Article body

1. Three questions of healthcare justice

According to Dennis McKerlie (1989; 1992; 2001; 2013), discussions of egalitarian justice have mainly engaged with two central issues, at the expense of a third of equal importance. The first issue is the currency of justice: which good or goods we should distribute in an egalitarian way. The second issue is the distributive pattern; within egalitarian thought, this typically concerns whether to aim for strict equality (egalitarianism), to prioritize the worst off (prioritarianism), or to make sure people have “enough” (sufficientarianism).

The neglected issue is the “temporal subject” of justice—that is, when or over what timescale(s) justice applies. It has risen in prominence since McKerlie (1989) framed it in this way, though it is still overshadowed by questions of currency and pattern. And, although the topic has been the subject of important work in healthcare justice specifically (e.g., Daniels, 1988, 2008; Jecker, 2013, 2018), it has received considerably less attention than the other two in this area. Yet a full theory of justice in healthcare must have something to say about the temporal subject because it has significant implications for the allocation of resources.

The main aim of this paper is to advocate a specific stance on the temporal subject with respect to healthcare. That answer is that justice demands a concern with particular moments in people’s lives, which is sufficientarian. This is “momentary sufficientarianism.” I make the initial case for a momentary view in section 2. This view is related to McKerlie’s own advocacy of a principle of justice applied to different “segments” of people’s lives, but differs in that he aims to provide an account of justice’s temporal structure and pattern from a purely welfarist perspective. My view is that different approaches to the pattern and temporal subject may be warranted depending on our choice of currency, and that there is no one currency that commands the undivided attention of justice. This means that the case for any particular answer to the questions of pattern and temporal subject demands attention to the nature of the good being distributed.[1] This is true even within a single sphere of concern: while section 3 makes the case for adopting a momentary view with respect to some central aims of healthcare, healthcare has multiple aims, and they may suggest different answers to the temporal question. Section 3 also considers two versions of an objection to my argument: that the intuitions elicited by the cases on which McKerlie relies can be explained without appealing to time-relative distributive principles.

Section 4 then adds the question of pattern to this momentary view of healthcare and makes the case for momentary sufficientarianism. This is a second point of departure from McKerlie (2013), who rejects sufficientarianism with almost no discussion,[2] instead advocating prioritarianism as a time-relative principle.

Section 5 extends this discussion by considering a comparison between two important goods that healthcare aims at—life-extension and pain relief—and which seem to offer different perspectives on the temporal subject even within healthcare. Initially, pain relief seems naturally to suit a momentary principle, while life-extension appears to be best considered from the lifetime view. I suggest, however, that neither of these claims is as straightforward as it seems.

Finally, section 6 considers the relationship between lifetime and momentary views in the context of healthcare rationing. Unless we adopt one view exclusively, we need to know how to decide between them. I suggest that some ways of combining the two—lexical priority and a “tie-break” system—unreasonably relegate one principle to insignificance and argue that the best way to accommodate both kinds of claim is a macro approach to healthcare spending, rather than an attempt to micromanage individual patient claims.

2. The temporal subject

We can often find injustice when we look at how people’s lives go overall. That some live privileged lives while others barely manage to get by is unjust. McKerlie’s important observation (e.g., 2013, p. 6–7) is that we should also be troubled by distributions when we look at periods shorter than entire lives. He imagines a city block whose young citizens enjoy a high standard of living, but whose elderly residents are crammed into substandard retirement homes, isolated and neglected. If today’s elderly enjoyed a similar high standard of living when they were young, and today’s youth will themselves be consigned to poverty and exclusion when they are old, the city block achieves lifetime equality of welfare, opportunity, and resources. We can add, given the context of this paper, that the youth of this block have strong priority in terms of healthcare access. While older patients are entitled to some healthcare, services that target older patients and the conditions that affect them are a low spending priority, and the fairly minor complaints of younger patients are given priority over more serious issues for the elderly. In this case, we also have lifetime equality with respect to healthcare access, and one that aims at equality in health.

Such a distribution also passes lifetime priority and sufficiency tests. Since nobody’s life is worse than anyone else’s, we cannot advocate improving the elderly’s lives because they are absolutely worse off. Similarly, since everyone has the same quality of life, either everyone will have a sufficiently good life or nobody will. If the latter is true, then we have reason to improve everyone’s life, so that each one reaches sufficiency. But, according to a pure lifetime view, we have no stronger reason to target that improvement at the worse period of life, old age, rather than at the already high-quality period of youth. Both will lead to a better life overall.[3]

McKerlie claims that the unequal city block is clearly unjust because it neglects those who are worse off at particular times. As we’ll see, this has been challenged on several fronts. We can, however, make the case for a time-relative principle from a weaker claim. This claim is that, in some cases, people’s time-relative situations can give them claims on our prioritizing them, even if that reason is ultimately overridden by stronger claims from elsewhere. All that matters is that the time-relative claim is not reducible to something else, such as a lifetime claim.

Imagine a choice between two people whose lives are entirely synchronous, both living for eighty years. Call the respective halves of these lives T1 and T2, and assume that health is measured by something like the Quality-Adjusted Life Years (QALY) score (e.g., Williams, 1997; 2004). The QALY measure represents perfect health with a score of 1, death with 0, and various health conditions with decimal scores to represent how bad they are compared to these two extremes. Each potential added year is weighted according to the health conditions it contains. So, we can compare prognoses for patients that differ in both length and quality.

Adali suffers from a chronic condition that gives her extremely poor health (0.2 QALYs per year) for the first forty years of her life, but on her fortieth birthday is given a treatment that significantly alleviates her situation, giving her good health (0.8 QALYs per year). Bilal enjoys full health (1 QALY per year) for the first forty years of his life, but at forty suffers an accident that leaves him in extreme ill health (0.1 QALYs per year) for the second half of his life. We can represent our lifetime situation as follows, with the numbers representing total QALY scores over each forty-year period:

forme: forme pleine grandeur

In lifetime terms, there is an inequality in Bilal’s favour; his lifetime QALY score is 44 (or 0.55 per year, depending how we measure lifetime QALYs), and Adali’s is 40 (0.5 per year). Similarly, Adali’s life is overall worse than Bilal’s, and so she is the stronger candidate for lifetime priority. Finally, although we haven’t specified the sufficiency threshold, it’s clear than if Bilal’s life is below it, so too is Adali’s, and even more so.

However, at the beginning of T2, we can reasonably predict several things. First, without intervention there will be a large inequality in Adali’s favour. Second, over the next forty years, Bilal will be much worse off than Adali in absolute terms.

If lifetime equality were all that mattered, we would have reason, perversely, to make Bilal’s situation worse. Of course, almost all egalitarians would deny that this is what we ought to do. Temkin (2000, p. 155), for instance, would say that, while there would be something good about Bilal’s situation getting worse (since it would equalize things at the lifetime level), this is overridden by other moral principles. Nonetheless, all three lifetime views imply that, if we can offer treatment to only one of these two, we ought to choose Adali.

More importantly for my argument, however, they imply that there is no reason at all to prefer Bilal. People may have different intuitions about what we should do in this case. But all I require is the following claim: Bilal has some claim to priority over Adali, based on what his life will be like in T2. This is important, because this claim cannot be based on lifetime considerations, which point exclusively in Adali’s favour. As such, I suggest, we must recognize that there are distinctive distributive claims based on people’s situations at particular times.

The purpose of this section has been to motivate the need for a time-relative principle. Sections 3 and 4 say something more specific: First, the relevant period of time is the moment. Second, in the context of healthcare, the relevant principle should be sufficientarian. To motivate these claims, I argue in section 3 that different principles may be appropriate for different kinds of good, and that this depends in part on how that good features in our lives. I also consider two objections to my Adali-and-Bilal case, which both claim that we can explain the relevant intuitions without invoking time-relative claims. Section 4 makes the case for adopting a sufficientarian momentary view in healthcare.

3. The temporal subject in health(care)

According to some of its opponents, the time-relative view has an intuitive hill to climb, because the lifetime view is the only intuitively plausible timeframe for distributive justice. Bidadanure (2016, p. 245) calls the lifetime the “par excellence time unit of distributive fairness,” citing in support Lippert-Rasmussen (2015, p. 156) and Wagland (2012), and offers four reasons for thinking so. Though not all are addressed in this section, I mention them all, since Bidadanure sets out the intuitive challenge facing time-relative views especially clearly.

Firstly, the lifetime view explains why maximization of utility is acceptable within a person’s life, but not across lives (the “separateness of persons” thesis). I address this “prudence problem” when I consider Daniels’s prudential account, in section 3.2.

Secondly, there is a potential problem with arbitrariness in choosing the relevant sublifetime time frame (the “arbitrariness problem”). I address this in section 4.2.

The third reason, related to the first reason, is that the complete-lives view allows us to accommodate the idea of responsibility and compensation. That I am doing badly at one point in my life can be less objectionable if that is the result of choices I took earlier or if I am compensated at some other point in my life. This, the “responsibility problem,” is addressed in section 4.4.

Finally, the lifetime view corresponds to “a widespread metaphysical view” about persons, which is that the same person occupies an entire life. The rest of this section considers this “metaphysical problem.”

We can think of the lifetime and time-relative principles as coming from two distinct, independent perspectives on social goods. From the inside, human lives embody both a lifetime and a momentary perspective, and certain goods matter to us from both. For most of us, it is impossible to think of ourselves coherently except as temporally extended across our lives. We make projects and plans, care about access to goods now because of their future implications, make sacrifices for later gains, and accept benefits knowing they’ll require sacrifice later. This is the lifetime view.

Some goods seem to be most naturally considered from this perspective (though perhaps constrained by momentary considerations, too, as we will see below). For instance, Daniels (1988, p. 41) suggests that it is not problematic if a society has its young people start out less wealthy and then accumulate wealth, so long as all generations go through the same process. Similarly, it seems acceptable if, on average, young people have readier access to education than older people—though that does not preclude the right of older people to access education if they need it.

But we are also temporally embedded beings. We care about certain things quite apart from how they contribute to our lives, but simply because of how they feel or matter to us now. This is the momentary view. My current pain does make my life worse. But I care about it primarily because of how it feels now. Similarly, some debilitating forms of ill health matter because they make the sufferer’s life worse. Ill health at one time may therefore call out, particularly if it is severe, for compensation at another. But my ill health also matters to me because of the way it feels as it occurs, because of the way it limits my ability to be and do various things (see Nussbaum, 1992; Sen, 2009), and because of the way it limits my capacity to create and take advantage of opportunities that matter to me right now. Since many of these experiences and opportunities are neither fully interchangeable with others nor capable of being recreated at another time, my ill health can also generate distinctive claims to be treated at particular times. While compensation might be welcome, compensation often cannot make up for what I lost while suffering ill health.

This suggests that the temporal subject cannot be entirely disentangled from currency, because different currencies may naturally imply different temporal views. Many of those who have discussed the temporal subject have adopted a specific currency before offering their arguments. For instance, McKerlie (2013) adopts a welfarist view, and Daniels (2007) is concerned with having a “normal” level of opportunity for one’s age. My focus is on health as a currency of justice. But it is worth reiterating two points explicitly. First, I will not assume that health is the currency of justice, because I do not suppose that there is single, irreducible currency. Second, I do not assume that there is one single answer to the temporal question even when we focus solely on healthcare. This is because, even when healthcare is intrinsically valuable, it may be valuable in a pluralistic way; there is no obvious reason for us to assume that the intrinsic value of pain relief is the same as the intrinsic value of life extension. But both constitute healthcare.

3.1 Relational goods

Another set of goods that matter from a momentary perspective relate to social standing. Relational egalitarians argue that the egalitarian perspective must include, as well as (Wolff, 1998) or perhaps even instead of (Anderson, 1999) distributive concerns, a focus on our ability to relate to one another as equals, which includes a concern with oppression, exploitation, shame, and respect. These are all goods that can be affected by health and access to healthcare, and which also arise in a momentary sense. McKerlie (1989, p. 479) imagines another case, where two groups swap social standings throughout their lives, with one group occupying an exploitative, domineering position over the other, which is then reversed.

While there is a sense in which this society equalizes social standing over its citizens’ lives, there are deep social inequalities during both periods. Being somebody’s social equal is not something that comes out in the wash, considered on balance over the course of a lifetime. One reason for this is that social standing is in part about our relationships and interactions with others. To say that people interact on an equal basis is not to say that they interact on an unequal basis, but while taking turns on top.

Some of our concern with moments in the distribution of other goods, such as healthcare, may well relate to this relational concern, as Bidadanure (2016) argues. She suggests that our intuitive reaction to McKerlie’s unequal-city-block example can be explained by appealing to relational equality. On this view, momentary distributive inequalities in health and healthcare access do not matter in themselves, but matter because they contribute to, or constitute, relational inequalities. For instance, Bidadanure (ibid., p. 246) suggests that one problem with the city block is “not that there is a timeslice inequality in distribution as such, but rather that relationships of inequality may pertain at all times.”

Bidadanure is concerned here with rejecting a time-relative egalitarian view, understood in the narrow, comparative sense. Since my view, as outlined in section 4, is that the momentary principle should be sufficientarian, I am not in direct conflict with her central proposal. But her view does raise the prospect of a more general sceptical view, that any intuitions that seem to suggest specifically momentary distributive concerns can be explained by nondistri-butive considerations.

Kasper Lippert-Rasmussen (2015, p. 155-156) notes that, in describing his unequal-city-block case, McKerlie says that the elderly residents lack “dignity,” and compares their situation to a racial injustice, which is clearly not only a distributive problem (see also Segall, 2016, p. 87). So, even if these cases raise valid concern, that concern may be directed towards nondistributive issues.

It is certainly right to suggest that one reason to object to a momentary distribution in healthcare rests on its implications for other forms of justice, and indeed issues beyond justice. But these other areas do not matter only because of their relationship with relational inequalities. As Daniels (1981, p. 146; 2007, p. 18) suggests, a central reason to place special emphasis on health is its implications for opportunity. Momentary inadequacy in health may thus also have a serious effect on opportunity.

But we should also care about (at least some) momentary allocations in healthcare not only for the sake of their implications for further issues (such as relational inequalities), but also for their own sake. As I suggest in section 4.1, one reason for this is that some cases of the relevant good—healthcare need—matter irreducibly from a momentary perspective. But we can offer an initial, different motivation here. One potential issue with an appeal to relational equality is its capacity to explain why certain distributions exemplify relationships of inequality, while others do not. Relational egalitarians are, says Bidadanure (2016, p. 236), concerned with “oppressive relationships such as exploitation, domination, or exclusion.” On the relational view, a distribution is unjust only if it constitutes or otherwise contributes to such a relationship.

Similarly, Lipper-Rasmussen notes that McKerlie’s imagined elderly lack dignity. But to frame a distribution as constitutively exploitative, dignity denying or (unjustly) exclusive, we must have a sense of why the relevant good matters in a particular way. For instance, Bidadanure (ibid, p. 246) suggests that McKerlie’s unequal city block may be relationally problematic because elderly citizens are “segregated,” which, along with different levels of affluence, “easily become[s] associated with unequal status and unequal levels of respect.” But why should it be that markedly different levels of affluence, for instance, are often associated with disrespect for some? It is because one’s level of affluence matters for the quality of one’s existence that it can imply disrespect.

Similarly, one reason that having (avoidably) insufficient access to healthcare may be demeaning and suggest a level of disrespect—even if it is balanced out at another time—is because it is wrong for its own sake. That the society of which you are a member has allowed you to fall into severely ill health is something that it is reasonable to feel aggrieved and disrespected over, because of the fundamental value that health has. It is bad in itself to be in severe pain, or to suffer from significantly restricted mobility, or to be bedridden. That is why some levels of disparate healthcare access constitute disrespect.

Further, even if one thinks that a nondistributive explanation can be found for the intuitive force of the unequal city block, the same is not so obviously true for my narrower healthcare example. For one thing, the linguistic cues highlighted by Lipper-Rasmussen with respect to the unequal city block are absent here. All I appeal to is the fact that various points of Bilal’s future will be very bad indeed.

Lippert-Rasmussen assumes that, unless advocates of the momentary view can come up with an example for which no alternative explanation can be found, then we should not adopt the momentary view. In part, this is based on the claim that the momentary view is deeply unintuitive, so we should accept it only if it explains a case for which no other explanation is available.

This rests on two further claims. First, the momentary view is prima facie implausible, since it faces the four challenges mentioned by Bidadanure. Second, it holds no intrinsic attraction, so its only support is its ability to explain cases that no other view can. But, as I will argue, the four challenges Bidadanure mentions can be overcome. Furthermore, the momentary view does have intrinsic plausibility if understood correctly. For, if it is understood as a sufficientarian view, and simply as a claim about grounding claims that are nondecisive, then it amounts to the following: that we have some distributive reason to want to prioritize Bilal over Adali, and that this is grounded in his poor future prospects.

Of course, one might worry that Bilal’s poor health will have various effects that we should care about for nondistributive reasons. But this is not the fundamental reason to care about his poor health. Rather, our fundamental reason for concern is simply that his future will be very bad. The claim that there should be a momentary principle of justice is grounded, in this case, only on the thought that there is a distinctive claim grounded in how a person will fare at a particular moment, or during a collection of moments. So, while I accept that alternative explanations must be considered, I am not sure it is reasonable to expect that advocates of the time-relative view find cases where we can completely discount alternative, nondistributive explanations of our intuitions.

3.2 Prudence

A different alternative explanation is that we can explain the intuitions raised by cases like McKerlie’s and mine from within a solely lifetime framework. A central example of this approach is Norman Daniels’s prudential lifespan account (PLA) (e.g., 2013). Briefly, Daniels suggests that, since we all occupy multiple age groups during our lives, intergenerational justice should be based on what one would prudentially choose from behind a Rawlsian veil of ignorance, with various morally irrelevant facts about oneself obscured. Such deliberators might well suffer reduced healthcare in old age for increased access while younger, in order to increase the chance of them making it to old age. Crucially, while Daniels’s deliberators have their actual age hidden from them, he stipulates (2008, p. 475) that allocators “must assume they will live through each part of the life, accepting any tradeoffs they make.”

One might therefore worry, as McKerlie (1989) and Jecker (2013) do, that this will lead us to tolerate not only inequality but extreme hardship in old age. Citizens who must reason as if they will start life from birth, with a limited amount of resources to spend on healthcare, would allocate (almost) nothing to some levels of old age, since they are less likely to reach that stage of life. In addition, Lazenby (2011) and Davies (2018) suggest that the requirement that citizens live through all stages of life, and start at the beginning, unavoidably obscures important issues, such as that of individuals who suffer from incurable life-shortening illnesses.

Defending Daniels, Bidadanure (2013, p. 26) rejects such charges as focusing on only one part of the PLA. In addition, she says, the account also stipulates that “it would be imprudent to discount one stage of our life span by denying our younger or older self sufficient resources to live decently.” As such, the intuitions elicited by McKerlie-style cases are accommodated by adopting a lifetime principle based in prudence as well as in equality.

However, while this may redeem Daniels’s account, it can do so only by incorporating a time-relative sufficientarian principle. It is worth noting that, later in her discussion, Bidadanure suggests not that avoiding hardship would be imprudent, but rather that “it would not be fair to deny the elderly the necessary resources to live decently” (my emphasis). This difference is important, and the second formulation is more plausible. As Daniels has since noted (2008), it is actually quite difficult to stipulate a single view of prudence—even from behind a veil of ignorance—that will rule out some individuals seeing hardship during some periods as worth enduring to maximize lifetime welfare. Indeed, some may even see it as worthwhile to endure aged hardship even if it does not maximize lifetime well-being, because they simply prioritize their youth to a considerable extent. So, an appeal to hypothetical prudence is incomplete.

4. The pattern of momentary justice

If people may have momentary claims to healthcare, to what distributive pattern do these claims relate? Recall that there are three broad answers we might give to this question: egalitarian, sufficientarian, and prioritarian. In the context of a momentary principle for health, these ideas might turn out in something like the following way:

Egalitarian: People have a claim to the healthcare they need to be just as healthy as the healthiest members of their society at all times.

Prioritarian: People’s claims to healthcare become stronger, the worse their health is at a particular time.

Sufficientarian: People’s claims to healthcare are discontinuously[4] stronger whenever they fall below some threshold of health.

Deciding among these principles requires a further specification of the idea of momentary egalitarianism, which will offer a further reason for adopting a momentary view, rather than a more general “sublifetime segments” view. The next two sections advocate momentary sufficientarianism. This is the view that justice gives people discontinuously stronger claims to benefits when those benefits would either prevent them from not having enough (in terms of health) at a particular moment, or rescue them from not having enough. The argument is that the moment is the only nonarbitrary alternative to the lifetime, and that a concern with moments is best understood as a concern with sufficiency, at least with respect to some goods that are central aims of healthcare.

4.1 Arbitrariness and equality

I have argued that some justice-relevant goods are best understood as implying a distinct concern with moments. A common criticism of the call for a time-relative principle of justice is the arbitrariness problem: to find a nonarbitrary time frame, aside from a lifetime, that grounds concerns of justice. These ideas combine to give an argument for some goods being governed by momentary sufficiency.

A concern with periods of time shorter than a life could mean a concern with all time periods shorter than a life. In the case of Adali and Bilal, the sublifetime inequities I identified were over a period of forty years. This leads us to a problem. In the original example, I assumed that both Adali’s and Bilal’s health statuses would be uniform throughout each period.

But it is more realistic to assume that their health will fluctuate within those periods. Although Adali does better across T2 than Bilal, perhaps this is because she does much better during the final twenty years of life, but a little worse during years 41 to 60. Once we identify this possibility, more emerge: since we can continuously subdivide time periods, there is always the possibility of further distributive issues.

The arbitrariness problem is the challenge to find a nonarbitrary length of time, shorter than the lifetime, to which to apply a time-relative distributive principle. In my view, following McKerlie, the only nonarbitrary option is that the relevant periods are moments, since no intrasegment inequities can occur. For McKerlie, this raises a fundamental problem for time-relative egalitarianism (cf. McKerlie, 2013, p. 84-85), since a demand for strict equality at every moment seems excessive. For instance, some healthcare treatments themselves negatively affect our health in an unavoidable way. An operation to avoid cancer may cause additional, less serious, health issues. Similarly, an intervention that relieves long-term pain may nonetheless cause pain for a shorter period. If two people undergo similar such interventions, momentary egalitarianism tells us that we must, as a matter of justice, schedule those operations so that the two individuals suffer their unavoidable reduction in health at the same time; otherwise, there will be an avoidable inequity at various moments between those two. But this seems unnecessarily fussy.

It is worth acknowledging that, just as I argued that currency may affect the temporal subject, not all goods are equally vulnerable to this fussiness concern. For instance, the relational goods discussed in the previous section may be such that, as Axelsen and Nielsen (2015) argue, only equality can count as enough in these contexts. Since the basic demand of social standing is to be treated as an equal, it does not seem arbitrary to demand that people are treated as equals in terms of their social standing; since, as I have argued, this cannot be understood in a lifetime balancing-out sense, that implies that momentary equality is not an unreasonable demand in terms of social standing.

But this kind of thinking does not apply to healthcare; as Schramme (2007, p. 126) says, “Whether we need medical help does not depend on the life-conditions of our fellow citizens, but only on our own personal bodily or mental condition.” So, the fussiness complaint against momentary egalitarianism stands.

As McKerlie suggests (2013, p. 105-106), this implausibility does not affect prioritarianism, because it is only derivatively comparative. Hence, small changes in one person’s behaviour or situation need not have meddlesome implications for others. We can have stronger reasons to help those who are worse off at any particular moment, without having to coordinate collective behaviour. Although he does not consider it, similar logic applies to sufficientarianism: that one person does not have enough at a particular moment can be understood without reference to anyone around that person.

Opponents of a momentary principle might insist that, even if momentary sufficientarianism does not demand scheduling, a focus on moments is still overly precise. Consider S, who is very well off for her whole life except for one moment, when she suffers a moment of intense agony. Opponents might object that we have no reason to benefit S during this one moment, given how good her life has been otherwise; S’s suffering is simply too minor to be concerned about. Yet on my view, S’s claims to benefits are stronger with respect to that moment. To respond to this claim, I think it will help to locate the view I am advocating within a family of related views.

Before doing that, in the next section, I will offer one argument in favour of sufficientarianism rather than prioritarianism being our momentary principle with regard to healthcare—a further such argument comes in section 4.4, with my discussion of the responsibility objection. Schramme’s argument explains why we want a noncomparative view. To see it should be sufficientarian, note that, as well as being noncomparative, many healthcare needs are satiable from a momentary perspective. When you suffer severe pain, we can give you pain relief; your claim to pain relief is based on the fact that it will relieve the pain. There comes a point at which this claim is sated. This may be because the pain is gone, because it is no longer severe, or because the pain is such that further pain relief will do no good.

As Shields (2012, p. 35-36) argues, when it comes to satiable claims, “those who have not secured enough can call on the weight of more and therefore a different profile of reasons than those who have secured enough.” This requires that the basis of the claim be noninstrumental and weighty. Insofar as seriously poor health—including severe pain—is intrinsically bad, and the source of a weighty claim, then, this supports a sufficientarian approach. The sufficientarian threshold represents the point at which claimants can call upon a discontinuously weighty set of reasons to benefit them: reasons that are based on their not having enough.

Shields does not outline his view in terms of either a time-relative principle or healthcare. But his reasoning clearly applies to both. Satiability is typically a time-relative concept: many needs are satiable in a temporary sense, but not in an ongoing sense. And I have demonstrated how satiability relates to claims to healthcare.

4.2 Momentary sufficientarianism: What it is and is not

I will now explain how momentary sufficientarianism differs from some other nearby alternatives. That a concern for sufficiency should aim at particular times is not a new idea. For instance, Jecker (2013, p. 10) says that a healthcare principle that looks only at lifetimes will “miss the unique features of caring for the chronically disabled,” and proposes that an allocation is just if it “maintains basic functioning and capabilities at a sufficient level”; although she does not specify the precise nature of the temporal subject, it is clearly shorter than the lifetime. Similarly, Gosseries (2011) considers a “continuous sufficiency” view, which places considerable restrictions on the forms of compensation allowed between different points in a person’s life. Gosseries does not commit to a full picture of what sets these restrictions, but seems to endorse a view based on the idea of human dignity. One important point of difference is that Gosseries explicitly claims that such a sufficientarian view “would not be concerned about the ability of people to meet their basic needs every second.” This reflects the concern raised at the end of the previous section, that such a concern would be excessive.

In my view, momentary sufficientarianism is concerned about people falling below the threshold set by sufficiency, even for just a moment. To be sure, if we can predict that it really will be just a moment, then we may not be greatly concerned. All theories of justice may lead to a theoretical concern with apparently somewhat trivial matters: for instance, a strict lifetime egalitarian view will care whether two people have nearly identical lives, which are extremely happy, but one of them has a single moment of feeling a little down. Momentary sufficientarianism can explain why we should care more about longer periods of insufficiency for at least one reason: a longer period comprises more moments than a shorter period. So, it does not imply that we should be indifferent between someone suffering a single moment of insufficiency and someone suffering a year of insufficiency.

This leads to two potential points of misunderstanding. First, as with many related views, momentary sufficientarianism sees injustice (as opposed to tragedy or a cause for regret) only in avoidable deviations from our pattern. This again serves to make the requirement that we care about every moment less demanding, and so more plausible.

Second, momentary sufficientarianism does not claim that people are entitled to compensation for every moment of their lives spent below the sufficiency threshold. The fact that someone drops, or will drop, below the threshold in a particular moment is the grounds for their having a strong claim on the rest of us to avoid or end that situation. It doesn’t mean they are entitled to compensation.

The person who experiences agony for a moment thus has a claim to having their pain relieved. If it is a genuine moment of agony, it seems odd to deny that we have a significant reason to prevent it. But, as with Bilal, the relevant reasons are defeasible; that S has a stronger claim at that time does not mean we must do everything it takes to help her. There may be practical constraints that mean we typically ignore such brief agonies. In addition, as I discuss in section 6, the claims that derive from momentary insufficiency might be outweighed in some cases by the claims that derive from other principles, such as a lifetime egalitarian principle.

Since our reasons are defeasible, a rejection of momentary sufficientarianism is committed to saying that, if we can foresee S suffering terribly for just a moment, we have no distributive reason to prevent it. Contrary to Lippert-Rasmussen and Bidadanure, it seems to me that it is a rejection of such a view that requires strong justification. As McKerlie (2013, p. 82) says, suffering is morally serious, even if it lasts only for a moment. If it is also something we could prevent or relieve, then that is something others may demand of us.

4.3 The sufficiency threshold

In characterizing sufficientarianism, I adopted a view based on Liam Shields’s: the sufficiency threshold marks a point at which distributive claims become discontinuously weaker. The main issue that Shields’s sufficientarianism faces is how to define the threshold nonarbitrarily. His answer to this shows why momentary sufficientarianism is an appropriate principle in the case of health. Shields argues that the key concept underlying the sufficientarian ethos is the idea of satiability. As he puts it, if someone is in pain, or needs treatment for an infection, there is a level of resources that will complete that task. That person may have further claims to a different benefit, but must appeal to a “different profile of reasons” (2012, p. 113) to make that claim.

The idea of satiability seems most obviously applicable to resources. One might worry that sufficientarianism looks plausible only because we can set the threshold for resources by appealing to a different pattern, such as priority, for some more fundamental good, such as health: people have “enough” resources if their holdings enable them each to satisfy their prioritarian claims to healthcare. But satiability can also apply to health itself. Health is itself valuable because of its relationship to well-being. As Ram-Tiktin (2012, p. 341-342) argues, we can identify key thresholds within the range of health statuses a person might have on the grounds of their contribution to personal flourishing. Such a view can also make room for a plurality of thresholds, each of which instantiates a discontinuous weighting for priority of claims. For instance, Ram-Tiktin suggests a lowest threshold of personhood, below which “human lives are not possible, either because the person dies or loses some fundamental capabilities basic to human life, such as the ability to perceive or communicate (even by eye or finger movement) when, for example, a person falls into a coma” (ibid., p. 343; see also Huseby, 2009).

To some extent, this idea bases sufficientarian health thresholds on their instrumental contribution to a further idea of flourishing or opportunity. This is enough to show that sufficientarian reasoning has a role to play in allocations of justice, since a central part of the value of health is its instrumental value. But, at least in some cases, we can also talk about intrinsic thresholds related directly to health. For one thing, health is not only instrumental to human flourishing but is on some views partly constitutive of it (e.g., Nussbaum 2006; Chisholm, 2011, p. 189) or of some related idea such as well-being. Insofar as Ram-Tiktin’s health thresholds are constitutive rather than instrumental, we thus have a more fundamental role for momentary sufficientarianism.

For a more specific example, consider again the idea of pain relief. A basic reason that many forms of healthcare are valuable is their contribution to pain relief. Pain may not seem an obvious candidate for sufficientarian justice, because it can obviously be increased and decreased incrementally, and so invites a common complaint towards sufficientarian views: how can two people whose conditions are almost identical be—by virtue of falling just above and just below the sufficiency threshold—seen as radically different by a theory of justice?

Frankfurt’s (1987) discussion of resourcist sufficientarianism is grounded in the reasonable attitudes that people take towards their holdings. For instance, Frankfurt discusses the idea of contentment. Contentment does not mean that one’s position cannot be improved, but that improvement is a possibility about which one is not particularly concerned. On Frankfurt’s view, people no longer have claims of justice when they are content with what they have. But, while contentment may work as an upper threshold, it surely cannot operate as a lower threshold; it would be overly demanding to say that the only point at which our claims of justice become distinctively strong is when we feel discontent. Rather, one option for a lower threshold is the notion of tolerability.

Toleration is an attitude taken towards one’s situation. It is neither merely a psychological state that contributes to welfare, like happiness, nor an external description of a particular level of welfare. It is true that finding a situation intolerable may cause additional distress, and hence lead to further declines in welfare or barriers to flourishing. But finding something intolerable is not the distress itself. Frankfurt discusses contentment in similar terms, saying that to be content need not imply that one would not be happier with more, but that one feels no sense of urgency to acquire more. Similarly, finding your situation tolerable—or intolerable—is a kind of attitude. Frankfurt applies this idea of attitudes solely to resources, but it is also applicable to other goods, including one’s level of pain or one’s opportunities. At the margins, a small increase in pain can move a person from being in pain they can just about bear to unbearable pain. In relation to opportunity, the notion of tolerability can occupy a midpoint between the fairly demanding notion of flourishing and the fairly minimal idea of personhood. Health, and healthcare, can therefore be the subject of momentary sufficientarian claims on these bases.

4.4 Responsibility

The final of Bidadanure’s challenges is the responsibility problem, which also includes a concern with compensation. On the face of it, momentary views seem excessively willing to ignore the history of distribution (e.g., Bou-Habib, 2011, p. 291; Bidadanure, 2016, p. 244-245). If Chris prefers to spend his money at the cinema every night, and Jenny saves hers, what right has Chris to complain if he is badly off at the end of the month? Perhaps it is also unfair to provide healthcare to people who are unhealthy because of their own poor choices. Relatedly, people often willingly tolerate hardship for the promise of benefits later. If they autonomously chose hardship, why is that unjust?

On the other hand, it can seem overly harsh—in Bou-Habib’s terms, overly willing to tolerate hardship—if we simply refuse healthcare to all who bear some level of responsibility for their condition (see Fleurbaey, 1995). This is exacerbated by the fact that the same level of responsibility can, through sheer bad luck, result in significantly different outcomes depending on one’s luck, such that isolating the degree to which one is responsible for one’s situation is extremely difficult to determine (Wikler, 2002).

Momentary egalitarians and prioritarians seem bound either to ride roughshod over people’s free choices—whether wise or foolish—thereby ignoring responsibility and prudence, or to reject freely chosen hardships as grounding stronger claims of justice, even if those hardships are extreme. To adapt Fleurbaey’s (1995, p. 40) phrase to a different context, “if you freely and deliberately make the slightest mistake that can put you in a very hazardous situation, a society complying with [these brands of egalitarianism or prioritarianism] will quietly let you die.”

The momentary sufficientarian, however, can mark a threshold below which even freely chosen hardships must be corrected. For the sake of explanation, assume tolerability grounds the relevant threshold. If someone is poorer than others due to a spending spree on concert tickets, and hence suffers some level of ill health, this need not be a concern of sufficientarian justice. But, if someone has, as in Fleurbaey’s example (ibid, p. 40), gone motorbiking without a helmet and suffered a life-threatening crash, sufficientarians can insist that her position below the relevant threshold means that her responsibility is immaterial to her claim (see Gosseries, 2011, p. 474 for discussion related to this point). This ability to accommodate both responsibility and hardship in a principled way is a distinct advantage of momentary sufficientarianism over both pure lifetime views and other momentary views.

This strength is compatible with a range of views on responsibility. One might take different views on hardship that occurs through foolish mistakes, and hardship that comes about through careful planning. Perhaps it is excessively harsh to punish people with extreme hardship for making a foolish mistake, but fair to expect people to endure the hardship they carefully planned into their lives. So long as one wants to make a distinction for some forms of responsibility, momentary sufficientarianism is the best placed to do so.

5. Momentary justice in healthcare: Life extension and pain relief

My argument thus far has been that momentary sufficientarianism is an important principle in the context of many health-related goods. This section addresses the application of this idea to two key aims of healthcare: saving people’s lives and relieving their pain.

If lifetimes alone matter, we have reason to significantly target healthcare spending in particular directions—namely, towards conditions that affect children (Lazenby, 2011) and those who will suffer worse lives overall because of bad luck or social injustice (Overall, 2009). This is particularly true for life-threatening conditions; someone who dies in childhood, or after many years of deprivation, is irredeemably badly off in a lifetime sense, both absolutely and comparatively with others. But this will also apply to many cases where someone’s predicted recovery will only partially compensate that person’s earlier deprivation. If, on the other hand, moments matter as well, this will affect the priority, and perhaps the direction, of this targeting.

It is important to note that when we speak of targeting, we are not necessarily restricting ourselves to choices between individual patients. Many philosophical discussions of healthcare justice focus (perhaps even fixate) on choices such as that between Adali and Bilal. These choices are relatively straightforward and make it easier to tease out features of problems and proposed solutions; they have philosophical value. But it is rare for rationing decisions to be made at this level. The UK’s National Institute for Health and Care Excellence (NICE) is one example. NICE’s methodology for calculating cost effectiveness involves identifying the average improvement in the health status of individuals receiving the intervention, “over and above any other gains they might receive” (Rumbold et al., 2017, p. 112). A distributive view of healthcare must say something about choices broader than those between individual patients.

As I discuss in section 6, the targeting of spending also applies at this macro level. For instance, a strategy that prioritized the healthcare of children might set less demanding cost-effectiveness thresholds, or actively increase spending per capita, for conditions that predominantly affect children. It might also devote greater funding to research into childhood diseases, and to centres that care predominantly or exclusively for children. Finally, it might simply take being below a particular age as an additional factor in deciding treatment priority in nonemergency cases.

Life extension is an issue where lifetime impulses are keenly felt. Many people argue on egalitarian grounds that in allocating life-extending resources, we must prioritize younger over older patients (e.g., Battin, 1987; Kappel and Sandøe 1992, 1994; Callahan 1995; Hardwig 1997; Lazenby, 2011). This sentiment is also common among the general public, many of whom prefer to some degree to give a shorter amount of additional life to a younger patient than a longer time to an older patient (see, e.g., Cropper et al., 1994; Nord et al. 1996; Johannesson and Johansson 1997; Nord 1999, p. 57–61). Many justifications of this view are explicitly egalitarian, appealing amongst other ideas to the fact that older patients, by definition, have had more life than younger patients. Others are more apparently sufficientarian, but where the notion of sufficiency is applied at the lifetime level. For instance, the “fair innings” argument (Harris 1985, p. 91) suggests that it is legitimate to make the saving of older people’s lives a lower priority not because they have had more life than the young, but because they have had a sufficient amount of time alive.

I claimed in section 3 that whether we apply a momentary principle, a lifetime principle, or both, depends in part on the kind of perspective the relevant good is valued from. Whereas some goods, such as equal social standing, seem constitutively bound to a momentary perspective, others are more naturally viewed from the lifetime perspective. It may seem that support for age-based rationing is backed up by this, because the idea of extending someone’s life is surely a concept that makes sense only from an extended point of view; the very concept of extending cannot be made sense of only at one moment.

Contrast that with pain. It may seem equally obvious that pain matters only in the moment. While it is not exactly constitutive of pain that we think of it from the perspective of the moment it is occurring, pain is a harm whose badness is very much of the moment; I want pain relief not because it makes my life better, but because it hurts right now.

However, this apparent simplicity is open to challenge. Consider first the suggestion that pain should be considered solely from a momentary perspective. We could allocate pain relief from a lifetime perspective. Some people have relatively pain-free lives, even though they would benefit from pain relief at particular times; others suffer pain more frequently throughout their lives. A lifetime view of pain would take the cumulative pain people have felt across their lives, and prioritize access to pain relief for those with the worst lifetime-pain score. A purely lifetime view would not further consider the current pain someone is in, except insofar as it contributes to that person’s lifetime pain score. We could, at least in principle, get the result that someone who is currently in mild pain is prioritized over someone who is currently in agony because the former has suffered more lifetime pain.

While a purely lifetime view of pain may thus seem implausible, the idea that there is some injustice in the fact that people have radically different lifetime experiences of pain—and that this inequality is often clustered with inequalities in other areas—is surely not. In a context where pain relief is scarce (or where cuts in pain relief might serve to secure other fundamental goods), those who advocate the lifetime principle in some cases need to explain why we should not also take this approach for pain relief.

We can also challenge the idea that interventions that prolong someone’s life can be meaningfully viewed only from the lifetime perspective. It is certainly true that when we consider these interventions as life extensions (which they surely are), we are taking on a lifetime perspective. And that suggests that there is some truth to the egalitarian complaint outlined above. If we control for quality, older individuals clearly have had more of a foundational good (life) than younger patients.

But, to make a point related to Harris’s (2004, p. 528), life extension is also lifesaving. When we save people’s lives, we are not only giving them more of a good that they have been accumulating; we are also maintaining their access to a good that is of foundational value at particular times, in a way that is not merely reducible to its cumulative value. It is foundational because being alive is a prerequisite for enjoying all other goods, including those that make sense primarily from a momentary perspective. Even if we cannot understand human lives except as an interrelated whole, this does not mean that such a perspective fully encapsulates the way in which we care about having more life. Having our lives saved is also valuable to us because it allows us to enjoy relationships and projects (and simply being alive) at moments.

So, while there is something to be said for the idea that life extension is naturally viewed from the lifetime perspective, and pain relief from the momentary perspective, the motivation for adopting the corresponding principle is available for each. This raises a final question. If the momentary theorist does not abandon the lifetime view, but only supplements it, how do these principles fit together in the context of healthcare?

6. The relationship between momentary and lifetime principles

There are several ways the claims of lifetime and momentary egalitarian principles might be jointly assessed. One option is for claims of one kind to take lexical priority over claims of the other. But, depending on what kind of pattern the lifetime principle takes, this might mean that we end up entirely ignoring momentary claims. For instance, if lifetime claims are lexically prior and egalitarian, this would imply that we cannot meet any momentary claims until we have achieved lifetime equality. If we had achieved lifetime equality, on the other hand, we could not meet them if that would violate lifetime equality. But this would mean we could never meet any momentary claims; for once we had achieved lifetime equality, any nonderivative momentary claims would upset that equality.

Conversely, a view that gave lexical priority to momentary claims would also have some problematic implications. What lifetime views get right is the thought that there is a problem of justice in assessing claims to certain kinds of goods from a purely momentary perspective. For instance, a lifetime view can explain why, if we are faced with a patient of eighty, and a patient of eight, each of whom will die without treatment, it misses something important to see both as having the same kind of claim. A view that saw momentary claims as lexically prior would, I take it, say that if we cannot treat both patients, we should simply choose at random. This is because such a view would say that until all momentary claims have been met, we cannot even consider lifetime claims. This differentiates it from a tie-breaker view, outlined below.

A more plausible, though incomplete, suggestion is that each principle may act as a tie breaker in cases where claims from the other principle are equally weighted. In a microallocation healthcare decision between individual patients, this would mean that we treat reasons to prioritize a particular individual as additive. If two patients are equally badly off in a momentary sense—for instance, if they will both die without treatment—then we look to the strength of their lifetime claims.[5] This will effectively mean preferring younger patients and those who have had worse lives overall, just as we would if we ignored momentary claims entirely. But it would rule out, for instance, a policy of allocating pain relief automatically to the patient who has had the greatest lifetime burden of pain; this is because two such patients may not be in the same momentary position. In other words, this idea seems largely in line with the common-sense combination of age-based rationing for life-extending treatments and a predominantly momentary focus for treatments such as pain relief. One worry, however, is that it will not provide sufficient protection against insufficiency for the elderly, giving rise to problems similar to those that affect an appeal to prudence, discussed in section 3.2.

A final possibility, then, is to treat the two kinds of claim as distinct, and not fully commensurable, on the grounds of my earlier argument: that they represent two distinct kinds of perspective that are, for most people, not reducible to one another. One option is to weight chances to be selected for treatment according to both lifetime and momentary claims; this allows some role for lifetime egalitarianism, but also recognizes the importance of momentary sufficiency by giving all those with a particular level of momentary claim a chance to benefit.

The thought behind this is that if people have equal claims, they should receive an equal chance of being selected; a move to unequal, but still important claims should thus prompt us not to act as if one person’s claim has been eliminated entirely, as appealing to tie breakers would, but to move to unequal chances. For instance, if flipping a fair coin would in principle be a fair way to decide between two patients with equal claims—giving them a fifty-fifty chance—then, for two patients with equal momentary claims, but unequal lifetime claims, we would move to a weighted lottery that gave both patients some weight due to their momentary claims, but greater weight to one patient proportional to the greater strength of his or her lifetime claim.[6]

Of course, there are practical issues of efficiency in running a weighted lottery every time a patient conflict arises for indivisible resources, particularly in emergency situations. It is more feasible to allocate greater weight for patients on the basis of lifetime priority for treatments that are allocated by waiting lists as one of a number of features that determine priority. However, there are still problems with applying weighting to such microallocations even in these more congenial circumstances. For instance, such a scheme seems likely to place excessive epistemic and bureaucratic demand on institutions in assessing entire lifetimes of welfare or opportunity for individuals. And as Wolff (1998) points out, egalitarian policies that depend on considerable amounts of personal information about the quality of individual lives might—despite those policies’ good intentions—exacerbate social stigma. Quizzing patients on intimate details of their lives, or even categorizing them as having had bad lives, may be humiliating and degrading even if it is not intended as such. But just as we should be reluctant to abandon momentary claims simply because of practical obstacles, it would also be wrong to focus only on such claims because we are worried about the potential invasiveness of trying to determine people’s lifetime claims.

If we cannot determine lifetime claims reasonably at the microlevel, due to a combination of practical constraints and competing normative concerns, an alternative way to include a concern for lifetime priority is in weighted macroallocations among different groups. When we are deciding how to allocate funds across the entirety of our healthcare budget, giving some weight to proportionality suggests that we ought to give what might otherwise appear to be a disproportionate level of resources (considered simply in terms of the number of people affected or in terms of the total health burden) to conditions that predominantly affect the lifetime worse off, or to target funding at health initiatives for groups who tend to be less well off in lifetime terms, or in geographic areas containing those groups. This allows some considerable role for lifetime egalitarian claims, without those claims “swamping” the role of momentary sufficiency (as an absolute tie break would), and without explicitly picking out particular individuals for lower priority. The cost, of course, is some level of inaccuracy; some individuals picked out in this macroallocation will not be the worst off. But where accuracy competes with intrusiveness and efficiency, trade-offs are inevitable one way or the other.

Even if some microallocations (such as additional criteria for waiting lists) could be framed to avoid this problem, a focus on macroallocations seems at least a plausible minimal role for lifetime priority. Treatments that will predominantly benefit the best off in lifetime terms are a lower research and funding priority than they would be if considered solely on the basis of the number of lives saved, QALYs added to a population, or some other consideration of efficiency. But this does not mean that we should entirely ignore or defund treatments that address momentary claims of the lifetime best off. For they still have considerable claims to those treatments on momentary grounds. In other words, the perspective of momentary justice shows us what is wrong with the claim that we should abandon life-extending treatments for patients above a certain age; it also shows us why egalitarian reasoning does not lead us towards an absolute preference for the worse off in lifetime terms.


As the idea of egalitarianism in healthcare grows more popular in both government policy and social demand, it is vital that we clearly delimit what exactly we mean by “equality.” If we focus only on lifetimes, the language of equality has significant potential to be used to deprive some people of access to fundamental forms of healthcare. At the very least, proponents of equality in healthcare must justify some position on the temporal subject, even if it is to reject anything beyond the pure lifetime view. I have suggested that, for some central goods at which healthcare aims, the most plausible position will incorporate momentary sufficientarianism, and will do so in a way that does not allow it to be over-shadowed—either in principle or in practice—by lifetime egalitarianism. We cannot ignore momentary claims because they represent a distinctive and irreducible aspect of how we view our lives from within. This does not preclude the importance of lifetime egalitarianism, but should instead complement it.