WHO, national sovereignty and who really makes the decisions when disease crosses borders.
A new respiratory virus is spreading internationally. The World Health Organization convenes experts, governments exchange information, and the WHO Director-General declares a Public Health Emergency of International Concern and subsequently determines that the event constitutes a pandemic emergency. Recommendations are issued, and countries begin introducing measures of their own: one closes schools, another does not; one introduces additional border controls, another concentrates on surveillance; vaccination begins in some countries before others; and governments make different decisions about masks, gatherings, testing, isolation and healthcare. Who is actually in charge? It sounds like a simple question. It is not.
The international system governing pandemics deliberately separates international coordination from national authority. WHO has substantial influence during a major health emergency — it can receive information, assess international risks, convene experts, declare the highest levels of international health alert, publish technical guidance, coordinate scientific activity, support countries and issue recommendations under the International Health Regulations — but what it cannot do is govern a country. The distinction matters because much of the argument surrounding international pandemic governance begins by confusing four different things: international law, WHO recommendations, international coordination, and national government decisions. They are not interchangeable, and understanding who is actually in charge requires separating them. This is the closing question of our One Health Security series on the rules of outbreaks.
Start with the obvious answer: governments
Within a country, the principal authority over public-health measures remains the state. Domestic legislation determines which institutions can exercise public-health powers, what those powers are and what legal safeguards apply; governments and legislatures determine how health services are organised; and national or subnational authorities may possess powers concerning isolation, quarantine, disease notification, business restrictions, vaccination programmes and emergency measures, depending on the constitutional system concerned. Those powers do not originate from WHO. A national government does not acquire the legal ability to close a school because WHO declares a PHEIC, nor does WHO acquire the power to close that school — the authority comes from domestic law. This is the first principle of international pandemic governance: an international emergency does not erase national sovereignty.
Then why do we need international law?
Because sovereignty does not stop infection crossing a border. Imagine two neighbouring countries: Country A identifies an unusual pathogen but delays sharing information, so Country B does not know that infected travellers may be arriving; Country B later detects cases but does not share its genomic information; and a third country introduces arbitrary restrictions that interfere with international travel while doing little to reduce transmission. Every government can claim to be acting within its sovereign territory, yet collectively the system fails. International law therefore creates rules through which sovereign states agree to cooperate — an important distinction, because international agreements do not necessarily represent the surrender of sovereignty; they can represent the exercise of sovereignty through agreed rules. States choose to accept obligations because problems extending beyond their borders cannot be solved effectively through unilateral action alone. The International Health Regulations are one example; the WHO Pandemic Agreement is intended to become another.
The IHR are legally binding
The International Health Regulations (2005) are binding international law for their States Parties, which gives them a fundamentally different status from an ordinary WHO technical guideline. States Parties have accepted obligations concerning matters including surveillance capabilities, the assessment and notification of public-health events, international cooperation and points of entry, and the Regulations also establish the mechanisms through which WHO can determine that an event constitutes a Public Health Emergency of International Concern and, under the amended IHR, a pandemic emergency. The fact that the IHR are legally binding sometimes produces a misunderstanding: people assume that if the Regulations are binding, WHO’s instructions under them must also be binding. That is not how the system works.
Binding rules can produce non-binding recommendations
When the WHO Director-General determines that a PHEIC exists, WHO can issue Temporary Recommendations under the IHR that may address appropriate public-health responses — but they are explicitly not legally binding. This distinction is important: states are legally bound by the international framework they have accepted, and within that framework WHO provides expert recommendations intended to guide their response to the particular emergency, but the recommendation itself does not become a domestic law. Governments still decide what measures they will implement, and must do so through their own legal systems. WHO therefore possesses international convening, assessment, coordination and recommendation authority; it does not possess general domestic executive authority.
WHO cannot order a lockdown
This point became so politically contested during negotiations over the 2024 IHR amendments and the Pandemic Agreement that WHO and Member States addressed it explicitly. WHO cannot impose a lockdown on a country’s population; it cannot order a country to introduce a vaccine mandate; it cannot require a government to close its borders; it cannot dictate national therapeutic or diagnostic policy; and it cannot rewrite national legislation. WHO’s explanation of the amended International Health Regulations states explicitly that the amendments give it no ability to impose health measures, including lockdowns or other restrictions, on national populations. The Pandemic Agreement goes further by writing the principle directly into the text: Article 24 states that nothing in the Agreement should be interpreted as giving the WHO Secretariat, including the Director-General, authority to direct, order, alter or prescribe national laws or policies, or to mandate measures such as vaccination requirements, travel restrictions or lockdowns. This is not merely WHO promising not to exercise a particular power; the Agreement says that the power is not being conferred.
So what does WHO actually do?
Quite a lot — because a lack of domestic coercive authority should not be confused with a lack of importance. WHO sits at the centre of the international information and coordination architecture. During an emerging outbreak it can receive notifications through the IHR system, collect information from other sources, request verification from governments, convene specialist expertise and assess whether an event presents an international threat; it can convene an IHR Emergency Committee; the Director-General can determine whether an event constitutes a PHEIC, and under the amended Regulations that the higher threshold of a pandemic emergency has been reached. WHO can issue recommendations, publish epidemiological and technical assessments, coordinate international scientific work, support countries lacking particular capabilities, and help organise access to supplies and expertise — and under the emerging Pandemic Agreement architecture it is also intended to play roles in areas including pathogen access and benefit sharing and access to pandemic-related health products. None of this is trivial: coordination can alter the course of an international emergency without requiring command authority.
Information creates another form of power
There is also a subtler source of influence, because WHO can define the international understanding of an event. When the Director-General declares a PHEIC, governments notice — and so do markets, airlines, hospitals, researchers, funders and media organisations; a pandemic-emergency determination is intended to create an even stronger global signal. The declaration does not legally command all those actors to behave in a particular way; it changes the information environment within which they make decisions. This is epistemic authority: influence arising from the ability to assemble evidence, convene expertise and produce a recognised assessment of what is happening. During an uncertain biological event, that can be enormously powerful — but it is still different from executive power.
Governments can disagree with WHO
Because national governments retain decision-making authority, responses can differ. A WHO recommendation may be considered appropriate by one country but unnecessary in another because epidemiology, health-system capacity or local conditions differ, and countries may also disagree with WHO’s assessment. This is not automatically evidence that the international system has failed, because pandemic risk is rarely identical everywhere: a small island state, a densely populated city-state and a large continental country may reasonably require different operational responses. The problem appears when national variation is driven not by legitimate contextual differences but by delayed information, political incentives, weak capacity or a failure to cooperate — so the challenge is to allow national adaptation without losing international coherence.
Sovereignty can protect good decisions — and create collective vulnerability
National sovereignty is sometimes portrayed solely as an obstacle to international pandemic response, but that is too simplistic. Local decision-making can be valuable: governments understand their healthcare systems, public-health agencies understand local epidemiology, and constitutional arrangements, risk tolerance and healthcare capacity all differ, so a measure appropriate in one country may be ineffective or disproportionate elsewhere. A central international organisation attempting to prescribe identical measures everywhere could therefore produce poor policy; subsidiarity has value.
The opposite problem is equally real, however, because an outbreak cannot always be contained within the decisions of one government. A state that fails to detect an emerging pathogen creates risk elsewhere; a government that delays notification can cost other countries valuable preparedness time; a country that cannot perform genomic surveillance may leave part of the international picture invisible; a state that restricts access to biological information may slow countermeasure development; and a government that hoards scarce health products can undermine response elsewhere and allow transmission to continue. Sovereignty therefore exists within an interconnected biological system, and a country’s decisions can create external consequences for populations that had no role in making them. International health law exists partly to manage those externalities.
The real system is shared authority
Pandemic governance therefore does not operate as a simple hierarchy running from WHO down to national governments and then to citizens. A better representation is a network: WHO provides international assessment and coordination; national governments exercise sovereign authority and implement international obligations; regional organisations may create additional layers, as the European Union does through ECDC, EWRS and EU health legislation; subnational governments and public-health agencies may possess substantial operational powers; hospitals and laboratories generate and act upon critical information; regulators approve medical products; manufacturers control production capacity; international organisations such as WOAH and the WTO govern aspects of animal health, trade and related systems; and researchers and genomic networks produce the knowledge on which decisions depend. Pandemic response is therefore a system of distributed authority, and the central governance question becomes how those authorities connect.
One Health makes the hierarchy even less realistic
Now introduce a zoonotic pathogen. The first signal comes from an animal, so the agriculture ministry becomes involved and veterinary authorities investigate while WOAH receives animal-health information; a human infection appears, and the health ministry and WHO become relevant; the pathogen is sequenced, and international genomic networks become involved; a vaccine is developed, and medicines regulators and manufacturers enter the system; animal-product restrictions appear, and trade ministries and WTO rules become relevant; and wildlife remains part of the reservoir, so environmental authorities may have a role. Who is in charge now? There still is no single answer, because the biological event crosses several systems of authority. This is why One Health governance cannot simply mean identifying a senior organisation and putting it at the top of an organisational chart; the problem is coordination across legitimate authorities. And the same problem exists inside countries, where a zoonotic threat may involve agriculture, health, environment, trade, transport, local government and emergency management, each with statutory responsibilities, different information, different databases and different thresholds for escalation. A national government can therefore be sovereign while still having difficulty governing the biological event coherently. The problem is not sovereignty; it is system integration.
Distributed control of the response
Ask who decides whether there is a pandemic emergency, and the IHR allocate a specific international decision: the WHO Director-General determines whether an event constitutes a PHEIC or pandemic emergency, taking account of the required information and the advice of an Emergency Committee. That determination has international significance, but it does not transfer control of domestic response — WHO determines the international alert; governments determine domestic action; and the effectiveness of the system depends on those two processes remaining connected, because an international alert that governments ignore achieves little, while national action taken without international information can become inconsistent or counterproductive. Ask who controls vaccines, and it is not WHO alone: vaccines may be developed by private companies, public research institutions or partnerships; manufacturing capacity belongs to particular facilities; national and regional regulators decide on authorisation; governments procure products; international mechanisms may help aggregate demand or allocate supplies; and the proposed PABS system would add further mechanisms — so no single actor controls the entire pathway. Ask who controls the border, and national governments generally exercise the relevant legal authority, subject to their domestic and international obligations, with WHO able to recommend travel-and-health measures under the IHR and WTO SPS rules relevant where animals, plants, food or related commodities are involved. Authority is distributed: a border decision may be national, the evidence supporting it international, the legal constraints treaty-based, and the biological consequences shared across several countries.
What happens when nobody agrees?
This is perhaps the most difficult scenario: WHO recommends one approach, several governments choose another, scientific advice is contested, and countries interpret incomplete evidence differently while international organisations operate within different mandates. There is no global cabinet capable of resolving every disagreement; the system depends instead on transparency, evidence, diplomacy, legal commitments and political accountability. This can appear untidy — and it is — but the alternative would require states to transfer extraordinary domestic powers to an international authority, something the current system does not do and governments have explicitly declined to do through the Pandemic Agreement. The realistic policy objective is therefore not eliminating disagreement; it is ensuring that disagreement occurs around shared evidence and functioning communication rather than because different actors cannot see what the others know.
The accountability problem
Distributed authority also creates a difficult question after an emergency: who is accountable when the response fails? WHO may have issued advice a national government ignored; or WHO may have assessed a threat incorrectly; a government may have lacked laboratory capacity; manufacturers may have been unable to increase production; international supply mechanisms may have distributed products slowly; and local authorities may have struggled to implement national policy. When authority is distributed, responsibility can become diffuse, and everyone can point somewhere else. Good pandemic governance therefore needs explicit responsibility before the emergency: who owns surveillance, who owns escalation, who decides on stockpiles, who coordinates animal–human information, who communicates with WHO, who controls emergency procurement, and who has authority to introduce particular domestic measures. The answers will differ between countries; the important thing is that they exist before the crisis.
Coordination needs to be designed, not requested
One of the least useful instructions during an emergency is that all relevant agencies should coordinate closely. Of course they should — the real preparedness question is how: which data can be shared, which organisation convenes the meeting, who has decision authority, what happens when ministries disagree, which thresholds trigger escalation, how decisions are recorded, how quickly a veterinary signal can reach public health, how international information reaches local healthcare, and how manufacturers receive reliable demand signals. Those are design questions. Coordination is infrastructure.
This does not necessarily mean creating another permanent organisation. For major cross-sector biological threats, countries could establish a predefined One Health incident governance model in which the responsible agencies retain their statutory powers, but when agreed escalation criteria are reached a temporary cross-government incident structure activates, establishes a common operating picture, assesses human, animal, environmental, genomic and supply-chain information together, and designates a lead coordinating authority — without pretending that it has absorbed every other organisation’s legal powers. The objective would be unity of understanding without artificial unity of authority. Internationally, the same principle applies: the global system does not need a world health government to improve; it needs better interoperability between the institutions that already possess legitimate authority — strong WHO–WOAH interfaces for zoonotic threats, workable arrangements between public-health and biodiversity systems for pathogen sharing, shared access between health and trade systems to changing epidemiological evidence, National IHR Authorities capable of connecting domestic institutions, exercises that test these interfaces under time pressure, and measured governance latency.
So who is actually in charge?
At global level, nobody is in charge in the way a national government is in charge within its constitutional jurisdiction. WHO coordinates, assesses, alerts, convenes, recommends and supports; states govern, legislate, implement, fund, regulate and decide which domestic measures are lawful and appropriate; other international organisations govern particular pieces of the wider system; and companies, laboratories and healthcare institutions control capabilities on which governments and international organisations may depend. Pandemic governance is therefore not a chain of command. It is a network of authority. That may sound like a weakness, but it does not have to be — networks can respond rapidly when information moves easily, responsibilities are understood and escalation pathways are designed in advance; they fail when everyone waits for somebody else. The most useful question during the next pandemic may therefore not be “who is in charge?” It may be: does everyone who is in charge of something understand how their decision connects to everyone else’s? That is a very different test of preparedness — and a much more One Health way of thinking about power.
Related One Health Security analysis
This piece is part of our Rules of Outbreaks series on global health law. It draws together The International Health Regulations, What Is a Pandemic Emergency?, The WHO Pandemic Agreement and Can You Close a Border to Disease?, and connects to Who Governs a Global Outbreak?
Correct at the time of writing (August 2026): the 2024 IHR amendments — including the pandemic-emergency category — are in force; the WHO Pandemic Agreement was adopted but had not yet entered into force, and its Article 24 sovereignty safeguard forms part of the adopted text.
Questions & Answers
Is the WHO Director-General’s PHEIC declaration legally binding on countries?
The International Health Regulations themselves are binding international law, but the Temporary Recommendations WHO issues under a PHEIC or pandemic emergency are explicitly not legally binding — governments decide what to implement through their own legal systems.
Can WHO order a country to impose a lockdown?
No. WHO’s own guidance on the amended IHR confirms it cannot impose lockdowns or other restrictions on national populations, and Article 24 of the Pandemic Agreement explicitly denies the WHO Secretariat authority to direct, order or prescribe national laws or mandate measures such as vaccination requirements, travel restrictions or lockdowns.
If WHO can’t enforce anything, why does it matter?
Because it exercises real influence through coordination and information: receiving notifications, convening expert committees, issuing recommendations, and creating an “epistemic authority” that shapes how governments, markets and institutions understand an emerging threat.
Who actually controls a country’s border during an outbreak?
National governments generally hold that legal authority, subject to WHO’s IHR recommendations and, where relevant, WTO SPS rules on trade in animals, plants and food.
Why does a zoonotic pathogen make “who’s in charge” harder to answer?
Because it crosses agriculture ministries and veterinary authorities, WOAH, health ministries and WHO, genomic networks, medicines regulators and trade bodies simultaneously — no single authority sits over the whole event.
What does the article propose instead of a single chain of command?
A predefined One Health incident-governance model in which agencies keep their statutory powers, but a temporary cross-government structure activates at agreed escalation triggers to build a shared operating picture and designate a lead coordinator — “unity of understanding without artificial unity of authority”.
References and further reading
- World Health Organization. International Health Regulations (2005), as amended — the international legal framework for surveillance, notification, assessment and response to serious cross-border public-health threats.
- World Health Organization (2024). International Health Regulations: amendments — questions and answers — confirming that the amended IHR give WHO no authority to impose lockdowns or other measures on national populations, and that Temporary Recommendations are not legally binding.
- World Health Assembly (2025). WHO Pandemic Agreement (WHA78.1) — including the Article 24 sovereignty safeguard. Adopted 20 May 2025.
Key Takeaways
- At global level nobody is "in charge" the way a national government is: WHO coordinates, assesses, alerts, convenes, recommends and supports, while states legislate, implement, fund and regulate — the IHR are binding international law, but the Temporary Recommendations issued under them are not.
- International agreements are the exercise of sovereignty through agreed rules, not its surrender: the amended IHR and the Pandemic Agreement — whose Article 24 explicitly denies WHO any power to direct national law or to mandate lockdowns, vaccination or travel bans — exist to manage the externalities one country's decisions create for others.
- Pandemic response is a network of distributed authority — WHO, national and subnational governments, regional bodies such as the EU's ECDC and EWRS, regulators, manufacturers, laboratories, WOAH and the WTO — and a zoonotic pathogen crosses all of them, so the real question is not who is in charge but whether each authority understands how its decision connects to the others.
- That makes coordination something to design, not request: pre-agreed escalation triggers, a common operating picture, a designated lead coordinator that does not pretend to absorb others' powers, National IHR Authorities that actually connect domestic institutions, and measured governance latency — "unity of understanding without artificial unity of authority".