Development impact bonds have generated a great deal of buzz in recent months, but until now, there has been little evidence of if and how well they work. Today, one of the first DIBs is releasing data from its first year that points to signs of success.
The Educate Girls bond is a results-based financing mechanism, through which the UBS Optimus Foundation provides upfront funding for a program angling to get more girls in school and improve overall educational attainment in 140 communities in Rajasthan, India. If the project succeeds in reaching 18,000 children, then the Children’s Investment Fund Foundation — as the outcome payor — will repay UBS’ investment, plus a return depending on performance.
When the Educate Girls bond was first discussed more than two years ago, those involved knew they would be trailblazing, trying to learn about and define a new financing instrument for the industry. The pressure was on, and the journey has not been easy. Designing a new financial instrument meant a host of negotiations to come up with the right contract that measured the right goals and set the right targets. Even after implementation challenges with a government agreement and condensed timeline made things difficult.
Yet after its first year, those involved are pleased with the progress towards the overall goals. Educate Girls has already enrolled 44 percent of all out of school girls in the region and while they’ve only met 23 percent of total targets for learning progress thus far, the organization is confident that now that some of the startup challenges are behind them progress should speed up in the next two years. The progress to date means that UBS will have already recouped about 40 percent of the initial $267,000 investment it made a year ago.
In the beginning
How does a development impact bond work?
Investors put up the money to fund a development project, a service provider or implementer carries out an intervention to tackle a project, and then an outcome payer or group of payers (in this case development finance institutions, donor agencies or foundations) pays for the success of the intervention. Investors are only paid if the intervention implemented by the service provider succeeds based on a set of predetermined metrics.
The Educate Girls bond traces back to 2012 when Safeena Husain, the executive director of the India-based NGO Educate Girls reached out to Avnish Gungadurdoss, co-founder and managing director at Instiglio, a company that specializes in providing technical assistance for the design and implementation of results-based financing. At the time, the U.K. Department for International Development had launched a Girls Education Challenge that Husain wanted to apply for; she needed help with the proposal.
Instiglio helped write the proposal, but after several rounds of the challenge, DfID decided it wouldn’t fund projects in India. Husain was left with a results-based financing proposal she was eager to test but short of funds to implement.
She spent about a year of pitching to investors, including undertaking a roadshow across India as well as reaching out to individuals, foundations and corporates. More than a few people asked Husain why she didn’t give up.
“We’d done nine months of work so we were married to the idea of wanting to test this,” she said. “The reason for being a headbanger is very different than financial. It’s about impact at the end of the day.”
Things began to fall into place after Husain spoke at a conference organized by the UBS Optimus Foundation. The foundation’s CEO Phyllis Costanza said was impressed with Husain and the organization.
So when Instiglio called to talk about the idea of funding Educate Girls through a DIB, Costanza was intrigued.
Costanza didn’t have to think too long about the DIB to decide it was the right opportunity at the right time for UBS Optimus, she said.
Costanza, who had worked at the Children’s Investment Fund Foundation, a U.K.-based philanthropic organization working to help poor and vulnerable children, asked them if they would be interested in being the outcome payor.
CIFF had been following some of the research about social and development impact bonds and thought it was an exciting idea. It was less clear how a DIB would work in practice, and what it meant to have a nongovernment outcome payor, said Robin Horn, the director of policy and advocacy of CIFF. The foundation’s board went back and forth about the decision debating whether the DIB was viable. They ultimately said yes.
The negotiation and design
Once the partners were in place, the hard work began. There was no template for this type of impact bond.
The Educate Girls DIB
Where: Mandalgarh and Jahajpur blocks in the Bhilwara district of Rajasthan, India
How: Programs in 166 government schools in 140 villages
What: Reach approximately 18,000 children — getting more girls in school and improving lerning outcomes
Payment: Maximum $422,000 based on outcomes, 15 percent maximum internal rate of return, paid at the end of three years
Although social impact bonds have been growing in popularity in the U.K. and U.S. for the last six years, development impact bonds were, and still are, new. The Educate Girls DIB is one of the first. Among the specific challenges were negotiating a riskier, developing country environment and designing a bond with a foundation, rather than a government, as an outcome payor. Those involved felt particular pressure knowing that they were among the first to try this sort of intervention.
As the intermediary, Instiglio helped design the bond, determining which outcomes to measure, how they should be priced and measured and how to structure the needs of each partner so they would agree to the terms. The group — Educate Girls, CIFF, UBS and Instiglio met regularly to negotiate the details of the deal, at times running into disagreements.
One of the key sticking points was over how to evaluate and measure progress, with the outcome payor and investor wanting an intensely rigorous evaluation and Educate Girls wanting to use a more scalable form of measurement.
“I thought it might not happen at that point,” Gungadurdoss said. “We had to find a way to bridge the gap.”
In the end all parties agreed to do a randomized control trial in an effort to use one of the most rigorous tools available given the scrutiny the bond would be under.
Dozens of other technical details also had to be agreed upon, for example how to devise a payback system that would differentiate between harder to reach girls and those who were easier to get into school. An advisory group made up of a range of experts, which was convened early on to consult on the DIB, helped provide input to navigate those decisions.
As discussions played out, lawyers from CIFF and UBS wanted to provide input about the potential legal implications of any deal. What they came up with was a contract for a $267,000 investment from UBS with a goal of reaching 18,000 children, with CIFF paying out a roughly 10 percent internal rate of return if they meet those targets, and up to a 15 percent IRR if they surpass them. And Educate Girls is also incentivized to perform well: They will receive an incentive payment from UBS Optimus of 32 percent of the payment the investor makes after recouping its principal investment.
By the times contracts were completed last year, it was a mere three months before the school term started, so the project implementation needed to move fast.
A preparation process that typically takes Educate Girls nine months had to be done in a third of the time. They quickly rented an office, recruited, hired and trained 200 people, and got government permission. Signing a memorandum of understanding with the government — the programs are all implemented in government schools — held up the process.
Before starting the program, Educate Girls did a censuslike door to door survey to create a list of out-of-school girls in 34,000 households. The data would be used as a baseline for targets related to increasing the number of girls in school. The independent evaluator, ID Insight then tested that data for accuracy.
The next step was to evaluate the educational level of students in the schools where they would be working. A team of about 160 community volunteers was deployed to both get more girls in school and supplement lessons in the equal number of government schools where the DIB is being implemented.
These schools are often little more than a mat on the floor and a blackboard with chalk. Most don’t have electricity or desks and chairs. But through the program, about three times a week, an educated person from the local village will come in to teach English, Hindi, and Math through songs, activities and games that help them gain critical reading and writing skills.
About halfway through the learning activities, ID Insight, the third-party evaluator, did a quick test of how kids were performing and identified points where they were getting stuck.
Through that testing it became clear that English performance gains had plateaued and were not on track. Educate Girls went to the community volunteers to find out why. They learned that many of the teachers were not confident in their own English skills and sometimes struggled with the lesson plans. Educate Girls offered additional training and modified lesson plans so the teachers could teach with confidence.
“This was the first time that we were using data not just to aggregate at each level but actually sharing all data,” Husein said. “Sharing data and information with front-line staff helped us understand what are [the] barriers.”
Despite some of the growing pains and first year challenges, stakeholders told Devex they are happy with the first year’s progress.
The results go beyond achieving the specific measurable goals; they may also have a lasting impact on how Educate Girls works more broadly. DIBs like this create a “culture of performance in a sector that badly needs it,” Gungadurdoss said.
While the original plan was to keep the DIB separate from the rest of the organization, it hasn’t quite turned out that way. Other Educate Girls employees were interested in the DIB and wanted to absorb some of the lessons. District managers across the organization now want a performance management system, for example.
“It’s bleeding into the rest of the organization in organic ways,” Husain said. “The DIB actually forces us to really codify everything much stronger and easier for field staff.”
While the first year’s achievements are on track, she said she expects the second and third year performance to be even better.
And in retrospect, the program could have done some things better from the beginning. DIB should have been bigger, maybe a million dollars, because that scale would have brought the overhead cost ratio lower, Costanza said. For future DIBs UBS Optimus Foundation will certainly look to work on bigger bonds.
The process was incredibly complex and streamlining it in a way that didn’t engage all actors at every step could have sped things up. Originating at the top and not with the service provider also may be better, Husain said.
While they chose an RCT because they wanted to use the most rigorous form of measurement, they all agreed that it may not have been necessary. A cheaper methodology could have given UBS the confidence to invest, Costanza said. And a similar-sized test could have been used for a much larger project, Horn added.
DIBs shift the relationship with donors, which has to be acknowledged, and eventually will be a more hands off process than grant giving. They also push both donors and service providers to move away from a focus on activities, some of which are expensive and may only produce a marginal benefit.
The group is keen to continue working together to inform the next wave of development impact bonds. For Educate Girls true success would be to see this DIB graduate to being a social impact bond, with the government as the outcome payor.
For the broader development community curious about the potential impacts of DIBs, this small test case may only offer limited answers. But it may pave the way for other DIBs — there are several in the final planning stages — that will help provide further answers to some of the critical questions about how to develop DIBs, when they should be used and how results-based financing may change development.
For more on where development impact bonds stand today, look out for an article later this week.
Join Devex to network with peers, discover talent and forge new partnerships in international development — it’s free. Then sign up for the Devex Impact newsletter to receive cutting-edge news and analysis at the intersection of business and development.