Machlup’s presidential address is a methodological reassessment of the postwar “marginalism controversy” over profit maximization, marginal productivity, and the adequacy of neoclassical price theory. Returning to the 1946–47 American Economic Review debate, he asks not whether one theory of the firm is realistic in all respects, but what each theory is meant to explain.
The chief issue, of course, was whether marginal analysis was invalid and ought to be discarded, especially as far as the theory of prices, cost, wages, and employment in manufacturing industry is concerned.
The essay’s governing distinction is between the firm as an analytical construct and the firm as an actual business organization. In competitive price theory, “the firm” is not a miniature sociology of General Motors, nor a description of managers’ psychology. It is a deliberately simplified agent used to infer how prices, outputs, and inputs change when wages, taxes, interest rates, technology, or demand conditions change. Machlup’s defense of marginalism therefore rests on fitness for purpose: the abstraction is legitimate when it answers the question for which it was designed.
To confuse the firm as a theoretical construct with the firm as an empirical concept, that is, to confuse a heuristic fiction with a real organization like General Motors or Atlantic & Pacific, is to commit the “fallacy of misplaced concreteness.”
This does not make Machlup hostile to behavioral or managerial theories. He accepts that real executives may seek prestige, security, leisure, staff, sales growth, discretionary expenditures, or social responsibility. His claim is narrower: under effective competition, such motives usually do not overturn the qualitative predictions of price theory. A competitive firm need not be psychologically realistic if the model’s purpose is to trace market-level adjustment.
Machlup’s automobile-driver analogy clarifies the point. A theory may predict aggregate consequences without reproducing each actor’s deliberations. Drivers differ in temperament and information, yet traffic outcomes can still be modeled for some purposes. Likewise, marginalist price theory does not require an exhaustive account of corporate administration when it is used to predict the direction of change after a specified shock.
I repeat: In the theory of competitive price the “real existence” of firms is irrelevant; imaginary (postulated) agents pursuing a simple (postulated) goal react to assumed changes in conditions and thereby produce (or allow us to infer) changes in prices, inputs, and outputs [24, pp. 13-14].
The managerial theories of Baumol, Marris, Williamson, and others are accordingly treated not as refutations of marginalism but as extensions or alternatives for different problems. In monopoly and oligopoly, where discretion is greater and internal organization matters, sales maximization, growth maximization, and expense-preference models can be useful. When they specify objective functions and constraints, they often remain recognizably marginalist in form.
Thus, instead of a heated contest between marginalism and managerialism in the theory of the firm, a marriage between the two has come about.
Machlup also resists the claim that imperfect information destroys marginal analysis. The theoretical firm need not know everything; it must possess only the information assumed relevant to the model’s explanatory task. Behavioral work on satisficing, search, and organizational procedure is valuable when the object is the concrete decision-making organization, but it is not automatically a superior substitute for price theory.
The essay culminates in a pluralist taxonomy: the firm may be a competitive reactor, innovator, welfare-theoretic unit, oligopolistic strategist, administrative organization, accounting entity, legal person, or census unit. Many controversies arise because writers slide among these meanings. Machlup’s enduring contribution is thus a disciplined methodological pluralism: abstract marginalism for competitive price adjustment, richer managerial marginalism for discretionary firms, and behavioral analysis for internal organization and advice. His defense of marginalism is conditional rather than dogmatic; models survive when judged by the questions they are built to answer.
This work was divided into 19 sections when it entered the library's research corpus—an apparatus for search and citation, not necessarily the author's own table of contents. Each title opens its summary.
Put a question to this work; the Librarian answers from its 19 sections and cites the passage.
Ask the Librarian