With the exception of the United States, individual cartelists are rarely subject to criminal proceedings. However, it cannot be ruled out that a managers may obtain private gains from their cartel participation, and therefore that they might have a personal incentive to set them up.

This article, available here, analyses the incentives for a manager to engage in a cartel by mobilising the theoretical framework of the ‘economics of crime’. It also examines the various solutions – both at company and public authority level – to limit individual incentives to engage in this type of practice.

Section II looks at the costs and benefits for a manager of participating in a cartel.

In most detected cartels, individuals who participated in the practice held relatively high positions within their company: they were often commercial directors, and sometimes even general managers or CEOs. This means that, typically, a cartelists’ remuneration includes a large variable part linked to the achievement of short-term objectives. Managers may also want to take risks to outperform their peers internally, and progress more quickly within the company or a sector.

When considering participation in a cartel, a manager might consider three types of costs. The first is the existence of criminal sanctions against individuals in the jurisdiction where the practice is implemented. In practice, only the United States pursues criminal proceedings against individuals, including prison sentences. Outside the United States, few jurisdictions in the world prosecute individuals. Countries such as Ireland (in 1996), the United Kingdom (in 2002) or Australia (in 2009) have introduced criminal sanctions but their effectiveness remains limited in practice. Secondly, while it is difficult to estimate the psychological cost associated with fraud, the degree of stigmatisation of cartels by society can also influence the manager’s behaviour. Surveys on cartel perception conducted in countries such as Australia (2011), the United States, the United Kingdom, Germany, Italy (2014) or France (2018) reveal that a majority of citizens dislikes these practices. Finally, the manager will take into account any internal costs or sanctions that might be imposed by their company, such as a dismissal or reprimand. This last cost depends on the degree of commitment of the company’s shareholders to comply with the law.

Section II also looks at psychological biases that can influence a manager’s perception of costs.

When a manager assesses the costs of participating in a cartel, his assessment may be affected by ‘psychological biases’ which lead him to underestimate their magnitude. For example, as soon as a company’s hierarchy approves an illegal practice, this might convince the manager that its action is well founded. The perception of illegality of a practice may be reduced if breaking the law is presented as unavoidable’ to achieve a legitimate purpose (e.g. make a profit). Perceptions of a cartel’s illicit nature may also be influenced by the fact that a cartel is a collective practice, carried out by individuals who perceive themselves as ‘peers’. The unlawful nature of a cartel can be blurred by managers’ poor knowledge of competition law, particularly in small businesses. When participation in a cartel is progressive and organised in several stages, from general discussions to pricing meetings, this gradualism may also reduce perceptions of illegality. Finally, the fact that the damage caused by the cartel is not directly visible to the perpetrators facilitates the adoption of unlawful behaviour, unlike offences targeting specific persons.

Section II then looks at the cost-benefit analyses pursued by managers participating in cartels.

After comparing gains and costs, a manager will also assess the risk that the cartel will be detected and condemned by antitrust authorities. The probability of detection has been estimated by the literature at 10% to 30%. However, managers may underestimate the objective probability of being caught due to two cognitive biases. The first is ‘availability bias’: the small number of cases handled each year by antitrust authorities, combined with their low media exposure, can lead a manager to underestimate this probability, or even think it is zero. The second behavioural bias is ‘overconfidence’: an agent, faced with decisions whose outcome is uncertain, tends to underestimate the occurrence of adverse outcomes, think that he will do better than others and overweigh favourable events. After all, the probability of detection of a cartel is not independent of the ability of the agents to conceal it. Detection is in part ‘controllable’, leading to overconfidence.

Section III looks at corporate mechanisms to discipline managers.

As long as the perceived individual costs of participating in a cartel are limited (no criminal sanctions in Europe) or reduced (behavioural bias), and while the anticipated gains are high, managers may be encouraged to cartelise. The question then is how to ‘discipline’ them.

If shareholders do not wish to engage their company in a cartel practice – out of interest or conviction – they can mobilize several internal levers. First, companies can modify the conditions for granting or exercising variable remuneration. Rather than limiting performance-based remuneration (which can have positive effects on productivity and risk taking), it is more effective to provide for bonus or capital gain refunds if the manager has breached competition law. Another mechanism is to include internal sanctions in employment contracts, up to and including dismissal, in the event of an infringement of competition laws. Finally, compliance programs and internal audits are part of a culture of self-regulation and can reduce the cognitive biases identified above, in particular by preventing the development of values within the company that are contrary to antitrust law.

Section IV looks at public enforcement against individual managers.

Corporate mechanisms face serious limitations – due to agency problems, monitoring costs, or even due to the absence of shareholder incentives to prevent collusion.  As a result, it may be optimal to implement public policies targeting individuals.

First, if managers are to be discouraged from engaging in cartel practices, it is useful to develop a ‘competition culture’ to increase the ‘psychological cost’ of violating competition rules. This is a long-term process requiring efforts through multiple channels, such as communication campaigns, better media coverage of cartel sanctions, political initiatives and the training/teaching of business leaders. Second, public authorities may also introduce sanctions against individuals. In addition to overcoming companies’ inability to pay ‘optimally deterrent’ fines, criminal sanctions can enhance the attractiveness of leniency programs, while sending a strong signal about the social stigmatisation of cartel practices. However, imposing fines on individuals is unlikely to be effective, and the imposition of prison sentences has met serious obstacles concerning in particular their perceived lack of proportionality and legitimacy. Disqualification measures may constitute a middle way: an executive or corporate officer who has directly contributed to violating competition rules or who has ‘allowed’ managers to act in full knowledge of the facts may be prohibited from exercising any position of responsibility within a company for a certain period.

Comment:

This paper provides an interesting discussion of the literature on individual sanctions for anticompetitive conduct. While not containing original theoretical insights, the paper systematises the available materials in a way that lends them practical usefulness. I found the section on behavioural biases to be particularly interesting, and very much enjoyed the balanced way in which the authors assess the potential effectiveness of each individual measure they discuss. 

Author Socials A weekly email with competition/antitrust updates. All opinions are mine

What do you think?

Note: Your email address will not be published