The Ethics of Using Vaccination Status as a Rationing Criterion: Luck Egalitarianism and Discrimination

This article refers to:

Rationing, Responsibility, and Vaccination during COVID-19: A Conceptual Map

In “Rationing, Responsibility, and Vaccination during COVID-19: A Conceptual Map,” Park and Davies interestingly lay out the discussion of employing vaccination status as a rationing criterion in times of emergency (Park and Davies 2024). In what follows, we first address whether rationing based on vaccination status should be understood as directly and/or indirectly discriminatory. Then, we consider some further luck egalitarian perspectives for the policies’ potential implementation.

Luck egalitarianism is a prominent theory about distributive justice. It holds that distributions are just, if and only if, they reflect people’s relative exercise of responsibility as opposed to differential luck. The theory has been extensively debated in the context of healthcare rationing. As Park and Davies’ discussion suggests, there seemingly is a straightforward implication for the question of rationing based on vaccine status. If an individual chooses not to get vaccinated and later becomes critically ill, then luck egalitarianism might suggest that society should ration treatment based on the individual’s past choice. This may be because the disadvantage incurred reflects the exercise of responsibility and thus limits the person’s claim to healthcare resources. Alternatively, if an unvaccinated person further spreads the disease, one could say that they create bad luck for others and should be asked to mitigate that (Thaysen and Albertsen 2017).

The first consideration we address is whether rationing based on vaccine status should be perceived as directly or indirectly discriminatory. Kasper Lippert-Rasmussen has proposed a generic definition of direct group discrimination. On this, an act or policy is directly discriminatory when it treats certain individuals worse than others because they are (or are perceived to be) members of a particular socially salient group (Lippert-Rasmussen 2013, 15). If vaccination status was used as a rationing criterion, then the unvaccinated would be treated worse than others because they belong to this particular group. The question is then whether the unvaccinated form a socially salient group.

Lippert-Rasmussen suggests that a “group is socially salient if perceived membership of it is important to the structure of social interactions across a wide range of social contexts” (Lippert-Rasmussen 2013, 30). Whether a group is socially salient is a matter of degree. It may “be anything from slightly important to wholly dominant in the structure of social interactions, and it may be important in more or fewer social contexts” (Lippert-Rasmussen 2013, 31). The emphasis on perceived membership means that even if people (often) do not know the vaccination status of the people they interact with, being unvaccinated would still be socially salient if perceived membership would structure interactions.

While the social salience of vaccine status may differ between vaccines and countries, it seems uncontroversial to say that vaccination status during the COVID-19 pandemic was a socially salient feature. Many implemented policies increased the importance of that feature for social interactions. Thus, being (perceived as) unvaccinated affected what you could do and how others would treat you. Vaccination status as a rationing criterion would only further manifest that. Thus, vaccine-sensitive rationing policies should be considered directly discriminatory under the employed definition of discrimination.

Similarly, we should explore whether such policies constitute indirect discrimination. Roughly put, indirect discrimination occurs when an allegedly neutral policy has a disparate and disadvantageous impact on members of a socially salient group (Lippert-Rasmussen 2013, 72). This seems to be the case. Park and Davies highlight how vaccine status might reflect broader preexisting inequalities. Rationing based on vaccine status might, therefore, also indirectly discriminate against some socially salient groups and communities.

However, the employed definitions are non-moralized. Saying that something is direct or indirect discrimination under such a definition does not imply that is morally objectionable. Luck egalitarianism provides a responsibility-sensitive approach to assessing when (if at all) such direct and indirect discrimination is wrong (Albertsen 2023b; Albertsen and Tsiakiri 2023). From this perspective, the above discrimination has several problematic features. They spring from the different ways a rationing policy sensitive to vaccination status may fail to be a responsibility-sensitive policy.

A crucial example of when vaccine sensitivity and responsibility sensitivity diverge is instances where the former is too inclusive regarding who it disadvantages. Under realistic assumptions, rationing based on vaccine status would disadvantage some who are not genuinely responsible for remaining unvaccinated. This is the case because unjust socio-economic inequalities may make it easier for some than others to get vaccinated and make us hesitant in deeming these responsible for their vaccination status—as Park and Davies point out (Park and Davies 2024). This would make some of the direct discrimination unjust because it disadvantages some who have not genuinely chosen to stay unvaccinated. It also explains why we might worry about the described indirect discrimination.

Even if we could apply a vaccine-sensitive policy only to those who genuinely chose not to get vaccinated, there might still be a course for luck egalitarian concern. The luck egalitarian concern with the unjust (economic) inequalities may also deem rationing based on vaccination status problematic because such a policy is inadequately attuned to the effects of preexistent social and natural inequalities beyond how they affect vaccination decisions. For COVID-19, our vulnerability in terms of getting the disease depends on natural or social factors beyond our control, and so does our risk profile if we get sick (Nielsen and Albertsen 2023). A truly fairness-based allocation of treatment would need to take such disparities into account as well (Albertsen 2023a).

Note, however, that for these verdicts, it matters whether luck egalitarianism in health is approached from an integrationist or an isolationist perspective. On the latter, we apply luck egalitarianism to health without considering distribution in other spheres (apart from how they influence responsibility for health choices). On the former, however, we care about whether people’s overall relative advantage reflects their exercises of responsibility across different spheres (Albertsen and Knight 2015). That may create reasons to refrain from disadvantaging some who are responsible for not being vaccinated if doing so would disadvantage people who are disadvantaged overall for reasons reflecting bad luck (Albertsen 2020).

In addition to this, as Park and Davies also invite us to consider elsewhere in the paper, other decisions, apart from getting vaccinated, determine people’s likelihood of becoming ill (Park and Davies 2024). Therefore, vaccination status might indeed be a highly indicative factor for that, but focusing exclusively on that may be insufficiently choice-sensitive.

Building on the above, under realistic circumstances, luck egalitarianism could not recommend implementing a policy that treats vaccine status as a sole criterion for denying treatment across the board. The main reason for this is that vaccine status would not track exercise of responsibility. A purer luck egalitarian approach would require efforts first to determine the genuineness of one’s choice to remain unvaccinated and insufficient efforts to protect oneself, and secondly, to protect the already disadvantaged from becoming even worse off. This would, arguably, leave room for a principled case for (some) employment of vaccine status as a rationing criterion—but also creates complex tasks in clarifying who fits into which category. One plausible solution would be to create a system of exemptions under which people could be exempted from the employment of vaccine status as a rationing criterion for reasons like one’s low socio-economic status and other coexisting medical conditions.

Assuming we could identify those who have genuinely chosen to stay unvaccinated, what are the appropriate luck egalitarian responses? Apart from denying treatment, other options would be co-payment or giving slightly lower priority in the assignment of ICU beds. These measures may be insufficient under extreme scarcity but remain responsibility-sensitive alternatives. One alternative mentioned by Park and Davies (Park and Davies 2024), and discussed during the COVID-19 pandemic, is to let the unvaccinated pay a fine for putting others at risk. Such a policy would be more attuned to the all-luck egalitarian approach. On this, everyone is owed the expected (moral) value of their choices (Knight 2021). Thus, those whose choices may disadvantage others might be asked to contribute.

Based on the above, we can say that employing vaccine status as a rationing criterion can be considered both direct discrimination against the unvaccinated and indirect discrimination against other socially salient groups. Such discrimination cannot be justified across the board from a luck egalitarian perspective because a vaccine-sensitive distribution is not necessarily a responsibility-sensitive one. In particular, the former is very likely to be influenced by social and natural circumstances and includes a too narrow conception of effort.

Furthermore, it may disadvantage some whose overall position reflects bad luck, even if their choice to stay unvaccinated was a genuine one. This suggests that there is only a limited luck egalitarian case for employing vaccination status as a rationing criterion. The difficulties of implementing such a policy may lead luck egalitarians to consider other responsibility-sensitive policies instead.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study is funded by Danish National Research Foundation (DNRF144).