For those trying to make budgets stretch further to fund international relief programmes, it might sound like a match made in heaven: private investors could fund projects that help people forced to flee conflicts or living in the midst of health or other emergencies, and those investors would be rewarded based on how well those projects were carried out.
The scheme might expand the donor pool (think Silicon Valley), and should appeal to a new generation of investors looking to make money and do good, and in the longer-term it should increase accountability and efficiency in the aid sector.
Impact bonds – complex loans whose cost is pegged to achieving social goals – now fund a range of causes from rehabilitating former convicts to saving rhinos. And yet, despite high hopes in recent years, this payment-by-results funding tool hasn’t taken off in the humanitarian sector.
What’s New: Impact Bonds
What they are: Invented less than 10 years ago, impact bonds are designed to harness private capital for public good. The better the outcome or results of a project, the better the return for the investors who funded the project.
How they work: Private investors put up money to pay for a project in socially focused sectors, including development and humanitarian work. If the project meets its targets, governments and foundations – traditional aid donors – repay the investors, adding a predetermined bonus or interest payment. If the goals aren’t met, investors may forfeit the bonus or capital.
Who uses them: The International Committee of the Red Cross is currently the only humanitarian aid organisation using an impact bond; the bond funds medical centres for people largely wounded in conflict. Various bonds fund 14 projects in developing countries, including early childhood education in South Africa and cataract surgery in Cameroon.
Impact bonds: The upside
- Donors share risk with private investors
- The model may, over time, attract private philanthropy and expand the pool of aid donors
- Long-running humanitarian-development needs may be a natural fit for results-based funding
- Accountability is increased by the need for highly measurable impact that can be used to assess whether the goals of a project are met
- New service providers may be attracted by wide calls for vendors, challenging established aid groups to innovate and modernise
Impact bonds: The downside
- Contractual set-up is complex, time-consuming, and expensive
- Ditto for monitoring and compliance – which could mean more money spent on administrative tasks, less on aid recipients
- Aid budgets – largely drawn from public tax monies – are tapped to pay profits to corporations, banks, and private investors
- Few new private donors have stepped up, so traditional aid donors still pick up the bill
- The crisis-induced, spur-of-the-moment funding needs of much humanitarian work are not conducive to the time required to set up and monitor impact bonds
Although an internet search will turn up many documents, articles, and speeches referring to the potential of humanitarian impact bonds, most lead to the same example: a project run by the International Committee of the Red Cross.
That, according to financial professionals and aid executives and analysts interviewed by The New Humanitarian over the past four months, is because it is the only one. The ICRC is not only an early adopter, so far it’s the only adopter in the humanitarian field. It is two years into a five-year bond, costing up to $25 million, that is paying for new medical centres in three countries.
Juan Coderque, head of new financing models at the ICRC, said the organisation had looked to financial innovation, like the bond, to “diversify… and bring additional resources”.
Impact bonds, invented less than 10 years ago, are designed to harness private capital for public good. Ideally, private investors and the public sector can both gain in funding new approaches to existing problems. They have been used to fund “social impact” projects, mainly in Europe and North America.
Despite their seeming promise, those involved in setting up and monitoring the bonds say there's so far little evidence they are suited to funding the needs of humanitarian aid projects, while critics allege that they are a "smoke and mirrors" contrivance that makes profit for investors at the expense of taxpayers and recipients.
Great expectations, slow start
Worldwide humanitarian needs have grown fast in recent years, driven by conflicts as well as natural disasters. In 2006, UN-led relief operations aimed to help 31 million people. In 2019, the figure is some 106 million. Funding has not kept pace: response plans coordinated by the UN missed fundraising targets by an average of $9 billion from 2015-2018. Meanwhile, income depends heavily on the European Union, the United States, and a handful of other wealthy nations.
One promise of impact bonds is in mobilising new sources of finance: an investor agrees to fund a project up front, and an “outcome funder” – usually a traditional humanitarian donor such as a government or foundation – pays back the investor at the end.
The Impact Bonds Working Group, a collection of major aid agencies, bankers, lawyers, insurance companies, foundations, and conventional donors believe impact bonds are an ingenious opportunity in development and humanitarian settings. The group’s website states that the bonds are a “powerful tool for mobilising some of the world’s most creative capital and know-how to make optimal use of scarce public resources”.
The working group, hosted by the World Bank, aims to increase the number and pace of deals and to pool outcome funder capital to help impact bonds and similar funding tools take off in the humanitarian and development sector.
Despite the slow start in the humanitarian segment, overall “the market has grown enormously”, said Emily Gustafsson-Wright, a fellow at the Brookings Institution who focuses on impact bonds. Five years ago, she said, not long after the concept of impact bonds first emerged in 2010, she was monitoring 38 of them. Now, the Brookings Institution keeps track of over 165, 14 of which are in developing countries, with $408 million in up-front capital.
Yet, she said, relative to “impact investment” as a whole, “the market is quite small, in terms of [the] number of impact bonds, and also in terms of [the] size for each of them”. And, she noted, commercial investors are still few and far between.
A UK-funded study on impact bonds in the development and humanitarian sector, including the ICRC one, has only recently started but has presented mixed evidence of success so far.
A common characteristic of such bonds, an early preview of the study found, is that they are “complex to design and expensive to set up”. In one example, a project to remove cataracts in Cameroon, set-up and legal costs were very high: up to 21 percent of total spending. Impact bonds, the researchers suggested, could also introduce new and different risks to conventional grant funding.
Sceptics say the model simply makes projects more expensive for traditional donors while directing money to private investors. In the ICRC’s project, for example, if it’s a success, the taxpayers of Switzerland, Belgium, and other funders will pay up to $25 million for a project that cost only $18 million to complete.
One senior humanitarian analyst, speaking on condition of anonymity to preserve professional relations, said the model was “smoke and mirrors” and enthusiasts were “confusing liquidity and capital”.
Nonetheless, Gustafsson-Wright and others said that impact bonds may offer indirect benefits to the humanitarian sector, including “a mindset shift towards outcomes” and more clarity on costs that is “not always there in traditional grant-based financing”.
In the end, most completed impact bonds tracked by the Brookings Institution did achieve their goals, Gustafsson-Wright said. However, she added that there’s been no rigorous evaluation comparing all aspects of impact bonds to other forms of financing.
How they work
In the case of the ICRC’s bond, if the project meets all its targets – including patients being successfully treated at a rate 80 percent faster than in existing centres – the investors will get their money back, plus interest payments. If not, they can lose 40 percent of their investment.
By the end of the five-year project (2017-2022), the bond requires the ICRC to construct three new centres in three conflict -affected countries, and meet efficiency targets based on the number of clients treated per staff member.
- Investors supplied about $18 million.
- If the project meets its goals, outcome funders will repay the investors their original $18 million plus seven percent annual interest, or about $25 million in total.
- If targets are missed, penalties are imposed on a sliding scale. In the worst-case scenario, investors would get 60 percent of their principal back: 50 percent from the outcome funders and 10 percent from ICRC.
Twice throughout the life of the bond, a third party will assess whether the project meets its goals: once after the construction phase and again at the end.
The investors are led by insurance firm Munich Re and others brought on board by Swiss bank Lombard Odier. The outcome funders – traditional donors – are Switzerland, Belgium, Italy, and the UK as well as the La Caixa Foundation. The Netherlands also contributed at the initial stages.
A neutral third party, usually an independent consultancy, (Philanthropy Advisors in the case of ICRC) will assess the success or failure of the project against specific targets.
Under the terms of the ICRC bond, if the project goes badly off track, the outcome funders pay no interest and only pay back 50 percent of the investor’s principal – about $9 million. In this case, ICRC would also have to pay back a further 10 percent of the investors’ original stake.
In a typical aid scenario, said Francois de Borchgrave, managing director of Kois Invest, donors accept the risk that a humanitarian project may not be completely successful – but still have to pay anyway. His firm was closely involved in setting up the ICRC bond and has advised on impact bonds in other sectors.
By contrast, de Borchgrave said, the bond offers the opportunity to fund solely on performance: there is little risk for the outcome funders. “You pay a bit more, but when it doesn’t succeed, you don’t pay,” he said.
About 38 percent of the total expenditure on the ICRC project – mostly from aid budgets – could go to Munich Re and other investors, not to the ICRC or the populations it serves.
For donors, it’s not been an easy model to adopt: their financial budgeting is rarely well-suited to conditional payments deferred for several years.
Coderque, of the ICRC, said the impact bond project “required some very interesting gymnastics” for its outcome funders, including Belgium, which had to change its law governing aid spending to fit the needs of the scheme.
In addition, the complex contracts required for the bonds mean legal, monitoring, and other advisory costs add up: they have been estimated at 10 percent of the final cost of the ICRC bond.
Independent humanitarian finance analyst Lydia Poole feels the framing of risk is both misplaced and undersells one of the potential advantages of the bond. Donors and aid agencies usually face no real financial risk for a poor quality aid project, she said – unless it's so bad a media scandal erupts.
In fact, Poole believes, the project beneficiaries bear the true burden of the risk of failed projects, as they may not get the help they need. This lack of accountability bedevils the aid system, she said.
One of the few compelling elements in the impact bond model, Poole said, is if the aid agency implementor pays a penalty for poor performance some of the risk is shifted onto actors in the delivery chain, creating a real incentive to achieve results.
In the ICRC project, for example, ICRC is liable to pay investors about $1.8 million if the project is a major failure.
People seriously injured in warfare conflicts may need amputations, rehabilitation, and specialist medical care. In 2016, the ICRC helped 330,000 people with disabilities. This included fitting thousands with artificial limbs and physiotherapy – often over the course of years. Yet projects that provide these services are often funded on a year-to-year basis and housed in ill-suited facilities that were never designed for the job. ICRC’s impact bond is designed to address these inefficiencies with purpose-built facilities, training and workflow. The bond funds three purpose-built physical rehabilitation and orthopaedic units in war-affected countries: the Democratic Republic of Congo, Mali, and Nigeria.
Ahmad received artificial limb in @ICRC_af Orthopedic center, he shows his emotion with dance after getting limbs. He come from Logar and lost his leg in a landmine. This is how his life changed and made him smile. pic.twitter.com/Sg7jJbUD2V
— Roya Musawi (@roya_musawi) May 6, 2019
Principles and stumbling blocks
To justify the added expense, the bonds should meet a particular financing problem that can’t otherwise be addressed with cheaper sources of funds, suggests Poole. The ICRC could, she believes, have found ways to structure traditional grant funding rather than embark on the bond – which, Poole says, is perhaps mislabelled as a “humanitarian” bond, since the long-term nature of the project and services could equally be tagged “development”.
If Belgium could change its funding rules to work with the ICRC bond, Poole says, then it could probably also have found a way to fund the ICRC’s needs over a five-year period without needing to commit to the additional costs of an impact bond. Belgium reports that more than 70 percent of its humanitarian funding is already issued as multi-year agreements.
There are political and ideological questions, too. A May report on the results of a three-year study into impact bonds in Canada, the United States, and the UK, carried out by academics at Canada’s York University, stated that critics view the model as “a slippery slope towards the privatisation, marketisation, and financialisation of the social sector”.
As Poole puts it, “there’s an ethical dimension” when “experimenting” with public money. In the case of the ICRC project, she said: “I have yet to be convinced the thresholds are met to justify the additional cost.”
Backers counter that in the long run, the productivity gains justify the premium paid by donors (and their taxpayers). In the UK, for example, the government tried a social impact bond to reduce re-offending rates of convicts. If successful, the government would save more on future prison expenditure even taking into account the extra payout.
In the humanitarian sector, similar logic applies: if the ICRC’s new physical rehabilitation facilities really are 80 percent more efficient, every donor dollar for the ICRC will go further.
Poole was co-author of a 2015 study on humanitarian finance prepared in advance of the World Humanitarian Summit that said impact bonds were most likely to work “in stable settings”, not in conflicts or fast-moving crises. However, Poole noted that even today the number of impact bonds for development projects was also low – the Brookings study says only 14 have been attempted in developing countries.
Humanitarian impact bonds appear best suited to protracted situations – projects dealing with health infrastructure or refugee services, for example – that would benefit from a better alignment of development and humanitarian funding, a strategy sometimes known as the “nexus”, Colderque said.
In addition, Gustafsson-Wright said, it may not be feasible to use the impact bond as an instrument for short-term emergency funding because it takes too long to set up. But she does see potential in projects with a longer-term horizon, such as refugee integration. For example, impact bonds support refugee integration and employment projects in Germany, Switzerland, and Finland – areas not usually described as humanitarian.
Measuring the results can also be difficult in complex situations with “a lot of moving parts,” Gustafsson-Wright noted, adding that too many variables or unclear targets would make an impact bond hard to structure. “If outcomes are not clearly measurable, then it’s hard,” she said. And, she added, some crisis situations will be risky by definition, and finding investors willing to take on such risk could be difficult.
Toward the future: Billionaires, bandwagons, and silver bullets
The hope among some analysts in the sector is that the model, as it becomes more widely understood, could attract unconventional funders, such as Silicon Valley billionaires.
But so far there is little new money. De Borchgrave says some Silicon Valley investors have shown interest, but declined to provide details, citing client confidentiality.
He and other analysts of the embryonic bond segment say it has also not yet unlocked new sources of capital – outcome funders – to pay for humanitarian relief.
The market for the unusual nature of the risks is very young, although a handful of foundations and nation-state donors, like those behind the ICRC project, are testing the potential of a future market.
These potential outcome funders could appreciate the option to “buy” a measurable, socially beneficial result without risk of failure. One impact bond is being designed to help eradicate cholera from Haiti, for example.
The ICRC impact bond is formally titled a “Humanitarian Impact Investment.” In recent years, other terminology has come to mean more or less the same thing as “social impact bond”, explains Emily Gustafsson-Wright of the Brookings Institution. Those terms include “pay-for-success financing”, “social benefit bonds”, or “social outcomes contracts”.
At least one other humanitarian impact bond is on the near horizon: a bond to develop the skills and small businesses of lower-income residents and refugees in Lebanon and Jordan. De Borchgrave said it may launch later this year. It’s priced at $21 million and implementation will be run by the Near East Foundation, a US-based NGO working on education and economic development in the Middle East and North Africa. Interest is pegged at up to eight percent, and the IKEA Foundation has pledged to be one of the outcome funders.
“Every other day” humanitarian organisations contact his firm, wanting to understand how impact bonds work, de Borchgrave said. Yet he knew of few projects in the pipeline.
Is he disappointed by the slow uptake for funding humanitarian work? It’s “a bit disappointing”, he said, but “in a way I’m happy that not everybody is jumping on the bandwagon”.
“Impact bonds are not the silver bullet for every single humanitarian issue,” he added, noting that the design of the bond can be “uncomfortable”, both for the aid agencies and donors. “It’s something that forces people to be very transparent about results, about metrics”.
That doesn’t always mesh well with humanitarian culture, which he said isn’t driven primarily by costings. “They would fly to the moon to help one person,” he noted.
Despite all the hype, Coderque told TNH recently that, “there is quite a bit of distance between the talk and the walk”.
Although the ICRC’s first bond had been complicated to set up and manage, he said, the organisation is pursuing other such bonds. One involves a city-wide sanitation project in a “biggish” city in Africa, he said, declining to give more specifics.
The bond concept may also widen the field of organisations that take part in relief or development projects: donors tend to contribute based on tenders, or Requests for Proposals (RFPs), de Borchgrave said. “The one who wins is the one who’s actually very good at giving an answer to an RFP, not necessarily the one who’s the best at delivering on the ground.”
Looking back now that the first impact bond in the humanitarian sector is well under way, was the experience worth it to the ICRC? “Absolutely,” Coderque said, noting that his organisation had not turned to the bond as a last resort for a difficult funding situation but had instead opted to use the bond as a tool for innovation: “We had a solution, an impact bond, and we were looking for a problem.
“There is a moral imperative to look at new things,” he said. “Not looking at new things is, I think, a moral failure.”
(*An earlier version of this story incorrectly named the consultancy as Philanthropy Associates.)
(TOP PHOTO: A funding tool called a Humanitarian Impact Bond is financing the construction of three rehabilitation and orthopaedic centres to be operated by the International Committee of the Red Cross (ICRC). Here the centre in Kinshasa, Democratic Republic of Congo is seen under construction in June 2019. Two others will be built in Nigeria and Mali in a first-of-its kind scheme running from 2017-2022.)
It was The New Humanitarian’s investigation with the Thomson Reuters Foundation that uncovered sexual abuse by aid workers during the Ebola response in the Democratic Republic of Congo and led the World Health Organization to launch an independent review and reform its practices.
This demonstrates the important impact that our journalism can have.
But this won’t be the last case of aid worker sex abuse. This also won’t be the last time the aid sector has to ask itself difficult questions about why justice for victims of sexual abuse and exploitation has been sorely lacking.
We’re already working on our next investigation, but reporting like this takes months, sometimes years, and can’t be done alone.
The support of our readers and donors helps keep our journalism free and accessible for all. Donations mean we can keep holding power in the aid sector accountable, and shine a light on similar abuses.