The paper starts by describing a strong argument in favor of the “selfinterest hypothesis” that we find in economics. This argument argues that any realistic political theory should “economize on virtue”. The present paper criticizes this argument in two ways: first, supposing that people are exclusively motivated by self-interest can have (four) socially bad consequences. Second, the argument that realistic political theories need to economize on virtue can be turned against the self-interest hypothesis; such hypothesis is (in four ways) not realistic enough. The concluding part of the paper then suggests that an appropriate conception of political realism does not support the self-interest hypothesis.
As Martin Hollis convincingly formulates it, a political theory always offers a recipe for a “socializing syrup” that is expected to render a particular human nature apt for social life.
Recipes for the Good Society […] have produced many classic dishes in political theory. […]
[T]he magic formula for the socializing syrup varies with the analysis of human nature. For
instance, if men are essentially greedy egoists in pursuit of riches, fame and honour, then the
syrup will be a blend of repression through fear and reward for cooperation. If men are born free,
equal and good, they need only to be stewed in Enlightened education amid democratic institution.
If men are by nature the sinful children of God, then a conservative chef will distil his brew from
notions like law, authority, tradition, property and patriotism, tinged with distrust of reason. [
Many arguments have been given to justify such a conception of human nature. For
example, it has been argued that it is a descriptively accurate first approximation of
human behavior, or that it enables us to achieve a mathematical representation of
choice [
As a consequence, concerning possible human motivations, a good social planner should base his decisions and predictions on a representation of human nature that “economize on virtue”, that is, that does not rely on the (utopian) expectation that citizens and economic actors will act from some kind of moral motivation or from some interest for the public good.
Compliance with morality, even one that the agents regard as ‘true’, simply cannot be taken for
granted. ‘If men were angels’, the economist is inclined to argue, things might be different; but
in the real world we must determinedly set aside heroic conceptions of human nature and deal
with human behavior as it is, warts and all. To do so commits the economist to a particular
interest in institutions, which, as economists often put it, ‘economize’ on virtue. […][
Such an argument argues that a legislator who hopes that people will act as they are morally required, even when that implies giving up some personal advantages, is committing what we can call a “moralistic fallacy”. As Ingo Pies expresses it concerning firms:
A moral fallacy occurs when we expect firms to act differently without having changed
incentives, rather than to change incentives in a way that makes it imperative (and actually
attractive) for them to act differently. The alternative at issue here can be formally formulated:
change of attitude versus change of conditions. [
A legislator should not wishfully believe that, just because it is morally required not to act in a way that damages the public good, economic actors will be motivated not to act this way. Such a reasoning derives an “is” (i.e. “economic actors do not act against the common good”) from an “ought” (i.e. “economic actors ought not to act against the common good”). Such a derivation is logically illegitimate – just as the converse “naturalistic fallacy”, deriving an “ought” from an “is”, is logically illegitimate. Therefore, governments should never expect agents to behave as they morally should. Rather, they need to expect them to be exclusively motivated by their personal, private advantage. As a consequence, society should be organized in a way that makes individual self-interest serve the public good.
Economics insists on the necessity to make sure that individual objectives are aligned with
collective objectives. [
The underlying idea is: if a society can work and materially prosper without any moral motivation, it is more sustainable than a society that is wholly dependent on people’s dispositions to care about the public good.
The rest of the paper is divided into two parts. The first shows that representing agents as exclusively self-interested can lead to dangerous consequences for society. We will explore four such consequences. We conclude that, given those bad consequences, it cannot just be taken for granted that self-interest is a more realistic hypothesis than, say, universal moral motivation. Instead, this thesis needs to be proved by an argument.
The second part of the paper shows that, besides its dangerous social consequences, the “self-interest hypothesis” fails by its own light – i.e. it is not a very “realistic” (or “not-too-optimistic”) hypothesis regarding human motivations. We will consider four objections of this second kind. We will then conclude that, since the hypothesis fails by its own light, it is very important to look for other plausible motivational basis for a viable society.
Representing citizens and economic actors as motivated solely by self-interest leads to four possible dangerous consequences: (a) moral considerations might simply disappear from private consciousness, (b) agents might become slowly unable to think about the moral dimension of their actions, (c) developing an incentive-based society might be connected with high costs, both in in terms of money and in terms of liberty and (d) moral responsibility for failures to achieve the common good might move from private actors to the State.
The risk that moral values disappear from individual consciousness arises when one is allowed to pay for getting the right to do an immoral action. When a monetary price is attached to an immoral action, it can be interpreted as buying the right to perform an action that was first morally forbidden. Said differently, the action changes from having the property “morally wrong” to having the property “morally permissible” when one pays the monetary penalties attached to it.
The disappearance of moral value has been observed, for example, in a case where a nursery decided to introduce a fine for parents being late at the end of the day.
The fine seems to have undermined the parent’s sense of ethical obligation to avoid
inconveniencing the teachers and led them to think of lateness as just another commodity they
could purchase. [
The existence of a price made the “sense of ethical obligation disappear”. This fact, i.e. that paying is considered as providing a right to do an immoral action, is by itself problematic. There is however an even more serious reason to worry. Indeed, the risks runs that the total number of bad actions increases rather than decreases after the introduction of “moral” penalties– contrary to expectations. Once one can buy a “right to do wrong”, the wrongness dimension of the action disappears from consciousness and people feel comfortable doing bad actions. Consequently, people no longer feel any moral pressure not to do the action, and just start doing it or, for those who already did it, start doing it more frequently. In the case of the nursery, not only did the number of people being late increase, but the time of lateness also increased.4
We could hope that, after having realized the self-defeating effect of monetary incentives, we could come back to the initial state of affairs by suppressing them; if the number of cases has increased by introducing fines, then the number of cases should decrease by suppressing fines again. Unfortunately, evidences show rather that we can expect the number of cases to increase to an even higher point. The fact that taxes have made the moral aspects of action disappear from consciousness explains the fact that, when such actions become free again, their moral dimension nevertheless remains invisible and the number of cases increases again. In other terms, the action had first a moral value, then a price, and, at the end neither a moral value nor a price.
The action had therefore become, from a self-interested point of view, very attractive. As a conclusion, incentive-based policies that aim at directing self-interest toward the public good can be self-defeating and therefore lead to socially bad consequences.
The risk that people lose their capacities for moral reflection comes from a slightly
different aspect. When a State does not expect its citizens to be motivated by something
else than their personal well-being, citizens can legitimately believe that the
responsibility for moral reflection lies completely in the State’s hands. They know the
State has the function of making sure incentives provide the correct direction for
individual actions. They thus come to believe that they can quietly follow their
selfinterest, because the State ensures that it serves the common good. Discharged from the
4 The same results have been observed in situations of the “tragedy of the commons” type; the
introduction of penalties made the situation even worse. [
By claiming exclusiveness for his private interest, the economic Subject, in liberal theory, is
not aware that he damages the interests of other economic subjects. He even knows that it is be
following his interests as a producer or consumer that he contributes to the general prosperity.
[
Economists who analyze the self-interested reasons for cheating and who explore their
implications while ignoring or dismissing “moral” reasons may, perhaps unintentionally, wind
up justifying the practice. If there is sufficiently large expected return, then cheating is rational
from the point of view of an individual concerned only with personal net income. From the
perspective of such materially self-interested individual, the only thing wrong with cheating is
the risk of getting caught. On the assumption that everyone is materially self-interested, those
found evading their taxes are either incautious or unlucky. [
Monitoring individual behavior in order to make self-interest serve the common good can cost to the State more than to let people sometimes act against the common good. Indeed, monitoring economic activities in a way that guarantees that everyone always respects the rules requires an omniscient and omnipotent State, with the capacity to reprimand every wrong action and/or to reward every good action. Besides the high implausibility of such a god-like State, this solution can be very costly both in term of individual liberty and in term of money.
From the point of view of legal norms, we can already doubt that they will be obeyed only
through the use of external constraints. Without a minimum of honesty and decency would a
huge State supervision and tracing apparatus become necessary, to force citizens to comply
with the law. Such a device would be very costly […] and quite probably little effective. [
The problem is that rewards require a lot of money and cannot always be put in place. Indeed, rewards have to be sufficiently high to guarantee that doing an illegal action ceases to be an option for economic agents. This is very demanding. Moreover, many actions cannot be accounted for in terms of “reward” and can only be associated with “penalties”. For example, what would it mean to “reward” a corporation for not having taken advantage of a monopolistic situation or of informational asymmetries? 5 Translated from French.
6 Translated from German.
Would that mean offering to the corporation as much money – or even more – as what it would have gained by violating market rules? But if so, that means society has lost at least as much money as it would have lost had the corporation violated market rules. It becomes, therefore, more costly to reward corporations than to let them violate market rules. That means the only solution in these situations is to use penalties rather than rewards. But penalties have their own deficiencies. Indeed, even though it might happen that punishing bad actions is cheaper than rewarding good actions – this is a highly contingent proposition – punishing would require an involvement of the State in economic activities that might not be realizable and that is incompatible with a “freemarket” economy.
In conclusion, rewards might be less intrusive than penalties, but they are costly and difficult to put into practice. On the other hand, penalties might be less costly that rewards, but they require a very intrusive State control, which appears both implausible and undesirable.
Finally, the “self-interest hypothesis” is associated with the very dangerous consequence of making moral responsibility move from economic actors to political actors. Indeed, under that assumption, the fact that some individual undertakes a bad action can only mean that State incentives were badly organized.
To the degree that economists assume that the only reasons to be sought in explaining the
firm’s behavior are self-interested reasons […] their analysis will tend to excuse or justify the
firm’s behavior and to locate responsibility for the pollution not with the firm but entirely with
the government for failing to set the firm’s incentives properly. [
Economic actors can no longer be held responsible for the socially bad consequences of their self-interested actions. The State is fully responsible. This consequence is more than a logically possible one, but is actually much visible in economic writings. For example, Jean Tirole’s book L’Economie du bien commun is full of passages where some morally bad behavior on the part of economic actors are taken to be nothing more than bad incentive management from on the part of the State.
Relationships between employers and employees are very bad in France…except when they
agree with one another at the expense of the [State] employment insurance. But, as always,
economic agents react to the incentives they face. It is therefore our institutions who are guilty
in this regard, because they encourage concerted manipulations between employers and
employees within the firm. [
In conclusion, we have seen that the economic conception of human nature may lead to some socially undesirable consequences. Such undesirable consequences cannot count, by themselves, as reasons to give up the “self-interest hypothesis”, but they offer reason to consider seriously the question as to whether relying on people’s moral motivations really cannot be a “realistic” or “feasible” option.
7 Translated from French.
Consider now a situation where the “self-interested hypothesis” is not associated with dangerous social consequences. It would still be possible to raise objections against this hypothesis. The objections we are going to study take the “no-optimism” – or “feasibility” – argument in favor of self-interest and turn it against itself. We will cover four objections of this kind. They all support the conclusion that a motivational conception of human nature as exclusively self-interested is, in various respects, still much too optimistic. This open the way for an inquiry into more realistic conceptions of human motivation – where “realistic” means “corresponding to reality” and does not simply mean “pessimistic”.
The first way in which self-interest appears as still a too naive conception of human nature relates to the fact that human beings are driven by other, much anti-social and anti-moral, motivations, such as jealousy, envy, racism, nationalism, sexism, domination, power, etc. By ignoring all motivations that are related to the well-being of others, “self-interest” as depicted in economics ignores not only positive care for the well-being of others, but also negative motivations toward the well-being of others. 8 Hence, if “being realistic” means “taking men as they could be at worst”, then economics should rely on anti-social and anti-moral motivations rather than on a “disinterested motivation”.
There is a ready answer to such worries. It says that self-interest should be precisely conceived as the only possible passion capable of opposing or retaining the most destructive passions of mankind. The role of the social planner is to find a way to restrain the wildest passions and, in this regard, self-interest fares much better than, say, moral or impartial considerations or benevolence. As Dennis C. Rasmussen indicates, the first economists looked at self-interest as the only passion that could prevent more dangerous passions from threatening social life.
Extensive commerce might be incompatible with strict republican virtue, [Hume and
Smith]acknowledged, but they also believed that a focus on material self-interest would help to
replace dangerous and divisive passions such as xenophobia, religious intolerance, and the thirst
for military glory. Moreover, they argued that commercial society helps to promote the
“bourgeois” virtues of reliability, decency, cooperativeness, and so on – moral and social good
that were comparatively lacking in pre-commercial societies. [
The second objection against self-interest as a “realistic” hypothesis is the
wellknown libertarian objection according to which there exist no principled reasons to
limit the scope of the self-interest hypothesis to economic agents without extending it
to political agents. There is no ground to make a motivational distinction between
8 The fact that indifference excludes both benevolent and malevolent motivations is often explicitly
recognized in economics: “I shall usually assume that people’s preferences are free of both benevolence and
malevolence. That is, the fact that one person is enjoying the consumption of a good does not in itself
increase or decrease anyone’s utility” [
9 Translated from French. economics and politics; in both spheres self-interest should be considered as being the exclusive motive of action. Supposing that political actors are preoccupied by the common good just because that is what they ought to be preoccupied with is to commit again the “moralistic fallacy”.
An individual’s nature does not change just because he moves from the private to the public
sector. It is always the same individual with the exact same motivations and concerns, just
doing another job for another employer. There is no reason to believe that by working for the
public sector he will cease to be motivated by his self-interest and act only according to the
public interest, up to the point of neglecting his own interests. [
This objection rightly points to some arbitrary distinction between the moral dispositions of private and public actors – private agents having no moral dispositions and public agents being moved only by moral considerations. The mere fact that public actors ought to care about the common good is taken as a reason to believe that they are indeed motivated by the public good. Yet the “feasibility constraint” requires not picturing people as if they were naturally disposed to act as they ought to, but rather picturing them as self-interested beings. Therefore there is no reason to arbitrary limit self-interest to the market and not to extend it to the public sphere.
The third objection turning the feasibility constraint against the “self-interest hypothesis” of economics is slightly different: it considers that the kind of self-interest described in economics, that is, as indifference to the well-being of others, is a very weak conception of egoism as compared to the real-world egoism of human beings. Indeed, the value people take themselves to possess is not appropriately described by the “self-interested hypothesis”. The real value that human beings believe to have is much greater: they take their own existence to be something of value in the world, or even in the universe.11 Yet this incredibly high importance we take to possess cannot be described by self-interest. As Thomas Nagel formulates it, the importance we believe to have is far bigger than just being able to satisfy our desires. We consider that we have value in ourselves and not just for ourselves.
I believe, as did Kant, that what drives us in the direction of universalizability is the difficulty
each person has in regarding himself as having value only for himself, but not in himself. If
people are not ends in themselves—that is, impersonally valuable—they have a much lower
order of worth. Egoism amounts to a devaluation of oneself, along with everyone else. [
11 “Human beings take themselves generally to be the center of the world.”. [
From the point of view of real-life egoism, we should “distinguish the desirability
of not being tortured [or murdered] from the desirability of its being impermissible to
torture [or murder] us”. [
The last objection is also quite common. It points to the fact that the market works only if market actors do actually possess some basic moral motivation. Indeed, the market is based on one basic moral norm which holds that economic actors ought not to take advantage of situations of market deficiencies.
The regulatory framework governing commercial activities is itself very imperfect. It cannot wholly
correct market deficiencies and thus cannot eliminate the whole set of morally reprehensible practices
that can emerge in market relations. In this case, market actors have an obligation not to take advantage
of non-corrected market deficiencies. [
For example, as noted by Kenneth Arrow regarding informational asymmetries between firms and consumers or workers, the ideal conditions of market efficiency require that consumers and workers are perfectly informed about relevant aspects of their decision. Yet in the real world this condition is not realized – i.e. prices do not necessarily reflect every relevant piece of information. A free market does not on its own provide the whole relevant information for consumers and workers. Therefore it is a duty of firms to realize those efficiency conditions by delivering to consumers and workers every useful piece of information.
There is clearly an obligation to reveal […] truth, even at the expense of profits, for the market
will generally do very poorly in sorting out the facts when the buyers are uninformed. […]
Similarly, the firm knows, by experience, the safety conditions in its plants more than workers
can. Hence, the conditions that the market works are violated, and moral obligation should take
its place. [
In the real world, firms are better informed about their products and about safety conditions on the work place than consumers and workers – and that violates ideal 12 Translated from French. market conditions. Therefore this creates for firms a moral obligation to inform both consumers and workers about their products and working conditions, even at the expense of profit maximization.
The market thus works under the assumption that market actors do not violate the fundamental market norm of “not taking advantage of market deficiencies”. This requires agents on the market to be disposed to respect a moral norm even when that implies giving up some possible “easy profit”. That is, market agents need to possess a fundamental moral motivation – exactly the kind of motivation that economics was supposed to “economize” on. In this sense, economics fails by its own lights, that is, fails to meet the “feasibility constraints” that justified the “self-interest hypothesis”.
The fact that without such basic norm the market in real circumstances – where
ideal conditions are not satisfied – does not lead to collective prosperity is a
wellrecognized fact in economics. [
A possible explanation of why respect for rules is taken for granted might come from a kind of misconception of rules in economics, where legal norms are treated as if they were laws of nature; both are taken as constraints on choice, among other kinds of constraints.14 Such confusion appears quite often, as we can observe in the following quotation, where respect for institutional rules is taken for granted:
Individuals pursuing their own self-interest within an institutional setting of property, contract,
and consent will produce an overall order that, although not in their intention, enhances the
public good. [
In the quotation, property contract and consent seem to operate exactly in the same way as do natural constraints. Yet natural constraints – such as scarcity – and legal constraints – such as property rights – do not operate in the same way at all. Whereas the first cannot – physically – be violated, the second by definition – as norms rather than necessities – can be violated, manipulated or modified.15 And to believe that firms do not try to violate and manipulate rules, or to modify them by political pressure, constitutes a very naive conception of self-interest.
The moral motivation required in this context – when, for example, firms ought to
give up an opportunity for “easy profit” – is not a kind of self-abnegation, a
renunciation of every kind of personal advantage. The motivation here in play is only a
moral limitation of self-interest in the name of social prosperity. Firms are still allowed
13 Some economists explicitly accept that they treat rules as exogenous factors of choice. This can be
justified in some circumstances, but not when a self-interested conception of human nature is justified by the
argument that we should “economize” on virtue. [
15 As Pierre-Yves Néron points it out, the moral obligation of firms that they should not take advantage
of market deficiencies does not have implication only for the way in which firm interact with other agents on
the market, but also have political implications. The main one concerns lobbying practices: “[M]arket actors
have a moral obligation not to create market deficiencies through their political activity (or not to oppose
new regulations that aim at correcting such deficiencies.” [
We believe that the attractiveness of self-interest comes from the idea that the only alternative to a system based on self-interest is a system that leaves no room for individual pretentions – and communism represents this anti-individualist system.16 In other words, self-interest gains his attractiveness from the fear of communism. However, such fear is based on a false dichotomy between self-interest and selfabnegation; if we allow a mid-way between self-interest maximization and no selfinterest at all, that is, if we allow for a morally constrained self-interest, we can start contemplating the possibility of a society based on the moral motivation of its members.
The appeal of the “self-interest hypothesis” comes from the idea that avoiding a misplaced optimism about human nature requires attributing to human beings (morally) bad motivations. That is, being “realistic” requires being pessimistic. Yet we see this conception of political “realism” as mistaken, as long as one has not provided independent support for the idea that moral motivation cannot play the most important role in sustaining a viable society. According to us, supporting a realistic conception of human nature requires to support a conception of motivation that is both true of human nature and that is capable of making society sustainable. It says nothing about whether these motivations are good or bad, noble or corrupted. As Bernard Williams expresses it, considering the motivations that people have in terrible and unusual conditions as being the most representative of human nature is a deeply flawed conception of realism.
If the test of what men are really like is made […] in conditions of great stress,
deprivation, or scarcity (the test that Hobbes, in this picture of the state of nature,
imposed), one can only ask again, why should that be the test? Apart from the unclarity
of its outcome, why it the test even appropriate? Conditions of great stress and
deprivation are not the conditions for observing the typical behaviour of any animal nor
for observing other characteristics of human beings. If someone says that if you want to
see what men are really like, see them after they have been three weeks in a lifeboat, it is
unclear why that is any better a maxim with regard to their motivations than it is with
regard to their physical conditions. [
The work to be done now consists in exploring arguments showing that some good or positive motivations really belong to human nature and are susceptible of making life in society viable.
As a final remark, our presentation only showed two things: that the “self-interest
hypothesis” can have significant social dangerous consequences and that this
hypothesis does not completely satisfy its own criterion of “feasibility constraint”. Yet
the fact that economics might not be able to get rid of moral motivation does not yet
16 “Soviet Union thinkers originally supposed that their system could forgo incentives, which are so
central in a free economy and take their source in the individual’s egoism. They thought socialist enthusiasm
and conscience could substitute for them. This conception was soon confronted with the real data about
human nature, which is far from being guided by some sublime abnegation.” [