After the Collapse. What the End of USAID Actually Means.

The international development community is navigating a structural break. Domestic nonprofits think they are watching someone else’s crisis. They are not. USAID is the case study. Dependency is the story.

I’ve been in a lot of rooms where civil society was being built. I’ve also been in the rooms where it was being dismantled. They feel different. They start the same.

What I’ve Seen

In Uganda, I worked with a coalition of young farmers building the organizational infrastructure to advocate for their own communities. They had passion, purpose, and USAID-funded support that gave them the training, governance systems, and sustained engagement to become something real. In Serbia, Kosovo, and Bosnia, beginning in 2010, and in Albania and Moldova since 2018, I worked with civil society organizations doing the unglamorous, essential work of judicial reform and institution building in societies where that capacity had been deliberately destroyed. In the Caribbean, I supported USAID-funded programs working to reduce youth violence in communities that had few other options. In Bolivia, I helped an agricultural development foundation build financial sustainability after its USAID funding was set to expire.

I list these to place myself inside this story, not to catalog a resume. Most of it was live work until the agency stopped operating. Every one of those organizations built something real with USAID support. Imperfectly, yes. The system had serious flaws that deserve honest examination. And still they built real entities. Trained real people. Designed real systems. Developed real capacity. Served real communities.

I was three weeks from flying to Harare. The Zimbabwean consultants we had already put through Common Ground's Consulting Academy were waiting for the advanced workshop that would prepare them to do this work without us, and the six months of mentoring behind it. The whole point was to lift up local talent and make ourselves unnecessary. Then the stop-work order arrived and took it all away. Those local experts turned consultants are still in Harare. The project that would have enabled them to support local organizations is not. I wish I could say we continued the work without USAID. Those consultants have families to feed and bills to pay. They moved on to other work. So did I.

All of that work is largely gone now. Not reduced. Not restructured. Gone. And the communities those organizations served didn’t get a transition plan.

The Numbers

Start with what actually happened, because the shorthand is wrong in a way that matters. USAID was not abolished. On July 1, 2025, it ceased implementing foreign assistance, and its surviving functions moved to the State Department.¹ Congress never repealed the statute that created it. The agency ended by executive action against a law still on the books, which is worth holding onto, because it shows how quickly the ground can move under an organization that assumed the ground was fixed.

In March 2025, Secretary of State Marco Rubio announced the cancellation of 83 percent of USAID programs.² Read that number carefully. It counts awards, not dollars, and the canceled awards skewed small and old. The Center for Global Development put the cut at roughly 34 percent of FY2024 and FY2025 obligations.³ The smaller number is the more damning one, because a third of the largest aid program on earth disappearing in six weeks is not a rounding error, and quoting the bigger figure invites an argument that lets people avoid the real one.

By July 1, roughly 94 percent of USAID staff had been laid off.⁴ U.S. foreign aid disbursements fell to $47.3 billion in fiscal 2025 from $73.5 billion the year before, the lowest level in inflation-adjusted terms since 2004.⁵ Measured as official development assistance on the OECD’s calendar-year basis, the drop was 56.9 percent in real terms, the largest single-year reduction by any donor on record.⁶

The employment damage is harder to pin down and warrants careful wording. A March 2025 survey of 981 humanitarian operations logged at least 12,000 terminated staff contracts and found that more than a quarter of international NGOs receiving U.S. terminations had closed at least one office in the country where they worked.⁷ An industry tracker maintained since the cuts began now puts global development job losses above 250,000, self-reported by 183 organizations.⁸ The first number is conservative and solid. The second is the best estimate available, self-reported and voluntary, and worth naming as a tracker rather than treating as a count.

The human cost is not abstract. In eastern Congo, the M23 offensive destroyed water infrastructure, and the suspension of USAID-funded water and sanitation programs removed a main source of replacement supply. In Goma, residents drew untreated water directly from Lake Kivu.⁹ Cholera followed. The World Health Organization recorded 71,147 cases and 2,071 deaths in the Democratic Republic of Congo in 2025, the country’s worst outbreak since 1994. The case fatality ratio more than doubled, from 1.4 percent to 2.9 percent.¹⁰ Cases rise for many reasons. A doubling in the share of cases that end in death is what a collapsed response looks like.

SECTOR MATH: In FY2024, USAID provided assistance to about 130 countries on well under one percent of federal spending. A peer-reviewed analysis in The Lancet estimates that USAID-supported programs prevented more than 91 million deaths between 2001 and 2021, including about 30 million children under five. The same research group now projects that continuing current funding trends adds 9.4 million deaths by 2030, and a severe scenario adds 22.6 million. Those projections are modeled scenarios rather than forecasts, and the authors say so plainly. The retrospective number is the one to sit with.

That 91 million figure has drawn methodological criticism, and I am not going to pretend otherwise. The authors themselves write that their findings offer strong support for a causal relationship without constituting definitive evidence, and at least one respected data organization declines to rely on the paper alone.¹¹ I use it because it is peer-reviewed, because the criticism is about magnitude rather than direction, and because there is no serious argument that the number is small.

Numbers still miss what was actually lost, which is much harder to restore than a budget line.

The Dependency Architecture

Here’s the honest structural analysis our sector has been reluctant to face. The USAID model created a dependency architecture that made the organizations it funded vulnerable to exactly this kind of catastrophic withdrawal. That doesn’t justify what happened. It explains why the damage is so severe.

For decades, USAID didn’t just fund programs. It shaped what civil society looked like in dozens of countries. It defined what counted as legitimate organizational capacity, what governance structures were acceptable, what outcomes were worth measuring, and what kinds of organizations were worth investing in. Organizations that aligned with USAID’s frameworks thrived. Organizations that didn’t align struggled to find other support. Over time, entire civil society ecosystems were organized around USAID’s priorities, reporting requirements, and funding cycles.

This created two structural vulnerabilities we understood in theory and never fully addressed in practice.

The first is mission drift. When a single large funder defines what counts as legitimate work, organizations gradually reshape themselves around that definition. Pragmatically, one grant cycle at a time. Over years and decades, the cumulative effect is an organizational identity that is partly the organization’s own and partly the funder’s. When the funder disappears, the organizations are left asking what they actually are without it.

The second is infrastructure fragility. Again and again, I encountered organizations with sophisticated program delivery capacity and surprisingly weak financial sustainability infrastructure. They could manage complex projects under international reporting requirements while having very little practice raising money from anyone other than USAID and its implementing partners. The localization movement of the last decade, which aimed to shift power and resources to local organizations, somewhat addressed the first vulnerability. It barely touched the second.

The numbers bear that out. USAID set a target of 25 percent of its funding going to local organizations by 2025. In 2023, the share of eligible funding reaching local partners fell to 9.6 percent, from 10.2 percent the year before.¹² On the humanitarian side, it is starker: local and national actors directly received 0.6 percent of first-level funding from Grand Bargain donors in 2023, against a 25 percent target set in 2016.¹³ Too often, localization produced organizations that looked more local while remaining financially dependent on the same external institutions. When those institutions disappeared, so did the infrastructure.

Secretary Rubio called this a “globe-spanning NGO industrial complex at taxpayer expense.”¹⁴ The framing is cynical and the political intent obvious. Buried inside it, though, is a kernel of structural truth we have been too defensive to examine: the model did create dependencies that served the implementing organizations as much as the communities they served. That’s a reason to build something different now, while there is still something to build from. It was never a reason to dismantle the system overnight without a transition plan.

Who Is Bearing the Cost

The cost of the USAID collapse is not being distributed equally, and our sector’s conversation about it has not adequately acknowledged that.

Large U.S.-based implementing organizations lost revenue and laid off staff. That’s real, and some of them did genuinely important work. They also have advantages that their local partners often do not: diversified funding relationships, institutional reserves, established donor networks, legal capacity, and the organizational infrastructure to restructure.

Independent consultants like me lost significant projects and income. I won’t minimize that. We have other clients, transferable skills, and the ability to redirect our work domestically or toward other funding streams. The disruption is significant. It is not existential, though the overnight cancellation of more than a million dollars in contracts sure made it feel that way for much of the last year.

U.S.-based nonprofits (and even U.S. farmers) that received USAID subgrants or purchase orders faced serious funding gaps. Many have had to reduce programs, lay off staff, and scramble for replacement funding. Foundation and individual donor support have helped some of them. Not all.

Among those bearing the most severe and least recoverable costs are local civil society organizations in the Global South. Those in Uganda, Moldova, Albania, Bolivia, Zimbabwe, and dozens of other countries built their organizational capacity around USAID-funded programs over decades. Many have little diversified revenue. Many have few institutional reserves. They often lack the donor relationships that allow larger organizations to redirect quickly. Many of them will not survive this. The staff they trained, the systems they built, the community trust they accumulated over the years: gone. And the communities they served don’t have a backup plan.

This is the part our conversation keeps underweighting. The organizations that will present at conferences, publish reports, and appear in press coverage about navigating the post-USAID landscape are mostly the large implementers and U.S.-based advocacy groups. The local organizations that were supposed to be the beneficiaries of a decade of localization rhetoric are, for the most part, too busy surviving to participate in the conversation about their own situation.

What Is Replacing It

For most of the last year, our sector has talked about this as an abandonment. That framing is now out of date, and holding onto it will cost organizations that need to see what is actually being built.

In September 2025, the State Department published the America First Global Health Strategy. The instrument is a bilateral memorandum of understanding signed directly with a partner government, and the stated goal is that a majority of 71 supported countries “will transition to full self-reliance during the term of the agreement.”¹⁵ As of March 2026, 27 of those agreements had been signed. By early August, the count reached 34, covering at least $24.2 billion in committed health funding through 2030, roughly 40 percent of it from partner governments.¹⁶ Kenya signed a five-year, $2.5 billion agreement in December 2025. Ethiopia followed that month, Côte d’Ivoire days later, and the Democratic Republic of Congo in February.¹⁷

So the question our sector has been asking, when is foreign assistance coming back, is the wrong one. Assistance is being rebuilt. It is being rebuilt around governments rather than around us.

Read the strategy’s own language, and the intent is not hidden. Global health assistance is described as “not just aid” but “a strategic mechanism to further our bilateral interests around the world.”¹⁵ That is the sentence to show your board.

Two things complicate the picture in ways worth knowing. The first is that NGOs are being reshuffled rather than eliminated. In March 2026, the State Department opened a platform worth up to $4.5 billion for roughly 100 projects, open to international and local NGOs, faith-based organizations, universities, and companies. In August, it announced nearly $2 billion to faith-based and community organizations, the largest such allocation in more than 20 years, including $850 million over five years to a single organization.¹⁸ Secular international NGOs are losing position. Faith-based organizations and recipient governments are gaining it.

The second is that this architecture is signed rather than running. Implementation plans were due at the end of March 2026 and most remain unpublished. The Center for Global Development’s assessment in early August was that the State Department is behind its own timeline.¹⁹ And several governments have said no. Ghana rejected a $109 million agreement over health data sharing. Zimbabwe walked away from a $367 million pact it called asymmetrical. Zambia rejected an agreement after a leaked draft showed extended U.S. access to national health data.²⁰ Kenya signed an agreement that was later suspended by its High Court over data privacy and awaits a final ruling this autumn.

The money is also less. Analysis by KFF found that combined U.S. and Global Fund funding in the agreement countries falls $4.3 billion through 2029, a 24 percent decline, with U.S. funding alone down 29 percent.²¹ Countries are being asked to move away from dependence on external support while simultaneously increasing their own spending, often without the fiscal room to do so.

Europe is not the answer either, and the numbers have moved since most of us last checked. The Global Fund’s Eighth Replenishment closed in February 2026 at $12.64 billion against an $18 billion target.²² The European Commission pledged €700 million, down from €715 million for the previous cycle.²³ France cut its main development-aid budget line by 37 percent in 2025, and its total aid fell 10.9 percent in real terms. German aid fell 17.2 percent in 2024 and a further 17.4 percent in 2025. The Netherlands is imposing what its own government calls a structural €2.4 billion reduction from 2027.²⁴ Aid from EU member states fell 9.8 percent in real terms in 2025, and aid from the EU institutions fell 13.8 percent. Across all OECD donors, the drop was 23.1 percent, the largest annual decline in the history of official development assistance.²⁵

Practically, this means the organizations that survive will be the ones that build genuine financial diversification now, while some options still exist, rather than waiting for a replacement system that arrives in a shape with no obvious seat for them.

The irony is hard to miss. A system dismantled in the name of ending dependency is being replaced by another dependency architecture. The difference is where the leverage sits. NGOs once depended on USAID awards. Under the emerging model, national governments negotiate directly with Washington over money, data, domestic spending, and, in some cases, terms they believe touch national sovereignty. Dependency did not disappear. It just moved.

The Domestic Preview

If you run a domestic nonprofit and have been reading this as an international development story, stop. You are not reading about someone else.

On January 13, 2026, the Substance Abuse and Mental Health Services Administration terminated more than 2,000 active grants worth close to $2 billion in behavioral health funding.²⁶ The money was congressionally appropriated and in many cases explicitly authorized by statute, some of it reauthorized on a bipartisan basis under the SUPPORT Act.²⁷ The stated justification was that the awards no longer aligned with agency priorities and that no corrective action was possible.

Roughly 24 hours later, the terminations were rescinded across the board. Grantees were told their awards would remain active under their original terms and to disregard the prior notice.²⁸ The Department of Health and Human Services never explained either decision. More than 60 House members and 20 senators demanded answers. As far as I can determine, none arrived.

The reversal changes what happened next. It does not change what the episode revealed: an administration was willing to terminate roughly $2 billion in behavioral health funding abruptly, affecting more than 2,000 grants, before political pressure forced a reversal within a day.

And the reversal did not undo the damage. Organizations acted on the letters. One Texas provider had already laid off 50 employees. As a group of senators put it afterward, services were paused, staff were let go or put on notice, and trust between patients and providers was undermined in ways that do not come back on a rescission notice.

Here is the part that matters most for planning, and it is not the part most people remember. The grants were honored. There was no second effort to cut them. Two weeks later, though, HHS launched an affirmative initiative under its own banner and has been awarding money under it ever since, including more than $700 million in June for mental illness, addiction, and homelessness.²⁹ Those awards carry conditions. Recipients may not use housing first approaches or the harm reduction services SAMHSA now prohibits, which since April include fentanyl test strips.³⁰ Funding did not stop. It was redirected, and the redirection carries terms. If your scenario planning only imagines the funding disappearing, you are planning for the less likely version.

The live threat is structural. The President’s FY2027 budget again proposes eliminating SAMHSA as an independent agency and folding its programs into a new Administration for a Healthy America. Three separate block grants, community mental health, substance use, and state opioid response, would merge into a single $4.5 billion Behavioral Health Innovation Block Grant.³¹ The consolidated grant is level-funded, which is the line that will get quoted back at you. What matters more is that merging three targeted streams into one flexible formula grant removes the guardrails directing money to specific purposes, while total mental and behavioral health spending falls by $576 million as other lines are eliminated.

Congress rejected the elimination for FY2026, funding SAMHSA at $7.44 billion in a bill enacted on February 3, 2026, three weeks after the termination episode. House appropriators rejected it again on June 9, 2026, though their FY2027 bill still trims SAMHSA by two percent, with substance abuse prevention absorbing the steepest cut.³² No FY2027 bill has been enacted. The House and Senate have instead passed competing continuing resolutions that would carry current funding into December.

None of this is hypothetical for the sector at large. About a third of U.S. nonprofit service providers experienced a disruption in government funding in the first half of 2025, and 21 percent lost funding outright, with agreements canceled or committed funds pulled back.³³

The pattern in international development and the pattern in domestic federal funding are the same: sudden, structural, without transition plans, aimed at organizations that built their models around government funding treated as stable. The organizations overseas now fighting for survival were built around USAID funding that looked permanent for decades. Domestic nonprofits built around SAMHSA, HHS, and other federal streams are in the same structural position. The only variable is the timeline.

What Comes Next

The honest answer is that the international development landscape will not return to what it was. The USAID architecture is gone, and its successor is being built around ministries of health and finance rather than around implementing partners. That is a different world, not a smaller version of the old one. As I said in the very first issue of The Social Prophet, "Stop planning for recovery. The thing you’re recovering to doesn’t exist anymore. Plan for what’s next."

For organizations in the Global South, the path forward is the one localization promised and rarely delivered: genuine financial autonomy, built through domestic resource mobilization, regional funding relationships, and organizational models that don’t require external validation to be legitimate. This is harder than it sounds. Domestic resource mobilization in low-income countries requires organizational capacity, donor relationship infrastructure, and time that most of these organizations don’t have right now. It is also the only path toward resilience, and organizations that start building it now will be positioned differently in five years than those that wait.

For U.S.-based international organizations, the restructuring requires genuine diversification, not the rhetorical diversification that appeared in funding strategy plans for years. That means European bilateral donors, multilateral mechanisms, private foundations with international portfolios, and, in some cases, earned revenue through consulting and technical assistance. It also means learning to work with governments as principals rather than as hosts, because that is where the money now sits. It means smaller organizations and fewer of them, focused more narrowly on what they do distinctively well. The era of the large U.S.-based international NGO doing everything everywhere on USAID money is over.

For domestic nonprofits, the lesson is to build before the crisis arrives at your door. The January termination letters were a preview. The FY2027 budget proposals are a signal. Organizations with real funding diversification, strong individual donor relationships, and financial reserves will be far better positioned to navigate the disruptions of the next three to five years than those without. The window to build that resilience is now, before the letter arrives. I wrote about this in Issue #3.

The organizations that will shape what civil society looks like on the other side of this period are the ones that treat the current disruption as a restructuring to participate in rather than a crisis to survive. That requires honesty about dependency, courage to build differently, and the willingness to let go of organizational forms designed for a world that no longer exists.

THE VERDICT: The USAID story’s real subject was never USAID. It is what happens when organizations build their existence around funding systems they mistake for permanent, and what happens next when those systems are rebuilt around something else. That lesson arrived in international development first. It arrives domestically next. The organizations that understand this are already building differently. The ones that don’t may still be surprised by something the pattern has made completely legible.

THE QUESTION

Which of your organization’s funding sources could disappear overnight without warning, and have you built anything that would survive that?

Sources