Simulation Models For The Quantification Of Cartel Damages

In the June issue of the German journal Wirtschaft und Wettbewerb (Economy and Competition), Managing Director Nicola Tosini and Felix Bönisch analyze the use of economic simulation models for quantifying cartel damages in litigation in the European Union, taking a German court’s (OLG Stuttgart’s) 20 November 2025 judgement in the European bathroom fittings cartel case as a starting point. In this landmark decision, the court developed a five-step framework for estimating cartel damages. While much of the prior public debate on the estimation framework has focused on the first three steps—establishing the existence and magnitude of a cartel-induced overcharge in the upstream market—this article concentrates on the fourth step, in which the court relied on an economic simulation model to assess how much of the overcharge was passed on to indirect purchasers in the downstream market.

At its core, cartel damages quantification is a problem of missing data. The relevant counterfactual—what prices and quantities would have prevailed absent the cartel—is inherently unobservable. This places cartel damages analysis within the broader economic practice of causal inference, for which outcomes must be reconstructed rather than directly observed. The authors explain that there are two methodological approaches to solving this problem. Comparison methods (also known as reduced-form approaches) infer cartel effects directly from observed data, for example by comparing prices between time periods or markets. Structural approaches, by contrast, specify and estimate an economic model of market behavior and use that model to simulate counterfactual outcomes. Both approaches rely on assumptions, and simulation models differ mainly in that their assumptions are explicitly grounded in economic theory and made transparent.

The article outlines the three steps required to apply simulation models. First, an economic model that captures key market features, including demand conditions, cost structures, and the mode of competition (such as price or quantity competition), is formulated. Second, the model parameters are estimated or calibrated using available data so that the model reproduces observed market outcomes. Third, the simulated outcomes under a competitive scenario are compared with observed or simulated cartel outcomes to quantify the damages.

Dr. Tosini and Mr. Bönisch then illustrate how simulation models can be adapted to case-specific circumstances. They discuss cartels with incomplete market coverage, which can be modeled as a dominant firm facing a competitive fringe, allowing for the analysis of price umbrella effects. They also consider “two-stage mechanisms,” in which cartels not only raise prices but also increase rivals’ costs, thereby weakening external competitive pressure on their members. Finally, they address cartels with asymmetric participants, arguing that joint profit maximization may be unrealistic in such cases and that Nash bargaining models may better reflect internal incentive constraints. 

Turning back to the OLG Stuttgart judgment, the article explains how the court used a simulation model to estimate the passing-on of the upstream overcharge to downstream customers. The court focused on factors such as demand and supply elasticity, competition intensity, and the duration of the infringement. The article notes, however, that in imperfectly competitive markets, the cost pass-through rate depends less on demand elasticity than on demand curvature, a parameter that is difficult to estimate empirically and was ultimately assumed rather than estimated by the court.

The authors conclude that, even if specific modeling choices in the OLG Stuttgart judgement are open to debate, the decision demonstrates that simulation models are practically applicable in court proceedings. Given the intrinsic lack of data on counterfactual outcomes, simulation models provide a coherent, theory-based, and transparent framework for cartel damages quantification that allows courts and experts to assess how assumptions translate into results.