BLOG POST

No, 90 Percent of Aid Is Not Skimmed Off Before Reaching Target Communities

A viral claim, spread and amplified by Elon Musk, suggests that only 10 percent of USAID money reaches its intended beneficiaries. This is a wildly incorrect and misleading interpretation of a different statistic—that 10 percent of USAID payments are made directly to organizations in the developing world. The remaining 90 percent includes all the goods and services that USAID, American companies, and faith-based organizations deliver in kind, from HIV drugs to emergency food aid, malaria bed nets, and treatment for acute malnutrition. And it is absolutely crucial to debunk this false claim because it is being used as part of the pretext to dissolve USAID in its entirety.

As background, over the weekend, Elon Musk took to his “X” platform to declare open warfare against US development assistance. Among a torrent of extreme attacks leveled against USAID, he shared one very specific claim. He quotes a PBS segment with Walter Kerr, co-executive director of Unlock Aid, a DC-based non-profit that aims to improve American aid effectiveness:

A screenshot of Elon Musk's tweet highlighting the alleged inefficiency of foreign aid dollars through USAID.

In the day that followed, variants of the same claim were spread and amplified by commentators on the right, Republican lawmakers, and even El Salvadoran President Nayib Bukele. Its viral quality is perhaps unsurprising: if this factoid were even close to true, it would indeed be a scandal—and those amplifying the claim are using it as evidence of not just inefficiency but rank corruption and grift.

The good news for the US taxpayer—and the beneficiaries of US foreign assistance—is that this figure is very, very wrong. The purported statistic is wildly inaccurate and not remotely what Unlock Aid’s own research actually shows. So let’s have a look at where the 10 percent figure originates, why it doesn’t show what the critics claim, and why it offers zero pretext for Musk’s current course of action.

US foreign assistance and the role of “prime” implementing partners

To understand the actual meaning of the 10 percent figure, you first need to understand how US foreign assistance is administered.

At the highest level, funding is first appropriated by Congress, often with directives for a specific program (e.g. the President’s Emergency Plan for AIDS Relief [PEPFAR] or food aid), country (e.g. Ukraine), or multilateral organizations (e.g. Gavi or the Global Fund). For a large share of bilateral aid, congressional appropriators will refrain from specifying a recipient, so USAID (or other agencies, where applicable) will usually conduct a competitive bidding and review process to select a “prime” implementing partner, who is directly awarded funding from the government and is responsible for running a distinct aid program.

There are many different kinds of “prime” implementing partners, including American nonprofits, including faith-based organizations; multilateral agencies like the Global Fund, Gavi, and UN agencies; big international NGOs like the Red Cross; development banks like the World Bank; for-profit companies who specialize in aid delivery; and foreign governments or country-level NGOs. These “prime” partners often work with many other partners, who receive “sub-awards” from the prime, and who themselves might be local organizations, suppliers, nonprofits, for-profit companies, and so forth.

Needless to say, organizations who are the “prime” on a US aid award do not get to keep the money. They are tasked to deliver goods and services—HIV drugs, bed nets, food aid, education programs, IT systems, etc., to developing countries.

What the 10 percent figure means–and what it very much does not

With this background, you can now understand where Unlock Aid’s 10 percent figure originates: it represents the portion of US foreign assistance awarded directly to a “prime” local partner (government, NGO, or company) in the target country. This factoid is neither a secret or a scandal. Similar figures have been reported in USAID’s own localization reports, which were readily available before its website was shut down over the weekend; see here for FY2024, for example. 

We went back to the data from foreignassistance.gov, and looked at all US foreign aid, excluding military aid, for fiscal years 2021-2024. We then flagged the portion going directly to foreign entities [1] which is the best proxy we have in this data for “local” organizations. (Remember, foreign is good here in Musk’s reckoning, and contracts to American companies are bad.)

Over this time period, awards to foreign governments, companies, or nonprofits as the prime partner accounted for about 11 percent of total US foreign assistance by value, as shown in the figure below. The remaining aid flowed through a variety of channels, including American companies and nonprofits (31 percent); multilateral agencies like the World Bank, World Food Program, and the Global Fund (46 percent); and US agencies themselves (12 percent). 

Infographic showing what happens to US foreign aid that doesn't go directly to foreign organizations

To be clear: though this is all public knowledge, the portion going directly to local organizations has always been lower than it really ought to be, and it remains persistently low despite USAID-wide ambitions to “localize” far greater portions of its funding. Increasing the share of funding allocated directly to local groups is indeed a worthy but thus far largely elusive goal for US aid reform, which deserves genuine scrutiny from policymakers.

But to be very clear: the portion going to foreign organizations is not equivalent, or even close, to the total portion of US foreign assistance reaching partner communities—because US-based and international prime implementing partners deliver lots of aid in-country. The other 89 percent of aid is chock full of some of our most effective, flagship programs, reaching hundreds of millions of people on the ground. For just a few examples:

  • About 7 percent of US foreign assistance went to Gavi, the Vaccine Alliance, which provides vaccines to the poorest countries; and the Global Fund to Fight AIDS, Tuberculosis, and Malaria, which invests in country-led programs in its three disease areas. Contributions to both organizations are specifically earmarked and authorized by Congress. These organizations, in turn, spend about 93 percent and 92 percent[2], respectively, directly on programmatic expenses–that is, delivering supplies and health services directly to countries.

  • Bilateral malaria, HIV, and TB programs—channeled through US-based companies and nonprofits—account for another 6 percent of the pie. This slice accounts for most of PEPFAR and the President’s Malaria Initiative, and includes procurement contracts that purchase and distribute bed nets, AIDS drugs, and other essential medicines for use in partner countries.

  • Another 14 percent supports humanitarian and emergency relief, with 6 percent going to US-based organizations as the prime implementing partners and 8 percent channeled through the World Food Programme. Cumulatively, the US Emergency Food Security Program provided essential support to 77 million people across 56 countries in FY2023. A quick scan of the major non-local implementing partners suggests a mix of large NGOs and faith-based organizations like Catholic Relief Services and World Vision.

  • Though not explicitly pulled out in our chart, in FY2023 Title II Food Aid accounted for $1.9b in USAID spending–essentially all channeled through US-based implementers and international organizations, and reaching 45 million people across 35 countries. The great majority of this amount (74 percent) was spent directly on buying food (including surplus from US farmers and producers) and distributing it in-country, or providing cash assistance to beneficiaries so they could purchase their own food locally. (Much of the remainder–13 percent—accounts for the freight expenses to ship US-purchased food to the target country.)

Rather than any serious criticism of foreign aid, Musk’s claims reflect a basic misunderstanding of how USAID safeguards taxpayer money

USAID is far from perfect. Yes, more money should flow directly to local partners. Yes, overhead rates at some US implementers are too high, and layers of sub-awards can result in progressively fewer aid dollars making it to beneficiaries. Yes, there is lots of room to reform how the US delivers aid to make sure more of the value makes it to intended beneficiaries—for example by reforming food aid to support greater local procurement.

But a key reason USAID relies on American and multilateral intermediaries is to protect against fraud and corruption. The agency demands that its partners provide a detailed paper trail to account for every dollar spent—creating administrative hurdles that few local organizations can clear. Musk has this totally backwards.

This cursory review of the data, conducted over just a few hours, shows just how off-base the recent social media caricature of US foreign assistance has been. The 90 percent of the aid budget that Musk and others claim is wasted is, in fact, helping to feed, house, and provide basic health care to some of the most vulnerable people on earth. There is no valid pretext to burn USAID to the ground over clear misinformation.


[1]  Here, “foreign” recipients include any organization that is not a multilateral, but is headquartered outside of the United States. Based on prior USAID localization reports, which are not currently available, we expect the majority of these recipients will be local organizations. However, the 11 percent figure also includes organizations headquartered in third-countries; for example, a theoretical NGO based in Switzerland would qualify as a “foreign” channel even though Switzerland is not a recipient of US assistance, and we can safely assume the aid is being targeted to a third country. 

[2] Calculated by considering the Global Fund’s total operating costs ($390m per year) and dividing them by annualized country allocations for 2023-2025 ($4.3b per year). Both numerator and denominator exclude strategic initiatives and multi-country catalytic programs, which are relatively minor outlays.

Disclaimer

CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.


Image credit for social media/web: John Hanson Pye/ Adobe Stock