DEE/Cade publishes study on the application of Agent-Based Modeling (ABM)


CADE’s Department of Economic Studies (DEE/CADE) launched the working paper titled, “Agent-Based Modeling applied to antitrust.” Drafted for didactic purposes, the paper makes a broad review of the existing literature on the subject, in order to analyze its applicability to antitrust.

Agent-Based Modeling (ABM) is a methodology that describes an economic system based on the representation of the individuals that compose it and their respective behavior. Once the agents capable of interacting with each other have been identified, a system is created that simulates the results from their interaction.

For example, agent-based modeling can be used to analyze cartels, given its origin in the interactions and cooperation between economic agents, as it allows for the creation of controlled conditions, which are necessary to evaluate their formation and effects. This methodology also allows for better understanding of the effects of an interaction between economic agents on factors such as market price. Ultimately, this methodology works as a means of representing economic phenomena and can be used to delineate behavior patterns in various markets.

Despite the potential utility of AMB, the DEE concludes that the work concerning the topic is quite diverse, while still being scarce in terms of antitrust application. For this reason, it believes that more studies are necessary to evaluate the possibility of using this methodology in the area, since, depending on the object of the study, it is necessary to determine sub-models capable of explaining the behavior of individual companies for the observed phenomena, which would require additional independent work.