CONSUMER (Social Science)

The consumer plays a central role in mainstream economic analysis. This analysis explains economic behavior in terms of individual optimization, and the consumer, who buys and uses goods and services to maximize utility or seeks his or her highest level of satisfaction given the constraints he or she faces, is a central locus of decision making. Mainstream economics is often defined as the subject that examines how scarce resources are allocated to alternative uses to satisfy human wants, and these wants are embodied in the preferences of the consumer. Many of the key ideas of mainstream economics—such as the existence of general equilibrium in a market economy, and its property that this equilibrium is socially efficient—can be illustrated with economic models in which the only economic agents are consumers with given endowments of goods, interacting with others through market exchanges (without the producer, the other important agent in economics).


The basic theory of the consumer examines an economic agent with a preference ordering over bundles of consumer goods. It is assumed that this ordering is complete (i.e., that the consumer can order all possible alternatives), reflexive, and transitive, and that more of any good is preferred to less (Barten and Bohm 1982). Under these assumptions the consumer’s preferences can be represented by a utility function that assigns a real value to commodity bundles


where utility, U increases with the quantity of each of N goods consumed, q.. The consumer is also assumed to have a budget constraint given by pxqx + p2q2 + ••• + p q < Y, where p. denotes the (money) price of good i, exogenously given to the consumer, and Y is the (money) income of the consumer. The consumer is also assumed to maximize utility by choosing quantities of goods consumed, q.. The solution of this maximization problem is usually shown for the case of two goods (N = 2) with an indifference curve diagram as shown in the figure above, where the line BB is the consumer’s budget line, which shows combinations of consumption bundles that satisfy the budget constraint with an equality (with the consumer spending his or her entire income) and the curves marked U are indifference curves, each one denoting a particular level of utility of satisfaction for the consumer, and drawn "smooth" and convex to the origin to show that consumers prefer "mixed" baskets to baskets containing a large amount of one good.

Figure 1

Figure 1

The consumer maximizes utility or reaches his or her highest level of satisfaction at the bundle E, which is the consumer’s equilibrium. At equilibrium, the consumer equates the price ratio, pj/p2, the slope of the budget line, which shows the rate at which the consumer can substitute between the two goods at their prevailing prices, to the marginal rate of substitution between the two goods, the slope of the indifference curve, which shows the rate at which the consumer prefers to substitute between them to leave utility unchanged. In the general case with more than two goods, the consumer’s equilibrium requires that the price ratio between any two goods is equated to the marginal rate of substitution between them.

This basic theory has a number of implications. First, given prices and preferences, an increase in income shifts the budget line outwards without changing its slope and increases the consumption of all goods if they are "normal" goods, and increases the consumer’s utility. For some goods it may be that the quantity demand falls with income; these are called inferior goods. Second, other things remaining constant, a fall in the price of a good rotates the budget line outwards and increases the consumption of that good, provided it is a "normal" good. This increase occurs both because the relatively lower price of the good makes the consumer substitute away from the other good along an indifference curve (the "substitution effect"), and because the lower price in effect increases the real income of the consumer, inducing him or her to buy more of all goods (the "income effect"). This yields the consumer’s demand function and curve showing the relation between the price and quantity consumed of a good. This, when aggregated over all individuals in the market, provides the market demand function and curve of the good. If the good is inferior, it is possible that the fall in the price of the good can lead to a fall in the quantity of the good consumed, with the consumer buying less of the good because he or she feels richer. Third, if the prices of all goods and the consumer’s income change by the same proportion, the budget line will be unchanged, and the consumer will choose the same bundle. In other words, consumer demand is homogenous of degree zero in income and prices, which is known as the homogeneity property. Fourth, a shift in preferences favoring a good shifts the indifference curves towards the axis that measures the quantity of that good, and increases the consumption of that good.


This discussion has used utility and the level of satisfaction interchangeably. However, the early analysis of the consumer provided a hedonic interpretation of utility, identified it with pleasure, and took it to be cardinally measurable (Stigler 1950). It was assumed that utility increases with the level of consumption of a good, so that marginal utility—the change in utility due to the change in the quantity of a good consumed—is positive, but increases at a diminishing rate, so that marginal utility diminishes with increases in consumption. In this approach consumer equilibrium occurs when the marginal utility obtained from a commodity divided by its price is the same for all commodities. This equilibrium implies that an additional unit of money is allocated over different goods in such a way that it yields the same additional utility.

Concern with cardinal measurability led to the development of the ordinal utility theory of the consumer, with early contributions by Vilfredo Pareto (1909) and Eugene Slutsky (1915), and with later developments by John R. Hicks (1934) and Roy G. D. Allen (1934). In this theory, the concept of diminishing marginal utility is not meaningful, and is replaced by the assumption of the diminishing rate of marginal substitution (which states that the amount of a good that the consumer requires to stay at the same level of utility when he or she gives up a unit of another good increases as the consumer has more of the latter good), which is equivalent to indifference curves being convex to the origin. The implications of changes in income, prices, and preferences are the same with this approach as with the earlier one, without requiring the stronger assumption of cardinally measurable utility. The marginal rate of substitution between two goods is equivalent to the ratio of their marginal utilities.

The assumptions required for analyzing consumer behavior using Paul A. Samuelson’s revealed preference approach (1938) are even less stringent, with no utility functions being assumed. Only the implications of observed behavior and the consistency of preferences are used to develop consumer theory.

Despite this trend towards requiring fewer assumptions, some analyses of consumer behavior, for example the examination of decision making in risky situations, and the evaluation of the well-being of the consumer, require stronger assumptions about the measurability of utility.


The basic theory of the consumer has found many applications in economics (and even outside it) by suitable interpretations of "goods." The choice made by the consumer between consumption and leisure treats leisure as a good and takes into account the time constraint faced by the consumer that is allocated to work (which yields income) and leisure (which provides satisfaction but not income). This analysis is used to examine the labor-supply decision of the consumer. The choice made by the consumer between consumption in one period and consumption in the next (assuming a two-period life of the consumer, for simplicity) addresses the intertemporal dimension of the consumption-saving decision, with the relative price of consumption this period being determined by the interest that is yielded by savings. The additional amount of the good in the next period with which the consumer must be compensated (so that the consumer’s level of utility does not change) for giving up a unit of the good this period is assumed to be greater than one by an amount that is called the consumer’s rate of time preference, which measures the consumer’s degree of impatience. In this theory, an increase in the interest rate reduces consumption in this period and increases saving through the substitution effect by making consumption this period more expensive, but it may reduce saving through the income effect of making the consumer increase consumption now. More generally, the consumer can be assumed to have a life-time horizon of longer than two periods. Sometimes the time horizon is taken to be infinitely long. The approach is used to examine choice when the consequence of choices is not certain. It is assumed that the outcome of a choice has a probability distribution, and the consumer maximizes expected utility, that is, the utility derived from each possible outcome weighted by its probability of occurrence. The approach is also used in the analysis of the portfolio choice among different assets for the asset-holding consumer. For example, if the consumer prefers higher returns to assets and lower risk, the consumer is taken to have a utility function over return and risk. Given the return and risk characteristics of different assets, the consumer can then choose the overall return and risk of his or her portfolio, and hence the allocation of wealth among different assets.


The empirical analysis of consumer behavior traditionally has been conducted using econometric methods, with the estimation of demand curves for particular products and systems of demand functions. This type of empirical work usually refers to an aggregate of consumers—for example, to all consumers in a country. Because the theoretical implications regarding individual consumers do not carry over to aggregates, additional theoretical assumptions, implying the existence of a representative consumer, are sometimes used to translate the theory regarding individual consumers to aggregate data. More recently, empirical estimation at the individual level has been conducted using survey and experimental data, as mentioned below.

The inverse relation between quantity demand of a product and its price has been repeatedly confirmed by the evidence. Charles Davenant’s 1699 estimation of an inverse relation between the price of corn and the amount harvested is perhaps its earliest example. Subsequent attempts at estimating the demand functions for individual products and systems of demand functions yield the same result. Although the so-called "Giffen good," for which the quantity demand rises with the price, has been much discussed, no such empirical relationship has actually been documented.

This, however, should not be taken to imply that basic theory of consumer behavior is validated by empirical evidence. As mentioned earlier, the theory does not necessarily imply that quantity demanded falls with the price. Moreover, as shown by Gary Becker (1962), the inverse relation between price and quantity demanded can be explained in terms of individual behavior that is totally random (though restricted by the consumer’s budget constraint) or determined by habit (as long as the consumer can afford to be governed by it), rather than by optimization, as assumed in consumer theory. Some other aspects of consumer theory do not seem to be consistent with empirical evidence. For instance, the homogeneity property is not usually confirmed. Survey evidence suggests that levels of utility as measured by self-reported indicators of happiness do not rise with the level of consumption and income, at least beyond a certain point. This casts doubt on the assumption that increases in consumption imply increases in utility. The usefulness of empirical work on consumer behavior arguably lies not in testing the validity of consumer theory but in exploring empirical regularities such as the well-established Engel’s law, which finds that people spend a smaller proportion of their budget on food as their income rises.


The theory of the consumer has been criticized from numerous angles, some of which have prompted its modification. Three may be briefly mentioned.

First, it is not clear who the consumer, the decision-making agent, really is. Given the preoccupation of mainstream economic theory with the optimizing agent, it is natural to interpret the consumer as an individual person. However, in studies of consumer behavior the consumer is often taken to be the household, or family, or even in some cases the family dynasty; this often leads to ignoring intrafamily differences. Even the idea of the individual as the consumer may be problematic if individuals are thought of as having multiple selves (Elster 1986).

Second, the notion that consumers can actually choose optimally over all possible consumption bundles has been criticized on the grounds that (1) they do not have easy access to all relevant information; (2) in uncertain environments they are unable to calculate realistic probabilities of outcomes; and (3) they do not have the information-processing capability of computing the optimal outcomes even if they had all the necessary information. Herbert A. Simon (1955) has argued that consumers and other economic agents are "satisficers" (a term he coined) rather than optimizers, in that they try to achieve satisfactory rather than optimal outcomes, and that they have, at best, bounded rationality. Because what is satisfactory is not as well defined as what is optimal, this approach has led to research on how consumers actually behave, using survey and experimental data. This research suggests that actual consumer behavior can deviate from what is considered optimizing behavior. For example, consumers give more weight to actual money costs than to opportunity costs (contrary to much of consumer theory), are more averse to gains than to losses, give greater weight to their own experiences and those of others that they know personally than to more objective data in making decisions in uncertain situations, and are myopic about the future.

Third, the notion that consumers have exogenously given preferences over commodity bundles flies in the face of evidence that people are affected by advertising, that they form habits that are difficult to break, and that they are affected by the behavior of other consumers. Many of these ideas can be incorporated into consumption theory by modifying it suitably. For example, it has been argued that a consumer’s utility depends not only on the absolute level of consumption of the consumers, but also on what he or she consumes relative to other consumers (especially peers), and this may occur because of information issues, network externalities, and status. Although this modification can be made within the optimizing framework of consumer theory, it may have implications very different from it. For example, individual optimization may be inconsistent with social optimality (i.e., everyone could be better off if they consumed less and enjoyed more leisure, but they are prevented from achieving this outcome because they value their relative consumption level), and people may not be significantly (or even not at all) better off with significant increases in consumption (as mentioned earlier) when others consume more as well (Frank 1999).

Next post:

Previous post: