyze and re-strategize.

Well after starting project, the team realized that baseline number of girls not in school had actually been lower than as originally assessed; this required edits to the contract and a subsequent decrease in amount paid per student. This imparts a key learning for our PFS work in the US: build flexibility into contracts. Such a discovery could have rendered the DIB void, but instead, owing to the openness of all parties to revisit and restructure the contract, the program was able to continue, keeping the focus on improving education outcomes for girls in India.

The outcomes payor in this case found it difficult to “take a back seat” and not get more involved in execution of intervention. This is a completely understandable position on the part of the outcomes payor, who has a large financial interest in the outcome of the DIB. It is a concern that needs to be addressed ahead of time so that the outcomes payor will be 100% comfortable with the arrangement and give service providers space and independence to complete their work.

The architects of this DIB found it easier to find investors than to find outcomes payors. This is an important lesson to apply to PFS feasibility studies in the US, as it’s clear that these two processes should not necessarily run in parallel. Seeking out possible outcomes payors sooner rather than later is a smart strategy to ensure eventual feasibility. We at ICS have also seen this trend in several PFS projects, though are heartened that this seems to be changing as the field ages and innovates. It is still important in the U.S. to leave time for adequate education of both potential outcomes payors and investors.

Another applicable finding is that the learning curve was steep for all participants—but it was particularly steep for service providers. Anecdotally there have been similar experiences here in the US, and this speaks to the need to develop tools to make the transition smoother for service providers. In some cases, the amount of data collection might be at levels service providers are unaccustomed to, so whatever steps can be done to ease this will help in executing PFS transactions. This is a key piece of ICS’s feasibility study model. Our colleagues at the Nonprofit Finance Fund have also developed tools to help providers this through their readiness.

The high evaluation costs relative to the size of this DIB presented a challenge. With a total investment of $267,000, this DIB is significantly smaller than any PFS transaction thus far in the US—where investments are in the millions of dollars. Since evaluation costs were substantial in this case, the ratio of evaluation costs to potential payout was quite low. A key takeaway for efforts in the US is to closely evaluate this relationship in the feasibility process. If the evaluation costs exceed a certain threshold, perhaps it would be prudent to explore other investment vehicles. Stakeholders in a project may be surprised by the cost of evaluation and can benefit from experts in this area explaining their decisions behind a specific methodology.

Many parties agreed that transaction costs need to be lowered to make DIB’s viable in the long run. This is something we hear time and time again with regard to PFS, and it is something the field must continue to research and work out to arrive at a viable solution.


Leveraging these learnings from India in our PFS efforts here in the US—particularly those in education—can help us sharpen the saw to make these tools more effective for the most in-need communities. Each incremental improvement that we can make to improve efficiency in PFS transactions helps us improve outcomes for the lives of children in the US.