Cost and Choice: An Inquiry in Economic Theory

James M. Buchanan.
Buchanan, James M.
(1919- )
Display paragraphs in this book containing:
First Pub. Date
Indianapolis, IN: Liberty Fund, Inc.
Pub. Date
Foreword by Hartmut Kliemt.

Chapter 2The Origins and Development of a London Tradition

Wicksteed and the Calculus of Choice


Wicksteed deserves recognition for having shifted cost theory away from its classical, objective foundations. Although Jevons can with justice be labeled as precursory, the major advance beyond Marshallian conceptions was made by Wicksteed, who tied opportunity cost quite directly to choice. He stated that cost-of-production, "in the sense of the historical and irrevocable fact that resources have been devoted to this or that special purpose, has no influence on the value of the thing produced."*14 In this respect, cost-of-production does not affect supply. What does affect supply is anticipated cost "in the sense of alternatives still open which must now be relinquished in order to produce this specific article";*15 this cost "influences the craftsman in determining whether he will produce it or not."*16 Here, the critical relationship between any measurement of cost and the act of choice is established. At any moment in time, one can look either forward or backward. One looks backward in time in a perspective of foreclosed alternatives. One looks forward in time in a perspective of alternatives that still remain open; choices can be and must be made. With this elementary clarification, cost tends to be a part of choice among alternatives, a choice that must be subjective to the chooser. Cost does not hold the direct relationship to commodity or resource units that it does in both the classical and neoclassical discussion.


Wicksteed fully recognized the many ambiguities that surrounded the usage of the term "cost," and he provided excellent examples.*17 But when all is said, Wicksteed, too, remained less than wholly clear. When put to it, he chose to define cost-of-production or cost-price as "the estimated value, measured in gold, of all the alternatives that have been sacrificed in order to place a unit of the commodity in question upon the market"*18 (italics supplied).


There is no doubt that Wicksteed was a major formative influence on the cost theory that emerged in the late 1920's and early 1930's at the London School of Economics. And, as I hope to show, traces of the "correct" theory of cost are found in the acute observations of Wicksteed. This may not emerge full-blown in Wicksteed for the simple reason that he need elaborate no further in answering his own questions. One feels that Wicksteed's cost theory, as that of Marshall, could have been made fully equal to the challenge presented by the new range of issues of the 1930's.

H. J. Davenport


Herbert J. Davenport was an American economist and, roughly, a contemporary of Wicksteed. His influence was limited to a relatively small group of followers, none of whom were major figures in the history of doctrine. Davenport's insights on opportunity cost, however, if read from the perspective developed in this book, suggest that it is appropriate to place his name between that of Wicksteed and Knight in this summary review.


Davenport's emphasis was on what he called "entrepreneur's cost," and he clearly defined this in a utility dimension. "That is to say, cost as a margin determinant is purely a matter within the personal aspects of entrepreneurship, a managerial fact, a subjective phenomenon, in which all the influences bearing on the psychology of choice between different occupations and leisure have their place."*19 Furthermore, Davenport explicitly recognized that cost is related to the particulars of the choice situation, and, indeed, his emphasis on entrepreneur's cost stemmed from his criticism of other writers, notably Marshall, who confused this with what Davenport called "collectivist cost."*20


Embedded in Davenport's treatise, Value and Distribution, is a conception of opportunity cost that is almost as sophisticated as that developed by Wicksteed. The failure of Davenport's ideas to have more influence than they did have was due, apparently, to his failure to articulate these ideas and also perhaps to his petulance toward the idols of the profession in his time. Davenport would have surely come into his own had he been able to criticize the more flagrant confusions in cost theory that emerged only after the 1920's.*21

Knight on Cost as Valuation


It is interesting that before he had written any of the papers previously cited, Frank Knight had explicitly referred to cost estimation as a valuation process inherent in choice itself. "[T]he cost of any value is simply the value that is given up when it is chosen; it is just the reaction or resistance to choice that makes it choice"*22 (italics supplied). Having made this tie-in between opportunity cost and the decision process, however, Knight tended to confuse the fundamental issues in his later emphasis on alternate-product value, a value determined, presumably, not by the chooser, but in the whole market process.

Robbins, 1934


In a basic paper published in 1934,*23 Lionel Robbins reacted against the emphasis by Knight and others on an alternate-product conception of opportunity cost, much as Knight himself was led to do in his 1934 and 1935 papers. In so doing, Robbins provided the basis for an opportunity-cost conception that later came to be identified with the London School of Economics. Neither Knight in his 1924 paper nor Robbins realized the importance of the distinction they were making, and Robbins considered himself to be merely clarifying certain ambiguities that had arisen in connection with the emerging Austrian orthodoxy, the source of which he attributed to Wieser. Specifically, Robbins argued that cost must be defined in terms of displaced value and not in terms of displaced real product. He demonstrated that once beyond the Smithian deer-beaver model, displaced real product has little meaning. His illustrative examples were those of final goods produced with wholly different inputs or with the same inputs but in differing and fixed coefficients. Shifts in demand generate shifts in cost under such conditions, and these cannot be interpreted in terms of displaced real-product alternatives. Costs are changed because the relative values of the inputs change, values derived from the demand for final products.


Although these clarifications were useful and represented the main thrust of Robbins' argument, they are not the subject of interest here. More or less as asides, Robbins introduced several statements that involve a different basic notion of cost. He apparently did not think of these statements in this light, perhaps because they were especially obvious to one who had fully learned his Wicksteed and perhaps because, in another sense, they were immaterial to his central theme. I refer to his explicit linking of cost to the act of choice itself. "The process of valuation is essentially a process of choice, and costs are the negative aspects of this process" (p. 2, italics supplied). "[I]t is the central requirement of any theory of cost that it should explain the actual resistances which production in any line of industry encounters" (p. 5, italics in original). "The isolated producer thinks of the sacrifice that he is making by not producing something else" (p. 5, italics supplied).


Unfortunately, after having interspersed these highly provocative remarks in his discussion, Robbins proceeded, almost simultaneously, to obscure their potential impact. In this, Robbins seems to have proceeded much as Knight had a decade before. On the page immediately following the last two statements cited above Robbins said: "The condition that prices shall be equal to cost of production in the value sense is as essential a condition of equilibrium in the Walrasian system as the condition that marginal products shall be proportionate to factor prices" (p. 6). The subtle but essential distinction between cost in the act of choice and cost in the predictive theory of economic behavior has disappeared in this apparently orthodox neoclassical statement.

Mises, Robbins, and Hayek on Calculation in a Socialist Economy


As I suggested in the Preface, latter-day Austrians can with some legitimacy claim that the concept of opportunity cost, here attributed to Jevons, Wicksteed, Davenport, Knight, and ultimately developing into a London tradition, was independently developed by later Austrians and notably by Ludwig von Mises. In his monumental, polemic, and much neglected treatise, Human Action,*24 Mises advances a theory of opportunity cost that is, indeed, almost equivalent to the full-blown LSE conception to be described later. Mises' explicit treatment of cost in Human Action will also be discussed later. At this point it is noted only that the German treatise that provides the basis for the English-language work was not published until 1940. For the period in question, therefore, Mises' earlier writings must be examined. In this connection, specific reference must be made to his 1920 paper, in which he argued that economic calculation in a socialist society is impossible,*25 and to his book which followed in 1922.*26


A modern reading of these early contributions by Mises suggests that some of the intuitive force of his argument stemmed from a more sophisticated conception of opportunity cost than he was able to make explicit at that time. Mises' attack on the possibility of socialist calculation is wholly consistent with the conception of opportunity cost that emerged more fully later, both at LSE and in his own writings. Although he did not center his early argument directly on the cost problem, per se, the general tenor of Mises' discussion is clearly and quite closely related to later developments in cost theory, and his contribution to that theory surely deserves recognition alongside those of Wicksteed and Knight. Quite apart from the importance of Mises' own works is his influence on the work of Lionel Robbins and F. A. Hayek, the transplanted Austrian who became one of the central figures of the LSE tradition.


In addition to writing the 1934 paper previously cited, Robbins also participated in the great debate over the possibility of socialist calculation.*27 An assessment of his contribution at this stage must be closely connected with the assessment made of Mises' work. Robbins' argument might well have been based on a more sophisticated notion of opportunity cost than that which is explicitly discussed, but one senses in a modern reading that, along with Mises and Hayek, he could have been much more effective if he had been able to make more articulate the distinction between objectively measurable cost and cost as an element of a decision process.


Hayek's specific contribution to the development of cost theory that is contained in his part of the debate on socialist calculation is a peculiarly mixed bag. In his "Introduction" to the famous collection of essays,*28 Hayek foreshadows his later and more explicit methodological emphasis on the necessity of distinguishing between the subjective apparent sense data of the person who chooses in the economic process and the objective data that are available to any external observer. As we shall see, this methodological step is essential to any genuine understanding of cost. It appears, however, that Hayek had not in 1935 incorporated this methodology fully into his own basic theory. In his essay, "The Present State of the Debate," included in the collection, he suggests clearly that cost of production becomes difficult to calculate in a socialist setting primarily because of the absence of the conditions of competitive equilibrium where "cost of production had indeed a very precise meaning."*29 This emphasis, which was also evident in Robbins' work, left the way open for Lerner's effective reply which argued simply for the adoption of a rule for setting prices at marginal opportunity costs, regardless of the state of the world.*30


My purpose here is not to evaluate the discussion of socialist calculation, but only to examine that discussion for the contributions it contains to the pure theory of opportunity cost. With the exception of Lerner (whose insight was much more profound, and who was himself a part of the developing LSE tradition), those who argued that socialist calculation is possible accepted an objective definition of cost without any serious critical examination of the issues that such definition might raise.

Hayek, Mises, and Subjectivist Economics


F. A. Hayek was appointed Tooke Professor of Economic Science and Statistics at the London School of Economics in 1931 and held this chair until 1950. Along with Robbins, he must be credited with providing the source of much of the LSE tradition in cost theory, a tradition that seems to have emerged gradually over these two decades. As suggested above, Hayek's contribution was primarily that of providing the underlying methodological basis for the more explicit works on cost by others. He presented the methodology of subjectivist economics with convincing authority; his essays remain recommended reading almost thirty years after their initial publication. And economic theory, generally, should certainly have avoided some major modern confusions if Hayek's essays had been more widely read and understood.


A distinction must be made between the orthodox neoclassical economics which incorporates the subjective-value or marginal-utility revolution in value theory and the subjectivist economics of the latter-day Austrians, notably Mises and Hayek. The dependence of price (value) on marginal utility, subjectively determined, can be fully recognized, while essentially an objective theory of cost is retained. In Jevons' famous statement, marginal utility depends on supply which, in its turn, depends on cost of production. As stated, this theory is wholly objectivist in character, although, of course, the valuation of buyers and sellers is incorporated as a part of the objective data. Costs are objectively determinable, although the theory does not say that costs alone determine value. As contrasted with classical theory, one-way causality is missing, but not the objectivity of the explanation. It is this latter objectivity that is jettisoned, wholly and completely, by both Mises and Hayek. In this respect, they differ sharply with earlier Austrians, although they do not seem fully to sense the distinction. In many respects, they seem much closer to Wicksteed than to Wieser.


There seems no doubt that subjectivist economics was explicitly introduced at LSE by Hayek. In a paper of major importance, published in 1937,*31 he laid down the central features of the subjectivist methodology, features that he elaborated in considerably more detail in later works.*32 In his 1937 paper, Hayek gives credit to Mises for the latter's background work,*33 published in German in 1933, but made available only much later (1960) to English-language scholars. Hayek's initial paper provides, in one sense, the "classical" methodology of the subjectivists, a methodology which is central to a theory of cost that is related directly to choice and that is to be contrasted with the theory of cost embodied in neoclassical orthodoxy.


The subtle distinction between the economics of subjective value and the subjectivist economics espoused by Hayek and Mises was quite naturally obscured so long as the task of economic theory was largely limited to the explanation of market interactions. The famous Jevons statement about supply serves as an illustration. So long as individual producers, responding to the demands of consumers, are the actors whose behavior we seek to explain, there is really no need of inquiring as to whether the costs of production are subjective or objective. Costs are obstacles to the choices of producers, and economists can discuss "laws of cost" in this context without presuming that objective measurement is possible.


With the advent of "welfare economics," regardless of how this might be defined, such previously admissible methodological fuzziness no longer passes muster. If idealized market interaction process—pure or perfect competition—is used as the standard for deriving conditions which are then to be employed as norms for interference with actual market process, the question of objective measurement must be squarely faced. If prices "should" be brought into equality with costs of production, as a policy norm, costs must be presumed objective, in the sense that they can be measured by others than the direct decision-maker.


Only Hayek and Mises seemed to be completely aware of this problem and of its major importance, although many other economists seem to have been vaguely disturbed by it. Subjectivist economics, for Hayek and Mises, amounts to an explicit denial of the objectivity of the data that informs economic choice. The acting subject, the chooser, selects certain preferred alternatives according to his own criteria, and in the absence of external change he attains economic equilibrium. This personalized or Crusoe equilibrium is, however, wholly different from that which describes the interactions among many actors, many choosers. In the latter case, the actions of all others become necessary data for the choices of the single decision-maker. Equilibrium is described not in terms of objectively determined "conditions" or relationships among specific magnitudes, e.g., prices and costs, but in terms of the realization of mutually reinforcing and consistent expectations. The difference between these two approaches, the objectivist and the subjectivist, is profound, but it continues to be slurred over in the neoclassical concentration on the idealized market interaction process in which all individuals behave economically. In an unchanging economic environment populated by purely economic men, the two approaches become identical in a superficial sense. In a universe where all behavior is not purely economic, where genuine choice takes place, the important differences emerge with clarity.


At this juncture in the development of economic theory, we must, I think, ask why the convincing arguments of Hayek exerted so little weight. Without question, objectivist economics continues to carry the day, and few of its practitioners pause to examine critically its methodological foundations. There are, no doubt, several reasons for this failure, but undue attention paid to the definition of equilibrium, although of immense importance in itself, may have retarded acceptance of the more general subjectivist notions. Neutral readers of the impassioned debates on socialist calculation might have been led to think that the central issue was really one that involved the possibly erroneous derivation of policy criteria from stationary equilibrium settings. Indeed this is an issue, but the subjectivist critique is obscured here. As noted earlier, this concentration on equilibrium, of which Hayek, Robbins, and to a lesser extent Mises, all are guilty, left the way open for Lerner to drop all references to general equilibrium in his derivation of the policy rules that explicitly require the introduction of objectively measurable costs.

The Practical Relevance of Opportunity Cost: Coase, 1938


Alongside the more abstract and methodological contributions to cost theory made by Robbins, Hayek, and Mises, other elements of perhaps a more authentic LSE tradition emerged in the 1930's. These reflect the direct application of some of the basic Wicksteed notions to problems that confront the businessman. This "common sense" approach had its roots at LSE in the work of Cannan who continually insisted on starting with problems as they exist. Cannan does not seem to have made specific contributions to cost theory, although he accepted opportunity-cost notions readily.*34


This practical-business approach was further promoted by Arnold Plant who seems to have contributed significantly, but indirectly, to the development of the London tradition. Plant did not, to my knowledge, treat cost theory explicitly in any of his published works, but the contributions made by his students and colleagues reflect his influence. Both R. H. Coase and G. F. Thirlby, whose contributions are summarized below, were Plant's students.


The contrast between the accountant's definition-measure of cost and that of the neoclassical economist is standard fare. But this contrast—when the full meaning of opportunity cost is incorporated—takes on features that are even now outside the orthodox economist's kit of tools. This can be seen clearly in a series of articles by R. H. Coase, published in 1938, which were written specifically for the enlightenment of accounting practitioners.*35 These papers remain known to relatively few modern economists despite their exceptionally clear discussion of definitional problems involved in using the term "cost" and their necessary and emphatic insistence that cost be related to the choice process.


"The first point that needs to be made and strongly emphasized is that attention must be concentrated on the variations which will result if a particular decision is taken, and the variations that are relevant to business decisions are those in costs and/or receipts" (p. 106). "It should be noted that accounting records merely disclose figures relating to past operations. Business decisions depend on estimates of the future" (p. 108). "[C]osts and receipts cannot be expressed unambiguously in money terms since courses of action may have advantages and disadvantages which are not monetary in character, because of the existence of uncertainty and because of differences in the point of time at which payments are made and receipts obtained" (p. 116). "The cost of doing anything consists of the receipts which could have been obtained if that particular decision had not been taken. When someone says that a particular course of action is 'not worth the cost,' this merely means that he prefers some other course—the receipts of the individual, whether monetary or nonmonetary does not matter, will be greater if he does not do it. This particular concept of costs would seem to be the only one which is of use in the solution of business problems, since it concentrates attention on the alternative courses of action which are open to the businessman. Costs will only be covered if he chooses, out of the various courses of action which seem open to him, that one which maximizes his profits. To cover costs and to maximize profits are essentially two ways of expressing the same phenomenon" (p. 123) (italics supplied).


A careful, modern reading of these early papers by Coase indicates that the concept of cost embodied in them is conceptually distinct from the neoclassical paradigm. Coase quite explicitly ties cost to choice, and he rejects any attempt to classify costs into categories—e.g., fixed and variable—independently of the identification of the decision under consideration. Perhaps the most significant contribution, for our purposes, is contained in the italicized statement cited above. Any profit opportunity that is within the realm of possibility but which is rejected becomes a cost of undertaking the preferred course of action. Despite the necessity of accepting this straightforward result of apparently consistent opportunity-cost reasoning, economists were—and are—extremely reluctant to take this step. To include all foregone profits as costs plays havoc with the whole cost-curve apparatus that is a part of our stock-in-trade. And without this how could we teach elementary price theory?


In the strict neoclassical model, costs are distinguished sharply from foregone profits because they are not tied directly to choice. Costs are objectively measurable outlays, approximated by the value of alternate product. It is useful to keep the classical foundations of the analysis in mind here. Costs, to the extent that they are objective and, hence, externally measurable by an outsider who stands apart from the choice process, provide the basis for a predictive hypothesis about the behavior of acting individuals (firms) and, through this, a hypothesis about prices. The neoclassical objectivist world and the London-Austrian subjectivist world cannot readily be reconciled.


Confusion was confounded by the Robinson-Chamberlin and related contributions in the early 1930's just when the more basic notions in cost theory seemed on the way to being clarified. These contributions elevated the theory of the firm to a position of undue importance in a model that apparently embodied the objectivist rather than the subjectivist notions of cost. If the purpose of analysis is to "explain" the behavior of the firm, choice must be the subject of attention, and costs cannot be objectified. The whole marginal-revenue-marginal-cost apparatus, strictly speaking, remains a part of a central logic of choice and nothing more because, to the effective decision-taker, both costs and benefits are evaluated in purely subjective terms. It is little wonder that modern developments in the theory of the firm have been concerned with relaxing the artificial and apparent objectivity of cost and revenue streams by substituting more plausible, even if largely nonoperational, utility indicators.


Coase's early work on the theory of the firm was within a choice-explanatory context and without the constraints of the more widely acclaimed contributions of the imperfect and monopolistic competition models. In this context, Coase was fully correct in his argument that foregone profits must be included in opportunity cost and in his insistence that cost be considered as that which can be avoided by not taking a particular decision.


Despite his major contribution toward clarifying the concept of opportunity cost in the context of the theory of the firm, Coase did not in his 1938 papers fully incorporate the "subjectivist economics" of Hayek and Mises into his analysis, nor did he draw the distinction between his concept and that embodied in neoclassical orthodoxy.*36

G. F. Thirlby and "The Ruler"


Academically, both Vienna and Capetown were close neighbors to London in the 1930's, and, as a consequence of the influence of transplanted LSE economists, the next major contribution to the theory of cost emerged in Capetown. Primarily under the influence of Arnold Plant and W. H. Hutt, an oral tradition developed at Capetown which expanded the London approach. The published results appeared in 1946 in two papers by G. F. Thirlby. In these papers, Thirlby, who had been trained at LSE and who returned to London a few years later, carried forward the process of clarification. He continued until 1960 his attempts to convert other economists to what he considered to be a more acceptable and consistent view of opportunity cost, but his argument seems to have been largely neglected.


In his first 1946 paper, Thirlby, like Coase, related the economist's notions on cost to those of the accountant.*37 Thirlby had fully incorporated the subjectivist economics of Wicksteed and the latter-day Austrians in his analysis, and his emphasis was on the subjectivity of costs. Citation at some length from this early paper seems warranted here:

To the subjectivist, cost would be understood to refer to the prospective opportunity displaced by the administrative decision to take one course of action rather than another. Cost is inevitably related to the behavior of a person. The person is faced with the possibility of taking one or other of (at least) two courses of action, but not both. He considers the relative significance to him of the two courses of action, and finds that one course is of higher significance than the other. He 'prefers' one course to the other. His prospective opportunity of taking the less-preferred course becomes the prospective cost of his taking the more preferred course. By deciding to take the preferred course, he incurs cost—he displaces the alternative opportunity. The cost is not the things—e.g., money—which will flow along certain channels as a result of the decision; it is the loss, prospective or realized, to the person making the decision, of the opportunity of using those things in the alternative courses of action. A fortiori, this cost cannot be discovered by another person who eventually watches and records the flow of those things along those channels. Cost is not something which is objectively discoverable in this manner; it is something which existed in the mind of the decision-maker before the flow began, and something which may quite likely have been but vaguely apprehended....
Cost is ephemeral. The cost involved in a particular decision loses its significance with the making of a decision because the decision displaces the alternative course of action (italics in original).*38


Thirlby's emphasis on the ephemeral nature of cost distinguishes his paper from earlier contributions in the LSE tradition. And in this early paper, Thirlby himself wavers in his adherence to this conception which his later writings were to stress. Note that he says "prospective or realized" at one point: he fails to see that the very notion of realized cost produces a contradiction, as he was later to demonstrate. Similarly, his reference to cost being "vaguely apprehended" implies that something different from that which is apprehended emerges at a later point in the decision-action sequence that might be called cost.


The extension of Thirlby's rigorous opportunity-cost reasoning to the question of the relevance and practicability of the so-called "rules" for pricing was straightforward, and this was the content of his second paper, "The Ruler."*39 Thirlby made it clear that he was relatively uninterested in the much debated "which rule" question, one that obscures the analysis of the "any rule" problem. As in his other paper, stress was placed on the fact that cost was not "an objective something in the sense that it can be scrutinized." The standard definitions and measurements were held to omit valuations of the "lost opportunities," and Thirlby argued that unless these were taken into account, no rule could ever be applied to ensure the satisfactory meeting of people's preferences.


He rejected the orthodox distinction between "long run" and "short run," and he was explicit in saying that "cost occurs only when decisions are made, that is, in planning stages" (p. 259). He clarified the distinction between what we might call the decision, budget, and accounting levels of calculation. Cost is relevant to decision, and it must reflect the value of foregone alternatives. A budget, however, reflects the prospective or anticipated revenue and outlay sides of a decision that has been made. It is erroneous to consider such prospective outlays as appear in a budget as costs. The budget must, however, also be distinguished from the account, which measures realized revenues and outlays that result from a particular course of action. This clarification is a simple one in itself, but it is highly useful for our purposes. It shows that the forward-looking or ex ante framework is not, in itself, sufficient to ensure the adoption of the appropriate cost concept. The budget is, by definition, a planning document, an ex ante projection of events; it does not, however, balance anticipated revenues against anticipated cost in the relevant opportunity-cost sense. The "cost" side of a budget measures anticipated outlays which are to be made as a result of a particular course of action's having already been chosen. It cannot reflect the value of alternative courses of action which might have been selected save in the exceptional case where no alternative revenues in excess of anticipated outlay could be secured.


There is, or should be, no difficulty in convincing critics that cost must be subjective at the moment of choice. But one can fully accept the subjectivist point of view in this respect and still think that, after decision, cost becomes objective and, hence, measurable. In his earlier paper, Thirlby might not have fully sensed the instantaneous vanishing of cost upon decision. In "The Ruler," however, this point is emphasized. "[T]he cost figure will never become objective; i.e., it will never be possible to check whether the forecast of the alternative revenue was correct, because the alternative undertaking will never come into existence to produce the actual alternative revenue" (p. 264).


Following these two 1946 papers, Thirlby continued to present his ideas on cost, for the most part in the context of a theory for business organization and decision. Although most of the central ideas were developed in the two early papers, some shifts of emphasis are worth noting. In a 1952 paper, Thirlby argued plausibly for a more widespread recognition of a time dimension in economic analysis, particularly with respect to the decision process. "[A] period of time elapses between the making of the decision and the achievement of the results.... A mental deliberation or planning operation, followed by a decision, precedes the business operations which are so planned." Recognition of this would "keep in front of our minds the high degree of subjectivity in the maximization process.... It would prevent our attributing a false objectivity to the cost and revenue figures."*40


In his latest paper, published in 1960, Thirlby suggests that subtle shifts in the definition of cost lead to confusion about social cost. "The subtle change in the meaning of cost, from the valuation of his own [the entrepreneur's] displaced end-product to the money input required for the selected course of action, is a change leading to still another conception, which carries with it the suspicion that it is to be regarded as a social cost. It resembles the first meaning of cost, in that it is supposed to be an alternative value displaced, but differs from it in that it is not the entrepreneur's own valuation of his own displaced end-product, but other people's (consumers') valuations of products which might have been produced by other entrepreneurs had they not been displaced."*41 This statement accurately summarizes the distinction between the London conception of opportunity cost and the orthodox conception that is currently held by most economists.

Mises' Human Action


As I noted earlier, Ludwig von Mises was one of the chief sources for the subjectivist economics expounded at LSE by Hayek, and his work was an influence as well on both Robbins and Thirlby. Mises' earlier work on the possibility of socialist calculation has been mentioned; some reference must now be made to his treatise, Human Action,*42 published in English in 1949, but based on a work in German published in 1940. In this book, Mises does discuss cost explicitly, even if briefly, and his basic conception is similar to that London conception that is best represented in Thirlby's work. Generically, "costs are equal to the value attached to the satisfaction which one must forego in order to attain the end aimed at" (p. 97). "At the bottom of many efforts to determine nonmarket prices is the confused and contradictory notion of real costs. If costs were a real thing, i.e., a quantity independent of personal value judgments and objectively discernible and measureable, it would be possible for a disinterested arbiter to determine their height.... Costs are a phenomenon of valuation. Costs are the value attached to the most valuable want-satisfaction which remains unsatisfied" (p. 393).


Mises' ideas on cost have been further developed by two of his American followers. In his two-volume treatise, Man, Economy, and the State, Murray Rothbard adopts a subjectivist conception of cost that is closely akin to that advanced by G. F. Thirlby.*43 And perhaps the single most satisfactory incorporation of a choice-related notion of cost into a general price-theory context is found in Kirzner's Market Theory and the Price System.*44

The Death of a Tradition?


At the London School of Economics, the ideas on cost developed by Robbins, Hayek, Coase, Thirlby, and others formed a part of a developing oral tradition which included many participants. Modern adherents to this tradition seem scarce, however, and only Jack Wiseman fully qualifies. In two basic papers published in the 1950's, Wiseman tried as others had earlier tried to apply LSE opportunity-cost logic to the long-discussed problems of marginal-cost pricing, applying this logic both as general criteria for organizing a collectivist economy and as the specific criterion for public-utility enterprise.*45


Wiseman shifted from LSE to York in 1963 and Thirlby retired from active academic life in 1962. There remain, no doubt, residues of the opportunity-cost tradition at LSE, but this does not inform the mainstream in either the teaching of economic theory or in the scholarly contributions of staff members. In the United States, the influence of Mises and his latter-day Austrian followers seems peripheral to the modern mainstream of economic theory. The concept of opportunity cost which emerged from both the subjectivist-Austrian and the common-sense approaches—the concept that blossomed for two decades at LSE—seems to have lost in its struggle for a place among the paradigms of modern economics. Along with other conundrums in intellectual history, this is not easy to explain. The arguments have not been refuted, and within their limits they surely remain valid. Hopefully, this book will generate a partial resurrection by delineating the methodological foundations of the two parallel theories of economic process.*46

Appendix to Chapter 2: Shackle on Decision


A survey of London contributions would be incomplete without reference to the work of G. L. S. Shackle. The problem of integrating intellectual constructions within one's own thought is well illustrated in his case. Shackle was a student at LSE in the years when the opportunity-cost conception was being developed. In several of his papers, Thirlby expresses indebtedness to Shackle; and it is immediately evident that Shackle's treatment of the decision process is wholly consistent with the London doctrine of opportunity cost. Yet—and surprisingly—Shackle does not, to my knowledge, make the obvious linkage between his provocative and important work on decision, uncertainty, and time and the work on opportunity cost carried forward by his LSE counterparts. In his general treatment of cost itself, Shackle reverts to orthodoxy.


His contributions to the theory of decision can, nonetheless, be helpful in clarifying the theory of cost, and some selected excerpts from one of his books*47 seem worth presenting:

When a number of actions, distinguished from each other by the sets of outcomes respectively assigned to them, are available and choice amongst them is open to the decision-maker, the sets of outcomes, each considered as a whole, are mutually exclusive rivals of each other. Within each set, the members also are mutually exclusive rival hypotheses. Thus these outcomes cannot be matters of fact but are things imagined by the decision-maker. They exist in his imagination, not after but before his commitment to a particular act; their existence is within the moment of decision and forms part of that act [pp. ix, x].
Decision means literally a cut ... a cut between past and future [p. 3].... We assume that choice amongst a set of rival available acts will be made in view of consequences associated in some manner and degree by the decision-maker with the acts. We also assume that the only consequences relevant for this choice are experiences of the decision-maker.... For three separate reasons ... they cannot be experiences coming from outside the decision-maker's mind from sources of stimulus observable in principle by others; they cannot, that is to say, be what we ordinarily speak of as 'real' experiences requiring the intervention of sense perceptions of the external world; they cannot be 'news' [p. 8].
... Outcomes are figments and imaginations [p. 9]. For it is the contention that the outcomes, by comparison of which decision is made, are figments of the individual mind (no matter whether in some later actuality they shall be observed to have come true: nothing could be more irrelevant) [p. 10].


These and many other statements by Shackle could be inserted almost without change in the cost discussions of Coase and Thirlby. Shackle's failure to make the shift of these relevant ideas to his own—though much more elementary—discussion of cost indicates that the classically based predictive theory can exist alongside the logical theory of choice in the thought patterns of a single economist, even though the two theories are incompatible with each other.

Notes for this chapter

Philip H. Wicksteed, The Common Sense of Political Economy (London: Macmillan, 1910), p. 380.



"These reflections will explain the great ambiguity of the term 'cost price.' ... [M]embers of the same trade ... will use the word in different senses. One will declare that he is 'making no profits at all,' but is 'selling at a loss,' and another will say that 'things are bad enough with him, but not quite so bad as that,' when they both mean to indicate exactly the same state of affairs. Men will declare in good faith that they are 'selling below cost price,' and yet will never think of suspending operations." Ibid., pp. 380-81.

Ibid., p. 385.

Herbert J. Davenport, Value and Distribution (Chicago: University of Chicago Press, 1908), p. 273.

See esp. ibid., p. 404.

For a summary of the history of cost theory in which Davenport's ideas are prominently featured, see Bob M. Keeney, "The Evolution of Cost Doctrine" (Mimeographed, Midwestern Economics Association, November 1967).

Frank H. Knight, "Fallacies in the Interpretation of Social Cost," Quarterly Journal of Economics, XXXVIII (August 1924), 592f., reprinted in F. H. Knight, The Ethics of Competition (London: Allen and Unwin, 1935), p. 225.

L. Robbins, "Certain Aspects of the Theory of Cost," Economic Journal, XLIV (March 1934), 1-18. Robbins' interest in the issues here was indicated in his earlier paper, "On a Certain Ambiguity in the Conception of Stationary Equilibrium," Economic Journal, XL (June 1930), esp. 209-11.

Ludwig von Mises, Human Action (New Haven: Yale University Press, 1949).

"Die Wirtschaftsrechnung im sozialistischen Gemeinwesen," Archiv für Sozialwissenschaften, XLVII (1920), reprinted as "Economic Calculation in the Socialist Commonwealth," in F. A. Hayek (ed.), Collectivist Economic Planning (London: Routledge, 1935).

Ludwig von Mises, Die Gemeinwirtschaft (Jena, Germany: Gustav Fischer, 1922). The second German edition appeared in 1932. Mises added an epilogue to this edition at the time of its translation as Socialism (New Haven: Yale University Press, 1951).

Lionel Robbins, The Great Depression (New York: Macmillan, 1934), esp. pp. 143-54.

F. A. Hayek (ed.), Collectivist Economic Planning, op. cit.

Ibid., p. 226.

A. P. Lerner, "Statistics and Dynamics in Socialist Economics," Economic Journal, XLVII (June 1937), 253-70.

F. A. Hayek, "Economics and Knowledge," Economica, IV (1937), 33-54, reprinted in Hayek, Individualism and Economic Order (Chicago: University of Chicago Press, 1948), pp. 33-56.

Additional essays that appeared in 1940, 1941, 1942, and 1943 are included in the two volumes, Individualism and Economic Order, op. cit., and The Counter-Revolution of Science (Glencoe, Ill.: The Free Press, 1952).

Ludwig von Mises, Grundprobleme der Nationalökonomie (Jena, Germany: Gustav Fischer, 1933), translated by George Reisman as Epistemological Problems of Economics (New York: Van Nostrand, 1960).

See esp. his review of Henderson's Supply and Demand, reprinted in Edwin Cannan, An Economist's Protest (London: P. S. King, 1927), pp. 311-14.

R. H. Coase, "Business Organization and the Accountant," The Accountant (October-December 1938). These articles are reprinted in David Solomons (ed.), Studies in Costing (London: Sweet and Maxwell, 1952), pp. 105-58.

Along with Coase, other members of Plant's group of young business economists were R. S. Edwards, R. F. Fowler, and David Solomons. This group was interested in making economic theory of greater practical relevance for business operations and especially for accounting practice.

In his later, and more widely known, paper on marginal-cost pricing, Coase's argument for the multi-part tariff was informed throughout by the conception of opportunity cost developed in his earlier papers. His emphasis, as it has been interpreted by later writers, was, however, on the familiar conflict between marginal-cost and profitability criteria. His opportunity-cost defense of multi-part pricing has been largely overlooked. See R. H. Coase, "The Marginal Cost Controversy," Economica, XIII (August 1946), 169-82. In a comment on Coase's paper, G. F. Thirlby criticized the implied objectivity of cost. See G. F. Thirlby, "The Marginal Cost Controversy: A Note on Mr. Coase's Model," Economica, XIV (February 1947), 48-53.

G. F. Thirlby, "The Subjective Theory of Value and Accounting Cost," Economica, XIII (February 1946), 32-49.

Ibid., 33-34.

G. F. Thirlby, "The Ruler," South African Journal of Economics, XIV (December 1946), 253-76.

G. F. Thirlby, "The Economist's Description of Business Behavior," Economica, XIX (May 1952), 150.

G. F. Thirlby, "Economists' Cost Rules and Equilibrium Theory," Economica, XXVII (May 1960), 150.

Mises, op. cit.

See Murray Rothbard, Man, Economy, and the State (New York: Van Nostrand, 1962), esp. Vol. I, pp. 290-308.

See I. M. Kirzner, Market Theory and the Price System (New York: Van Nostrand, 1963), esp. Chapter 9.

Jack Wiseman, "Uncertainty, Costs and Collectivist Economic Planning," Economica, XX (May 1953), 118-28; and his "The Theory of Public Utility Price—An Empty Box," Oxford Economic Papers, IX (February 1957), 56-74.

The manuscript for this book was completed before I had access to the article on "Cost" by A. A. Alchian in International Encyclopedia of the Social Sciences, III (New York: Macmillan, 1969), pp. 404-15.

G. L. S. Shackle, Decision, Order and Time in Human Affairs (Cambridge: Cambridge University Press, 1961), pp. ix, x.

End of Notes

Return to top