On the Behavioral Foundations of the Law of Supply and Demand: Human Convergence and Robot Randomness

In a seminal series of papers, Gode and Sunder [1993, b, 1996] have explored the relationship between limited rationality, market institutions and the general equilibration of markets to the competitive equilibrium. Their fundamental discovery is that within the classical double auction market institution only the weakest elements of rationality need to be present for markets to exhibit high allocative efficiency and price convergence. While Gode and Sunder place more emphasis on allocative efficiency than on price convergence, the apparent price convergence increases the agreement between their simulation results and observed price convergence in single isolated periods of double auction markets with humans. Their 'Zero Intelligence' [ZI] agents, are governed by completely random choice and constrained only by a budget constraint, are coordinated by market forces to the competitive equilibrium. The results are closely related to the results by Becker [1962] that the budget constraint alone, in the presence of randomly behaving agents, assures that demand curves will be downward sloping.