The Methodology and Multifaceted Role of Models in Economic Theory: Between Abstraction and Application

Models have become an essential to contemporary economics, but the methodological underpinnings of these models and their philosophical position are still being debated. This paper critically examined the methodological foundations of economic modelling, and specifically through the lens of abstraction, assumptions, formalization, and calibration versus econometric estimation. The article discussed economic models as an explanatory tools that elucidate the processes of causation, predictive tools that help forecast future, policy laboratories to simulate interventions, and conceptual models that help speculate the theoretic implications. Meanwhile, some important limitations were identified. The weaknesses of model-based reasoning were underscored by over-simplification, the issues brought up by the Lucas Critique as well as the philosophical arguments on realism versus instrumentalism in general. The analysis puts these tensions in the context of the poles of abstraction and application implying that the strength of models is not on their correspondence with reality but rather on their ability to organize the economy in a logic manner and to direct one to its inquiry. The article concludes with the identification of new methodological horizons such as agent-based modelling and machine learning that could change the role of models in economic theory.