Simulare per spiegare. Il contributo metodologico degli agent-based models all’economia

Agent-based modelling (ABM) has become increasingly popular in economics over the past decades, offering a powerful framework to capture complex phenomena through an emergentist and micro-founded perspective. Yet questions remain about their methodological contribution. In this paper, I situate ABMs within the broader debate on scientific explanation, arguing that they possess genuine explanatory power. Specifically, ABMs provide how-possibly explanations by reconstructing plausible generative mechanisms and making transparent the causal dependencies between micro-level behaviors and macro-level outcomes. Unlike nomological-deductive approaches, they express causal relations without relying on universal laws, emphasizing processes, interactions, and aggregation. This approach allows ABMs to illuminate how economic phenomena could emerge, highlighting their distinctive methodological and explanatory strengths.