Agent-based model in labor market as a third way to micro/macro relation: a new paradigm?

This paper presents an agent-based model where is shown in which circumstances cooperation appears from a heuristic process and how equilibrium endogenously emerges from the decentralized interactions of adaptive and autonomous agents. This modeling strategy is presented as a promising way to avoid the burdens of traditional rational agent conceptualization, for instance, the link between micro and macro level, a fragility overtaken by the idea of “aggregation” and the ubiquitous concept of “representative agent”.
A human subject experiment inspired this agent-based computational model in labor market. The model defines two sorts of agents – firms and workers – with different payoff functions and learning abilities which allow them to develop a behavior rules repertoire selectively activated. In a frame of incomplete contracts each wage offered by the firm will be associated to an effort level chosen by the employee. According to the obtained payoffs, rules will be redefined and new payoffs will arise in the next iteration.
This process is repeated a large number of iterations. Then, slight changes will be introduced to the model (e.g., reciprocity, diversity and unemployment) for the purpose of comparing outcomes. How do different features create an efficient coordination among divergent micromotives? Agent-based computational simulations can help explaining the complex relations between micro and emergent macro realities and, consequently, give another perspective about interaction structures, their heterogeneity and agents evolutionary and adaptive processes. This work broad aim is to show that bridging the analysis of individual agents and system behavior is where the promises of gains are.
This work is organized as follows: Section 1 introduces a methodological and
conceptualization overview. Section 2 brings in Labor Computational Economics.
Section 3 introduces the formal model. Section 4 provides the implementation details.
Section 5 exposes the simulation results. In Section 6 we summarize.