It is a commonplace that corruption is the major stumbling block to aid effectiveness: in the U.S., the average response to the question “how much aid ends up in the pocket of corrupt officials?” is 60 percent, while in the U.K., more than half of people surveyed said corruption is the single most important reason why poor countries stay poor.
The heads of aid agencies respond accordingly. Take, for example, World Bank President Jim Kim, who said in 2013 that “We will never tolerate corruption, and I pledge to do all in our power to build upon our strong fight against it.” Development agencies treat corruption as a problem they can measure and improve through tighter control and centralized delivery. But we can’t really measure corruption, and our fear of the threat or perception of corruption drives responses that are detrimental to aid effectiveness and, consequently, hurt the world’s poorest people.
The idea that a culture of corruption determines the effectiveness of aid is pervasive, but it does not stand up to scrutiny. Research by my former colleague David Roodman at the Center for Global Development overturned a World Bank study that had suggested countries receiving aid with weak institutions and poor policies failed to turn that aid into economic growth. Cevdet Denizer, Daniel Kaufmann, and Aart Kraay — researchers based at the World Bank — looked at the bank’s own projects to find that about 80 percent of the variation in project outcomes between success and failure occurs across projects within countries, rather than between countries. If a national “culture of corruption” had been the dominant force behind failure, which country the project was in would have had a far bigger impact on success.
"Our fear of the threat or perception of corruption drives responses that are detrimental to aid effectiveness and, consequently, hurt the world’s poorest people."—
Nevertheless, fear of corruption has led some aid agencies, including the Millennium Challenge Corporation to reduce or deny aid to countries they perceive as corrupt — even if they might be home to some of the world’s poorest people. Or look at Afghanistan, where the U.S. special investigator general for reconstruction wanted to close down USAID’s support for the health ministry in the country despite historically unprecedented health progress simply because the ministry’s accounts weren’t perfect. The investigator general had no evidence of corruption, but he felt messy paperwork was enough of a red flag to stop financing for vaccinations and antibiotics.
And corruption concerns are a big reason — or at least excuse — why so much aid is run out of donor capitals, passed through consulting firms in those capitals and integrated so poorly with the investment plans of recipient countries. For U.S. aid, considerably less than $1 in $10 is spent using recipient country procurement systems. In some developing countries, the end-run around government is almost complete. In 2010 and 2011, the Haitian government directly received about $23 million in aid. That’s less than the U.S. government contracted with each of Chemonics International, Lakeshore Engineering Services, Development Alternatives Inc., PAE Government Systems, and Management Sciences for Health — all private contractors for USAID — for their work in Haiti.
"The idea that a culture of corruption determines the effectiveness of aid is pervasive, but it does not stand up to scrutiny."—
But it isn’t just an American problem: in 2014, general budget support to recipient countries accounted for less than 1 percent of bilateral aid flows according to OECD data. All of this despite a strong international consensus shared by donors and recipients alike that aid spent through recipient governments is far more effective at delivering development.
And even on the rare occasions when money is funneled through recipient governments, corruption concerns have led to a strong focus on process and financial oversight. That is why the World Bank spends considerably more each year on procurement and financial management specialists alongside investigators looking for fraud and corruption than it does on evaluating if its projects achieve outcomes in terms of improving wealth, health, and well-being.
It is time for donor agencies to fundamentally rethink their anti-corruption approaches. Rather than larding projects with oversight procedures and paper-pushing, they could respond directly to fears that corruption might stop aid from delivering development by better measuring and ensuring that quality goods, works, and services have actually been delivered in projects. If the road project is built on time, on budget, and to quality standards or if all the children who should be vaccinated by an aid-financed project are vaccinated, then we can say with some confidence that the project worked and corruption did not derail it. To reduce corruption and improve effectiveness, it is time to direct our zero tolerance toward aid that doesn’t deliver — and focus on results, not receipts.
Join the Devex community and access more in-depth analysis, breaking news and business advice — and a host of other services — on international development, humanitarian aid and global health.