What We Know—and Don’t Know—About the Trump Administration’s Global Health Agreements

The State Department has announced an initial set of bilateral health cooperation agreements, signing eight memorandums of understanding (MOUs) with governments in sub-Saharan Africa. These compacts are positioned as a central pillar of the Trump administration’s new approach to global health assistance—and offer the clearest signal yet of how the United States intends to operationalize the country-partnership goals envisioned in the America First Global Health Strategy.

After months of disruption following the dismantling of USAID and the resulting interruptions to global health assistance, it is encouraging to see renewed US engagement—and even the prospect of continuity with some long-standing priorities.

But press releases and early headlines only tell part of the story. The compacts mark a sharp departure from how the United States has delivered global health assistance over the past two decades. They combine US funding reductions, ambitious co-financing expectations, and a shift toward direct government-to-government assistance, all under a newly reconfigured approach. While each of these arguably represents long-needed changes, they carry tremendous risks to service delivery and hard-won public health gains.

The details of how agreements are implemented, financed, and monitored will matter enormously, with lives at stake. Operationalizing a reconfigured approach to US global health assistance—particularly direct government assistance—at this scale and speed is unprecedented, and each potential point of failure risks lives. Ultimately, whether the compacts represent a constructive next chapter will depend less on stated intent than on how they balance country ownership with fiscal realities and credible accountability.

Here’s what we know so far—and what remains unclear, but likely incredibly consequential for the continuity of life-saving services, as this next chapter unfolds.

What we know so far: Understanding the lay of the land

Scope and scale of the agreements

Secretary of State Marco Rubio signed the first health compact with Kenya, alongside President William Ruto. Seven additional MOUs followed, bringing the total to eight. Each compact pairs a five-year US funding commitment with a negotiated pledge from the partner government to increase domestic health spending over the same period.

Figure 1 illustrates the total dollar value of signed compacts by country at the time of publication. Hover to see the breakdown.

Figure 1. US bilateral health agreements

Despite differences, the topline is that these compacts represent a reduction in total US health spending for each country. Relative to FY24 health obligations, the agreements amount to a 49 percent decrease, on average, in annual US financial support.

Further, while year-by-year co-funding details are not available for all compact countries—and first-year reductions may appear modest—the trajectory is clear: over the five years, even the poorest countries face steep declines in US financial support compared with historic funding levels. Mozambique and Liberia (both low-income countries) would see reductions of 39 percent and 63 percent, respectively, relative to FY24 levels. In comparison, Kenya—a relatively wealthier country—faces a 28 percent cut when the compact value is annualized (see Figure 2).

Figure 2. Projected five-year declines in US health funding by country

What the money will be spent on

While we understand that either party may make these agreements public, only three MOUs—Kenya, Uganda, and Liberia—are currently available, the provenance of which are unclear. Although each compact will vary by country, the agreements still offer a useful reference point for what to expect in others.

The three MOUs reveal investments across six areas: health commodities, frontline health workers, data systems, disease surveillance and outbreak response, laboratory systems, and a flexible “strategic assistance” category intended to support shared priorities and innovation. The agreements also outline shared annual performance and process targets.

The compacts appear to preserve (some) continuity with disease-specific investments. The bulk of funding under the commodities and health worker categories is likely to support life-saving treatment, particularly for HIV/AIDS. At the same time, reporting suggests a relatively rapid drawdown of US investments in areas such as tuberculosis and malaria, while support for surveillance and outbreak response remains comparatively steady. Notably, the compacts would almost eliminate maternal and child health funding. They also appear to signal greater “verticalization” around major infectious diseases rather than integrated health systems support (even if the funding modality may be more fungible than past approaches).

Who pays for what

Country governments are expected to shoulder a growing share of financing. The MOUs available so far do not precisely define domestic government health expenditures, kicking the can down the road to a later implementation phase; they specify that funds from other donors or multilateral organizations (presumably including World Bank financing) would not count toward co-investments.

Further, the MOUs for Kenya, Uganda, and Liberia explicitly allow the US to reduce or cease funding if governments fail to meet co-investment commitments. Per Uganda’s MOU, for example, the US plans to reduce funding at a 2:1 ratio if the government does not make good on committed annual increases to health spending.

Even if countries meet their targets, US funding could be subject to shifting congressional priorities or future funding disruptions. Congress will need to appropriate the funds to deliver on the compacts’ precise line items. This raises uncertainty about the durability of multi-year compacts and whether countries can confidently plan around them. Moreover, these agreements are not legally binding like a contract—the MOUs are formal statements of intent.

Accountability risks

In principle, embedding co-financing into health agreements can strengthen sustainability and align incentives. In practice, such arrangements hinge on credible accountability mechanisms. A shift toward greater reliance on government service delivery could reduce US visibility into spending and performance—raising concerns about how to maintain accountability without burdensome reporting requirements.

This tension has long been a sticking point for Congress in adopting a government-to-government assistance model. Well-designed compacts will need to clarify evolving oversight roles, rely on and strengthen national data systems, and preserve US capacity for independent verification.

Fiscal realities and sustainability

Finally, co-investment targets will need to be grounded in fiscal reality. Recent World Bank analysis, for instance, shows that government health spending in low-income and lower-middle-income countries has stagnated and remains far below levels needed to finance universal health coverage. Between 2024 and 2030, projected increases in government health spending in most low-income countries (LICs) and many lower-middle-income countries (LMICs) are expected to be outweighed by steep donor cuts—leaving total health spending (government + donors) falling in roughly 80 percent of LICs and 40 percent of LMICs.

Table 1. US bilateral global health agreements

Source: State Department and country government press releases and public statements. 2025 World Bank Group Government Resources and Projections for Health report. International Monetary Fund World Economic Outlook April 2025. Author calculations. Rounding errors are possible.

*The State Department has not shared a dollar amount associated with the government of Mozambique’s co-investment. The press release noted Mozambique commits to increasing its domestic expenditures on healthcare as a percent of its government budget by nearly 30 percent over the next five-years. Thirty percent of Mozambique’s FY24 health expenditure is approximately $137 million.

What don’t we know: Understanding potential risks

Much (still) remains unclear about how these compacts will be implemented in practice. Below is a non-exhaustive list of key areas we’ll be watching:

How will transactional pressures shape priorities?

One key uncertainty is the extent to which compacts can advance broader foreign policy objectives. When health assistance becomes tightly linked to non-health priorities—such as diplomatic alignment, cooperation on unrelated policy issues, or advancing business interests—life-saving programs risk being leveraged to undermine health outcomes. Recent reporting on the delay in Zambia’s compact signing, seemingly informed by interest in mining sector engagement, underscores the risk transactional dynamics can play when lives and public health gains are at stake.

How will oversight, evaluation, and accountability actually work?

The administration is operating with a notably lean structure, transferring only a fraction of USAID’s former Global Health Bureau to design and manage this new assistance architecture. With limited in-house procurement and contracting capacity at the State Department—and no USAID missions managing awards and relationships with sectoral government counterparts on the ground—how compact authorities and accountability systems function in practice will be critical.

Equally important is whether elements of the traditional monitoring and evaluation architecture are preserved, and how they might steer investments toward cost-effective interventions and support the generation and use of evidence to inform program decisions.

What funding mechanisms will be used?

The US has historically supported government systems through a range of models: milestone- or results-based awards, direct cash transfers to national budgets, and multi-donor trust funds administered by the World Bank. These approaches vary widely in fiduciary risk and risk tolerance.

It is unclear which mechanisms the administration will deploy for the compacts (likely varying by country). The agreements acknowledge a need to channel some resources through implementing partners (off-budget) while planning to increase the amount of assistance flowing directly through a partner government’s treasury (on-budget). Both the Kenya and Uganda MOUs commit to a mix of on and off-budget assistance and to incrementally increasing funding through government-to-government mechanisms over the course of the compacts.

One promising option for the initial years of implementation in some contexts is the Millennium Challenge Corporation approach. This is where the agency obligates funding upfront to a locally established entity (Millennium Challenge Account [MCA]) responsible for oversight, procurement, and compliance. While often embedded in local government structures and shaped by local actors and priorities, funds do not flow through the central treasury. Particularly as the US seeks to transition commodity procurement to partner governments, an MCA-like approach could combine country ownership with strong US oversight and long-term institutional capacity building.

What safeguards exist for continued service delivery?

Reporting suggests compact funding is tied to milestone-based conditions. While such mechanisms can strengthen accountability, the administration’s approach to managing underperformance is not yet clear. If a country misses a milestone, what safeguards will prevent disruptions to life-saving services? And how will treatment continuity be protected if funding is paused or withdrawn? Moreover, in the event of a major crisis, coup, or geopolitical shift, what contingency mechanisms are in place to preserve shared health system investments?

What’s the plan where compacts aren’t viable?

In some settings, there may be no credible or stable government counterpart with which the United States can negotiate or implement a compact—even though the country remains a priority for global health security or humanitarian response. How the administration intends to operate in fragile or conflict-affected contexts where collaborating directly with a government may not be feasible (e.g., Haiti) is an open question. Alternative delivery models will be essential where needs are acute, but state capacity is limited.

With thanks to Rachel Bonnifield, Erin Collinson, and Charles Kenny for useful feedback