Dissertation Abstract: Decision-Theoretic Models of Market Equilibration
- 48 Downloads
There has been a surprising dearth of decision-theoretic approaches to market equilibration within the extended Arrow-Debreu-McKenzie framework. While game theory is replete with learning and evolutionary models in which Nash equilibria are not necessarily the only stable states of a game, general equilibrium theorists have generally been content to accept the attainment of competitive equilibrium as a matter of faith. The first two chapters of this dissertation advance the study of behaviorally plausible models of market equilibration. First, we develop the epsilon-intelligent competitive equilibrium algorithm. According to this model, the actions of minimally sophisticated agents based on local information will lead an exchange economy to approximate competitive equilibrium in a larger set of economics than Walras’ tatonnement. The algorithm also supports a behavioral interpretation of Negishi’s existence proof of competitive equilibrium. The second chapter is an empirical analysis of laboratory markets designed to test the extent to which human behavior is consistent with the algorithm’s behavioral restrictions, and suggest alternative hypotheses. The chief finding is that while subjects sufficiently sophisticated to consistently secure competitive utility for themselves exist, the majority are satisficers who follow small modifications of a simple utility-improvement rule.
While general equilibrium theory has lacked a behaviorally plausible foundation of price equilibration, behavioral economic models typically ignore general equilibrium implications. In the third chapter of this dissertation, I present the surprising result that loss aversion and diminishing sensitivity for gains imply a greater redistribution of endowments than when otherwise identical preferences are reference-independent. The result is surprising because loss aversion was developed in part to account for the status quo bias, whereby people tend to value a good more when it is in their possession than when it is not. One might reasonably suspect trade to be inhibited by this bias. The counter-intuitive result is driven by the fact that, given the axiomatization of loss aversion and diminishing sensitivity for gains, the acquisition of some quantity of a good increases one’s taste for that good, thus perpetuating a taste for more trade.