Deferred prosecution agreements (DPAs) are alternative ways to settle criminal cases via deals reached between public prosecutors and corporations charged with a criminal offense (Garrett, Reference Garrett2014; Reilly, Reference Reilly2015, Reference Reilly2018; Arlen, Reference Arlen, Søreide and Makinwa2020). These negotiated settlements allow corporations to avoid trial and defer prosecution subject to agreed conditions such as the admission of the facts, the payment of a reduced fine and compensation of victims, and the reform of the corporation’s compliance programs (Arlen & Kahan, Reference Arlen and Kahan2017; Reilly, Reference Reilly2015). If the agreed conditions are not met, prosecution will resume (Ryder, Reference Ryder2018). If they are met, the case is closed without prosecution and criminal conviction (Hess, Reference Hess2009; Greenblum, Reference Greenblum2005; Reilly, Reference Reilly2015; Spivack & Raman, Reference Spivack and Raman2008).
DPAs date back to the early 1990s within the US federal system (Reilly, Reference Reilly2015; Spivack & Raman, Reference Spivack and Raman2008). Their increased use in the US has been linked to the default of the Chicago-based accounting firm Arthur Andersen LLP (one of the world’s largest multinational enterprises [MNEs]) resulting from its criminal conviction, later overturned by the US Supreme Court (O’Sullivan, Reference O’Sullivan2014; Ryder, Reference Ryder2018). Arthur Andersen’s sentence and subsequent collapse followed its refusal to settle the case through a DPA (Markoff, Reference Markoff2013). Since then, MNEs and US enforcement agencies, to avoid corporate death and overcome the devastating economic consequences of what has come to be known as the “Andersen effect,” have increased the use of these agreements instead of embarking on costly trials and convictions (Markoff, Reference Markoff2013). Markoff (Reference Markoff2013: 807) estimates that the number of DPAs increased from “a handful per year before Andersen to an average of approximately 30 per year,” and the majority of business organizations that reached such settlements were MNEs.
The growing uptake of DPAs stems from their US adoption in lieu of trials, especially (but not solely) for cases of international corruption by MNEs from different jurisdictions (Makinwa, Reference Makinwa2015; OECD, 2019; Søreide & Makinwa, Reference Søreide and Makinwa2020). Over the last decade, these pretrial diversion agreements have been increasingly transplanted into other national legal systems, such as England, Wales, and France (Arlen, Reference Arlen, Søreide and Makinwa2020; Grasso, Reference Grasso2016). The spreading of DPAs also allows MNEs to reach global settlements through agreements with multiple law enforcement agencies for cases of foreign bribery of public officials.
A recent prominent case concerns Airbus SE. On January 31, 2020, this French, German, and Spanish-owned MNE, agreed to pay—through a global DPA settlement—US$3.9 billion to the Department of Justice (DOJ) in the US, the Serious Fraud Office (SFO) in the UK, and the Parquet National Financier (PNF) in France to resolve a bribery case of foreign public officials (DOJ, 2020). Significantly, though, two years later, the Tribunal of Paris approved another DPA between Airbus SE and PNF for yet another claim of transnational corruption (Parquet National Financier, 2022).
The American, English, and French experiences, in addition to those of other countries such as Canada, have attracted the attention of scholars and legislators alike worldwide (Arlen, Reference Arlen, Søreide and Makinwa2020). Yet, the important conversation about DPAs is still in its infancy and, thus far, has been based mainly on legal grounds (Arlen, Reference Arlen, Søreide and Makinwa2020) to the neglect of ethical ones. Existing academic research about DPAs, mostly hosted in the pages of law journals, focuses primarily on the merits and demerits of the jurisprudence of these agreements and their specific implications for criminal enforcement, with insufficient attention paid as to whether DPAs can be regarded as morally acceptable.
Doubtless, the diffusion of such compromise-based agreements raises significant ethical concerns. At its simplest, when translating legal discourse into the ethical domain, two broad lines of thought can be sketched. The first, in stark opposition to DPAs, would suggest that these settlements are deontologically unethical as they offer a far too lenient and hence unfair alternative form of criminal punishment and justice thereby favoring corporate misconduct. Specifically, the fact that MNEs culpable of foreign corruption, can—thanks to DPAs—reach agreements with enforcement agencies to reduce the impact of criminal prosecution and conviction, makes such agreements inherently immoral. The second line of thought, in support of DPAs, is that a utilitarian perspective would nonetheless justify their use. After all, MNEs may not self-disclose violations and cooperate in their investigation if it were not for DPAs, thus decreasing the probability of a criminal conviction if the case went to trial. Accordingly, by foregoing formal convictions and imposing the reform of compliance programs aimed at preventing recidivism, DPAs would reduce the costs and increase the benefits for both law enforcement agencies and MNEs.
Despite this fundamental dichotomy, only sparse, short reflections on the ethics of DPAs can be found in the business ethics literature (Bertels, Cody, & Pek, Reference Bertels, Cody and Pek2014; Weismann, Buscaglia, & Peterson, Reference Weismann, Buscaglia and Peterson2014).
The aim of this article is to subject DPAs to critical, ethical scrutiny and, in so doing, contribute, albeit in small part, to what Ryan (Reference Ryan2000: 337) described as the “greatest ethical challenge for the 21st century,” namely, the need “to combat and conquer corruption and develop a true global ethic” (Ryan, Reference Ryan2000: 337). Corporate corruption or the misuse of resources or power for private/corporate gain, matters. Although it is often portrayed as a “victimless crime” (Ruggiero, Reference Ruggiero and Ruggiero2001: 106), it has devastating and wide-ranging effects, not only on the healthy operation of the economic system of production but also on the very fabric of society by undermining sound political and social decision-making. It is no coincidence that “Substantially reduce corruption and bribery in all their forms” also features as one of the targets of the United Nations (UN) Sustainable Development Goals (see SDG 16.5, United Nations, n.d.) that were agreed in 2015 by all UN member states as a global call to action to achieve a better future for all.
In our present endeavor, we address two fundamental ethical dilemmas: Is a DPA, an agreement that provides culpable MNEs with a “privileged exit route” from prosecution, morally tolerable from a deontological perspective? And, from a utilitarian perspective, is a DPA capable of fostering compliance with the law and ethical business behavior? Or are DPAs to be deemed unethical from both perspectives?
To address these questions, we begin by unpacking the ethical issues underlying the normativity of DPAs while also delving into their specific implications for MNEs culpable of international corruption. The focus on MNEs allows us to consider key ethical issues at stake, especially since DPAs typically involve large and powerful corporations in what have been described as battles between David and Goliath (Garrett, Reference Garrett2014), namely prosecutors and MNEs. The overarching aim is to stimulate a philosophical discussion about DPAs particularly as they apply to MNEs thought to be “too big to jail” (Garrett, Reference Garrett2014).
The reason for our focus on foreign corruption by MNEs is to better understand the persistence of corrupt practices within a business world increasingly dominated by MNEs. Moreover, despite the promotion of MNEs’ ethical behavior through regulation and enforcement, evidence from the literature across the globe alerts to the key role MNEs play in transnational corruption (see, e.g., Zhu, Reference Zhu2017, for China; and Luiz & Stewart, Reference Luiz and Stewart2014, for Africa). Spencer and Gomez (Reference Spencer and Gomez2011) also found that in their East European sample of MNEs, even those from home countries participating in the OECD Convention for Combating Bribery, were susceptible to corruption. By and large, the business literature suggests that corporate corruption is pervasive and not solely associated with exploiting institutional voids (Bahoo, Alon, & Paltrinieri, Reference Bahoo, Alon and Paltrinieri2020).
After outlining our approach, the following two sections of the article constitute the fulcrum of our ethical evaluation. Deontologically, we find it ironic that legal agreements developed to enhance the juridic toolkit in the fight against corporate corruption (and other financial crimes), embrace an immoral, pragmatist, relativistic philosophy, which makes the MNEs perpetrating these crimes evade proportionate punishment, in large part. Guided by normative ethical theory and, in particular, Kantian moral philosophy, we argue that DPAs are antithetical to, at least, a Western notion of justice as retribution—that is, a fair judgment according to which the wrongdoer receives punishment commensurate with the severity of their crime.
Of course, the ethical legitimacy of DPAs could still be argued on utilitarian grounds, by focusing on outcomes. To this end, following a consequentialist Machiavellian logic, we go on to investigate whether—guided by Bentham’s utilitarian criminal law principles of deterrence and reformative justice—the ends of DPAs may justify the means, also through an analysis of data on corporate recidivism. We find no evidence in support of the view that DPAs deter corporate misconduct, prevent recidivism, and foster ethical business behavior.
By initiating a discussion about the ethics of DPAs, our contribution responds to several calls from the business ethics community. The study broadens “the intellectual base” of thinking about DPAs from an integrated, “interdisciplinary” perspective as highlighted by Greenwood and Freeman (Reference Greenwood and Freeman2017: 1), with its primary focus centering on the intricate interplay between business ethics and law. Islam and Greenwood (Reference Islam and Greenwood2021) argue that the way forward in the context of business ethics should entail an integrated analysis between philosophical thinking and the social world. Based on such a recommended approach, we analyze the impact DPAs have on business ethics, integrating theoretical and practical perspectives.
ETHICAL ISSUES UNDERLYING THE NORMATIVITY OF DPAs
Characteristics, Legitimacy, and Effectiveness of DPAs
DPAs fall into the broad category of settlement agreements between prosecutors and defendants, also known as “alternative dispute resolutions” (Reilly, Reference Reilly2015: 307). DPAs share affinities with non-prosecution agreements (NPAs) and plea bargains (Alexander & Cohen, Reference Alexander and Cohen2015; Reilly, Reference Reilly2015), all resulting from negotiations between parties to solve criminal cases in place of trials (Alexander & Cohen, Reference Alexander and Cohen2015). Opting for a trial generally results in more severe sanctions and can lead to significant negative consequences for the corporation. The use of DPAs has grown significantly since the early 2000s, also due to the Enron scandal and the collapse of Arthur Andersen.
Although Markoff (Reference Markoff2013: 797) argues that “there is no evidence to support the existence of the ‘Andersen effect’ and the much-hyped corporate death penalty”, the specter of corporate death through conviction remains and its significant social and economic consequences are potentially severe (for further reading on the Andersen effect, see, e.g., Markoff, Reference Markoff2013, Garrett, Reference Garrett2014, and Koehler, Reference Koehler2015). But it is equally clear that DPAs, as currently administered, offer a very lenient way out for MNEs charged with foreign corruption. This is the reason why corporate power and MNEs’ lobbying campaigns to change the sentencing rules for corporate crime in favor of DPAs, are becoming increasingly evident in other jurisdictions. This is occurring in spite of any potential reputational damage for the MNE stemming from an air of suspicion surrounding both the integrity of their business operations, and their desire for more lenient sentencing via a DPA. SNC-Lavalin, one of the largest engineering, procurement, and construction Canadian MNEs, is a case in point. In 2015, the MNE was charged by federal prosecutors with foreign bribery of Libyan government officials and defrauding Libyan organizations of over C$100 million (Cochrane, Reference Cochrane2019). If convicted, the MNE would be barred for 10 years from federal government contracts, effectively putting the very existence of SNC-Lavalin at risk. So, SNC-Lavalin started a multi-year lobbying campaign to lean on the Canadian Government to change the Criminal Code in order to introduce DPAs (Cochrane, Reference Cochrane2019; Acorn, Reference Acorn2021). The MNE’s efforts partly paid off, with changes made to the Canadian Criminal Code to allow for a regime under which Canada’s Attorney General can approve a DPA, though in 2019 Canada’s Director of Public Prosecutions refused to pursue a DPA in the SNC-Lavalin case. In 2023, SNC-Lavalin changed its name to AtkinsRéalis.
Although plea bargaining and NPAs belong to the same genus of DPAs, they feature significant legal differences, raising further ethical problems. Taking the US experience as a reference, plea bargaining is an agreement between a prosecutor and an individual or entity approved and overseen by the judiciary (Garrett, Reference Garrett2017). Unlike plea bargaining, a DPA does not require the defendant to plead guilty, though such agreements do expect MNEs to admit to the facts of wrongdoing in exchange for more lenient treatment. DPAs and NPAs are both pretrial diversion agreements (PDAs) that do not entail a firm’s conviction (Arlen & Kahan, Reference Arlen and Kahan2017: 332–33), but while under NPAs there is no charge against the company, under DPAs there is a charge though the prosecutor commits to refrain from pursuing a conviction (Arlen & Kahan, Reference Arlen and Kahan2017). Unlike DPAs, NPAs are not subject to the review of a court (Reilly, Reference Reilly2015; Garrett, Reference Garrett2017). These agreements often include the need for corporate compliance reform and the appointment of corporate monitors (Hess & Ford, Reference Hess and Ford2008; Weismann & Newman, Reference Weissman and Newman2007), which are more prevalent in the case of NPAs and DPAs than plea agreements (Alexander & Cohen, Reference Alexander and Cohen2015).
DPAs have advantages and disadvantages (Parker & Dodge, Reference Parker and Dodge2023). They are expected to prevent collateral consequences linked to convictions (the Andersen effect), hold corporations to account without resource-intensive trials when evidence is insufficient (Parker & Dodge, Reference Parker and Dodge2023), and improve the compliance programs to prevent future misconduct (Nasar, Reference Nasar2017). They are also conducive to MNEs’ cooperation with prosecutors thereby aiding law enforcement (Nasar, Reference Nasar2017). On the other hand, DPAs are thought to be insufficiently specific to be effective. They allow prosecutorial excess without adequate judicial oversight, and do not hold corporations and individuals accountable (Nasar, Reference Nasar2017).
DPAs aim to respond to corporate criminal liability problems but also cause a weakening of them (Campbell, Reference Campbell2019). Some authors contend that DPAs adequately deal with corporate wrongdoing (e.g., Parker & Dodge, Reference Parker and Dodge2023) while others state they are an ineffective deterrent to corporate crime and impede the societal condemnation typically associated with criminal prosecutions (Uhlmann, Reference Uhlmann2013). It has also been argued that DPAs are intrinsically unfair (Reilly, Reference Reilly2015), erode corporate criminal liability, undermine the rule of law (Uhlmann, Reference Uhlmann2013) and have exhibited limited effectiveness in dealing with corporations that persistently reoffend (Ryder, Reference Ryder2018), with calls for abolishing DPAs based on the erosion of the public trust (Bourjaily, Reference Bourjaily2015).
Stakeholders’ Roles, Powers, Rights, and Abuses
DPAs arise from a dynamic interplay involving multiple stakeholders. They primarily rely on the active involvement of defendants (alongside their lawyers) and prosecutors. DPAs are predominantly, if not exclusively, granted to corporations (Amulic, Reference Amulic2017). In the US, there has been criticism that prosecutors do not adequately consider the collateral consequences of convictions for the individuals involved, as opposed to corporations, when deciding on the eligibility for DPAs (Amulic, Reference Amulic2017).
Significant issues, therefore, arise about the accountability and rights of both the corporations and individuals involved, not only in the US but also in other legal systems such as England and Wales (Garrett, Reference Garrett and Arlen2018; Ryder, Reference Ryder2018). DPAs have traditionally generated limited individual accountability of high-level executives (Copeland, Reference Copeland2016; Garrett, Reference Garrett and Arlen2018). Several authors have highlighted the need to address this problem (Copeland, Reference Copeland2016; Garrett; Reference Garrett and Arlen2018; Werle, Reference Werle2019) and recent US DOJ’s guidelines direct greater attention toward individual accountability when investigating corporate misconduct (as we discuss later in the article).
The controversial issue of corporate accountability is closely intertwined with the legitimacy and effectiveness of DPAs in deterring especially business organizations considered to be “too big to jail” (Garrett, Reference Garrett2014). In underscoring the lenient treatment of corporations through DPAs, Reilly (Reference Reilly2018: 1113) goes as far as labeling DPAs as “sweetheart deals” that undermine the integrity of the US criminal justice system. To substantiate this argument, Reilly (Reference Reilly2018: 1113) emphasizes the lack of meaningful judicial review that is granted by other, alternative dispute resolution agreements (such as plea bargains) and cites several cases in which federal district courts expressed misgivings about having to ratify (in compliance with appellate judicial decree) DPAs that “they would otherwise have likely rejected for being overly lenient.”
In terms of the protection of the fundamental rights of firms, the main legal-philosophical issue is the potential violation of the presumption of innocence (Husak, Reference Husak2014). DPAs lead to the imposition of a financial penalty and an order of compliance “without the corporate defendant having gone through a formal trial proceeding at which the criminal guilt of the defendant was established” (Shiner & Ho, Reference Shiner and Ho2018: 709). Nevertheless, there is no conclusive condemnation of DPAs on this basis (Husak, Reference Husak2014).
The prosecution authorities represent the counterpart of the firms that often wield significant power, as is evident in the US, where the issue of abuse of discretion by federal prosecutors is augmented by limited judicial scrutiny (Lawlor, Reference Lawlor2019). Federal prosecutors’ discretionary use of DPAs, with minimal judicial oversight, raises concerns as it may undermine the public interest and weaken the separation of powers (Reilly, Reference Reilly2017). Because of this, it has been suggested to limit prosecutors’ discretion by strengthening regulatory and control mechanisms (Spivack & Raman, Reference Spivack and Raman2008; Reilly, Reference Reilly2018).
The role of the courts poses significant challenges, particularly in the US, where judicial scrutiny is notably constrained (Davis, Reference Davis2022), with adverse effects on stakeholder interests (Greenblum, Reference Greenblum2005). For this reason, it has been suggested that the concept of public interest should serve as a guiding principle for adopting and approving DPAs (Garrett, Reference Garrett2017). Unlike the US, in countries such as the UK and Singapore, judicial oversight of DPAs plays a significant role, requiring that such agreements are reached “in the interests of justice and that their terms are fair, reasonable, and proportionate” (Chua & Chan, Reference Chua and Chan2020). The critical analysis of DPAs in England and Wales has led to three interdiscursive mechanisms that serve to justify the approval of DPAs in the area of transnational corporate bribery (Lord, Reference Lord2022). The first considers the company’s conduct as deviant yet acceptable in order not to proceed with prosecution (“deviance elastication,” Lord, Reference Lord2022: 856). The second allows the corporation to distinguish itself from the entity responsible for the offense (“corporate dissociation,” Lord, Reference Lord2022: 860). The third argument considers the collateral effects of the conviction, particularly debarment (“anticipatory offsetting,” Lord, Reference Lord2022: 862).
Cooperation, Compliance, and Monitorship of DPAs
The use of DPAs is marked by cooperation between corporations and public prosecutors across all stages of the DPA process. The reasons for corporations’ cooperation with the government (starting with self-disclosure) are manifold, but the promise of less burdensome consequences is pronounced (Bohrer & Trencher, Reference Bohrer and Trencher2007).
However, limitations around cooperation have been highlighted. For example, corporations, even if innocent, to avoid negative consequences, may be under pressure to reach a DPA (Meeks, Reference Meeks2006). Especially in the immediate aftermath of the Enron scandal, the governmental cooperation of firms in the investigation and prosecution of individuals raised concerns about the right to non-self-incrimination (Griffin, Reference Griffin2007), with such cooperation turning “corporations into agents of the state” (Bohrer & Trencher, Reference Bohrer and Trencher2007: 1481).
In the US, DPAs increasingly include provisions for reforming compliance programs, concerning business and board changes as well as monitoring processes to oversee the implementation of agreed reforms (Kaal & Lacine, Reference Kaal and Lacine2014). These measures effectively result in prosecutors’ influence on corporate governance (Kaal & Lacine, Reference Kaal and Lacine2014; Robinson, Urofsky, & Pantel, Reference Robinson, Urofsky and Pantel2005) with US prosecutors being criticized for imposing corporate reforms on the basis of their broad discretion, which creates issues of (in)consistency with the rule of law (Garrett, Reference Garrett2007; Arlen, Reference Arlen2016).
DPAs and (International) Corruption
The US has developed an enforcement practice against domestic and international corruption through the Foreign Corrupt Practices Act 1997 (FCPA) by significantly leveraging on DPAs and NPAs (Koehler, Reference Koehler2015). Koehler (Reference Koehler2015) points out that between 2004 and 2014, roughly 85 percent of DOJ’s FCPA criminal enforcement actions were resolved through NPAs and DPAs.
Some authors have suggested that failure to deter bribery at the international level stems precisely from the growing adoption of DPAs, which has turned wrongdoing into a “cost of doing business” (Weisman, Buscaglia, & Peterson, Reference Weismann, Buscaglia and Peterson2014: 591–92). One further problem with DPAs relates to the fact that some nations (England, Wales, France, and Switzerland) seldom receive compensation for being victims of international bribery (Capus & Brodersen, Reference Capus and Brodersen2022).
APPROACH
Following the pathway suggested by Dunfee (Reference Dunfee1996), we conduct an ethical analysis of legal issues related to the use of DPAs and discuss implications for business ethics. These disciplines are equally important in the business world (McCarty, Reference McCarty1988) and afford an integrated, holistic view of the phenomenon under scrutiny (Dunfee, Reference Dunfee1996). Based on such an approach, we carry out an evaluation of DPAs from the perspective of Kantian deontological theories and the utilitarian theories of Jeremy Bentham. While the deontological evaluation is purely theoretical in nature as it is based on values, the utilitarian appraisal also benefits from an analysis of information and data on DPAs and their consequences, thus providing an evidence-based contribution to understanding whether DPAs effectively deter corporate crime and prevent recidivism.
The analytical review focuses on available studies and data in the US, where evidence on corporate recidivism following DPAs makes it possible to conduct such a study. It must be acknowledged that data limitations prevent a systematic statistical analysis of all possible cases of MNE recidivism, as this would require not only a precise and uniform definition of the concept of company recidivism—which can be complex and subject to varied interpretations—but also a complete, detailed set of statistical data on a global scale that, to date, remains the exclusive preserve of public authorities across many national jurisdictions. We therefore concentrate on a qualitative analysis of identifiable cases of MNEs’ recidivism focusing on multiple disposition types against the same company in the US.
We start by drawing on data about DPAs published in the Corporate Prosecution Registry (CPR), including federal organizational pleas and prosecution agreements (Garrett & Ashley, Reference Garrett and Ashley2023). On July 25, 2023, using “DP” (i.e., DPA) as a “disposition type” for the default period of 1992–2023, the CPR returned 301 companies that had accessed DPAs (a staggering number that, in itself, alerts to the weakness of DPAs’ leniency as a deterrent for corporate misconduct). For each of these corporations, we checked whether there was any other disposition type (acquittal, declination, dismissal, NPA, plea, trial conviction) in the Registry. This analysis was carried out on a textual basis and aimed at understanding whether the same companies—that is, not various companies in the same group, for which there are already public studies (Public Citizen, 2019)—had been subjected to other disposition types. This initial textual search revealed 28 companies that accessed at least one DPA. Next, we performed a mix of textual analysis and further research, which revealed five further companies (for a total of 33) that accessed at least one DPA.
It should be noted that a text search relying on the exact company name, may not produce precise results. For example, a textual search for “Société Générale SA” showed a DPA and a plea on the same year. By analyzing the documents, these disposition types concerned different companies in the same group, as the DPA was reached by “Société Générale SA” while the plea referred to “SGA Société Générale Acceptance, N.V.”
The identification of such cases could be influenced by various factors, such as a company name change or limitations in the textual information of the database, which would make it difficult to correctly identify a specific business organization. Corporate changes—even rebrands—could have the effect of preventing the same company from being identified. Such limitations in data collection made it unadvisable to conduct statistical evaluations, favoring instead a qualitative analysis of the identified cases. The 33 companies which make up our list are reported in Table 1.
Examples of Multiple Dispositions by Same Company Including at Least One DPA
Note. Data from Corporate Prosecution Registry (latest access: July 27, 2023).
We also undertook an additional analysis of recidivism related exclusively to enforcement actions in connection with the FCPA matter in the US, which involves both bribery and accounting violations. In searching for reliable primary data sources for this additional analysis, we followed the excellent study presented by Koehler (Reference Koehler2019) on repeat offenders within the context of the FCPA. Koehler’s (Reference Koehler2019) table (see table 1 in Koehler Reference Koehler2019: 1305–1308) contained enforcement actions related to business organizations that resolved more than one FCPA enforcement action, regardless of whether it was brought by the US DOJ or the US Securities and Exchange Commission (SEC), and regardless of the form it took (e.g., plea agreement, NPA, administrative order, etc.). Over a sample period from 1989 through to 2018, Koehler identified 13 cases of FCPA recidivism.
Following Koehler’s analytical blueprint, we examined data publicly available on FCPA from both the US DOJ database (www.justice.gov/criminal-fraud/corporate-enforcement-actions, last update: January 31, 2023) and the US SEC database (https://www.sec.gov/enforce/sec-enforcement-actions-fcpa-cases, last update: June 23, 2023). We then reviewed the two lists of companies that were subjected to the enforcement actions by DOJ and SEC over the period 2000–2022. Based on data available on July 25, 2023, we identified 15 cases of MNEs’ FCPA recidivism (see Table 2).
Considering the selected time frame, we excluded General Electric and Goodyear analyzed by Koehler (Reference Koehler2019), as their first enforcement actions took place in 1989 and 1992, respectively, years outside our reference period. Additionally, the textual analysis of enforcements against corporations with the same name did not allow us to identify Aibel Group/Vetco explicitly as indicated in Koehler’s analysis, even though a DOJ press release indicated that Aibel Group entered into a plea after it was not in compliance with a DPA. Finally, our analysis of enforcements against the same company did not offer the opportunity to verify the case of Alcatel-Lucent SA and Alcatel-Lucent France SA. As a precaution, therefore, we excluded this corporation. However, Koehler (Reference Koehler2019) listed it; we can therefore take our number of identified “repeat offenders” as a more conservative estimate.
DEONTOLOGICAL EVALUATION
In this section, we address the question of whether DPAs can be deemed ethical from a deontological perspective in light of Kantian theories of punishment and justice (Kant Reference Kant, Denis and Gregor2017, orig. 1797; Kant Reference Kant and Gregor1996, orig. 1793), as revisited by further scholars and re-adapted to the context of modern corporate criminal justice.
Kant is one of the major proponents of deontology, retributive theory, and the principle of equality. As argued by Alexander and Moore (Reference Alexander, Moore and Zalta2021, para. 2.4) in the entry on “Deontological ethics” of The Stanford Encyclopedia of Philosophy, “[i]f any philosopher is regarded as central to deontological moral theories, it is surely Immanuel Kant.” In the same Encyclopedia, Walen (Reference Walen, Zalta and Nodelman2023, para. 3.6) argues that the notion of punishment developed in the lex talionis (law of retaliation) present in Babylonian and early Roman law was also “endorse[d]” by Kant, albeit in the form of “invoking the principle of equality for punishment.” Hence, in the absence of previous studies on the ethics of DPAs, Kant’s theories appear to us to be fertile ground for cultivating a deontological evaluation.
Of course, we do acknowledge that Kantian theories have themselves been subjected to critical analysis and diverse interpretations.Footnote 1 But, surely, this would not be a reason for anyone, especially in the field of ethics and philosophy, to be prevented from taking Kant’s theories as a benchmark for a deontological appraisal. Moreover, such a choice does not preclude other ethical frameworks to be adopted in future studies once an essential Kantian perspective has been cultivated. Another potential caveat to dispel about our endeavor to employ Kant’s moral philosophy to subject DPAs to deontological scrutiny is that Kant connected the theory of retribution to human sufferance and the theory of justice to individuals. Yet, we believe that intersecting such a deontological vision and ideas of both punishment and justice with criminal offenses by MNEs offers an invaluable holistic lens through which to assess the morality of DPAs for society at large in the context of modern corporate criminal punishment and justice.
Based on these premises, we explore whether DPAs run counter to the notions of fair retribution (O’Connell, Reference O’Connell2014) and fair justice (Rawls, Reference Rawls1971, Reference Rawls1980; Pogge, Reference Pogge1981), which reflect the central tenets of substantive and procedural criminal justice. The former refers to the fairness of the rules of a legal system while the latter to the fairness of how the system of law is applied. Although we make a distinction between substantive and procedural justice, they are nevertheless closely connected, allowing us to engage in both an analytic and a holistic process of reflection on the ethics of DPAs as far as MNEs are concerned.
DPAs and Retributive Justice
We begin by examining whether DPAs align with the retributivist Kantian theory of punishment. Kant argues that the “law of punishment is a categorical imperative” (Kant, Reference Kant, Denis and Gregor2017, orig. 1797, 6: 331), with the result that the lack of punishment is morally “wrong” (O’Connell, Reference O’Connell2014: 480). As argued by O’Connell (Reference O’Connell2014: 477), from the perspective of Kant’s retributivism, “punishment is warranted as a means to promote proportionality between well-being and virtue”; in other words, “the amount of suffering inflicted on wrongdoers should be proportional to the wrongdoing for which they are being punished” (O’Connell, Reference O’Connell2014: 479). Thus, in the realm of deontology, wrongdoers should face retributive punishment even if it does not produce any other positive outcome (such as re-educating or preventing recidivism) to hold them accountable for their actions. According to Buell (Reference Buell2020: 30), retributivism aims to “fulfill a moral imperative—that the wrongdoer must be punished. Whether such punishment produces benefits or harms is irrelevant to a ‘pure’ or ‘full-throated’ retributive theory.” These theories would suggest that DPAs are morally wrong as they challenge the categorical imperative of proportionate punishment.
Although a reflection on plea bargaining, in particular referring to individuals, has already been made from the perspective of retributive justice (Slobogin, Reference Slobogin2016), an analysis of DPAs reached by MNEs under this lens is lacking. Such a reflection may be lacking for two reasons, which we address in turn. First, Kant developed his theory of punishment referring to the responsibility of individuals. However, we would argue that this qualification alone should not preclude the necessity of exploring how DPAs in criminal business justice can be interpreted in the light of Kantian retributive theory.
Second, and related to the first point, discussing retributive justice (or “just desert,” i.e., the punishment that one deserves) in relation to corporate liability can be seen as problematic as the connection between the two constructs is enshrined in the dogma societas delinquere non potest (meaning legal persons/entities—i.e., corporations—cannot commit a crime). As Buell (Reference Buell2020: 45) argues, since corporations cannot suffer deprivations of personal liberty like natural persons, they could not “be punished on retributive grounds.”
Nevertheless, we are of the view that since MNEs are capable of acting under civil law, there is no reason why they would not be capable of acting under criminal law, be culpable and accountable. Moreover, as noted by Rich (Reference Rich2016: 98), even if corporations “cannot suffer” like individuals, they “have the capacity to understand moral matters and make moral judgments. It follows from this that retributivism must be included as a component justification of corporate sentencing” (Rich, Reference Rich2016: 98). Indeed, despite the legalistic approach to punishment in its distinction between legal and natural persons, sanctions against an undertaking are collective sanctions directed in a certain way against all subjects of an undertaking. As Hasnas (Reference Hasnas2010: 77) points out “Corporate punishment necessarily falls indiscriminately on the innocent as well as or in place of the guilty.” It is certainly true that, as noted more than two centuries ago by Edward, First Baron Thurlow (1731–1806), corporations have “no soul to be damned, and no body to be kicked” (Coffee, Reference Coffee1981: 386). However, a punishment that results in the dissolution of a company results in collective responsibility. For example, in the Arthur Andersen case, many employees of the convicted firm were dismissed, and the firm died (Hasnas, Reference Hasnas2010: 78). It is no coincidence that Andersen’s conviction, with its collateral damage, was the reason behind the growth of DPAs. Fisse (Reference Fisse1983: 1182–83), in considering retribution and deterrence as “concurrent goals,” frames the issue well by stating that “justice as fairness” is the foundation of corporate crime punishment and, as such, it should be a means “to restore and reaffirm the community’s core values” (Robson, Reference Robson2010: 144). In conclusion, “retributivism rests on the idea of imposing punishment in response to past wrongdoing” (Rich, Reference Rich2016: 116) even if the “true punishment” is seen by some as “impossible in the corporate case” (Rich, Reference Rich2016: 116).
Accepting that retributivism is a theory that can also be applied to corporations and must be understood, from a Kantian perspective, as a moral imperative to punish corporations in line with the principle of proportionality,Footnote 2 two conflicting views can be noted. The first, as purported by Shiner and Ho (Reference Shiner and Ho2018: 707), is that the conditions that a corporation agrees to in settling a DPA constitute “obligations” that are, in and by themselves, “a reasonable retributive response to a breach by that corporation of the community’s laws.” The second view, which is the one we espouse, is that—assuming the corporation is indeed guilty of the crimeFootnote 3—DPAs fall short of the demands of retributive justice and are intrinsically unfair and immoral. In developing arguments in support of the latter view, we draw on the ideas of the “moral duty to punish” and “quantum” (or proportionality) of punishment.
The first argument in favor of the immorality of DPAs relies on the “non-conviction” of MNEs perpetrating international corruption; a concept closely intertwined with punishment. Essentially, DPAs allow MNEs culpable of foreign corruption not to formally plead guilty or be convicted. In a holistic view of criminal law and procedure interpreted in light of retributive justice, “conviction does matter” irrespective of goals. Buell, who advocates retiring corporate retributivism, asked “why conviction for corporations should matter when the criminal process cannot impose sanctions on corporations that are different in kind, or even theoretically in degree, than civil lawsuits and regulatory enforcement actions” (Buell, Reference Buell2020: 44). In response to Buell, we would argue that the first reason why conviction should matter for corporations, is that convictions involve delicensing and debarment from public contracts, which are not contemplated by DPAs. The second reason is that convictions, as acknowledged by Buell (Reference Buell2020: 45) himself, “express legal consecration of the corporation’s wrongdoing and blameworthiness to a degree that DPAs and NPAs cannot.”
We are of the view that to consecrate, deontologically, and give teeth to the legally binding nature of the various international conventions against bribery (see, e.g., the OECD Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, OECD, n.d.), for very serious corporate crimes such as corruption and bribery of foreign public officials, it is the delicensing and debarment that would be stipulated through a conviction sentence in a court of law that sets the benchmark for the “proportionate punishment” that would satisfy the categorical imperative. See, for example, the SNC-Lavalin case discussed earlier, according to which, if convicted, the corporation “would be slapped with a 10-year ban on receiving federal government contracts” (Cochrane, Reference Cochrane2019). A punishment which would be, by law, and deontologically, justifiable. As to the quomodo of justice and the issue of blameworthiness, as recognized by Buell (Reference Buell2020: 45) himself, unlike prosecutions, DPAs also “do not involve an independent judicial officer entering a judgment of conviction on a court’s docket and performing the ritual of imposing a ‘sentence’.” So, the question begs, why should a DPA be allowed to express “blameworthiness” less than a formal conviction sentence, especially when considering that the ritual of imposing a sentence is, in itself, seen by many members of society as a critical element of retributive justice? In the words of Uhlmann (Reference Uhlmann2013: 1302), DPAs “limit the punitive and deterrent value of the government’s law enforcement efforts and extinguish the societal condemnation that should accompany criminal prosecution.”
The second argument supporting the idea that DPAs do not satisfy the criteria of retributive justice relates specifically to the fact that DPAs do not ensure that punishment is proportionate to the wrongdoing (“quantum” of punishment). In a world characterized by massive incarceration of street crime, it seems questionable at best that it is MNEs and white-collar criminals who end up benefiting from privileged avenues of case resolution. By adopting Brocardo’s biblical-derived precept malum passionis propter malum actionis, proportionality would require corporations to be at least liable for their deed by reason of the seriousness of the offense. However, in light of prevailing opinion and empirical findings, DPAs are best characterized as a very lenient response to severe criminal offenses committed by MNEs and their executives.
DPAs appear to be a shield for the criminal liability of individuals at the helm of offending MNEs, as the British and American experiences show (Garrett, Reference Garrett2015, Reference Garrett and Arlen2018, Reference Garrett2020; Ryder, Reference Ryder2018; Werle, Reference Werle2019; Hawley, King, & Lord, Reference Hawley, King, Lord, Søreide and Makinwa2020). Traditionally, the law attributed criminal liability to a corporation through the actions of its directors; a system that—somewhat perversely—encourages board members to distance themselves from the criminal business’s operations (and thus from direct knowledge of individual wrongdoing in corporate misconduct). In many countries, DPAs are available to corporates, not individuals, even for cases of economic/financial offenses such as bribery and fraud. Subject to the MNE adhering to the conditions of the DPA, the offenses (at both corporate and individual officer level) are not prosecuted. As eloquently put by Garrett with reference to the DPA experience in the US, “[t]he corporation appears to be a kind of a scapegoat: perhaps not entirely blameless, as in the traditional concept, but literally impossible to actually jail—yet capable of receiving the brunt of blame … while the individual culprits go free” (Garrett, Reference Garrett2015: 1789). Similar considerations have been developed concerning the English system (Hawley, King, & Lord, Reference Hawley, King, Lord, Søreide and Makinwa2020). According to retributive theory, an MNE and the individuals responsible for triggering corporate liability should be punished because they deserve it based on the principle of “just desert.” However, DPAs, as they work in practice, are associated with highly inadequate punishment and thus are morally indefensible.Footnote 4
DPAs and Equal Justice
The conclusion that DPAs are immoral also finds support when taking a procedural perspective that employs Kant’s theory of justice as an ethical lens. As underscored by Rosen (Reference Rosen1993: 14), Kant’s universal principle of justice relies on a crucial proposition: “equality of each with all others as a subject.” In Kant’s perspective, “legal equality” is strictly connected to the “principle of civil freedom” (Rosen, Reference Rosen1993: 26) and implies treating all equally by excluding privileges. Despite the inevitable limitations in crudely transposing such a theory into the realm of corporate criminal justice, we nevertheless consider that legal equality thus conceptualized is an essential lens through which, from a deontological perspective, DPAs can be ethically appraised. Reconnecting to the notion that corporations and natural persons are different subjects, the main question, following the approach suggested by Shiner and Ho (Reference Shiner and Ho2018: 712), then becomes whether they are treated “in a manner sufficiently analogous.”
Taking this proposition as our conceptual arena, a case can be made for three levels of potential legal (and ethical) inequality resulting from DPA-based corporate enforcement strategies. The first level of inequality concerns the limitations for individuals versus corporations in accessing these pre-trial diversion schemes. A second level of inequality stems from corporations’ prerogative to defer prosecution, with the trial being replaced by an agreement between parties, and the judicial oversight that should ensure equal independent treatment often being inadequate. The third level of inequality relates to the absence of a formal conviction at the end of the proceedings. This last argument, in our view, highlights an injustice that spans across the closely connected substantial and procedural issues of fairness.
Regarding equality in terms of access to justice, in the US experience, historically the use of DPAs concerning individuals appears to be limited to crimes such as juvenile offenses (Greenblum, Reference Greenblum2005). Progressively, DPAs have been extended to corporations and only rarely to individuals, and these instruments for the resolution of criminal cases appear to entail favorable treatment of corporations and, in particular, MNEs (Garrett, Reference Garrett2014). In other legal systems where DPAs have recently been imported, such as France, England, and Wales, current legislation does not even allow individuals to reach these agreements. The preclusion of individuals (different from corporate officers) in accessing these settlement agreements seems to be one of the key features of inequality that underline the biased nature of these instruments; a bias that sways the ethical assessment of DPAs for MNEs’ crimes further toward a verdict of immorality. To this conclusion, one could still object that natural and legal persons are different entities. Hence, there should be a moral difference between them that justifies a delimitation of the golden rule of Kantian thought, namely the principle of equality. Nonetheless, we see no plausible reason to argue that justice (of either the substantive or procedural kind) should be more favorable in granting access to DPAs to large corporations than individuals, as it is the case in most jurisdictions such as England, Wales, and France.
Critically, DPAs imply not only deferring but exempting prosecution (Lord, Reference Lord2022) through the replacement of the trial in public with a settlement agreed upon by the parties. As is the case for plea bargaining, DPAs ensure rewards, while the trial, which constitutes a right, entails more significant penalties (Lippke, Reference Lippke2011). This paradox aside, negotiating punishment before conviction (Bronitt, Reference Bronitt, Colvin and Stenning2018) raises additional ethical concerns, including the presumption of innocence of corporations, itself the subject of a heavily contested debate in the legal literature (Husak, Reference Husak2014; Shiner & Ho, Reference Shiner and Ho2018). DPAs thus appear to violate the “due process” principle underlying the deontological theory of justice. No end can justify the means from a deontological perspective, and DPAs, by their very nature, stand in stark conflict with both the need for a fair trial—a cornerstone of the jurisprudence of international law—and the need to ensure equal treatment before justice in its Kantian dimension. Finally, DPAs lack transparency, a core ethical principle in business ethics. They are not always subject to effective judicial scrutiny and are developed through negotiations between the parties that flow into the statement of facts and agreement. Negotiation can affect the content of such agreements (which may also include non-disclosure stipulations) without judicial scrutiny being able to identify or verify any underlying criminal issues.
The fact that DPAs are often subjected to limited judicial scrutiny in the US experience has significant implications regarding (in)equality. There are two opposing views on the involvement of the judiciary in the procedure for enforcing DPAs. Some scholars argue that the role of the judge is “minimal” within DPAs and there is a case for expanding this role to defend corporations (Greenblum, Reference Greenblum2005: 1869). Other scholars contend that the role of the judiciary in DPAs should be expanded to protect the public interest in the name of justice (Garrett, Reference Garrett2017). The judge, expected to be an impartial subject, should effectively review what has been negotiated between the parties in the interest of society and ensure equal and consistent treatment of everyone. Yet, despite different premises and logics of argumentation, both camps seem to converge on the conclusion that it is essential to improve the review process of DPAs (Daniels, Reference Daniels2017). As commented by Reilly (Reference Reilly2015: 307), as they stand, DPAs make “a mockery of the criminal justice system” as they generate “unfairness, double standards, and potential abuse of power.”
With respect to the third level of inequality, DPAs do not seem to adhere to the values embedded in the legal mantra “equal justice for all” as they offer favorable treatment to big corporations by helping them avoid prosecution and sentencing rather than letting them face the full consequences of their illegal conduct. Garrett (Reference Garrett2014: 6) argues that the logic of “too big to jail” prevails in US corporate justice, and DPAs are the most relevant expression of such a logic. Considering the lack of prosecutions of many Wall Street banks after the financial crisis of 2007–2008, Garrett (Reference Garrett2014: 1–2) stated that some corporations “are considered to be so valuable to the economy that prosecutors may not hold them accountable for their crimes,” and usually, individuals did not go jailed (Garrett, Reference Garrett2014: 13–14). The conclusion of this study highlights how individuals (different from white collars) and large corporations are treated differently. Garrett (Reference Garrett2014: 14) asks: “How is a corporation punished? Not by relying on strict and narrow sentencing guidelines, as with individuals, but by using more flexible guidelines that may give the biggest fish the best deals.” Although decades have passed since Sutherland (Reference Sutherland1940) called attention to the significance of white-collar crime, it could be argued that DPAs represent the modern manifestation of (in)justice for crimes of the powerful and the blatant unfairness of contemporary corporate criminal justice.
Summary
Using the lens of Kantian deontological theory, and employing ideas based on both the moral duty to punish and the “quantum” of punishment, DPAs would appear to have failed ethical scrutiny. They do not lead to a formal criminal conviction, with a punishment entailing the mere application of pecuniary sanctions and compliance reforms but, crucially, without involving delicensing and debarment. Regarding quantum, DPAs do not entail a proportionate response, as they significantly cause underenforcement against MNEs. With respect to the quomodo of justice, even though such agreements typically expect MNEs to admit the facts of wrongdoing, they do not require an explicit admission of guilt in court.
To sum up, the ethical evaluation carried out so far leads us to state that such settlements pose significant ethical concerns. They allow large MNEs to enjoy privileged treatment in relation to serious criminal offenses such as foreign corruption and, in most cases and jurisdictions, even exempt liability of the natural persons involved in such offenses. They also raise serious ethical concerns in terms of procedural justice by excluding formal conviction and avoiding blameworthiness.
Of course, it must be acknowledged that our deontological evaluation of DPAs presented above, rests largely on the Kantian premise of justice interpreted as “retributive justice,” one fundamentally based on the principle of “just punishment.” We recognize that there are other philosophical traditions that would reject an ethical review premised on this basis. For example, pragmatic ethics—as advanced by Charles S. Pierce (1839–1914)—would suggest the impossibility of determining a priori a universal ethical principle on the basis of which to undertake deontological ethical assessment. By rejecting any form of absolutism, pragmatism is not based on principles, yet it is not unprincipled. It fosters a form of relativism that establishes the crucible of human experience and practical interests as the moral basis for judgment and evaluation. As such, it holds for ethical principles being social constructs, including resulting actions (“action” in practice, in the sense of the Greek word “pragma”) of compromise (the negotiated agreement embedded in DPAs in our case), to be evaluated in terms of their usefulness.
This ethical lens of relativistic pragmatism connects and opens the door to our further ethical evaluation of DPAs through a consequentialist, utilitarian moral perspective according to which the rightness or wrongness of an action (a settlement via DPA) depends only on the total goodness or badness of its consequences. To this end, in the next section we investigate whether “the ends” of DPAs may “justify the means.”
A CONSEQUENTIALIST UTILITARIAN EVALUATION
Utilitarianism constitutes one of the main consequentialist theories, in the sense that conduct is evaluated on the basis of the general goodness of the consequences produced (Driver, Reference Driver, Zalta and Nodelman2022). We undertake such an evaluation by delving into the influential realm of Jeremy Bentham’s utilitarianism and its profound implications within the domain of penal justice.
Despite some difficulties in reconstructing Bentham’s theories of punishment in light of the fragmented material available, Bentham’s main theses have been excellently summarized by Bedau (Reference Bedau2004) and Crimmins (Reference Crimmins, Zalta and Nodelman2023). These texts, therefore, constitute our essential reference point in the interpretation of Bentham’s philosophical thought. Crimmins (Reference Crimmins, Zalta and Nodelman2023, para. 7) emphasizes that Bentham’s utilitarian criminal law would have deterrence as its primary objective, but it would also entail “disablement, moral reformation, and compensation.” As highlighted by Bedau (Reference Bedau2004), Bentham would argue that since punishment is itself an evil, it should be minimized. The goals of punishment should be to prevent crime, correct offenders, and compensate victims (Bedau, Reference Bedau2004).
The approach of analyzing DPAs from this perspective might encounter criticism, given that Bentham’s viewpoint centered on the punishment of individuals (rather than corporations), as evident from his focus on issues such as imprisonment and the death penalty (Bedau, Reference Bedau2004). However, it is worth noting that Bentham lived in a significantly different time, where attention had not yet reached corporate crime. As indicated earlier, the extension of past philosophical theories to the present time is inevitably complex and challenging. However, there is no reason not to try to explore how DPAs can be evaluated using Bentham’s theses, especially since his philosophical ideas would require punishment to be “designed to promote the greatest happiness in society” (Sverdlik, Reference Sverdlik2019: 246), a contextual realm which nowadays bears a close relationship to the activities of MNEs.
Taking the main axioms of Bentham’s theory of punishment as a reference—that is, deterrence and rehabilitation—the question then arises as to whether, from a utilitarian point of view, DPAs successfully fulfill their crime-prevention and offender-correction functions.
Deterrence and Rehabilitation
Bentham represents one of the main philosophers who, alongside Beccaria, laid the foundation for deterrence as the basis of punishment (Paternoster, Reference Paternoster2010). Deterrence and rehabilitation are the primary justifications for corporate punishment in countries like the US, although doctrines emphasizing incapacitation are increasingly being added (Thomas, Reference Thomas2019). From a utilitarian perspective, scholars have argued that DPAs would prevent the “corporate death penalty” due to the collateral consequence of the conviction and, more significantly, would rehabilitate offending firms and prevent recidivism.
Regarding the prevention of companies’ demise due to prosecution (the Andersen effect), we have already highlighted that, based on studies conducted so far, although this effect is a consequence of a well-known previously reviewed case, other convicted corporations have indeed survived prosecution. As for preventing recidivism among companies settling criminal cases via DPAs, recent developments in enforcement in the US and available data seem to at least cast doubt on this argument.
Nonetheless, several scholars have pointed out that DPAs adopted in the US constitute forms of restorative justice aimed at preventing future crimes in the business realm while others have advocated for DPAs as suitable instruments to prevent corporate recidivism. Although the currently available data may not allow for a reliable and definitive assessment of deterrence and rehabilitation, it does appear that the notion that DPAs serve the function of preventing future corporate offenses can be called into question.
Recidivism and Beyond
Notable studies are now surfacing in the US, delving into the issue of recidivism in cases involving DPAs. One recent study by Public Citizen (2019) specifically examined the connection between DPAs (and NPAs) in the US and their effectiveness in deterring repeat offenders. This analysis holds transversal significance as it pertains to recidivism theories across various legal domains. Through this empirical investigation, Public Citizen (2019: 5) demonstrates that DPAs and NPAs “do not prevent corporations from breaking the law again.” Moreover, “most corporations that have faced multiple criminal enforcement actions, yet avoided prosecution, are large multinationals, and most of these have avoided prosecution more than once” (Public Citizen, 2019: 4). The study identified 38 corporations that received a NPA or DPA “but nevertheless did offend again, sometimes repeatedly” (Public Citizen, 2019: 6). It also noted that prosecutors appeared to punish firms to a limited extent when breaching agreements. According to the report, US prosecutors, by utilizing DPAs (and NPAs), would reach agreements prioritizing the protection of corporate profits over the prosecution of business organizations to hold them accountable for their legal violations (Public Citizen, 2019: 4). This approach also raises concerns about the protection of the victims affected by these crimes. Looking, therefore, at Bentham’s theory as summarized above, the usefulness of DPAs would have to be questioned as they do not prevent crime, correct offenders, and adequately compensate victims.
Of course, assessments of recidivism depend on multiple factors, including the specific operationalization of this construct. Yet the exact definition of this term has significant implications in the analysis of the phenomenon. It seems conceivable, therefore, to distinguish between various levels of analysis, also considering current studies on the subject. Our research, as shown in Table 1, presents all the hypotheses in which multiple disposition types were adopted and applied against the same company. As can be seen, the use of multiple disposal types in connection with different enforcement actions (parallel or consecutive) becomes relevant when the attention shifts from different companies within a multinational group to the same individual company.
The 33 cases we report in Table 1 provide a substantial body of evidence from which issues directly related to a consequentialist analysis can be raised. First, there are cases of parallel enforcement, like those concerning Aegerion Pharmaceuticals, Inc., in which the US government entered into a DPA, the terms of which seem to include compliance with a plea agreement. There are also instances of breaches of DPAs, followed by other agreements of a different nature. An example of this is Aibel Group Ltd, whose DPA reached in 2007 was dismissed after pleading guilty to the violation of the FCPA. Next, we observe cases in which DPAs come after other agreements, meaning they were used after an MNE had already reached other settlements. Finally, there are cases in which DPAs precede other disposition types connected to further enforcement actions for different offenses, highlighting the ineffectiveness of DPAs in preventing recidivism. These cases can involve either homogeneous (same matter or type of crime) or heterogeneous violations (different matters).
We also conducted a more fine-grained analysis by focusing on repeat offenders that have resolved more than one FCPA enforcement action to examine violations related to the same subject matter. In an analogous study carried out in 2019, Koehler identified 13 cases of repeat offenders over the period 1989–2018. The numerous instances of repeat offenders identified were such to lead Koehler (Reference Koehler2019) to conclude that his evidence of recidivism demonstrated, at best, the impossibility for large business organizations to guarantee FCPA compliance, and, at worst, “the ultimate failure of ‘soft’ FCPA enforcement” (Koehler, Reference Koehler2019: 1309), undermining the “FCPA enforcement agencies rhetoric” that instruments such as DPAs can significantly change the behavior of specific companies and, more generally, influence corporate culture on a global scale (Koehler, Reference Koehler2019: 1309).
Drawing on the data collected by Koehler (Reference Koehler2019) and supplementing it with repeat offender cases occurring after 2018 (see our “Approach” section), our present analysis, as shown in Table 2, corroborates his conclusions. In fact, since 2018, the number of FCPA repeat offenders has increased further, with yet new cases of recidivism being identified.
In contrast to Koehler (Reference Koehler2019), by “repeat offender,” we employ an even more stringent definition, by referring to the same-named company that has settled multiple enforcement actions related to the Foreign Corrupt Practices Act (FCPA), regardless of the enforcement agency (DOJ or SEC), the resolution method used (plea agreement, NPA, DPA, administrative order, etc.) or the specific FCPA violation. With reference to Table 2 it is important to highlight that there are cases not yet reported in the Corporate Prosecution Registry (from which we drew for our analysis reported in Table 1) as of the extraction date of the information published in Table 1. In this regard, it should be noted that ABB Ltd. reached a second DPA in 2022 in the FCPA matter, which does not feature in the Registry. Additionally, Koninklijke Philips N.V., formerly known as Koninklijke Philips Electronics N.V., was subjected to another SEC FCPA enforcement action in 2023, but we excluded it from Table 2 since the year falls outside our sample period.
It is interesting to observe that most instances of recidivism occurred after the intervention of the SEC, and paradoxically, DPAs and NPAs were both involved in such cases. Among the 15 FCPA-related repeat offenders identified, there are six instances where DPAs were used either before or after. For companies such as ABB Ltd., Biomet, and Orthofix International, further enforcement actions took place after a DPA. These data do not allow for hasty conclusions, of course, but clearly DPAs played a decisive role in many cases involving FCPA repeat offenders, either before or after another civil or criminal enforcement action was issued.
Although the avoidance of corruption looms large in the emerging demands on MNEs for responsible business practices, including legally binding international instruments aimed at controlling corporate behavior (see, e.g., OECD, 2023, the “OECD Guidelines for Multinational Enterprises on Responsible Business Conduct,” which includes Guideline VII. Combating Bribery and Other Forms of Corruption), our analysis provides ample evidence that DPAs are intertwined within the realm of real-world scenarios characterized by widespread and increasing MNEs’ (FCPA) recidivism.
Summary
Using the lens of utilitarian theory, DPAs raise significant ethical issues if they fail to effectively prevent further crimes or halt incessant corporate recidivism. DPAs seem to grant even the same companies within multinational groups preferential treatment, allowing them the opportunity to enter into another DPA or other settlement agreement after the first one, regardless of whether the crime is different or the same. This seriously undermines the notion that DPAs may serve a utilitarian objective.
CONCLUSION
This study has assessed the ethics of DPAs using both deontological and utilitarian theories. Our conclusion is that DPAs fail ethical scrutiny from both perspectives. From a deontological standpoint, DPAs do not uphold the principle of just punishment, they lack proportionality and undermine procedural justice. From a utilitarian perspective, DPAs grant preferential treatment to MNEs while failing to effectively deter corporate crime and prevent recidivism. In short, the “ends” of DPAs do not appear to “justify the means.”
Of course, this does not mean that all agreements are ineffective. The complexity of the business and legal context can influence outcomes, and recidivism could be linked to various factors, such as the specific corrective measures adopted by the MNEs involved or the dynamics of the industry and countries in which they operate. Nevertheless, our evidence raises important questions on the lenience DPAs warrant to large corporations that are caught repeatedly violating the law. The breadth of such cases is such that it appears reasonable to infer that corporate corruption has indeed become normalized, and even, possibly “priced into” MNEs’ strategy as a cost of doing business given the enforcement authorities’ weak response through DPAs. The rinse-and-repeat cycle of corporate crime, even for serious cases of international corruption, erases any confidence in the belief that the penalties threatened by DPAs can deter the temptation to offend and corroborates the suspicions of government regulators’ complacency in perpetuating the axiom that large corporations are indeed “too big to jail.”
Some readers may still argue that our identification of 33 instances of corporate recidivism (based on a single US data Registry sample according to which 301 corporations had already failed to be deterred to offend by the threat of DPAs), falls short in cogently proving the case against DPAs’ effectiveness in fulfilling their crime-prevention and offender-correction functions. Yet, what our evidence shows is, first, the poor deterrent effect of the threat of punishment posed by DPAs, second, that a non-trivial number of MNEs that benefited from the lenient treatment granted by DPAs as a carrot for self-disclosure and cooperation, continue to repeatedly violate the law and play the judicial system. We leave it to the reader to contemplate what the outcome on recidivism for those 301 MNEs with DPAs would be, within a hypothetical counterfactual scenario where such corporations had been found, in the first instance, criminally guilty and sentenced in a court of law with delicensing or debarment.
The significance of our findings aside, a final caveat is in order. The very complex nature of DPAs lends itself to a myriad of ethical evaluations premised on different philosophical and judicial notions of justice and moral judgments according to numerous alternative normative systems concerned with the standards that define principles of ethical behavior. As such, we do not claim to have provided a definitive verdict on the issue nor of having exhaustively explored all the theoretical and philosophical lenses of analysis available to subject DPAs reached by MNEs culpable of foreign corruption to ethical scrutiny. Given the large breadth of scholarship traditions in moral philosophy and applied ethics (ranging from Rawlsian fairness considerations through to stakeholder ethical theory) we could have drawn from, and the fact that within the confines of a journal article we had by necessity to be selective in our treatment of issues, we leave this task to future research. This caveat notwithstanding, the significance of our contribution also lies in having started what we see as a much-needed debate that has to date been absent in relevant literature and, in so doing, having opened a conversation of research on the ethics of DPAs that could profitably extend beyond the range defined by the theoretical lenses we used in this article.
We are also aware that legislators, judges, or prosecutors pragmatically concerned with the business of making laws, administering punishment, or reaching settlement agreements may be somewhat circumspect of our philosophical discussion on the principles which make DPAs morally intolerable. As Hart (Reference Hart1960: 2) noted more than half a century ago, “A judicial bench is not and should not be a professorial chair.” However, given the demonstrated Devil’s pact-like ethical connotations of DPAs in the context of corporate crime from both deontological and utilitarian perspectives, magistrates and legislators alike are best advised to at least reflect on the moral inadequacy of DPAs, particularly as they apply to theories of deterrence, retribution, or reform against which all questions about the justification of settlement (not just punishment) are to be answered.
Acknowledgments
We wish to thank two anonymous referees and the associate editor, Denis Arnold, for many helpful comments on previous drafts. The usual disclaimer applies.
Glauco De Vita is Professor of International Business Economics in the Centre for Business in Society (CBiS) at Coventry University (UK). He is also a strategic director of the CBiS PhD program. He has published widely in the business and management field and has a keen interest in international business ethics and corporate social responsibility.
Donato Vozza (donato.vozza@roehampton.ac.uk, corresponding author) is a senior lecturer at Roehampton Law School, University of Roehampton (UK), and a member of the Research Centre for Sustainability and Responsible Management. His fields of expertise include corporate crime and business integrity.
Open access
