Given the weaknesses of the neoclassical theory of consumption outlined in The Calculus of Hedonism, one hardly needs attacks by critics to demolish the notion of utility maximisation. Nonetheless, of course, criticisms abound. Some of the most cogent were made by the American economist Thorstein Veblen at the turn of the 20th century.
Writing of “The Limitations of Marginal Utility” in 1909, Thorstein Veblen conceded that the premises of marginal utility theory “commend themselves to all serious and uncritical persons at first glance”, since it is evident that human behaviour is purposeful, and marginal utility theory appears to provide an explanation for purposeful behaviour. But Veblen saw crucial logical inconsistencies and intellectual shortcomings in the “Hedonistic Calculus”, which invalidated this superficially plausible account of human behaviour.
Economic theory purports to explain how a consumer, with given tastes, will spend his/her income. How these tastes are formed is completely ignored, as if they came into existence via an act of immaculate conception. It is an explanation of “what” a consumer will do, but it is not an explanation of “why”. It is thus a teleological explanation (from the Greek “telos” meaning “end”) when a more interesting theory would also be evolutionary, explaining how consumer tastes are formed, and how they change over time.
When confronted with this criticism, economists once claimed the intellectual equivalent of the Fifth Amendment: the formation of consumer tastes is an issue for Marketing, Anthropology or Cultural Studies, which are separate disciplines to economics. Thus Jevons remarked that
This perspective could be acceptable if the question of how tastes develop over time was truly economically unimportant. But it is quite obvious that the process of technical change over time interacts with the development of consumer tastes over time. The fact that consumers have taken to computers as much as they have, for example, certainly influences the relative level of investment in the development of computer technology, compared to, say, investment in alternatives to the internal combustion engine.
Since technical change is the largest factor in determining economic growth, then the question of how consumer tastes develop over time is clearly an economic issue. If the manner in which economists represent consumer behaviour prevents economists from examining this issue, then far from being a tool to analyse individual behaviour, the economic theory of the consumer is a hindrance to understanding economic processes.
The economic manner of analysing consumption omits all these interactions between consumers and the economic system, when in fact these interactions are what makes the market economy dynamic. Veblen savagely satirises this hermetic approach:
Veblen instead argued for an evolutionary approach to analysing economics, in which consumer tastes would not be taken as given but rather seen to “co-evolve” along with changing culture and technology.
Veblen’s next criticism points out a paradox: if individuals are to behave as hedonists, and the market economy is to function as economists assume, then all behaviour cannot possibly be motivated by hedonism. Instead, consumers must be schizophrenic: they must be hedonistic with respect to consumption, but utterly selfless with respect to the key institutions of the market.
For a market system to function properly, there must be at least two social institutions: the concept of individual property, and the concept of legal exchange. If people instead acquire the commodities they want by theft, then force rather than prices and incomes will determine what people consume–and, of course, the whole system of production will break down. An orderly market system therefore requires that, in the main, people respect the institutions of property and exchange by means of market-determined prices.
However if people are truly hedonistic, then why should they unflinchingly obey the rules of legal exchange? Why not instead obey the law only up until the point at which the marginal disutility (“pain”) from breaking the law equates to the marginal utility (“pleasure”) gained from breaking it? If people did behave in this fashion, then the system of exchange–and much else besides–would start to break down.
Bowles and Gintis (1993) point out that among the things which break down is conventional economic theory itself. The presumption that everyone accepts property rights without question implies that the market is populated by “Victorian gentlemen”, who may bargain to get what they want, but once a bargain is struck, “a handshake is a handshake”. However, modern economics has replaced this genteel conception of “Homo Economicus” with
This universal picture of self-interest introduces costs into every market transaction which prevent those transactions from being set at prices that maximise social efficiency. Thus an employer will pay her employees more than their “marginal product”, so that the threat of the sack is a credible way to enforce employees to work hard; a banker will ration credit, and charge borrowers less than the market rate for loans, so that borrowers are encouraged to maintain their relationship by not defaulting. Complete rationality at the individual level is thus incompatible with the original economic vision of a market economy which impartially maximises everyone’s utility.
Therefore, Veblen alleges, hedonism alone cannot explain individual behaviour in the real world, even when our purview is restricted simply to the issue of how consumers decide which commodities to consume:
Clearly people do not, in the main, behave this way. Instead, they obey the conventions of civil society, while at the same time not being driven solely by hedonistic calculation in their individual behaviour. This non-self-interested behaviour is crucial to enabling a market society to function with reasonable efficiency. As Bowles and Gintis put it,
efficient and otherwise desirable solutions to coordination problems often are facilitated by social norms valuing such things as cooperation, truth-telling and non-aggression towards others... Kenneth Arrow once suggested that norms of social behaviour, including ethical and moral codes, might be “reactions of society to compensate for market failures” or “agreements to improve the efficiency of the economic system … by providing commodities to which the price system is inapplicable”. His example was trust. Indeed, the enforcement costs of a society without trust would be monumental...
Stiglitz (1993) commented that Bowles and Gintis actually understate the impact of their critique, leaving out other ways in which completely self-interested individuals are inimical to the functioning of a capitalist economy, and not sufficiently emphasising the way in which this vision backfires on economic theory:
The issues of how the people behave in a market system, and how a market system affects how people behave, are therefore important and interesting ones. But they are topics on which conventional economists have made and can make almost no contribution, because their model rules out the very social relationships on which actual exchange in society depends. As Hirschman observes, this is truly ironic, because it means that economists are unable to proclaim one of the possibly profound benefits of a commercial culture:
There is a final irony which explains why so many socially conservative individuals are also critics of economic theory. If the dominant philosophy of our society successfully encourages the view that “it’s every man/woman for him/herself”, then this will undermine respect for institutions, and encourage lawlessness. Thus some social conservatives correctly see economic theory as a force which is undermining civil society. Rising crime rates and the erosion of once prevalent customs may be in part a by-product of an utterly individualistic–and utterly false–perception of individual behaviour in a market economy.
Veblen argues that a proper economic theory of consumption should consider not only how demand is determined, but how the rate of change of demand is determined. But economic theory treats demand as an entity which can either actually be at rest, or which can be justifiably treated as constant even though, in reality, it is continually changing.
Both economic assumptions are incorrect. Demand is a function not simply of quantity (as economic theory argues) but also of time. By leaving time out of their equations, economists ignore the truly important questions of what determines how demand changes over time.
They also presume that the time interaction of demand and supply is unimportant: the equilibrating mechanism of the market is taken as being so effective that the time path of demand and supply is irrelevant. Why worry about the dance of the demand and supply curves if the dance is over so quickly that, before you know it, the lovers are once again seated in perfect equilibrium?
Unfortunately, as the chapter Chapter 8 shows, the dance of demand and supply is never ending: economic equilibria are almost certainly unstable, and the dancers of supply and demand will never take their equilibrium seats.
Marginal utility theory argues that the ultimate objective of all actors on the market economy stage, from the poorest to the richest, is consumption. Thus both the poorest wage-slave and the richest tycoon work simply in order to be able to satisfy their desire for consumption.
To this, Veblen says “nonsense”. This proposition is reasonably applicable when applied to workers, but patently invalid when applied to businessmen. The wealth the most successful businessmen accumulate–the Rockefellers of the 19th century, the Gates and Murdochs of the 20th–far exceeds any practical level of consumption, and unless bankruptcy intervenes, they die with the vast mass of their wealth unconsumed. The prime motivation for these ultimate capitalists–and their lesser brethren–is thus not consumption, but wealth itself, and Veblen argues that is absurd to imagine otherwise. It is worth quoting Veblen at length on this issue:
This adds an additional level of criticism to the earlier derision of the assumption that all individuals are identical. Some individuals are so poor that all they can consider is food; most Western consumers are sufficiently well-off to consider consumer durables and the occasional indulgence; and a minority are so wealthy that individual consumption could never be the sole objective of the wealth, and their marginal expenditure will bear no resemblance to the expenditure of the poor or the merely affluent. We therefore need to consider income differences when trying to account for individual behaviour: a one-size-fits-all analysis will not do. Once again, the Classical approach of analysing the behaviour of different groups in society differently is preferable.
Veblen’s critique and the internal inconsistency of utility theory do not exhaust the problems with the economic conception of individual behaviour. Four more problems are worthy of mention.
The economic definition of rationality ceases to be rational as soon as you consider one essential aspect of the real world: time. When time is taken into account, two of the four fundamental building blocks of the economic concept of rationality (completeness, transitivity, non-satiation, and convexity) are clearly irrational: non-satiation and transitivity. The other two (completeness and convexity) are also suspect.
It is not true that “more is always preferred to less” when we consider consumption as a function of time.2 Consider the prime example used in this chapter: bananas. Two bananas per day may well be preferred to one; three per day is probably pushing the envelope for most humans; and you would have to be a monkey to, for example, prefer twenty bananas in a day to nineteen. Most humans would kill rather than consent to eating a twentieth banana in a day. Thus, when we consider consumption as a function of time, anyone who behaves in a fashion which economists call rational–always preferring more bananas per unit of time to less–is clearly insane.
Transitivity also ceases to operate when time is part of the argument. A consumer may well prefer bundle A to bundle B at one time, and B to C, and yet at another time he/she may prefer bundle C to bundle A. Economists might protest that these comparisons are only valid when they are all made at the same instant, but that is precisely the problem: economic theory requires that everything happens at once, when in the real world an individual’s consumption is sequential, not simultaneous.
Completeness is needed to be able to argue that tastes are independent of the consumer’s income, since if a consumer’s tastes altered as his/her income rose, the “indifference map” would change with every movement in the budget line and individual utility maximisation could not work. But the failure to aggregate effectively from the individual to society means that this very problem recurs at the social level, so avoiding it at the individual level is not much use. In reality, of course, our tastes are shaped by our incomes, and a theory of consumer behaviour should recognise this. As discussed below, completeness also requires an impossible level of computation by the individual consumer.
Convexity is also a by-product of the timeless analysis of consumption, which permits economists to ignore the manner in which the consumption of one commodity interacts with the consumption of another. Convexity implies that, to keep a consumer’s utility constant, it is necessary to subtract units of one commodity while adding units of another. But if the commodities are “sugar” and “tea”, and we are considering the act of drinking tea, then it is not true that a reduction in the amount of sugar can be compensated for by an increase in the amount of tea.
Secondly, as I will argue in more detail in Chapter 8, the whole notion of a demand “curve” only makes sense in an unchanging, static world. In the changing, dynamic world in which we actually live, what matters is not simply demand as a function of price now, but how that demand is changing over time. While it might seem that it is harmless to focus simply upon demand as a function of price, and to ignore demand as a function of time, in fact a time-based analysis leads to a very different view of how the economy functions. The whole focus upon the “demand curve” itself is misplaced.
Just as the economic treatment of consumers treats all people in a “one size fits all” manner, it does the same thing to commodities–the decision to eat food is treated as no different than the decision to attend an opera. Yet there are some goods without which life is merely less interesting, while others without which life is impossible.
Psychologists are familiar with the concept of a hierarchy of needs, as first developed by Abraham Maslow. Maslow argued that there were 5 major levels to human needs, starting with food, then clothing, shelter, self-esteem, and finally self-actualisation.
Whereas neoclassical economics treats the consumer’s preferences as being entirely independent of income–so that even the poorest consumer’s “indifference map” includes Rolls-Royces–Maslow’s approach intimately links income and tastes. A poor consumer is dominated by needs at the bottom end of Maslow’s hierarchy–the need to eat, to obtain the necessities of life. For a rich consumer, on the other hand, the necessities of life are minor points, while the consumption decisions which matter relate to the fourth and fifth levels of the needs hierarchy.
Maslow’s analysis has obvious implications for the classification of commodities and the analysis of consumer behaviour, but obviously the concept has played no role in neoclassical economics. In particular, it provides additional support for Kirman’s argument, derived from the aggregation failure of conventional economics, that consumers should be divided into social classes when analysing consumption. This treatment is an underlying concept in Post Keynesian economics (see Lavoie 1994), and a key concept in the fledgling school of evolutionary economics (see Earl 1995).
The economic example of a consumer choosing between different bundles of two commodities makes the process of forming a set of indifference curves seem quite easy (even if one of the two “commodities” is “all other goods”). But in the real world there are not two commodities, but tens (and perhaps even hundreds) of thousands of products. To actually construct a set of indifference “surfaces” between these tens of thousands of products, a consumer would need time to decide between such combinations as “10 pieces of tofu, 1 hamburger (and given quantities of 10,000 other commodities) and “9 pieces of tofu, 2 hamburgers (and given quantities of 10,000 other commodities)”, or “9 pieces of tofu, 1 hamburger, 13 softdrinks (and given quantities of 9,999 other commodities) and “9 pieces of tofu, 2 hamburgers, 12 softdrinks (and given quantities of 9,999 other commodities).”
Each such decision involves choosing between two minor variations on combinations of ten thousand commodities, and the choice has to be repeated for each of the possible variations, and for each of the possible quantities of each commodity. The number of choices rapidly escalates into the trillions of trillions of trillions, so that even if each decision was made in a millionth of a second, a consumer would be older than the known universe before she was ready to shop. Even this is being generous, since experiments have found that, when forced to choose between just eight commodities the way economists assume consumers do, experimental subjects took 30-40 minutes to make their decisions (Sippel 1997). The economic vision of the consumer is therefore not an ideal, but a sheer impossibility.
As an illustration, consider a consumer faced with a choice between two commodities, when she can purchase between zero and ten of each. The number of combinations is 100--from (0,0) to (10,10) and everything in between. That's just 100 combinations which, to utility maximise while staying within a budget constraint, the consumer would have to:
calculate the utility derived from the consumption pair;
calculate the budgetary cost of the pair
decide whether the pair fits within the budget
work out whether the utility of the pair is greater than all others previously considered.
These four decisions won't take too long for just two commodities. But if there are 3 commodities to consider, then there are 1,000 possible combinations; if there are 10, there are 10,000,000,000 combinations.
In reality, a weekly shop in the local supermarket results in between 20 and 40 different items in the typical shopper's basket. Let's consider the midpoint of 30 commodities, and imagine that we resricted ourselves to between zero and ten items of each.
If you could speedily rule out 99.9% of these combinations on the grounds that they either exceeded or fell well below your budget, and if each set of decisions above took you just one billionth of a second, you could decide which of these 1,000,000,000,000,000,000,000,000,000,000 combinations maximised your utility in 31 billion years--roughly one and a half times the maximum estimated age of the universe.
Clearly, consumers don’t behave the way economists assume they do. Instead, they compartmentalise their shopping decisions–with the largest and most important categories being “food, clothing, shelter”–and form habits within these compartments which are constrained by their incomes. This behaviour sensibly restricts the choices they have to make to a manageable set, but of course results in consumers not responding instantly to new relative prices, as economists expect them to.
This is one reason why experimental tests of the theory of consumption normally find that consumers are not “rational”, as economists define the term. For instance, Battalio et al’s study of consumer responses to changes in prices at a psychiatric hospital (I kid you not!) found that their behaviour violated GARP, the “Generalized Axiom of Revealed Preference”. According to GARP, if a consumer buys bundle A when both bundle A and B are affordable, then he should always buy A rather than B at other prices when both are affordable, since A must lie on a higher indifference curve than B. But these experimental consumers frequently made purchases which, to an economist, meant that they were voluntarily choosing a lower indifference curve than they could afford.
The researchers attempted to explain this “irrational” violation of economic theory as “the residual effects of the previous period’s consumption patterns” (Battalio et al. 1977). But instead, what is irrational is not the consumer, but the economist. Even patients undergoing psychiatric care are rational enough to rely upon habit to reduce the bewildering range of choices to a more manageable set. This realisation–that economic agents must, in some systematic manner, reduce the range of options considered from the near infinite number that computationally exist to a number which is manageable, underlies Herbert Simon’s argument that agents evince “bounded rationality” and therefore “satisfice”, rather than “optimising” with infinite rationality, as postulated by neoclassical economics (Sent 1997).
The sole reason why economics does not acknowledge that one person’s welfare affects anothers is that this would make it patently impossible to argue that society’s welfare is simply the sum of the welfare of its individual members. Since economists have proven that, even if we abstract from these aspects of our humanity, society still cannot be reduced to the sum of its individuals, then economics can no longer justify ignoring the undeniable interpersonal and social aspects of human behaviour. Indeed, some critics go so far as to argue that, far from being something which economists should ignore, the ethical traits of humanity are the attributes which most strongly identify us as human–as strongly as the spinning of webs identifies a spider.
Once we accept that one person’s actions will affect another’s “utility”, and that people do care about what happens to other people, then two results follow. Firstly, the pure self-interest which economics takes as the essence of human behaviour cannot possibly lead to the maximisation of social welfare. Secondly, to explain how markets actually behave, we must acknowledge the ethical, non-hedonistic aspects of our humanity. As Wilber (1996) argues:
This ingrained morality cannot be reduced to the individual “pursuit of pleasure and avoidance of pain” which Bentham believed was at the root of all behaviour, because as the evidence of history shows, belief in what is ethical has time and time again impelled people to put their very lives at risk for what they perceive as the common good. Nor does ethical or altruistic behaviour obey the laws of the market. Quoting Wilber again,
This approach amplifies Veblen’s comments about the schizophrenia behind “the calculus of hedonism”. Far from ignoring ethics, a well-functioning economy requires that people display ethical behaviour and obey the rules of market exchange.
There are few direct policy implications from the calculus of hedonism, but many indirect and sometimes contradictory ones. For example, you would probably expect that studying economics makes people more self-interested; but the evidence on this is mixed. Some studies show that economics students are more likely to propose apparently self-interested behaviour when deciding how to share between themselves and others (Marwell & Ames 1981, Frank, Gilovich & Regan 1993, 1996), but other studies claim that economics students are more altruistic in their actual behaviour than others (Carter and Irons 1991, Yezer, Goldfarb & Poppen 1996). And, of course, economists themselves often dispute whether this selfishness, if proved, would actually be a bad thing (Hirshleifer 1993).
These apparently contradictory results make sense when you appreciate the fact that followers of economics actually believe that their theory shows how society could be improved to the benefit of all its members. Ironically, though economic theory preaches hedonism, it does so on the basis that this is the best way to organise society.
Because this faith is misplaced, economics predisposes its believers to accept that the market is fair (Whaples 1995), and to promote policies which make society more unequal, more unfair, while at the same time encouraging them to behave relatively altruistically in this not quite the best of all possible worlds. It is this predisposition to accept hedonistically-oriented policy which is perhaps the most insidious effect of this aspect of the economic creed.
The fact that this creed is based on erroneous logic means that, in reality, economics should encourage socially cohesive behaviour, and policies which promote cohesion, rather than “altruistically” championing the pursuit of inequality.