Making cartels pay for umbrella damages may be desirable, but not for the reasons you think…

A right to compensation for umbrella effects of cartel formation, or harm resulting from non-members’ economic reactions, was established by the EU Court of Justice ten years ago (Case C-557/12 Kone AG v ÖBB-Infrastruktur AG). Members of a cartel are consequently accountable for overcharging for both their own sales and those of their rivals who are not violating the law. According to the 2014 ruling and similar rulings from US and Canadian courts, this accountability structure deters antitrust violations and makes “a significant contribution to the maintenance of effective competition.” New research by Stefan Napel, Professor of Economics at University of Bayreuth, and Dominik Welter, Adjunct Lecturer at University of Bayreuth and competition consultant argues against this: making identified cartels pay for umbrella losses incentivizes the formation of more encompassing cartels. This tends to increase prices and can, on balance, harm rather than help consumers.

To illustrate this, let’s suppose that a cartel comprised only three out of four producers in an industry when firms were not liable for umbrella losses. The prospect of now having to pay damages also to customers of the non-member, in the event of detection, reduces the cartel’s marginal returns on its price increases. This lowers optimal prices and expected profits of the cartel. Assume that this actually renders the infringement unprofitable. Then liability for umbrella losses has deterred the given cartel.

However, a cartel’s size is endogenous: all firms in the industry will adapt to changes in the legal regime. In particular, the cartel outsider who could previously raise its prices under the umbrella of the cartel faces a new situation. It could formerly not be convinced (and perhaps was not even asked) to become a member because undercutting cartel prices was more lucrative than its prospective share of collective profits. But the cartel of three is no longer profitable, cannot form, and cannot be benefitted from. The non-member is now better off joining the others and forming a – still profitable – industry-wide cartel. The same holds if the original cartel remains viable but its re-optimized prices fall sufficiently below the (unchanged) prices that an all-encompassing cartel can sustain.

The problematic feature of liability for umbrella damages is that a smaller market coverage means a larger number of non-member customers who can claim losses. This hurts expected profits of a cartel more, the smaller its market share. Umbrella compensation thus destabilizes small cartels and, conversely, increases the structural stability of big cartels. The latter commonly set prices closer to the monopoly level, and this can reduce consumer welfare even after accounting for expected compensations. In particular, liability for umbrella losses leaves profits of an industry-wide cartel unchanged and lowers them for all partial cartels.

Most formal statements in the study of Napel and Welter are derived in a model of Bertrand-Edgeworth competition (see Bos and Harrington, Rand J. Econ. 41(1), pp. 92-117, 2010). But the highlighted cartel size effects and their tendency to reduce rather than encourage competition rest on three rather general conditions: (i) stable partial cartels can emerge in the default regime; (ii) the new regime in which cartels must compensate umbrella losses does not prevent all cartel formation; and (iii) bigger cartels set higher prices ceteris paribus. These conditions are likely to be met in a wide range of industries. Thus, when courts or policymakers give reasons for making cartels compensate umbrella damages, they should better emphasize legal principles and distributional justice – not deterrence and effective competition.

(Full paper published in IJIO Volume 91, December 2023)