March & Simon: Early Socialist Calculation Revisionists
| Nicolai Foss |
It is now commonly recognized that the majority of the economics profession for about four decades held an erroneous view of the nature of the “socialist calculation debate.” In particular, the nature of the arguments put forward by Mises and Hayek were misconstrued.
Revisionism took off in the mid 1980s with the work of Peter Murrell (e.g., here) and Don Lavoie (e.g., here). From a mainstream perspective, Murrell argued that the Austrians had developed sophisticated insights in property rights economics and the agency problem and had applied these insights to the problem of socialist calculation. Lavoie highlighted the distintive Austrian knowledge argument in the calculation debate, in particular emphasizing Hayek’s contribution. A bit later, Salerno and others emphasized the distinctiveness of Mises’ contribution. Thus, whereas Mises stressed the need for a distributed process of entrepreneurial judgment in the context of a private ownership economy characterized by uncertainty, Hayek put more of an emphasis on the impossibility under socialism of harnessing and processing massive amounts of knowledge, particularly under dynamic conditions.
Between the 2nd Wold War and these works, there are four decades in which the dominant opinion held that the Austrians had been thoroughly defeated by the formal demonstration, particularly in Oskar Lange’s work, of the possibility of combining socialism and efficient allocation. A pertinent question is whether there were (non-Austrian) dissenters from this dominant view. To those well steeped in libertarian social theory, names such as Trygve Hoff and G. Warren Nutter come to mind. But apart from these, it would appear that it is not until the mid 1980s that the distinctiveness of the Austrian arguments in the calculation debate concerning knowledge becomes recognized.
However, an early contribution, unknown to most economists, that fully recognized the distinctiveness of the Austrian arguments, is the 1958 book Organizations by James G. March and Herbert A Simon, a book that many would regard as the seminal contribution to organizational theory and a milestone in the evolution of organizational theory (I have heard organization theory scholars remark that all org theory in the last five decades is just variations over March & Simon themes).
The discussion of the socialist calculation debate takes place in the final chapter, “Planning and Innovation in Organizations,” the main purpose of which is to “… contrast the concept of rationality that has been employed in economics and statistics with a theory of rationality that takes account of the limits on the power, speed, and capacity of human cognitive faculties” (1958: 172) — in other words, bounded rationality.
This theme is taken through a number of variations, one of which is the theory of planning, understood as both “national planning and intrafirm planning” (p. 200). March and Simon (1958: 201) argue that even if motivational problems can be solved, there are still planning (coordination) problems remaining. They note, echoing Hayek (1945), that if one person or group of persons possessed “… all the relevant information connecting possible courses of action with the utilities resulting therefrom, he or they could discover which course of action was best for the organization” (p. 201). An alternative is to make use of the price mechanism, for example, through the Barone/Lange idea of consistent marginal cost pricing throughout the organization. March and Simon note a number of difficulties with this proposal, such as the requirement that externalities be absent. More seriously, perhaps, they note that it is not clear how to make a choice between the alternatives of central planning and pricing, since modern welfare economics, including the Lange/Barone proposal, does not give any positive reason for preferring the one to the other.
This is where the Austrian arguments in the socialist calculation debate enter the scene. These are placed under the heading “The principle of bounded rationality” (p. 203), and, accordingly, March and Simon note that the “… argument of von Mises and Hayek (we will use the latter’s version) depends crucially on the limits of information available to humans and their abilities to use information in their computations.” In other words, Hayek argues that “given realistic limits on human planning capacity” (italics removed) a decentralized system will work better than a centralized one.
Thus, March and Simon present a sympathetic reading of the Austrian — mainly Hayekian — positions in the socialist calculation debate. In the context of Organizations, March and Simon also criticize the “Robbinsian” characterization of decision-making in mainstream economics (to use Kirzner’s terms) – that is, the given’ness of means and ends — and they stress that the understanding of behavior should be broadened to include the process of discovering choice alternatives. These two observations are related, for it is arguably exactly because March and Simon are critical of the conceptualization of behavior in mainstream theory that they are so appreciative of the Austrian positions in the calculation debate.