Care for a Profit?

We vindicate the widespread intuition that there is something morally problematic with for-profit corporations providing care to young children and elders. But instead of putting forward an empirical argument showing that for-profit corporations score worse than not-for-profits when it comes to meeting the basic needs of these vulnerable groups, we develop a philosophical argument about the nature of the relationship between a care organisation, its role-occupants, and care recipients. We argue that the correlation between profit and lower-quality care is a result of intrinsic features of a for-profit model, combined with conceptual features of meaningful caring relationships, such that non-profits are the most reliable institutional providers of adequate care. Our claim is that care requires a kind of commitment that for-profit institutions are constituted to avoid, and that non-profit institutions are constituted to embrace.

T he COVID-19 crisis has highlighted many socioeconomic challenges facing contemporary societies. One issue that has been brought to the fore with immense force is the need of care for those at early and late stages of their lives, and the fact that many societies have failed to create the institutional conditions for adequate caregiving to take place, especially in times of crisis. Indeed, the high number of deaths in aged care centers in the United States and Europe, the role of insecurelyemployed carers in spreading the disease in parts of Australia, and the immense burden placed on working parents once childcare centers closed around the world have raised the question of whether the "caring architecture" of industrialized capitalist states is adequate for performing the essential work of caring for those who cannot care for themselves.
While the crisis has highlighted these issues, they were already coming under scrutiny before. Even prior to 2020, many governments had commissioned reports investigating the standards that govern the care sector, raising questions around the ratio of workers to residents or children, workers' qualifications and working conditions, as well as inquiring into the procedures centers have in place to avoid young children and elders suffering abuse and neglect. 1 But the COVID-19 crisis has led many commentators to raise a more fundamental critique: that problems with the care sector arise when it relies on a forprofit model of caregiving (Star 2020; Bachelard 2020). Because there is empirical evidence suggesting that there is an inverse correlation between quality of care and profit (with non-profit centers performing better than for-profit on average), such commentators have insisted that societies can only become more caring by moving away from a for-profit model of care (Stall et al. 2020). The empirical claim behind this concern was highlighted in Spain in 2020, where the worst outbreaks of COVID-19 took place in for-profit aged care centers, in which elderly sick residents were abandoned by their caregivers and left to die by themselves, as opposed to in non-profit centres (Rada 2020).
This critique is good as far as it goes, but it does not go very far. The alleged correlation between profit and inadequate care does not in fact establish that the problem can only be addressed by moving away from a for-profit model of care. Instead, the inverse correlation may be a result of the fact that governments do not have the right legislative framework in place and do not divert enough tax revenue into the system. And, of course, if the problem is simply that profit takes away resources from caring, then all that is needed is a government that caps profit and pours additional resources into the sector. More government funding, combined with legislation that forces care providers to spend those resources on caregiving, will then solve the problem at hand. The upshot of this critique is that when a care organization fails to provide adequate care to young children or the elderly, they do so because they are poorly regulated or underfunded. 2 In this essay we highlight a deeper problem with the for-profit model of care. Unlike the critics of the forprofit model who make an empirical claim about the negative effects of profit, we develop a conceptual argument against profit in care that withstands the injection of additional resources or revenue-cap regulation. We argue that the correlation between profit and lowerquality care is a result of intrinsic features of a for-profit model, combined with conceptual features of meaningful caring relationships, such that non-profits are the most reliable institutional providers of care because they have the right kind of robustness. 3 Our claim is that-philosophically and conceptually-care requires a kind of robust commitment that for-profit institutions are constituted to avoid, and that non-profit institutions are constituted to embrace. 4 Our essay is structured in four parts. We begin by discussing the caring needs of the very young and very old, and show that they go well beyond having one's basic needs met. 5 We argue that a general focus on "care" must sometimes be replaced with a specific focus on (what we call) "meaningful caring relationships." We then argue that adequate meaningful caring relationships conceptually require robust caring attitudes and commitments on the part of caregivers-which, in an organizational context, requires robust caring attitudes and commitments from the organization itself. Next, we show that due to intrinsic features of the for-profit model, it cannot provide these robust caring attitudes and commitments. Finally, we discuss an important objection to our argument by rejecting the claim that a pecuniary motive in itself is wrong in the provision of care. Instead, we argue that such a motive blocks the conditions for the provision of meaningful caring relationships when the motives are held by a collective agent as opposed to an individual.
The upshot of our discussion is that if we want citizens at early and late stages of their lives to be well cared for, we need to eliminate for-profit care organizations from the picture, and create a system geared towards the development of a meaningful caring relationship between caregivers and care-receivers, where the former are adequately supported in developing robust caring attitudes towards the latter.

Good Lives
Childcare centres and aged care centres provide the very young and the very old with care so that such age groups can have their basic needs met, even when family members or other loved ones (most notably parents and adult children) are unable or unwilling to fulfill their needs. Although these settings differ when it comes to some of the additional goals they have been set up to pursue (e.g., education in the case of children and housing in the case of elders), they both share a core mission. This mission is ensuring that citizens, who are unable to meet their own basic needs due to cognitive or physical constraints associated with a given life stage (and who cannot have these needs met primarily or partially by loved ones), are able to have those needs adequately met by paid caregivers.
Workers in these centres assist those under their care to perform daily activities that many citizens take for granted. Central biological activities such as eating and drinking, taking medicine, getting adequately dressed for the weather, sleeping, and disposing of one's urine and faeces form the core of what takes places in such centres. Care facilities thereby play a vital role in the larger socioinfrastructure developed over the years for the purpose of ensuring that people receive life-sustaining care at early and late stages of their lives.
The fact that the very young and very old need support so as not to die of hunger, thirst, cold, heat, or preventable illnesses, however, might obscure the fact that these groups have a non-biological interest that should also be protected and promoted in such centres: the interest in leading good lives as a member of a particular age group (Gheaus 2015). Simply put, young children have an interest in leading good lives qua young children and elders have an interest in leading good lives qua elders. These interests are present irrespective of how a child's life will go once she grows up, and irrespective of the sort of life an elder has already had. A child who will go on to achieve amazing things as an adult still has an interest in her childhood going well. Similarly, an elder who has lived a wonderful life since birth has an interest in her final years being good.
But what does it mean for these groups to lead good lives qua children and qua elders? The question of what it means for a life to go well is notoriously contentious. Here, we follow Susan Wolf in thinking that the ability to engage with some meaningful projects and relationships is necessary for a good life (1997a; 1997b; 2010). For Wolf, a "meaningful" project or relationship is one that is both subjectively attractive and objectively good (ibid). Importantly, "objective goodness" can take on a huge variety of forms (Wolf 2010, 47). Thus, a commitment to objective goodness (in both projects and relationships) is consistent with proper attendance to the Rawlsian "burdens of 626 Perspectives on Politics judgment" (Rawls 1996, 36-37, 55-57). Here we will assume that the opportunity to engage with meaningful projects and relationships at each life stage is important: a child or elder who has zero opportunity for meaningful projects and relationships leads a bad life at that time, even if there are numerous meaningful projects or relationships in their past or future. Such individuals deserve what support society can reasonably provide them to find meaning.
Subjective attraction and objective goodness can usually be relativised to a life-stage. For example, the projects and relationships that are subjectively attractive and objectively good for a three-year-old differ from those of a thirty-yearold, which differ again from those of a ninety-year-old. Despite this relativisation, meaningful projects and relationships must be extended through some period of time: engagement should not be sporadic, but rather ongoing and open-ended. This is partly because the life stage as a whole must benefit from meaningful projects and relationships, and partly because the projects and relationships that give meaning to our lives tend to extend over time. By way of illustration, consider a young child's extended project of learning to write the alphabet, or an elder's life-long relationship with a sibling.
Clearly, many meaningful projects and relationships extend across multiple life-stages (as the siblinghood example demonstrates). Moreover, the neat division of a life into "stages" is somewhat arbitrary. Still, we operate with a rough notion of life-stages, and with the idea that meaningful projects and relationships are usually somewhat relativised to life-stages, in order to enable a focus on the early and late years of life as distinctive. As a rough approximation, we focus on children under the age of five and aged care residents who have undergone significant cognitive or physical decline.
When thinking about meaning in early and late life, one obvious problem is that the same intrinsic features of those life stages that put them at odds with a life of independence also mean that the very young and the very old are constrained in the number and type of meaningful projects they can pursue. Because of their cognitive or physical abilities, these groups struggle to engage with many projects that are both subjectively attractive and objectively good. That is, these two groups are often cognitively or physically unable to engage with the sort of intellectual, physical, and artistic enterprises that often give meaning to adult lives.
And yet it is implausible that such groups lead fulfilling human lives by merely engaging with pleasurable activities that are devoid of meaning. Most people would be appalled by a system that plugged young children and elders into an "experience machine" full of pleasurable sensations, until they became more independent or died. This is because members of these groups have an interest in engaging with life in ways that are not only subjectively attractive, but also objectively good, that is, worthy of such attraction.
If we are right in suggesting that pleasure will not suffice in early childhood and the final stages of adulthood, whilst recognizing that those living those life stages have open to them only a narrow range of meaningful projects, how do we ensure that they can lead good lives? The answer lies in the opportunity for enjoying meaningful relationships. The relationships we have in mind are those that are subjectively attractive and objectively good (again, with a broad and varied interpretation of objective goodness), such as the relationship between two elderly siblings we already alluded to. Indeed, two siblings who enjoyed each other's company in early childhood and who continue to do so in old age enjoy a relationship that was meaningful at the start of their lives, and that continues to be meaningful now in their final years.
But meaningful relationships are not only relationships between intimates. Another meaningful relationship is that between a child and a carer, where the latter delights in the former's exploration of the world, and attends to her basic needs with attention, sympathy, patience, affection, and full recognition of her worth. Another example is the relationship between an elder and a carer who is respectful, empathetic, genuinely interested in how she is feeling on the day, and who wants her to feel at ease as she negotiates the challenges of progressively becoming less physically and cognitively capable. For many children and elders, relationships with carers take up a large portion of their waking hours, compared to other relationships. At these life stages, then, caring relationships loom large for three reasons: first, the care recipient is constrained in the number of meaningful projects they can pursue; second, the caring relationship is inescapable insofar as the child or elder cannot meet their basic needs by themselves; and third, caring relationships are relatively time-consuming for these groups.
Still, it is important to emphasize that we are not suggesting that caring relationships should be imposed on those who are not subjectively attracted to them. If a child is very shy and would prefer very little social contact, carers should follow her lead. Similarly, if an elder would rather have a more impersonal relationship with her carers, her aged care center should facilitate that. After all, such meaningful caring relationships only contribute to a good life if they are endorsed by those who partake in it, and not if they are forced from the outside (Wolf 1997a; see also Raz 1986;Olsaretti 2005;Kagan 2009). What matters for our purpose is that such relationships are available for those in child and aged care centres-so that those who are in fact subjectively attracted to them, can in fact enjoy them during the life stage in question. Furthermore, such relationships should be available in all care organizations, not just in some of them. This is because any care recipient's subjective attractions might change-the shy child might become lonely; the reserved elder might start to crave a deeper connection-and it is extremely difficult to move between caring organizations, especially for elders. 6 We also don't mean to suggest that caring relationships are typically as valuable, and by implication as meaningful, as the relationships young children and elders may enjoy with family members or friends, including those who attend the same care institution. The point is simply that carers who have meaningful caring relationships with these care recipients contribute not only to the protection of basic biological needs, but also to a meaningful life. And the more constrained a person is in terms of undertaking projects that are meaningful, the more important it is to partake in relationships that make life go well. Similarly, the more time one spends in such caring settings, the more important it is to enjoy meaningful relationships with those one necessarily interacts with. 7 Having said that, we don't mean to imply that it is morally wrong for any given carer to choose not to develop and maintain meaningful caring relationships with those they care for. One can refrain from being abusive and neglectful, without developing a fully rounded caring relationship with those for whom one performs caring activities. The point is simply that in such cases, the very young and the very old are being provided with a service, rather than partaking in a caring relationship which adds meaning to their life. Whereas the former can take place in ways that are morally acceptable, and that have minimally positive effects on well-being, only the latter contributes to a meaningful life. As we will explain later, the organization can ensure that each care recipient has the opportunity for meaningful caring relationships, without each and every care giver in the organization having to stand ready to provide them.
To be sure, the distinction between a service and a relationship is difficult to draw precisely. One key test is replaceability: if identical actions were performed by a different carer, would those actions bring just as much value to the recipient's life? The more the answer is "yes," the more what's being provided is a service rather than a relationship. Another mark of the distinction concerns concrete versus diffuse goals: in a service, there are concrete and measurable deliverables; relationships are more diffuse and open-ended in their goals and outputs. The difference between a relationship and a service is more of a continuum than a sharp distinction. The variables underlying the continuum warrant more attention in future work. Nonetheless, we suggest that the opportunity for relationships (not just services) is of crucial importance for a meaningful life.
One might wonder: if a person lacks the cognitive and physical capacities necessary for many meaningful projects, then won't they also lack the capacity for meaningful relationships? In response, we wish to make two points.
First, we do not claim that children and elders can engage in zero meaningful projects: learning to spell one's name (for children), or completing a craft project (for elders) are meaningful projects. We simply suggest that these groups are greatly limited (compared to other age groups) in the meaningful projects that are options for them. This limitation makes meaningful relationships all the more important for these groups, and so produces a strong imperative that society creates opportunities for meaningful relationships insofar as this is feasible. Second, projects and relationships often require different capacities. Relational capacities include the capacity for shared emotion, for requesting help, and for physical touch. Project-focused capacities include the capacity for building up a bank of knowledge, planning and persevering, and recalling specific memories (of earlier projectstages). Thus, relationships are an important source of meaning for the young and old, because they are worthy of attraction, and because relational capacities are often more accessible to these groups than project-focused capacities.

Caring Organizations
We suggested earlier that carers are important potential participants in meaningful relationships with those they care for, particularly when care recipients are young or elderly. However, in the context of childcare centres or aged-care homes, we must ask not just how care occurs at the individual level, but also how it occurs at the level of the organization. After all, as we will discuss, the organization guides and constrains individual carers.
Care ethicists have done important work that helps map this dynamic. Perhaps most relevant is Joan Tronto's influential taxonomy of four "phases" of care (Tronto 1993, esp. 105). These phases are "caring about," "taking care of," "caregiving," and "care receiving." Caring about "involves the recognition in the first place that care is necessary. It involves noting the existence of a need and making an assessment that this need should be met" (Tronto 1993, 106). Next, "taking care of " involves "assuming some responsibility for the identified need and determining how to respond to it" (Tronto 1993, 106). Third, caregiving is the actual hands-on work of fulfilling the need. Finally, care receiving involves a mutual recognition "that the object of care will be affected by the care it [she, he, they] receives" (Tronto 1993, 107; see also Kittay 2014 on "the completion of care").
Tronto's taxonomy appears easily applicable to the organizational context of well-functioning childcare and aged-care centres, where the needs of young children and elders are recognised and attended to, and where different parties play different roles in the provision of care-from those who interact directly with care receivers, to those who cook meals or clean their physical environment. Following Tronto, one might think that the individual 628 Perspectives on Politics caregivers perform the caring about and the caregiving, whilst the organization itself takes care of young children and elders in the sense of taking responsibility for ensuring that there are sufficient caregivers and support workers on the ground.
We believe that the role of the organization is more pervasive than this. Notice that individual caregiverswhen they are employees in care organizations-are not autonomous and resource-rich. Instead, the individual caregiver is playing a role in the organization's division of labour. Here, it helps to consider how organizations operate as agents in the world. Organizations are not freefloating entities that can be ontologically separated from those who enact their plans and policies. Instead, organizations are physically constituted by the individuals who occupy roles in the organizational chart (Hess 2018;Hindriks 2021;Collins n.d.). When these role-occupants perform their roles, the organization is performing an action (Collins 2018)-at least when the role-occupant construes themselves as such (Lackey 2018). Consider that when a professor teaches a unit, her department teaches a unit; when a lawyer gives advice, her firm gives adviceand so on.
If this ontological picture of organizations is correct, then the organization is not just the one that takes care of the care recipient. Instead, the organization should also be attributed role-occupants' actions (at least, role-occupants' actions when they are performing their roles and intend to act on behalf of the organization). Thus, the organization is also the one that cares about care recipients and gives care to them. After all, when a role-occupant intentionally does these things with the organizational "hat" on, then the organization does these things. In this way, the organization is more present throughout in the caring phases than one might think. Many particular (token) attitudes, activities, processes, and practices are attributable both to the individual role-occupant and to the organization in which she has a role. The organization is there throughout the caring process, orchestrating and constraining individual role-occupants, whose role-bound activities are attributable to the organization.
The term "constraining" is particularly important for our purposes. The organization's rules, resources, and procedures limit the ways in which role-occupants can perform their organizational roles. Thus, it's not just that the organization acts when the role-occupant acts, or gives role-occupants the resources with which to act. It's also that the organization greatly constrains and influences the ways in which the role-occupant can act-and, indeed, the attitudes a role-occupant can legitimately develop and hold, while she is enacting her organizational role. These include not just formal codes of conduct and employment contracts, but also informal norms, expectations, hopes, institutional culture, and presumed standards of success-all of which can subtly and slowly infiltrate a role-occupant's approach to her activities (Schein 2010;Herzog 2018).
Because of the organization's role in shaping the conditions for meaningful caring relationships-and because of the constitutive relationship between role-occupants and organizations-there is a sense in which the organization as a whole needs to stand in a meaningful caring relationship with the care receiver, if the individual roleoccupant is to stand in such a relationship. 8 A roleoccupant is significantly constrained in their ability (qua role-occupant, that is, during working hours) to stand in a meaningful caring relationship with a care receiver unless the organization itself is fully committed to supporting such a relationship. This has implications for how careproviding organizations should be structured, if care recipients are to enjoy meaningful caring relationships with caregivers. We will illustrate these implications by taking "caring about" and "caregiving" in turn.
First, caring about. We have quoted Tronto's characterisation of caring about as "the recognition in the first place that care is necessary." Arguably, though, one can care about someone not just by recognising that care is necessary, but also by having certain emotions while caregiving, or even having certain emotions or other attitudes (for example, worry) during times when one is not caregiving. Caring attitudes are, broadly, attitudes that reflect one's desire that a person's interests be fulfilled (Collins 2015). These attitudes are valuable both instrumentally (if you care about someone, you are likely to be good at caregiving for them) and intrinsically (caring about someone is valuable for its own sake). On all care ethical accounts, the attitudes of caring about are central to a meaningful caring relationship.
How can organizations care about their charges? Michael Slote and Virginia Held have each distinguished between two types of caring about. Slote parses the distinction as that between an "intense personal caring towards people one knows," on the one hand, and "a general humanitarian caring or concern about people one only knows about (as part of a group)," on the other (Slote 1999, 2;likewise Held 1993). Slote argues these two types of caring about must be integrated in any "morally decent person," but that this integration need not be conscious or deliberate. Rather, the morally decent person will simply "go about their lives, … sometimes acting from humanitarian concerns and sometimes acting out of concern of the perceived needs of people they know" (Slote 1999, 3).
As important as general humanitarian caring might be, Slote's "intense personal caring" is the crux of the meaningful caring relationships. The caring attitude at issue in these relationships is particularised caring about a concrete other. Yet, it's doubtful that organizations can engage in such intense personal caring (Noddings 1984, 103;Björnsson and Hess 2016). Organizations have plans and policies that take care recipients as a group, but these are necessarily untailored to the circumstances of specific recipients. They fall under Slote's "general humanitarian caring." Even at their very best, organizations cannot experience the felt emotions at issue in intense personal caring. As a result, an organization's intense personal caring can occur only via role-occupants. There is a gap, here, between the generalised caring about that might be exhibited by an organization (in its plans and policies) and the particularised caring about that is manifested by the role occupant (requiring emotions and attitudes of a specific caregiver in a specific context). To help secure intense personal caring on the part of role-occupants, the organization must create an institutional culture that encourages and celebrates such caring on their part. Moreover, once such intense personal caring obtains, the organization must act by leaving scope for discretion, nuance, empathy, and time on the part of role-occupants. Thus, for an organization's generalised caring to manifest in a roleoccupant's intense personal caring, the organization's goals must both encourage and permit this kind of leeway on the part of the role-occupant.
Another aspect of "caring about" is commitment. This will be crucial in our argument to come. As Joseph Walsh (2017) has argued, successful practices of caring are partly constituted by the caregiver's commitment to the care recipient. An attitude of commitment entails an intention to continue caring about the care recipient across a range of different possible futures. Most importantly, the attitude of commitment requires not just that the carer is committed to providing care across a large number of possible futures, but also that the carer is committed to providing care across possible futures in which the carer has no instrumental (i.e., broadly self-interested) motivation for doing so. That is, meaningful caring relationships require that the carer would continue to provide care simply for the care recipient's own sake, if required. Such a commitment is part of what it means to value and recognise the care recipient as someone who has final and intrinsic moral worth, who is worthy of meaningful care even if this is not in the narrow self-interest of the caregiver. In an organizational context, such a commitment concerns the range of contexts in which the organization would continue to facilitate care. The constraints of the organization (including its other core goals-such as profit-seeking) determine the number and kinds of possible futures that are in the range of its commitment to continue providing (or supporting) meaningful caring relationships. Againjust like the intense personal caring discussed earlier-the commitment aspect of caring attitudes therefore requires a kind of open-minded permissiveness on the part of the organization.
The value of commitment demonstrates why it is helpful to view the organization itself as a participant in the caring relationship. We usually forgive a care worker who leaves their job, thereby terminating some meaningful caring relationships. But we do not so easily forgive a caregiving organization that abandons its care recipients. That is, in an institutional context, the "commitment" component of care cannot reasonably be expected to come from workers. It is reasonable to leave one's job as a caregiver. So, in an institutional context, the commitment component of care can only reasonably be expected to come from the organization itself, not from its members. This commitment component of care is a core component of caring relationships. If this core component of caring relationships can be provided only by the organization itself, then we have reason to view the organization itself as a participant in the caring relationship.
So much for caring attitudes. What about caregiving? There are numerous characterisations of this. For example, Diemut Bubeck defines care as "the meeting of needs of one person by another where face-to-face interaction between carer and cared for is a crucial element of overall activity and where the need is of such a nature that it cannot possibly be met by the person in need herself" (1995,129). Eva Feder Kittay gives two conditions for an action to count as care: first, "we are motivated to concern ourselves with the well-being of another for that other's own sake," and second, "such motivated actions contribute to the person's flourishing" (2011, 614;similarly Kittay 2014). Nel Noddings describes care as action that is in accordance with a person's attention and motivation being focused on another person's needs (2002,19).
When considering meaningful caring relationships within organizations, two things stand out about these definitions of caregiving. First, caregiving requires latitude, just as we saw earlier with caring attitudes. To see that caregiving requires permissiveness on the part of the organization, consider that Bubeck describes care as an interaction between giver and receiver; Kittay describes it as acting of concern for wellbeing; Noddings describes it as involving attention and motivation. These are not the kinds of activities that can be prescribed in detail by an organization's policies or procedures. If these activities are to arise in an organizational context, individual caregivers require the latitude and time to develop emotional connections with care receivers.
Here, it is worth noting that meaningful caring relationships contain some built-in inefficiencies. This is largely because of the commitment involved in caring attitudes and in caregiving. As explained earlier, the commitment of care implies a willingness and ability to continue providing care across a range of different possible futures, including those in which caregiving is not in the caregiver's narrow self-interest. For instance, a childcare centre needs to take into account the possibility of a baby who might become sick or go through a major milestone and desperately want to be held at all times. But if childcare workers are so timepoor that they cannot give additional attention to a baby 630 Perspectives on Politics who may require additional physical contact, then they cannot in fact provide care for her. Planning for adequate care requires planning for children who might need additional physical contact or attention even if there will be many days and weeks where no child will find herself in that situation.
This example illustrates the idea that caring resources such as time and training must be kept "in reserve" by the organization, in case those circumstances arise. Of course, if the circumstances end up not arising, then the reserved resources will not be used. This inefficiency follows from the commitment of care, which is a commitment of the organization. Thus, caring organizations must provide those on the ground with ample resources to cultivate caring attitudes and caring practices-including additional resources for when additional needs arise, and even though such resources will often go to "waste." ("Waste" here in scare quotes, because the resources are crucial to the commitment, and therefore do produce value in the meaningful relationship, even if they are not ultimately used for caregiving.) One might wonder what extent of reserves are necessary. We answer: all else being equal, it is better if more, rather than fewer, resources are kept in reserve.
To summarise: when meaningful caring relationships arise in an organizational context, it is informative to view the organization as central to such relationships, across all four phases of care. This centrality mandates a focus on how organizations can be structured as partners in meaningful caring relationships. When we look at what's involved in caring attitudes (including intense personal caring and commitment) and in caregiving, we see that caring emotions and attitudes must be encouraged on the part of role-occupants, and that time, latitude, and even inefficiencies must be granted to them by the organization in the actual process of care-giving. In the next section, we examine how this conception of organizational care manifests in a for-profit organizational model.

For-Profit versus Non-Profit
In this section, we show that the for-profit model is in tension with the organizational conditions for meaningful caring relationships between organizations (or their roleoccupants) and care recipients. The result is that nonprofit caring organizations are the most ethically defensible feasible option for organizational care.
To begin our case, consider how the for-profit model operates. As an illustration, we will focus on corporations. These are entities in which there is a separation between managers and shareholders. Shareholders provide funds that secure them a financial interest in the corporation's profit. Managers aim to operate the corporation in a way that is in the financial interests of shareholders. Some business ethicists have gone so far as to argue that managers' sole moral obligation is to maximise profit to shareholders (Friedman 1970). Corporations are thus a particularly stark case of the salience of profit in a for-profit model. That said, not all for-profit caregiving organizations are corporations, since not all for-profit entities divide shareholders from managers. Nonetheless, the basic dynamics of profit-pursuit are the same across for-profit organizations. We focus on corporations as a paradigm example.
Of course, many business ethicists dispute the idea that corporate managers' sole moral obligation is to pursue profit. This dispute can be seen in the ascendancy of "stakeholder" theory over the last few decades. Stakeholder theory insists that corporations have moral obligationsperhaps weightier than their obligations to shareholdersto attend to the interests of (say) customers, civil society, or the environment (Freeman et al. 2010). Still, even for stakeholder theorists, the pursuit of profit is central to corporations' existence. Any other obligations are limiting constraints on the pursuit of profit, or (at best) considerations to be balanced against the pursuit of profit. 9 To be sure, organizations can include other values in their charters. And individual managers might try to champion particular moral causes. But those values and managers will be constrained by the profit imperative.
There are several ways in which profit-pursuit is in tension with the aspects of caring relationships we outlined earlier-including intense personal care attitudes, commitments, and caregiving. First, the pursuit of profit mandates an institutional culture of efficiency and costeffectiveness. However, as we explained earlier, caring well is resource intensive and requires inefficiencies in time and resources, because of the commitment that it includes. Naturally, many for-profit organizations have built-in inefficiencies. (One need only consider the rates at which supermarkets throw out food.) But such inefficiencies are instrumental to the profit goal, not to the goal of creating the conditions for customers to lead good lives. Even if governments introduced regulations that demanded inefficiencies, those inefficiencies would not be robust in the specific way required by the commitment of care, when compared with the feasible alternative of non-profit care. As explained earlier, the commitment of care requires the inefficiencies to be robust across circumstances where the inefficiencies are not in the organization's narrow selfinterest. (We return to this robustness point later.) Second, the profit motive requires that the time and resources spent on any one task be minimized. This is different from the previous point about inefficiency. While inefficiency concerns resources for tasks that might never be performed, minimization is about how much resources are spent on tasks that are performed. The imperative of minimization creates a strong incentive to take intense personal caring out of the equation and to only do the things required for meeting basic needs. In other words, the pursuit of profit pushes the corporation towards being merely a service provider rather than relationship participant. Again, regulation will have difficulty solving this, because of the nature of caring relationships. Government regulators will struggle to mandate that organizations provide the opportunity for a relationship rather than a mere service. Recall that the relationship-versus-service distinction is partly marked by diffuse and open-ended goals, versus concrete and measurable ones. Regulation works excellently for the latter, but poorly for the former. The former is best produced by an organizational ethos or culture that reflects certain values-including a devaluing of cost-minimization.
To this, one might reply: surely for-profit caregiving organizations have strong incentives to provide meaningful caring relationships. After all, the elimination of intense personal care would affect the organization's profits, since the family of care recipients will prefer to place their loved one in an environment that is conducive to meaningful caring relationships. Thus, the families of care recipients will "vote with their wallets." Meaningful caring relationships will result.
We think this paints an overly optimistic picture of families' abilities to tell whether their loved one is being provided with a meaningful caring relationship as opposed to a mere service, and to act on that information if they receive it. Young children, for instance, are incapable of articulating whether they are receiving a relationship or a service. And as Rutger Claassen explains: "In many public markets, consumers are vulnerable or weak agents (think of patients, children, addicts, etc.). In these cases, they may not be able to compare offers made by providers and make effective consumer choices" (Claassen 2015, 266).
Many elders might be able to articulate this, and consequently, be able to compare offers, but the process of changing one's nursing home is drawn-out and often fraught with physical and psychological risks. The family might reasonably decide that "voting with their wallet" is too costly and risky in the case of a frail relative. What's more, even if families have the information and ability necessary to choose a relationship-providing for-profit carer, still: such care organizations will charge accordingly, which will put meaningful caring relationships out of financial reach for most families. Meaningful caring relationships, we suggest, are too important to be left to the unreliable market mechanisms of supply and demand, especially when the market has proved to be notoriously poor at adequately securing public goods to all of those who have an interest in receiving them (Schwarzenbach 1996, 124). And as we noted earlier, even elders who autonomously don't desire meaningful caring relationships might eventually change their minds. Because it is difficult to move between caring organizations, all such organizations should stand ready to provide meaningful caring relationships to those who desire them. This is not to deny that, in general, giving citizens more options in terms of service provision is superior to giving them fewer options. We agree with Claassen when he argues that the restriction of choices in access to important services has to be carefully argued for, such as with the argument that choice will lead to unjust inequalities (Claassen 2015, 270-271). We also agree that for many goods, "it is appropriate that they are provided through the market and through one or more non-market alternatives simultaneously" (Claassen 2009, 421). But as our discussion has made explicit, the for-profit option is a poor provider of the specific good we are concerned with, which is the good of meaningful caring relationships. We will argue this more fully later, in our discussion of robustness. For now, note that our argument against for-profits still allows for the plurality of providers Claassen advocates for, since meaningful caring relationships will still be provided informally by friends and family members and formally by government agencies and non-governmental organizations. In other words, we are not arguing against having a plurality of providers in the domain of care; we are arguing that one of the current choices in the menu falls short of providing the good in the manner deemed valuable (because such provision is not robust across falls in profits).
Let us now consider the fourth phase of care: care receiving. Here, there is an interaction between caregiver and care recipient, in which the care recipient responds to the care and the caregiver incorporates this response into their ongoing practices and dispositions. In an organizational context, meaningful care receiving requires ongoing and detailed communication between different parties in the organization. In particular, it requires bottom-up communication about the success or failure of care on the ground, a time and resource-intensive exercise, which can detract from the pursuit of profit in many corporations. 10 Again, regulation cannot fully solve this: workplace democracy is partly a matter of culture and ethos, which must be reflected in the self-constituting values of the organization itself (rather than imposed from the outside). Workplace democracy can co-exist with buoyant profits (Dow 2003), but democracy is likely to stay only for as long as it is profitable.
This brings us to a more general point, which underlies the earlier points about efficiency, cost minimization, and democracy. The general point concerns the robust requirements of care. Plainly, it is possible for many aspects of meaningful caring relationships to be found in for-profit caregiving organizations: under the right conditions, forprofits can be appropriately inefficient, cost-permissive, and democratic. But such organizations' caregiving propensities are insufficiently robust. Many for-profit organizations would cease to be caring with relatively minor changes-for example, changes in which individuals occupy which roles, or changes in the extent of need 632 Perspectives on Politics presented by those in their charge. 11 Even if caring relationships were "baked in" to the organization's procedures and policies, still those procedures and policies would have a propensity to change if caring relationships were no longer compatible with profit. This is the crucial point. In such organizations, caring relationships occur only at the behest of profits. Caring relationships are not robust across circumstances in which caring relationships are no longer profitable. Most importantly, caring relationships are not robust across circumstances in which the organization must respond directly to the care recipient's intrinsic worth as a vulnerable human worthy of meaningful caring relationships.
Of course, care might be somewhat robust within forprofit organizations. But we suggest care is not adequately robust. The bar for "adequacy" should be set by the feasible alternatives: if there are alternative feasible arrangements that are more robust-and in particular, that are more robust across circumstances in which care must be provided purely for the recipient's sake-then those arrangements should be implemented instead. The less robust arrangements are therefore rendered "inadequately" robust. The feasibility of non-profit caring institutions-in which care relationships are provided more robustly, in circumstances where profit doesn't mandate them-renders for-profit caring institutions inadequately robust.
There is even a case to be made that a radically nonrobust caring relationship is no caring relationship at all. To see this, recall that commitment entails an intention to continue with an attitude or practice across a range of different possible futures. More generally, a practice, relationship, or valuable constituent of life is only robust if it would be present across a range of possible ways the world might go in the future (or across a range of possible ways the world could have gone at present). Following Philip Pettit (2015), we can think about robust goods along the following lines. There is some good-such as intense personal caring, emotional warmth, and attentiveness to one's well-being-which a person might enjoy in the actual world. Call this a "thin" good. In addition, there is a conceptually separable "thick good" that one enjoys when the following is true: one would continue to enjoy the thin good, would a range of non-actual circumstances to have been the case.
For example, consider friendship. A "fair-weather friend" is often considered to be no friend at all. If a person is not facing any problems in their life, then a fair weather friend will provide that person with the thin goods of friendship-including intense personal caring, emotional warmth, and attentiveness to one's wellbeing. But if circumstances were just slightly different-if the person needed money, or needed high levels of emotional support, or moved cities-then the fair-weather friend would cease to provide the person with the "thin" goods in question. In this way, the person does not enjoy the thick good of friendship, even though she does (in the actual, fairweather circumstances) enjoy the thin goods of intense personal caring, emotional warmth, and attentiveness to one's wellbeing.
It is plausible that meaningful caring relationships are thick goods, not thin goods. Meaningful caring relationships exist only if the four phases of care would continue within that relationship, across a range of possible futures and across a range of possible but not-actual ways the present might have been. In for-profit entities, the provision of the "thin" goods of care is robust across circumstances in which profit is secure, but is fragile across circumstances where profits become precarious. The caring relationship is too contingent. Like the fair weather friend, in extreme cases of non-robustness, this amounts to no caring relationship at all, even if the "thin" goods are enjoyed in actual circumstances. Perhaps not all for-profit caregiving is so radically non-robust as to be not caring at all. But plausibly, some is. And even somewhat robust forprofit care lacks the robustness of feasible non-profit alternatives.
Importantly, governments cannot force robust meaningful caring relationships to develop within a for-profit entity (or anywhere else), even if governments can cap profits and put additional resources into the caregiving system. For one, meaningful caring relationships require mental states that are not apparent to bureaucrats standing outside the organization, rendering enforcement by the government a fraught enterprise. Moreover, once abuse and neglect are not at issue, the quality of any caring relationship cannot be reliably assessed by parties who are not themselves in the relationship. But most importantly, in such a scheme, there's no pledge in the constitutional charter of the corporation that guarantees that the additional government-provided resources will really be spent on ensuring the existence of robust meaningful caring relationships. Rather than going to profit, such funds can be spent on executive remuneration, flashy buildings, advertisement, and so on. Things will be even worse if the corporation falls below the profit cap-in that case, focus will fall back on the efficiency, minimization, and hierarchy described earlier. Even in a capped system, caring relationships are not adequately robust across circumstances in which the corporation enjoys profits that are below the cap.
It is important to emphasize that our argument is not that non-profit care providers will provide care no matter the circumstances. They are liable to the exploitation of workers (including those involved in meaningful caring relationships), self-dealing, misappropriation of funds, and so on. Sometimes, their provision of care will be contingent on such nefarious practices. Less nefariously, nonprofits might find it difficult to be robustly committed to care because they have unstable funding, or an inability to give high salaries to retain a good workforce, or cannot resist business-like management structures. 12 Thus, some might think that non-profits are also unable to provide the good of meaningful caring relationships in a robust way. Specifically, one might worry that these organizations can provide care that is only as robust as its donors: if the donations stopped flowing in, then any charitable caregiving organization would cease to provide the thin goods of care.
These are important concerns, but it is important to recognize that they arise largely because of the existence of for-profit organizations. Donations and government funding are much less likely to stop or decrease if the state knows that there are no for-profit alternatives. A prohibition on for-profits thus would make the provision of meaningful caring relationships more robust, since it would make funds for non-profits more robust, while also removing those organizations whose care-provision is worryingly contingent on profit. And of course, if forprofits were not part of the institutional mix, workers wouldn't have the option to work for a for-profit organization in virtue of its ability to offer higher salaries. 13 Still, one might worry that state-provided or philanthropic caregiving is only as robust as a certain political or moral ideology, such that caregiving would cease if a different political party were in power, or if donors adopted a different worldview. In general, the worry goes, it's not the case that non-profits or states have a blank check to provide care as robustly as possible.
In response, it is worth emphasizing that although charitable and state-provided care are not perfectly robust, the limitations on their robustness derives from external circumstances rather than from the internal values of the caring organization itself. Analogously, difficult life circumstances such as debilitating illness can make it impossible for an individual to provide another individual with the thin goods of friendship. In the case of the for-profit organization-like in the fair-weather friend-the roadblocks to robustness comes from the values that guide the organization's constitution. This poisons the caring relationship, even in those favorable circumstances where the 'thin' goods can fortuitously still be provided. By contrast, if fickle donors, lenders, or electors take a charity or state off its caring course, then the problem is not internal to the caring relationship itself. Instead, the situation is analogous to a friendship that becomes difficult because one's friend faces new and challenging life circumstances. This is not the same as-and not as problematic as-the fairweather friend.
Still, one might insist, the contingency of non-profit care is worrying-even if that contingency arises from "external circumstances" rather than "internal values." To this, we respond that our point is purely comparative. All types of caring organizations face constraints because of others' financial choices or societal attitudes. Non-profit care is not perfectly robust; far from it. But non-profit organizations are more robust in caring relationships than for-profit ones-particularly across circumstances in which the care recipient must be cared for for their own sake. This is primarily because the for-profit model centers obligations to shareholders (or other profit-recipients), where these obligations are in tension with the provision of meaningful caring relationships across circumstances in which the only reason for care provision is the wellbeing of the care recipient. By contrast, non-profit care organizations have constitutional charters that mandate a response to the well-being of the care recipient (as long as resources allow).
The role of an organization's motive in producing robustness can be seen particularly starkly by considering an extreme case: Japan's rent-a-family industry (Batuman 2018;Brownlee 2020). There are several companies in Japan that hire actors to play roles in front of a customer's friends and family. Customers rent wedding guests and shopping buddies; they also rent family members. Sometimes customers are motivated by the pressure to display a vibrant social life. Other times, customers are motivated by loneliness. The latter cases will sometimes involve meaningful caring relationships being sold for a profit. Yet clearly-and this is our point-these relationships are significantly less valuable than the real thing. By contrast, if an organization is committed to provide meaningful caring relationships-if that is its founding purpose and core mission, as it can be under a non-profit model-then the resulting relationships increase in the relevant kind of robustness, and in value. 14

Objections and Complications
In response to our discussion so far, a critic might agree with our conclusion and yet offer a different explanation for what is at stake with for-profit care providers. She might argue that the pecuniary motive on the part of forprofit organizations is problematic because it fails to show appropriate respect for care receivers or for the caring relationship itself. In other words, a critic might suggest that even if for-profit care organizations could somehow be motivated to enable robust caring relationships, it would still be deeply problematic for them to commodify human relationships and to charge money for something that human beings have a fundamental interest in receiving (On commodification, see Anderson 1990;Satz 2010).
Such a criticism could generalize to other forms of caring activities, such as health care. Although we agree that a just society guarantees health care services, and that no one should be denied heath care due to their economic situation, our argument has not shown that it is morally problematic for doctors to work for a profit. Doctors provide a service, not a relationship. And although markets are bad at securing public goods for all persons without partnership with, and oversight by, democratically elected 634 Perspectives on Politics governments, they may have a limited, albeit important, role to play in the provision of services that are needed for the protection of basic human interests. Nothing we have said demonstrates otherwise.
A critic might instead argue that a for-profit model for relationships shows a lack of respect to human beings by treating them as mere commodities. The idea here is that for-profit care organizations fail to show respect to those they provide care to. But again, the critic may continue, this line of attack will generalize and include other forprofit corporations such as private schools, which arguably do provide (some) relationships. Such a generalization appears to undermine our argument: although there might be something morally problematic with private education in an egalitarian society, it is not clear that pupils are treated as mere commodities by their private schools. 15 So, the critic presses, the pecuniary motive can have a role to play in valuable relationships.
We agree that the pecuniary motive does not always prevent valuable relationships from arising. In fact, we think that it can even play a role in a meaningful caring relationship. However, we believe that this only holds if the relationships' participants are individuals (rather than organizations) and if further conditions are met.
To begin with, consider a babysitter who accepts a salary to look after the child of a family member or friend on a full-time basis. This babysitter has a pecuniary motive: she wants to earn a living. Yet she is willing to enter and maintain a meaningful relationship with the child she provides care for independent of the salary. Given that she is motivated by both the desire to earn a salary and the desire to care for that child in a diffuse and open-ended way, and given that she has full control over her actions, she is capable of ensuring that the pecuniary motive does not jeopardize the robustness and quality of the caring relationship. Babysitters who have salient caring motives guiding their actions can work as a surrogate parent to a child and be robustly disposed to attend to her needs despite receiving a salary for that. 16 More specifically, if we go back to the discussion of what a caring relationship entails, we can see that a babysitter partly motivated by the desire to enjoy a caring relationship with a child (even while partly motivated by salary) can exercise a sufficient level of discretion, nuance, empathy, and invest sufficient time in the relationship. It is up to her whether she spends the time playing or reading to the child, or if she merely minds the child without working on the development of the relationship. There is nothing about her receipt of a salary that discourages her from robustly attending to the needs of the child, or that incentivises her to only meet the child's basic biological needs. Such a babysitter would remain motivated to have a meaningful caring relationship with the child, even in the absence of a salary.
By contrast, consider a babysitter who wants to make a living by looking after a child but lacks any other caring motive. She sees the child as a job to be performed and not as a party to a meaningful caring relationship. Imagine also that this babysitter takes her reputation very seriously and does not want to be perceived as a lousy worker. Despite having motives that support adequate care, the enjoyment of a meaningful caring relationship will not be sufficiently robust. (At least, the non-caring babysitter is "insufficient" if the caring babysitter is available: as we mentioned earlier, the bar for "sufficiency" is set by the feasible alternatives). The pecuniary or reputational motive, when not accompanied by caring motives (i.e., the pull towards caring for the child for the child's own sake), is also likely to lead to a situation where the babysitter focuses on the provision of a service rather than the development of the relationship. This may be fine if a babysitter looks after a child for a couple of hours a week. Our argument is silent on the value of the profit motive in cases where the care-recipient is not dependent on the carer for a meaningful caring relationship. Instead, the babysitter's purely self-interested motives become a problem when the child is under the sitter's care on a full-time basis or for significant periods of time. In this case, the child is missing out on an important ingredient of a good life: a robust meaningful caring relationship with the person she spends most of her time with. The fact that the child spends most of her time with the sitter means that the child comes to depend upon the sitter for a meaningful caring relationship.
The fact that babysitters who are motivated by care are psychologically capable of developing robust meaningful relationships with their charges shows that the pecuniary motive is not itself the problem. A pecuniary motivate can co-exist with a meaningful caring relationship, but that is only possible in the case of individuals, because only individuals are constitutionally capable of bringing caring, pecuniary, and reputational motives into an equilibrium. In the institutional case, however, there is serious concern with a pecuniary motive on the part of the organization, since that motive will play too large a role in shaping the way the organization approaches the role they play in the lives of those under their care. Organizations lack the capacities for empathy, physical touch, and so on. These capacities keep the pecuniary motive in check in individuals. Because of organizations' unavoidable lack of human fellow-feeling, the pecuniary motive threatens to run amok in their procedures and policies. Better if it is absent entirely.
Thus, our argument is not just a repeat of the familiar claim that for-profits are "psychopaths" that care only about their profits. 17 Our claim is that the profit motive has special problems in particular contexts. These are contexts where (1) the recipient needs the organization for (2) the opportunity for a meaningful caring relationship. If the organization provides such an opportunity, but the client does not need it (e.g., because the organization plays a somewhat peripheral role in the client's life), then being profit-driven may not be such a problem. Likewise, if the client needs the organization, but only for the provision of a service, then being profit-driven may not be so bad. In this way, our argument does not aim at a full-blown indictment of the profit motive. The argument is more circumscribed and less simplistic than that. (But nor do we positively claim that the profit motive is morally good in non-dependent or non-relationship contexts: we are simply silent on such cases, leaving examination of them for another day.)

Conclusion
Our discussion vindicates the widespread intuition that there is something morally problematic with for-profit organizations providing care to young children and elders. We argued that the problem lies in the mission of a forprofit organization and the nature of the relationship between that organization, its profit-recipients, its roleoccupants, and care recipients. Because for-profits are in the business of making money out of care services, they are not set up in a way that enables the development of sufficiently robust meaningful relationships between role-occupants on the ground and those they provide care for. The upshot is that we ought to move away from a forprofit model, and support organizations that are better placed to create the conditions for young children and elders to enjoy meaning in their lives.

Notes
1 The most comprehensive of such reports is the Thematic Review of Early Childhood Education and Care Policy by OECD countries (http://www.oecd.org/ education/school/earlychildhoodeducationandcare. htm). 2 A different argument might run: given that for-profits will inevitably be poorly regulated and underfunded, the only way to address the problem is to move away from a for-profit model of care. We take no stand on the plausibility of this argument. We advance a different critique, outlined in the next paragraph. 3 Non-profits here include government-and charityrun centers. Note that we are neutral on which, if either, of these provides higher quality care, but we recognize that there could be an additional case in favor of government-run services due to the costs associated with the privatization of public services. Chiara Cordelli, for instance, argues that when NGOs provide essential services on behalf of the state, they might invertedly undermine social justice due to their inability to factor in all the relevant consequences of their actions. They might also violate fair equality of opportunity by distributing opportunities on the basis of affiliation or conformity to the organization's worldview. Finally, privatization threatens value pluralism by preventing those organizations from distributing goods according to their distinctive ethos. See Cordelli 2012;2013. See also Dorfman and Harel 2013. 4 The key distinction between for-profit and non-profit models is that in the former, excess funds are typically transferred to shareholders or other specific rightsbearers; while in the latter, any excess funds are funneled back into pursuing the organization's activities and goals. We take the most fundamental difference to be whether the organization is run for the financial benefit of a particular group of people. Legally, this plays out differently in different jurisdictions, particularly regarding taxation; these differences do not affect our argument. 5 Our discussion also applies to citizens who live with certain kinds of disability. However, we think that because disability is not a life stage, and therefore, not typically temporary, it raises different moral issues, and must be discussed separately. 6 But given the fact of reasonable disagreement, does the mere creation of an opportunity to enjoy meaningful caring relationships count as a problematic form of perfectionism on the part of the liberal state? We don't believe so. After all, meaningful caring relationships may be understood as a primary good in the Rawlsian sense, and so as part of a thin theory of the good. See Claassen (2011a, 7) on why this does not amount to perfectionism. Or they can be understood as a basic capability. See Claassen 2011b for the claim that care counts as a basic capability. But even if the creation of an opportunity to enjoy meaningful caring relationships requires a commitment to perfectionism, this may not be such a problematic result given that these age groups are severely constrained in the amount of autonomy they can exercise in their lives. For the argument that perfectionism for children is compatible with liberal justice, see Fowler 2014. And of course, Fowler's argument can be extended to elders who have suffered significant cognitive decline. As he explains in the case of children: "[children] have lesser physical and mental capacities than adults, implying that they are especially vulnerable to threats. Furthermore, the way in which we should help children differs 636 Perspectives on Politics from adults because children are especially vulnerable to their own poor choices." He adds that "we have special obligations to children that are different in kind as well as degree compared with our duties to adults" (2014,315). We thank an anonymous reviewer for raising the important issue of perfectionism. 7 The upshot here is that workers who enjoy meaningful caring relationships with those they care for make the lives of young children and elders significantly more meaningful, even if young children and elders enjoy loving relationships with people who cannot or will not meet their basic biological needs for some or most of the time. 8 For those who are uncomfortable with the idea that the organization itself stands in the caring relationship, another way of expressing our point is that the organization is disposed to fully support caregivers on the ground to offer meaningful caring relationships to care receivers. Talking of an organization's "standing in the relationship" here will count as figurative or metaphorical. Such a view still supports our overall conclusion against for-profit care, pending our argument below that for-profits cannot "fully support" meaningful caring relationships in the right way. That said, we will later provide further reasons for viewing the organization itself as standing in the caring relationship -specifically, we will argue that care is partly constituted by a particular commitment that can reasonably be held only by the organization itself. 9 The obligation to pursue profit is often viewed as fundamental because it is plausibly seen as deriving from a promise made by managers to shareholders when shareholders buy shares. 10 In a similar vein, Tronto (2013) emphasizes the importance of workplace democracy in a care ethical context; our discussion has strong affinities with Tronto but is a broader point about profit per se. Our argument also dovetails with Roberto Frega's (2019) recent interpretation of democracy as a "paradigm normative concept" that demands application across the whole of social life. Our claim here, though, is the specific one that bottom-up decision-making enables the discretion that is conceptually necessary for successful care. 11 To illustrate, think of the precariousness faced by students when high quality education is solely dependent on the hard work of one head-teacher, or when students have gone through a traumatic event and need additional mental health support to keep up with their studies. 12 We thank an anonymous reviewer for this point. 13 Does this problematically restrict the freedom of care workers to work for a higher salary? We don't think so: it restricts that freedom for the sake of a worse-off group, namely, care recipients who require the opportunity for meaningful caring relationships. Furthermore, if the government and public faced the imperative to adequately and robustly fund such organizations (because of the lack of a for-profit alternative), it's likely working conditions within them would improve. 14 We don't claim that non-profit organizations could provide real familial relationships. In our view, caring relationships and familial relationships are different: the former can be provided by non-profit organizations; the latter, most likely, cannot. 15 Private education also does not seem to express a lack of respect to the teacher-pupil relationship, relationships which can be meaningful to those who partake in them. So it cannot be that no kinds of meaningful relationship (such as meaningful educational relationships) can be provided by a for-profit organization. 16 Obviously, the babysitter is not primarily motivated by parental love, which means that the relationship is much less robust than the relationship the child enjoys with her parent (Ferracioli 2014). Still, the caring relationship between babysitter and child is robust in a way that the relationship between for-profit care organization and child is not. 17 That familiar claim is made by Bakan 2004 and Bakan 2020; we put "psychopaths" in scare quotes because it is arguable whether this representation is true to psychopathic humans.