Synonyms

Amnesty Programs

Definition

A leniency program consists in canceling or reducing any fine or other sanctions against companies, engaged in an illegal cartel, when they report strong evidence of their activity to a Competition Authority.

Leniency program’s objectives

Across the world, one of the main objectives of antitrust authorities concerns the fight against illegal cartels. For that purpose, they conceive new tools to prosecute them and to increase their effectiveness. For more than 30 years, leniency programs constitute a major innovation applied by many competition authorities to fight cartels. These programs were first applied in the USA in 1978: the Amnesty Program. Currently, following the success of the American system, different leniency programs are in effect in most advanced countries (most of the European member states, e.g., Germany since 2000 and France since 2001, Japan, Australia, New Zealand, Canada, China, South Korea, etc) and also in some developing countries. More than 50 countries have adopted different leniency programs.

For competition authorities, the main objectives behind this tool are:

  • To increase their effectiveness in order to uncover cartels by relying on testimony from cartel’s firms (enhance cartel detection, reduce the cost of prosecution)

  • To prevent or reduce cartel formation by making cartel less stable, providing a greater incentive to deviate from cartel agreement and to immediately cooperate with authorities (enhance cartel deterrence)

History

The first American Corporate Amnesty Program, implemented in 1978, granted amnesty to the first company that gave strong evidence of its cartel before any authority’s investigation. This program was unsuccessful (one case by year) for two reasons. Firstly, the amnesty depended on the discretion of the attorney general and required full cooperation of the applicant. Secondly, before any investigation by an authority, the individual risk of conviction is low, and the first American program gave no sufficient incentives to be effective. Hence, in 1993, the US Antitrust Division of the Department of Justice decided to review the program. The current program also provides full amnesty to the first company revealing information even after the authority’s investigation has already begun and until the authority has sufficient evidence to condemn the cartel. Since this reform, there are one or two leniency applications per month.

In 1996, the European Union has also implemented its own leniency program. However, after 5 years of implementation, results appeared to be mixed. There was no certainty that the first applicant would obtain complete immunity. In fact, less than half immunity was granted. In addition, once the Commission had begun an investigation, it was only possible to obtain a maximal fine’s reduction of 75%. That is why the policy was first improved in 2002 and in 2006, with the objective of a greater transparency and certainty.

Current conditions are as follows:

  • The Commission cancels any fines to the first applicant that provides substantial evidence of the cartel, allowing the authority to implement an investigation pursuant to Article 14 (3) Regulation No. 17 or to prove an infringement of Article 81 of the Amsterdam Treaty.

  • In addition, if another company provides further information, it may receive a fine reduction (first firm to cooperate, reduction of 30–50%; second firm to cooperate, 20–30% reduction; other firms to cooperate, reduction of up to 20%).

Direct and Indirect Effects

Theoretical analyses have shown that leniency programs can increase the incentive to cooperate with authorities and may have a deterrence effect on collusion. However, they may also have a pro-collusive effect (Motta and Polo 2003). Indeed, leniency programs can increase the benefits of a cartel’s member as they reduce the expected cost of the antitrust sanction. Hence, it is possible that the number of cartels increases because of leniency. Different empirical studies seem to conclude that the positive effect on cartel destabilization and on cartel deterrence dominates the pro-collusive effect. However, empirical identification is only derived with data from detected cartels (Brenner 2009; Miller 2009). Therefore, if the success of a leniency program depends on the number of uncovered cartels, it may mean that there are more current cartels and not necessarily more efficient prosecution.

To reduce this pro-collusive effect and to increase the effectiveness of a leniency policy, it is necessary to have higher sanctions against cartel’s members and to have a real power of investigation. Indeed, if fines are weak and if the probability of an investigation is low, the expected sanction will be low. Then, cartel’s members will have no incentive to report to the competition authority because the risk of a conviction is weak. That is why, since 2002, the European Commission has decided to increase the level of fines against cartel’s members. The maximal fine was doubled and represents 10% of a company’s global turnover. This policy has increased the incentives to apply for leniency.

In contrast with the European program, the US program only grants exclusive amnesty to the first confessor. This strict program may create a “preemption effect” and may increase the deterrence effect (Harrington 2013). Indeed, even if one member thinks that the competition authority has a low likelihood to undermine the cartel, he may also think that another member may report and apply for leniency. In these conditions, he can have a strong incentive to quickly cooperate with authorities to preempt other members and to benefit from the fine’s reduction. With the European program, second and third firms revealing strong evidence receive reductions in sanctions as well, but lower than the first confessor. Hence, the “preemption effect” exists but is less powerful than the one in the American program. It is also important that the gap in the reduction of fines be sufficiently great to increase the incentive to be the first to cooperate.

Can a Ringleader Apply for Leniency?

In the current US leniency program, a firm may only be eligible for amnesty when it is not a “ringleader.” This clearly means that it is not at the origin of the cartel and that it has not invited any other firm to take part in the cartel (organize meetings, communication between cartel members, etc.). The first European leniency program, between 1996 and 2002, applied discrimination to the ringleader, too, but it was less strict: a ringleader was not illegible for full amnesty but could only have smaller fine reduction between 10% and 50%. However, the current program, following two revisions in 2002 and in 2006, is more accurate and allows any cartel’s firm, even a leader in the cartel, to benefit from full immunity provided it is not the cartel’s instigator (which organizes and initiates the activity).

What is the economic impact of excluding the ringleader? If a firm, as a ringleader, is excluded from amnesty, it increases his potential sanction’s expected cost in comparison with other members. Then, this ringleader can ask the others for a monetary compensation. This can create an asymmetry between this firm and the others and may decrease the sustainability of the cartel. Indeed, the result depends on the probability to be reviewed by the competition authority (Bos and Wandschneider 2011). If this probability is rather low, including the ringleader in the program may increase the probability of a successful prosecution. In contrast, if this probability is rather high, it may be better to exclude the ringleader to reinforce asymmetry with others.

How to Improve Leniency Programs?

Since the first American leniency programs, many measures have been adopted to improve their efficiency.

Leniency Program for Multimarket Firms

Multimarket contacts can facilitate the sustainability of collusion (Bernheim and Whinston 1990). In practice, cartels may concern many markets. That is why in 1999, the USA has adopted the Amnesty Plus program. Its objective is to convince a firm convicted in a first market to report evidence on a collusive agreement in another market. Hence, this firm could obtain amnesty as a cartel member in the second market if it is the first one reporting this cartel and a fine’s discount as member in the first cartel. Theoretical conclusions on the Amnesty Plus reform are not clear. Indeed, it has two opposite effects on companies’ incentive to collude. It may reduce the deterrence effect by increasing the sustainability of multimarket collusion. However, it can reduce the cartel’s expected duration by increasing the risk of report after the detection of the first cartel (Lefouili and Roux 2012). If the program is rather lenient, it would seem that the procompetitive effect is stronger.

Leniency Program and Fight Against International Cartels

In practice, many cartels concern international markets and may be prosecuted by different jurisdictions. In the USA, 90% of recent fines for antitrust infringement result from international cartels. Information sharing among competitive authorities of different jurisdictions could increase fight against cartels, reduce the cost, and improve the effectiveness of an investigation (reducing time to obtain strategic information, improving exchange of strategic documents). However, information sharing between different authorities could reduce the effectiveness of leniency programs as well (Choi and Gerlach 2012). Indeed, confidentiality is essential and is a necessary condition to be credible and to protect the integrity of leniency applications.

A More Incentive System: Reward for Whistleblowers

In April 2005, the South Korean Fair Trade Commission has introduced a reward system to give more financial incentive for outsider whistleblowers to reveal information and provide evidence on secret cartels. Hence, in June 2005, an anonymous person who revealed a welding rod cartel, providing strong evidence, received a reward of $63,700. In the USA, through the False Claims Act, there is a similar financial reward mechanism to fight fraud against the government, but not for violations of competition law. In Europe, no reward system for informers exists. Theoretical predictions suggest that rewarding cartel members for reporting is a powerful tool to deter cartel formation (Aubert et al. 2006; Buccirossi and Spagnolo 2006). It increases the risk of an individual deviation, bringing more incentive to report. It can also increase the cartel’s expected cost because firms will have to “bribe” individuals to prevent them to report. However, a first experimental analysis concludes that rewarding whistleblowers does not further deter cartel formation in comparison with traditional leniency (Apesteguia et al 2004). But a second one, assuming a dynamic setting, predicts that rewards increase cartel detection due to self-reporting (Bigoni et al. 2012).

Leniency Program and Administrative and Criminal Enforcement

In the USA and in South Korea, there are criminal sanctions for individuals engaged in cartel activities (custodial sentences). In the USA, since 1994, a criminal leniency has been implemented. Moreover, since 2004, an individual involved in a cartel can be imprisoned for up to 10 years. Most of European member states have the possibility to impose criminal sanctions on individuals for cartel conduct. For example, in France, individuals can face jail time for up to 4 years and may have a criminal fine for up to 35,000 euros. Moreover, some European states (UK, Ireland, and Austria) have both administrative and criminal leniency programs. In these countries, the competition authority will not prosecute an individual if he satisfies conditions for leniency. Other member states (Belgium, France, Germany, Slovenia, Estonia, Ireland, etc.) do not have any criminal leniency program. In practice, jail sentences are indeed rarely imposed. Moreover, currently, the law of the European Union does not provide for criminal sanctions, and each member state is free to impose its own system of penalties for antitrust infringements. In contrast, in the USA, there were many jail sanctions against chairmen of companies engaged in cartel activities. The idea is that jail sentences are a great disincentive for individuals to participate in a cartel. In Europe, some officials think that high administrative fines are sufficient to deter cartel conduct. Both mechanisms are indeed complementary. With an administrative leniency, the efficiency only depends on the incentive of cartel companies. With an individual criminal leniency, the impact depends on the objectives of an employee concerned by a criminal sanction, but it may also concern his company itself. Indeed, if the company anticipates the risk of report by its employee, it could have more incentives to apply for a corporate leniency.

Cross-References