The Economics of Welfare

Pigou, Arthur C.
Display paragraphs in this book containing:
First Pub. Date
London: Macmillan and Co.
Pub. Date
4th edition.
30 of 73

Part II, Chapter XV


§ 1. A CONDITION of monopolistic competition exists when each of two or more sellers supplies a considerable part of the market with which they are connected. In these circumstances it can be shown that there is no tendency for them to devote to the industry in which they are employed that amount of resources which I have called the ideal investment, namely, that amount which will make the value of the marginal social net product there equal to the central value of marginal social net products in general.*63 A demonstration of this proposition, possible differences between social and private net product of the type examined in Chapter IX. being ignored, can be given in ordinary language as follows.


§ 2. Let us first ignore all forms of action which aim, by sacrifice in the present, at obtaining an advantage against rivals in the future. We have, then, to do with the pure problem of "multiple monopoly." This problem assumes its simplest form when two monopolists only are supposed to be present; and, in this form, it has been much discussed among mathematical economists. Cournot decided, as is well known, that the resources devoted to production under duopoly are a determinate quantity, lying somewhere between the quantities that would have been so devoted under simple competition and under simple monopoly respectively. Edgeworth, on the other hand, in an elaborate critique, maintained that the quantity is indeterminate. In more recent discussions there is apparent some measure of return towards Cournot. If, it is held, each of two monopolists, in regulating his action, assumes that the other will not alter his output in consequence of what he does, the quantity of resources devoted to production by the two together is determinate at the amount calculated by Cournot. If each monopolist assumes that the other will not alter his price in consequence of what he does, then, in a perfect market, the quantity of resources devoted to production by the two together is determinate at the amount proper to simple competition; in an imperfect market—that is to say, a market in which some of the buyers have a preference for one monopolist over the other—this quantity is determinate at an amount less than that proper to simple competition, and falling short of it more largely the more imperfect the market is.*64 More generally, if each seller makes and holds to any definite assumption about the conduct of the other, it would seem that the quantity of resources devoted to production by both together is determinate at some amount not greater than that proper to simple competition, and not less, in a perfect market than the Cournot amount, in an imperfect market than the amount proper to simple monopoly. In real life, however, it is, I suggest, very unlikely that either seller will hold any consistent view about his rival's state of mind. His judgment will be variable and uncertain. As in a game of chess each player will act on some forecast of the other's reply; but the forecast he acts on may, according to his mood and his reading of that opponent's psychology, be one thing or another thing. Hence, as it seems to me, we may properly say that the aggregate amount of resources to be devoted to production is indeterminate in the sense that from a mere knowledge of the demand conditions and of the cost conditions affecting the two monopolists—whether the cost conditions are independent or inter-linked—we cannot foretell what it will be. The range of indeterminateness extends over a distance which is larger in a perfect than it is in an imperfect market, and in either sort of market is diminished as the number of monopolists is increased above two. In no case can the aggregate investment be greater than the quantity proper to simple competition. We have learned, however, from Chapter XI. that, except in industries in which imported materials subject to increasing supply price are employed, this last quantity cannot be greater than the ideal investment. It follows that, except in these industries, the investment which is forthcoming under multiple monopoly cannot be greater than the ideal investment. It may, however, easily be, and, in view of what has been said, in general seems likely to be, substantially less than this.


§ 3. Hitherto we have specifically excluded the effects of price warfare designed to secure future gains by driving a rival from the field or exacting favourable terms of agreement from him. The indeterminateness just described exists under monopolistic competition, even though neither of the monopolists "hopes to ruin the other by cut-throat prices."*65 In many instances of monopolistic competition, however, price warfare—or cut-throat competition—does, in fact, take place. It consists in the practice of selling at a loss in order to inflict injury on a rival. It must be distinguished carefully from the practice of reducing prices down to, or towards, prime cost, which frequently occurs in periods of depression. This latter practice may involve large reductions of price below the "normal," and it is certain to do this when demand is variable and prime cost is small relatively to supplementary cost; but it does not involve "selling at a loss" in the strict sense. Cut-throat competition proper occurs only when the sale price of any quantity of commodity stands below the short-period supply price of that quantity. When it occurs, the range of possible aggregate investment no longer has, as an upper limit, the quantity proper to simple competition, but is liable to exceed this quantity to an extent determined by the opinion entertained by each of the combatants about the staying power of his opponent and by other strategical considerations. There is, obviously, no tendency for it to approximate to the ideal investment; but we can no longer say, as we could when cut-throat competition was ignored, that it is likely to be less than the ideal investment.

Notes for this chapter

Cf. ante, pp. 223-4.
G. Hotelling, "Stability in Competition," Economic Journal, March 1929.
Edgeworth, Giornale degli economisti, November 1897, p. 405.

Part II, Chapter XVI

End of Notes

30 of 73

Return to top