New perspectives on realism, tractability, and complexity in economics

Fuzzy logic and genetic algorithms are used to rework more realistic (and more complex) models of competitive markets. The resulting equilibria are significantly different from the ones predicted from the usual static analysis; the methodology solves the Walrasian problem of how markets can reach equilibrium, starting with firms trading at disparate prices. The modified equilibria found in these complex market models involve some mutual self-restraint on the part of the agents involved, relative to economically rational behaviour. Research (using similar techniques) into the evolution of collaborative behaviours in economics, and of altruism generally, is summarized; and the joint significance of these two bodies of work for public policy is reviewed. The possible extension of the fuzzy/ genetic methodology to other technical aspects of economics (including international trade theory, and development) is also discussed, as are the limitations to the usefulness of any type of theory in political domains. For the latter purpose, a more differentiated concept of rationality, appropriate to ill-structured choices, is developed. The philosophical case for laissez-faire policies is considered briefly; and the prospects for change in the way we ‘do economics’ are analysed.