Designing a Social Impact Bond in the Tropics

Over the past year, Instiglio has been designing some of the first social impact bond (SIB) projects to be implemented in developing countries. Although this work is preliminary and ongoing, we would like to share five lessons that have come out of this process so far.

The SIB model was piloted by Social Finance UK starting in 2010. The New York City SIB may now be the second best-known project because Goldman Sachs has agreed to make a $9.6-million investment in it. Currently, the national governments of Canada, Ireland, Scotland, and Israel, as well as numerous local governments, are exploring the possibility of implementing social impact bond projects.

Skeptics may argue that increased risk related to investing in emerging economies makes developing country SIBs unattractive. Yet we have observed significant interest for implementing SIBs in emerging economies on behalf of governments, service providers, development agencies, foundations, and yes, potential investors.

Our exploratory efforts have included conversations across North and South America, Africa, and India. As a result of these conversations, the government of the state of Antioquia in Colombia has decided to partner with us to explore and design an SIB. Antioquia may therefore soon be the first public entity in Latin America to have a social impact bond.

Through the process of “selling” the idea of social impact bonds in developing countries, some lessons and conceptual questions have begun to emerge. We hope these reflections may be useful for others designing social finance ventures in developing countries.

1. The ups and downs of legal innovation

One of the most significant obstacles in getting various stakeholders on board the social impact bond bus is uncertainty about how the model can be implemented in compliance with local laws. Government officials especially are wary of signing on to projects that may lead to judicial investigations later on. Their concerns are justified. Public servants are not used to signing contracts that pay out of future years’ budgets, much less contracts that may not pay out at all.

Yet every jurisdiction has different regulations that make SIBs possible, even if some require alterations to the original model. Partnering with top legal advisory firms to streamline the legal design process has proved to be a crucial component of the early stages of designing a SIB.

In the federalized United States, laws differ by state, yet there are sufficient similarities between state laws to believe that upfront investment in designing the legal structure for the first SIB will pay off in the design of later SIBs in different states. In the same way, Latin American countries have a similar civil law tradition that makes investment in the first country beneficial for later implementation in others. The process of legal innovation, therefore, does not need to start from scratch, but can cross borders to some extent. The extent to which it is able to cross may drive SIB adoption in the region.

2. Savings are unlikely to significantly offset the cost of the program

The SIBs that have been implemented in the UK and the US have been lauded as mechanisms that may reduce government budgets by refocusing investment on preventative services. If SIBs designed to lower recidivism, such as the one being implemented Peterborough (UK), are successful, the state would save by having to process and care for a smaller prison population. If SIBs designed to reduce chronic homelessness, such as ones being created in Massachusetts and London, achieve their goals, the government should save resources from reduced demand of social safety net programs that benefit the homeless. There are two reasons, though, that explain why SIBs may not lead to the same cost savings in developing countries:

First, developing countries in general have smaller social safety nets than developed countries. As a result, the costs associated with vulnerable populations are absorbed by the population as a whole and, in particular, by vulnerable populations, rather than by the state. SIBs may therefore play a particularly important role in developing countries by improving welfare, but the service they provide is less likely to lead to future cost-savings for the government.

We can observe this by looking at the homelessness example. Colombia offers fewer services to the homeless than Massachusetts, yet the impact of homelessness is just as damaging to the individuals involved and to the population as a whole in both places. While the Massachusetts SIB provides the government with an opportunity to spend funds now to save expenditures on social services later, the situation is slightly different in the developing country context. For the Colombian government, for example, spending money now to prevent homelessness implies future social savings for the population, but not for the government.

Second, the success of SIBs depends on the implementation of robust program evaluations. While social programs tend to be cheaper in developing than in developed countries, the opposite is true for robust evaluations. These can be much more expensive in developing countries due to the lack of administrative data. As a result, the implementation of a SIB is relatively more expensive, and therefore not as attractive from a public budget perspective, as it could be in a developed country.

A concrete example may come from the initiatives in Peterborough (UK) and New York City aimed at reducing criminal recidivism. While evaluators in the US and the UK can look at local and national databases to identify whether treated and control individuals have ended up back in prison, evaluators in a developing country are more likely to have to periodically follow the individuals to gather the necessary data for impact evaluation.

This does not mean that SIBs in developing countries are unattractive, but rather that interested governments must have greater vision if they are to sign on to an SIB project. Interested policymakers must understand and value impact evaluation as part of their long-term plans for improving the population’s welfare.

3. Significant opportunity for development agencies

Due to the high upfront costs related to the design and implementation of SIBs, development agencies can play an important role in getting the ball rolling. If they can come in as early funders of legal analysis and feasibility studies, they may be key players in allowing governments to overcome some of the early hurdles in the process.

In the longer term, development agencies may play an important role as joint financial sponsors of SIBs. Once again, development agencies may be able to help governments overcome hurdles they may encounter in this stage due to lack of resources. Joint financial sponsorship is a common arrangement, and applying it to SIBs could drive their expansion.

Development agencies also have an instrumental role in the SIB variant known as the Development Impact Bond. In this arrangement, development agencies, rather than governments, pay service providers or intermediaries only if the latter achieve predetermined outcomes. The Center for Global Development has created a working group to explore that arrangement.

4. Opportunity to shake up service providers

To implement a successful SIB, a service provider must be identified that has its house in order and an innovative approach to solving a social program that can be scaled up. Finding such a provider can be a challenge in the developing world for multiple reasons. Service providers in developing countries are used to operating with limited discretion, as the government and international development agencies alike prefer to hire them through rigid contracts that allow little room for program model deviation. They are also often less institutionalized than similar organizations in developed countries, and have less access to resources, due to a limited culture of philanthropy and a less developed social sector.

What this means for the implementation of SIBs in developing countries is that the financial intermediary has the opportunity to take on a more active role in the implementation process. The SIB would constitute a larger portion of the service provider’s revenue and operation, giving the intermediary greater leverage over the project to be implemented. Furthermore, the limited access to funding means that SIB implementation may provide a unique opportunity for a service provider to scale an innovative project. While this implies a more hands-on and time-intensive approach, it also means that the intermediary has a greater opportunity to have a positive social impact role.

5. The necessity of multitasking

Once on the ground, Instiglio has had to work simultaneously on many fronts, including legal, procurement, financial design, and stakeholder engagement. One challenge potential intermediaries will face is to get government officials on board before selecting a service provider. On the other hand, it can be hard to engage a service provider without a concrete government entity backing the project. We are using both approaches on different projects.

Our continued work and the work of other organizations bringing this results-based financing innovation to international development will soon yield deeper and broader insights than the ones we outline here. We hope that this article, in a small way, advances the ongoing discussion around the theory and practice of the social impact bond model.

Read more stories by Sebastian Chaskel & Michael Belinsky.