Serial Collusion by Multi-Product Firms

William E. Kovacic, Robert C. Marshall, and Michael J. Meurer

That companies sometimes cheat is a given.  But what happens when the cheating becomes part of a company’s long-term business strategy?  New research shows that current enforcement strategies and mechanisms are insufficient to prevent serial collusion by large multi-product firms and that these firms routinely exploit the weaknesses in the system to their benefit and the detriment of consumers.  This paper examines why and how firms get away with their serial collusion and offers some policy recommendations for combating it.

Analyzing decades of European Commission (EC) decisions on collusion in various industries, the paper shows that many firms have been repeatedly caught colluding in several, and as many as sixteen, products.  These cases, too, represent only those instances where the colluders have been caught; there are doubtless others that have escaped detection.

Current enforcement strategy relies on the assumption that each cartel is the work of a rogue employee. However, the typical responsible managers the EC identified in these investigations tend to be top officers at the firms.  The EC has also found that one firm has served as a consulting firm to every chemical firm found guilty of collusion more than once, which suggests that not only is the “rogue employee” scenario false, but decisions to engage in collusive behavior have the full support of the firm as part of its business strategy.

Firms engaging an outside consultant to facilitate their serial collusion shows how much the current enforcement regime has failed.  It also suggests that top management and perhaps even shareholders sometimes welcome cartels as tools for increasing profits — even after factoring in the expected costs of detection and punishment.  If the top management is repeatedly involved (and caught) in collusive behavior, then clearly the deterrence in place is ineffective, and tougher sanctions on firms and corporate officers may be necessary. However, any policy reform must account for the clever, malleable, adaptability of experienced cartel firms.

As a first step, enforcement agencies need to make better use of the data already available in investigating and sanctioning collusive behavior, such as judicial decisions, government reports, and academic research, and look for larger patterns of cartelization and not assume that each cartel is the work of a single employee.  

Next, just as the Federal Aviation Administration will examine the factors leading up to an airline incident in great detail in order to understand the incident to avoid its recurrence, so too should the Department of Justice (DOJ) work to reconstruct a cartel to understand how it worked, who was responsible, and what other markets could be affected, either by that particular cartel, or by possibly related cartels.  This reconstruction would, ideally, identify the features and links that might lead to collusive behavior in the future.

Third, the DOJ should include close monitoring provisions as part of any sentence, settlement, or plea arrangement.  Serial cartel participants should be subject to progressively more stringent monitoring by law enforcement. Monitoring would help enforcers identify red flags of collusion and guide merger review where it intersects with anti-cartel enforcement.

Fourth, the DOJ should reform current leniency programs and add bounty programs to reward smaller firms for reporting a cartel.  Serial multi-product colluders are currently able to manipulate the leniency system, appearing to be contrite in one cartel while simultaneously continuing another cartel and exploiting smaller single-product firms.  Enforcement agencies, because they tend to view cartels as the act of rogue employees, often fail to see the connection between simultaneous cartels, and grant leniency in such cases. Such an approach needs to change, and there needs to be greater incentive for small single-product firms to be whistleblowers against the multiproduct firms, such as by paying them a percentage of the fines recovered from the other cartel firms.  

Fifth, any proposed merger involving a firm that has engaged in serial collusion in the past must be carefully reviewed for the possibility of coordinated effects as a result of the merger.  The results of cartel reconstruction and settlement and monitoring should be reviewed before any merger is approved involving any serial cartelist.

Finally, cartel facilitators should be subject to the same penalties as the other participating firms, especially because they are able to act more broadly, and longer term, applying know-how acquired in one cartel in another later.  They should also have to help with reconstruction.

The current approach to investigating and punishing serial multi-product cartels does not work.  Although no system will be able to discover all collusive behavior, the current system’s failure to consider that some firms consider collusion to be a means of doing business, and thus engage in serial collusive agreements across multiple products, benefits these firms and harms smaller companies and ultimately the consumer.  Some straightforward policy changes, and an eye on the bigger picture, can help reduce this collusive behavior.

Full text in Journal of Antitrust Enforcement, vol. 6, no. 3, 296-354 (Oct. 2018).