Markets and Prices: Superseding the Theory of Perfect Competition

In his dissertation, reviewing the cost structure of the typical firm in the British cotton industry, Kahn argued that in the short run the average cost curve of the firm is not U-shaped, but is an inverted L, meaning that the average cost remains constant up to a point corresponding to the maximum obtainable level of production. Given this form of the average (and marginal) cost curve and the assumption of a horizontal demand curve in perfect competition, the only way to determine the level of production at which the firm maximizes profits is by postulating imperfect competition, or in other words that the firm is facing a downward-sloping demand curve. In this way Kahn explains why the degree at which productive capacity is exploited during a period of economic downturn falls below full capacity for all the existing firms. His account tallied with the empirical evidence, which revealed that all the firms in question worked full time for some days of the week and suspended production for the rest, but clashed with the theory of perfect competition where it stated that only efficient firms surviving the competitive struggle would be able to continue production, and at a level corresponding to full capacity, while the others would stop production altogether (Marcuzzo 1994).

The dissertation formulates two fundamental questions that arise once the hypothesis of perfect competition is abandoned: (1) If the firms are no longer subject to perfect competition but decide the price themselves, how will they take into account the behaviour (the reactions) of the other firms in the same industry?, and (2) Is the pursuit of maximum profit the best way to describe a firm’s behaviour? Three papers by Kahn, ‘Imperfect Competition and the Marginal Principle’ (Kahn 1932b), ‘The Problem of Duopoly’ (Kahn 1937), and ‘Review of Oxford Studies in the Price Mechanism edited by T. Wilson and

PW.S. Andrews’ (Kahn 1952a), represent successive stages in the evolution of his thought and the contemporary literature that developed on the subject. However, what they have in common is the same approach in answering these two questions.

As to the first question, in the Principles Marshall had addressed the problem of the determination of monopoly equilibrium, but he had not ventured very far into the effects of the behaviour of rival firms on a firm’s demand elasticity, or, indeed, into the effects of one firm’s price policy on the behaviour of rival firms. Despite the works by Cournot and Edgeworth on duopoly, at the time that Kahn’s dissertation was being written, a decisive solution to the problem had yet to be arrived at and integrated into general analysis of imperfect competition. There are two aspects of particular interest marking out the dissertation from the previous literature on oligopolistic markets. The first is Kahn’s invention of a way to measure the degree of market imperfection assuming linearity in the demand and average cost curves, through the so-called annihilation coefficient (Kahn 1989: 121), which is conceptually identical with that measurement that was subsequently to become known as the ‘degree of monopoly’.

The second interesting aspect is the approach taken to the effects on firms’ demand curves when one of the rival firms changes its price. A whole section of the dissertation is dedicated to this issue, and here Kahn also criticizes the conclusion Sraffa had arrived at in his 1926 article, to the effect that in an imperfect market the final equilibrium price is the price that would be arrived at if the market were entirely controlled by one single monopolist. Kahn argues his point on the basis of the idea that the slope of the individual demand curves reflects the conjectures that each firm makes about the behaviour of the others. Kahn demonstrates that whatever the hypothesis on the conjectures, the relation between the individual and market demand curves is such as never to produce equilibrium at the monopoly price, but at a lower price, and consequently with a greater quantity than would have been chosen by a monopolist in the same technical conditions (ibid.: 117).

Thus, we must reckon Kahn as among the inventors of the kinked demand curve, even if he defines his treatment of the problem as ‘primitive and incomplete’ (ibid.: xx). However, it had the merit of introducing ‘conjectures’ into demand curves, in the form of elasticity values, for analysis of price and quantity produced in a duopolistic or, more broadly speaking, oligopolistic markets. In ‘The Theory of Duopoly’, which became a classic on the subject, without any mathematical tools, Kahn goes through the whole complicated discussion on the effects of the actions of one duopolist on the behaviour of the other, and thus the role of the strategies implicit in every move of each firm, anticipating the later development in the theories of oligopoly.

As to the second question—the validity of the hypothesis of profit maximization—and in contrast to the traditional marginalist analysis, Kahn unfailingly considered it as the one certain rule upon which to build economic analysis, but interpreted the pursuit of maximum profit by firms as tantamount to a ‘trial and error’ method rather than as a manifestation of an optimizing rationality on the part of entrepreneurs. But, again marking himself out from approaches alternative to those based on marginal analysis, Kahn never accepted explanations of the price-formation mechanism of the descriptive type or based on a hypothesis of non-‘rational’ behaviour.

From the theoretical point of view, the leading advocate for the abandonment of the theory of perfect competition and of the hypothesis of profit maximization in favour of price-formation theory based on (fairly steady) mark-ups on (fairly steady) prime costs was Michai Kalecki, who formulated his theory subsequent to a study of the prices, proceeds, and costs of a number of British industries (Marcuzzo 2011: 189). Unlike Joan Robinson, who subsequently adopted it, Kahn remained averse to mark-up pricing, maintaining to the very end,

I think that the concept of the horizontal short period supply is exaggerated. I do not hold altogether with the ideas of Wynne Godley about the fixed write up margin. What determines the size of this margin? And of course it varies between firms in accordance with their differences in prime costs (Kahn quoted in Marris 1991: 184).

Similarly, and unlike Robinson, Kahn was never convinced by Sraffa’s long-period production prices, prompting Robinson to state bluntly: ‘Cannot we agree on Piero’s prices for the long run and on Keynes’s prices for the short run and leave it at that?’ (Robinson to Kahn, 19 May 1961, RFK 13/90/6/199-200).