Kenneth Arrow and What’s Important

Print/Save PDF

In my considered opinion (with apologies to my MIT colleague Paul Samuelson), Ken Arrow (1921-2017) was the most accomplished economic theorist of the second half of the 20th century. I am especially impressed that he tackled only really important problems. It was like anything else was a waste of time. Steve Pressman’s essay on Arrow in Fifty Major Economists (2006) identifies four areas of research in his “Works by Arrow”: Social Choice and Individual Values (1951); “Existence of Equilibrium for a Competitive Economy,” Econometrica (1963); Essays in the Theory of Risk Bearing (1970); and The Limits of Organization (1974).

All four are fundamental contributions to economic theory. The brief descriptions that follow are, for convenience, taken out of order so as to better emphasize their connections to the GEM Project. Social Choice works through Arrow’s proof and demonstrated consequences of the his famous impossibility theorem. He shows that it is not possible to aggregate individual preferences in a way that always preserves each person’s rank ordering. “There cannot be a completely consistent meaning to collective rationality.” Economics can usefully explain individual choice but it cannot always explain group decision-making. It makes sense that heterogeneous preferences cannot be merged. The impossibility theorem, however, does not compromise the use of rationality to motivate generalized-exchange behavior. GEM modeling takes care to establish effective employee-preference priority attached to equitable treatment by management.

Essays collects Arrow’s pioneering work on what turned out to be the most consequential subject in economic research in the 20th century, i.e., information asymmetries and related problems. He illustrated his analysis mainly with health-related markets. Imperfect information needs no elaboration for readers of the GEM Blog.

Existence is fundamental to mainstream New Keynesian macroeconomics, which is constructed on market-centric general-equilibrium theory. Arrow elegantly demonstrated the existence of single-venue general equilibrium (SVGE). It was a great achievement in economic theory that had eluded theorists since Adam Smith ably outlined the analysis of rational marketplace exchange. Mainstream theorists occasionally point to Arrow’s analysis as a justification for their assertion that today’s consensus market-centric model is settled theory. But that reflects a fundamental misunderstanding of what Arrow accomplished. He made clear that the essential task of his existence demonstration was to identify the assumptions needed to support market-centric general-equilibrium.

Arrow’s conditions assuring the SVGE system of excess-demand equations has a market-clearing solution include: (a) the absence of increasing returns; (b) the absence of joint products or externalities in production or consumption; (c) universal price quoting (implying the existence of forward markets for all final goods and services); (d) universal price taking (implying the absence of pricing power in all product and factor markets) informed by perfect, symmetrical information; and (e) gross substitutability of all goods (implying that an increase in the price of one good always produces a positive excess demand for at least one other good).

Arrow, of course, never argued that the SVGE-existence requirements actually characterize modern economies. He saw them as necessary conditions, not axioms upon which operational theorems can be derived. Yet, as the market-centric general-equilibrium model class has been inexorably pushed forward as the proper framework for policy-useful macro analysis, the conditions have morphed into shadow axioms. SVGE limitations on the capacity to explain labor pricing and use are instructively illustrated by the universal price-taking and price-quoting that require that all goods and services be priced in competitive markets. That restriction is inherent in the ubiquitous search/match model class that has been widely put forward as stabilization relevant. It is, however, broadly understood by practitioners that worker cooperative input on the job, in the context of inherently costly, asymmetric workplace information, cannot be adequately observed, priced, or exchanged in the marketplace. Labor hours, which markets can measure, are an acceptable proxy for cooperative input (Έ) only if employee on-the-job behavior is cost-effectively monitored (Źj=Źɱj) or otherwise effectively managed. By definition, mainstream theory has no capacity to model the generalized rational exchange that developed in the aftermath of the Second Industrial Revolution. The market-centric general-equilibrium model is no more than a special case of the Project’s two-venue analysis, making it a deeply misleading description of actual macrodynamics.

The three fundamental Arrow works elaborated upon so far rank among the best known in the economics literature. By contrast, the last (The Limits of Organization) is virtually unknown. It is never assigned as graduate-school reading, never quoted in mainstream publications. Oblivion has resulted from “settled” market-centric theory not being able to accommodate Limits. As GEM Blog readers know, a great deal of critically important analysis has been obscured by the Ptolemaic insistence that model-building to be wholly conducted inside the marketplace box.

Unlike modern New Keynesian theorists, Arrow did not ignore the specialization and scale that generate costly, asymmetric workplace information, making voluntary labor cooperation with management objectives a necessary part of the effective pursuit of profit. From Limits (1974, p.25): “A firm, especially a large corporation, provides another major area in which price relations are held in partial abeyance. The internal organization is again hierarchical and bureaucratic… Internally, and especially at lower levels, the relations among its employees are very different from the arm’s length bargaining of our textbooks. As Herbert Simon has observed, an employment contract is different in many ways from an ordinary commodity contract; an employee is selling willingness to obey authority, a concept of central importance to which I will return in a later chapter.” Arrow seeks, with admittedly mixed success, to inform the economics of rational employer-employee exchange that occurs in large, bureaucratic firms. There is pleasing symmetry in his attempt to do for firms what he did for markets.

What is most pleasing to me is Arrow, the intellectual godfather of SVGE theory who only tackled really important problems in economic theory, recognized the importance of the central argument of the GEM Project. Breaking rigorous economics out of its market-centricity straightjacket is necessary to the development of useful macroeconomics.

Blog Type: Wonkish Saint Joseph, Michigan


Write a Comment

Your email address will not be published.