What Should Economists Do? A Historical Perspective

Much has been said about this article, yet some underappreciated aspects ought to be emphasized to understand more clearly Buchanan’s political economy and the distinction he made in his article between choice and exchange. This is indeed why “What Should Economists Do?” has become so famous, the opposition Buchanan drew between two alternative ways of defining economics. Buchanan thus insisted that,

• The theory of choice must be removed from its position of eminence in the economist’s thought processes. The theory of choice, of resource allocation, call it what you will, assumes no special role for the economist, as opposed to any other scientist who examines human behavior. (217)

Rather, economists “‘should’ concentrate their attention… on exchange” (p. 217). They should see their discipline as “a sophisticated ‘catallactics'” (p. 214).

Buchanan not only opposed choice in favor of exchange, even more surprising, he set himself against two traditions in economics, starting with Lionel Robbins. He was particularly virulent toward Lord Robbins whom, Buchanan wrote, “I take… as an adversary” (p. 214)3 because his “overly persuasive definition of economics has served to retard rather than advance scientific progress” (p. 214). He did not hesitate to suggest that the economists who followed the path opened by Robbins had not remained “within the ‘strict domain of science'” (p. 213). By contrast, since he wanted to adopt a scientific approach, Buchanan chose to travel with Adam Smith.

Buchanan made two major points against the choice perspective. The first was a slippery slope argument. If economics were defined as a science of choice, and groups as well as individuals were supposed to face allocation problems, it was easy to shift from analyzing individual choices to examining the choices made by groups, “to slip across the bridge between personal or individual units of decision and “social” aggregates.” (p. 215) Adopting Robbins’s definition made this slip not only possible, but actually too easy to make. Buchanan disagreed with this tendency. In his mind, groups do not exist independently from individuals; groups do not make decisions (about the allocation of resources, or any kind of decisions). Only individuals do.

Buchanan’s second point was that a science of choice ignores uncertainty or assumes that individuals “know what [they] want” (p. 217), and therefore reduces choice to a “purely mechanical” (p. 217 & 218) behavior. In the absence of uncertainty, or if individuals are assumed to be certain of what they want, their choices are not genuine. Individuals choose what they are “predetermined” to choose. A computer or any external observer could easily determine the solution to the problems raised by choices in such situations or under such assumptions. Accordingly, these are problems for scientists or social engineers, but not for economists. Thus, when adopting such a framework, economists transform their discipline into a computational and a technical science, into applied mathematics or managerial science (p. 216). They therefore deal with problems they should not address, no longer being social scientists by having become social engineers.

Such opposition to choice might seem surprising under the pen of someone who, with Gordon Tullock, had written in the recently published The Calculus of Consent, “that they were proposing a theory of political choice (p. 4, emphasis added), and even a theory of praxeological choice. (p. 29, emphasis added)

The contradiction is apparent only. The choice Buchanan was mentioning was different from the choice economists were commonly using, and his economic analysis of politics—actually about constitutions—had not much to do with choosing an outcome. In his mind, collective action and the political choice he was talking about in The Calculus of Consent was not about choosing an outcome, but about devising institutions to make decisions. Buchanan was developing, and had long been developing, a form of economics that focused on the ways in which individuals organize themselves collectively to deal with the problems they cannot solve privately.

Buchanan’s opposition to choice and the idea that economists should try to solve the problem of allocating scarce resources (means) efficiently to reach alternative or competing ends first came from his rejection of using social welfare functions to determine resource allocation; he preferred Knut Wicksell’s approach, centered on unanimity and Pareto optimality [see Vilfredo Pareto]. He had made the point in his criticism of Kenneth Arrow’s views on institutions as a social welfare function, because he viewed democracy as a means to make decisions but to promote discussion to reconcile conflictual interests (1954b).

But why such a strong opposition to Robbins? To understand it, one must refer to the exchange Buchanan had with Israel Kirzner just after “What Should Economists Do?” was published. Kirzner indeed disagreed with Buchanan. To Kirzner, and by contrast with what Buchanan had written, exchange and choice were therefore not separable. And by extension, Robbins’s work complemented Ludwig von Mises’s. This was what Kirzner argued in his dissertation, written under Mises’s supervision. Kirzner explained that Mises had developed the praxeological dimension that Robbins had failed to fully recognize. Robbin’s definition was insufficient, according to Kirzner, but it nonetheless was a step in the right direction. And Mises linked the science of human action, praxeology, to exchange or catallactics; economics or catallactics was the “best-developed part” of “the theoretical science of human action, praxeology” (1944, p. 527; see also Mises, 1949, p. 3). Kirzner wrote a comment on “What Should Economists Do?” that he sent to Buchanan in March 1964 with a letter in which he noted that Buchanan’s “excellent” paper “had caught [his] notice”. Kirzner, however did not understand why there was no mention of Mises or Friedrich Hayek in the paper. His point was that what Buchanan thought economists should do was exactly what he, Mises, and Hayek were already doing; they were focusing on exchange but remained within a Robbinsian (choice) framework. Kirzner (1965, p. 258) even found it,

• … obvious that Buchanan’s own characterization of economics as concerned with the implications of the human propensity to truck, can, without strain, be subsumed under Robbins’ economics. After all men do seek out exchange opportunities in the course attempts to avoid “wasting” their resources.

More problematic, by separating choice from exchange, by refusing to start with choice, Buchanan refused to assume that human beings have plans. Economics, Kirzner argued, had to start from individual choice and then analyze exchange and the market process as a consequence of these choices. This is what Mises, Hayek, and Kirzner himself were doing. With the exception of Kirzner, Buchanan mentioned none of them.

Buchanan replied to Kirzner that he had a section in which he cited Hayek, Mises, and Kirzner “in the first draft of [his] paper”, which he eventually “left out in later drafts for space reasons.” (Buchanan to Kirzner, 25 March, 1964) Indeed, Buchanan had written one page to clarify the differences between their respective positions; he was aware that “there [wa]s an obvious affinity here, that deserves some clarification.” (1964b, 23) He thus copied and pasted the section that had been removed into his letter to Kirzner.4

In this section, Buchanan admitted that “the Hayek, Mises, Kirzner, argument does represent a significant improvement over the open-ended “logic of choice” conception of Robbins” (ibid.). However, he was also explaining that he opposed a definition of economics centered on choice because it implied that economists should or could study any kind of human choice—Buchanan had perceived what would become a reality 25 years later, when economics started to expand beyond its traditional boundaries. In his mind, boundaries should limit the subject matter of economics.

In addition, Buchanan did not want to study exchange that would result from choice, as Mises, Hayek and Kirzner did. To him, political economists should not study exchange when it results from choice, but exchange when it is the primary purpose individuals have. To Buchanan, an individual who trades because he needs what others have treats others as a resource and therefore as a means (see Marciano and Meadowcroft 2025). The individual Buchanan was interested in is someone who is primarily exchanging with others, and therefore treats others as ends and not as means. Buchanan wanted to stress that political economy is a social science, a science that studies how individuals cooperate with others, how they organize collective action.

Buchanan thus distinguished between an exchange that results from choice and the exchange that comes from a desire to exchange or, as Smith had said, “the propensity to truck and barter and the manifold variations in structure that this relationship can take” (p. 214). His focus was on the later, not on the former. It is thus not as decision-makers (as choosers), or because they make decisions (choices), that individuals are an object of study for economists, but “in so far as… they exchange, trade, as freely contracting units” (p. 221).

This explains why Buchanan did not refer to Mises but to Archbishop Richard Whately, who supposedly coined the term ‘catallactics’ in his Introductory Lectures on Political Economy published in 1832. Among other important things, Whately had said that it is as “[a]n animal that makes Exchanges… alone that Man is contemplated by Political Economy” (Whately, 1832, 7). Not the acting man of Mises nor the economizing man of Robbins. As Buchanan said in his reply to Kirzner: “I concentrate only on that individual choice behavior that involves conscious cooperation among individuals, and upon those institutions that evolve as a result of such behavior.” (Buchanan to Kirzner, 25 March 1964)

This is indeed what Buchanan explains in the part of his article devoted to exchange and catallactics. An economist for whom economics is a science of choice, for example, would focus on the choices Robinson Crusoe makes when he deals with his environment. In that case, Robinson Crusoe has a computational problem to solve (p. 217)—the kind of problem economists should not study. Robinson Crusoe also had computational problems when he interacted with Friday as if the latter was part of his environment, as part of nature. Crusoe then does not treat Friday as if he was a human being with whom interactions could develop, but “simply as a means to his own ends” (p. 218) Transactions could take place, obviously. They were exchanges, as if the two individuals were choosing goods and not exchanging with other human beings. These were purely computational problems; they were mechanical and involved no uncertainty. These were not problems economists should study.

To Buchanan, economists (i.e., political economists) should study the interactions Robinson Crusoe and Friday have when they are primarily trying to exchange, when they do not treat one another as means but as ends. They should thus study the trades, exchanges that take place when individuals explicitly and consciously interact with one another. When Robinson Crusoe and Friday face each other, as long as they do not treat each other as a means, “a wholly different, and wholly new, sort of behaviour takes place, that of exchange, trade, or agreement” (p. 218).

Here, one finds another issue of particular importance—conflict (versus cooperation). When Robinson Crusoe treats Friday as part of nature, as if they were totally independent from each other, he makes choices that are not part of the subject matter of economics because they are conflictual. Economists, Buchanan said, should study only the transactions that are cooperative. That was precisely what ‘catallactic’ meant to Buchanan. The research program Buchanan presented in “What Should Economists Do?” revolved around economics defined as a science of cooperative collective action.

To discuss this aspect of economics, Buchanan introduced a second concept, “symbiotics”. “Symbiotics” means “the study of the association… mutually beneficial… between dissimilar organisms” (p. 217) and “conveys, more or less accurately, the idea that should be central to our discipline” (p. 217), the idea of a discipline that studies a “relationship” based on “association” and “cooperation” (p. 217). Thus, when Crusoe treats Friday as part of nature, “a ‘fight’ ensues, and to the victor go the spoils. Symbiotics does not include the strategic choices that are present in such situations of pure conflict.” (p. 218) Economics as catallactics, and thus as symbiotics, was not only a science of exchange, it was also a science of cooperation, a science that studies how individuals cooperate to exploit mutual gains from trade. It was also an optimistic science of cooperation since, as Buchanan noted, “Individuals are observed to cooperate, to make agreements, to trade.” (1964b, p. 219; italics added) Buchanan did not doubt that individuals would cooperate, trade, and devise arrangements to organize these trades.

Certainly, Buchanan was no dreamer; cooperation was not always warranted. But he believed that individuals would always find a solution, if a solution exists. This is also in contrast with economists who believe in market failures. The latter believe that the failure to allocate resources efficiently is the end of the process. Buchanan believed that failure is only a step towards a solution. According to Buchanan, once a failure has been observed, if private arrangements fail, then individuals will nonetheless “search voluntarily for more inclusive trading exchange arrangements.” (p. 220) And if this kind of arrangement, private and collective, did not work, and if a solution was required, individuals then look at a higher stage, accepting “the need for transferring, again voluntarily, at least at some ultimate constitutional level” (221) the activities that were raising problems at the private level.

From this perspective, “What Should Economists Do?” is also a paper that demonstrates Buchanan’s optimism and belief in human nature, his conviction that individuals were “reasonable men, capable of recognizing what they want, of acting on this recognition, and of being convinced of their own advantage after reasonable discussion” (1959, p. 134).

This optimism progressively faded away. Barely two years later, in “Economics and Its Scientific Neighbors,” (1966) Buchanan was already reconciling choice and exchange—economics as a logic of choice and as a science of exchange (see also “Retrospect and Prospect”, 1979). Buchanan was still building his research program. This ultimate evolution makes the 1964 article even more important, as a moment in Buchanan’s trajectory, as a step towards a refined view on economics.

References

Buchanan, James M. (1954). “Social Choice, Democracy, and Free Markets.” Journal of Political Economy, 62 (2): 114-123.

Buchanan, James M. (1959). “Positive Economics, Welfare Economics, and Political Economy.” Journal of Law and Economics 2 (October): 124-138.

Buchanan, James M. (1964). “What Should Economists Do?” Southern Economic Journal 16(2), 168-174. Reprinted in What Should Economists Do? Indianapolis: Liberty Fund.

Buchanan, James M. (1966). “Economics and Its Scientific Neighbors.” In What Should Economists Do? Indianapolis: Liberty Fund, pp. 115-142.

Buchanan, James M. (1979:1). “Retrospect and Prospect.” In What Should Economists Do? Indianapolis: Liberty Fund, pp. 279-284.

Buchanan, James M. (1979:2). What Should Economists Do? Preface by Geoffrey Brennan and Robert D. Tollison. Collection of essays. Indianapolis, Liberty Fund, Inc.

Buchanan, James M., and Gordon Tullock. (1962). The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor: University of Michigan Press.

Kirzner, Israel M. (1960). The Economic Point of View. An Essay in the History of Economic Thought.

Kirzner, Israel M. (1965). “What Economists Do.” Southern Economic Journal, 31(3), 257-261.

Marciano, Alain and J. Meadowcroft. (2025). “Buchanan’s theory of emancipation: Artifactual man in perspective.” in M. Novak (ed), Liberal Emancipation: Explorations in Political and Social Economy, Springer, forthcoming.

Mises, Ludwig von (1944). “The Treatment of ‘Irrationality’ in the Social Sciences.” Philosophy and Phenomenological Research, 4(4), 527-546.

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

Whately, Richard. (1832), Introductory Lectures on Political Economy, London: B. Fellowes.

Footnotes

[1] James M. Buchanan, What Should Economists Do? Preface by Geoffrey Brennan and Robert D. Tollison. Collection of essays, including Buchanan’s “What Should Economists Do?” Indianapolis, Liberty Fund, Inc.

[2] James M. Buchanan (1964). “What Should Economists Do?” Southern Economic Journal 16(2)

[3] In a first draft, he had noted “an adversary” then switched to “my adversary”, making the claim even more personal, before switching back to “an adversary”.

[4] In particular, Buchanan had noted, “I want, if possible, to be able to develop conceptually refutable hypotheses about this particular set of human behavior problems” (1964a, p. 24). This seems to show that Buchanan was aware of the criticisms that Kirzner could (and did) raise against his position.