Agent-based computational economics:

Agent-Based Computational Economics: Studying the Effect of Different Levels of Rationality on Inflation and Unemployment
Ahmed Okasha and Colin Johnson
knowledge, utilities and so on. Finally, economic theory assumes that agents are anonymous, that is, there is an auctioneer, union or invisible hand (Adam Smith) that determines the market price and then all agents, consumers and producers, mustfollow that price. Economic theory depends on these assumptions in order to study inflation. An Agent-based Computational Economics models (ACE) assume that agents are heterogeneous, boundedly rational [1], [16], [18] and interact with each other. Each agent interacts and communicates with other agents in its neighbourhood. Also, agents follow simple rules in their interactions and communications.Therefore, ACE is a bottom-up study of economic phenomena, starting from the individual level. Macro-properties or macro-level regularities emerge from these interactions. Economic theory ignores the communication and direct interaction among agents. ACE modelling provides the modellers with facilities to model social behaviours which is not an easy task when economic theories are used.Furthermore, ACE modelling gives modellers the ability to engage and model social communication skills. This paper studies demand-pull and cost-push inflation by using the ACE modelling technique. The paper investigates the effect of different levels of rationality on the inflation and unemployment rate. Moreover, the model relaxes the homogeneity and anonymity assumptions, which are the key assumptions ineconomic theory when studying inflation. This paper is organized as follows: Section 2 describes the characteristics and benefits of an agent-based computational economics model and explains assumptions underlying this approach. Section 3 describes in brief inflation and the main causes of it. Moreover, it explains demand-pull and cost-push inflation. In Section 4, bounded rationality is investigated.The ACE model is described in Section 5 while results of the simulations are explained in Section 6. II. AGENT-BASED C OMPUTATIONAL E CONOMICS M ODEL (ACE) ACE refers to a set of computational methods which simulate how autonomous agents, interact, communicate, learn, evolve, and make complex decisions. Agents interact locally with each other and with their environment; these local interactionsgive rise to global regularities [7], [18]. The challenge is to explain how these global regularities arise from local interactions of autonomous agents, and to determine the rules that exhibit certain behaviour of the problem under consideration.

Abstract— This paper presents an agent-based computational economics model (ACE) to study demand-pull and cost-push inflation. Moreover, it studies theeffect of different levels of rationality on the equilibrium price and unemployment rate. The model examines three different economies. In the first economy workers choose firms randomly, in the second economy there is loyalty between workers and firms. In the last scenario workers are persistence to find jobs. Simulations show that there is a positive relationship between equilibrium price and levelof rationality while there is a negative relationship with unemployment rate. Moreover, the model is able to reproduce the behaviour of demand-pull inflation and cost-push inflation without homogeneous and perfectly rational agents assumptions.

I. I NTRODUCTION

I

N the real economic world, the dynamic behaviour and interactions between economic agents (for example, firms, consumers andbanks) are very complicated. Economic agents are heterogeneous (i.e., different), autonomous (that is, they have no central authority that controls them), and behave irrationally. Moreover, the growing use of communication technologies affect the market mechanism itself, increasing the dynamic fluctuations of economic systems. Therefore, economic systems are becoming more complex in terms of their...