Robert Hahn is director of economics at the Smith School, University of Oxford, and a senior fellow at the Georgetown Center for Business and Public Policy. Peter Passell is a senior fellow at the Milken Institute and the economics editor of the Legatum Institute's Democracy Lab. They are cofounders of Regulation2point0.org, a web portal on regulatory policy.
Pretty much everybody who's been paying attention has an inkling about why economic aid to developing countries has largely proved ineffective. Too often, it can be summed up in a word: incentives.
Donors are usually governments, international agencies, and philanthropies run by bureaucrats whose employment security and salaries aren't closely tied to project success. Indeed, those in charge often have strong incentives to stay with the herd regardless of outcomes. Much the same goes for both recipient governments and aid workers on the ground.
Over the years, battle-hardened donors have gotten better at delivering aid efficiently, cutting out the middlemen whenever possible, and demanding strict accountability. But the incentive problem is a tough nut to crack. Hence of lure of an idea pioneered in Britain—the social impact bond.
In 2010 a British nonprofit called Social Finance teamed up with the U.K. Department of Justice to try a new approach to solving a notoriously difficult problem: prison recidivism. Rather than spending on the same old, same old, they raised some $7 million from investors to pay for privately delivered support to some 3,000 men released from Peterborough Prison. The government will repay the investors with tax money according to the percentage who stay out of prison and for how long, with a maximum return on investment of 13 percent annually for eight years.
Note that the agreement is not a "bond" in common parlance. It's a pay-for-performance contract between the government and private investors to provide outcomes rather than services—a contract that puts the risk on the investors who, in turn, hire the service providers. If the investors don't think the project will succeed, they have no incentive to risk their money. And if contractors screw up on delivery, the investors have incentives to fire them.
While this market-friendly approach to getting the best out of service providers is too new to have a track record, it does mesh neatly with center-right ideas about getting more out of government for less. And not surprisingly, it is already being copied in the United States (New York, Massachusetts, Ohio) as well as Australia and Canada.
Which brings us back to developing countries. Britain's Department for International Development is backing a deal in Ethiopia in which payment will be determined by the numbers of high school students who pass a specific achievement exam, along with parallel projects in Rwanda and Uganda in healthcare. And a U.S. nonprofit called Instiglio is working with the Colombian state of Antioquia to design model social impact bonds. Meanwhile, the research-oriented Center for Global Development is trying to drum up more support for what it terms "cash on delivery" projects.
Is the social impact bond (or whatever you want to call it) the future of development aid? It certainly makes sense on paper. But the path from here to there may be tortuous. Are there enough investors around with the knowledge and deep pockets to make a difference? Can incentive contracts be written with clear, meaningful outcomes in mind? Will donors be able to contain fraud on the part of investors and service providers? Will the bureaucracies that control the aid money put up with the degree of transparency implied in the contracts? (From their perspective, after all, success may be better than failure—but invisible failure beats transparent failure every time.) Even cynics ought to agree, though, that this is one of those ideas that just has to be given a chance.