Universal Health Care and Enforced Beneficence

I examine Allen Buchanan’s arguments for enforced beneficence and express a number of worries concerning his attempt to justify coercive distributive policies that guarantee (basic) health care services for all citizens. The central objection questions whether, given Buchanan’s own stipulation of universally-instantiated attitudes of moral beneficence amongst all society members, his arguments from, first, the coordination problem and, second, the assurance problem successfully establish a justification of enforced contribution. I defend alternative, non-coercive, responses to the aforementioned problems and show that a particular kind of institution (an “information service”) provides all citizens with the sufficient and reliable epistemic resources so that they can effectively help the sick and needy. I notice that Buchanan’s difficulties with justifying coercion can be regarded as providing indirect support for the view that developing a justice-based conception of moral health care rights remains, pace Buchanan, an important task to be completed.

I examine Buchanan's arguments for enforced beneficence and express a number of worries concerning his attempt to justify coercive policies. The central objection questions whether, given his own stipulation of universallyinstantiated attitudes of moral beneficence amongst all society members, Buchanan's arguments from, first, the coordination problem and from, second, the assurance problem successfully establish a justification of coercive contribution. I defend alternative, non-coercive, responses to the aforementioned problems and show that a collective agent/institution might well provide all citizens with sufficient information concerning the question of how to effectively help the sick and needy. My criticism does not deny that Buchanan has identified a set of important collective action problems regarding the universal provision of health care. However, given Buchanan's own premises, his argument falls short of vindicating his ambitious conclusion concerning enforcement as the remedy for these problems. Coercive and non-voluntary transfers of resources are simply not necessary in a society in which moral attitudes of beneficence are shared by all and when this fact is acknowledged by all members of the stipulated society. I notice that Buchanan's failure to justify coercion can be regarded as providing some indirect support in favor of the claim that developing a workable conception of moral health care rights remains, pace Buchanan, an important task to be completed (but not taken up in this paper).

Universal Basic Health Care Without Moral Rights
The diversity of ethical issues involved in debates about health care policies has grown in both academic and non-academic discourse. Technological innovations in medical science, and the related rising costs of medical practice, have led to an intensified scrutiny of the notion of health care rights and of the question of what "equity" with regard to access to health care services amounts to in scientifically-advanced capitalist societies. The problem that is central to this paper concerns the efficacy of collective (and to-be-coordinated) acts of providing health care for all citizens, and, most fundamentally, how enforced contributing to such policies can be justified on grounds of this desired efficacy even if positive moral rights to health care are called into question.
The idea that society (and its citizens) have an enforceable obligation to contribute to public health care endeavors, even if the sick and needy do not have a corresponding moral right to health care, lies at the heart of Buchanan's work; work that has been influential partly because it rightly reminds us not to take moral rights as an unquestionable given. Buchanan does not only call into question liberal and social democratic conceptions of universal welfare rights; Buchanan also criticizes conservative and libertarian strategies of categorically rejecting any legally-guaranteed welfare provisions in the name of supposedly absolute property rights and liberties. The latter dimension of his writings is the central topic of the following reflections.
In his "The Right to a Decent Minimum of Health Care" Buchanan (1984, pp. 59-66) criticizes three prominent philosophical (liberal) proposals and shows that they cannot establish a positive right to health care (utilitarianism, Rawls' "Justice as Fairness," and Norman Daniels' (1985) attempt to derive a right to health care from a Rawlsian principle of fair equality of opportunity). Its unsuccessfulness notwithstanding, Buchanan claims that we should nevertheless be careful when we consider the implications of this three-fold failure. Just because all these attempts to justify a positive right to health care fail, this failure does not lead, by default, to the libertarian triumph consisting in the successful refutation of any coercive arrangements designed to secure health care. In his discussion of rights' enforceability Buchanan (1984, pp. 56-7) notes: Indeed, the surprising absence of attempts to justify a coercively backed decent minimum policy by arguments that do not aim at establishing a universal right suggests the following hypothesis: advocates of a coercively backed decent minimum have operated on the assumption that such a policy must be based on a universal right to a decent minimum. The chief aim of this article is to show that this assumption is false. 2 2 One issue that I can only address in passing is the important question of why Buchanan's challenges of coordination and mutual assurance (extensively discussed below) do not support a more generous universal healthcare system, but merely the mentioned "decent minimum of health care." (I am indebted to a referee for this journal for raising this question.) Recently, Buchanan has directly addressed this issue by stating that "obligations of beneficence are traditionally understood to be limited by the proviso that rendering Even if all rights-based theories fail, Buchanan continues, there is an alternative (pluralistic) justification of coerced contribution to health care provision available. Buchanan begins his positive case by noticing that politicians and philosophers who are attracted to rights language in health care debates attribute significant importance to the object of that right, viz. a secured minimum of medical care for all. Why should they insist on this project being realized if and only if we are able to tag the label of "rights" on the respective policy? Buchanan (1984, p. 66;my emphasis) suggests that they should not: "My suggestion is that the combined weight of arguments [none of which is based on antecedent moral rights] is sufficient to do the work of an alleged universal right to a decent minimum of health care." Buchanan is quick in assuming that his beneficence-based account will be able "to do all the work" that a justice-and rights-based account can do, without running into the aforementioned problems that come with moral rights. Even if we restrict ourselves to a consequentialist perspective, that focuses on the outcomes of the two competing justificatory approaches, Buchanan's claim seems overly optimistic for a variety of reasons (on top of those that I spell out in the next sections). Especially with regard to the issue of the subjective experience of social and economic security, a universal and publicly guaranteed positive entitlement in terms of fundamental moral rights appears to contribute in ways that cannot be fully accounted for when society leaves everything to charitable impulses. As we will clarify in a moment, Buchanan (1984, p. 57) defends his charity-based approach by claiming that "[t]o the morally virtuous person the imperatives of charity may be as urgent as those of justice." First of all, this claim may be unconvincing to those who do not already believe that Buchanan's envisioned moral community can be realized and consequently think that meeting the basic health care needs of all must be backed by an irreducible appeal to rights and social justice. (Kantians, like Ripstein (2009) for example, have good arguments for such claims.) In addition, even if aid to the needy is not to be unduly burdensome to the benefactor. Consequently, the enforced beneficence approach avoids objections to which more demanding egalitarian concepts of the right to health care are vulnerable" (Buchanan 2009, p. 74). I do not find this quick clarification fully satisfactory but a detailed analysis of this aspect of Buchanan's view has to be postponed for another occasion. all citizens were morally virtuous persons (but, in principle, retain the freedom to change this attitude), and gave enough to meet the health care needs of all, it may nevertheless be a constitutive feature of a guaranteed basic minimum that the beneficiaries know that their health care needs are taken care of as a matter of justice-based entitlements and rights. Of course, this first set of critical remarks is not a conclusive argument in support of such a moral right's existence; Buchanan will probably (and rightly) highlight this point. However, it calls into question Buchanan's optimistic initial assumption that beneficence can do all the work that rights-based approaches do. I put aside these worries. Having said that, especially the second worry from the good of rights-based public guarantees leads us to the core of Buchanan's ambitious defense of enforced beneficence. To this central innovation we now turn.

Buchanan's Pluralistic Justification of Coerced Contributing
Buchanan's strategy for justifying an enforceable principle, guaranteeing a decent minimum of health care for everyone, is pluralistic. It establishes citizens' access to basic medical services in the framework of four independent considerations. The combined weight of these considerations is supposed to provide an argument justifying a centralized agent using coercion in order to meet the moral obligation of beneficence (as distinct from an obligation of justice) to help those in need of medical assistance and services. A crucial element of Buchanan's strategy is to present the obligation of beneficence in question as a collective one. Buchanan (1984, p. 70) justifies this shift from individual to collective beneficence by means of pointing to the significant financial and organizational efforts that are necessary to realize the goal of providing minimal medical services for all members of society. 3 A decent basic minimum of health care is such a collective good. 4 These goods come with a particular set of problems (other examples of such collective goods are national defense, environmental 3 Cf. Buchanan (1991, pp. 173-177). 4 I am indebted to a referee for highlighting that the good in question is not properly referred to as a traditional "public good." Hence, I replaced the language of public goods in many contexts. The good in question is health care that private and distinct individuals (the patients) enjoy. However, the crucial point is that we are considering the option of realizing and guaranteeing these goods in a public manner. "Publicly-realized-personal-goods" is probably the best label for what is at stake. protection, etc.) that must be confronted by societies in order to successfully generate these goods.
Before dedicating the remainder of this paper to these problems let me briefly present the other three considerations that Buchanan's (1984, pp. 66-68) pluralistic strategy employs. The first argument focuses on "special rights" and is concerned with the rectification of past and present injustices (e.g. health problems related to discriminatory policies), the requirements of compensation (e.g. health problems related to a third party's negligence), and entitlements to health care based on extraordinary sacrifices that citizens provide for their society (e.g. impairments due to compulsory military service).
The second consideration argues that a lack of certain kinds of collectively provided (and enforced) basic health care can be regarded as the violation of a specific negative right. This argument provides a justification for health care measures such as public sanitation and immunization programs and can be summarized under the heading of "harm prevention". It is, for example, in everyone's interest that certain infectious diseases are controlled by reliable public agents and services; collectively implementing (and financing) immunization programs (also, and especially 5 , for those who could not otherwise afford the vaccine) is in everyone's interest. 6 Thirdly, there is a number of prudential arguments in support of a guaranteed decent minimum to health care.
These arguments emphasize a healthy population's collective benefits such as a more productive labor force and fitter soldiers.
Buchanan is confident that these three arguments present strong support for the claim that every citizen should have access to a decent minimum of health care. He also reminds us that the three arguments do without any appeal to universal (as opposed to special) positive health care rights. In addition, and this 5 I say "especially" because historically it was the worst off members of societies that had been exposed to diseases that are now controlled by comprehensive preventive measures. Especially in today's circumstances (globalization and urbanization) the better off can hardly argue at this point that they are not required to contribute to vaccination programs because they are free to avoid close contact with the poor. 6 It is worth noting that this second argument by Buchanan challenges radical libertarianism in two ways: not only does it ask citizens to contribute, at least some amount of resources, to the collective sanitation and/or immunization program. In addition, Buchanan's argument seems to imply that a collective agent is justified in forcing citizens to undergo this immunization, if this is necessary to prevent harm not merely for the person in question but for those around her. I cannot pursue these issues here. Cf. Francis (2005). is the major difference to Daniels' and Rawlsian approaches, there is no need for a comprehensive theory of justice in order to support this entitlement to basic medical care.
The major component of Buchanan's pluralistic strategy is still underdeveloped at this stage of the argument though. After all, Buchanan claims that a health care regime based on beneficence and charity (as opposed to justice and positive moral rights) can do all the work that a coercive (welfare) state could do, viz. to guarantee a decent minimum of health care for each individual member of society. So far Buchanan has established the justification for such a legal entitlement either for some subgroups of the citizenry only (e.g. former military staff), or for the populace at large, but merely with regard to an extremely minimal subset of essential medical services (e.g. vaccinations against some contractible diseases). I now turn to Buchanan's argument for enforced beneficence that is supposed to solve this problem of providing the resources necessary for such a guarantee without thereby being committed to a universal moral right to health care. This is the fourth and final argument of Buchanan's pluralistic strategy. Buchanan spends by far the most amount of time and effort on defending the argument from enforced beneficence. It is crucial for the success of his pluralistic strategy.

Buchanan's Argument for Enforced Beneficence
Buchanan claims that reasonable secular and religious moral outlooks accept the existence of a moral duty of beneficence to help fellow humans in dire need. According to Buchanan, even (most) libertarians are committed to this duty. The latter can do so because this duty does not seem to imply any positive moral rights on part of the beneficiaries. With regard to the issue of providing a decent minimum of health care, however, discharging this obligation to help those in dire need consists, primarily, in contributing to the collective endeavor of providing basic medical services for those who cannot otherwise afford them.
As mentioned above, this assumption is crucial. The most effective way to discharge the obligation in question, according to Buchanan, is assumed to be a collective regime, as opposed to individual, small-scale initiatives.
With regard to this collective effort, Buchanan asks us to envision two scenarios. First, a situation in which all agents are actually morally motivated to discharge their duty of beneficence to help those in need effectively. Secondly, we imagine a situation in which an individual benefactor cannot be sure that all others are equally motivated by this duty of beneficence as she is. Buchanan's conclusion is that in both scenarios, beneficent and rational individuals have a decisive incentive not to contribute to the collective effort of providing a decent minimum to health care. However, acting on this rational incentive results in the most efficient policy not getting realized. This is so despite the fact that the agent is perfectly beneficent and consequently wants to help as effectively as possible.

The Coordination Problem
The first scenario envisions a society of individuals, all of whom are motivated to act in accordance with their duty of beneficence to help those who cannot purchase a decent level of health care. It is exactly because of this universally-present genuinely beneficent motive that the agents in question want to discharge this duty effectively, i.e., they want their individual contributions to improve the situation of the poor to the greatest possible extent. Buchanan assumes that this motivation expresses itself in the fact that each beneficent agent maintains a willingness to provide (a portion of) the means necessary for the successful implementation of a decent minimum of health care policy. 7 This is so because the agent is aware of the significant financial resources that are needed in order to achieve this aim. Buchanan presumes, for the sake of argument, that the agent in question is, in principle, willing to direct her individual contribution to this collective project because it promises to be the most effective way to discharge her endorsed duty of beneficence to help those in dire need.
There is one inescapable problem though, according to Buchanan. Since the benefactor is obligated (and willing) to help effectively, and since her individual contribution is going to be marginal (in comparison to the large number of individual contributions needed to realize the collective aim) she will conclude (for reasons detailed in the next paragraph) that the most rational thing 7 If we assume that our duty of beneficence must be discharged in an impartial manner and in accordance with some minimally egalitarian intuitions then a successful decent minimum policy gets all beneficiaries above the threshold of access to basic medical care. Once this goal is realized then the specific duty of beneficence, that Buchanan is concerned with, "disappears" so to speak.
to do is not to contribute to the collective endeavor. To the individual contributor, this conclusion appears to be the most rational, exactly because her contributing will very probably result in a contribution that less than maximally helps those in need. The rational and beneficent agent will, therefore, rather direct her individual contribution to small scale (but, overall, less effective) projects that aim to alleviate the health problems of those who cannot do so themselves, for example, on the level of local health care initiatives. Consequently, and assuming that all other beneficent agents are deliberating in a similar fashion, the most effective way to discharge the obligation in question will not be realized. Why does this paradox result? After all, are we not assuming that all agents are motivated by proper moral motives and genuinely want to help the poor as effectively as possible (and all society members know this about one another)? According to Buchanan, this first problem amounts to a variety of the "coordination problem".
The dilemma that each potential contributor faces is that either her contribution unnecessarily adds to the good of universal health care because enough others have already contributed or she gives her resources when not enough others contribute. 8 Buchanan (1984, p. 70) concludes: "In either case, my contribution will be wasted. In other words, granted the small scale of the investment required and the virtually negligible size of my own contribution, I can disregard the minute possibility that my contribution might make the difference between success and failure." In both cases the beneficent agent's contribution is wasted and would have been of more effective (and more beneficial) use if it had been spent on individual and small scale projects, despite the fact that these latter projects turn out less effective overall in comparison to the collectively-provided decent minimum policy. Again, since the most rational thing to do appears to be not to contribute to the collective policy it will not be realized despite its acknowledged superior efficacy (the very fact that would make beneficent agents contribute to it in the first place). The next step is the crucial one: Buchanan concludes that there is only one way to resolve this problem and to ensure that all citizens contribute to the decent minimum regime in a well-coordinated manner. Buchanan (1984, p. 70; my emphasis) says: "But if everyone, or even many people, reason in this way, then what we each recognize as the most effective form of beneficence will not come about. Enforcement of a principle requiring contributions to ensuring a decent minimum is needed." It is at this point that one wonders if the justification of enforced beneficence is in fact successfully established by Buchanan's previous argument.
Keep in mind that we are working under the assumption that all individuals are motivated by the stringency and force of an accepted moral obligation. Conflicts between individual self-interest on the one hand and duties of beneficence on the other are not the problem afflicting the scenario discussed. Given this reliable and society-wide presence of beneficent motives it appears ad hoc to claim that coercive mechanisms are needed and justified in order to overcome the coordination problem. My main objection to Buchanan's first argument (and as it will turn out, also to his second) for enforced beneficence is that it does not establish a clear link between the coordination problem and any sufficient justification of collectively-imposed coercion. Given the current discussion of the coordination problem, it is difficult to see why other (non-coercive) mechanisms are incapable of overcoming this particular problem. The problem in question is "information based", as opposed to being a challenge that requires coercion for its solution. Let me clarify and illustrate this objection to Buchanan's first attempt to vindicate the enforcement of contributions to collective health care endeavors.
I will then reply to objections and potential defenses of Buchanan.
Consider this alternative "mechanism." The state, the government, or some private institution may provide a service that solves the coordination problem without using coercion by determining the amount of each individual contribution to the collective project of guaranteeing the decent minimum of health care (again, a good that all beneficent individuals are presumed to want to realize together). In returning to Buchanan's discussion of the rational incentive that each of these agents has for not contributing, we can make my proposal clearer. The beneficent individual that we are asked to imagine is in a state of epistemic, rather than a motivational and normative "uncertainty." The task of my suggested institution, let us call it the "information service," would be to remove this epistemic uncertainty and to determine each individual's contribution that is necessary to achieve the publicly-guaranteed good of universal access to health care. Notice that the ultimate step of actually transferring the contribution, as determined by the information service, is then nothing anybody needs to be coerced to in a scenario of universally-maintained attitudes of beneficence. After all, thanks to the imagined highly-reliable information service, all individuals know that the contribution they give won't be wasted. They voluntarily give their contribution that is needed to realize the most effective policy for helping the sick, i.e., the collective decent minimum policy, as envisioned by Buchanan.
Two objections to my proposal emerge immediately. Firstly, it appears unrealistic that the envisioned institution can determine the very exact amount of individual contributions in a way reliable enough to overcome the coordination problem. After all, in order to determine the size of the individual contributions, one has to know the exact overall budget necessary to realize the decent minimum policy. Considering the unpredictability of advancements in medicine and the similarly unpredictable rise or decline of the number of worse off citizens who must be covered by the decent minimum policy (and the poor's diverse medical conditions), it seems rather unlikely that the information service can confidently guarantee each individual agent that her contributions are precisely as large as this is needed in order to avoid that her contributions are wasted.
One first thing to notice in response is that this same practical problem seems to apply to Buchanan's policy of enforced beneficence. A real-world Benefactors who are acting from a sincere duty of beneficence and have access to the accurate (and trustworthy!) information that is used by the centralized agency to determine the amount of needed contributions, do not have to be coerced to contribute to begin with.
Secondly, it is an empirical question how my information based approach can avoid these problems of recommending slightly too little or slightly too much voluntary contributions. One expects this empirical issue to be settled by continuous political and practical processes. 9 In addition, one might propose that the beneficent and non-coercive contributions, determined by the information service, are deliberately set slightly above the expected budget that is needed to ensure the decent minimum for all. This seems to be justified because of the specific features of the collective good in question, i.e., the unpredictable nature of health-care-related public policy goals (mentioned above). The resulting surplus will then not count as "wasted." It can be used in the following year(s) and/or for other basic minimum projects that have an indirect and long term impact on the society's overall health situation (such as dietary initiatives in public schools, etc.).
Moreover, one can argue that it is part of the idea of a guaranteed decent minimum, briefly mentioned in section one, that there are at least some additional reserves available in case the information service was too conservative in its projectionssomething that will always remain a possibility, regardless of how well the information institution is set up. According to this rejoinder, contributing to an already sufficiently-financed decent minimum program turns out not to be wasteful after all; on the contrary, it contributes to this minimum being guaranteed for all potential patients in the face of the inherently unpredictable variables characterizing the complex project in question. 9 Another issue is that the "right" amount of individual contributions is dependent on a particular society's conception of the decent health care minimum. Determining what services count as basic enough and what medical resources are to be dedicated to the implementation of these services (e.g., decent technology vs. the best technology available regarding cancer treatment) is not a valueneutral and apolitical project. I assume for the purposes of the discussion in the text that societies have settled that issue, that is, the decent minimum policy is reasonably codified and limited and its administrators have a precise idea of what this policy is going to cost. Many have stressed the point that continuous democratic deliberation is called for in order to settle these complex questions in real-world circumstances in a legitimate way. Cf.: Gutmann (1982, pp. 556-8) and Bole (1991, pp. 10-7).
A different way to formulate an objection to the suggested non-coercive solution to the coordination problem highlights that Buchanan repeatedly emphasizes the, seemingly unavoidable, negligible size of the individual contributions. 10 Why should the information service make a difference regarding this particular dimension, given that any morally-motivated agent, faced with the negligible impact that her contribution potentially has, will again judge that her contribution does more good if transferred to small scale and local initiatives and policies? We seem to run into the initial problem, the presence of the powerful information service notwithstanding.
As mentioned above, it must be acknowledged that this version of the objection also calls into question the applicability of the information-servicesolution to real world circumstances, with all their political and empirical complications. The service would have to provide an enormously precise, detailed, and (morally problematic) intimate set of information, determining exactly what the individual contributions would have to be in order to realize the publicly-funded health care infrastructure. The information has to be that detailed, exactly in order to make sure that the individual contributions in question are never negligible. If (and yes, it remains a big "if") such a service is delivered in this reliable and unambiguous manner, no contributor would ever be justified in judging her potential contribution a negligible one. The information service would make sure that the contribution is exactly as it ought to be, relative to the goal of efficaciously-guaranteeing the target endowment needed for the most effective strategy of helping those in need. 11 Again, all this calls into question the practical feasibility of my proposal. At the same time, this vagueness is permissible at this point in the development of the argument, if one keeps in mind 10 I am indebted to a referee for this journal for framing the objection in question in terms of the negligible size of the individual benefits. 11 There's an interesting further complication arising from this response (that I put aside in the paper). Does my proposal, counterintuitively, lead to the discrediting of any additional (voluntary) contributions into the public system, because these unforeseen contributions would then invalidate the information that the information service attempts to generate? I postpone this further complication until a later opportunity. However, the next paragraph in the text hints at the likely response, defending the moral praiseworthiness of contributors who, in a supererogatory spirit, opt to transfer even more towards the publicly-financed health care regime than would be required to precisely guarantee its realization. At first, all this had looked like a severe blow to the information service proposal (and this problem has readily been acknowledged). However, the unpredictable and "unplannable" nature of individual human health works even more decisively in the other direction, rendering problematic a contributor's personal judgment not to contribute to the public project due to some vague worry regarding the potential negligibility of her resource transfer. Realistically, the (real world) information service will provide a certain, reasonably-broad, spectrum of projected overall costs, taking the unpredictability of individual health into consideration (for, e.g., a certain number of fiscal years, based on assumptions regarding population development, life expectancy, etc.). Hitting one of the many reasonable targets within that range, will then be considered a satisfactory outcome. Given the reliable availability of the information (service) regarding the spectrum and range of efficacious outcomes, individual and private judgments in favor of non-contribution, based on the negligibility of one's contribution, are even harder to justify. 12

The Assurance Problem
Buchanan's second scenario appears more promising with regard to justifying enforced beneficence. We are still deliberating whether to contribute or not from the standpoint of a beneficent agent, i.e., an agent who accepts her duty of charity to help the poor with their health care needs and, a fortiori, wants to do so as effectively as possible. In contrast to the first scenario, however, the beneficent agent does now seem to have a new incentive not to contribute, namely that others might be prone to free-ride. She does not know whether or not enough others will actually contribute. This time the reason why these others might fail to contribute is not the inherent paradox that comes with the universal presence of the beneficent motive to help the poor as effectively as possible but, supposedly, it is the possibility of this very motive getting overpowered in others' practical deliberation by self-interest.
Unfortunately, Buchanan does not consistently motivate this shift towards morally-deficient agents in the set up of his thought experiments. I therefore proceed in two steps in order to execute my discussion of the assurance variety of the enforcement argument: First, I grant for a brief moment that we are now considering an imagined society, in which at least some agents might deflect from contributing to the collective health care policy because they end up being overpowered by self-interested, non-beneficent, motives. Second, however, I highlight that this very scenario is not consistent with most of the passages in Buchanan's writings, i.e., assumptions that remain committed to the idea that we are considering a society of universally shared attitudes and motives of beneficence.
It turns out that regardless of which of the two readings we endorse, the remainder of the argument from the assurance problem is relevantly similar to Buchanan's argument from lack of coordination discussed above and, therefore, open to an epistemic (and non-coercive) resolution. Buchanan begins the relevant argument with the reasonable claim that without the assurance that enough others actually contribute, the most rational thing to do, again, appears to be to direct one's individual contributions to (suboptimal) projects of dispersed and small-scale health care endeavors. 13 According to Buchanan, the assurance worry suggests that in order to achieve the optimal and most effective outcome, an agency must be established that enforces contributions in order to disarm the beneficent agents' (rationally-warranted) incentive not to contribute. Once the beneficent agents rest assured that their morally-deficient, i.e., narrowly selfinterested, co-citizens are forced to contribute, the former will regard their contribution as not being wasted and will contribute their share.
Granting for a moment this way of setting-up the assurance problem in the context of Buchanan's wider argument, one first observation concerning his reflections is that they seems to assume that each individual's contributions are strictly fixed. In particular, Buchanan appears to presume that beneficent individuals are not willing to contribute even slightly more than their "fair and equal share" of the overall sum that would have been sufficient to realize the decent minimum regime if all others had done the same right thing. If we imagine a small scale society of ten, equally well off, members and stipulate that a universal decent minimum policy costs one hundred dollars, then each individual's fair and equal share amounts to ten dollars. If one out of the ten is overpowered by self-interested motives and prefers to keep these respective ten dollars, then the other nine would waste nine times ten dollarsas long as they remain unwilling to contribute more than the fixed amount of ten dollars. Does this fact by itself establish Buchanan's conclusion, according to which the society in question is justified in forcing the one to be "beneficent" and to contribute her fair and equal share? Not necessarily, it seems to me. One should stress at this point the potentially problematic aspect of the assumption mentioned above, viz., that the potential benefactors are depicted as inflexible (and unwilling!) when it comes to giving even slightly more than their fair share. In the case of our model society, the fact that the one imperfectly-moral agent fails to give results in asking the beneficent nine for individual contributions of a bit more than eleven dollars (assuming that the medical services comprised by the decent minimum remain available to all, including the one deflecting member). If, and Buchanan is committed to this assumption, the nine others are ready to act from genuinely beneficial motives (as opposed to, for example, justice-based or contractarian motives of reciprocity) then they will not, at least as long as free riding remains rare (and it has to, according to Buchanan's own presentation of the assurance problemeven according to the first of my two readings), pull out of the collective endeavor, because a small minority of morally deficient individuals refuses to contribute its fair shares. And the above-introduced information service will again stand ready to alleviate that deeper epistemic worry, pertaining to determining the precise level of contributions and the number of active contributors that are now necessary, in the face of non-ideal levels of beneficent compliance. Crucially, coercion remains unwarranted, even if we currently go along with Buchanan and accept that the envisioned hypothetical society consists of both, benevolent and self-interested agents.
A qualification such as "as long as a certain point of widespread refusal is not reached" is, of course, a critical feature of my proposal. I admit that. If a large number of individuals is constantly overpowered by self-interested motives and stops being motivated to meet their obligations of beneficence, then the decent minimum policy breaks down. But notice that this would be a rather unsurprising outcome and not original news at all (and it is partly for this reason that I believe that this first interpretation of Buchanan's assurance problem cannot be what he had in mind. In line with Buchanan's other assumptions, mentioned above, such a society would be deeply morally deficient, a point granted by all major philosophical and religious doctrines and playing a major role in Buchanan's argument in the first place (further discussed below). Under stable moral conditions of widespread beneficent attitudes, agents are not going to be obsessed with ruling out every possible occurrence of the free rider problem as a necessary precondition for them to contribute to the project of realizing the collectivelypursued good of decent health care for all.
Be that as it may, my main response to Buchanan's treatment of the assurance problem is more fundamental, namely that his overall presentation of the collective action problems appears to rule out the assurance problem in its traditional formulation from the get go. When Buchanan describes the society that we are supposed to imagine for the sake of his arguments, its central features contradict those versions of the assurance problem that would be required to vindicate coercion and enforcement (as opposed to establishing the information service).
Just consider that Buchanan (1985, p. 74) explicitly states that in regard to both (!) collective action problems he "proceeds on the assumption that the individuals in question are motivated by a desire to be charitable, not simply by a desire that the needy be provided for (by someone or other)." But given this assumption that applies to his argument for enforced beneficence as a whole, Buchanan's version of the assurance problem ultimately collapses into the epistemic and knowledge challenge, initially introduced in response to the coordination problem above. Hence, given Buchanan's own assumptions about the motivational states of the individuals populating his envisioned libertarian society, even the "assurance problem" falls short of constituting the kind of challenge that we need to start vindicating coercively-enforced beneficence.

The Three Brothers' Problem and Buchanan's Assumptions
We can see this more clearly when we discuss another, closely-related, objection to my discussion of the assurance problem, namely the so-called "three brothers' problem" in evolutionary theory; also discussed in the economics of altruism. 14 This scenario poses a challenge to my critique of Buchanan because it describes another situation in which altruistic individuals seem to be rationally compelled not to engage in an action that they all acknowledge as necessary for generating a universally-desired collective outcome. We are supposed to imagine three brothers, one of whom (the "recipient") is in some dire emergency. Let us say he fell into a pond, can't swim, and would drown if no one helped him. Each of his two brothers (the two potential "donors") is equally far removed from him, can swim, but is able to rescue him only by incurring some non-negligible risk to his own life.
Evolutionary theorists highlight that even if each of the two donor brothers is basically altruistic and acknowledges that his genetic endowment (shared by the recipient brother in need) will be maximally promoted only when both he and the recipient survive, this attitude alone seems to fail to get the rescuing action going in the case at hand. The result might well be, the presentation of the puzzle concludes, that the brother in need drowns in the pond, leaving only two instead of the maximum three relatives alive and in a position to pass on their genes. 14 I am indebted to an anonymous referee for this journal for drawing my attention to the three brothers' problem and to the literature discussing it. Two varieties of the problem are discussed in two important papers by Eshel and Motro (1988a;1988b). My discussion focuses on the presentation of the problem in Cohen and Motro (1990, p. 56). Obviously, this constitutes an outcome that all three had rationally acknowledged to be less advantageous than the one that would have been feasible.
Why do the three brothers end up with this sub-optimal outcome? The answer is that from the narrow perspective of each of the two donors' rationality, the individually best outcome is to stand by and let the other brother jump into the pond, let him incur the risk of drowning, and to have the three brothers survive (who then have their shared genetic endowment promulgated to the maximum extent). The evolutionary biologists' take home lesson is that this scenario seems to always (!) support the less altruistic brother in terms of her "inclusive fitness" over other relatives who end up engaging in the risky, life threatening, rescuing action. In addition to having his brother getting rescued, the less altruistic brother enjoys the additional evolutionary benefit of having his personal genetic endowment not getting endangered by any risky rescue. In summary, Cohen and Motro (1990, p. 56) state, "this [the rescuing brother's decision in the face of all other potential donors remaining passive] entails an even greater increase in the inclusive fitness of the relatives which decided not to offer their help. It seems, therefore, that if there is any altruistic relative in the vicinity natural selection will always favour the other selfish relatives." At first sight, the three brothers' problem appears to support Buchanan (and undermine my information focused proposal) because it presents at least one case, in which some coercion and enforcement (not mere information services) seem unavoidably necessary to bring about the optimum outcome, in order to overcome the impact that evolutionary forces have on kin selection in the presence of more than one relative. Coercing one of the brothers to rescue the one in dire straits seems necessary in order to realize the outcome of one's kin's genetic endowment being maximally spread. Similar to old-fashioned prisoner's dilemmas, without any enforcement mechanism, a merely suboptimal collective outcome gets produced when all agents act in accordance with what seems the most rational thing to do (from the individual perspective), i.e., to wait for others to take the risk involved in rescuing the brother. This aspect of the three brothers' problem parallels Buchanan's description of what is happening in the case of altruistic individuals failing to provide a shared good that they all deem worth realizing but, due to one or the other collective action problems, are only capable of realizing if an external enforcement mechanism compels them to contribute.
In response to the three brothers' problem, and in concluding my investigation, let me apply another time the crucial distinction between scenarios in which assurance is absent because of some collective knowledge deficit or, alternatively, because the motivational states of the agents involved are unpredictable, unreliable, and unstable. Recall my above reflections on Buchanan's assurance problem as well as on his coordination challenge: With regard to both Buchanan (1985, p. 73) presumed that we are dealing with "a society of morally upright, altruistic libertarians," i.e., a group of individuals, with respect to which "the barrier to successful collective action is [neither] egoism [n]or self-interest in any significant sense." My central proposal has been that in these and in many other passages, Buchanan commits himself to a crucial and consequential presumption. If his argument for enforced beneficence includes this presumption from universal altruism, it undermines his argument for coercion and centralized enforcement mechanisms. The "morally upright libertarians" in question need institutions that overcome the distinctively epistemic deficits characterizing their predicament. Once a planning and knowledge agency provides the exact pieces of information regarding the empirical facts of what each person has to contribute in order to hit the target of effective health care provision for all, the universally-shared and acknowledged altruistic motives take care of the rest. No coercion and enforced contributing enter the picture at all. Hence, no argument is even necessary to justify such practices to begin with.
Alternatively, and this is the second horn of what we might call "Buchanan's dilemma", if we allow that some (many?) members of Buchanan's envisioned libertarian society are prone to free riding, deception, etc. then this not only contradicts many other things that he says (and that I quoted above) but, more problematically, this alternative set of premises lets his argument run into the standard problem that the enforcement in question will be executed against the preferences (and "the will") of non-consenting others, who will then simply reject the claim that they are members of "a society of morally upright, altruistic libertarians" (as it is defined by Buchanan). In that case the issue of coercion indeed becomes a relevant one and an enforcing, not just information-providing, authority must be introduced to realize the collective goods in question by ensuring that enough others contribute. However, framing the collective-actionchallenges this way would amount to Buchanan engaging the controversy concerning the enforcement of (controversial) virtues and actions; a debate in which the libertarian will readily insist that imposing beneficent actions and policies on dissenting agents is a morally impermissible thing to do on part of public institutions. Moreover, Buchanan's writings that I currently examine do not challenge the libertarian on that front. This, in turn, lends further support to my claim that Buchanan's overall argumentative strategy must be interpreted as resting on the alternative assumption of universally-shared beneficence amongst all parties. Now a similar Buchanian dilemma emerges when we revisit the three brothers' problem. While I cannot fully develop an analogously-structured response to that problem here, it should be clear at this point that, given the above reflections and claims, the three brothers case must be further specified in order to really present a challenge to the alternative solution of Buchanan's two collective action challenges. We have to ask, if the two brothers' (that is, the two potential donors') problem is an epistemic predicament or a matter of internal motivational deficiencies? If the latter, then it can be readily agreed that the only way to overcome their hesitance to help their brother is an external enforcement mechanism, forcefully "coordinating" the rescuing effort of their brother and countering the evolutionary pull to free-ride by simply waiting for the other brother to take care of the risky rescue.
As I tried to highlight throughout this essay, this does not at all appear to be a formulation of the three brothers' problem that fits Buchanan's analogous scenario regarding health care provision and its collective action hurdles. A parallel version of the three brothers' problem would presume universally shared altruistic attitudes and motives on part of all brothers, ruling out the desire to free ride from the beginning. Rather, the two potential donor brothers must be envisioned as facing a variety of the above described epistemic problem. That problem, however, can be resolved by the acquisition of information concerning the required act of rescue; again, an act that both are presumed to be willing to undertake. I call this a "variety" of the epistemic problem because in the case of the three brothers we need not merely an information gathering and distributing agencyand it is also for this reason that discussing the three brothers' problem is an enlightening exercise. Different from the good of collectively-provided health care in Buchanan's argument, rescuing the third brother is an indivisible good (that is, it can only be realized by each of the two donors individually).
Hence, in addition to the information concerning the exact contribution that is required to realize the desired collective good in an effective expression of beneficence, the two brothers need an unambiguous procedural mechanism that determines whose turn it is, so to speak. A lottery, for example, might be one way of settling the question of which of the two brothers actually ends up performing the rescue. Again, under the assumption of universally-shared attitudes of beneficence, this lottery is not insisted upon because the brothers distrust each other regarding their attitudes and motives. Rather, they need to generate a specific kind of belief content in order to overcome their currently vague situation in terms of their actions. After all, it would be an irrational waste of resources if both brothers were to jump into the pond, together overdetermining the act of rescue through their uncoordinated individual decisions. 15 These additional issues regarding the three brothers' problem are certainly important and more work needs to be done to spell out the details and their relevance to Buchanan's dilemma. However, the main response continues to consist in the observation that given (!) the assumption that all relevant parties are predisposed altruistically, also the three brothers' problem is susceptible to a non-coercive solution. Enforced contribution (enforced rescuing) is only necessary in case (some) parties' beneficent motives and attitudes are unpredictable and unreliable.
Both, Buchanan's society of "morally upright libertarians" and a three brothers scenario in which all are genuinely and reliably benevolent, indeed present collective action problems that ask for shared solutions -Buchanan has done a lot to correctly identify that point and its relevance for the health care debate. However, the problem, as described and contextualized by Buchanan, is open to getting resolved by entirely non-coercive means. On the other hand, if Buchanan's society and the three brothers were prone to deflection and freeriding, coercion would indeed be necessary to realize the shared goods in question. However, this latter scenario would then shift Buchanan's argument into the familiar territory of standard political-philosophical debates on the justifiability of enforcing contested public moralities on non-consenting members of society. 16