Reading: An Unhappy Pursuit of Happiness


A- A+
Alt. Display


An Unhappy Pursuit of Happiness


Adam Oliver

Department of Social Policy, LSE, GB
X close


Over recent years, academics and policy makers in several countries have been advocating for measures of utility and happiness to replace gross domestic product as indicators of development. Yet the notion of utility has a somewhat confused history, meaning different things to different people at different times. Hume, for instance, aligned utility with public usefulness, Bentham with hedonic feelings of pleasure and pain, and Mill and modern welfare economists with pretty much anything, and this confusion has not been absent from contemporary calls for utilitarianism to inform public policy design and intervention. A possible reason why there are many different meanings attached to the concept of utility is because many people, much of the time, are not driven to maximise utility at all, irrespective of how it is defined. That is, the pursuit of utility does not drive all desires, but rather desires are antecedent; in those circumstances, at most, some conception of utility is a possible consequence of achieving one’s desires. Desires are multifarious and vary across people. The policy maker’s role over the private realm of individual decision-making should not therefore be to strive to maximise utility, but rather to put in place conditions that facilitate people in the pursuit of their own conception of a desired life.
How to Cite: Oliver, A., 2021. An Unhappy Pursuit of Happiness. LSE Public Policy Review, 1(3), p.3. DOI:
  Published on 05 Mar 2021
 Accepted on 05 Feb 2021            Submitted on 14 Nov 2020


Over recent years, there have been growing calls from some quarters to replace gross domestic product as an indicator of a country’s development with utility or happiness indices. Beyond a particular national income threshold, so the argument goes, increases in income appear to generate little if any additional aggregate utility, and thus public policy ought to focus more directly on improving the happiness of citizens. As I will detail later, many of those advocating this neo-utilitarian approach initially appeared to hold ‘pleasure and pain’ as the standard for utility but have since moved closer to “general life satisfaction” instead (and some are attempting to combine these with measures of ‘meaningfulness and fulfilment’ also). These apparent uncertainties regarding what exactly utility comprises are nothing new; indeed, I will argue in this article that the concept of utility, hereby treated interchangeably with welfare and happiness, has a long and confusing history in economic thought.1

Before surveying the competing definitions that have been attached to the concept of utility, however, it might be instructive to highlight how the neo-utilitarians, like the classical utilitarians before them, would, in broad terms, inform the design and delivery of public policy with their doctrine. In a nutshell, they would call on policy makers to seek to maximise utility within society, subject to resource constraints: in essence, it is a rationalist social planning perspective. Although social planning makes sense over public sector services, such as health care and education (so long as the outcomes produced by those services are broadly accepted by a society’s citizens—e.g., health from health care; literacy and numeracy from education), one may question the encroachment of standardised outcome-based prescriptions, whether these be utility-focused or otherwise, on the private realm of individual lives. We must, I argue towards the end of this article, respect the multifarious desires that individual people wish to pursue in their private lives—so long as they impose no substantive harms on others—and resist the urge to steer people towards some standardised general objective. But that is a discussion for later. For now, let us consider how utility has been variously defined.

Public Usefulness, Hedonism, and Beyond

David Hume, one of the earliest writers on utility, equated it with public usefulness. He argued that social virtues, such as cooperation and reciprocity, derive most of their merit from their promoting social utility, which he viewed as the ultimate goal of humanity [1]. In a later work, Hume referred to the ‘happiness of society’ as being the basis of morality and proposed that there are three components to happiness: action, pleasure, and indolence [2]. Human actions are driven, according to Hume, by a sense of public usefulness and are undertaken in the pursuit of our enlightened self-interest. Hume therefore saw the pursuit of the social good as being consistent with the pursuit of self-love, although he viewed self-love to the point of avarice as reprehensible. However, as implied above, he acknowledged that too much action may be counterproductive, with happiness also dependent in part on respite. While Hume accounts for pleasure within his theory of utility, it is just one of three elements,2 distinguishing him from the eighteenth and nineteenth century British utilitarians, who restricted themselves to pleasure and pain alone.

Jeremy Bentham, widely believed to be the founding father of classical utilitarianism, famously argued that actions are governed by the sovereign masters of pleasure and pain (the so-called hedonic measures). For Bentham, these two elements constitute utility and are experienced on a continuum that allows them to be compared, and he maintained that societies ought to strive for the greatest happiness for the greatest number [3]. Objections to Benthamite utilitarianism are legion; Sen summarised three of the most prominent criticisms as follows [4]. First, there is the ‘happy slave’ argument, where a person may have adapted to, and perhaps accepted, circumstances that are objectively bad. If so, then a Benthamite utilitarian would propose that nothing needs to be done to change that person’s circumstances. Second, Benthamite utilitarianism overlooks the importance of addressing inequality; that is, there may be individuals or groups within society who are disadvantaged to the point where, from a social justice perspective, we may wish to target policy action towards them, even if other less disadvantaged groups could gain a greater increase in utility than that experienced by the disadvantaged, so increasing the overall happiness of society to a greater degree, if they were the target of policy action instead.3 Third, there is the argument that consequences should not supersede all considerations of rights. For instance, all else equal, a lower gain in utility might be expected from a health care intervention that saves the life of a 45-year-old compared to that of a 35-year-old simply due to the fact that, with treatment, the older person has a lower expected number of years of life left, and yet it is likely that there would be broad support for the statement that these individuals have an equal right to treatment.4

Sen argued that rather than attempting to maximise hedonic utility across society, policy makers ought to try to move societies towards a just distribution of basic human functionings (e.g., health, literacy) through access to the capabilities that help people to achieve those functionings (e.g., decent health care and education services). Thus, unlike a person’s subjective assessment of their own feelings of pleasure and pain, Sen’s functionings are chosen via an objective assessment of what is believed necessary for people to flourish (admittedly, this objective assessment is somewhat subjective, but it is not the subjective assessment of one’s own feelings about oneself). Sen’s approach can serve as a justification for the domains over which one might introduce a public sector service. For instance, if it is decided that health and literacy are basic functionings that cannot be delivered justly or efficiently through private actions, then there is a strong argument for State intervention to ensure adequate provision of the services that deliver these functionings. But for personal objectives that are not basic functionings (e.g., the desire to own a nice car or to always keep one’s promises or to meticulously read all of the writings of an LSE academic), there is no need for State intervention. Indeed, the freedom to pursue one’s personal desires is in itself a basic human functioning that interventions to attempt to ensure the maximisation of hedonic utility may negatively impact upon.

The restrictive implications of the hedonic approach were, in fact, well recognised by Bentham’s early protégé, John Stuart Mill, who eventually sought to refine utilitarianism. Mill argued that pleasure and pain are not the sole constitutes of utility, as illustrated by his famous quote that it is better to be Socrates dissatisfied than a fool satisfied (assuming that he was equating satisfaction with pleasure). Mill expanded the scope of what utility encompassed, perhaps even more so than Hume. As Isaiah Berlin put it, Mill

became not so much an open heretic from the original utilitarian movement, as a disciple who quietly left the fold, preserving what he thought true or valuable, but feeling bound by none of the rules and principles of the movement. He continued to profess that happiness was the sole end of human existence, but his conception of what contributed to it changed into something very different from that of his mentors, for what he came to value most was … diversity, versatility, fullness of life – the unaccountable leap of individual genius, the spontaneity and uniqueness of a man, a group, a civilisation [5].

In Utilitarianism, Mill maintained that people should be impartial as to their happiness and that of others and consequently willing to sacrifice their happiness for others’ if the circumstances warranted it [6]. As such, Mill still seemingly thought that interpersonal cardinal utility (i.e., a utilities that reflect the intensity of different people’s feelings for goods and that can be compared and thus summed across those people) could be assessed with some degree of accuracy. But in coming to believe that a concern with one’s own feelings of pleasure and pain are an excessively reductive view of what drives, and what ought to drive, human action, while at the same time wishing to retain the consequentialist notion of utility, Mill left himself open to the charge of trying to cram too many considerations into a single index. Berlin suggested that the notion of happiness

is “complex and indefinite” in Mill because he packs into it the many diverse (and, perhaps, not always compatible) ends which men in fact pursue for their own sake … In J. S. Mill’s writing happiness comes to mean something very like “realization of one’s wishes”, whatever they may be. This stretches its meaning to the point of vacuity [5].

Yet while the definition of happiness becomes vacuous if pretty much everything is included within it, this does not mean that it is meaningless to build a policy framework on the notion that people have diverse desires if it is accepted that the pursuit of happiness does not necessarily drive their desires and that policy makers should not be striving to maximise utility (on which more later).

By the end of the nineteenth century, the idea of measuring interpersonal cardinal utility was in retreat, with most economists and philosophers resigning themselves to the view that they were impossible to measure. Therefore, although the general ethos of utilitarianism perhaps had a policy impact in some important areas (e.g., in the criminal justice system, where it might have been more readily recognised that the ‘pleasure and pain’ of the poor ought to count for as much as that experienced by the more privileged), an accurate calculation of interpersonal cardinal utility as a means to guide policy action appeared to be a fool’s errand. However, most economists retained the normative assumption that people ought to want to, and indeed—errors aside—will, maximise some notion of utility. Since they believed that the measurement of cardinal utility was beyond them, they turned to a new analytical framework.

The Ordinal Turn and the Cardinal Sidestep

The birth of modern welfare economics is based not upon cardinal utility but upon ordinal utility. That is to say that all that matters is the direction of preference (i.e., that an individual prefers one thing over another, from which it is implied that the individual gains more utility from the former than from the latter thing in question). The intensity over which the individual prefers the former to the latter, which is key to cardinal utility, plays no part in this analysis. Underpinning welfare economics is the concept of Pareto efficiency, which states that a point of efficiency has been attained if an alternative position cannot make any one person better off without making at least one other person worse off.

While welfare economics follows the tradition of classical utilitarianism in assuming that people will and ought to seek to maximise utility, it does not aim to define, measure, or observe utility. Instead, it infers utility from choices. The theory assumes that as long as people make broadly consistent choices across time and place, it does not matter what they choose. For some, utility could be pursued through pleasure (or even, for a masochist, pain), while others may pursue Humean and Millian pro-social objectives. Once more, however, this is vulnerable to Berlin’s critique of Mill’s theory that if utility can encompass anything, it becomes vacuous. If all that is required from welfare economics is for people to be consistent in their choices, does utility need to be mentioned at all?

The dominant position that welfare economics has held in the field of economics since the beginning of the twentieth century inevitably goes hand in hand with it being highly influential in informing economic policy, both in the past and the present. The ideas that underpin it suggest that the competitive market is a means by which people will maximise their utility, which, according to the principles of welfare economics, is what people ought to do. The competitive market, with adjustments for factors that might interfere with its efficient working (e.g., externalities, information asymmetries, non-excludability in public goods) has been central to the theory and application of welfare economics. Moreover, the policy influence of welfare economics has not been restricted to economic policy; it has also informed social policy. For instance, cost-benefit analysis, a tool that is used in many countries to assess the value for money of publicly financed projects and interventions, is underpinned by welfare economic theory, and over the past thirty years those sympathetic to the assumptions underlying welfare economics have advocated successfully for the introduction of choice and competition within some public sector services, including health care and education, in a number of countries. However, welfare economics does not impact on the private domain of people’s lives in the manner that Benthamite and neo-Benthamite approaches might. For example, hypothetically, if it were discovered that married persons are, on average, happier than the unmarried and that the childless are, on average, happier than parents, then a social planning utilitarian might advocate for policies that incentivise marriage and disincentive pregnancy. Welfare economists would presumably be opposed to such acts of social engineering.

In parallel with the development of welfare economics, a group of mathematicians and economists laid the foundations for expected utility theory [7]. This theory is focused on choice in circumstances of risk (i.e., over decisions where outcomes have a probabilistic chance of occurrence) and offered the potential to resurrect the utilitarian dream of eliciting cardinal utility. If people comply with the theory’s assumptions (or axioms) when making their decisions, then this allows us to use what is known as the certainty equivalence method to measure the cardinal utility of the possible outcomes they face. Given its original use within the field of economics, the certainty equivalence method was introduced with money as the outcome of interest, but it can (in theory, at least) be used over any outcome domain.

The certainty equivalence method is not dissimilar to the methods that are currently being proposed to elicit indicators of happiness by some of the neo-Benthamites. To illustrate how the method works, consider this example: imagine that an individual is presented with a gamble that offers a 50% chance of winning £100 and a 50% chance of winning nothing. The individual must then say what sum of guaranteed money would be, for him, equally preferable to the gamble. If the individual enjoys risk, he will nominate an amount over £50, while if he is averse to risk, he will state an amount below £50. The guaranteed money amount is the certainty equivalence of the gamble. Assume that his certainty equivalence is £45. Because £100 and £0 are the best and worst outcomes that he faces in this scenario, we can normalise those with values of 1 and 0, respectively, on a zero to one utility index. This therefore implies that this individual’s cardinal utility of £45 is equal to 0.5*1 + 0.5*0 = 0.5. Through similar processes, one could elicit the cardinal utility of every money amount between £0 and £100 in this example, and by altering the best and worst outcomes in the gamble one can elicit utility on a zero to one index for larger money amounts, or indeed over different types of outcomes (e.g., health status, literacy rates, etc.).5

Like utilitarianism and welfare economics, expected utility theory assumes that people ought to and, aside from making random errors, will maximise some notion of utility. But like welfare economics, it leaves utility undefined. Given the use of money in the original certainty equivalence method, the focus was seemingly on the relative pleasure experienced by someone from receiving a guaranteed amount compared to gambling for a higher or lower amount. It is true that the purposes for which people would use these money amounts remain unstated, and it is possible that those exposed to the certainty equivalence method may be thinking of using the money for prosocial objectives or for any other aim that requires a monetary contribution (an assumption that is consistent with welfare economics). However, the narrative that follows expected utility is that people are selfish utility maximisers, and thus it seems legitimate to assume that the concept of utility here is broadly consistent with Benthamite hedonic feelings. If so, then the two dominant economic theories developed over the past one hundred years function not only with different forms of utility (ordinal versus cardinal), but also hold different assumptions as to what utility constitutes.

Welfare economics and expected utility theory do, however, share the notion that people should be consistent in their choices. That people are in fact often inconsistent in their choices across different decision contexts gave rise to behavioural economics. Behavioural economists have discovered that people’s behaviours, decisions, and choices are highly context-dependent in ways that cannot be attributed to random error. Nonetheless, most behavioural economists still assume that individuals ought to maximise utility; this is why they perceive the common violations of the assumptions of expected utility theory as biases and errors of judgment. The perspective that people should but often do not make choices that would maximise their utility underpins the justification for forms of paternalism in the burgeoning field of behavioural public policy, with the so-called nudge approach being the thus far dominant manifestation of these developments. The argument underlying nudge policy is that because people make predictable errors in choices due to how their environment is structured and framed (e.g., people may buy more alcohol than they really want if alcohol is placed at salient positions at the end of supermarket aisles) their environments ought to be readjusted so that these predictable errors are minimised (e.g., supermarkets ought to be encouraged to place alcohol in less salient positions).6 However, it may not necessarily be the case that inconsistent choices are tantamount to errors. Rather, the ‘errors’ might be the result of evolved processes in human decision-making that facilitate people in achieving objectives that they personally desire, objectives that may typically be at odds with maximising any notion of utility. Of course, the contemporary neo-Benthamites referred to earlier disagree strongly with this suggestion, and now is perhaps the right place to consider their perspective in a little more detail.

The Neo-Benthamite Perspective

As noted at the beginning of this article, neo-Benthamites, embracing Bentham’s greatest happiness principle as the standard for all laws and all rules of morality, have been active in their attempts to influence policy over both the public sector and the private domains of people’s lives over recent years. For the most part they depart from welfare economics and expected utility theory because of the evidence that demonstrates that people struggle to reliably evaluate both retrospectively and prospectively how much total utility they experience or expect to experience from an episode [8].7 The neo-Benthamites have thus advocated for a variety of alternative methods that are intended to measure utility in the moment, including sliding scales to record the continuous mood of those being assessed, requiring people to record their mood every minute or so during relatively short episodes, and the day reconstruction method, which asks people to record their current mood and their activities periodically during the day.

Hedonic feelings of pleasure and pain initially appeared to form the focus for many of the neo-Benthamites, but perceptions of life-satisfaction currently seem to be viewed as the best measure of utility [9]. Life satisfaction is generally measured through surveys that ask people to respond, on a 0 to 10 scale, to the question, ‘Overall, how satisfied are you with your life nowadays?’, with 0 signalling ‘not at all’ and 10 signalling ‘completely’. These surveys are therefore probably not for the most part measuring hedonic feelings of pleasure and pain. Nor do they correspond neatly with the Aristotelian eudemonic form of happiness that promotes a meaningful and virtuous life, which may correspond to some of Hume’s and Mill’s ideas on what utility constitutes [10]. For instance, I might be perfectly satisfied with my life within the rather narrow context in which I live it nowadays, but if I were really to think about it, I might recognise that there is no special meaning to my life and that it could be significantly more worthwhile (hypothetically speaking). This is not to argue that a satisfied life is any better or worse than a meaningful one; it is merely to point out that these two notions of utility differ from one another.8

Pleasure, satisfaction, and fulfilment are thus different forms of utility, and yet scholars within the field of economics (let alone without this field), as I have tried to explain in this article, have different views on what it ought to mean to live a happy life. To be fair, it should be acknowledged that there are scholars working in the field of happiness studies who recognise the differences between the various notions of utility and who advocate for the use of surveys that try to capture purported measures of each.9 A possible problem with that approach, however, is that we return to the issue of almost everything potentially constituting utility, and the inevitable trade-offs that would have to be made in trying to engage and promote the different components at a population level may suit nobody at an individual level. To illustrate in simple terms, if one of three people each prefers cranberry juice, pineapple juice, and tomato juice in a choice of three fruit drinks, a cranberry-pineapple-tomato cocktail is likely to be less desirable to each of them than their principal preferred flavours. In the domain of utility, the analogy extends to pleasure, satisfaction, and fulfilment, which would present a quandary even in the absence of any claim that desires might often be driven by considerations that transcend these notions.

Despite the apparent shift in emphasis away from hedonic utility, the neo-Benthamites mostly retain the Benthamite emphasis on experienced utility (i.e., that (almost) moment-to-moment experienced utility should be measured and aggregated to derive the total value of an episode). However, the claim that such an episode—which could be a whole life—can be legitimately valued as such is contentious. The philosopher Daniel Hausman, for instance, has written that ‘a good life is not a sum of the net goodness of its moments … The same sum of momentary experiences can add up to a wonderful life or an incoherent and mediocre one, depending on how the experiences are ordered and what overall narrative they sustain’ [14]. Following this line of argument, it is the ‘shape’ of an episode (i.e., how the individual moments in an episode fit together) that gives an episode its meaning and is probably the reason why prospective and retrospective evaluations, the very instruments that the neo-Benthamites distrust, offer a different, and perhaps more accurate, assessment of the impact that an event will have or has had on a person than aggregating the ‘in the moment’ measures of utility.10

Given what has been discussed thus far, what, then, should policy makers be pursuing? Must all laws and all rules of morality really be based on a single consequentialist principle?

The Desired Life

Policies that are informed by any notion of utility maximisation are based on population averages. Even if we assume that utility is all that drives actions, behaviours and choices, it is still the case that for many individuals within the population the policy directions that might be optimal on average will be suboptimal for them. For example, if we return to the earlier hypothetical example of there being evidence that married people are on average happier than the never married and assume further that there is evidence to suggest that they are also on average happier than the divorced, then the utilitarian policy implication is that interventions ought to be introduced that encourage marriage and discourage divorce. Again assuming that happiness, however construed, is the only relevant consideration of all concerned parties, further incentivising marriage and disincentivising divorce as policy strategies may lead to personal choices that compound the unhappiness of a great many people. Prioritising a general outcome objective over the private domain of individual decision-making as a policy strategy has the potential to inflict untold harm, and such tension between the aggregate and the individual appears to be what Friedrich Hayek had in mind when he wrote that ‘The welfare of a people, like the happiness of man, depends on a great many things that can be provided in an infinite variety of combinations. It cannot be adequately expressed as a single end’ [15].

I concur with Hayek on this point, and I would suggest that a third party (i.e., me, you the reader, or a policy maker) cannot possibly discern what each individual wants from life—what drives their decisions and behaviours. People have multifarious legitimate reasons for their actions, both interpersonally and, across time and context, intra-personally, and to assume that any notion of utility maximisation is an appropriate universally applicable normative goal is, as claimed by Sugden, ‘a view from nowhere’ [16].11 As we have seen, over the past 250 years economists and others have not even been able to agree on what utility means, reflecting the fact that different people have differing views on what drives and what ought to guide human actions and behaviour.

However, I would also suggest that the inconsistencies in the definition of utility should not concern us too much, because our normative and descriptive framework ought to be built on something else instead. Mill’s godson, Bertrand Russell, in ruminating on whether utility as pleasure drives people’s choices, wrote that

if what is meant is that, when I desire anything, I desire it because of the pleasure that it will give me, that is usually untrue. When I am hungry I desire food, and so long as my hunger persists food will give me pleasure. But the hunger, which is a desire, comes first; the pleasure is a consequence of the desire. I do not deny that there are occasions when there is a direct desire for pleasure … But the actions thus determined by the direct desire for pleasure are exceptional and unimportant. Everybody’s main activities are determined by desires which are anterior to the calculation of pleasures and pain… Anything whatever may be an object of desire; a masochist may desire his own pain. The masochist, no doubt, derives pleasure from the pain that he has desired, but the pleasure is because of the desire, not vice versa [18].

Thus, to Russell, desires usually come first. Everyone has desires, such as the desire for pleasure, to be publicly useful, to be generally satisfied, or to feel some sense of worth in one’s life, but there could also be the desire to feel nothing or to do nothing, to be solitary, to renounce ambition, to never search for meaning, to always keep one’s promises, to be true to oneself irrespective of the personal and professional consequences, or even to write articles and books that hardly anyone will ever read (among countless other desires). To me, the viewpoint that multifarious desires that differ across people ought to be respected is compelling as it allows people to pursue any of the various types of utility that have been advocated by different scholars throughout the history of economic thought if they so wish, while also allowing people to seek to satisfy any other desire that they might have instead (so long as they are not harming others). Policy makers, at least in the realm of personal lifestyle choices, must to my mind therefore limit themselves to supporting the general conditions that might best help people to lead their lives—to follow their desires—as they themselves see fit.

To loop back close to where I began this article, with David Hume, it is my contention here that the best way for policy makers to facilitate people in the pursuit of their privately held desires is to nurture—or at the very least to try to ensure the non-crowding out of—the cooperative, reciprocating instincts that evolved to serve group and, by extension, individual interests (for a detailed discussion of reciprocity, see [19]). Space constraints in this article limit me to mentioning very briefly just three ways of doing this.

First, there is the power of policy rhetoric. In order to encourage people to act and behave cooperatively and reciprocally, it is important that policy makers give the impression that they genuinely believe that these pro-social tendencies serve individual interests. Messages that proclaim that ‘greed is good’ or that ‘there is no such thing as society’ can have a powerful influence in bringing to the fore aspects of human nature that crowd out those which might better serve people’s long-term self-interest, and it would be wise for political and policy leaders to frame their messages with these considerations in mind.

Second, governments should act to reduce the extreme concentrations of income, wealth, and accompanying power enjoyed by small numbers of people within many of the world’s nations—concentrations that have intensified greatly over the past four decades. If a small number of people are allowed to continue to harvest off for themselves a high and increasingly disproportionate share of the income and wealth that a society creates, then this might serve to disillusion many of the potential benefits of societal cooperation, and it may serve as a signal that it is aspirational to be selfish. In short, these income and wealth concentrations may serve to incentivise behaviours that are ultimately harmful for societies, which might in turn impact negatively upon most individuals in their pursuit of their own privately held desires.

Third and finally, governments should embrace and encourage local control and experimentation in all manner of decision making, rather than serving to guide or manipulate people in the direction of a centrally imposed conception of the general good. Social planning over the private sphere of decision-making—whether utilitarian-informed or otherwise—is not the way in which to value the ‘diversity, versatility, fullness of life’ that Isaiah Berlin, as earlier noted, suggested was a central feature of Mill’s thought.

To conclude, I have argued that we have multifarious desires that differ from person to person and that the fulfilment of these desires results in our own conception of a flourishing life. Policy makers can help to create the conditions by which we might have the best chance of fulfilling our privately held desires, conditions that nurture the intrinsic human tendency to reciprocate with our fellow citizens. This is not necessarily a framework that facilitates the happy life (although it does if all that you desire is some notion of happiness); it is a framework that facilitates the desired life.


1Although, as we shall see, many scholars at various times did not equate the whole of utility with our own individual happiness (or pleasure). 

2It has to be admitted, though, that Hume, great thinker and writer that he was, did not lack entirely the capacity to confuse. For instance, he elsewhere states that pleasure (and the avoidance of pain) is the ultimate end: ‘Ask a man why he uses exercise; he will answer because he desires to keep his health. If you then enquire, why he desires health, he will reply, because sickness is painful. If you push your enquiries farther, and desire a reason why he hates pain, it is impossible he can ever give any. This is the ultimate end and is never referred to any other object. Perhaps to your second question, why he desires health, he may also reply, that it is necessary for the exercise of his calling. If you ask, why he is anxious on that head, he will answer, because he desires to get money. If you demand Why? It is the instrument of pleasure, says he. And beyond this it is an absurdity to ask for a reason’ [1]. However, it is not self-evident that utility, happiness, or pleasure are the ultimate ends for all human beings; they are only the ultimate ends (for you) if you believe them to be so (and even then your beliefs may be mistaken). 

3Embedded within utilitarianism is the assumption that people have a declining marginal utility of income/money/wealth. That is, £100 gives more utility (pleasure) to a person who earns £10,000 per annum than if he earned £50,000 per annum, and gives more utility in the latter case than if he earned £500,000 per annum etc. Therefore, on the general subject of inequality, if it is assumed that everyone has the same marginal utility of income curve, then utility would be maximised by everyone having an equal income. But that is a strong assumption. 

4There is likely to be more support for the consequentialist claim if the individuals differ more markedly in age—for instance, a 15-year-old versus an 85-year-old. 

5There is a theoretically identical alternative method to the certainty equivalence method called the probability equivalence method. With the latter, in an example analogous to the one just given, an individual is informed that he could accept, say, £50 for certain or take a gamble that offers a probability, p, of winning £100 but that also offers a 1-p chance of winning nothing. The individual is then asked for the p required for him to be indifferent between the guaranteed outcome and the gamble. If he states that he would require a 60% chance of a positive payoff from the lottery, then again normalising the utilities of £100 and $0 at 1 and 0, respectively, the utility of £50 for this individual is 0.6*1 + 0.4*0 = 0.6. As intimated already, in health policy, for instance, the outcomes do not have to be money—they can be states of health, with the endpoints defined as full health for a statistical lifetime and immediate death. Indeed, in health economic evaluation, with which the value for money of health interventions and programmes is estimated, the probability equivalence instrument is one of the ways in which units of benefit called quality adjusted life-years, or QALYs, are typically in large part derived. In that context, the probability equivalence method is called the standard gamble and is used in practical applications that serve as inputs into decision-making that literally concern matters of life and death. It is from these types of practices that some of the neo-Benthamites, in trying to develop cardinal indices of what they call ‘wellbeing’, have taken inspiration. 

6Not all of those working in behavioural public policy believe that paternalistic intervention (i.e., to reduce the harms that people impose upon themselves) is the best way forward for this field. Some hold the view that behavioural-informed regulations to address externalities (i.e., to reduce the harms that people impose on others) is a more appropriate path to take. 

7I say ‘for the most part’ because recently, as mentioned in footnote 5, some of the neo-Benthamites do not appear to be averse to using the utility elicitation methods implied from the axioms of expected utility theory. 

8Implicitly in relation to the different definitions of utility, the cognitive psychologist Steven Pinker writes: ‘People who lead happy but not necessarily meaningful lives have all their needs satisfied: they are healthy, have enough money, and feel good a lot of the time. People who lead meaningful lives may enjoy none of these boons. Happy people live in the present; those with meaningful lives have a narrative about their past and a plan for the future. Those with happy but meaningless lives are takers and beneficiaries; those with meaningful but unhappy lives are givers and benefactors. Parents get meaning from their children, but not necessarily happiness. Time spent with friends makes a life happier; time spent with loved ones makes it more meaningful. Stress, worry, arguments, challenges, and struggles make a life unhappier but more meaningful. It’s not that people with meaningful lives masochistically go looking for trouble but that meaning is about expressing rather than satisfying the self: it is enhanced by activities that define the person and build a reputation’ [11]. Of course, Pinker restricts his definition of happiness here to life satisfaction and perhaps a bit of immediate mood hedonism rather than eudemonic happiness (which he terms meaningfulness), but the quote demonstrates how these different notions can clash with each other. 

9See work undertaken by Benjamin et al. and Helliwell [12, 13]. 

10Some economists may maintain that the preferred shape(s) of a utility profile, as a part of utility that might be related to, for example, the sense of fulfilment or meaning that a person gets out of an experience, is just an additional argument that can be included in the utility function. One gets the sense that whenever economists (and others) who adopt this perspective hit a bump in their theoretical road, their stock response is to widen the utility function. But then, going back to Berlin, such widening risks vacuity. 

11Actually, I am of the view that it is not exactly a view from nowhere, even if there is no real empirical support for the proposition. As we have seen just in this article, the concept of utility maximisation has long held a prominent position in the history of intellectual debate, and although Sugden used the term ‘the view from nowhere’ pejoratively, he borrowed the phrase from the philosopher Thomas Nagel, who believed utility maximisation to be an appropriate position to take in moral reasoning [17]. As is clear, however, my sympathies lie with Sugden in that I maintain that the definition of utility is confused and confusing, and that to assume that it drives and ought to drive all human action is inconsistent with what many people do or want to do. 

Competing Interests

The author has no competing interests to declare.


  1. Hume D. 1751. An Enquiry Concerning the Principles of Morals. In: Coventry A, Valls A, (eds.). David Hume on Morals, Politics, and Society. New Haven: Yale University Press; 2018. DOI: 

  2. Hume D. 1777. Essays and Treatises on Several Subjects. Printed as Essays, Moral, Political and Literary. In: Coventry A, Valls A, (eds.). David Hume on Morals, Politics, and Society. New Haven: Yale University Press; 2018. 

  3. Bentham J. The Principles of Morals and Legislation. New York: Prometheus Books; 1781/1988. 

  4. Sen A. Development as Freedom. New York: Random House; 1999. 

  5. Berlin I. Introduction. In: Mill JS, (ed.). On Liberty and Utilitarianism. Oxford: Oxford University Press; 1863/1969. 

  6. Mill JS. Utilitarianism. On Liberty and Utilitarianism. Oxford: Oxford University Press; 1863/1969. p. 111–172. 

  7. Von Neumann J, Morgenstern O. Theory of Games and Economic Behavior. Princeton: Princeton University Press; 1944. 

  8. Kahneman D, Wakker PP, Sarin R. Back to Bentham? Explorations of Expected Utility. The Quarterly Journal of Economics. 1997; 112: 375–405. DOI: 

  9. Frijters P, Clark AE, Krekel C, Layard R. A Happy Choice: Wellbeing as the Goal of Government. Behavioural Public Policy. 2020; 4: 126–165. DOI: 

  10. Aristotle. The Nicomachean Ethics. Oxford: Oxford University Press; 350BC/1998. 

  11. Pinker S. Enlightenment Now: The Case for Reason, Science, Humanism and Progress. London: Allen Lane; 2018. 

  12. Benjamin D, Cooper K, Heffetz O, Kimball M. Self-Reported Wellbeing Indicators Are a Valuable Complement to Traditional Economic Indicators but Are Not Yet Ready to Compete With Them. Behavioural Public Policy. 2020; 4: 198–209. DOI: 

  13. Helliwell JF. Three Questions About Happiness. Behavioural Public Policy. 2020; 4: 177–187. DOI: 

  14. Hausman DM. Valuing Health: Well-Being, Freedom and Suffering. Oxford: Oxford University Press; 2015. DOI: 

  15. Hayek FA. The Road to Serfdom. Abingdon: Routledge Classics; 1944/2001. 

  16. Sugden R. The Community of Advantage: A Behavioural Defence of the Liberal Tradition of Economics. Oxford: Oxford University Press; 2018. 

  17. Nagel T. The View from Nowhere. Oxford: Oxford University Press; 1986. 

  18. Russell B. History of Western Philosophy. London: George Allen and Unwin Ltd; 1946/1996. 

  19. Oliver A. Reciprocity and the Art of Behavioural Public Policy. Cambridge: Cambridge University Press; 2019. DOI: