The Shortcomings of the Leniency Agreement Provisions of Brazil’s Clean Company Act

If the CEO of a corporation operating in Brazil learns that her company has committed an unlawful act of corruption, should she order the corporation to self-report and negotiate a leniency agreement with the Brazilian authorities under Brazil’s 2013 Clean Company Act, which authorizes such settlements? In most of the cases, the corporate legal department would probably advise against it. Indeed, the number of leniency agreements based specifically on Brazil’s Clean Company Act has been much smaller than expected.

Several factors drive companies away from cooperating with Brazilian public authorities under the Clean Company Act:

Continue reading

Colombia’s Harsh Criminal Penalties for Corruption Are an Illusion. Here’s How To Fix That.

Whenever a new corruption scandal comes to light, many politicians instinctively react with strong punitive rhetoric, and this rhetoric often translates into action, usually in the form of amendments to criminal codes that make penalties for corruption offenses harsher. Latin America supplies plenty of examples of this (see here, here, here, and here.) Yet despite this emphasis on punishment, many corrupt politicians avoid justice altogether, and in the rare cases where they are found guilty, many end up doing only short stints in comfortable detention centers. Consider, for example, Colombia, which has unusually good public data on corruption convictions and sentences thanks to the work by the Anticorruption Observatory of the Secretary for Transparency. According to this data, between 2008 and 2017, criminal courts in Colombia have convicted 2,178 individual defendants for corruption (51.2% for bribery, 23% for embezzlement, and the remainder for other corruption-related offenses), but only about one-quarter of these convicted defendants actually went to prison. Approximately half of these defendants received suspended sentences, while another quarter were sentenced to house arrest. And of those who did go to prison, the time served was only about 22 months on average, much lower than the penalties on the books for corruption offenses. No wonder many Colombians believe the criminal justice system is too lenient.

The reason that actual Colombian sentences end up being so light, despite the penalties on the books being so heavy, is that Colombian law includes a set of provisions that allow for a variety of sentence reductions if certain conditions are met. For example, a defendant who accepts guilt can receive a 50% reduction in his prison term. Inmates may also reduce their prison term through work, with very generous terms: An inmate reduces his sentence by one day for every two days of ordinary work (8 hours of work per day), or for every four hours of work as a teacher. An inmate can also reduce his sentence through in-prison education, with  six hours of study translating into one day of sentence reduction. Furthermore, once an inmate has served 60% of his sentence, he can petition for release for good behavior. 

This excessive leniency needs to be addressed, not only in corruption cases but in all cases. Specifically, Colombia should adopt the following revisions to its criminal laws: Continue reading

How Anticorruption Enforcement Can Undermine Antitrust Amnesty Programs, and What To Do About It

One of the most important law enforcement techniques that has emerged in the last few decades to combat cartels (anticompetitive collusion between competitors) is the use of programs that promise automatic amnesty to the first member of a cartel to self-report the illegal enterprise. These amnesty programs enable law enforcement authorities to gather the evidence they need to build strong cases against other members of the scheme, and, perhaps more importantly, these amnesty programs destabilize cartels—and might even deter their formation—by taking advantage of the incentive that individual cartel members have to cheat on each other. Since the 1990s, after the success of the amnesty program pioneered by the Antitrust Division of the U.S. Department of Justice (DOJ), antitrust amnesty programs have been replicated in many jurisdictions, leading some to declare a “leniency revolution” in competition law.

But the existing amnesty programs have a weakness They usually only offer protection for violations of antitrust laws, leaving even the firm that self-reports the antitrust violations potentially liable for other unlawful conduct that the cartel members engaged in as part of their anticompetitive scheme. And many of these anticompetitive schemes turn out to involve corruption, especially in the public procurement context. Cartels often bribe the official in charge of the procurement process, because a corrupt official can monitor and punish defections from the cartel, facilitate the exclusion of non-aligned competitors, and ensure an equal distribution of cartel profits. A firm that hopes to take advantage of an antitrust amnesty program might have to report all of this to qualify for amnesty, as often the programs require, as a condition for amnesty, reporting on the involvement not only of other cartel members, but of any public officials who may have facilitated the collusive conduct. But the fact that a self-reporting cartel member is not guaranteed amnesty from prosecution for corruption or other associated wrongdoing (such as money laundering) complicates the operation of antitrust amnesty programs, because this lack of guaranteed amnesty weakens the incentive of cartel members to self-report in cases where the cartel has engaged in bribery. The problem is especially pronounced when the penalties for bribery are much more severe than those typically imposed in cartel cases.

This is less of a problem in jurisdictions where anticorruption and antitrust authorities are departments of a single agency, as with the US Department of Justice (DOJ). But in many other jurisdictions, such as the EU, Brazil, and Mexico, competition law enforcement—and administration of the antitrust amnesty programs—are handled by enforcement agencies that do not have authority to prosecute corruption cases. From a potential self-disclosing company’s perspective, this poses a challenge: Disclosing participation in a bribe-paying cartel to the competition authority may also trigger an enforcement action by the separate agency responsible for prosecuting corruption, meaning the company will have to negotiate with both agencies, with the anticorruption agency not bound by the antitrust amnesty program. Indeed, in many countries anticorruption agencies may not have the same authority as antitrust agencies to grant leniency to self-reporting companies. In Brazil, for instance, though an antitrust amnesty program has been in place since 2000, settling corruption cases only became possible in 2014. In Mexico, the antitrust amnesty program was created in 2006, but a program for self-reporting bribery cases only entered into force in 2016. In both countries, although there is an established process for settling corruption investigations, there is no immunity provision for self-reporting; a discount in the applicable fines is often the best a firm can hope for. And even when both the antitrust agency and the anticorruption agency have authority to settle and grant leniency, the mere fact that a company knows it will need to enter into two or more separate negotiations increases the uncertainty and costs associated with self-disclosure, undermining the effectiveness of the amnesty program.

How should this problem be addressed in those countries where merging authority over antitrust and anticorruption enforcement in a single agency is not feasible or desirable? There are several possibilities:

Continue reading

Leniency Agreements Under Brazil’s Clean Company Act: Are They a Good Idea?

Brazil’s 2013 Clean Company Act, the country’s first anti-bribery statute applicable to companies, has grabbed Brazilians’ attention due to its recurrent use in the context of the so-called Car Wash operation. The Clean Company Act has provided the main legal basis for Brazilian public authorities (especially federal prosecutors) to sign leniency agreements with construction corporations whose top executives stand accused of bribing officials in exchange for contracts from Petrobras, Brazil’s state-owned oil giant. Under the Act, Brazilian authorities may enter into a leniency agreement as long as the company admits its participation in the illicit act, ceases any further participation, provides full restitution for damage caused, and cooperates fully and permanently with the ongoing investigation. In exchange, the fines can be reduced by up to two-thirds and, more importantly, the cooperating company may be exempted from judicial and administrative sanctions, including suspension or debarment from public contracts. Over the course of the Car Wash investigation, Brazilian authorities have already signed five leniency agreements with some of Brazil’s largest engineering firms, and at least twelve more companies are currently negotiating leniency deals with Brazilian authorities.

But do these sorts of leniency agreements provide for sufficient deterrence of corrupt behavior? And are they consistent with the interest in punishing those companies that have committed a serious crime? Those who defend Brazil’s increasing use of leniency agreements emphasize that a similar approach has proven to be effective in countries like the United States, one of the most successful countries in the world in the fight against corruption. Indeed, the leniency agreements authorized by the Clean Company Act were modeled on the Non-Prosecution Agreements (NPAs) and Deferred Prosecution Agreements (DPAs) used by US authorities in white-collar criminal law enforcement. However, Brazil is following the US model precisely at a time when the widespread use of NPAs and DPAs is becoming more controversial, in part because of concerns that these sorts of agreements fail to deter economic crimes and allow high-ranking executives to escape accountability for their crimes (for a summary of the criticisms of those agreements, see here and here). Perhaps more importantly, even if one views the US experience with NPAs and DPAs as successful overall, there are several reasons why this model might be more problematic in the Brazilian context. Continue reading