On 19 September 2025, a set of amendments to the International Health Regulations quietly entered into force. They add a new alert level above the familiar public health emergency of international concern: a “pandemic emergency,” for an outbreak that is spreading widely across countries, overwhelming health systems and causing major social and economic disruption.
Four months earlier, on 20 May, the World Health Assembly adopted the WHO Pandemic Agreement with 124 votes in favour, none against and 11 abstentions, after three years of negotiation.
On paper, the world is better prepared than it was in January 2020. It has a new legal tier of alarm, a treaty text, and a commitment that manufacturers participating in a new pathogen-sharing system will make a share of their real-time production available to WHO.
And yet anyone who has sat in a health ministry since 2023 knows the mood has changed. The emergency budgets are gone. The surge staff have been let go. The stockpile contracts are up for renewal and nobody is fighting for them.
My argument is that this is not random. Preparedness has kept the things that cost little to maintain, like laws, frameworks and signatures, and lost the things that need a budget line every single year. That pattern is predictable, and it should shape where advocates spend their energy.
What stuck: rules that don’t need a line item
Legal and institutional changes are sticky because, once made, they persist without annual renewal. The IHR amendments were adopted by consensus in June 2024 and are now binding on most countries. They create national IHR authorities and a clearer escalation ladder. Those will outlast any particular minister.
The Pandemic Agreement is stickier still in symbolic terms, but it is not finished. Its central bargain, the Pathogen Access and Benefit-Sharing annex that governs how countries share samples and genetic data in exchange for access to vaccines and treatments, was left for further negotiation. The agreement opens for signature only once that annex is adopted, and enters into force after 60 ratifications.
So even the most visible success is a frame waiting for its most contested part. Signing a treaty is cheap. Writing the annex that decides who gets vaccines first is where the real fight has been deferred to.
What faded: anything that needs recurring money
The OECD’s data on health spending tells the other half of the story. At the peak of the pandemic in 2021, almost 10% of public health spending across OECD countries was directly related to COVID-19, and prevention’s share of health spending roughly doubled compared with 2019. By the OECD’s end-2024 review, prevention had fallen back to around its pre-pandemic share.
Some of that fall is healthy. Mass testing and vaccination sites were never meant to be permanent. But the collapse took with it things that were meant to last: surveillance teams hired on short contracts, laboratory capacity funded by emergency appropriations, public health workforces that were expanded and then allowed to shrink by attrition.
The reason is structural. A treaty provision costs nothing to keep in force. A genomic sequencing lab costs money every year, and in a year with no outbreak it looks like an idle asset. Finance ministries are designed to find idle assets.
The geopolitical shock
One more thing changed that no planning document anticipated. On 20 January 2025, the United States issued an executive order to withdraw from the WHO, pause US funding to it, recall US personnel and stop participating in negotiations on the pandemic agreement.
Whatever one thinks of that decision, it exposed how much of the global preparedness system depended on one country’s financing and technical staff. It also meant that the agreement adopted in May was negotiated and adopted without the participation of a country that is home to several of the world’s major vaccine makers and was, until this year, the WHO’s largest state funder. The legal architecture is broader than ever. Its financial base is narrower.
The uncomfortable incentive
Preparedness has a political problem that no treaty solves. Success is invisible. A pandemic that is caught early looks, in hindsight, like a scare that never needed the money. A lab that sits ready for five years looks wasteful for all five of them.
Politicians are rewarded for responding to crises, not for preventing them. That is why the parts of preparedness that survived are the ones that let governments look responsible without spending: signing, amending, committing.
The mechanism: preparedness has no natural constituency
Most public spending is defended by someone who benefits from it every day. Hospitals have patients and staff. Schools have parents. Roads have drivers and construction firms. Each budget line has a group that notices when it is cut and complains.
Preparedness is different. Its benefits are probabilistic and delayed. In a normal year, a well-prepared surveillance system produces reports that nobody outside a small technical community reads. A stockpile sits in a warehouse. An emergency operations centre runs exercises. Nobody experiences these things as a service. The people who will benefit most, those who would have died in the next outbreak, do not know who they are.
That is why preparedness gets funded in waves. After an outbreak, the political cost of being unprepared is vivid, and money flows. As memory fades, the absence of any visible benefit makes the spending look like waste, and it drifts down. Public health researchers have described this for decades as a cycle of panic and neglect. COVID-19 did not break the cycle. It simply made the panic phase larger.
A scenario: the lab that looked idle
Consider a national public health laboratory that, in 2020 and 2021, received emergency funding to sequence virus samples. It hired bioinformaticians, bought sequencing machines, trained staff and built links with hospitals to send samples in quickly. At the peak, it was processing samples every day and informing decisions about restrictions and vaccines.
By 2023, the emergency funding had ended. The machines were still there. The staff were on contracts that were not renewed, because there was no recurring budget line. Some moved to universities or private firms that paid more. The sample-sharing links with hospitals lapsed, because nobody was paid to maintain them.
On paper, the country still has sequencing capacity: the equipment is on the asset register. In practice, it would take months to rebuild the team and the networks needed to use that equipment at scale in a new outbreak. That gap, between capacity on paper and capacity in practice, is what the next pandemic will expose.
The solution in this case is not to keep the lab running at pandemic levels forever. It is to give it a routine job, such as sequencing for antimicrobial resistance, foodborne outbreaks or influenza, that keeps the team employed and the networks active, and that can be scaled up quickly when needed.
The strongest counterargument
A reasonable finance official would push back on my argument. Much of the pandemic spending was never meant to last. Mass testing centres, emergency field hospitals and temporary staff contracts were surge measures. Keeping them would have been wasteful. A good preparedness system does not maintain pandemic-level capacity at all times; it maintains the ability to scale up quickly.
This is right, and it is the core of good preparedness design. The model that makes sense is a “warm base”: a smaller standing capacity doing useful routine work, plus contracts, plans and trained reserves that can expand quickly. That is cheaper than permanent surge capacity and more robust than nothing.
The problem is that the warm base is itself what is being cut. It is not the field hospitals that have vanished, which is fine. It is the core surveillance staff, the laboratory networks, the logistics contracts and the trained reserves, which are exactly the things the warm base model depends on. Those are small budget lines, and they are the easiest to trim without anybody noticing.
Second-order effects
The fading of preparedness has consequences beyond the next outbreak.
Trust erodes. Public health institutions that made unpopular decisions during COVID-19 are, in many countries, less trusted than before. Those institutions now face budget pressure at the same time as they need to rebuild credibility. That combination makes it harder to argue for resources, and harder to persuade the public to follow advice when the next threat arrives.
Workforce knowledge disappears. The people who ran contact tracing, vaccine logistics and emergency operations learned lessons the hard way. If they leave the public sector, those lessons leave with them. After-action reviews help, but they are not a substitute for people who have done the job.
International coordination weakens. The US withdrawal from the WHO removes a major funder and a large pool of technical expertise from the multilateral system. Other countries and philanthropies may partly fill the gap, but the result is a system with more legal commitments and fewer resources to carry them out.
The benefit-sharing negotiation is also exposed to this cycle. The longer it takes to finalise the annex, the more the political urgency fades, and the more likely it is that the final text reflects the interests of countries with manufacturing capacity rather than those who need access.
What this means for each actor
- For health ministries: identify the standing capacities that the next response will need and give each of them a routine, everyday job. Capacity with a daily purpose is much harder to cut.
- For finance ministries: treat core preparedness as insurance. Insurance premiums are paid in years without claims. That is the point.
- For funders: avoid funding preparedness only through time-limited emergency grants. Support transition to domestic budgets from the start.
- For clinicians and hospital leaders: keep emergency plans, stock and training current. The hospital-level warm base matters as much as the national one.
What to watch next
The key milestones over the next year are the negotiation of the benefit-sharing annex, due to be considered at the next World Health Assembly; the pace of ratification once the agreement opens for signature; and how countries implement the amended IHR, particularly whether they designate and fund national IHR authorities. Just as important, and far less visible, are the next budgets for surveillance, laboratories and public health staff. Those numbers will say more about preparedness than any treaty text.
Where the effort should go now
If this diagnosis is right, the advocacy agenda should shift from new frameworks to protecting recurring capacity:
- Tie preparedness funding to existing services. Surveillance labs that also do routine diagnostics and antimicrobial resistance testing are harder to cut than labs that only wait.
- Fund people, not only equipment. Machines bought in 2021 are useless without the staff to run them in 2027.
- Finish the annex before memory fades further. The benefit-sharing terms are the part developing countries care most about. Delay favours those who already have manufacturing capacity.
- Report preparedness spending separately. If it is not visible in budgets, it will be cut without anyone noticing.
The world did learn from COVID. It learned how to write better rules. It has not learned how to keep paying for the dull, standing capacity that makes those rules mean anything when the next outbreak arrives.
What piece of pandemic-era capacity in your country has already disappeared, and could it have been saved?
