Article Text

Download PDFPDF

Love thy neighbour? Allocating vaccines in a world of competing obligations
  1. Kyle Ferguson,
  2. Arthur Caplan
  1. Division of Medical Ethics, New York University Grossman School of Medicine, New York, New York, USA
  1. Correspondence to Dr Kyle Ferguson, Division of Medical Ethics, New York University School of Medicine, New York, NY 10016-6402, USA; kyle.ferguson{at}


Although a safe, effective, and licensed coronavirus vaccine does not yet exist, there is already controversy over how it ought to be allocated. Justice is clearly at stake, but it is unclear what justice requires in the international distribution of a scarce vaccine during a pandemic. Many are condemning ‘vaccine nationalism’ as an obstacle to equitable global distribution. We argue that limited national partiality in allocating vaccines will be a component of justice rather than an obstacle to it. For there are role-based and community-embedded responsibilities to take care of one’s own, which constitute legitimate moral reasons for some identity-related prioritisation. Furthermore, a good form of vaccine nationalism prioritises one’s own without denying or ignoring duties derived from a principle of equal worth, according to which all persons, regardless of citizenship or identity, equally deserve vaccine-induced protection from COVID-19. Rather than dismissing nationalism as a tragic obstacle, it is necessary to acknowledge that a limited form of it is valuable and expresses moral commitments. Only then can one understand our world of competing obligations, a world where cosmopolitan duties of benevolence sometimes conflict with special obligations of community membership. Once these competing obligations are recognised as such, we can begin the work of designing sound ethical frameworks for achieving justice in the global distribution of a coronavirus vaccine and developing practical strategies for avoiding, mitigating or resolving conflicts of duty.

  • ethics
  • philosophical ethics
  • distributive justice
  • international affairs

This article is made freely available for use in accordance with BMJ’s website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.
View Full Text

Statistics from

Many writers are condemning what they call ‘vaccine nationalism’ as an obstacle to justice in the future distribution of a safe and effective coronavirus vaccine. We argue that, under the right conditions and subject to important limits, allocating a vaccine in a nationally self-interested way will be a component of justice, not an obstacle to it. A fair allocation scheme ought to reflect the fact that there are role-based and community-embedded obligations to take care of one’s own. Although there are global justice–related limits to prioritising on the basis of such associative ties, these role-based, relational and communitarian responsibilities must be acknowledged rather than dismissed out of hand. Once these responsibilities are recognised as such, the conflict between vaccine nationalism and vaccine cosmopolitanism can be seen for what it is: not a conflict between obligations and obstacles, but rather a conflict of duties.

In the growing discourse about justice and how it is at stake in the future distribution of a coronavirus vaccine, there are two kinds of voice. One voice is prescriptive. It commands actions, assigns responsibilities and wishfully works out a plan to achieve the goals it envisions. The other voice is descriptive. It neutrally recounts the past and disinterestedly observes the present. It predicts what may happen in the future, and with great reportorial objectivity explains why those forecasted events might unfold. While the prescriptive voice speaks of what we ought to do, the descriptive voice speaks of what we will do. The prescriptive voice offers reasons; the descriptive voice, causes. It is as if the two voices answer different questions: ‘Who should get the vaccine first?’ and ‘Who will get the vaccine first?’ From the prescriptive voice, we hear that equal global access to a coronavirus vaccine is an obligation of justice; from the descriptive voice, that nationalism is brute-fact obstacle to justice. The two voices conspire to tell us that vaccine cosmopolitanism is the best of all possible worlds and that vaccine nationalism is our selfish world’s tragic reality.

By ‘vaccine cosmopolitanism,’ we mean a view of distributive justice for vaccines, according to which community membership—in particular, citizenship or belonging to a nation—is ethically irrelevant. On this view, justice demands that vaccine allocation schemes disregard potential recipients’ national identities and associative ties, and instead identify other allocation criteria. According to the emerging consensus regarding domestic allocation, just distributions of initial supply prioritise high-risk healthcare workers and individuals whose medical conditions that put them at high risk of morbidity and mortality due to COVID-19.1 Vaccine cosmopolitanism scales up this type of allocation framework so that it encompasses the global population. Once scaled up, the global allocation framework disregards national identities. The guiding intuition in this grand analogy is that small-scale, nation-sized justice provides a useful, miniature image of global justice. It begins with a vision of justice in a smaller-scope community (e.g., a nation-state), and extends that vision to the global community. The WHO SAGE Values Framework, for example, calls for distributing vaccines to countries according to their number of healthcare workers, the proportion of their populations over 65 years old, and the proportion of its population with comorbidities.2 Here, allocation criteria are the same as those in domestic allocation schemes, but the whole world is in view.

It is perhaps natural to think of global justice for vaccine allocation in this way: Since national identity is medically irrelevant—that is, this trait does not cleanly or systematically map onto triage-relevant traits like severity of risk, urgency of need and efficacy of allocation—any allocation scheme that makes it a relevant factor leads to arbitrary and therefore unjust distributions. Thus, vaccine cosmopolitanism, as we have characterised it, sees injustice in the rush of activity by various states—high-income countries (HICs) like the USA, Canada, the UK, Italy, the United Arab Emirates or collectives of European Union countries—to secure the initial supply for their own citizens. Those countries are making relevant something that justice ought to make irrelevant.

Those who use the term ‘vaccine nationalism’ use it in only a pejorative and accusatory sense. Vernon Silver describes it as ‘the jockeying of governments to secure doses of promising candidates for their citizens.’3 The ‘jockeying’ takes many forms: investments in research, advance-purchase agreements, production deals and even spying and hacking. In ‘The Tragedy of Vaccine Nationalism,’ Bollyky and Brown describe it as ‘a “my country first” approach to allocation.’4 They write:

Absent an international, enforceable commitment to distribute vaccines globally in an equitable and rational way, leaders will instead prioritize taking care of their own populations over slowing the spread of COVID-19 elsewhere or helping protect essential health-care workers and highly vulnerable populations in other countries.4

Yamey et al. avoid using the term, but have the same worries and cite recent history as a cause for concern:

The need for COVID-19 vaccines is global, although the need is differentially distributed within populations.… High-income countries must not monopolise the global supply of COVID-19 vaccines. The risk is real: during the 2009 influenza A/H1N1 pandemic, rich countries negotiated large advance orders for the vaccine, crowding out poor countries. Such an outcome would result in a suboptimal allocation of an initially scarce resource.5

No matter the exact definition, the offence lies in vaccine nationalism’s spirit: competition, contest and capital. Vaccine cosmopolitanism, on the other hand, has the virtues of coordination, cooperation and solidarity. Given that vices and virtues so neatly align on opposite sides of the cosmopolitanism–nationalism divide, justice would stand against jockeying. A just approach to allocation must have the whole globe in view and require that national identity and the associative ties of citizenship be ignored.

In his article, ‘“Vaccine nationalism” threatens global plan to distribute COVID-19 shots fairly,’ Kupferschmidt explains how some countries’ desires ‘to protect their own citizens’ may lead them away from fully participating in the COVAX Facility, an effort led by Gavi, the Coalition for Epidemic Preparedness Innovations, and the WHO, aimed at global procurement of coronavirus vaccine candidates.6 It is worth noting three things about the COVAX Facility.i First, despite its cosmopolitan view of justice (i.e., that fair distributions are based on need and not on citizenship), the relevant actors in the COVAX Facility are nation-states. Our world is filled with Westphalian agents: independent nation-states with exclusive sovereignty over their territories. This is not to say that our world only contains nation-states. For there are other kinds of agent: transnational regulatory networks, international courts, WHO, the WTO. Despite this diversity, the nation-state framework tends to dominate as evidenced by the COVAX Facility’s signatories. Nation-states, acknowledging responsibilities to take care of their own, are the agents that will determine what global distribution of the vaccine looks like. Even those who advocate vaccine cosmopolitanism and denounce vaccine nationalism see the main characters in the story as nation-states or collections of them. Second, as they condemn nationalism, and with it the responsibility to protect one’s own, these writers fail to appreciate that nationalism drives countries to join the COVAX Facility in the first place. That is, joining is itself an expression of nationalism: HICs self-interestedly avoid the risk of betting on the wrong vaccine, andlow- and middle-income countries (LMICs) self-interestedly secure supplies for their own populations. Third, the agreement concerns access in the long run, not how to allocate initial supplies. Thus, the COVAX members’ populations vastly outstrip the supply that will be distributed in the first wave of potential allocations.

The reality is that even if we limit eligibility to the global population of essential healthcare workers and the most vulnerable, initial supply will be so scarce that the global demand cannot be met. Given this scarcity, and given that the goals of vaccine cosmopolitanism cannot be accomplished with such low supply, we need a view of how the initial supply should be distributed. Our question is: Would it be unjust to allocate in a way that reflects leaders’ commitments to their own citizens?

Critics of vaccine nationalism would think so. They would find a distribution unjust if it included, say, American essential healthcare workers and excluded their non-American counterparts. On their view, justice would demand that we flip a coin or establish a lottery system to decide which essential healthcare workers receive the scare resource. In the absence of any compelling reasons to prioritise some would-be recipients over others in the global population of essential healthcare workers, the allocation decision should be made arbitrarily. Or, via reductio ad absurdum, perhaps no allocation should be made. If the guiding principle is ‘Some for all, not all for some,’ then perhaps there should be none for any until manufacturers produce enough supply to meet the needs of first-phase recipients worldwide. The cosmopolitan view seems to imply that it is unjust for any country to protect its first-phase population until there are enough vaccines to be distributed globally to all who satisfy first-phase allocation criteria. Thus, perhaps countries should wait for supply to increase until impartial, global distribution becomes possible, rather than protecting their own first-phase populations when they can. This is a standard which cosmopolitanism may imply but which we reject.

There are compelling moral reasons to think that limited national partiality is justified. These reasons have to do with the associative ties of community and the roles occupied within communities.

A community is a set of individuals who think of one another in terms of ‘we’ and as each being ‘one of us.’7 8 Each person belongs to multiple communities, resulting in nested and overlapping identities. Nested identities allow for one to be a philosopher who is a moral philosopher. Overlapping identities allow for one to be a philosopher and a Democrat. Perhaps more familiarly, one can simultaneously be a father and a son or a mother and a daughter, an employee and a colleague, a member of a club and a citizen of a nation-state. Each of these identities is a matter of community membership, standing in relations to comembers and occupying roles within those webs of relations. One thing that cosmopolitanism stresses is that in addition to the aforementioned smaller-scope communities, each of us belongs to a global community. We agree. But it does not follow that cosmopolitan obligations automatically outweigh obligations one bears by virtue of one’s membership to other, smaller-scope communities. Our claim is that within a nation-state, there are legitimate moral reasons to procure and allocate vaccines in a self-interested manner.

A just distribution of the vaccine’s initial supply can reflect these associative ties and moral commitments to one’s community and its members. That is, there is a kind of vaccine nationalism that leads to justice, not away from it.ii But to see why, one must first acknowledge the moral weight of commitments to protect one’s own. At the core of the right kind of vaccine nationalism is this type of special obligation.

In general, belonging to a community brings both benefits and burdens. Such memberships and relationships are familiar features of moral experience. As moral agents, we regularly occupy roles, stand in relations to one another, and belong to communities—all of which create moral reasons for us to act in the interest of particular others rather than any others. In the context of the coronavirus vaccine, this means that some will have the privilege, not the right, to receive the resource. It also means that some will have the primary obligation, though not their only obligation, to privilege their own. Just distributions of the initial supply of a coronavirus vaccine can and should reflect these special relations and associative ties.

In acknowledging this familiar fact of moral life, we depart from the family of ethical theories known as utilitarianism. Our view is that there are genuine, agent-relative reasons found in the roles decision makers occupy in their communities. There are special obligations, obligations that are arise from one’s belonging to a community and occupying certain roles within it. Utilitarians reject this idea. Singer, for instance, writes:

No doubt we do instinctively prefer to help those who are close to us. Few could stand by and watch a child drown; many can ignore the avoidable deaths of children in Africa or India. The question, however, is not what we usually do, but what we ought to do, and it is difficult to see any sound moral justification for the view that distance, or community membership, makes a crucial difference to our obligations (p.202-3).9

Singer thinks that helping members of our community is merely a matter of instinctual preference or natural biological affinity. We think it is a matter of obligation, that we have good moral reasons to do so. And those reasons consist in or are derived from the associative ties that bind individuals living in a community. This does not mean that we can or ought to ignore the lives and welfare of those outside our community. We continue to have obligations to them. But in addition to those, we have obligations to our communities and their members. Utilitarianism can indeed help us discern some of our obligations. But it is not the only way discovering what we ought to do. Its analyses fail to capture all dimensions of moral life. In the context of distributive justice, it omits the ethical relevance of community membership, the commitments members have to each other, and the special obligations arising from occupying roles within them.

As we dismiss the view that utilitarianism is the only source of moral obligation, we align ourselves with a constellation of figures in the history of ethics. One such figure is Ross, who saw that utilitarian thinking ‘oversimplifies the moral life (p.189).’10 Ross reminded us that when we engage in moral deliberation, we not only consider future consequences, but we also weigh heavily the promises we made and commitments we undertook in the past.11 (p. 17) Ross also objected to utilitarianism for suggesting that the only morally relevant relation ‘in which my neighbours stand to me is that of being possible beneficiaries of my action (p.19).’11 Those with whom we share communities are much more than identity-lacking receptacles or empty vessels for the good we might give them. What we owe to each other often depends on the associative ties the define our relation.

In addition to the general obligations or duties ‘which we owe to all men alike,’ there are ‘special responsibilities that we have undertaken to particular men (p.27).’11 I might owe beneficence to any and all, but only owe reparation or gratitude to some. The latter depends on the particular relations we have to one another.iii Our claim is that community membership and intracommunal roles create special obligations. These, in turn, can justify partial allocations of vaccines or at least elevate those considerations to the status of moral reasons. National leaders have made promises or undertaken commitments to protect their own.iv These concerns can and should drive their rush to procure promising candidates for use by their citizens.

At the core of vaccine cosmopolitanism is the commitment to the equal worth of all persons regardless of their identities or community memberships. This insight is expressed by Singer in the passage above and by classical utilitarians like Jeremy Bentham and John Stuart Mill. Utilitarian thinking demands impartiality in weighing at-stake interests. As Mill put it, ‘… one person’s happiness, supposed equal in degree …, is counted for exactly as much as another’s (p.98).’12 Utilitarianism’s connection to justice can be captured by the dictum Mill attributed to Bentham: ‘everybody to count for one, nobody for more than one (p.99).’12

It is important to see that we can make our point about nationally self-interested allocations without denying the principle of equality, according to which all persons are equally worthy or deserving of our benevolence. In general, people do not always get what they are worthy of. Sometimes this is because one’s worthiness does not create an obligation in others; other times, that worthiness creates an obligation, but one that is outweighed by some other obligation in others. Individuals rarely receive all that they are owed simply because we owe each other a great deal but can only do so much. We agree that all persons equally deserve health, protection from coronavirus, and the benefits these goods would bring in tow. But we also assert that there is an obligation to prioritise one’s own. This means that there are moral reasons both to allocate the vaccine in a nationally self-interested way and to pursue its global distribution. The former is a special obligation; the latter, a general obligation. When these obligations point in opposite directions, we encounter a conflict of duties. The real ethical challenge is deciding how best to resolve those conflicts or designing ways to achieve both. It is not to deny their existence.

It is helpful to distinguish between three types of vaccine nationalism: the good, the blind and the ugly. Ugly vaccine nationalism denies the equal worth of persons, saying that the lives and interests of one’s own citizens are the only valuable ones or are always more valuable than those of others. Here, the most trivial interest of one’s own is worth more than the lives and vital interests of outsiders, even large numbers of them. Blind vaccine nationalism accepts the equal worth of persons, but never acknowledges duties directed at those beyond one’s own borders. The blind vaccine nationalist sees outsiders as worthy of the vaccine, but never sees himself as responsible for progress towards that that end. Good vaccine nationalism endorses the equal worth of persons and recognises obligations to persons and communities globally. The ugly reject cosmopolitan responsibilities, the blind fail to see them, and the good recognise and appreciate them. But good nationalists also feel the weight of reasons to take care of their own, even when that means agonising while awaiting the propitious moment to realise their cosmopolitan responsibilities.

When does that moment come? In the vaccine context, assuming some HICs discover or purchase all initial supplies of a safe and effective vaccine, would it be fair for them to consume the resource internally to protect every one of their citizens before sharing it with the rest of the world? We do not pretend to know where the limits of acceptable national partiality lie. But there is, to be sure, a limit to justified nationalism. Our main point is simply that there is a good form of vaccine nationalism, one that sees a real conflict between the cosmopolitan duties of benevolence, where one owes the good to others because the others are persons, and the special obligations of membership to specific communities.

The most difficult ethical challenges concern balancing these two sets of responsibilities and delineating the moral limits of acceptable partiality. Marking the limit will be complicated and difficult. It will likely turn on empirical matters that we are only just beginning to understand. Justice will not be achieved by a simple algorithm, and its demands will evolve as supply increases and the pandemic’s grip on the world changes shape.

We have argued that limited national partiality in allocating a coronavirus vaccine can be justified on grounds of the associative ties and special obligations of community membership. We have not tried to locate the limits of good vaccine nationalism. Nor have we claimed how best to balance competing obligations. Our claim is simply that there is a balancing act to perform and that on both sides there are moral reasons. Unfortunately, the current stilted discourse surrounding coronavirus-vaccine justice is not presenting the challenge in this way. This is because vaccine nationalism’s most vocal critics fail to appreciate the relevance of associative ties and the moral responsibilities arising from them. We would do well to settle into a position of pluralism, where many ethical frames of reference are recognised as valid and regularly employed. But such settling in requires getting more comfortable with conflicts of duty that inevitably arise from taking multiple perspectives at once.

We have not suggested how to resolve these conflicts of duty. Instead, we have argued that they must be recognised as such. The right kind of vaccine nationalism gives moral reasons for prioritising one’s own, and just distributions of the vaccine ought to reflect these commitments and their limits.


The authors thank two anonymous reviewers for the time and effort they devoted to their careful, critical readings of our earlier manuscript and for their valuable feedback.


View Abstract


  • Contributors KF contributed to the conception and design of the work, reviewing and interpreting the literature, drafting the article and revising the article. AC contributed to the conception and design of the work, critical revision of the article and gave final approval of the version to be published.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • The editors of The Lancet recently commended but raised important questions about the COVAX Facility.13

  • Ezekiel Emmanuel and colleagues discuss possible ethical justifications for national partiality and make room for those values in their Fair Priority Model, a proposal for fair global distribution of a coronavirus vaccine.14 In their model, the COVAX Facility is valuable tool for global justice and limited national partiality is identified as a possibly legitimate value rather than dismissed as a tragic obstacle.

  • In addition to Ross’s point about special obligations, we are gesturing towards Bradley on role-based obligations in ‘My Station and Its Duties’15 and even Socrates, who in the Crito connects his obedience to Athens to his gratitude for the benefits of citizenship. For an overview of the literature on special obligations, see Jeske.16

  • Who counts as ‘one’s own’? It is important to realise that associative ties exist between nation-states as well as within them. In addition to the transnational associative ties that create a global community, which we acknowledged above, one country might prioritise its allies over other countries, or one country might prioritise those it once colonised over those it never wronged. Throughout this article, we are thinking of ‘one’s own’ in terms of national citizenship, and we are claiming that the associative ties of citizenship can justify limited national partiality in allocating vaccines. But we also acknowledge that there are noncosmopolitan, international associative ties that might justify partiality by one nation-state towards another. We thank Henry Richardson drawing our attention to these complexities.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.