Measuring the Deterrent Effect of European Cartel Law Enforcement

This article proposes a new approach to measuring the deterrent effect of cartel law enforcement by combining a game-theoretic model with Monte Carlo simulations. The game-theoretical analysis shows which type of perfect Bayesian Nash equilibria is obtained depending on the parameter setup: perfect compliance, imperfect compliance or zero compliance. For each equilibrium, we also derive the probabilities of type I (false-positive) and type II (false-negative) errors committed by the cartel authority. To account for the uncertainty and the vague knowledge concerning the model parameters, we perform Monte Carlo simulations based on parameter ranges extracted from the related literature. The simulations indicate that zero compliance dominates the picture and that the error probabilities are high for type II and negligible for type I errors. The results are fairly robust against correlation in the input parameters. Further robustness studies and interactive visualizations can be obtained with a supplemental web application.