Classic Objections to Utilitarianism
Cosmopolitan utilitarianism is a normative theory of ethics that can be applied at the individual and the institutional levels. In the former case, it describes how single persons ought to behave with each other, distributing duties and rights on the basis of diminishing marginal utility. On the other hand, when it is applied to institutions, the consequentialist scheme of the theory remains untouched, but the allocation of duties and rights is shifted to states, or, more generally, to collective and public decision-making bodies.
Although utilitarianism has been presented in several variants, we can identify some common characteristics that made up the backbone of the theory, and that make it possible to formulate it in cosmopolitan terms. Many of its proposers are divided by deep differences, especially when it comes to the definition of value. What is really valuable for a utilitarian? Happiness? Fulfilment of desire? Pleasure? This is an old story in the utilitarian field. However, this practical difficulty, which pertains mostly to the second-order issue of what produces utility and of its measurement, should not dishearten us about the compactness of the general theory. There are some clear and widely accepted elements in the utilitarian tradition of thought that guarantee its robustness and coherence.
According to Bentham, the great merit of utilitarianism is recognizing that mankind is subject to “two sovereign masters”: pleasure and pain. He describes the principle of utility as one, “which approves or disapproves every action whatsoever, according to the tendency which it appears to have to augment or diminish the happiness of the party whose interest is in question: or, what is the same thing in other words, to promote or to oppose that happiness” (Bentham 2000:14).
Utilitarianism is the ethical theory of consequences. When a person is asked or autonomously decides to make a choice or take a decision, he should pick, from among all the available options, the one that guarantees the greatest benefit for all. The idea is that every person is endowed with a utility curve. Any action taken in the world can have a positive or negative effect on the utility curve of every living and even future individual. Obviously, the magnitude of the impact of someone’s action on my utility curve depends on spatial and temporal propinquity, and, more generally, on the degree of interdependence between us.
If my neighbour decides to move his car from his garage and park it in front of mine, he is causing me a loss of utility on my curve. He is preventing me from using my car for a purpose that I deem valuable - for example, going to a concert. If we assume that I was supposed to go to that concert with a friend, my neighbour is also affecting negatively the utility curve of my friend who will be obliged to spend two hours in the concert hall without the company he was expecting, which I would have offered him, had I been able to drive my car out of my garage. My friend and I have been damaged by the negative consequences, even if the utility we have been deprived of is different. He has been forced to see the concert alone, but I have completely missed it; so, in a way, I am worse off than him with respect to the concert.
Nevertheless, the consequences of my neighbour’s action are not limited to the single event of the concert. By turning on the engine, he has produced a given quantity of carbon dioxide. The small pollution he has caused in his very brief journey will combine with the other carbon dioxide that is floating in the air and, which, next week, will surpass the maximum-tolerated threshold, thus inducing the mayor to close the entire neighbourhood to car traffic until the legal parameters concerning the presence of CO2 in the air are not met again. The inevitable action of the mayor - that is, the result of the sum of other deliberative individual actions - will produce another inevitable array of consequences on the lives of the people living in the area, affecting (negatively) their utility curves.
But the consequences of this apparently insignificant action performed by my neighbour do not stop here, neither spatially nor temporally. The pollution released by his exhaust pipe will erode an infinitesimal quantity of the one trillion tonnes of CO2 we human beings can still emit before causing a global warming of 2C above pre-industrial levels (Shue 2011).
Apparently, the action performed by my neighbour is a negative one. He has driven down my utility curve in a substantial way. Something similar has happened to my friend who was waiting for me at the concert hall. Furthermore, his action, together with those of other people living in the area, has contributed to a loss of welfare among the inhabitants of the neighbourhood. Moreover, even if in infinitesimal terms, it has had a negative effect on the utility curves of future generations living in the most disparate regions of the world.
Nonetheless, for a utilitarian, these empirical facts would not be enough to label the action as a negative one, as something to be avoided. Everything depends on why my neighbour performed his action. There might be a trade-off between the negative consequences listed so far and some positive consequences that it yielded. If my neighbour left the car in front of my garage because he was too lazy to open the door of his garage and drive the car inside, we can assume that he registered a utility increase on his curve. But this positive effect is, in absolute terms, too small to compensate for the loss of utility he has caused me and my friend. So, if this was the reason behind his decision, we can conclude that, in utilitarian terms, his action is wrong because it has failed to maximize global utility.
On the other hand, if we learn that my neighbour moved his car from his garage to the entrance of mine because he had called the ambulance to rescue his son, who was very sick, but the door on the stairs was too small to let the gurney enter and, hence, the only available solution was to let it pass through the garage door, we would have to change our judgment about his action. In this case, a utilitarian would not hesitate to approve this action as a positive one. Obviously, the reason is that the only thing that matters to a utilitarian is overall utility. This opens the way to an infinite number of welfare trade-offs between single individuals, or between groups and single individuals.
Amartya Sen has strongly criticized what he defined as the “sum-ranking” component of utilitarianism. According to him, the exclusive focus on general well-being risks being insensitive to inequalities in the distribution of utility (Sen 1999: 59-63). A huge increase in the utility of a wealthy person can outweigh a small increase in the utility of a group of destitute people.
Moreover, this problem is exacerbated by the psychological phenomenon of “mental conditioning and adaptive attitudes” (Sen 1999: 62). A man who was born in poverty might be quite happy with his condition and may not wish for an increase in wealth. Or, his desire for an increase in wealth might be inferior to that of a richer person for the same proportional increase in wealth. In this case, the right distribution - meaning the distribution that maximizes global utility - is the more unequal one because the happiness we can produce by allocating the same amount of goods to a rich or to a poor person is higher if we allocate it to the rich person.
Finally, Sen highlights that the utilitarian theory neglects rights (Sen 1999: 62). Even fundamental human rights, I would add. Imagine a man who was born in slavery and has spent all his life serving his master’s family. Since he has never experienced a different condition, serving his master may have become his life’s mission. He might also feel some pleasure in doing it. In this scenario, we may even imagine that freedom from slavery would cause him a loss of utility. When rights are concerned, utilitarianism has some problems in reconciling them with utility maximization and, more generally, with the meta-ethical point made by Bentham about pleasure and pain as the “two sovereign masters” of mankind.
In his discourse on utilitarianism and global justice, Charles Jones has raised similar objections to the utilitarian theory of international moral obligation. He presented the two ways in which utilitarianism is deemed as “too permissive”. The first of these can be traced to Sen’s argument about the distributional indifference of the utilitarian principle that I have briefly sketched above. Jones tries to save utilitarian theory from this criticism by limiting the scope of the utility principle to “the satisfaction of basic wants or vital interests” - that is to say “the necessary conditions of a recognizably human life” (shelter, water, clothing, basic education and basic healthcare) (Jones 1999: 25).
Through this move, he prevents trade-offs between the people who live below the threshold of basic wants fulfilment and the rest of society. Since the only thing that is valuable for “basic interest utilitarianism” (BIU) is the satisfaction of basic wants, as long as there exists even one person whose basic wants remain unfulfilled, the modified utility principle will trigger a redistribution mechanism that is deeply egalitarian, in the sense that moves every resource towards the bottom of society. One criticism against BIU is that the restriction in the scope of the utility principle (the neutralization of the utility-generating capacity of everything that is not vital) is an ad hoc move. Jones accepts that it might seem so, but says that, in the current international scenario, there are strong empirical reasons for focusing on basic wants. First, basic wants are the most cherished by human beings. Second, the enjoyment of the most fundamental human rights depends on the previous fulfilment of subsistence rights. This is the classic argument used by Henry Shue to uphold the existence of economic human rights. As he wrote in Basic Rights, “no one can fully, if at all, enjoy any right that is supposedly protected by society if he or she lacks the essentials for a reasonably healthy and active life” (Shue 1996: 24). Third, it can be incredibly difficult to assist people living abroad in the fulfilment of no basic needs. This difficulty is both organizational and economic. Therefore, it is much more practical and realistic to focus on the satisfaction of vital needs. Lastly, Jones says that this restricted scope of the utility principle can better sustain “the utilitarian’s ultimate aim of promoting human well-being” (Jones 1999: 27-28).
The second reason why utilitarianism is accused of being “too permissive” can be summarized in Stuart Hampshire’s “objection from historical consciousness” (Jones 1999:40). Political and social arrangements find a consistent part of their justification in backward-looking reasons, but consequentialist theories like utilitarianism fail to recognize these undeniable references. Utilitarians respond to this attack by saying that if backward-looking elements re-emerge in the desires of living people, then the utility principle would inevitably take account of them; if they do not, then we ought to accept that there is no place for them in ethics. Jones thinks that both positions (Hampshire’s and classic utilitarianism) suffer from the same drawback. Both have to find a place in their moral schemes for such desires as racism. Hampshire would have to do it on the basis of “the equal claims of memory and imagination to supply moral direction”. Classic utilitarianism, on the other hand, is incapable of providing a qualitative distinction between individual wants. Hence, even here, Jones holds that the only way to deal with the excessive permissiveness of classic utilitarianism is to shift to BIU (Jones 1999: 40-41). For if we limit the scope of the utility principle to basic interests, then the moral value of the historical consciousness of a community will never be traded off with the vital needs of the individuals, both internal or external, to that community.
These were the accusations of excessive permissiveness levelled against utilitarianism: indifference to equality in the redistributive process and indifference to backward-looking justifications of political and social arrangements. Nonetheless, utilitarianism has also suffered a strong objection of the opposite nature: excessive demand. This criticism has mostly been developed around Peter Singer’s cosmopolitan interpretation of the utility principle. Singer has presented a moral argument that is incredibly powerful in its simplicity:
suffering and death from lack of food, shelter and medical care are bad … [therefore] if it is in our power to prevent something bad from happening, without sacrificing anything of comparable moral importance, we ought, morally, to do it (Singer 1972: 231).
The expression “comparable moral importance” makes Singer’s theory one of the most radical global redistributive principles ever proposed. If I assume that in this moment there are n people in the world dying from a lack of economic resources and that each of these n people could be saved through a $10 donation, I should analyse every good that happens to be in my possession and ask whether its moral importance is comparable to the moral importance of the lives that might be saved by its economic value. None of the goods I own would pass the test. At the end, I would be morally obliged to sell everything I own, up to the point where this process would leave me worse off than the persons I am trying to save.
Obviously, when Singer presented his theory more than 40 years ago, he was not expecting humankind to put this moral principle in practice. Not even he did it. What he really wanted was to demonstrate that the life we live is structurally wrong from the moral point of view. The setting of the moral goal, even if it is practically unachievable, serves to trace a direction. The more human beings get close to that goal, the better.
In 2009, Singer recognized that upholding moral principles that nobody will ever fully adopt because they are too demanding risks making the people “question the point of striving to live an ethical life at all” (Singer 2010: 151). Therefore, he has recently proposed as a public moral standard a progressive system of taxation that involves the richest 10 per cent of the society, without drawing resources from the remaining 90 per cent (Singer 2010: 160-169).
This more “realistic approach”, as Singer calls it, is surely less problematic, since it is designed to raise the money needed to prevent easily preventable deaths, without altering the standard of living in developed countries. Moreover, apart from the quantitative aspect, the new public standard proposed by Singer introduces an institutional element in the redistributive mechanism. The progressive taxation he has devised requires backing by a state law and, thus, is compulsory. Once it has been agreed upon by the political majority none can escape it. This institutional move softens the problem of the excessive demands of the utility principle in the case (very likely) in which only a restricted group of people is willing to stick to the utilitarian moral commitment. In utilitarian interactional logic, the smaller the group of people that is willing to undertake the morally correct action, the more burdensome it is for them to undertake that action. If all the wealthy people in the world were to prevent deaths from hunger at the same time, it would cost less money per person. If, on the contrary, only a few people are ready to redistribute resources, they would be obliged (according to the utility principle) to renounce to all their wealth and, probably, would not even manage to solve the problem they were addressing.
Nonetheless, as Charles Jones rightly points out (Jones 1999: 37-38), the shift from individuals to institutions changes the nature of the moral obligation, but does not save the theory from the objection over its excessive demand. For if it is true that within an institutional approach single individuals are not burdened with the moral duty of unilateral redistribution, they still have the moral duty to put in place those institutions that are supposed to impose and coordinate a collective redistribution of resources guided by the utility principle.
Local Inaction in a Globalised Context
The arguments exposed above are the classic objections levelled against utilitarianism. Over recent decades, the proponents of the theory have managed to soften its more radical aspects by limiting the scope of the utility principle, thus introducing a sort of “priority clause”, and by shifting from an interactional to an institutional approach. Nonetheless, a further objection that has been underestimated concerns the case in which the only way to maximize utility in a global context is to renounce action at the national or local level. This problem is caused by the principles that make up the utilitarian theory; hence, it can also emerge at the local level, but is exacerbated in the cosmopolitan version. My thesis propounds that this difficulty, stemming from a misalignment between the decision-making body and the persons whose utility curve is affected by the action/inaction of that body, risks creating a dangerous moral confusion in the electorate that threatens the global applicability of utilitarianism as a political theory.
Imagine the following case. A car company is going to close down an industrial plant in a city - let us call it Old City - of a developed country. This decision has been taken because, given the current situation, it is much more convenient for the company to delocalize the car production to another city - let us say New City - in a developing country. In Old City, standard wages are too high, the taxation level is relatively unbearable and the security requirements are highly demanding. New City can offer a relative advantage in all three areas of interest.
If the industrial plant closes down, 1,000 workers will lose their job. They will not suffer from hunger because, having worked under favourable conditions during previous years, everyone has managed to save a sufficient amount of money. Moreover, their country offers a sound welfare programme; hence, from the day after the plant closes down, they will start receiving unemployment help. They will have to reduce their living standards by half but will still be able to lead a decent life.
Meanwhile, if the company inaugurates a new plant in New City, 1,000 persons who are currently unemployed, not assisted by the state and, thus, living on the edge of starvation, will get jobs. Their children will start going to school. They will be given accommodation and gain access to a decent life.
Obviously, workers in Old City do not want to lose their jobs. They go on strike and ask the city council to intervene. The mayor has a powerful ace up his sleeve. Every year, his municipality receives a remarkable amount of money from the federal government as funds for industrial development. It is completely up to him whether to use it or not. If he does not use the money in a given year, it goes back to the federal government and the municipality loses it completely. If the mayor uses the fund to subsidize the car company, the CEO will not delocalize and the workers will save their jobs. If, on the contrary, he does not do so, then the industrial plant will be moved to New City and Old City will lose the money in the fund.
What should the mayor do?
According to cosmopolitan utilitarianism, the rational choice in private life as well as in public policy consists in picking among available alternative actions the one that maximizes the happiness or the utility of humanity as a whole. In the case I have presented, the best maximizing option for the mayor is to renounce his political action. A cosmopolitan utilitarian would be obliged to accept that the best thing for the mayor to do is keep the money in the fund and let the industrial plant fail. In order words, he must lose the money and indirectly drive 1,000 fellow citizens into unemployment.
The reason, clearly, is the classic utilitarian argument of decreasing marginal utility. A job contract in New City can generate a much higher utility than in Old City. Workers in New City are much poorer than workers in Old City, and every additional dollar coming from a job contract will create a higher utility in New City than in Old City. If workers in New City lose their jobs, they will reduce their living standards by half, but their life will still be more than acceptable. In contrast, workers in New City are desperately looking for jobs and live close to starvation levels; a job contract will enable them to access the basic goods needed to start a decent life.
The problem with this extreme conclusion is that cosmopolitan utilitarianism does not simply suggest that the policy decision maker should act only in the way that maximizes global utility independently from local or national affiliations. In this case, cosmopolitan utilitarianism requires the politician who has the means and the power to help his community to renounce his agency. He should do nothing and opt for inaction because here inaction is the maximizing action.
This example can seriously challenge the universal applicability of utilitarian ethics in public policy. If a utilitarian approach that results in a cosmopolitanism of action can still be promoted as a normative theory, then the extreme case of the inaction that maximizes the utility by neutralizing local political activity risks being structurally rejected by public opinion.
When discussing cosmopolitan utilitarianism, we tend to take into consideration positive actions. A standard example may be the following. Two boats are going to sink off the coast of Old City. On the first boat are four people from Old City. On the second one are five people from New City. Unfortunately, the municipality of Old City has only one rescue vessel at its disposal. Since the two sinking boats are far from each other and there is not enough time to reach both of them, the mayor must decide which boat to rescue.
According to cosmopolitan utilitarianism, the mayor should opt for the action that maximizes global utility. He should not take into consideration things like citizenship or local affiliation. Every utility curve has an equal value, independent of the gender, sex or nationality of the relevant person. Saving five lives is better than saving just four. Hence, the mayor should send the vessel to rescue the people from New City, letting his four fellow citizens die.
Obviously, many people would reject this conclusion. Some would question the basic assumptions of utilitarian ethics. Some others would claim the existence of special duties towards compatriots. But the cosmopolitan utilitarian can still defend his argument. He holds that the public decision maker should exercise his political power in one way rather than another. He should do A (saving lives irrespective of ID) instead of B (saving lives taking into consideration special duties toward compatriots). Put in this way, as an alternative theory of action, cosmopolitan utilitarianism can find its own credible place in the political debate.
But the classic case of action A instead of action B does not exhaust all the possible scenarios that a political decision maker faces during his term in office. In a globalized world, a case like that of the car company is common, and the consequentialist soul of cosmopolitan utilitarianism risks binding the politician to a series of inactions that contradict his own job and mandate. If we look only at the consequences of our actions, then delocalization to New City is clearly the best option. But the cosmopolitan utilitarian recommends to the mayor not to do A instead of B (as in the case of the sinking boats), but rather to do nothing. Just letting things happen, delegating the decision to other private and public (foreign) decision bodies. Not helping his own fellow citizens that have chosen him as their political representative because from their suffering a greater happiness will emerge abroad.
Inaction in Local Supererogation
The problem of political inaction emerges more clearly and more often in the cosmopolitan formulation of the utilitarian theory than in the classic one. This happens because, in the former, the principle of decreasing marginal utility is freed from every local or communitarian boundary. When Bentham described how institutions were to apply the utility principle in the provision of public policies, he gave this simple definition:
a measure of government (which is but a particular kind of action, performed by a particular person or persons) may be said to be conformable to or dictated by the principle of utility, when in like manner the tendency which it has to augment the happiness of the community is greater than any which it has to diminish it (Bentham 2000: 15).
Here, Bentham holds that the sum-ranking logic, intrinsic to every form of utilitarianism, is limited to the community. In another passage the British philosopher describes the principle of utility as the one that “approves or disapproves of every action whatsoever according to the tendency it appears to have to augment or diminish the happiness of the party whose interest is in question” (Bentham 2000: 14). Russell Hardin maintains something similar when he writes that utilitarianism is “the moral theory that judges goodness of outcomes - and therefore the rightness of actions insofar as they affect outcomes - by the degree to which they secure the greatest benefit to all concerned” (Hardin 1988: xv).
Both definitions leave room for extension or restriction of the moral scope. Who are the persons whose utility the government has to consider as part of the total sum? Who are the members of the “community”? Who are “all [the people] concerned” or “the party whose interest is in question”? Peter Niesen holds that there are two possible responses. The first one consists in the “all-affected interpretation” of Bentham’s passage: “the happiness of the party whose interest is in question”. According to this interpretation, when evaluating the consequences of a specific action the actor should take into consideration “all those causally influenced by their actions” (Niesen 2012: 6-7). If we read Bentham in this way, his utility principle is clearly cosmopolitan, because the scope of the moral judgment about the action is given by the extent of its consequences, regardless of any local or political affiliation.
The second reading of the passage consists in the “all-subjected” interpretation proposed by David Lyons. From this perspective the interests that the government ought to maximize are those of the people that are subjected to its power. The non-residents are thus excluded from the moral calculus (Niesen 2012: 7).
We may hold that the government ought to maximize the total utility of its political community, meaning national citizens. Or we may maintain that every human being is part of a community that ought to be taken into consideration when evaluating the consequences of a specific action. If we adopt the latter view, as cosmopolitan utilitarians do, we determine a misalignment between the electoral body and the persons that the government ought to consider while evaluating the consequences of a given policy.
Obviously, such misalignment exacerbates the problem of forced inaction by increasing in number the circumstances in which it can occur. But its origins can be traced to the intrinsic characteristics of the utility principle and of decreasing marginal utility. A domestic case in which utilitarianism compels the decision maker to an inaction that is difficult to be accepted both from a political and from a purely intuitive point of view concerns the performance of a supererogatory act by a person exercising a public office.
Imagine this second scenario. We are in a university. Every professor teaching in this institution is required, besides lecturing activities, to receive students in his office once a week. The day should be fixed in advance before the semester starts. During that day, the office is open for three hours. All of a sudden, an announcement of competition is published. It says that one of the current students has the opportunity to spend one year abroad in a prestigious university. This experience will probably change for good the life of the student who will be enrolled in the exchange programme. In order to participate in the selection, interested candidates have to submit their curriculum and a presentation letter from their supervisor.
Many students submit the application, but there are only two students with the right grades to win over the others: John and Philip. Both have the same supervisor, Professor Stuart, who has fixed office hours on Monday. John reads the announcement as soon as it is published, weeks before the deadline. So, he has all the time to prepare his application. Philip, however, does not see it in time because he was spending some months abroad doing research and his classmates do not inform him of the announcement because they think him to be a too dangerous competitor. Philip reads the notice of competition three days before the deadline, on Tuesday. He needs the letter from his supervisor, but Professor Stuart is not required to serve office hours until the next Monday, which is after the deadline. However, he is completely free. As the semester is almost over, he has no lectures or exams, and is supposed to sit in his office every day of the current week.
Should Professor Stuart open the door and sign the presentation letter for Philip?
He knows that Philip has the best grades in school and that if he gets his letter signed, he will be selected. Looking at the issue from a different point of view, he knows that if he signs Philip’s letter, John, who at the moment is the best applicant, will lose the competition. Philip comes from a wealthy family. Getting enrolled in this prestigious programme and starting an academic career is his dream, but it is not as economically important for him. He can keep on studying relying on his high family income while waiting for the next opportunity. On the other hand, John, the second-best student, depends on a scholarship that will expire within few months. If he does not get this opportunity now, he will probably not get a second one.
Opening the door, receiving Philip and writing a letter for him would take 30 minutes. Professor Stuart is a good man and completely free during these days. He would be willing to offer 30 minutes of extra work to the state to do a good deed. But is helping Philip compete with the others a good deed?
Philip cannot claim the right to have the letter signed in the name of a more general right to free and fair competition for public office. The university rules are very clear. Every professor is required to receive students only once a week, on a specific day (in this case Monday) that is fixed at the beginning of the semester. Not having realized the existence of the notice of completion in time to knock on the door of Professor Stuart on Monday, he has lost the chance to take part in the competition. On the other hand, Professor Stuart is willing to bring him back on track, offering 30 minutes of extra work and is also free to do it. Receiving Philip would not divert human resources from other worthy activities.
Helping a student by offering him some minutes of extra work seems to be a good deed. The limit of “one day of consulting hours per week” has been set to guarantee the professor against working performances that go beyond his job contract. It has been decided that given his contract he cannot be asked to open his office door to students more than once a week. The professor is the only beneficiary of this guarantee. Students like Philip cannot infringe upon it, because if they were able to do so, the guarantee would have no value.
Since the professor is the sole beneficiary of the guarantee, he is the only one who can depart from it. If Professor Stuart is not coerced, not threatened and well-informed, he is free to perform a supererogatory act by opening his office door to Philip on a day that is not Monday. Some people may resist the claim that this represents a good deed. They may wonder why Professor Stuart is willing to extend his consulting hours. And they may find out that he does not really care about helping Philip, but is doing it only for narcissistic reasons - perhaps to promote his own positive image in the department. However, intuitively, nobody would condemn it, just as nobody would condemn the school teacher who grades the written exams at home, after dinner, because doing it only at school would take up too much time and she does not want her students to wait so long; or the magistrate who spends the evening in her office because she wants to prepare all the documents before the first hearing of the appeal.
These people are over-performing at their jobs. If we assume that their job tasks are right (grading exams, judging the accused, and so on), then we cannot condemn a supererogatory act that goes in the same direction. But with utilitarianism, the picture becomes much more complicated and we are forced to question such an act that has easily passed an intuitive test.
If we return to Philip’s case, a utilitarian cannot limit his moral analysis to the intrinsic characteristics of the supererogatory act performed by Professor Stuart; rather, he needs to look at the consequences. If Professor Stuart opens his door when he is not required to do it, Philip will surely win the competition. He will be enrolled in a prestigious and well-funded exchange programme that will probably be the starting point for a promising career.
But the utilitarian would delve deeper and ask: Who is Philip? The answer is that he is from a wealthy family. He does not really need to win this competition to change his life. He can keep on living on his family money, waiting for the next train, which might even be better than the current one. The second question that the act utilitarian would ask is: Who are Philip’s competitors? There are hundreds of them. But the utilitarian analysis is concerned with knowing that before Professor Stuart opens the door, the virtual winner of the competition is John. Who is John? He comes from a low-income family, his scholarship is running to an end, and he desperately needs a new one. If he is enrolled in the exchange programme, he will really have the opportunity to change his life. If he does not, he will probably be forced to abandon, at least temporarily, his life plans. It is clear that the utility John would get from the exchange programme is much higher than Philip’s.
We have, as the utilitarian would say, a status quo situation (John wins the competition) and a situation B (Philip gets his presentation letter signed and wins the competition). By performing a little supererogatory act (opening the door to Philip on Thursday), Professor Stuart causes a direct move from status quo to situation B. This move is highly inefficient from the point of view of utility. It subtracts a huge utility from John’s utility function, while allocating a much smaller one to Philip’s utility function. From the point of view of universal utility maximization (that is what really matters to a utilitarian), the status quo is preferable to situation B. So, if Professor Stuart really wants to do good, he should just remain inactive - he should sit in his office for hours (since as we said above during that specific week he has nothing to do) and pretend he does not hear Philip knocking.
Sticking to the utilitarian argument in the case of Professor Stuart, to perform a supererogatory act in the management of his public office is wrong. The conclusion is that to fulfil a certain task, to receive students, is a good action up to a specific threshold (in this case, measured on the time scale). Once the time limit has passed, the same action that was deemed as good below the threshold becomes a bad action.
We can recognize that there is clear logic beyond this conclusion. Hence, speaking of a contradiction would be misleading. But the point is that this apparent contradiction in the moral analysis of the tasks assigned to a person who holds a public office is difficult to reconcile with the evaluation of the content of the supererogatory act and risks creating a dangerous moral confusion.
Obviously, the problem with inaction in the case of a supererogatory act does not indicate a more general friction between the law and utilitarian ethics. Had Philip read the competition announcement soon enough and knocked on Professor Stuart’s door on Monday, even the utilitarian would have agreed that it is Professor Stuart’s duty to open the door and sign Philip’s presentation letter. He would not be allowed to balance consequences against a procedural right that upholds Philip’s claim. Once rules have been established, the utilitarian does not require the decision maker to maximize utility against the rules. Were it so, utilitarianism would result in a comprehensive anarchical doctrine. The conflict emerges when the decision maker is not obliged by the rules to act in a specific way but wants to conform his actions to the general rules concerning his office (in the case of Professor Stuart, to guarantee his students equal access to competition).
The more the world becomes globalized through faster movements of people, information and capital, the more cosmopolitan utilitarianism shows its limits as a universal normative theory of public policy. By asking the decision maker to opt for the alternative that produces the best consequences, cosmopolitan utilitarianism compels him to refrain from easily achievable actions that may promote the well-being of his political community. For an external observer, this may seem the logical thing to do - to reduce utility in a place in order to create a much bigger utility in another place. But this strongly contradicts the fundamental aim of public institutions. A normative doctrine claiming that, in certain situations, the best thing a politician can do is to remain inactive in order to reduce the well-being of the political community he is ruling over risks being politically unsustainable.
A cosmopolitan utilitarian may respond that, in the car company instance examined in this article, his theory requires the politician to act against the well-being of his fellow citizens, but not necessarily against their interest. In fact, utilitarianism is first an ethical theory, and if the majority of the political community accepts it, then they cannot but wish for delocalization. In this case, the mayor’s inaction operates against their well-being, but not against their interest. For their interest would be to reduce their well-being in order to maximize world utility. As per Peter Singer, the man bumping into the drowning child should be willing to reduce his well-being (by wasting his brand new shoes) in order to uplift the utility curve of the dying child (Singer 2010: 3-5).
A situation in which the majority of the community members embraces utilitarian cosmopolitan ethics is unrealistic. However, it may happen that the utilitarian comes to power or, rather, that the ruling politician persuades himself of the desirability of a utilitarian policy. But implementing public policies inspired by cosmopolitan utilitarianism in a polity where cosmopolitan utilitarianism is not widely accepted from an ethical point of view will prove politically unsustainable.
Nevertheless, let us try to accept a hypothetical counterargument according to which a cosmopolitan utilitarian is not seeking to impose (or propose) cosmopolitan utilitarian policies on a political majority that does not accept the basic tenets of the theory. Rather, he wants to persuade the majority of the theoretical robustness and desirability of his theory, until the moment in which the community consciously elects a politician who is supposed to implement a number of cosmopolitan utilitarian policies.
We can assume that this result has been achieved and that the majority of the citizens living in Old City has recognized the theoretical superiority of the utilitarian normative theory. They warmly welcome things like cutting funds to private schools in order to divert public money toward overcrowded state schools. These are actions that can be easily reconciled with the public perception of political action. The municipality stops performing action A (funding private schools) because it deems preferable directing economic resources towards action B (creating better conditions in state schools).
But the problem of global or local inaction continues to be troublesome even in this utilitarian Eden. The first reason is that, as we have seen already, there are cases in which a public decision maker should stop (or abstain from) performing action A in order to let a foreign decision maker perform action B. This means that there are some circumstances in which the public decision maker should abstain on purpose from increasing, or keeping constant, the well-being of his community in order to create the best conditions for another decision maker to promote the wellbeing of another community.
The second reason is that the commitment to look at the consequences of every political decision may result in a very long process of action delegation. The instance of the car company is a basic example of a two-country scenario. But what if we introduce other cities in the scheme - New City 2, New City 3, New City 4, New City 5, and so on - where the numbers (1, 2, 3, 4, etc.) indicate an increase in their poverty level? If capital flows easily across borders, New City Municipality should consider the consequences of accepting a foreign direct investment from the car company. If prospective workers in New City 2 are more destitute than in New City, then the best thing for New City Municipality to do is renounce the investment or act against it, because, if it can be assumed that given a constant quality of work performance, capital seeks the lowest labour costs and labour costs are lower where the income level is lower, then we can reasonably expect that if New City renounces the investment, the car company will open its industrial plant in New City 2.
Why not go directly to New City 2 if the cost of labour is lower in comparison with New City? This is because New City 2 has more corruption and bureaucratic costs are higher. This makes New City the best option on the list for the car company, which wants to delocalize away from Old City. But if New City eliminates itself from the list, the car company will necessarily move to the second best option, which is New City 2.
The cosmopolitan utilitarian would recommend that New City Municipality leave its own fellow citizens unemployed in order to let capital flow to New City 2. The same thing would hold true for New City 2 towards New City 3, and for New City 3 towards New City 4, and so on. As long as there is some other decision maker who may benefit from others’ inaction nobody should undertake a political action. The process of action delegation stops when we reach the bottom, where the effects of a fixed number of positive opportunities (in our case 1,000 job contracts) on the utility curves are higher.
This problem can appear even at the local level. In the article, I have given the example of the supererogatory act performed in the exercise of a public office. The fulfilment of a task related to a specific job may be acceptable up to a threshold that corresponds to its compulsoriness. Once this threshold has been surpassed and the moral rightness of a given act ceases to be guaranteed by its rule-based compulsoriness, a supererogatory performance of the same act may become morally unacceptable because of the principle of decreasing marginal utility.
My thesis is that such a shift in moral judgment over the same act risks creating a dangerous moral confusion. This problem does not emerge as often at the national level because of an extensive regulation of the interaction between social and decision-making bodies. In contrast, within the global context, relations between decision-making bodies belonging to different countries are generally unregulated, and there is a misalignment between the electorate and the persons that every decision-making body reaches through the consequences of the policies it adopts. These two elements expose cosmopolitan utilitarianism to a constant risk of international action delegation that threatens the domestic political acceptability of this political theory.
 Luis Cabrera, for example, has argued that an halt at satisfying basic wants would be prevented by the same intrinsic logic of utilitarian reasoning (Cabrera 2004: 35-36) .
 Singer recognizes the theoretical existence of what he defines as ‘eccentric positions’ on this point. He accepts that we cannot deny the existence of a person who does not deem bad a death from starvation or the suffering of illness. Nonetheless, he thinks that such cases are simple exceptions that should not threaten the general validity of his philosophical premise.
 Remember Singer, “if it is in our power to prevent something bad from happening, without sacrificing anything of comparable moral importance, we ought, morally, to do it” (Singer 1972: 231).
 See also Jones 1999: 23.
 We can assume that also an act utilitarian (an utilitarian that applies the utilitarian principle to particular acts) would agree that Professor Stuart should respect Philip’s right to get the letter signed, if he asks for it on Monday, even if this action will lower overall utility. The reason is that the negative effects on overall utility caused by the infringement of Philip’s right (in terms of erosion of trust in the university system) would outweigh the positive effect of John winning the competition. Anthony Ellis made a similar point by writing that “even when he [the act utilitarian] thinks that it would maximize utility to deviate from the rule he may be justified in placing greater trust in the rule and following it” (Ellis 1992: 171).