Thinking Through Waste, Fraud and Corruption in US Foreign Assistance

The US administration has presented a weak case for its argument that US foreign assistance was so riddled with abuse and fraud that the only answer was to shut down USAID and reboot. The lists of terrible projects and “appalling waste” don’t show any evidence of fraud or abuse, and I’d argue most of them don’t even seem wasteful: circumcision to reduce HIV infection rates, early childhood development programs and so on. As much to the point, they account for a fraction of a percent of foreign assistance spending.

I think the White House could have done a better job at making the argument: US assistance is certainly not as efficient or effective as it could be, and it can’t be doubted that hundreds of projects involving many hundreds of millions of dollars have failed to help deliver the long-term development results hoped for by their designers. But at the same time, US aid has also delivered incredible, life-altering, life-saving results at huge scale, with extremely little evidence of diversion or malfeasance. The big reason the administration’s case looks weak is because it is wrong.

Start with the claims of abuse, fraud and corruption: money not getting where it was intended, siphoned off into offshore bank accounts and the like. Between January 1, 2012, and December 31, 2020, the Special Inspector General for Afghan Reconstruction (SIGAR) issued 176 financial audit reports covering $8.5 billion in costs incurred for Afghanistan reconstruction projects. About 0.4 percent of audited costs were eventually disallowed—about $3 million a year, a fraction of SIGAR’s own annual budget, which was $55 million in 2019. Similarly, in Ukraine, audits and spot checks have found essentially no missing aid expenditures all the way through to final intended recipients in the country. Even in the cases of an impoverished fragile state and a country at war there still simply isn’t evidence of diversion of aid at any scale.

Perhaps the most fraught examples are where heading off humanitarian catastrophe involves working with those who bear some or even considerable responsibility for the crisis. For example, since the Taliban victory, the US has been supporting humanitarian activities in Afghanistan that at least indirectly benefit the Taliban regime by providing access to foreign exchange and tax revenues—although it also means that people living in one of the most fragile countries in the world get access to life saving supplies.

But the vast majority of US foreign assistance doesn’t involve flying in pallets of dollar bills to pay for food and medicines and, as the Afghanistan case also illustrates, US assistance comes in for considerable oversight and control. That includes the Offices of the Inspectors General, the Office of Management and Budget, the Government Accountability Office, the Office of the Director of Foreign Assistance, the Special Inspector Generals, the 1,977-page Federal Acquisition Regulations, regulatory controls over overhead rates, Congressional oversight committees, and the 450-page Foreign Assistance Act.

(A side note: one of the many things that isn’t getting done as the administration deconstructs USAID is the oversight and compliance efforts associated with all of these rules. As the USAID Inspector General reported before he was fired, consultation terminations and placing staff on administrative leave have “curtailed USAID's ability to vet humanitarian assistance awards for potential terrorist ties and monitor aid deliveries in high-risk environments” as well as “constrained USAID’s ability to receive and respond to allegations of misconduct.” And as a judge authorizing a temporary injunction against shutting down USAID noted, the dismantling has likely disrupted compliance with “The Federal Information Technology Acquisition Reform Act, the Federal Information Security Management Act, the Privacy Act, the E-Government Act of 2002, and the Government Performance and Results Act, among others.”)

The baroque complexity of oversight and control mechanisms helps to explain why of US assistance contracted to firms and nonprofits worldwide in 2020, four fifths still went to US firms (at least as the prime contractors) and only about 2 percent of aid is provided as direct budget support (cash) to foreign governments. Working through contractors with deep bureaucratic expertise in following all of the regulations is simply the best way to ensure there isn’t risk of noncompliance.

And that brings us to the “efficiency” element of waste. This massive regulatory and oversight counter-bureaucracy (to use former USAID Administrator Andrew Natsios’ term) comes with a price: not least the overhead associated with a contracting chain that so often runs through metro-DC institutions on to subcontracted local suppliers. USAID contractor and grantee indirect costs sometimes equal 30 percent of direct project costs.

The US government’s contracting apparatus certainly generates profits, and sometimes generous payoffs. But to be fair, US foreign assistance simply doesn’t generate the kind of contractor rents created by other parts of the government. The fat cats of US development are emaciated kittens next to the lion kings of defense contracting, for example. Compare the top three USAID awardees to the top three defense contractors and look at CEO pay: for aid contractors it averages less than three percent of CEO pay for defense contractors. The contracting chain may not be the best way to deliver foreign assistance, but the level of inefficiency is comparatively limited.

I'd still trade off the (I think largely illusory) risk of one percent more lost to malfeasance for five percent savings on bureaucracy and overhead alongside working directly with governments and local contractors. But maybe, despite some evidence of bipartisan support for taking on a little more risk for greater efficiency, that isn't a politically realistic choice. Similarly, I’d rather see cash and locally purchased food used to respond to food and nutrition needs in the wake of humanitarian disasters than US crops shipped on US boats at considerable cost, but again, the politics are what they are.

In fact, the way US foreign assistance operates, including its inefficiencies , are the result of US politics far more than any unique features of developing countries—as demonstrated by extremely successful and comparatively streamlined aid programs that work directly with developing country governments as implementing partners, like the World Bank’s International Development Association. But even with the inefficiencies demanded by domestic politics, US assistance often delivers incredible results—which brings us to waste through ineffective use of resources.

Again, there are many clear cases of US assistance projects that failed to deliver on (permanent) development. In Afghanistan, the governance and human rights work supported by USAID over two decades obviously didn’t lead to sustained progress. Even if you hope the roots are still under the ground, the flowering of gender equality has been mowed down in the last four years. Regarding stability, “money as a weapons system” had clear limits in the country. (Mind, you the considerably more spent on weapons as a weapon system in Afghanistan had the same limited return in terms of sustaining a vaguely liberal democratic regime. Iraq appears a potentially somewhat more positive example for both in that regard.)

Far from all the aid (even) in Afghanistan was “wasted” in this sense, however. Not least, tens of thousands of mothers and hundreds of thousands of children are still alive because maternal and child mortality rates both considerably more than halved during the period 2001–2020, when US assistance was flowing into the country at a high rate. There is evidence of a rising mortality rate since the Taliban came back to power, but nothing to suggest a complete reversion. Or look at infrastructure: while the percentage of the country with access to electricity is estimated to have fallen from 98 percent to 85 percent of the population between 2021 and 2022, that compares to an estimated access rate of about four percent in 2000.

It is not just in fragile states that aid for policy, institutional, and cultural change is a gamble on development compared to the certain payoffs of vaccines and medications and the more likely results of infrastructure construction. It is a gamble that depends on local leadership, politics, and luck. Maybe you think that gamble isn't worth making, or at least we are bad at making the right bets. But taking the whole US assistance portfolio, the gamblers can claim some success. After all, there are countries where US assistance for broad-based development has been significant and that have made sustained progress toward wealth and the protection of human rights of late—Liberia is one example (GNI per capita has approximately doubled since 2007 and is up about eightfold from its trough in the late 1990s).

And that brings us to another way we know waste or corruption aren’t rampant in the aid system. We know a lot of aid is being delivered as intended because we can see it in the results. Again that’s clearest for health programs where repeated estimates of PEPFAR’s impact suggesting millions of lives saved over the program, for example. If the program wasn’t working as intended, there would be somewhere between 1.6 to 2.9 million more deaths to report each year. Put all US lifesaving programs together and you get somewhere in the region of 2.3 to 5.6 million or more.

US foreign assistance could work far better: too many parts of the government provide overlapping assistance, there are unwarranted regulations and Congressional mandates with the overheads and reliance on contractors that result, excessive limits on providing assistance where it could work best and too few on spending it where it is an irrelevance. But it still works—sometimes in miraculous ways, saving lives at the cost of a few dollars. There’s simply no evidence of massive waste nor significant fraud or abuse. That means cutting assistance will lead to lives lost and livelihoods ruined. The US administration and Congress should confront that reality, not hide behind the pretense that setting out to destroy America’s foreign assistance institutions has no real cost.

Note: We received an email in response to this blog post from Richard Gardella, a public affairs specialist at SIGAR. Mr. Gardella shared the following updated information:

* Between January 1, 2012, and December 31, 2024, SIGAR’s 284 financial audits examined $10 billion in amounts awardees spent under awards funded by DOD, State, USAID, and USDA.

* During that same time-frame, SIGAR questioned more than $540 million in incurred costs and recommended that the funding agency officials determine if the costs were allowable and if other amounts were payable to the U.S. government, such as interest.

* SIGAR financial audits prompted the responsible contracting and agreement officers to determine the allowability of and recover, as appropriate, the more than $540 million in questioned costs identified in our reports.

* As of December 31, 2024, funding agencies had disallowed $31.33 million in questioned amounts, which are then subject to collection.

* An additional $27.98 million questioned in SIGAR financial audits are awaiting the funding agencies’ determinations on their allowability.

* To date, agencies have demonstrated efforts to recover nearly $30 million in disallowed costs. SIGAR has recommended that agencies attempt to recover more.

It is up to agencies to recover costs; SIGAR can only recommend that agencies attempt to do so.

I’m grateful to Mr. Gardella for the more recent data. To update my estimate that disallowed funding was about $3 million a year and 0.4 percent of funding, the more recent data suggests that $2.4 million a year was disallowed, which is 0.3 percent of funding.