Oligopoly, the economist’s analogue to oligarchy in political science, is defined as a market situation where independent sellers are few in number. The origin of the term is not clear, but it is known to have appeared in the original, 1518 Latin version of Thomas More’s Utopia. Common usage of the term in English writings, however, dates from the 1930s (see Chamberlin 1957). The central analytical problem with which the theory of oligopoly is concerned is how each of the few sellers reacts to the economic activities of its rivals in order to bring about determinate equilibrium solutions. The theoretical problem is therefore conceptually different from, and vastly more complex than, that of either pure monopoly or pure competition. The monopolist’s revenue function is identical with the market demand schedule; that of the pure competitor is easily derivable from a given market price which, by assumption, cannot be affected by any one of the numerous competing firms. Hence, neither the monopolist nor the purely competitive firm must consider how alternative actions by rival firms will affect its own revenue possibilities—the monopolist has no rivals and the competitive firm has so many that it can ignore any one of them. Oligopolists not only have rivals, but they have so few of them that each is large enough to affect the others significantly. Their prices, outputs, and other relevant variables are therefore interdeterminate.
The problem of interdeterminacy among oligopolists can easily be shown by considering the simple maximizing equations of contemporary economic theory. It is generally assumed, at least as a first approximation to reality, that business firms seek to maximize profits. The validity of the assumption very likely varies directly with the degree of competition in the market. Highly competitive firms earn only a competitive return on capital even when they pursue profits maximization. Hence, should they pursue any other objective, they would earn less than a competitive rate of return. Firms with some degree of market power may at least seek another objective, such as maximum revenue or market share, subject to a minimum profits constraint, and still survive (Baumol 1959, pp. 54-68).
Profits, π, are defined as total revenue, R, less total costs, C. In turn, R and C are, respectively, price and unit costs multiplied by the quantity produced and sold. That is, R = pq, and C = cq. For a monopolist the JR schedule for various alternative q’s is readily determinable, since the price under given demand conditions depends entirely on the quantity the monopolist offers on the market: p = F(q).Hence, the monopolist’s equilibrium output and price are determined when π = R — C is maximized. Under pure competition the same equations apply, except that in R = pq, p is determined by the market and is treated by each competitor as a parameter.
For each oligopolist, on the other hand, the price it receives is dependent not only upon the quantity it supplies (q1) but also upon the quantities supplied by its rivals (q 2 ,qs, . . ., q n ). Therefore,V = P(tfi + q2 + q3 +... + q n ) and Ri = pq x. Accordingly, unless each oligopolist makes certain assumptions about the quantities its rival will supply, its own revenue schedule (R1) is formally indeterminate. Moreover, the interdependency of oligopolists encompasses far more variables than their respective outputs: in the modern industrial market such variables as pricing, advertising, and innovational activities and strategies are also determinants of equilibrium. Theoretically, and as a practical matter, each oligopolist must evaluate these variables for all its rivals before it can rationally select its own course of action so as to maximize its profits. Furthermore, in the selection process it must allow for whatever effect its own course of action will have on the variables it must evaluate. This complex interdependency of firms’decisions with respect to the important market variables is the essential and distinguishing feature of oligopoly. It also explains, as will be demonstrated below, why oligopoly theory predicts less satisfactorily the behavior of business firms to which it applies than do the theories of monopoly and competition.
The unsatisfactory and inconclusive state of contemporary oligopoly theory leaves an important gap in our knowledge of the operative mechanics of modern industrial economies containing a significant ingredient of private capitalism. Since World War II the more developed countries, especially Canada, Japan, and those of western Europe, have placed increased reliance on competitive market forces to regulate their respective national economies—greater reliance than ever before. To secure at least a workable minimum of competitive market forces Austria, Belgium, Canada, Denmark, France, West Germany, Ireland, Japan, Luxembourg, the Netherlands, Norway, Sweden, Switzerland, and the United Kingdom have enacted legislation limiting cartel activities; the Italian parliament is committed eventually to enact similar legislation, on which it has deliberated for several years. Both the European Coal and Steel Community and the European Economic Community(European Common Market) treaties contain provisions prohibiting certain cartel activities and restrictive business practices. In the United States antitrust legislation has been strengthened, and the laws have been administered with increasing vigor. Clearly, in much of the advanced industrial world the pace and direction of economic activity are left to market forces preserved by laws designed to prevent private firms from frustrating these forces through cartels and restrictive business practices.
[SeeAntitrust Legislation; Cartels and trade Associations.]
Yet in all these countries many industries conform to some variant of oligopoly. In the United States, concentration ratios—the share of output accounted for by the four largest firms—exceed 50 per cent in many manufacturing industries. The level of concentration in manufacturing in Canada and the United Kingdom is generally higher than in the United States (Rosenbluth 1955, pp. 71-79),as is probably the case in most other industrialized countries. In Denmark, West Germany, and the United Kingdom, dominant firms (usually defined as single firms accounting for as much as 25 per cent of a relevant market) are sufficiently evident to be given special attention in legislation governing restrictive business practices. The prevalence of oligopoly, and the present inadequate state of oligopoly theory, leaves a serious void in our understanding of how market forces govern a significant portion of the economic activity in those countries which rely on them. The void persists in spite of the heroic efforts made since the 1840s to close it.
The Cournot solution
The first formal solution to the problem of oligopolistic interdependence is associated with Antoine Augustin Cournot, the French economist and mathematician (1838). Contemporary theory textbooks often identify Cournot only with the classical duopoly (two-firm) solution. It is true that his examples consisted of two-firm markets, but this seems clearly to have been a matter of simplifying the presentation. At each point he extended his analysis to the three-, four-, and n-firm cases.
Essentially, Cournot resolved the problem of oligopolistic interdependence by postulating that each (firm) independently sought maximum profits under the assumption that its rivals’ outputs were fixed. That is, given the output of its rivals, each oligopolist supplied that share of the remaining demand which maximized its profits. (The similarity in this regard of Cournot’s oligopoly model and the much more recent models of dominant-firm price leadership will become apparent later on in the treatment of price leadership.) However, until they reached a stable equilibrium, any change in the quantity offered by one firm necessarily forced the others to alter their offerings too. Accordingly, while each rival proceeded to maximize profits on the assumption that it had no direct effect on the outputs of other firms, the outputs of all were made interdependent through the indirect effect; i.e.,each move by one forced a countermove by the others.
Cournot’s greatest contribution to the problem was his recognition of the strategic interdependence of oligopolists and the set of reaction curves he designed, which, under the foregoing assumptions, yields a stable equilibrium solution (see Figure 1).Reaction curve m1 n1 is defined as those profitsmaximizing quantities firm i will offer for sale for all the possible quantities supplied by firm ii; firm n’s analogous reaction curve is m2 n2. Thus, if firm i were to offer the quantity x′, which would be its profits-maximizing quantity if firm II offered the quantity y4, firm n would then react by offering
the quantity y 3 but then firm i would react by adjusting its quantity to x′ forcing firm n in turn to offer quantity y2, The reaction process would continue until the quantities supplied by firm I and firm II were respectively x and y, the equilibrium output for each firm. The equilibrium is stable, since the outputs of the two firms will eventually settle at x and y, irrespective of the assumed initial offering of either firm.
The implications of Cournot’s oligopoly solution for welfare theory were extremely important, although rarely explicitly stated. He carefully pointed out that his reaction model, built upon formal independence among the participating oligopolists, yielded a lower price and a larger output than would be the case were they to enter into an agreement( 1960, pp. 79-80). Moreover, the larger the number of participants, the lower the price and the greater the output. Cournot’s generalized price equation for markets comprising identical firms yielded an equilibrium when D + np(dD/dp) = 0, where D is quantity demanded, n is the number of firms, and p is price. When oligopolists entered into a collusive agreement, n became unity and p became the monopoly price. The larger the number of independent firms, the more closely p approximated the purely competitive price.
The defects in Cournot’s model for predicting oligopoly behavior in the real world have been pointed out by many. Bertrand, Marshall, Pareto, and Edgeworth all argued that it failed to allow for the incentive confronting each firm to lower its price and enlarge its sales (Chamberlin 1933, pp. 36-37 and appendix A). Marshall noted that this incentive would be particularly strong under conditions of increasing returns. The most systematic criticism of the theory of oligopoly as developed up through the first quarter of the twentieth century, however, is associated with Edward H. Chamberlin.
The Chamberlin model
Chamberlin’s criticism of the Cournot solution and its variants stemmed from their essential shortcoming: none of the solutions conformed perfectly to the hypothesis that each seller maximized its profits. Realistically, a profits-maximizing oligopolist must take account of its indirect as well as its direct effect on price. As Chamberlin put it, “When a move by one seller evidently forces the other to make a counter move, he is very stupidly refusing to look further than his nose if he proceeds on the assumption that it will not” (1933, p. 46). Chamberlin sought to remedy this defect by having oligopolists at the outset recognize their mutual dependence; he thus gave birth to the notion of conjectural interdependence or oligopolistic rationalization.
When each firm takes into account its total effect on price, under certain rigid assumptions, the oligopoly solution is the monopoly price. This price provides the maximum profits for the group as a whole and, when sellers recognize that their fortunes are mutually dependent, the maximum profits for each. None of the oligopolists will reduce its price below the monopoly price, since it knows beforehand that others will be forced to follow, and thereby reduce the profits of all. Chamberlin emphasized that this results when each firm independently seeks to maximize profits, provided the firms are so few in number that each must consider both the direct and the indirect effect it has on market price. When firms are sufficiently numerous to ignore their indirect effect, price will settle at the level given by the Cournot solution; when they are sufficiently numerous to ignore their direct effects as well, price will fall at once to the competitive level. Hence, in the Chamberlin model, unlike Cournot’s, there is no gradual descent to the competitive price as the number of firms is increased. Instead, a discontinuity occurs somewhere along the spectrum of firm numbers between one and many, whereby the oligopoly solution changes radically from the monopoly price to the competitive price.
The Chamberlin oligopoly solution was received with enthusiasm. Students of industry had often observed monopolistic behavior in industries that were not, by the classical definition, monopolized. Chamberlin’s theory provided an explanatory hypothesis for such behavior. In the United States, antitrust practitioners quickly translated Chamberlin’s oligopolistic rationalization into the legal doctrine of “conscious parallelism,” and in several cases the government successfully prosecuted oligopoly as illegal monopoly under the Sherman Act (Nicholls 1949). While such victories were applauded by some, they were viewed with alarm by others. Chamberlin’s oligopoly solution rested on the rational actions of enlightened oligopolists seeking maximum profits. Moreover, the incidence of oligopoly in the United States, and elsewhere, was high. Hence, the judgments handed down in those cases implied either of two conclusions, both highly disturbing: business firms must stop behaving rationally or a significant percentage of United States industry would be declared in contravention of the antitrust laws.
In time, recognition of the restrictive assumptions underlying the Chamberlin solution revealed that much oligopoly behavior was still left unexplained. Chamberlin had assumed static conditions in which oligopolists confronted identical cost functions and considered only price and output as variables. In the real world, as Chamberlin himself carefully pointed out and elaborated, these highly restrictive assumptions are seldom fulfilled. The modern business firm may typically confront relatively few rivals, but it has available a wide variety of business policies and strategies to employ against them. Advertising and research budgets, product change, marketing methods, and nonprice terms of trade, as well as prices, are variables over which most oligopolists exercise some degree of discretion, and how they exercise it is the essence of the oligopoly problem. Empirical studies of oligopolistic industries have yet to reveal a systematic pattern of behavior with respect to such variables. Accordingly, models of oligopoly confined to only two of them, e.g., price and output, can scarcely be expected to provide an explanatory hypothesis valid for oligopoly generally. As Fellner (1949) and Markham (1959) have pointed out, in oligopoly the struggle for competitive advantages over rivals logically centers on nonprice activities. The new product, process, or advertising campaign is much less susceptible to immediate detection and imitation than the new price schedule is. The significant contribution of Chamberlin’s work in the subject is not that oligopolists can generally be expected to behave as a monopoly but, rather, that they will select competitive strategies more complex and more subtle than simple price competition.
The game theory model
Recognition of the variety of competitive strategies open to oligopolists has led in recent years to a search for formally determinate solutions through the application of the mathematical theory of games [seeGame Theory].Since game theory is expressed in terms of matrix algebra, it has at least the pedagogical advantage over Cartesian diagrams of being able to handle a large number of strategies and counterstrategies simultaneously. The decision problem confronting each oligopolist in the game theory model consists of choosing an optimal strategy from some finite number of strategies, where a strategy is defined as the selection of a particular value for the variables over which the oligopolist has some degree of control. For each strategy selected by one oligopolist, there is a corresponding set of payoffs, or profits, one for each strategy selected by other oligopolists. An example of such a set of payoffs for an industry comprising two firms, A and B, each having available three possible strategies, may be put in the form of Table 1, where the pairs of
values represent the profits A and B will earn for all combinations of strategies they can use; e.g., if A uses strategy 1 and B uses strategy 3, A sustains a loss of 2 and B earns profits of 8. (The illustration is a nonzero-sum game. Zero-sum games, in which the algebraic sum of all gains and losses equals zero, have limited applicability to oligopoly.) Which strategy each firm will in fact use depends entirely on its propensity to gamble and on the intelligence it accords its rival. In the above example, if both A and B could assume that each would cooperate with the other so that the combined profits of both would be maximized, A would use its strategy 1 and B would use its strategy 1. This would be the cooperative equilibrium solution. However, if each rival distrusted the other, neither would use its strategy 1, because in doing so, B could earn a profit of 8 and force A to take a loss of 2 by using its strategy 3, and A could earn a profit of 7 and force B to take zero profits by using its strategy 2. Hence, if A and B each assume the worst, both will use strategy 3 and earn a minimum profit of 3 each. The strategies are compatible, and, so long as these conditions prevail, the industry is in equilibrium.
Theory of games models can be made much more complex, involving mixed strategies whereby each oligopolist seeks to maximize its minimum level of expected earnings by shifting from one price strategy to another, in accordance with some predetermined selection scheme. Such an optimum mixed strategy may be determined by linear programming techniques [seeProgramming]. But it is sufficiently clear from the foregoing simple purestrategy model that formally determinate game theory solutions assume a much higher order of knowledge than business firms possess. If it is true, as economists and business spokesmen frequently assert, that the typical firm can only roughly approximate the demand schedule for its product, it is virtually certain that its knowledge of all the possible strategies available to its rivals and their respective payoffs must be extraordinarily incomplete. Entrepreneurs, however prudent they may be, pursue different strategies when confronted with incomplete knowledge than they do when all the possible outcomes are known. In truth, in a dynamic oligopolistic industry the process of creating competitive strategies is as essential to an understanding of industrial behavior as is how they are employed. Moreover, as game theory models are expanded to more nearly approximate actual oligopolistic situations, they become so complex that they are virtually meaningless. Perhaps the most heroic effort yet made in applying game theory to oligopoly markets can be credited to Shubik (1959).But as one reviewer of Shubik’s work (Kaysen 1960) has observed, the gap between his game theory model of oligopoly and the United States automobile and cigarette industries, to which he applied it, is enormous. On the other hand, a model capable of handling five or six automobile producers, each having a half-dozen or more strategies, would have to be of such tremendous size that virtually nothing meaningful could be said about it.
The Cournot, Chamberlin, and game theory models do not exhaust the list of deductive generalizations on oligopoly behavior; but they are landmarks in the development of oligopoly theory since the 1840s and adequately illustrate its strengths and weaknesses. In its evolution the theory has unquestionably been strengthened. To Cournot oligopoly was simply an intermediate range on the spectrum of firm numbers between one and many and was distinguishable in no fundamental sense from classical monopoly or competition. Chamberlin made oligopoly a distinctive form of market organization guided by its own modus operandi. Game theorists have recognized this distinctive character and have sought to design models capable of handling the variables over which oligopolists exercise discretion. But a theory of oligopoly generally valid for all industries comprising a few firms has yet to be developed. The absence of such a theory is attributable to the inherent complexity of the problem and to the heterogeneity of oligopoly itself.
Oligopoly theory includes explanatory hypotheses for limited aspects of oligopoly behavior as well as the foregoing more general models. The apparent rigidity of oligopoly prices and the uniformity of price among oligopolists have been accorded particularly close attention since the world-wide depression of 1929-1933. The theoretical construction generally offered as a rationale for rigid oligopoly prices is the kinked demand curve;that for uniform prices, dominant-firm price leadership, or Chamberlinian conjectural interdependence as expressed in some form of barometric or conspiratorial price leadership.
The kinked demand curve hypothesis, advanced almost simultaneously by Paul Sweezy (1939) and a group of Oxford economists (Hall & Hitch 1939),rests on the assumption that each oligopolist visualizes two distinct demand schedules for its output, one that obtains if all rivals meet any new price it sets and another that obtains if rivals do not meet such new prices. The second is more elastic than the first because the sales of any one oligopolist will respond much more to price changes that rivals do not match than to those they do. However, if oligopolists reason that their rivals will meet all price reductions but will not follow price increases, a kink occurs in the demand schedule at the currently prevailing price, since the first demand curve is relevant for all contemplated price reductions and the second for all contemplated price increases(see Figure 2). The kink in turn produces a gap in the firm’s marginal revenue schedule, the gap being proportional to the differences in the elasticities of the upper and lower segments of the demand curve D1D′2. Because of the gap, costs may fluctuate over a significant range without disturbing the equality between marginal cost and marginal revenue. Accordingly, a given price may prevail for long periods of time.
This theory of rigid oligopoly prices is inconsistent with the various theories of price uniformity among oligopolists, such as Chamberlin’s theory and the several theories of price leadership (Markham 1951). Under dominant-firm price leadership, the large firm, or partial monopolist, sets a price and lets the rest of the industry sell as much as it wishes at that price. In the absence of a dominant firm, price leadership may emerge as a vehicle for carrying out a tacit price conspiracy, or simply because, by custom, one firm serves as a price barometer. Under any one of the three forms of price leadership, the presumption is that price followers match the price leader’s increases as well as its decreases, and hence the oligopolist’s demand curve contains no kink.
Stigler’s analysis of oligopoly prices (1947) casts considerable doubt on the validity of the kinked demand curve hypothesis. In his price comparisons he found that oligopolies with dominantfirm price leaders generally tended to have less flexible prices than those possessing no price leader. If the kink existed, the opposite should have been the case. Moreover, according to Stigler, there is no logical basis for assuming the existence of such kinks. To exist, they must be created by entrepreneurs, and to create artificial obstacles to profitable price increases would be inconsistent with the objective of maximum profits.
Rigid oligopoly prices, often referred to as “administered” prices, still stand in need of a satisfactory explanation. However, in recent years precisely what needs to be explained has itself changed. Until the end of World War II oligopoly prices demonstrated considerable inflexibility, upward and downward, in comparison with competitive prices. In fact, a leading American economist who served as a price control official during the war has argued convincingly that oligopoly prices were relatively easy to administer by public authority because typically they were administered privately (Galbraith 1952). In
the postwar inflationary period certain oligopoly prices have tended to demonstrate considerable upward flexibility in business upswings but very little downward flexibility in business downswings. Accordingly, the so-called administered price group has, at least in a statistical sense, contributed significantly to the postwar inflation in the United States.
At least four independently conducted studies of the problem have concluded that such price behavior is attributable much more specifically to market power than to administered prices (Adelman 1961; Galbraith 1957; Stigler 1962; Markham 1964). In fact, the term “administered price” probably does more to cloud than to clarify the issue. Administered prices have generally been defined as those prices arrived at within the firm as an integral part of its decision-making process, as distinguished from those generated entirely by the impersonal interplay of market forces. Nearly all prices communicated to the public through published price lists are, by this broad definition, administered. But the oligopolistic prices contributing to the postwar inflation have been relatively few, confined principally to highly concentrated industries that deal with strong labor unions. Clearly, neither oligopoly nor administered prices are massive undifferentiated wholes but, instead, exhibit wide varieties of behavioral patterns.
Retrospect and prospect
George Gaylord Simpson, commenting on the social theorist, once observed: “Each student thus actually put his particular theory into the data, and it is not surprising that each then gets his own theory out of these data when he is through” (1949, p. 139).
In certain respects oligopoly theory has developed like this. There are many varieties of industrial structures, each exhibiting its own peculiar behavioral pattern, that can technically be defined as oligopoly. Some conform to Chamberlinian conjectural interdependence, others to one of the several price leadership models; some give factual support to the kinked demand curve hypothesis, others exhibit reasonably flexible prices; some oligopolists appear to use sophisticated strategies, others to rely heavily on price concessions. Oligopoly is far less homogeneous than the term “few sellers” might imply.
There is a great need for developing the important empirical facts on oligopoly which will permit breaking down an unsatisfactory and meaningless aggregate into classes useful for analytical purposes. Indeed, until this is done, the plethora of theories and subtheories must remain untested, and the operative mechanics by which large sectors of most private enterprise economies are governed must remain obscure. The large number of industry studies published since the 1940s in the United States, Canada, the United Kingdom, Japan, and western Europe have provided a substantial beginning in amassing the relevant facts. When appropriately classified and subjected to the powerful tools of modern statistical analysis, they should reveal the stochastic laws of various types of oligopoly. For the important purposes of administering the growing body of antimonopoly laws in capitalistic societies, of effectively using monetary and fiscal policies to combat inflation and unemployment, and of designing and administering policies to stimulate economic growth, such predictive laws are obviously essential. Oligopoly, in its variety of forms, is and will remain a significant sector of private enterprise.
Jesse W. Markham
[See alsoGame Theory; Industrial Concentration; Monopoly; and the biography ofCournot.]
Adelman, Morris A. 1961 Steel, Administered Prices and Inflation. Quarterly Journal of Economics 75:16-40. → A comment by G. Macesich and a reply by Adelman appear on pages 498-500.
Baumol, William J. 1958 On the Theory of Oligopoly.Economica New Series 25:187-198.
Baumol, William J. 1959 Business Behavior, Value and Growth. New York: Macmillan.
Bishop, Robert L. 1960 Duopoly: Collusion or Warfare? American Economic Review 50:933-961.
Chamberlin, Edward H. (1933) 1956 The Theory of Monopolistic Competition: A Re-orientation of the Theory of Value. 7th ed. Harvard Economic Studies, Vol. 38. Cambridge, Mass.: Harvard Univ. Press.
Chamberlin, Edward H. 1957 On the Origin of “Oligopoly. “ Economic Journal 67:211-218.
Clark, John M. 1940 Toward a Concept of Workable Competition. American Economic Review 30:241-256.
Cournot, Antoine Augustin (1838) 1960 Researches Into the Mathematical Principles of the Theory of Wealth. New York: Kelley.→ First published in French.
Cyert, Richard M.; and March, James G. 1955 Organizational Structure and Pricing Behavior in an Oligopolistic Market. American Economic Review 45: 129-139.
Fellner, William J. 1949 Competition Among the Few: Oligopoly and Similar Market Structures. New York: Knopf.
Galbraith, John Kenneth 1952 A Theory of Price Control. Cambridge, Mass.: Harvard Univ. Press.
Galbraith, John Kenneth 1957 Market Structure and Stabilization Policy. Review of Economics and Statistics 39:124-133.
Hall, Robert L.; and Hitch, Charles J. 1939 Price Theory and Business Behaviour. Oxford Economic Papers No. 2:12-45.
Kaysen, Carl 1960 [A Book Review of] Strategy and Market Structure: Competition, Oligopoly, and the Theory of Games by Martin Shubik. American Economic Review 50:1039-1040.
Markham, Jesse W. 1951 The Nature and Significance of Price Leadership. American Economic Review 41: 891-905.
Markham, Jesse W. 1959 Changing Structure of the American Economy: Its Implications for Performance of Industrial Markets. Journal of Farm Economics 41:389-400.
Markham, Jesse W. 1964 Administered Prices and the Recent Inflation. Pages 145-175 in Commission on Money and Credit, Inflation, Growth, and Employment. Englewood Cliffs, N.J.: Prentice-Hall.
Nicholls, William H. 1949 The Tobacco Case of 1946. American Economic Review 39, no. 3:284-296.
Rosenbluth, Gideon 1955 Measures of Concentration.Pages 57-95 in Universities-National Bureau Committee for Economic Research, Business Concentration and Price Policy: A Conference. National Bureau of Economic Research, Special Conference Series, No. 5. Princeton Univ. Press.
Rothschild, K. W. 1947 Price Theory and Oligopoly.Economic Journal 57:299-320.
Shubik, Martin 1959 Strategy and Market Structure: Competition, Oligopoly, and the Theory of Games.New York: Wiley.
Simpson, George Gaylord (1949) 1961 The Meaning of Evolution: A Study of the History of Life and Its Significance for Man. New Haven: Yale Univ. Press.
Stigler, George J. 1947 The Kinky Oligopoly Demand Curve and Rigid Prices. Journal of Political Economy 55:432-449.
Stigler, George J. 1962 Administered Prices and Oligopolistic Inflation. Journal of Business 35:1-13.
Sweezy, Paul M. 1939 Demand Under Conditions of Oligopoly. Journal of Political Economy. 47:568-573.
Triffin, Robert 1940 Monopolistic Competition and General Equilibrium Theory. Harvard Economic Studies, Vol. 67. Cambridge, Mass.: Harvard Univ. Press.
Villard, Henry H. 1958 Competition, Oligopoly, and Research. Journal of Political Economy 66:483-497.
An oligopoly is an intermediate market structure between the extremes of perfect competition and monopoly. Oligopoly firms might compete (noncooperative oligopoly) or cooperate (cooperative oligopoly) in the marketplace. Whereas firms in an oligopoly are price makers, their control over the price is determined by the level of coordination among them. The distinguishing characteristic of an oligopoly is that there are a few mutually interdependent firms that produce either identical products (homogeneous oligopoly) or heterogeneous products (differentiated oligopoly).
Mutual interdependence means that firms realize the effects of their actions on rivals and the reactions such actions are likely to elicit. For instance, a mutually interdependent firm realizes that its price drops are more likely to be matched by rivals than its price increases. This implies that an oligopolist, especially in the case of a homogeneous oligopoly, will try to maintain current prices, since price changes in either direction can be harmful, or at least nonbeneficial. Consequently, there is a kink in the demand curve because there are asymmetric responses to a firm's price increases and to its price decreases. That is, rivals match price falls but not price increases. This leads to "sticky prices," such that prices in an oligopoly turn out to be more stable than those in monopoly or in competition; that is, they do not change every time costs change. On the flip side, the sticky-price explanation (formally, the kinked demand model of oligopoly) has the significant drawback of not doing a very good job of explaining how the initial price, which eventually turns out to be sticky, is determined.
Airline markets and automobile markets are prime examples of oligopolies. We see that as the new auto model year gets under way in the fall, one car manufacturer's reduced financing rates are quickly matched by the other firms because of recognized mutual interdependence. Airlines also match rivals' fares on competing routes.
In oligopolies, entry of new firms is difficult because of entry barriers. These entry barriers may be structural (natural), such as economies of scale, or artificial, such as limited licenses issued by government. Firms in an oligopoly, known as oligopolists, choose prices and output to maximize profits. However, firms could compete along other dimensions as well, such as advertising, location, research and development (R&D) and so forth. For instance, a firm's research or advertising strategies are influenced by what its rivals are doing. When one restaurant
advertises that it will accept rivals' coupons, others are compelled to follow suit.
The rivals' responses in an oligopoly can be modeled in the form of reaction functions. Sophisticated firms anticipating rivals' behavior might appear to act in concert (conscious parallelism) without any explicit agreement to do so. Such instances pose problems for antitrust regulators. Mutually interdependent firms have a tendency to form cartels, enabling them to coordinate price and quantity actions to increase profits. Besides facing legal obstacles, cartels are difficult to sustain because of free-rider problems. Shared monopolies are extreme cases of cartels that include all the firms in the industry.
Given that mutual interdependence can exist along many dimensions, there is no single model of oligopoly. Rather, there are numerous models based on different behavior, ranging from the naive Cournot models to more sophisticated models of game theory. An equilibrium concept that incorporates mutual interdependence was proposed by John Nash (1928–) and is referred to as Nash equilibrium. In a Nash equilibrium, firms' decisions (i.e., price-quantity choices) are their best responses, given what their rivals are doing. For example, McDonald's charges $2.99 for a Value Meal based on what Burger King and Wendy's are charging for a similar menu item. McDonald's would reconsider its pricing if its rivals were to change their prices.
The level of information that firms have has a major influence on their behavior in an oligopoly. For instance, when mutually interdependent firms have asymmetric information and are unable to make credible commitments regarding their behavior, a "prisoner's dilemma" type of situation arises where the Nash equilibrium might include choices that are suboptimal. For instance, individual firms in a cartel have an incentive to cheat on the previously agreed-upon price-output levels. Since cartel members have nonbinding commitments on limiting production levels and maintaining prices, this results in widespread cheating, which in turn leads to an eventual breakdown of the cartel. Therefore, while all firms in the cartel could benefit by cooperating, lack of credible commitments results in cheating being a Nash equilibrium strategy—a strategy that is suboptimal from the individual firm's standpoint.
Models of oligopoly could be static or dynamic depending upon whether firms take intertemporal decisions into account. Significant models of oligopoly include Cournot, Bertrand, and Stackelberg. Cournot oligopoly is the simplest model of oligopoly in that firms are assumed to be naive when they think that their actions will not generate any reaction from the rivals. In other words, according to the Cournot model, rival firms choose not to alter their production levels when one firm chooses a different output level. Cournot thus focuses on quantity competition rather than price competition. While the naive behavior suggested by Cournot might seem plausible in a static setting, it is hard to image real-world firms not learning from their mistakes over time. The Bertrand model's significant difference from the Cournot model is that it assumes that firms choose (set) prices rather than quantities. The Stackelberg model deals with the scenario in which there is a leader firm in the market whose actions are imitated by a number of follower firms. The leader is sophisticated in terms of taking into account rivals' reactions, while the followers are naïve, as in the Cournot model. The leader might emerge in a market because of a number of factors, such as historical precedence, size, reputation, innovation, information, and so forth. Examples of Stackelberg leadership include markets where one dominant firm dictates the terms, usually through price leadership. Under price leadership, the leader firm's pricing decisions are consistently followed by rival firms.
Since oligopolies come in various forms, the performance of such markets also varies a great deal. In general, the oligopoly price is below the monopoly price but above the competitive price. The oligopoly output, in turn, is larger than that of a monopolist but falls short of what a competitive market would supply. Some oligopoly markets are competitive, leading to few welfare distortions, while other oligopolies are monopolistic, resulting in deadweight losses. Furthermore, some oligopolies are more innovative than others. Whereas the price-quantity rankings of oligopoly vis-à-vis other markets are relatively well established, how oligopoly fares with regard to R&D and advertising is less clear.
see also Monopoly
Cournot, Antoine A. (1971). Researches into the Mathematical
Principles of the Theory of Wealth, 1838. New York: A.M. Kelley.
Friedman, James W. (1983). Oligopoly Theory. New York: Cambridge University Press.
Fudenberg, Drew, and Tirole, Jean. (1986). Dynamic Models of Oligopoly. New York: Harwood.
Goel, Rajeev K. (1999). Economic Models of Technological Change. Westport, CT: Quorum Books.
Shapiro, Carl. (1989). "Theories of Oligopoly Behavior." In Richard Schmalensee and Robert D. Willig (Eds.), Handbook of Industrial Organization, vol. 1, New York: North-Holland.
Rajeev K. Goel
What It Means
The word oligopoly means “few sellers.” In an oligopoly a small number of companies control the majority of the output, or the market, for a good or service. The few companies (which are usually large in size) produce goods that are slightly or somewhat different from each other.
An oligopoly is similar to a monopoly, in which one company alone exerts control over a market. An oligopolistic industry is more concentrated than a competitive industry (that is, a small number of firms operate within the oligopolistic market), but it is less concentrated than a monopoly, in which only one firm dominates the market. A duopoly is an oligopoly in which only two firms control a market.
In an oligopoly there are so few firms that the actions of any one of them will very likely affect the price of the good and therefore have an impact on other firms in the oligopoly. For example, the majority of automobiles in the United States are manufactured by three automakers: Ford, General Motors, and Chrysler. If Ford were to lower its prices, more consumers would buy Fords, and the other two firms would experience a drop in sales. To counteract this, the other firms would most likely cut their prices as well, which would in turn impact Ford. These automakers, like firms in all oligopolies, are able to set their own pricing policies exclusive of market forces (the interaction of supply and demand that affects a market economy).
When Did It Begin?
During the Industrial Revolution the production of goods and services and the competition between companies increased, as did the formation of both monopolies and oligopolies. In the 1900s several large companies dominating the U.S. automobile and steel industries were the first oligopolies. These oligopolies attracted the close scrutiny of the U.S. government, and, over time, the U.S. government made some forms of oligopoly illegal under laws known as antitrust laws. The objective of these laws is to encourage greater competition between firms by restricting trade practices that are considered to be unfair. For instance, antitrust laws enforce pricing that is determined by market forces rather than by price setting among a small and controlling group of corporations.
Large corporations have come to dominate the business world in the twentieth century and the early years of the twenty-first century. Many corporations are the result of the merging, or joining, of several smaller companies. Fewer, larger corporations exert much of their power by controlling the supply of a particular good, much as the large automakers and petroleum companies do.
More Detailed Information
There are two basic types of oligopoly. A perfect (or homogeneous) oligopoly is one in which all firms in the industry produce an identical good or service. For example, the oil, milk, coal, copper wire, and cement industries all produce goods that are identical or nearly identical within their respective industries. The breakfast cereal industry is an example of a so-called imperfect oligopoly, or one in which each firm’s product has different characteristics but is essentially similar to the others. Imperfect oligopolies are sometimes called heterogeneous or differentiated oligopolies.
An oligopoly may be international, as in the case of automobile manufacturing, in which six automakers from the United States and Japan (General Motors, Ford, Chrysler, Honda, Toyota, and Nissan) control a large majority of U.S. automobile sales. The breakfast cereal oligopoly is a national, U.S. oligopoly. Some daily newspapers in small U.S. cities operate in an oligopoly.
Because there are so few firms in an oligopoly and they operate in highly interactive and interdependent ways, each firm will know the percentage of the total market, called the market share, for the good or service it produces. The decisions of one firm influence and are influenced by the decisions of other firms, so that any alteration in price or percentage of market share on the part of one firm will affect the other firms. Once prices are established by firms in an oligopoly, they tend to remain fixed, since each firm is aware of the consequences for other firms when it makes price and production decisions.
One of the most distinctive traits of oligopolies is the restrictions they place on the entry of firms not established in the oligopoly group, as well as restrictions on the exit of firms already in the group. An outside firm attempting to gain entry into an oligopoly must have money to pay for the high costs of starting up and conducting the research activities necessary to becoming competitive. It also must have resources to advertise and time to build the loyalty of customers. Often the significant financial and technical resources needed to gain entry into an established group of interdependent firms prohibit any other firms from joining.
The collective effort to prevent new firms from joining is a key strategy of oligopolistic competition. Within an oligopoly, firms also devise ways to outmaneuver one another in order to gain a larger portion of market share. For example, while it may be that no single firm in an oligopoly is willing to upset the price equilibrium among all the firms by raising or lowering its prices, a firm might try to distinguish itself from the others by signaling to customers that the quality of its product is superior. One way for a firm to demonstrate confidence in the quality of its product is to offer customers an extended warranty (a type of guarantee). Within an oligopoly, a firm might also vie for a dominant position by concealing from other firms its efforts to innovate in hopes of taking the market by surprise with a new or improved product. Although illegal, some firms in oligopolies have bargained with one another, agreeing to keep prices high or to stay out of each other’s most profitable geographical areas. The field of game theory in economics centers around efforts to understand how individual firms engage in strategic competition within an oligopolistic, or interdependent, situation.
In the last part of the twentieth century and the early years of the twenty-first century the number and power of oligopolies have increased because of corporate globalization, or the growth of transnational companies (companies that operate and invest in many countries around the world). Transnational corporations dominate the global economy, and many of them are parts of oligopolies.
In addition to globalization, increased competition has caused the growth of larger and more powerful corporate concentrations in such industries as food, drug, and agriculture. For example, the world’s top 10 seed companies currently control roughly one-third of the seed industry. In the United States oligopolistic industries also include the music-recording industry and the aerospace industry, in which Boeing and Airbus form a duopoly to control the market for large passenger aircraft manufacturing. Within the heavily regulated wireless communications industry, many U.S. states will license only two or three providers of cellular phone services, and these companies form local oligopolies.
When firms in an oligopoly grow intensely competitive with one another, different outcomes will occur. The firms in the group may agree to come together and agree as a group, or collude, to raise prices and restrict production. If the firms involved come to a formal agreement in this way, they are called a cartel.
Oligopoly is a type of market structure where a few large suppliers dominate an industry. Oligopolies are neither purely competitive (with many producers) nor monopolistic (with one producer), but fall somewhere in between. In the United States there are plenty of examples of oligopolist industries (in which only a small number of firms dominate). Heading the list are the following industries: automobile, steel, rubber, copper, aluminum, tobacco, and breakfast cereal. Oligopolist industry products may be differentiated, as with cereals, or homogeneous, as with steel.
In some oligopolies one firm tends to be the price leader, such as U.S. Steel Company in the steel industry. In others, all firms may have similar power. It is generally very difficult for a new firm to break into an oligopoly market because of the huge initial investments required. When only a few large firms dominate the market they cannot act independently without causing a change in output, sales, and prices in the industry as a whole. Where only a few large competitive rivals exist, their interdependence is recognized. They strategize to better anticipate each other's decisions. Competition among oligopolies usually does not include price wars (which would damage potential for generating profit). While price fixing is outlawed, oligopolistic firms keep their prices close together and instead compete with advertising and product variation. Advertising is a means to gain a competitive advantage. By making the product's availability better known, demand for the product is increased along with sales. Product variation results in continuous expansion and redefinition of products. Oligopolists are always seeking the best-selling new model. The automobile industry is a classic example.
The United States economy has traditionally consisted of four industry types: purely competitive, oligopolies, monopolistic competitors, and monopolies. Monopolistic competitors compete for essentially the same customers with slightly differing products. Between the four industry types, in 1939 oligopolies contributed a 36 percent share of the national income while purely competitive firms accounted for 52 percent. By 1958 oligopolies still contributed a 35 percent share but by 1980 they accounted for only an 18 percent share of the national income with purely competitive firms accounting for a 76 percent share. The U.S. economy became more competitive over the years due to increased competition from imports, deregulation, and enforcement of antimonopoly laws.