In this chapter, we develop an agent-based Scarf economy with heterogeneous agents, who have private prices and adaptively learn from their own experiences and those of others through a meta-learning model. We study the factors affecting the efficacy of price discovery and coordination to the Walrasian Equilibrium. We also find that payoff inequality emerges endogenously over time among the agents and this is traced back to intensity of choice (a behavioural parameter) and the associated strategy choices. Agents with high intensities of choice suffer lower payoffs if they do not explore and learn from other agents.
Complex Systems Modeling and Simulation in Economics and Finance, Springer, pp.113-139