The Controversy on Marginalism in the Theory of the Firm and Markets

10.3.1. Critiques of the neoclassical theory of the firm

The position reached by traditional neoclassical theory in regard to the nature and objectives of the capitalist firm posed several theoretical and interpretational problems.

The traditional neoclassical view of the firm is based on three pillars. The first is the theory of perfect competition, a nucleus around which the analysis of other market forms has developed by difference. In such a context the firm is ‘a neuron of invisible hand’ (Marris, ‘Managerialism’, 1989, p. 4); and neurons, as is well known, never grow. The second pillar is the argument that the long run is nothing more than the summation of many short periods, so that the firm maximizes profit in the long run if and only if it manages to maximize it period by period. The third is the conception of the firm as a ‘technological black box’ which produces an output by combining inputs bought on the market with the specific resources of the firm. Thus, the problem of the firm’s economic performance is treated as one of an optimal combination of factors, while ignoring every organizational and institutional problem. Yet the internal structure of a firm is not independent of the structure of the market in which it operates.

With such premisses, traditional theory has developed a model of perfect competition that completely excludes any serious consideration of the dynamics and life of the firm. And in fact, this model, while having a great deal to say about the price system, has little to say about the process of competition among firms and their internal organization.

Yet this is not the kind of firm we observe in reality, in which a great variety of hierarchical and ownership structures can be detected. In fact, as early as 1932 two American scholars, Adolf Berle and Gardiner Means, published The Modern Corporation and Private Property, a book destined to become the reference point for all the modern theories of the firm. It was here that the concept of the separation of ownership and control was expressed and rigorously demonstrated: it is not capitalist owners but professional managers who control the large-scale firms; therefore power without owner­ship can exist.

In the same period, two English scholars, Robert Hall and Charles Hitch, published an essay entitled ‘Price Theory and Business Behaviour’ (1939). Their research was carried out in Oxford, and investigated the decision­making process followed by firms when faced with specific government measures.

Although formulated from empirical research and without aiming at pro­posing any alternative theoretical approach, Hall and Hitch’s conclusions represented a serious criticism of marginalism, the first, in fact, to come from empirical research; Sraffa’s 1925 and 1926 articles contained only a theoretical attack. Then, in 1939, Paul M. Sweezy published ‘Demand under Conditions of Oligopoly’, in which, by taking up the results obtained by Richard J. Kahn in ‘The Problem of Duopoly’ (1937), he formulated the famous ‘kinked-demand’ model in order to explain why prices in oligopolist markets tend to remain rigid and not to move in the ways predicted by traditional marginalist theory.

Various streams of research formed in the wake of this work in the 1940s and 1950s, and, although characterized by different points of view and intentions and having in common only the rejection of traditional neo­classical theory, they all aimed at giving theoretical depth to the empirical results of the Oxford Group. The most important results of the ‘marginalist controversy’ are two groups of theories: the post-Keynesian theories and the managerial-behavioural theories.

Then, from the late 1960s, the economic theory of the firm discovered problems which modified its content. Two main lines of research have emerged. The first has dealt with the nature of the firm as an organization, in particular, the question as to why the firm exists. In fact, if market transactions are the most efficient way of organizing economic activity, why do most economic activities take place within firms?—a question that Ronald H. Coase had already raised in 1937 in his pioneering article ‘The Nature of the Firm’.

10.3.2. Post-Keynesian theories of the firm

Even the scholars who accepted the assumption of profit-maximization had pointed out that the objective of long-run maximization does not imply, in itself, equality between short-run marginal costs and marginal revenues. Only if the decisions taken in each period are independent of each other will short-run profit-maximization also lead to long-run maximization. The discussion on long-run profit-maximization began in Oxford. Three issues of Oxford Economic Papers (1954, 1955, and 1956) contained the principal works on the subject: those by J. Hicks, P. Streeten, R. Harrod, F. H. Hahn, and H. R. S. Edwards.

The reasons for the abandonment of the marginalist principle of equality between marginal cost and marginal revenue are as follows. First, the firms do not precisely know their own demand curve; therefore, the marginalist rule cannot be applied owing to lack of information. Second, the main concern of a firm is price and not the quantity to produce; the firm sets its price on the basis of a certain criterion, and sells at that price whatever quantity the market can absorb. Finally, the application of the marginalist rule implies strong price variability, in the sense that any variation, even small, in the conditions of cost or demand causes a variation in price. But this is clearly contradicted by empirical evidence, which shows that prices of manufacturing goods tend to be rigid, despite the variations in demand and costs.

Of special importance is the case of oligopolistic markets, where the firms have both common and conflicting interests. In common they have an interest in the expansion of the sector in which they operate, since, once respective market shares have been specified, the profit of each firm grows with the width of the sector. On the other hand, it is precisely in the determination of market shares that conflict occurs. Now, among the various instruments of conflict, price competition is certainly the most dangerous. A price reduction on the part of a firm in the attempt to increase its own market share will lead to immediate retaliation, with negative effects on the level of profits of the whole sector.

The specific way in which non-price competition occurs depends on the characteristics and the history of the industry, the type of goods produced, the laws in force, and the general state of the economic system. In other words, the understanding of the specific forms of struggle among rival firms in oligopolistic markets cannot avoid explicit consideration of the institutional context. This means that it is impossible to understand the behaviour of the oligopolist firm only on the basis of the criterion of profit maximization—an argument that Nicholas Kaldor had put forward as early as 1934 in ‘The Equilibrium of the Firm’.

Large firms possess discretional market power because of phenomena such as vertical integration, product diversification, and oligopolistic co-ordination of the market. On the other hand, realities such as concen­tration, entry and exit barriers, and collusions are in contrast with the competitive mechanism conceived by neoclassical theory, as they imply that prices do not respond only to the differences in supply and demand.

Different answers have been given by different authors. The theory of entry barriers was invented by P. Sylos-Labini and J. Bain, who wrote, respectively, Oligopolio e progresso technico (1957) and Barriers to New Competition (1956). In this approach the level of mark-up is determined by the necessity of preventing the entry of potential competitors into the market. This is the limit-price theory, according to which the mark-up depends on factors such as the degree of industrial concentration, scale economies, product differentiation, and cost advantage of existing firms over potential competitors.

Post-Keynesian literature has stressed other factors as determinants of the mark-up. Attention, particularly during the 1970s, focused on the supply and demand for the finance necessary for investments; A. S. Eichner, the author of The Megacorp and Oligopoly (1976), has argued that, as it is self-finance, rather than external finance, that determines the growth of a firm in the long run, the mark-up will be calculated with the aim of expanding the firm. The idea of using investment to determine prices has enabled Eichner to lay down the bases for a new micro foundation of macroeconomic dynamics.

The post-Keynesian approach has, in recent years, found support in the work of the economic historian Alfred Chandler. In The Visible Hand: The Managerial Revolution in the American Economy (1977), he pointed out that the practice of mark-up pricing has been introduced by large firms as a technique of financial control over the huge amounts of fixed capital they have invested. By the end of the nineteenth century, large firms had already resorted to vertical integration and product diversification, above all in the areas producing primary goods. From the 1920s onwards, successive developments in this direction resulted in the diffusion of ‘multi-plant’ and ‘multi-product’ activities. In this context, mark-up pricing became the instrument with which large firms tried to decentralize their productive units into divisions or subdivisions: the price of each product must be enough to ensure a rate of return on the capital invested in the division producing that good, a rate which must be in line with the average rate of return on the total capital invested by the firm. In this way, according to Chandler, the U-form model has been supplanted by the M-form. Finally, it is worth mentioning the fact that the first historical evidence about the practices of price formation according to the full-cost rule goes back to 1878, the year in which a Manchester accountant, Thomas Battersby, published a paper on the pricing procedures followed by English firms of that period.

10.3.3. Managerial and behavioural theories

Another line of attack against traditional neoclassical theory has come from the observation of the decline of the family firm and the emergence of the limited company in modern capitalism. This has led to two important consequences. First, the control of the firm must be entrusted to professional managers. Second, since the optimal portfolio of each investor tends to be diversified among the shares of various companies, the shareholder can no longer be well informed about the events occurring in a single firm. This means that the managers of the large firms have the power to lay down company policy without any supervision by the shareholders. In an important work, The Theory of the Growth of the Firm (1959), E. Penrose pointed out that the firm, using internal and external resources, can grow up to a certain stage without any substantial barriers to expansion. In other words, it enjoys ‘dynamic’ economies of scale which can be realized through the acquisition of other firms or through market diversification. There is, however, a ‘growth curve’: beyond a certain stage, organizational costs rise and the growth rate of the firm begins to decline.

Penrose’s work was an important turning point in the theory of the firm. It replaced the traditional neoclassical viewpoint by a conception of the large firm as a pool of resources organized by the managers. This conception of the firm has initiated two distinct lines of research, the managerial and the behaviourial. Even though based on the same presuppositions and sharing the same ultimate goal of accounting for the ‘managerial revolution’, the two approaches diverge, especially in the area of methodology.

The first line of research, associated with the work of Baumol, Marris, and Williamson, maintains the principle of constrained maximization as a guarantee of the rationality of choices. The novelty, with respect to the traditional approach, is the specification of the target function. In the work of William Baumol (Business Behaviour, Value and Growth, 1959), for example, it is the growth rate of the firm which is maximized. But also, the influence of stock markets was explicitly taken into consideration. Robin Marris, in The Economic Theory of Managerial Capitalism (1964), also insisted on this point. The stock exchange is a source of finance and allows for continuous evaluation of the firm by means of the share values. On the other hand, as was pointed out by O. Williamson in Managerial Discretion and Business Behaviour (1964), managers do preserve a certain margin of discretion, and use it to pursue company policies which maximize their own utility. In this context, profit only acts as a constraint resulting from the need to remunerate the shareholders so as to avoid a fall in share values. The theory of takeovers developed by Marris accounts for one of the main ways in which capital markets can influence managerial behaviour: they produce a control effect. The threat of takeovers serves to discipline managers: the failure to maximize profits reduces the value of the firm on the stock market; and this may induce investors and external entrepreneurs to buy it and replace the management. The first formal study of the disciplinary role of takeovers was undertaken by S. G. Grossman and O. Hart in ‘Takeover Bids, the Free-Rider Problem and the Theory of the Corporation’ (1980), while C. M. Jensen and W. H. Meckling, with ‘The Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure’ (1976) were the first to analyse the important role played by ownership structures in determining the effectiveness of takeovers.

The second line of research, linked to the work of Herbert Simon, the pioneer of the modern theory of organizations, is characterized by the adoption of a different rationality criteria. The modern firm, Simon observed in Administrative Behaviour (1957), operates in a context of uncertainty and in a complex world where information and its speed of diffusion constantly increase. Its organization is the means of coping with these difficulties. More precisely, at the basis of Simon’s work is the argument that the modern firm is not a well-defined ‘individual entity’ but rather an ‘organization’, a set of individuals and power centres. The decisions, in this context, result from interactions and compromises between the various centres. Therefore, only an adequate study of the interactions among the internal components of the firm would allow a definition of the firm’s objectives.

Such a study includes various aspects. First, the firm, as a system, is made up of individuals, but these act as ‘role personifications’ interacting among each other, or, rather, as elements of an information network. Second, the reality of modern business cannot be explained by assuming that the only reason for economic action is egoistical. Simon observed that self-interest is often held in check by identification mechanisms that operate cognitively and motivationally. The spirit of identification is responsible for loyalty to the organisation and acceptance of its objectives, and plays a decisive role in the success of a business. Third, if the firm is an ‘organism’ made up of various power centres, each with its own specific objective, co-ordination and control are needed to reach univocal decisions. The control theory of complex sys­tems states that these are based on the principle of ‘homeostasis’, which is the ability of an organism to maintain its own structure through time, whatever the environmental variations and stimuli may be. As a ‘homeostatic organism’, the firm assumes forms that enable it to generate self-regulation processes capable of reacting to external changes so as to re-establish its internal equilibrium.

The idea is that the basic objective of a firm is ‘survival’. On the other hand, firms aim at a ‘satisfactory’, rather than optimal, solutions; they seek solutions good enough for all the internal groups; and, as survival is linked to profit, they aim at attaining a satisfactory level of profit. This is a typical example of what Simon has called satisficing behaviour, a concept justified by the observation that firms, in the same way as any agent, act on the basis of bounded rationality. Only in a stationary state will the differences in results between maximizing and satisficing behaviour tend to disappear, since, in such a situation, the profits which satisfy the firm will correspond, in the long run, to its opportunities to make the maximum profit. However, this is not the case in situations that change continually, and in which knowledge is incomplete. Furthermore, the behaviour of the firm on the market is no longer guided, according to Simon, by so-called substantive rationality, but by a procedural rationality, which consists in attributing rationality to the procedures of behaviour as guidelines for determining economic choices. The now popular notion of procedural rationality was presented by Simon in ‘From Substantive to Procedural Rationality’ (1976).

Simon’s ideas gave rise to the behaviourial approach to the theory of the firm, of which we will mention only the book by R. Cyert and J. March: A Behavioural Theory of the Firm (1963). The central aim of the work of these authors was the study of the decision-making process of large firms with multiple production, operating in conditions of uncertainty and in an imperfect market. In particular, interest in organizational problems created by large scale companies has led Cyert and March to a conception of the firm, not as a single decision-making unit with a unique objective, but as a multi-decisional organization with many objectives. This is because a firm is a coalition of stakeholders who are dependent on its activity in different ways and pursue different objectives. Thus, for example, the managers’ aim is to increase their power and prestige, the shareholders’ is to achieve the highest possible dividends, the workers’ to earn higher salaries and have better working conditions. For this reason, the complex structure of a firm cannot be likened to a perfectly programmable and controllable machine, as the marginalist theory would have us believe.

10.3.4. The neoclassical reaction and the new theories of the firm

Four lines of defence have been raised in support of marginalism during this controversy, the most strenuous of which was that of the Chicago School.

Milton Friedman has argued that the traditional theory of the firm has produced good and reasonable predictions, and that therefore it should be judged in a positive way. Such an argument derives from Friedman’s methodological view that the realism of assumptions is fairly irrelevant. By Friedman’s conventionalism, the only thing that really counts in evaluating a theory is the predictions that it enables one to make. Basically, the entre­preneur behaves, according to Friedman, as an expert billiards player who manages to pot the ball by putting the right angle and speed on the cue ball but knows nothing of the laws of physics or the rules of geometry. Of course, this line of defence is very weak. Who guarantees that the laws the entre­preneur unconsciously follows are precisely those invented by the marginalist economists?

Marginalism’s second line of defence has been to seek support in empirical research on the so-called ‘auxiliary hypotheses’, those on the form of demand and cost curves and on the time horizon of decisions (short or long run?). J. Earley, for instance, on the ground of a sample of 110 American com­panies, has found that modern methods of accountancy are able to supply information about marginal costs and revenues, and that this information is in fact used by the companies. This empirical investigation has reached conclusions completely opposite to those reached in the studies following in the wake of Hall and Hitch.

Other authors, such as A. Alchian, who wrote ‘Uncertainty, Evolution and Economic Theory’ (1950), have resorted to the Darwinian principle of survival of the fittest, to conclude that the strongest firms, which survive in competition and remain on the market, are those which aim at profit maximization. The economic environment selects the firms that maximize profit and eliminate all the others. Thus, the economist is justified, according to Alchian, in postulating that the best hypothesis about the firm’s behaviour is that of profit maximization.

Finally, a fourth line of defence claims that the hypotheses of the tradi­tional theory of the firm are, by and large, realistic. In ‘Marginal Analysis and Empirical Research’ (1946), F. Machlup pointed out that the empirical evidence against marginalism has too many shortcomings to be decisive. The declarations of businessmen that price is fixed at the level of average cost plus a mark-up does not represent any proof against the marginalist rule. This is so for the simple reason that the subjects interviewed, not knowing the language of economic theory, express themselves in an inadequate way. Moreover, there are psychological reasons why entrepreneurs should declare that the maximization of profit is not among their objectives; they wish to appear ‘honest’, and to show that their activities serve a social objective. The basic hypotheses of marginal theory, according to Machlup, are in the main plausible. It is true that the marginal cost and revenues are not known in an objective way by the firms; but this does not create a serious problem, given that a subjective evaluation of these curves is just as good.

In a successive work, ‘Theories of the Firm: Marginalist, Behavioural, Managerial’ (1967), Machlup made an attempt at reconciling marginalist, managerial, and behaviourial approaches. He believed that there is no radical conflict between the principle of full cost and marginalism, as that principle could be incorporated, as a special case, into marginalist theory. As long as the firm enjoys wide profit margins, there is room for the interests and wishes of the various groups which work in it, just as is theorized by the managerial and behavioural approaches. But when the wind of competition blows, and internal conflict reaches levels that erode the profit margins, to the point of threatening the survival of the firm itself, its behaviour must follow the familiar marginalist rules of maximization. The closing of the controversy in this way did not, however, prevent the vigorous resumption of interest, in the early 1970s, in the theory of full cost, nor did it prevent more recent criticisms of traditional theory from pursuing new and original paths.

One of these is the evolutionary approach to the theory of the firm, ini­tiated by the important work of R. Nelson and S. G. Winter, An Evolutionary Theory of Economic Change (1982), in which the behaviour of firms is explained in terms of adaptive mechanisms. These are well known in biology. In general, living organisms do not follow optimal paths. Optimal results can, perhaps, be reached, under special conditions, as asymptotic properties, but not as a direct consequence of the agents’ behaviour. The ‘memory’ of the firm is the basis of its behaviour. When the results are no longer satisfactory, the firm looks for new routines, either formulating them autonomously within the organization or imitating external ones. The evolutionary approach has recently become intertwined with the neo-Schumpeterian work on the firm and markets, for which it is worth mentioning N. Rosenberg, Inside the Black Box: Technology and Economics (1982).

Another interesting branch of research is the theory of industrial organ­ization. At the basis of this is the idea that the study of the firm and markets should be carried out in terms of optimal solutions in respect, not so much to the productive constraints, but rather to the contractual constraints between the interested parties. This approach considers the firm as a nexus of con­tracts in which a basic role is played by the transmission of information among the members of the organization. In fact, if productive specialization ensures an increase in the productivity of the employees, it will also give rise to a problem of co-ordinating the actions of the various members of the firm, many of whom possess different information sets. Moreover, interest groups are not limited to those represented by managers and the shareholders; workers, cadres, and other groups exist which have particular interests and which, because of the existence of asymmetric information, can pursue goals that are detrimental to the firm as a whole. Therefore, as J. Tirole has pointed out in The Theory of Industrial Organization (1988), it is necessary to investigate how it is possible to exploit information dispersed within the firm system by finding an equilibrium among the multiple centres of interest that operate inside the firm. This shows why one of the main analytical instru­ments used in the varied literature on industrial organization is the principal­agent model, introduced by S. Ross (‘Economic Theory of Agency: The Principal’s Problem’, American Economic Review, 1973) and by J. Mirrlees (‘Notes on Welfare Economics, Information and Uncertainty’, in Contri­butions to Economic Analysis, edited by M. S. Balch et al., 1974; and ‘Optimal Structure of Incentives and Authority within an Organization’, Bell Journal of Economics, 1976).

Finally, another research line is that of the ‘new industrial economics’, whose characteristic is—as has been declared by R. Schmalensee in The New Industrial Organization and the Economic Analysis of Modern Markets (1982)—the abandonment of the idea that the firm is an organization which adapts, more or less passively, to given conditions. This approach aims at going beyond the famous scheme of traditional industrial economics, based on the triad structure-behaviour-performance. Productive structures, market forms, and types of company organization are not simply the result of an efficient adjustment to an external order; and this is so because the firms are able to modify the environmental conditions by their behaviour. In other words, while traditional industrial economics sees the firm as a rational adapter under the selective processes of the market, the new industrial economics sees the firm as actively seeking domination strategies (such as the strategic creation of entry and exit barriers) aimed at increasing its market power.