An economically realistic asset exchange model

Abstract

Despite their highly idealized nature, certain agent-based models of asset exchange, proposed for the most part by physicists and mathematicians, have been shown to exhibit remarkable agreement with empirical wealth distribution data. While this agre- ement is comforting, there is widespread sentiment that further progress will require a detailed under- standing of the connection between these idealized models and the more realistic microeconomic models of exchange used by economists. In this paper, we examine that connection for a three-parameter asset exchange model, the Affine Wealth Model (AWM), that has demonstrated fraction-of-a-per cent agreement with empirical wealth data. We compare certain properties of this model with those of three great milestones of twentieth century economics, namely (i) Expected Utility Theory, (ii) General Equilibrium Theory and (iii) Prospect Theory. We find that the phenomenology exhibited by the AWM is fundamentally incompatible with Expected Utility Theory and General Equilibrium Theory, but very similar to that exhibited by Prospect Theory. Based on these observations, we argue that AWM transactions are, in a particular sense, an approximation to those described by Prospect Theory, and that Prospect Theory provides the sought-for connection between econophysics and microeconomics, at least for the topic of wealth distribution. This article is part of the theme issue ‘Kinetic exchange models of societies and economies’.