The impact bond market expanded in 2018, according to analysis by the Brookings Institution. As of Jan. 1, the institution’s global impact bond database tracked 134 contracts, 24 more than a year ago.

The vast majority of contracts — 127 — are social impact bonds, where the outcome funder is the government. Third-party funders pay for seven development bonds.

The social welfare sector with 47 contracts and the employment sector with 45 contracts account for 69% of impact bonds. Health, education, criminal justice, and environment and agriculture together account for the remaining 31%.

Brookings reported that 24 new impact bonds were contracted in 2018, down from 34 in 2017. Nine of the new contracts will focus on social welfare, six on employment, five on health and four on education.

Although most of the new contracts were in high-income countries, four were in low- and middle-income countries. Cameroon rolled out a development impact bond for cataracts, and India launched its third DIB, with three service providers offering different interventions to improve learning outcomes. In addition, South Africa contracted one social impact bond for youth employment and another for early childhood development.

Analyzing Outcomes

Although the majority of impact bonds are still in the implementation phase, an increasing number of the early deals are ending, with more than a quarter of the 134 contracts completed, according to the Brookings analysis. These provide evidence on the successes, challenges and lessons learned.

The impact bond market is growing steadily and has spread across many regions, but remains small at about $370 million invested over eight years. By comparison, impact assets under management last year amounted to $228 billion.

Brookings’ data on the completed deals shows that most have repaid investors their principal plus positive returns. Only the NYC Able Project for Incarcerated Youth made no outcome payments.

Most of the completed deals so far have been in the U.K., where a number of process and impact evaluations have been published.

The first education impact bond in a low- or middle-income country, Educate Girls DIB, launched in 2015, wound down in 2018. An evaluation by IDinsight found that the service provider enrolled 768 out-of-school girls and that achieved learning outcomes were 160% of the final target.

According to Emily Gustafsson-Wright and Izzy Boggild-Jones, the authors of the analysis, isolating the “impact bond effect” — that is, whether using the impact bond financing mechanism actually adds value — continues to be difficult.

They write: “Even for impact bonds where rigorous evaluations were used to measure outcomes, only the effectiveness of the intervention itself was captured, so we don’t know if the same results could have been achieved with input-based financing, traditional payment by results, or even just providing cash with no strings attached.”

They note that a key observation from their analysis is the role of data, and the need for service providers to be able to gather, analyze and respond to information to achieve outcomes. “If impact bonds are to increase in scale … the demand for timely, usable data, and the systems that support their use, will continue to grow.”