Was Hayek an ACE?

• Author: Vriend, Nicolaas J.
• Position: Agent-based computational economics

1. Introduction

   Hayek was without doubt one of the great minds of economics, and not only of economics. Obviously, this paper will not pretend to question his being an ace. The Ace in the title rather refers to agent-based computational economist (ACE). As Tesfatsion puts it on the ACE Web site:

   "Agent-based computational economics (ACE) is roughly characterized as the computational study of economies modelled as evolving decentralized systems of autonomous interacting agents. A central concern of ACE researchers is to understand the apparently spontaneous formation of global regularities in economic processes, such as the unplanned coordination of trade in decentralized market economies that economists associate with Adam Smith's invisible hand. The challenge is to explain how these global regularities arise from the bottom up, through the repeated local interactions of autonomous agents channeled through socioeconomic institutions, rather than from fictitious top-down coordination mechanisms such as imposed market clearing constraints or an assumption of single representative agents. ACE is thus a specialization to economics of the basic complex adaptive systems (CAS) paradigm." (Tesfatsion 1998)

   The descriptions used in this informal definition of ACE must look rather familiar to experts of Hayek. Given the easily recognizable affinity between Hayek and ACE it is no surprise that many Hayek experts seem interested in the recent ACE literature. At the same time, however, many ACE researchers seem hardly aware of Hayek's work. Every now and then somebody might mention that it would be interesting to have a closer look at Hayek's work, but that is about it. In this paper I will take up these suggestions. I present a concrete example of an ACE research project concerning the phenomenon of information contagion as a guide to address in great detail the question whether Hayek might have been an ACE avant-la-lettre. Apart from the purely intellectual motivation for such a study, some of the underlying questions motivating this project are: How could Hayek's insights and theories help to understand ACE research? And what, if anything, could we learn from current ACE research about Hayek's work? Far from offe ring definite answers to these questions, this paper will suggest that there might be some reasons to believe that a close encounter between Hayek and ACE has potential benefits that might work in both directions.

   This paper is organized as follows. Section 2 outlines the interest of Hayek in complex adaptive systems, and discusses some methodological issues concerning ACE modeling, relating it to Hayek's work. Section 3 presents an ACE model of the emergence of information contagion, whereas section 4 presents an analysis of the properties of the model. In section 5 I relate the specifics of my ACE model to Hayek's work on the division of knowledge and information aggregation, and section 6 concludes.

2. Hayek, Complex Systems, the Methodology of the Social Sciences, and ACE Modeling

   Hayek shared with ACE the belief that the economy needs to be understood from a bottom-up perspective. In this he stood out from both Keynesian macroeconomics and Walrasian general equilibrium theory, which came to dominate the field of economics during Hayek's life. He insisted on the need to consider a market economy as a truly decentralized system of interacting individual agents. One of the central questions Hayek analyzed was: "How can the combination of fragments of knowledge existing in different minds bring about results which, if they were to be brought about deliberately, would require a knowledge on the part of the directing mind which no single person can possess?" (Hayek 1948b, p. 54). In much of his work Hayek took the view that to explain such phenomena one must start the analysis from the level of individuals. His view of individual behavior was firmly rooted in the "antirationalistic" (Hayek 1948a, p. 8) approach of the English individualism as known, for example, from Adam Smith's Invisible Hand metaphor: "... true individualism is the only theory which can claim to make the formation of spontaneous social products intelligible" (p. 10), and "true individualism believes ... that, if left free, men will often achieve more than individual reason could design or foresee" (p. 11). With respect to general equilibrium theory, Hayek pointed out: "The equilibrium relationships cannot be deduced merely from the objective facts, since the analysis of what people will do can start only from what is known to them" (p. 44), and: "... the general question of why the subjective data to the different persons correspond to the objective facts. Our problem of knowledge here is just the existence of this correspondence..." (pp. 51-52). In this respect, Hayek clearly distinguished himself from Keynes: "Keynes' theories will appear merely as the most prominent and influential instance of a general approach to philosophical justification of which seems to be highly questionable. Though with its reliance on apparently measurable magnitude it appears at first more scientific than the older micro-theory, it seems to me that it has achieved this pseudo-exactness at the price of disregarding the relationships which really govern the economic system" (Hayek 1978, p. 289).

   Starting from his 'true individualism' and his skepticism concerning the approaches followed by Keynes and Walrasia general equilibrium theorists, and his view that what really mattered was something to do with the interactions between the individual agents, during the 1950s Hayek came to consider the economy as a complex adaptive system. A lucid account of the developments in Hayek's work in the 1950s and 1960s is given in Caldwell (2000), who describes how "(b)y the 1960s Hayek was seeing complex orders everywhere" (p. 19), with the underlying principles best understood from an evolutionary perspective.

   Hayek's research followed two tracks in the 1950s. First, his interest in the methodology of the social sciences led him back to his earlier work on theoretical psychology. In Hayek (1952) the sensory order of the human brain is described as an example of a self-organizing complex order, with linkages within the brain being strengthened or weakened as a result of feedback from the external environment. This work was one of the principal readings for Hayek's 1952 seminar at the University of Chicago on "Scientific Method and the Study of Society", in which people like Enrico Fermi and Sewell Wright participated, and that focussed on methodological issues concerning the study of complex phenomena.

   The second track followed by Hayek during the 1950s concerned his work on political theory. In Hayek (1960) the development of civilization is related to the growth of knowledge, where knowledge was seen broadly, including things such as habits, skills, emotional attitudes, tools, institutions, and even ethical and aesthetical principles. These various forms of knowledge evolve as a result of random variations (accidents) and "selective elimination of less suitable conduct" (p. 26). (1) Hence, Hayek started looking at cultural evolution as the evolution of a tradition of learned rules of conduct and social norms. "We understand now that all enduring structures above the level of the simplest atoms, and up to the brain and society, are the results of, and can be explained only in terms of, processes of selective evolution, and that the more complex ones maintain themselves by constant adaptation of their internal states to changes in the environment" (Hayek 1979, p. 158).

   Some of the methodological insights developed in Hayek's work on cultural evolution seem to be of particular interest to ACE research. Not so much because, starting from Hayek's work, ACE introduces novel methodological developments, but because ACE turns out to be a very advantageous way to actually apply the abstract methodological insights of Hayek and others. Moreover, acknowledging this, in turn, helps to put ACE research in the right perspective, facilitating a fruitful interpretation of its results as well.

   Social theory attempts to explain social phenomena, and as Weimer (1982) puts it: "explanation is modeling" (p. 271). According to Hayek (1948c), what we do is, "we construct hypothetical models in an attempt to reproduce the patterns of social relationships which we know in the world around us" (p. 68). In contrast to the natural sciences, in the social sciences "(e)xperimentation is impossible, and we have therefore no knowledge of definite regularities in the complex phenomena in the same sense as we have in the natural sciences" (Hayek 1948f, p. 126). That is, as Weimer (1982) explains, in the natural sciences there is the ability to simplify and control a situation to the extent that it can be repeated, either under identical conditions or those that we choose to vary systematically, such that we can isolate and identify the definite regularities in observed phenomena. (2) However, "(t)he empirical research in complex social phenomena consists in the construction of situations in which we demonstrate to ourselves that we can produce "facts" of which we are already well aware. Our demonstrations "test" our theoretical models only in the sense already noted; they compare the consistency of our theoretical model with an analogical knowledge of social phenomena, but they neither confirm nor refute them in any logical sense" (Weimer 1982, p. 252). And Hayek (1948c): "The theory itself, the mental scheme for the interpretation, can never be "verified" but only tested for its consistency. It may be irrelevant because the conditions to which it refers never occur; or it may prove inadequate because it does not take account of a sufficient number of conditions. But it can no more be disproved by facts than can logic or mathematics" (p. 73). And "... a simple theory of phenomena which are in...