Economic Man, Minimal Benevolence, and the Inadequacy of the Preference Satisfaction Theory of Well-Being.

Neil Haddow

The scope of neo-classical economic theory has greatly expanded the second half of the twentieth century. Gary Becker was awarded the Nobel prize for his work applying economics to all areas of life and many other works have followed suit, applying the economic method outside the sphere of economics. [1] The general driving thought behind 'economic imperialism' is that “economics is seen as the science of rational choice, and rational choice is seen as governing human action in all its departments” (Kirzner p. 259). But the model of economic man [2] (or homo economicus) ---man as a self-interested and calculating utility maximizer---has been thought to be either simplistic or false. The views of noted economists provide material for this criticism. For instance, here is Milton Friedman's thought on the 'private' interest of economic man:

[E]very individual serves his own private interest.... The great Saints of history have served their 'private interest' just as the most money grubbing miser has served his interest. The private interest is whatever it is that drives an individual (quoted in Machan (1990) p. 21).

This view is reinforced by F. Y. Edgeworth's claim: “The first principle of economics is that every agent is actuated solely by self-interest” (in Hollis p. 57). Similarly, Gary Becker argues: “The combined assumptions of maximizing behavior, market equilibrium, and stable preferences, used relentlessly..., form the heart of the economic approach as I see it” (in Machan (1990) p. 21). While the economic view seems to make self -interest tautological, since self-interest covers any action that is actually performed, does it really explain behaviour in an interesting way? Is it not plausible to argue that some actions are performed out of altruistic motives that need to be meaningfully distinguished from the broad notion of self-interest above?

Another implication of economic man is that, since subjective preferences are equated with welfare, we can say that trade between uncoerced persons is mutually beneficial. The idea of mutual advantage has been used to justify free trade between individuals. Milton Friedman states: “...both parties to an economic transaction benefit from it, provided the transaction is bi-laterally voluntary and informed “ (1962, p. 13; emphasis in original). A person receives an economic benefit if he is uncoerced, gets what he was promised, and acts with as much information as he could. Here is political philosopher David Conway's way of putting the point: “For the essence of exchange is that, where uncoerced, both parties are better off for having engaged in it” (in Machan (1990) p. 10). Yet, it seems easy to generate counterfactuals that throw this principle into dispute. As Tibor Machan argues, if a trades his ounce of crack for b's ounce of cocaine, is it not more plausible to claim that neither have benefitted? But economists do have a defense of the idea of mutual benefit; as we will see, to argue for economic man is to involve oneself with a certain normative notion of minimal benevolence–viz., that preference satisfaction is considered to be identical with welfare or well-being.

Yet, in spite of the seeming implausibility of self-interest, the model of man as a self-interested utility maximizer has a certain explanatory appeal. In his essay “In Praise of Commercial Fame”, economist Tyler Cowen argues that celebrities and fame serve as economic focal points in that they function to lower the costs of trading and communication. He notes that commercial fame has been much maligned with economic arguments claiming that commercial culture, in its demand of monetary profit, leads to pandering to the lowest common denominator–viz. that commercial culture attempts to use popularity to sell products to the hoi polloi. But there is no reason for thinking that these athletes and movie stars somehow make better the products they sell, or even know anything about the goods they endorse. So on the face of it, the use of celebrity endorsement seems a fallacious appeal to authority. But Cowen argues that, while Michael Jordan or Tiger Woods are not experts on most of the goods they endorse, celebrity endorsements can be trusted since it is in the celebrity's self-interest to know the quality of what he's endorsing. He argues,

For example, Tiger Woods is too busy playing golf to learn much about the most of the products he endorses. But he (and his agent) have every incentive to maintain his reputation by projecting the right image and not disappointing his fans. Thus, Woods' agent makes sure that he endorses only 'the right' products. Fans thus know they can expect a product of a certain kind when they see Woods' endorsement. The company, Woods, and the fans are all better off, as a result of this voluntary exchange (Cowen p. 2).

As we saw above, Cowen's conclusion regarding mutual benefit from trade does not follow; but his point regarding celebrity endorsements providing information to consumers is intriguing. Consumers, he argues, can trust the endorsements of celebrities since it is in the self-interest of the celebrity to hire people to inform him of the quality of the goods he endorses. Focusing on self-interest, economists can tell us an unknown function of celebrity endorsement: that celebrities, to keep their good reputations, will attempt to seek out the best goods they can to endorse. Contrary to many theorists, economists explain norm-following as a result not of good will, or a moral faculty, but as the result of the agent's search for material gain (which will be explicated later).

Gordon Tullock and Richard McKenzie argue that while the economic perspective might seem unrealistic, or not capture 'the fullness of human experience', its virtue is that its predictions are more accurate than can be obtained by other means (p. 7). [3] Since the driving force of the market is economic man we will examine the main components of his character: rationality, utility maximization, and preference, as well as the basic principle of normative economics, minimal benevolence. Economic man is meant not only as an explanation of why people act to achieve what is in their interest, but also attempts to provide an explanation of well-being in terms of the achievement of their interest. To illustrate the importance of this method to economists and philosophers [4] I will examine two arguments based upon the idea of economic man: one is von Mises' argument that the free market best satisfies the desires of agents; the other argument is Richard Arneson's justification of an egalitarian society based upon a preference satisfaction view of well-being. I will argue that these arguments fail because the idealized version of well-being as preference satisfaction requires a substantive notion of the good in order to make sense of the claim that a certain course of action is really in a person's interest. A substantive idea of the good is cognitivist in that it allows us to say that certain normative pronouncements are true or false, and not merely the expression of personal preferences. So, the attempt to guide public policy or explain well-being by the standard of what constitutes well-being or welfare for economic man is destined to fail.

Economic Man

Economic man is a model that attempts to explain how humans behave when confronted with different courses of action in relation to their beliefs and desires. Economics, then, is to a certain extent normative: it tells us how agents will act under such and such circumstances. For instance, given that people desire to get all that they can (acquire the bundles of goods for which the expected value is the greatest), they will buy more at lower prices and less at higher prices. This law of demand is considered to be one of the strongest laws a social science can have. An economist claimed if a test were ever to show that the opposite happened---demand increased when prices increased---that there would be something wrong with empirical testing (in Tullock and McKenzie p. 18). But the job here is not to assess the empirical status of such laws, but to explain how economic man is generally construed. As noted, economic man is partly a normative claim about how persons rationally, or self-interestedly choose, not just how they do choose. But the economic view restricts normativity to assessing means-ends relations, and not the propriety of the ends themselves. Economists take a person's preferences, which arise from a person's beliefs and desires, to be given data. The data themselves –the reasons or desires for preferring one thing over another– cannot be said to be rational (or good), but only the choices that are based on them. Choices are said to be rational when they are determined by a rational set of beliefs and preferences; choices are irrational when they are determined by an inconsistent set of beliefs and preferences (Hausman & McPherson p. 27). There are two basic components of rationality–transitivity and completeness–that help economists to determine when sets of beliefs and desires, hence preferences and choices, are rational.

Transitivity, or consistency, states that in order for an agent to be rational his set of preferences must be ranked in a consistent manner. For example, if agent R prefers x to y, and y to z, then R must prefer x to z. To say that R could have the ranking he does above and also prefer z to x would be to admit an inconsistency in his preferences. Hence, the agent in preferring z to x would be irrational. Hausman and McPherson (p. 28) claim that to admit an intransitive preference ranking could leave an agent open to manipulation. For instance, the 'money pump' argument holds that if an agent, Q, has intransitive preferences (prefers x to y and y to z, but z to x), already has some of y, and is willing to pay a penny to trade what she has for something she prefers, that she will pay a penny to trade y to get x, another penny to trade x to get z, another penny to trade z to get y, another penny to trade y to get x, etc. This could go on forever, unless Q realizes that her preferences are intransitive and allow her to be manipulated.

Another condition of rationality for economic man is that of completeness. This requirement states that an agent must have all of his preferences ranked. So, whenever confronted with x and y, the agent will always be able to choose since he has already ranked these two items. (Of course, the agent could be indifferent, and have ranked the two items equally; in such a case he might just flip a coin to decide which item he will choose). If an agent does not have a complete preference ranking then he may end up having an intransitive ranking; hence, to the extent that his choices are unranked and end up being inconsistent is the degree to which he can be said to be irrational. [5]

The view of rationality outlined here is an instrumental one, since it is only concerned with assessing the relation between means and ends, and not whether certain things are irrational apart from this means -end relationship. There are a couple of objections to instrumental rationality that should be noted. The first is the claim that there are irrational preferences apart from any consideration of consistency. Aristotle, Plato, Thomas Nagel and Derek Parfit have held this view. The latter philosopher argues that it is irrational to desire something that is in no respect worth desiring. He imagines the example of a person who is 'future Tuesday indifferent' (cited in Hausman and McPherson p. 29). This agent does not care what happens on future Tuesdays, therefore he is indifferent about anything that might happen on this day. When presented with a greater and a lesser pain, he will choose the greater pain on Tuesday and the lesser pain on Wednesday since the greater pain will not matter to him if it's on a Tuesday. But Parfit argues that this is an irrational preference since the fact that the greater pain will be on a Tuesday is no reason for preferring it. Since to prefer the worse of two pains for no reason is irrational, there seem to be preferences that are irrational a part from their relation in a set.

Another objection to the idea of transitivity is given by John Broome. He argues that the transitivity condition of irrationality either turns out to be an empty condition of rationality, or is in need of a substantive theory of preference. He argues that an agent Q might have an intransitive preference ranking, preferring x to y, y to z, and z to x, but still argue that she is rational. Q could charge that 'z compared with y' is a different thing altogether than when 'z is compared to x'. The first comparison could be called z(1) and the second z(2). If this can be pulled off, then Q could have a fully ordered preference ranking: z(1), x, y, z(2). Broome argues that this take on the transitivity of preference rankings can always be given, so transitivity turns out to be an empty condition of rationality. The defender of transitivity could argue that it is irrational to prefer z(2) (z when compared to x) to z(1) (z when compared to y), but this would require a substantive element be added to rationality that the instrumental theory lacks. For now we will ignore these problems and turn our focus on the normativity of economic man.

Completeness and transitivity establish a weak ordering of any finite set of alternatives (Hausman and McPherson p. 28). An economist can assign numbers that correspond to the agent's ranking. This is called an ordinal utility function, since the numbers that are assigned represent the order of the preferences, and not a cardinal measure of the value of the goods (or preferences). When economists talk about the agent maximizing his utility, there is no separation between utility and the agent's esoteric ranking (Hausman and McPherson p. 29). The ordinal measure of preferences leaves the value of the rankings a purely subjective matter based on the agent's beliefs and desires. A particular ranking of goods is based on the individual's own desires and beliefs and can not be compared to any other agent's ranking. This latter claim is a denial of interpersonal utility comparisons; it is controversial, but we will take it as true for argument's sake.

It should be noted that the economic man is a normative notion for several reasons. Firstly, some economists, such as Milton Friedman, hold that economic man is self-interested. This claim often seems so broad that it is descriptive and vacuous because it covers any and all behaviour, no matter how altruistic; in order for it to be a normative notion we would require some substantive account of what really is in a person's interest (see Machan (1990) for the claim of vacuity). Hausman and McPherson argue that if economists say that economic man is self-interested, they also require some substantive notion of what really is in a person's self-interest (p. 29). A minimal, normative notion would be that agents act to maximize their material well being. This principle embraced by most economists can be called the 'Axiom of modest Greed' (McCloskey p. 105). Hence, agents will choose actions that they believe make them wealthier; agents will tend always to pick up $100 dollar bills lying on the sidewalk. This allows us to say that sometimes people act in ways that do not maximize their utility, say if they do not pick up $100 bills. Some standard of an agent's self -interest must be kept so that we can say they at least believe that certain actions will make them better or worse off.

Also, the view of choice called 'revealed preference' should be avoided because it leads to a denial of the normativity of economic man. Revealed preference is an attempt to do away with any notion of connecting an agent's psychological state (beliefs and desires) with his choices. Whenever an agent acts he can on this view be said to have revealed his preference. Hence, we do not need to get inside his head and to show that his preferences are consistent with his choices–we do not even need to say that he has preferences. But Hausman and McPherson argue that the revealed preference theory is wrong to equate preference and choice; the two must be kept separate so that preferences can explain choices (p. 28). There is always the possibility that agents will choose actions that do not maximally satisfy their preferences; as Sen argues, we could hardly call a person rational who consistently chose actions that did not conform to his preference ranking (in Stewart) . Such a person would seem rational on the revealed preference theory, since he always acted consistently, even though his actions consistently thwarted his preferences. So, we have seen the importance of at least a weak notion of normativity –of a standard for deciding what is to count as rational, or self-interested.

The components of economic man come together to create a person who has complete and consistent preferences, and who is interested in getting for himself as much as he can for as little possible. Whether one wishes to call him selfish is not important. What is true is that economic man tries to get the most preferred goods for as little as he can. Such a claim requires that economics, in its concern with rationality, must also be concerned with how persons choose and reason. Therefore, economics is not just about the causes and consequences of the actions of agents but also the reasons for them (Hausman and McPherson p. 38). The appeal of explaining action using economic man is similar to the appeal of folk psychological explanations of human action. In both theories the beliefs and desires of the agent are said to 'conspire' to cause the actions the agent undertakes. To explain a certain action, one may cite either the agent's beliefs or his desires, or both. Often it will be a hard task to find the causes of an action because there will be numerous desires and beliefs that gave rise to it; but following Donald Davidson we can say that the reason for an action, the desire(s) or belief(s) most responsible for the action, can be said to be the cause of the action. In economics, we can say that an agent's desires and beliefs which make up his preferences cause his actions. This requirement that beliefs and desires be connected to action, and that certain reasons are the cause of his choice, allows for an evaluation of the agent's action as rational or irrational–of whether an agent's actions satisfy his preferences or not.

Hausman and McPherson argue that normative economics rests on three claims: some notion of self-interest; complete knowledge; and minimal benevolence. The first claim is simply an extension of the one noted above that agents will always attempt to choose what they prefer more over that which they prefer less. Put in a moralistic sense, people will attempt to choose what they believe to be better for themselves over what they believe to be worse (p. 42). The second assumption is that since agents have complete knowledge of their preferences then that person is the best judge of what will make him better off–or whether his preferences have been satisfied. The second assumption requires that preference satisfaction is tantamount to being made better off –or that welfare is equivalent to preference satisfaction. This could be stated as follows:

Preference-satisfaction theory of good. One thing A is better for a person than another thing B if and only if the person prefers A to B (Broome p. 3)

But, on the face of it, this claim is false. Agents often act on false information and make mistakes. As we saw above, we need some notion of what is really in an agent's interest. So, following Broome we will settle on the following definition:

Ideal-preference-satisfaction theory of good. One thing A is better for a person than another B if and only if the person would prefer A to B were she in ideal conditions (Broome p. 4).

This definition contains the idea that agents have their preferences satisfied when certain ideal conditions are satisfied: such as being well-informed and rational. This allows us to say that agents are well-off when they get what they really want. Hausman and McPherson note that many economists make the identification between preference satisfaction and well-being without recognizing that they are defending a controversial moral theory (p. 42).

Yet, once one has identified the satisfaction of preferences with welfare, all that is needed for a minimal moral theory from normative economics is the principle of minimal benevolence (p. 43). This principle states that, ceteris paribus, it is a good thing that people are better off. Hausman and McPherson say that this principle could be stated as a tautology: in general, it's good if there's more good (p. 43). While this principle seems innocuous, it is good to keep in mind that well-being is equated with preference satisfaction. Many economists argue, then, that it is a good thing if people's preferences are satisfied; and much of the debate in modern political economy is what sort of economy will best satisfy people's preferences. For instance, Pareto optimal situations–where it is possible to make someone better off without affecting anyone else adversely–are to be preferred. But 'better off' in this situation is a measure of the satisfaction of the preferences of agents. [6] So, the rationality assumption allows us to judge whether an agent is correctly satisfying his preferences, and the principle of minimal benevolence allows us to make some claims about the morality of people's actions–namely, that they are well-off if their preferences are satisfied, or a situation is good if the agents in question have their preferences satisfied.

Economic Man in Action.

It was briefly noted above the claim that when two persons trade this trade makes both agents well-off is not self-evident. This view seems open to numerous counterexamples, in which two agents do not make each other well-off. People do not always choose what is best for them, nor do they always spend their money prudently. But it seems possible that the adherent of mutual benefit can appeal to the idea of minimal benevolence to claim that allowing people to act in their own self-interest makes them better off. But this claim requires that one have a more substantive theory about why allowing adults to make up their own minds is a good thing. The principle of minimal benevolence rests only upon the idea that preference satisfaction can be equated with welfare and individual well-being. But this is too weak to support the moral claim that it is a good thing to allow people to live their lives as they see fit. This latter argument requires one to hold that people ought to be allowed to make mistakes since there is something about interfering with a person's autonomy, even when we coerce him to do what's right, that destroys the moral status of his actions.

One extension of the idea that there is something important about satisfying the preferences of individuals is implicit in Ludwig von Mises' argument against the possibility of a socialist economy efficiently satisfying consumer preferences. This argument was first proposed by von Mises in 1920 in his paper “Die Wirtschaftsrechnung im Sozialistischen Gemeinwesen”, and was later expanded into a full length book, published in English as Socialism . Often socialism was opposed because of an incentive problem: this was the claim that because socialism was to provide all of the goods any one could ever need, no one will have an incentive to work hard (Rothbard p. 409). But von Mises' argument was different. He granted socialists the premise that people would enthusiastically do what the planners asked of them but also argued that no planned economy would never work. Here is a summation of the argument by Don Lavoie:

The problem [the economic calculation argument] raises for the economics of socialism can e described briefly as the inability of any system of common ownership of the means of production to generate and disseminate the scattered and largely tacit knowledge, including knowledge of the relative scarcities of various consumers and producers goods, upon which advanced technological production depends. The argument claims that separate owners actively contending with one another for money profits are able to (undeliberately) impart knowledge to the system of relative prices, and in turn to orient their own actions by reference to their 'economic calculations' in terms of these prices, in such a way as to enable the millions of independent decision makers to coordinate their actions with one another. Without these prices to serve as 'aids to the mind,' as Mises called them, the planner would not know how to organize his or her commonly owned means of production with anything close to the degree of efficiency attained spontaneously by the rivalrous workings of the market process. Thus Mises's argument is that there is a practical problem involving the use of knowledge facing any society which attempts to deliberately (and thus non-rivalrously) plan its economic order (Lavoie pp. 483-84).

Mises argued that without the prices and other information that a free market provides, no centrally planned economy can succeed in efficiently providing consumers with goods [7] . The nonmarket economy which lacks a price system will not be able to accurately measure profits and loss, hence it will not be able to direct the production of goods. In his work on economic calculation, Trygve Hoff argues that the question of efficiency in economic calculation is whether the resources that are employed by a socialist economy in producing industries could have been better employed elsewhere; the test of whether economic activity is rational is if “[e]ach factor of production [is] employed so as to give the greatest return according to the ends” (Hoff p. 295). But socialism fails on this account because it does not have the needed knowledge of ends, or preferences, and so it fails to satisfy those ends and is irrational. This argument was vigorously opposed by socialist writers, such as Oskar Lange, but is generally accepted today, after the collapse of Eastern Europe's planned economies, as valid. [8] Another point of note is that Mises' argument has often been incorrectly thought to be the claim that a planned economy is only really, really hard because no one can solve the equations necessary for distributing goods. If we could only get a computer big enough we could find a way to makes a planned economy work. Mises' claim is that a planned economy is impossible because the needed knowledge to form any equations does not exist.

The calculation argument is thought to show that socialism and other planned economies are undesirable things. Mises thought that economics was a purely wertfrei science which did not offer ethical pronouncements about the goodness or badness of policies; but as Murray Rothbard points out, Mises implicitly assumed that the failure of socialism to do what it said it would was 'bad' (Rothbard p. 90). To make this argument one needs to assume something like the identification of well-being and preference satisfaction. If it is a good thing that preferences are satisfied in an efficient manner then it will be a bad thing to have a social arrangement which has a hard time producing the goods. But this argument assumes something that socialists would deny–namely, that the satisfaction of preferences is a desirable thing. As E. J. Mishan argues, the calculation argument:

Would be more compelling... if the declared aim of a Communist regime were that of stimulating the free market in order to produce much the same assortment of goods. We should bear in mind, however, that the economic objectives of a Communist government include that of deliberately reducing the amounts of consumer goods which would have been produced in a market economy so as to release resources for a more rapid build-up of basic industries (in Machan (1990) p. 120).

Against the claim that preference satisfaction is good, it seems open to the socialist to argue that while his system may not be as efficient as capitalism it provides a distribution that (according to his egalitarian notion of justice) is just. Socialists could argue that von Mises' argument for the efficient market coordination of activities-–of the satisfaction of preferences---rests upon the existence of private property rights (cf. Kirzner p. 138); but why should this initial starting point be granted by Mises' opponents? The initial starting point will help one assess whether a certain activity has been successfully completed. Different systems of rights will lead to different conceptions of coordination. While these different starting points may be presupposed by economic man, they certainly need a prior, moral justification in order for the whole claim regarding coordination to get off the ground.

The belief that there is a calculation problem “assumes, without justification, the importance of coming at least reasonably close to the satisfaction of the aggregate demand of those individuals who comprise the economic system” (Machan (1990) p. 119). But there is a justification of sorts, since in normative economics preference satisfaction can be equated with well-being. Since the market system best satisfies individual's desires, then it is the rational or preferred system. But the theory of minimal benevolence is not sufficient because there are questions about rationality, self-interest and formal views of well-being.

Economic man acts because he has reasons for acting. His reasons for his preferences are his beliefs and desires. In order to act rationally economic man must consistently connect his beliefs and desires with his actions, and he must choose his most preferred course of action. As we saw above, the instrumental theory of reason might be in need of substantive notions about why certain courses of action are rational and others not. Explanations in terms of reasons raise questions about whether the reasons are good ones. On the minimal view of well-being the agent is rational if he acts to satisfy his preferences. But if one sees an agent's actions as irrational, or not sufficient to explain his behaviour, then one requires further answers about the agent's reasons (Hausman and McPherson p. 46).

Often a non-sufficient reason for motivation could be the very motivational concern of economic man: material self-interest. Following Davidson, we can say that the reason that best explains the agent's action can be said to be the cause of the action. So, does the model of economic man acting out of self-interest seem to be the best explanation of action. There may be many reasons why Tiger Woods or Michael Jordan sell quality goods besides the idea of material benefit cited by economists. They may desire to do the right thing out of a feeling of duty to God for all we know. So, while it is certainly in their material interest to find good quality products to sell, we do not necessarily know if there are not other overriding---perhaps moral---reasons that best explain their behaviour. We can conclude that the claim economic man always acts out of a desire for material self-interest is false.

Finally, the preference satisfaction view of well-being seems implausible because of its formality. Theories of well-being can be classified as either formal or substantive. Formal theories tell us how one finds out what is good and substantive ones tell us what is good (Hausman and McPherson p. 72). So, to say that welfare or well-being is preference satisfaction is to offer a formal theory because it tells us not what is good but how we can find out—viz. by looking at what people prefer. The notion of economic man only contains a formal notion of well-being as preference satisfaction. Formal theories appeal to economists because they require less philosophical commitments such as a controversial substantive theory of the good (Hausman and McPherson p. 72). Therefore, formal theories seem to avoid paternalism, in that they do not tell people what is good (beyond their own preferences) and leave them to decide their fate on their own. But while this theory might seem open-minded and allow for agent autonomy, it often grants a validity to actions that we should not call moral, try to satisfy, or even prefer.

Another argument based upon the idea of preference satisfaction comes from the other side of the political spectrum; while von Mises argues in favour of the free market, Richard Arneson claims that welfare-egalitarianism, which will eliminate inequalities in welfare not due to personal choices, can be based on the idealized version of preference satisfaction. Arneson argues that all agents deserve an equal opportunity for welfare. Welfare is determined by the satisfaction of an agent's considered preferences. So, the amount of welfare a person is entitled to must be determined somehow by the individual's idealized preferences. As Arneson argues: “Anything whatsoever could be valuable for an individual provided he would come to value that thing after fully informed ideally extended deliberation” (in Schaller p. 303). But there seem to be two problems here.

The first problem is based on the fact that economic man always tries to maximize his utility. His desire for a certain good, such as money, could be insatiable. This leads to the problem that “insofar as [agents] have insatiable preferences, it is not possible to determine whether they have been satisfied to the same extent and so it is not possible to judge whether they enjoy equal welfare” (Schaller p. 303). One might want to argue that insatiable desires are irrational, and that an agent should avoid such 'materialism'. But to go in this direction is to go beyond the confines of economic man by adding a substantive notion of rationality—namely, that people are irrational to the extent that they desire more than they should.

The second problem is that we require a notion of the good in order to judge whether agents are enjoying equal welfare. We must be able to compare agents by specifying what kinds and amounts of things are good. In order to know what is good we require some sort of substantive account of good. There is no shortage of such theories; and most of them will conflict with some of our preferences. But for now it seems sufficient to point out that there are certain commonsense examples of actions that seem good and bad. For instance, actions that lead to our health are generally good and ones that are not conducive to health are bad. The defender of the preference satisfaction view of well-being would be hard pressed to claim that actions, such as inveterate drug use, lead to an agent's well-being. Using the theory of idealized preferences, the most we could say would be that agents believed that they were bettering themselves by satisfying their preference for drugs. But on our commonsense view that a morality must not allow habitual actions that interfere with our health, we can see that certain preferences do not lead to well-being.

So, a justification of the free market must provide a moral justification of a system in which outcomes might not be beneficial. [9] Only if we add a substantive notion of the good or well-being to idealized conditions would the claim that an agent was well-off when his preferences were satisfied work. That the market is satisfactory or harmonious according to the model of economic man will not convince those who have a different conception of justice and man's nature. Thomas Nagel (1991), for instance, argues that from an impartial viewpoint a rational agent will see that the inequalities in the capitalist system require a greater sacrifice from the poor than would a more just distribution cost the interests of the rich. Socialists, et al., have a different view of justice, a person's responsibilities, and even the degree of autonomy agents have in the marketplace. Many would disagree with the claim that agents have endogenous preferences that are unaffected by advertising. The so-called left-wing view of humans and society usually emphasizes the existence of social pressures that coerce people into buying what they do not really need. Of course economists do not claim to know what agents really need, since ends are treated as given, unanalyzable data; but in order to challenge paternalist views one must engage in substantive philosophical inquiry, since such inquiry will give us the ability to judge the moral truthfulness of people's preferences. A more credible view of man than that of homo economicus is needed.

Concluding Uneconomic Postscript

We have seen that economic man is in part a normative theory about what sorts of actions qualify as rational or leading to well-being. Given that rational agents always desire to get more of what they prefer, agents are rational and well-off to the extent that they efficiently satisfy their desires. Economic man is rational when his preferences are complete; and he is well-off when he chooses rationally and his preferences have been satisfied. Carried into realms other than economics, the model of economic man implies that agents who freely enter trades with one another are always better off. Also, given the presumption that men maximize their utility, the political and economic system that leads to the satisfaction of preferences in the most efficient manner is rational and leads to the well-being of its citizens.

The state of the market economy in which demands of agents have been efficiently satisfied is called an equilibrium state. In this state of an efficiently functioning economy, “the price system coordinates prices, including wage rates and the prices of other productive factors, so that there is never any shortage or unsold surplus” (Rothbard p. 135) [10] . In early economic theory the equilibrium state was thought to be due to divine providence, since it was not clear how disparate individuals could act separately and produce ends that benefitted everyone. But with Adam Smith there was a change in the director of the market: the divine hand of God became the invisible hand, in which the harmonious coordination of the acts of many different agents, each acting in his own self-interest, resulted in general utility. For Smith the market outcome may or may not be just; he believed that there was a difference between the money prices of goods and the actual labour value of goods. The market was good to the extent that prices reflected the labour or real price of the goods (Myers p. 118). But the modern interpretation, while rejecting the idea that there is any such thing as the labour value of a good, is still that equilibrium values have some special significance beyond simply being one of many possible outcomes (Clark p. 167). Mises' argument assumes that market-generated equilibrium states are preferable over other states that are a result of intervention. But in order to argue for this conclusion, a reliance on economic man is not sufficient. He must be supplemented with a more substantive nature in order for us to know more about what really is in his self-interest [11] .


Broome, J. (1999) Ethics From Economics. Cambridge: Cambridge University Press.

Caldwell, B. (1985) Beyond Positivism. London: George Allen and Unwin.

Clark, M. (1992) Economic Theory and Natural Philosophy. Cheltanham, U.K.: Edward Elgar.

Cowen, T. (2000) “A defense of Commercial Culture” online at

Davidson, D. (1968) Actions, Reasons and Causes” in (ed.) A. R. White The Philosophy of Action. Oxford: Oxford University Press.

Friedman, M. (1962) Capitalism and Freedom . Chicago: University of Chicago Press.

Haddow, N. (2000) “Is There a Problem with Economic Subjectivism?” Philosophical Writings. No. 13, Spring. pp. 29-40.

Hausman, D. and McPherson. (1996) Economic Analysis and Moral Philosophy. Cambridge: Cambridge University Press.

Herbert, A. (1978) The Right and Wrong of Compulsion by the State. Indianapolis: Liberty Fund Press.

Hoff, T. (1981) Economic Calculation is a Socialist Society. Indianapolis: Liberty Fund Press. Hollis, M. (1995) The Philosophy of Social Science. Cambridge: Cambridge University Press.

Kirzner, I. (2000) The Driving Force of the Market. New York: Routledge.

Lavoie, D. (1992) “Between Historicism and Formalism” in (ed.) J. L. Auspitz Praxiologies and the Philosophy of Economics. New Brunswick: Transaction.

Machan, T. (1990) Capitalism and Individualism . Hertfordshire: Harvester Wheatsheaf.

-----------. (1995) “Reason in Economics vs. Ethics”. International Journal of Social Economics. vol. 22, no. 7. pp. 19-37.

McCloskey, D. (1996) The Vices of Economists–The Virtues of the Bourgeoisie. Amsterdam: Amsterdam University Press.

Morse, J. (1997) “Who is Rational Economic Man?” Social Philosophy and Policy. pp. 179-206.

Myers, M. (1983) The Soul of Modern Economic Man. Chicago: University of Chicago Press.

Nagel, T. (1991) Equality and Partiality. New York: Oxford University Press.

Rothbard, M. (1997) The Logic of Action One. Cheltanham, U.K.: Edward Elgar.

Schaller, W. (2000) “Why Preference-Satisfaction Cannot Ground an Egalitarian Theory of Justice” Journal Of Social Philosophy. Vol. 31, No. 3. pp. 294-306.

Stewart, H. (1995) “A Critique of Instrumental Reason in Economics” Economics and Philosophy. Vol. 11, pp. 57-84.

Tullock, G. and R. McKenzie. (1978) The New World of Economics. Homewood, Illinois: Irwin.

von Mises, L. (1978) Socialism. Indianapolis: Liberty Fund Press.



[Note 1] I take the economic sphere to be concerned mainly with the demand and supply for market goods; also we could say that economics is concerned with the monetary cost and value of goods, in contrast with other sorts of costs. But it is arguable–as McKenzie and Tullock show–that the economic notion of cost, in terms of foregone opportunities, fits many other areas, such as crime, politics, and pollution.

[Note 2] I call economic man a model because 'he' is precisely that: an abstraction or ideal type that helps us to understand the behaviour of actual persons. So, to argue that economic man is unrealistic is somewhat beside the point; in a sense all scientific theories contain abstractions that are not realistic, but which help us to predict behaviour. McKenzie and Tullock argue that this ability to predict better than any other method or model is the test of whether economic man and economic rationality is valid (pp. 6-7). Whether economic man and economic rationality can be tested is another story; see Caldwell chapter 6 for a discussion of whether economic man and rationality is testable.

[Note 3] In the context of social science debates, I will be taking the side of those who argue for the explanation of action, rather than the prediction of it. This isn't to say that the two are mutually exclusive. But they may not always co-exist as Carl Hempel thought. See Caldwell for discussion of Hempel's view of the symmetry between explanation and prediction and also Scriven's criticism of that view.

[Note 4] The economic view of man and the idea of minimal benevolence seems to be accepted by many positivistic philosophers, such as Jan Lester, and also as normative notion in contractarian ethics. Lester and the contractarian school in general attempt to look at various (Pareto-optimal) states in which the greatest number of agents have the greatest number of their preferences satisfied–without coming up with a substantive notion of the good.

[Note 5] One problem that arises here is that real agents often change their preferences. Once an agent changes his mind about what he desires or believes he may end up having conflicting preferences. But does changing one's preferences mean that he is irrational? Most people would not think so.

[Note 6] Of course it's questionable whether a Pareto optimal situation will ever be implemented if everyone's preferences need be taken into account. For instance if there is only one person who objects to a certain state of affairs because it will affect him adversely then that state is not Pareto optimal. But what if this person is a crank? There is always someone who will object to any state of affairs–an anarchist, for example, will always object to an ordering in which the state still exists. But to criticize his preferences we would need to go beyond the idea that well-being is just preference satisfaction.

[Note 7] Admittedly, this is a very simplistic overview of Mises' argument. I leave out discussion of how the debate concerned matters of cost–for Mises, cost was a subjective matter; and for socialists, cost tended to be measured in terms of objective labour value. For more on the nature of von Mises' argument see Rothbard.(1997).

[Note 8] Long time socialist Robert Heilbroner says, in his 1990 New Yorker article on the collapse of communism, “It turns out, of course, that Mises was right” (p. 92). For an overview of the debate, see Hoff's work Economic Calculation in the Socialist Society (1981).

[Note 9] One justification for classical liberalism and the freedom of persons to trade with anyone they wish, was given by Auberon Herbert (1978). In his theory of 'voluntaryism' Herbert argues that agents own themselves hence they have a right to the objects that are produced with their bodies. Material objects become extensions of the primary and originally owned faculties of one's self. Whether this Lockean sounding theory works or not, it is at least an attempt at answering the question why people have a right to trade with others apart from whether it really benefits themselves or others.

[Note 10] Of course this view of there being no shortages of goods assumes that values are subjective or equated with preference satisfaction; I argue that on a moral reading the market– which is composed of different persons who may not choose what's right –may fail to produce values that ought to be produced (Haddow 2000).

[Note 11] For discussions regarding the potential for expanding the nature of economic man see Machan (1995) and Morse (1997).