Just Waiting: The Ethics of Queues and Waiting Lists

A. Access Determining Function

The access determining function uses the queue to determine who is eligible or entitled to receive a good or service. Your position in the queue determines whether or not you will receive a good or service. On this model, only those up to the nth number of people in the queue will get the good/service in question, where n is the number that represents the number of people entitled to access a good. Everyone else beyond the nth person gets nothing. The queue determines the answer to an either-or question: you either receive the good, or you do not.

A classic example is the release of limited tickets for a popular concert. The queue serves an access determining function for whether you will receive the concert tickets or not. Those up to a certain n will receive tickets, depending on factors such as stadium capacity and ticket release volume. The queue for a concert taking place in a stadium with a capacity of 10,000 would mean n = 10,000. The first 10,000 people (and only the first 10,000) would get tickets to the concert. The 10,001^st person would not get tickets to the concert. Queues, in this sense, serve an access determining function to determine who gets access to the good.

Queues that serve an access determining function are pervasive in the political sphere. In the United States, permission to employ or enter the US as a temporary non-agricultural worker (via an H-2B visa) is limited by an annual statutory cap. Applications received within a filing window are evaluated but once the cap is reached (the n number), no other applicants will receive an H-2B visa.⁸ In the United Kingdom, the allocation of council houses (publicly funded social housing) is limited by the supply of available lettings in a given time period (the n number), with demand greatly exceeding supply. Only the first n people on the housing waiting list will be offered a tenancy within a certain period, where n corresponds to the number of public housing lettings available. In Scotland, free university places are capped each year by the Scottish Government, and the list serves an access determining function. Only the first n applicants on the list will be eligible for tuition-free university studies every year.⁹

B. Sequencing Function

Queues may serve a more straightforward sequencing function. When queues serve a sequencing function, the same good will eventually be received by everyone in the queue, but the burden incurred in the process of accessing the good might differ. For example, even though all will eventually receive the good, waiting times may differ for individuals in the queue, adding a temporal burden to the process of accessing the good.¹⁰ Queueing for the canteen at lunchtime (where there is no risk of particular food options running out) is an example of a queue serving a straightforward sequencing function. Queueing to board a plane configured in an all-economy class cabin arrangement is another example. Everyone will eventually get on board and receive the same product (i.e. an economy-class seat) but the sequence in which you receive that good is different. Sequencing could be a practical response to dealing with a bottleneck when demand exceeds supply at a particular point in time. In their article on queues, Gil Hersch and Thomas Rowe have in mind what I am calling the sequencing function of queues.¹¹ Contra Hersch and Rowe, however, my argument is that queues are more than just about sequencing.

The political realm is filled with important examples of queues that serve a sequencing function. In most jurisdictions, the wait to receive official government documentation (such as passports and national identity documents) follows a sequencing function for all eligible applicants. Every eligible individual will eventually receive the document they apply for, and the waiting list simply serves a sequencing function. The queue to vote in person on election day in the UK serves a sequencing function rather than an access determining function. As long as eligible voters are in the queue before polls close at 10pm, they are entitled to vote even if they have not yet received their ballot papers.¹² The queue here serves a sequencing function to determine when (not if) you are able to vote.

Queues that serve a sequencing function provide identical goods to all recipients, measured in objective terms, but this is consistent with subjective variations in the queueing experience and in the subjective value of the good eventually received. Consider the wait experienced by victims of state wrongdoing for official compensation. Once the state agrees to set up such a scheme, all eligible victims can receive compensation, with the process typically governed by a sequencing function. Examples of such schemes include the Windrush Compensation Scheme in the UK and the Indian Residential Schools Settlement Agreement in Canada.¹³

On the one hand, waiting a long time might damage the subjective quality of the eventual compensation or apology to be received. The compensation may arrive too late, when the victims are now experiencing deteriorated health and therefore unable to make the most of the payments. On the other hand, the good could become subjectively more valuable as the time goes on. Some victims of state injustice may value institutional recognition and compensation even more over time as their need for closure intensifies. The eventual redress and apology may be experienced as more meaningful after prolonged state denial. More generally, in some cases there will admittedly be a feeling of inconvenience and impatience at having to wait and some subjective variation in the quality of the good to be received, but a queue that serves a sequencing function will mean that the actual good received by queuers will be identical viewed from an objective standpoint.

C. Prioritising Function

Queues can serve a prioritising function where, unlike the sequencing function, the queue determines the quantity or objective quality of the good to be received. The queue determines those who get ‘first dibs’ on something. The prioritising function of queues can therefore be used for a quality-prioritising function or a quantity-prioritising function (or often both these functions at the same time).

Queues that serve a quantity-prioritising function enable those at the front of the queue to get more of a good in question. An example of a queue that serves a quantity-prioritising function is the queue for a wholesale warehouse that sells only one type of good, say, planks of wood. Those at the front of the queue are able to purchase more planks of wood prior to the stock selling out than those at the back of the queue. Many local government authorities rely on a simple queue to serve a quantity-prioritising function when distributing emergency supplies following natural disasters, in part because of the complexities of setting up a more sophisticated queueing system in such circumstances. In the US, local authorities often set up centralised distribution points for people to pick up supplies such as drinking water after natural disasters, but explicitly note that these are only while stocks last and those arriving earlier can obtain more supplies than those arriving later.¹⁴ In the UK, council provision of sandbags for flood defences is generally on a first-come, first-served basis until supplies run out.¹⁵

Queues that serve a quality-prioritising function use the queue to determine the quality of the good that will be received by each person in the queue. Those at the front of the queue have access to higher quality goods than those further back in the queue. An example of a quality-prioritising queue is the line of shoppers at a department store’s big sale. Those at the front of the queue have the opportunity to access the best bargains and buy the best products before those at the back of queue. In the political context, one way of allocating council houses used by many councils and housing associations is the choice-based letting scheme.¹⁶ Under a choice-based letting scheme for council housing, those at the front of the queue have ‘first dibs’ or the right of first refusal over available properties. The queue, in this sense, serves a quality-prioritising function.

The quality-prioritising function of the queue can work in a less direct way where the quality of the good received, while ostensibly the same as what others have received further forward in the queue, is lower in quality the longer one waits in the queue. Consider a person far down on a waiting list for a kidney transplant. At first glance, it may seem that they are part of a queue that serves a sequencing function, because everyone on the list will eventually get the same good, namely, a viable and compatible kidney.¹⁷ However, extensive clinical evidence shows that the longer a patient spends on dialysis waiting for a donor kidney, the worse the outcomes will be for the patient after a transplant.¹⁸ While waiting for a kidney, patients experience cardiovascular, metabolic, and vascular damage that negatively impact their post-transplant outcomes.

Despite the queue ostensibly distributing the same good to all (i.e. a kidney), the waiting list for most organ transplants actually serves a quality-prioritising function. The good the patients on the waiting list are queueing to receive is not just the physical kidney, but the good of a ‘successful transplant’. Phrased in this way, we can see that the waiting list for a kidney transplant serves a quality-prioritising function rather than simply performing a sequencing function. Unlike the sequencing function, where only the subjective quality of the good changes, in this case, the objective quality changes. One task of medical waiting lists, in this sense, is to fairly distribute the burdens of deterioration.

Some theorists implicitly assume that queues are a useful proxy for strength of desire, with those waiting longer having greater need or desire for the good in question. Richard Zeckhauser argues that setting up tedious ordeals (such as queues) can weed out those who genuinely need the good in question from those who do not.¹⁹ This assumption may work in some queues (such as for the latest iPhone or concert tickets) but it is unlikely to hold in all queues. Waiting a long time for a council house may mean that one, in fact, no longer needs the property when one reaches the top of the waiting list (because one has managed to source alternative accommodation in the meantime) and waiting too long for an organ transplant may mean that the body is too damaged to survive an organ transplant procedure.

D. Summary

Table 1 below summarises the three ways queues can function and how they can be used to determine queuers’ claims and rights, noting that the examples provided in each category are designed to be illustrative rather than exhaustive or definitive:

A. Purely Temporal versus Non-Purely Temporal Queues

Organising principles for queues can either be purely temporal or non-purely temporal.²⁰ Purely temporal queues consider only time when deciding how to constitute the queue. In a simple queue governed by the organising principle of FCFS, time is the only relevant consideration when determining one’s place in the queue. Alice arriving before Bill is both necessary and sufficient for Alice to be in front of Bill in the queue. In purely temporal queues, non-temporal considerations such as need cannot be used to displace others in the queue. If Charlotte arrives after Alice and Bill, Charlotte cannot push in front of Alice and Bill simply because she has a greater level of need to be in front of the queue than Alice or Bill.

Non-purely temporal queues, in contrast to purely temporal queues, do not organise the queue based on considerations of time alone. Other considerations are factored in when deciding how to constitute the queue. Two of the most important non-temporal considerations include a person’s level of need and the costs they may have to bear in waiting. A person who joins the queue for the public toilets may be permitted to jump ahead if they are desperately in need due to a medical condition affecting their bladder control. A disabled person joining a physical queue otherwise governed by FCFS may be permitted to go to the front since the burden she would otherwise have to bear of standing in line is unreasonable.

Because it is still a queue, temporal considerations remain important, but other morally relevant considerations can be brought to bear to change the constitution of the queue along non-temporal dimensions. We can see the importance of temporal considerations in at least two ways, even in queues that are not purely temporal. Firstly, time can act as a tiebreaker. If two people in the queue are equal in all other respects (e.g. need and costs), then whichever one arrives first or has spent the most time waiting – a distinction I return to – would be put in front of the queue. The fact that time can act as a tiebreaker in this case highlights the importance it retains, even in non-purely temporal queues. Secondly, time can aggregate against other considerations (such as need) even in non-purely temporal queues.

A paradigmatic example of a non-purely temporal queue in the political context is the allocation process for council housing, which I have already alluded to in my conceptualisation of queues. Time spent waiting is taken into account, but it is only one among many factors taken into account. For example, Islington Council in London determines applicants’ priority on the waiting list by a combination of factors such as need, local links and time spent waiting, with time waited accounting for 5% of an individual’s ranking on the waiting list.²¹ The Scottish Government’s guidance document for allocating council houses uses a similar model, with waiting time considered among several other features and where time can act as a tie-breaker among those with similar levels of need.²² Council housing allocations across the UK more broadly is designed along a non-purely temporal queue.

The queue for an emergency department in a busy hospital is another familiar example of a non-purely temporal queue. Both temporal and non-temporal considerations are at play in organising this queue. Upon arrival in the emergency department, a preliminary medical assessment (known as triage) is done to evaluate your level of clinical need and the relative urgency of your condition. You are then triaged into a category (Category 1, 2, 3, 4 or 5).²³ Category 1 patients have the greatest need and are admitted immediately into the emergency department. These are critically ill patients with immediately life-threatening conditions that require prompt intervention. Categories 2, 3, 4, and 5 have varying degrees of need and urgency, with Category 2 being urgent and serious but not immediately life threatening and, on the other end, Category 5 being neither urgent nor life-threatening.

A Category 1 patient is immediately put into the front of the queue, skipping all other non-Category 1 patients who have been waiting longer. It is for this reason that the emergency department queue is not a purely temporal queue. If there are two Category 2 patients and only one bed is available, then the patient who has been waiting longer will be admitted first. This highlights the role of time as a tiebreaker, even in non-purely temporal queues. A Category 5 patient who has been waiting longer than a Category 4 patient may sometimes displace a Category 4 patient in the order of priority, because of the importance of temporal considerations alongside evaluations of need. A Category 5 patient who has been waiting for hours may be seen ahead of a Category 4 patient who has more recently arrived, for example, despite their condition being more serious and urgent.

Evidence for this kind of aggregation can be seen in the UK government’s ‘See and Treat’ policy for emergency departments, whereby patients with less serious complaints can be streamed into the system instead of following the traditional triage process of waiting until all those with more serious conditions are treated. The rationale for this arrangement is explicitly to ensure that less serious patients are ‘not constantly left waiting in favour of more serious patients’.²⁴ This speeds up the treatment of Category 5 (and 4) patients who will often be seen ahead of Category 4 (and 3) patients who have been waiting for less time, and reflects the way time can aggregate against need even in non-purely temporal queues.

B. Determining Time Zero

The appropriate organisation of a queue requires us to consider when to start counting time for individuals waiting (that is, determining ‘time zero’). Suppose Alice arrives before Bill in the queue for a walk-in clinic for minor health conditions. In a simple sense, Alice is in front of Bill in the queue and should be seen first. Suppose, however, that even though Alice arrives before Bill in the queue, Bill has been suffering with his minor ailment for two weeks, whereas Alice’s is a new injury. If we start counting time spent waiting from when someone incurs an injury, then Bill has spent more time with his condition than Alice, and has been waiting for treatment longer than Alice, thus putting Bill in front of the queue over Alice.²⁵ Simple queues that have a purely temporal organising principle may start counting time spent waiting only upon physically joining the queue, which may miss out those who have notionally been waiting longer prior to joining the queue. The normative significance of when we conceptualise the start of the waiting time and how we determine time zero will become clearer in Section V, when I explore the objection to queues on the grounds that they magnify injustice.

C. Parallel Queues

In many cases, multiple queues operate simultaneously. For example, there may be multiple queues at the supermarket for each checkout till. Parallel queues do not challenge my conceptualisation of queues thus far. The function and organising principle of the queue could operate at the level of each individual queue or at the level of the overall system. In the supermarket checkout example, each individual queue operates according to a sequencing function, with individuals having discretion over which queue to join. Once they join a particular queue, they obey the sequencing function of that queue, which may go faster or slower than other queues. In a simple airport security queue, passengers often start in one giant queue before being funnelled into parallel queues for different security lanes and scanners. In such cases, the function and organising principle of the parallel queues operate at the broader system level, regulating everyone’s overall sequence, rather than at the level of each individual security lane.

A. Temporal Justice

The first and perhaps strongest argument for the moral significance of queues is that time (in some form) is an issue that should be governed by principles of distributive justice. In his discussion of temporal justice, Robert Goodin argues that time should be a concern of justice.²⁶ Julie Rose defends the value of leisure and free time, arguing that these should be distributed according to principles of justice.²⁷ Elizabeth Cohen talks about the distributive dimensions of time within the political sphere.²⁸ Elsewhere, I defend the relevance of temporal burdens for theories of distributive justice.²⁹ Queues concern temporal considerations at least in some form, and they are morally significant for this reason. If time is independently valuable, then the amount of time people spend waiting is a burden that may engage questions of distributive justice that principles of queueing provide guidance on.

Some kinds of queues, it might be argued, do not raise considerations of temporal justice because the person does not need to be physically there and is not having their time encroached upon. For example, waiting in virtual queues or for online applications to be dealt with in the order they are filed may not encroach on the time of those waiting. However, seeing such queues as not engaging temporal justice considerations would be to understand temporal burdens in too narrow a sense. A physical queue involves costs in terms of foregone time and aching legs, but a non-physical queue still involves temporal costs in terms of uncertainty, having something ‘hanging over our head’ and needing to restrict our activities in preparation to making it to the front of the queue. We may hang around the telephone or letterbox in anticipation of an appointment letter or restrict travel to be able to quickly accept and move into a council house that becomes available. Waiting in a non-physical queue is therefore not costless and involves temporal burdens.

Nor can we brush aside issues of temporal justice on the grounds that queues equalise waiting time for everyone. This implausibly assumes that queues are equally long at each point in time, which is unlikely to hold in the real world. Someone arriving at the canteen in the middle of the lunch rush will be waiting longer than someone arriving at the tail end, and the waiting list for council housing is likely to be longer in some parts of the country compared to others. The costs of waiting also vary for different people. The idea that queues necessarily promote temporal justice by equalising temporal costs is therefore untenable.

B. Fairness

The second argument for the moral significance of queues relates to their propensity to realise fairness. Even if the practice of queueing itself does not have any intrinsic moral value, we have instrumental moral reasons to abide by its rules given its propensity to realise a fair and efficient allocation of vital goods. Tyler John and Joseph Millum argue that as long as the queue system is sufficiently just, we have a duty of fair play to abide by the system’s rules and wait in the queue.³⁰ John and Millum develop this argument by showing that waiting in the queue meets all of John Rawls’ criteria for an obligation of fair play. As Rawls notes:

The main idea is that when a number of persons engage in a mutually advantageous cooperative venture according to rules, and thus restrict their liberty in ways necessary to yield advantages for all, those who have submitted to these restrictions have a right to a similar acquiescence on the part of those who have benefited from their submission. We are not to gain from the cooperative labors of others without doing our fair share.³¹

With this characterisation in mind, a queue is an active scheme of cooperation, broadly defined.³² A queue involves a restriction of liberty because everyone except the person at the very front must have their freedom restricted and be prevented from simply going straight up. A queue is a mutually advantageous form of cooperation because, in general and especially over time, it allocates goods much more efficiently and effectively for all than if the system were a free-for-all. Queue-jumping is a form of free-riding, in the sense that the queuer-jumper is only able to get herself to the front because others are already forming an orderly queue instead of chaotically trying to access the good in question. The queuer-jumper cannot jump the queue if there is no queue in the first place. The opportunity to jump the queue, then, is contingent upon others abiding by a duty of fair play to queue in an orderly way and wait their turn.

C. Symbolism of Relational Equality

The third argument for the moral significance of queues relates to their expressive function and symbolism of relational equality. Queueing has an important expressive dimension, expressing respect to our fellow citizens. In his sociological study of queue-jumping in rural India, Stuart Corbridge captures the way queueing can come to signify inferior status:

Many poorer people – and poor women especially – can be seen waiting patiently outside government buildings. … It is not unusual for them to be kept waiting for hours, and sometimes for days. During the course of this wait they will often see local political bosses (netas) storm into the office of the highest-ranking local government official to demand an audience. … Waiting is something that poorer people do, and more than once we witnessed adivasi (tribal) women standing in the sun or rain for hours waiting their turn to see [government officials] sometimes refusing to go for lunch in case they ‘lost their place.’³³

The political elite use their status to jump the queue without the thought that those in the queue are their fellow citizens with equal claims to an audience with government officials. As Corbidge puts it, ‘members of the upper castes … simply do not recognise the claims to equality that standing in line would seem to imply’.³⁴ Waiting in line has an expressive dimension of connoting respect for our fellow citizens, whom we regard as having an equivalent moral claim to the good in question. Conversely, jumping the queue connotes a sense of superiority over others who are waiting their turn. This is contrary to the ideal of democratic citizenship and civic friendship and violates the notion of equal moral status.³⁵ Queues serve an important moral function in expressing this respect for fellow citizens.

It is for this reason that paying someone to wait in line for us, even if we do not directly push in, similarly expresses disrespect for others. It implies that we are different to those waiting in line and do not need to comply with the cooperative scheme of queueing. As Michael Sandel observes, such arrangements are often hidden or made to look as discreet as possible, because those paying for others to hold their place know that they are acting disrespectfully towards others in the queue.³⁶ At the same time, the queue is not unique in expressing respect to others. The expressive argument extends beyond the context of queues and covers other social and political situations, as pointed out by egalitarian theorists.³⁷

D. Abiding by Social Norms

A fourth and final argument for the moral significance of queues comes from the importance of social norms. Laura Valentini argues that we have a pro tanto reason to comply with socially constructed norms, independently of the consequences of us doing so.³⁸ We have content-independent reasons to abide by social norms because doing so respects fellow citizens’ agency, when those fellow citizens have made a commitment to publicly accept the norm on the basis of others doing the same. The queue is a social norm in some sense in most, if not all, parts of the world, governed by internal norms and expectations.³⁹ Similar to the arguments based on fairness and the expression of relational equality, the argument from social norms are based on respecting others who abide by the norms of the queue.

These four arguments motivate the case that queues are morally significant, both in terms of justice and non-justice-based considerations. My view is that all four reasons obtain, but even if only one of the foregoing arguments about the moral significance of queues is sound, this has important implications for our understanding of the normative significance of queues.

A. Injustice Objection

The first objection to the use of queues in distributive justice is that they are unjust. Govind Persad, Alan Wertheimer, and Ezekiel Emanuel argue that queues (which they use interchangeably with distributing by FCFS) should be rejected as a relevant principle for the allocation of scarce health resources.⁴⁰ Their objections are generalisable beyond health and seem to apply to distributive justice more generally. Persad and colleagues’ main objection to queues is that they are liable to be manipulated by those who are able to join the queue or manipulate their position within it. Doctors can exaggerate patients’ waiting time or put patients on waiting lists much earlier, to give them an advantage over those who join the queue later. ‘First-come, first-served allows morally irrelevant qualities – such as wealth, power, and connections – to decide who receives scarce interventions, and is therefore practically flawed.’⁴¹ Persad and colleagues also state that queues ‘ignore other relevant principles’.⁴²

Firstly, notice that Persad and colleagues’ arguments rely upon the view that a queue distributes only according to FCFS. That is, they assume that the organising principle for all queues is a purely temporal one. If queues distribute only according to FCFS, then of course, by stipulation, such a method of allocation would not consider other morally relevant principles. Persad and colleagues’ objections to queues, then, relies upon a mistaken understanding of the concept of queues and their view that a queue excludes other morally relevant features is merely an argument by definitional fiat.

Secondly, and relatedly, Persad and colleagues’ arguments should be seen as objections against the principle of FCFS and a particular way of measuring time spent waiting rather than against the idea of queues more generally. Some theorists note that FCFS may disadvantage disabled individuals, who will find it much harder to join a physical queue.⁴³ Persad and colleagues’ objections have a nugget of truth in it, capturing this concern. But their objection is limited to an attack on the principle of FCFS and not against the idea of queues more generally.

Thirdly, Persad and colleagues’ objections against queues, if true, would seem to apply to nearly all principles for allocating healthcare resources (and perhaps nearly all principles of distributive justice more generally). Patients (or doctors) might exaggerate their own (or their patients’) level of need in the same way they might exaggerate how long they have been waiting. Some patients might exaggerate how much pain they are in or play up the impact of the condition on their daily lives in an effort to get treated quicker. It is highly likely that those with greater wealth, power and connections will be in a better position to manipulate how they present their level of need. From this possibility, I doubt Persad and colleagues would say that distributing by the principle of need is practically flawed and should be abandoned. The fact that a principle is prone to be abused in practice does not mean that we should jettison the principle rather than think about how it could be better designed to protect against these abuses. Their objection to queues, therefore, is overinclusive and would apply to any principle prone to being abused in practice.

Olga Jonasson argues more forcefully against the use of waiting lists on grounds of injustice. She notes that significant numbers of patients will never make it on to the waiting list in the first place, especially poor and non-white patients.⁴⁴ James Childress also notes a similar concern around access to waiting lists in the first place, in many parts of the US.⁴⁵ Writing in the context of the waiting list for kidney transplants, Jonasson states that ‘far from objective, time on the waiting list is often abused by patients and by transplant centers’ and that ‘length of time on the waiting list is the least fair, most easily manipulated, and most mindless of all methods of organ allocation’.⁴⁶ Jonasson points to the issues with determining time zero, that is, when a patient becomes eligible for an organ transplant. More advantaged and affluent patients tend to be entered into the waiting list for organ transplants much earlier than those who are more disadvantaged. Using waiting time as a criterion for transplantation priority therefore means more advantaged patients are prioritised for organ transplants, who tend to be entered into the waiting list much earlier in their clinical journey than poor, non-white, and older patients.⁴⁷ Queues and waiting lists therefore simply mirror or magnify existing disadvantage and injustice.

Jonasson’s critiques, while powerful, are not fatal to the normative idea of queues. Firstly, despite Jonasson acknowledging the problems around determining time zero, her account nevertheless assumes that the waiting list is organised purely according to temporal considerations. The same response I made to Persad and colleagues can therefore also be offered here: Jonasson’s critique might land against purely temporal queues, but they do not necessarily land against the broader idea of queues as I have conceptualised them, where broader non-temporal considerations are relevant.

Secondly, Jonasson’s critique of waiting time does not focus on the inherent wrongs of using time as a consideration but instead identifies the contingent ways in which queues might be unjust given background social injustice. A proper reading of Jonasson’s critique should therefore see it as targeting background social injustice and revising how we determine time zero rather than waiting lists as such. If the background conditions of society were sufficiently just, and if patients were entered into the waiting list in accordance with criteria that were transparent, consistent, and fair, then Jonasson’s complaints would evaporate.⁴⁸ The faults Jonasson identifies are thus not inherent to queues or waiting lists, but merely justified complaints about the contingent ways in which queues can be abused in our non-ideal real world.

None of this is to trivialise the significant ways in which social injustice can manifest in how queues are designed in ways that impede our distributive claims. This is not unlike the observation that many institutions are less accessible to disadvantaged groups. From this observation, however, it remains an open question whether what ought to be done is for those institutions to be abolished or instead for access and fairness to be improved for those disadvantaged groups. In opposing institutional injustice of these kinds, we must be careful not to throw the baby out with the bathwater.

Thirdly, it is possible to design queues and waiting lists with the reality of background injustice in mind, including issues around access Jonasson has in mind. New Zealand had a brief policy where patients’ ethnicities were taken into account to decide their position on waiting lists.⁴⁹ Disadvantaged groups such as Māori and Pacific Islander patients were given higher priority on waiting lists for non-urgent elective surgery to combat known barriers these groups have in accessing primary care and their subsequently delayed referrals to secondary care. Regardless of whether or not we agree with this policy, the waiting list was designed in a way that was attuned to the observations of injustice that Jonasson notes. It is therefore possible, in both principle and practice, for queues to respond to background injustice and to be designed in a way that takes into account the concerns Jonasson has in mind.

B. Redundancy Objection

My response to the injustice objection above and my conceptualisation of queues more generally challenge the predominant view that queues are simply situations where FCFS applies. FCFS may be the default organising principle of a queue, but it does not follow that this is the only relevant moral principle that governs the practice of a queue. This conceptualisation, however, may open me up to an important objection around redundancy. If queues are not merely situations where FCFS applies, but instead situations where other normative considerations are at play, what value does the concept of a queue actually add? The queue would seem to be normatively redundant, where it simply refers to the strength of people’s different distributive claims – situations that are the bread and butter of standard theories of justice. This is what I call the redundancy objection.

Consider a queue governed by a non-purely temporal organising principle based on need. The queue is organised so that those with the greatest level of need are placed at the front of the queue and those with the lowest level of need at the back of the queue. If the queue simply tracks the strength of our needs-based claims relative to others, the objection could go, then what does the concept of a queue add over and above a needs-based distributive principle? The concept of the queue, it would appear, is normatively redundant. This is a powerful objection, but I shall show that it does not succeed. Far from being normatively redundant, queues add an important normative dimension from the standpoint of ethics and distributive justice.

Firstly, queues provide an organising structure for realising individuals’ distributive claims when not all of them can be realised immediately. A queue, especially the institutionalised variant of a waiting list, is a way to formalise and organise the nature of different people’s needs-based claims in a way that is action-guiding, reliable and transparent. If we treat needs-based justice as the ideal distributive principle, then the queue is the social or institutional mechanism by which individuals’ needs-based claims can be organised and fulfilled in turn. The queue is the institutional arrangement that operationalises the demands of distributive justice, whatever those demands might be. This shows that even sceptics of queues cannot reject them for being normatively redundant, when they serve an important organising function.

Secondly, a queue (or waiting list) governed by a needs-based organising principle is still distinct from a non-queue-based system that ranks individuals purely by their level of need. A needs-based queue must still consider temporal features, with time spent waiting able to act as a tiebreaker in cases where there is equal need. In certain cases, time spent waiting can also be aggregated against need, such that a person with a marginally lower level of need who has been waiting for a very long time could be put ahead of someone in marginally greater need who has been waiting for less time. A purely needs-based theory of distribution, in which people are ranked only according to need, would not capture the importance of temporal considerations. It would risk a bottomless pit objection, familiar to those who support absolute prioritarianism, whereby we are required to keep devoting significant resources to the worst-off individual until they are at the level of the next worse off individual, when those same resources could help countless other people who are slightly less worse off.⁵⁰

Queues, in virtue of focusing on temporal considerations, can provide a principled response to why we are sometimes permitted to prioritise a person with lower need over a person with greater need, when the former has been waiting for a longer time than the latter. The claim here is not that this is the correct way to organise the distribution of a particular good, but to point out that there is a normative distinction between a purely needs-based theory and a needs-based queue, and that the charge of normative redundancy therefore fails. The conceptual framework and typology of queues I developed in Sections I–III helps us respond to Jonasson and others, showing that the organising principle of a queue need not necessarily be purely temporal.

Thirdly, and most fundamentally, the moral norms that govern queues are not necessarily co-extensive with the demands of justice. Queues capture distinct considerations of normativity that are not necessarily matters of justice, so the objection that the queue is normatively redundant and merely captures considerations of distributive justice is mistaken. Consider the British practice of queueing at bus stops or waiting your turn to be served at the pub. It is considered polite to board the bus roughly in order of arrival at the bus stop, and to let those who arrive at the bar first to be served before you. Jumping the queue at the bus stop or pub will generate looks of disapproval or condemnation from fellow queuers, but it is not clear that an injustice has been committed by the queue-jumper. If the queue is simply justice operationalised, then an injustice has been committed by the queue-jumper – but the complaint is not one grounded in the language of justice, so there must be something else normatively at play.

One might insist that there is an injustice that has been committed, albeit of a very minor scale. After all, if free time is a relevant distribuendum of justice, then it would seem that even the pub queue-jumper has affected our distributive claim to this good. This objection is too quick. The fact that something is a distribuendum of justice does not necessarily mean that any and all instances where I am deprived of it warrants the language of injustice. If I am on a bus service running a few minutes late because the driver missed her alarm and arrived to work late, the fact that I am ‘losing time’ does not necessarily mean that I am the victim of an injustice. Similarly, money is a distribuendum of justice, but the fact that I am under-tipped occasionally as a waiter (in a context with a strong tipping culture and absent any discriminatory intent) does not necessarily mean that I have suffered an injustice.

The moral status of queue jumping illustrates the issue of the normativity of queues more generally. There are two broad sets of reasons why queue jumping may be objectionable, based on what we can call external and internal reasons. External reasons trace the moral wrong of queue jumping to the violation of an independent and more general principle of distributive justice. What is wrong about jumping the queue of a needs-based queue, on the external account, is that it is contrary to the principle of distributing by need. Internal reasons, by contrast, trace the wrong of queue jumping to the violation of the norms of the queue and the integrity of the ‘rules of the game’.⁵¹

The external account raises questions about the fairness of the queue. They concern questions of distributive justice about whether the queue is appropriately constituted in the first place – that is, whether it has the right organising principle. The internal account raises questions about fairness in the queue. They are concerned with the fact that, given that we have a certain organising principle for this queue, is that organising principle being complied with. Given that a certain procedure is in place, the fact that certain others are not complying with it could be an indicator of unfairness through free riding. The fact that these two considerations can diverge helps to show that the queue is not normatively redundant, in the sense that not all issues captured by queues are about external concerns of general distributive justice.

Jumping the queue for a bus or pint of beer may not trigger external reasons for moral condemnation, but they may nevertheless raise internal reasons for objections. As I explained, they may express the wrong attitude about the importance of social norms or be a form of objectionable free riding. On the other hand, attempting to bypass the queue for an organ transplant – for example, by rigging the process or dishonestly reporting your medical condition – would trigger both external and internal reasons for moral condemnation. It violates an external distributive principle (such as distributing by need) as well as the internal rule based on abiding by the fair rules of the system.

One might push further and note that the four reasons I outlined for the moral significance of queues in Section IV are not distinctly considerations about queueing and are instead parasitic on considerations that are external to the practice of queueing. If so, then it seems that the queue is normatively redundant and relies upon other considerations external to it. This relies upon an overly strong account of what it means for a theory not to be redundant. Consider two senses in which we can accuse a theory of being normatively redundant. The first sense is that a theory is redundant if it is not fully freestanding or non-derivative. The second sense is that a theory is redundant if it adds nothing useful to our conceptual or normative landscape.

The redundancy objection assumes that the first sense of redundancy is the correct standard by which my account should be judged, but this is an unnecessarily high bar. If political philosophy is simply ethics for the political sphere as some moralists believe, for example, then it would mean political philosophy is redundant – and I doubt we want to say that. The question is not whether a normative theory or concept is fully freestanding or entirely non-derivative, but whether or not they add something useful normatively, even if they are parasitic on, or derivative of, other normative values at the foundational level. My argument is that we should view the queue in this light. Even if the argument in its support are ultimately derived from other normative considerations, queues provide us with important normative insights and are a normatively important institutional arrangement. Queues have an important role in ethics and political philosophy, but they are not designed to be fully freestanding frameworks, separate from broader considerations about what we owe to each other.