Sage Journals: Discover world-class research

Abstract

Leniency offers corporations the possibility to come clean about their involvement in cartel conduct (for example, price-fixing, bid-rigging) in exchange for immunity or reduction of financial penalties. In Europe, nearly 60 percent of detected cartels are discovered through leniency. This makes leniency the most applied detection tool for uncovering cartel conduct violations. What are the considerations in applying for leniency or refraining from doing so? How do those considerations relate to private law enforcement through civil liability regarding business cartels? These questions are discussed based on semi-structured interviews (n = 34) with cartelists, competition lawyers and in-house legal counsel to study theoretical assumptions underpinning leniency arrangements in the Netherlands. This study investigates four scenarios on the use of leniency suggested in the literature and finds empirical support for only two. Strategic use of leniency and false confessions occur in the Netherlands, but to a lesser extent than the existing literature suggests. Moreover, various disincentives, and especially the rise of private enforcement, make leniency an unattractive and uncertain option for cartelists.

Keywords

Business cartels corporate crime regulatory enforcement whistle-blowing

Introduction

Cartel confessions are an essential source for competition authorities in uncovering cartel conduct violations (for example, price-fixing, bid-rigging). Leniency offers corporations the possibility to confess to their involvement in cartel conduct in exchange for immunity or a reduction in financial penalties. In Europe, nearly 60 percent of detected cartel cases are discovered through leniency applications, making it the most employed detection tool by competition authorities in uncovering cartels (Carmeliet, 2012). Over the last decade, more than 50 jurisdictions worldwide have adopted leniency arrangements in their competition and anti-cartel enforcement policies (Beaton-Wells, 2015). Leniency arrangements contributed to the detection of a substantial number of cartels across the globe. For example, owing to its amnesty programme, the US Antitrust Division secured more fines in 1999 alone than the sum of fines imposed under the Sherman Act since its adoption in 1890 (Aubert et al., 2006). Leniency also triggered the largest financial penalties imposed by the European Commission (for example, Vitamins, Auction Houses, Air Cargo) (see Carmeliet, 2012; Stephan and Nikpay, 2015). Recently, however, the number of leniency applications has declined worldwide, causing concern amongst experts about the dependency of competition authorities on this detection tool (Guttuso, 2015; ICN, 2014). Therefore it is important to understand more about the decision-making process and motivations regarding leniency applications. This study contributes to that understanding by empirically investigating motivations and disincentives for leniency applications. Does leniency work according to the theoretical assumptions underpinning it or do cartelists use leniency strategically and, if so, to what extent?

Previous research has addressed these questions through economic modelling (see Blum et al., 2008; Hinloopen, 2003; Hinloopen and Soetevent, 2008). This gives insight into potential interactions of cartels with leniency and theoretical scenarios on strategic use, but economic models lack empirical validation. Socio-legal scholars have conducted empirical research through single case studies or systematic document analyses of detected cartel cases (see Bergman and Sokol, 2015; Brenner, 2009; Stephan and Nikpay, 2015). Some studies also include interviews with cartelists, but these studies have not specifically addressed leniency (see Parker, 2013). The most extensive study regarding leniency is by Sokol (2012), who uses surveys and interviews amongst antitrust practitioners in the US. Through his study, Sokol (2012) was the first to probe the assumptions underpinning leniency by studying how leniency decision-making works in practice. However, Sokol did not interview cartelists themselves and the study is US based only. Although many studies have looked at the effectiveness of leniency, at optimal leniency and some at leniency in legal practice, most of these studies are conducted in the US (see Harrington, 2008; Sokol, 2012). It is known that the US antitrust policy is set in a more adversarial legal system compared with European institutions (see Kagan, 2006). Therefore, the findings of this research cannot be automatically generalized to the European context. This study contributes to extant research by adding a case from a European country and by adding first-hand experience from cartelists. Building on insights from these earlier empirical studies, this study looks at the specific Dutch context for leniency and investigates the theoretical assumptions behind leniency. The following questions are addressed: what are the considerations in applying for leniency or refraining from doing so? How do those considerations relate to private enforcement of business cartels? And to what extent do cartelists use leniency strategically? This article aims to answer these questions through interviews with competition lawyers, in-house legal counsel and cartelists in the Netherlands.

Theory: Assumptions underpinning leniency arrangements

Cartel agreements are based on mutual trust between cartelists. This trust entails the willingness to make oneself vulnerable to the risk of betrayal by others in the cartel (Leslie, 2006). However, explicit agreements, informal contracts and secret book-keeping are often also part of running cartels. Those can be seen as attempts to formalize or solidify this trust and as signs of an underlying lack of trust at the same time (Jaspers, 2017). Leniency plays into this trust dynamic, attempting to disrupt cartels by increasing the benefits of betrayal for cartelists (Spagnolo, 2000). Leniency facilitates this by offering immunity to the first firm that comes forward to the authority with evidence of their involvement in cartel infringements. In addition, leniency provides a reduction in financial penalties for firms with sufficient evidence that follow after the immunity-seeker. Policy makers and enforcement authorities rely on leniency to undermine potential trust in cartel agreements and induce distrust between cartelists, ultimately destabilizing cartels (Choi and Hahn, 2014). In combination with deterrent penalties and a credible threat of detection, leniency is presumed to instil distrust between cartelists and trigger a race to the competition authorities (Stucke, 2015). Extradition of cartelists to foreign countries and examples of cartelists sentenced to prison in the US presumably increase the credible deterrence (Crofts and Nylen, 2015; Thomas et al. 2017). In reality, cartels might also end for exogenous reasons such as changing market conditions, change of management, mergers and acquisitions (Levenstein and Suslow, 2006, 2011). Because we observe only discovered cartels, it remains unknown to what extent leniency programmes actually destabilize existing or deter potential cartels (Harrington, 2008).

Cartel enforcement is based on the Beckerian theory of optimal deterrence (Becker, 1968), also referred to as the Chicago School Approach (Sokol, 2012). The experimental support for the Beckerian Proposition is mixed, given different contexts and markets, and there is a lack of consensus on the appropriate level of deterrent fines. The underlying assumption is that firms balance expected gains from a violation against the expected punishment. In this theory, deterrence is a function of the size of the penalty and the probability of detection (Wils, 2007). Stephan and Nikpay (2015) identify three main theoretical assumptions underpinning leniency arrangements: (1) cartelists are rational and unified entities with centralized decision-making, (2) cartelists have accurate and predictable information about the expected benefits of the cartel, and (3) deterrent penalties and a credible threat of detection make leniency an attractive option for cartelists. These three assumptions – rationality, predictability and deterrence – will be compared with alternative explanations in the existing empirical research on leniency, resulting in three hypotheses for this study.

Rationality and unified entities with centralized decision-making

The first assumption underpinning leniency policy is that participants in a cartel are rational actors whose behaviour is primarily determined by profit incentives. Additionally, firms are believed to be unified entities with centralized decision-making processes and assumed to make decisions about when to enter, leave or report a cartel at the top level. Contrary to the idea of firms as unified entities, corporations are in fact complex and fragmented institutional structures with decentralized decision-making in various units and locations (Harding, 2013). Firms are complex structures owing to differences between individual and firm incentives, and firms may have difficulties in controlling and monitoring rogue agents (Sokol, 2012). Information deficits exist within corporations: different departments can fail to communicate and executives can be unaware of what happens on the shop floor. Differences exist between smaller and larger firms regarding decision-making about compliance (Parker, 2013): large corporations with many subsidiaries face the difficulty of identifying infringements and of managing improvement if cartel violations have occurred (Braithwaite and Makkai, 1991; Jamieson, 1994). However, large corporations have more resources available to adopt professional compliance programmes and legal advice. Smaller firms have fewer difficulties in monitoring cartel conduct: within smaller firms (for example, family owned, owner-director), directors or managers are more likely to be aware or actively involved in conduct. Reversely, small firms will have fewer resources available regarding compliance programmes.

Moral ambivalence and ambiguity are known to play a role in cartel conduct violations (Haines and Beaton-Wells, 2012; Parker, 2012; Whelan, 2013). Business people can find competition law abstract and complex or might not support the moral wrongfulness of cartel conduct (see Hertogh, 2010). Therefore, the expectation is that the size of firms and their legal consciousness (awareness and support for legal rules) influences to what extent they are aware of cartel conduct and inclined to act on this. Studies show that low legal consciousness towards competition law causes calculated interaction with leniency. The literature finds a paradox between cartel criminalization – enhanced enforcement, significantly high financial penalties – on the one hand, and fixed collectivistic sentiments and values in business cultures on the other (see Stephan, 2010). Within firms, the topic of competition law and cartel conduct is associated with moral ambiguity (Parker, 2012; Whelan, 2013). This is closely related to what Gilbert Geis (1977) noted in his seminal study on the heavy electrical equipment cartel cases. He quoted a Westinghouse executive rationalizing price-fixing: ‘Illegal? Yes, but not criminal. . . . I assumed that criminal action meant damaging someone, and we did not do that’ (1977: 67). This translates into the attitudes and self-perception of cartelists. They realize that certain economic action and business conduct is strictly speaking not allowed, yet they do not perceive it as unjust or morally objectionable.

The predictability of expected benefits and likelihood of detection

The second assumption underpinning leniency policy is that firms have accurate information about the expected benefits of the cartel, and adequately assess the likelihood of detection and the size and likelihood of punishment. Competition authorities are known to frame leniency as triggering ‘a race to the authorities’ between firms that struggle with the risks of their involvement in cartel conduct, distrust towards their ‘partners in crime’ and remorse (Stucke, 2015). Competition authorities actively present this frame within their regulatory communication (Van Erp, 2018). Through information on their websites, (social) media campaigns and movie clips, competition authorities portray leniency as an attractive choice for cartelists that are considering their options. However, the suggestion that leniency destabilizes active cartels has been challenged in research. Based on a case-study analysis of 40 international cartels, Stephan and Nikpay (2015) concluded that 53 percent of those cartels ended before parties applied for leniency and only 6 percent ended after they applied. More studies underline this conclusion (Brenner, 2009; Stephan, 2008) and also suggest that reporting is sometimes used strategically, challenging the idea that leniency ends existing cartel agreements (Miller, 2009). In addition, practitioners question the reliability of testimonies derived from cartel confessions. It is suggested that whistle-blowers have an incentive to exaggerate their testimony, in order to be granted immunity or a reduction in penalties (Snoep, 2009).

Harding and Edwards describe enforcement gaming as follows: ‘it may be, then, that another aspect of the impact of regulation and the use of sanctions in this context is an unintended and unwelcome . . . “reverse exploitation” or capture of the process of enforcement by firms subject to the regulatory process as a kind of business opportunity’ (2015: 206). Scholars suggest several ways in which cartelists use leniency strategically (Harding et al., 2015; Levenstein et al., 2016; Marvão, 2016; Marx et al., 2015). The large number of gaming strategies listed by Harding et al. (2015) include strategies to hurt competitors, to intentionally bust the cartel for personal gain or to divert attention from regulators away from a bigger cartel (Marx et al., 2015). Marx et al. (2015) use game theory to predict firms will apply for leniency for several products at a time. The likelihood of investigation and conviction will increase for the first product but decrease for subsequent products. Firms will have an incentive to form ‘sacrificial cartels’ and apply for leniency in less valuable products. Using this strategy, cartelists can reduce the likelihood of detection and convictions in more valuable cartels. This can also be referred to as a ‘throwing-the-bone strategy’ – diverting the attention of regulators away from a larger cartel. In other words, it is expected from the literature that firms use leniency strategically when they confess. However, for many of the theoretical scenarios it remains unclear to what extent empirical data support them and under which conditions. Therefore, this study will probe to what extent and in which ways leniency is used strategically in the Netherlands.

For this study, different theories on the use of leniency in the literature have been condensed into four scenarios: (1) the destabilizing effect (default, no strategic use), (2) the opportunistic response, (3) the anticipation effect, and (4) the throwing-the-bone strategy. Firstly, the destabilizing scenario is leniency by the book, similar to how competition authorities generally bring it forward in their communication: destabilizing active and current cartels. Secondly, the opportunistic response is when cartels have already failed for other reasons. In that case, firms might request leniency to put their former cartel members at a disadvantage. Thirdly, the anticipation effect signifies a cartel as an intentional and planned construction, only out to eventually damage other market players in the long run. When the firm is the first to ask for leniency it makes it immune from sanctions. Note that this means leniency does end the cartel in this scenario, although it is used in a strategic way. Fourthly, the throwing-the-bone strategy entails firms holding a large international cartel agreement in product X while also retaining an agreement on a smaller side market Y (or markets) for a different product. Now firms individually or collectively ask for leniency in the smaller cartel to distract the efforts of the authorities and to decrease the risk of detection of the larger cartel. These four scenarios are operationalized for this study using scenarios in the interviews (see the Methods section).

Deterrence and a credible threat of detection

The third assumption underpinning leniency arrangements is that they are effective only when combined with penalties of sufficient magnitude to deter and a credible threat of detection. In addition, competition authorities can exploit the uncertainty around the strength of an investigation by bluffing (Sauvagnat, 2015). However, leniency policies do not operate in a vacuum; they interact with other legislation and enforcement, such as criminal liability and private damage actions (Luz and Spagnolo, 2017; OECD, 2012). Although it may be beneficial to self-confess in order to be granted immunity or a reduction in penalties by public enforcement authorities, admitting to a violation and providing evidence for it can also be a disadvantage in follow-on civil damage cases because immunity from public sanctions does not extend to follow-on civil damage claims (Canenbley and Steinvorth, 2011; Cauffman, 2011; Guttuso, 2015; Swaak and Wesseling, 2015; Wils, 2009). Buccirossi, Marvão and Spagnolo (2015) demonstrate that damage actions could potentially improve the effectiveness of leniency programmes, but only if the civil liability of the immunity recipient is minimized and access to all evidence is granted to claimants. However, as Guttuso puts it: ‘the probability of exposure to private damages not only in the jurisdiction where leniency was obtained, but also in other jurisdictions, have been cited as key factors inhibiting potential leniency applicants from self-reporting’ (2015: 279).

Private enforcement in the form of follow-on damage procedures is the most important reason cartelists are increasingly cautious about leniency. These procedures are civil cases following after an infringement decision from the European Commission or national competition authorities. In the Netherlands, there have been developments in several cases where claimants are seeking redress from cartelists, such as TenneT/ABB, Sodium Chlorate, Paraffin Wax and Air Cargo. In TenneT/ABB, the court ordered ABB to pay TenneT €68 million in damages. In Sodium Chlorate and Paraffin Wax, the German litigation funder CDC (Cartel Damages Claims) was involved. The Netherlands is a popular forum for claiming cartel damages. This attracts professional litigation funders from abroad. Litigation funders gather claims from damaged parties in the market and litigate on their behalf in order to collect a return on their investment in the form of a percentage of the damage claims, if granted (see Maton et al., 2011). For example, in the Air Cargo cartel: two days after the dawn raids, 20 class action lawsuits were filed against Lufthansa alone in the US. Lufthansa ended up settling for US$85 million. In addition, Lufthansa’s outside legal costs relating to the cartel averaged US$12 million a year between 2005 and 2009 (Bergman and Sokol, 2015). In short, it is expected that disincentives and stakeholders outside of cartelists and the competition authority might influence the decision-making process of leniency for corporations.

Table 1 summarizes the main theoretical assumptions underpinning leniency arrangements and the alternative empirical explanations that challenge them. This study assesses whether there is support for these alternative explanations in the Netherlands, and to what extent, by answering the following question: does leniency work according to the theoretical assumptions underpinning it or do cartelists use leniency strategically and to what extent?

Table 1.

Leniency policy assumptions and empirical explanations.

Theoretical assumptions about leniency	Alternative empirical explanations
Firms are rational and unified entities with centralized decision-making	The distribution of responsibilities within firms complicates centralized decision-making and moral ambiguity distorts rational decision-making
Cartelists have accurate information on the expected benefits of the cartel	Firms use leniency strategically if and when they confess
Deterrent penalties and a credible threat of detection make leniency attractive	Leniency applications pose additional risks and uncertainties to cartelists and lead to increased exposure to private enforcement

Methods

For the purpose of this article, a qualitative study amongst corporate and legal professionals in the Netherlands was conducted (see Sokol, 2012). As mentioned, most socio-legal studies on leniency in cartel cases are based on single case studies or systematic case-file analyses. Unfortunately, leniency case-file reports – including transcripts of confessions, internal communication and documentation of leniency applicants – are classified. In the Netherlands, the national competition authority was unwilling to grant access to these files. Owing to both the limited availability of official documents as empirical material and the nature of the research question – which is directed at qualitative understanding of the interaction between leniency and cartelists – motivations and perceptions are studied through the use of interviews.

I conducted 34 semi-structured interviews with competition lawyers (17), in-house legal counsel (6) and cartelists (11), such as executives and directors. The reason for also selecting non-cartelists as respondents was threefold. Firstly, contacting and establishing trust with legal professionals in the field of competition law was essential in order to gain access to cartelists. The cartelists in this study were clients and professional contacts of these lawyers and in-house legal counsel. Secondly, cartelists have only a one-time experience with leniency, which is informative but means they lack more general and comparable knowledge of several cases (for example when it comes to assessing strategic gaming scenarios). Lawyers and in-house counsel provided for broader experience. Thirdly, when cartelists decide whether or not they will apply for leniency, legal professionals are influential agents in this process. Although a final decision on whether or not to apply for leniency lies with the firm, the expertise of legal professionals causes an information imbalance with their client that grants them power in the decision-making process. Therefore, their experience and attitude towards leniency have significant effects on the decision of their client.

The competition lawyers and in-house legal counsel all had extensive experience in consultation and defence in (alleged) cartel infringement cases. In addition, most of them (14) had experience with preparing and submitting leniency applications for clients. Because cartel cases both tend to be a specialist topic and are often considered high-profile cases with high financial and reputational stakes, a combination of a level of seniority and expertise amongst respondents was essential. Therefore, all respondents were selected on the basis of their professional experience and expertise. Respondents held relevant professional experience of a minimum of 8 years up to 41 years and all were experts in the field of competition law. The competition lawyer respondents all worked in one of the following occupations: partner, lawyer-partner, senior counsel or off-counsel. The in-house legal counsel respondents also had a relevant professional experience of a minimum of 8 years, up to 18 years. In addition, interviews were held with cartelists (11) – chief executives and directors with experience in the process of considering leniency related to (potential) cartel conduct violations in their company. Their professional experience in an executive or director role varied between 5 and 26 years. All the corporate professional respondents had a one-time only experience with cartel infringements and/or leniency. Two had a background in law, the others in business studies or economics. The cases involved the construction industry (4); heavy industry (3); the general services industry (2); and the financial industry (2). The average duration of the cartels was 3.4 years. The number of participants in the cartels varied between 3 and 7 firms. Typically, the cartelists physically met twice a year. Agreements were made through the use of minimum pricelists, customer lists and geographical market allocation. The cartelists (n = 11) interviewed for this study were involved in a wide variety of cartel types. These cartels initially started through meetings in market associations and industry connections through subcontracting. Cartel agreements were triggered by low margins, pressure by other companies in the supply chain or uncertain market conditions. The motives for starting the cartel and the conditions under which they took place had consequences for the decision to apply for leniency or refrain from doing so. Most notably, in several cases the fear of the reactions of others in the industry (social control) is a reason to refrain from applying for leniency. Six cases ultimately led to an administrative fine from the competition authority. Respondents pursued leniency in four cases. In two of these four cases, respondents indicated they wanted to change their standing business practices and leniency provided a way to do that. In two cases, cartelists used leniency strategically to improve their position by hurting their co-cartelists. However, these cases can be described mostly as an opportunistic use of leniency: the cartel had already ended or the cartelists no longer upheld the agreement at the time of the leniency application.

In order to gain access to the field and establish contacts with respondents to conduct interviews, the Dutch national association for competition lawyers was contacted. The board of the association forwarded a request in their newsletter, explaining the research and the call for respondents. Several respondents contacted me after the call in the newsletter, others were contacted after online searches for contact information, and others were approached through the so-called snowballing method (every respondent was asked to recommend additional respondents, who were contacted). Clients (in-house legal counsel, chief executives and directors) were also contacted through this method of snowballing. The interviews themselves took place at the offices of the respondents. Mostly, these interviews were held in the conference room, and sometimes in personal offices or coffee corners. Interviews typically lasted for 50 minutes, with outliers of 45 and 150 minutes. The interviews were audio recorded and transcribed afterwards.

During the interviews, the following topics were discussed: issues relating to fragmented institutional structure (for example, the role of legal professionals) and decentralized decision-making within companies relating to cartel infringements and leniency; incentives and disincentives to apply for leniency; and the role and influence of follow-on civil proceedings (damage claims, compensations) and the influence of professional litigation funders. Four scenarios on the use of leniency from the theoretical framework were presented to respondents for confirmation or falsification regarding their practical experience with leniency. The four theoretical scenarios on the strategic use of leniency were introduced to respondents through the use of short scenario descriptions (see the online appendix) and respondents were asked to comment on the accuracy of these scenarios in their daily legal practice and asked to give factual examples. The transcripts of the interviews were coded for the following topics: decentralized decision-making; moral ambiguity; framing and reliability; opportunistic use; civil damage claims. In addition, interviews were coded for verification or falsification on each of the four scenarios.

Limitations: Non-response and social desirability

About 30–40 percent of the potential respondents I approached declined to be interviewed for this study. This mainly concerned those who were contacted after the online search (‘cold calls’). All non-responses were followed up with a question about the reason(s) for non-response. Where indicated, reasons for non-response amongst lawyers were mainly a lack of time, interest or sufficient expertise in their own opinion. Reasons for non-response amongst in-house legal counsel, executives and directors mainly concerned issues regarding confidentiality. In-house legal counsel often sign internal contracts with their company, or between company and regulator in the case of settlements. This constrains them from speaking about or commenting on the case. In addition, the negative image of big business (especially multinational corporations) as non-compliant was noted as a reason not to cooperate in this study. Concerning non-response amongst cartelists, it was mostly multinational firms that declined to participate in the study, which might indicate a slight overrepresentation of more local and smaller businesses. On the other hand, the majority of interviews with lawyers were with lawyers in large international and transnational firms dealing with major multinational corporations as clients, so their experience is represented in the data from that end.

Semi-structured interviews have limitations. Firstly, it is not possible to generalize results to a larger population. Therefore, the results in this study are not representative of all cartels in the Netherlands. The purpose of this study is to understand the nature of the interactions regarding leniency. Research findings regarding the interaction between regulation and businesses can have a wider relevance, for example in other domains of law. They inform us about the nature of interactions between different types of professionals when it comes to the decision-making process around compliance. Secondly, the social desirability of respondents is an issue in interviews. Naturally, lawyers are advocates for their clients’ best interests and therefore, to a certain extent, also their conduct. To correct for anticipated social desirability, I used my extensive research through case studies (see Jaspers, 2017) on recent national cartel cases. In addition, during the interview I consequently invited respondents to provide specific examples from their own professional experience to prevent superficial answers, avoid generalizations and socially desirable responses.

Results: Leniency applications in the Netherlands

As per the theoretical framework, the results are categorized using the assumptions underpinning leniency policy: rationality, predictability and deterrence.

Rationality: Centralized versus fragmented decision-making

The results of this study confirm that fragmented institutional structure and decentralized decision-making within firms cause differences in awareness of cartel conduct between departments and demonstrates that the awareness of senior management in the event of cartel infringements diverges between smaller and larger firms (see Harding, 2013; Sokol, 2012). These differences can be attributed to the variety in firm size. The interviews show how large (multinational) companies with multiple subsidiaries struggle with ‘local’ issues more often when it comes to cartel conduct violations. Smaller and mid-size companies include active involvement by the executive director or owner-director more often than do large firms (see Jamieson, 1994). The difference in size also affects the decision-making process in applying for leniency. Smaller firms often entail the direct involvement of executives and owner-directors, so that the decision to apply for leniency will affect the executives directly and will have personal consequences. A confession means coming clean about your own personal responsibility for the infringement. This is in contrast to a chief executive of a multinational company that deals with a ‘local issue’ at a subsidiary. An executive will experience more personal distance in the decision-making process regarding leniency.

Larger firms will generally be better equipped to deal with compliance procedures and legal issues regarding competition law. Hence, they are better equipped than smaller companies to hire internal and external legal advice and consultation. This enables them to better organize, structurally incorporate and implement fair competition rules in their compliance programmes. Smaller companies have fewer resources and staff (or none) to organize compliance in a professional and structural way. The results indicate that infringements by small firms are often unintentional and a result of ignorance or a lack of professionalism. However, this does not explain why both small and big businesses are involved in cartel conduct violations. There is more to it than knowing the law. Parker (2013) demonstrates how knowledge about the law is less important to companies than their relationship with the law. Bigger ‘elite’ firms see themselves as intimate with the law and able to ‘game’ it in their advantage. Smaller companies perceive themselves as unknowing towards compliance. This perception influences internal support for the rules, otherwise known as legal consciousness.

Rationality: Moral ambiguity

Secondly, this study demonstrates that low legal consciousness towards competition law causes calculated interaction with leniency. The results show that cartelists present several reasons and rationalizations for their misconduct. These relate to tradition, market structure and mutual dependencies between competitors in a market. Lawyers also indicated how their clients often submit explanations or rationalizations regarding their cartel activities. Cartelists embrace these justifications and are even happy to explain their illegal conduct to competition officials:

They believe the market conditions justify it . . . that they. . . feel like they have to, because of the small profit margins. I notice this with my clients. For instance, a big problem in these interrogations with the competition authority is how clients are more than happy to tell them [the authorities] everything they did and why they did it, because they truly feel that what they did is the right thing. (R3)

This example illustrates how cartel conduct is integrated in traditions within certain markets. Cartelists have internalized justifications of these practices. In a few markets and industries, these justifications are institutionalized. The results demonstrate a paradox between increased awareness and enforcement on the one hand and sentiments of moral ambiguity on the other. From regulatory studies we know that a lack of support for rules or moral condemnation of conduct can result in an instrumental interaction with rules and regulation (see Blumenthal et al., 2001; Braithwaite, 1995; Gneezy and Rustichini, 2000; Van de Bunt and Huisman, 2007; Van Wingerde, 2012).

Competition lawyer respondents with the longest working experience (>20 years) point out the changing regulatory context over the past two decades (see Beaton-Wells and Ezrachi, 2011). They explain how the expansion of global cartel enforcement has induced elevated levels of awareness and increased professionalism in compliance programmes and special compliance departments regarding competition law. Today, heavy financial penalties make competition law a top priority in internal compliance programmes and training sessions. However, the topic simultaneously generates the most moral ambiguity amongst business people according to respondents. Topics such as bribery or privacy laws, also considered important issues in contemporary compliance, are more unambiguously supported according to the interviewees.

Predictability: Strategic use of leniency

The interviews demonstrate that firms use leniency strategically, but in a limited number of ways. Firstly, the results do not support more sophisticated scenarios, such as an anticipation effect – cartelists entering into a cartel with the intention of asking for leniency at a later stage to damage their cartel ‘partners’ financially (see Harding et al., 2015) – or the ‘throwing-the-bone strategy’ – distracting the attention and resources of regulators by asking for leniency in several product markets (see Marx et al., 2015). The results do not support these two scenarios for the strategic use of leniency in the Netherlands. However, the results do support the ‘opportunistic’ use of leniency and selective framing of confessions. Opportunistic use is either reporting cartels that have already ended, for example in the process of mergers and acquisitions (to avoid or reduce accountability at a later stage), or reporting in order to get ahead of competitors with more market power when market conditions change.

Examples of the opportunistic use of leniency are cartelists that are in a so-called hub and spoke construction with a more powerful market player. In a hub and spoke cartel, the dominant firm is the producer of one of the resources for example, so that its competitors are also its clients. These dominant firms set conditions for other players in the market. In two cases, companies used leniency to damage the dominant firm in order to get out of restricting contracts and market conditions. One cartelist explained why he applied for leniency:

After a few years this player made it increasingly difficult for us to do business . . . Every year I was presented with a list of prices for the upcoming year . . . And yes, as with any of our activities this was a way to push the other firm out of business, or at least to make their life as miserable as possible. (R28)

Concerning the strategic use of leniency, other studies point to elaborate gaming schemes and strategies employed by cartelists (Harding et al., 2015; Marx et al., 2015). However, this study found more subtle and nuanced forms of framing and using leniency in practice. By selective framing of confessions, cartelists present their testimony selectively, through either withholding certain information or overstating certain facts and circumstances in order to be granted leniency. For example, respondents refer to ‘over-confession’ when they describe how cartelists frame their confession to leniency officers. When, after internal deliberations, a firm has committed to applying for leniency, they are highly motivated for the application to be granted. Competition authorities set certain standards for the application, such as that substantial evidence should be provided and facts and circumstances must indicate a clear infringement. This can put a strain on the reliability of the testimony of leniency applicants. Respondents also noted that they have come across examples in practice where several leniency applications are contradictory and inconsistent in the same case, indicating false confessions.

Competition lawyers with extensive experience consider applying for leniency to be an exception to the rule. If they do apply for leniency this seems to be driven more by inevitability and opportunism. Internal investigations are part of the process of mergers and acquisitions. These internal investigations can result in firms applying for leniency. Change in management can create momentum for a fresh start or leniency is used to settle possible liability risks in the future. These leniency applications often entail ceased activities or ‘dead’ cartels, as one competition lawyer points out:

By a clean slate I mean in case of selling the company or when you are about to enter into a merger. Maybe there are some skeletons in the closet that you have to get rid of. And you can do this through applying for leniency. (R14)

Deterrence: Disincentives to apply for leniency

The fourth result from this study is that several disincentives make leniency an unattractive option for cartelists in most cases. In the literature, corporations involved in cartel conduct are often considered agents in a prisoner’s dilemma (in the context of game theory; see Blum et al., 2008) regarding the decision on whether or not to apply for leniency. However, when there is no reason to assume that information about the illegal agreements is reaching the authorities and there is sufficient mutual trust, the best option for all those involved is ‘not to play the game at all’ (Leslie, 2006). The results from this study indeed demonstrate a number of reasons for firms involved in cartel conduct not to play the game, namely: insufficient evidence of the conduct for an application, social control (reactions in the market), cultural objections, the costly and time-consuming internal investigations and preparation of leniency applications, and the influence of follow-on civil damages procedures after investigations and fines from public competition authorities.

First off, respondents explain that facts and circumstances might indicate an infringement, but too little evidence can be retrieved to assure that an infringement actually happened. If there is a lack of evidence after internal investigations, firms cannot provide the authority with sufficient information to be granted leniency. The following example by an in-house legal counsel illustrates this:

Internal investigations pointed out that regionally there had been deliberations and illegal agreements between competitors. This was mapped into detail and was discussed with head legal in Germany on further steps to follow. A draft leniency application was even drawn up. In the end we decided against leniency, because there was simply insufficient written evidence. Nothing but suspicions really. (R32)

Secondly, the reactions of others in the market are also a reason to remain silent in the case of possible infringements. The social control backlash from ‘telling on your peers’ varies greatly, especially depending on the type of market. Depending on how certain markets are structured, they induce great dependency between competitors. Damaging trust by telling the authorities about certain agreements that might qualify as an infringement can gravely damage that trust. Damaging trust might form a disadvantage to firms in future activities in the market. The following example from an in-house legal counsel of a mid-size firm illustrates this:

The market we operate in is defined by a number of legitimate collaborations on numerous activities. And you know, if you take these steps [apply for leniency] you will be treated like dirt by the others in the market afterwards. (R16)

Thirdly, respondents report cultural objections. Cultural objections have a social dimension, but should be understood as more principle-based objections towards leniency. For example, respondents refer to leniency as ‘telling on someone’ or ‘betraying others’, and they qualify this as immoral in and of itself. The idea of being disloyal to other businesses in the market, which are often considered colleagues in business, is often perceived as incompatible with personal values. A competition lawyer of a multinational firm explains:

I have had clients that simply say: ‘we will not betray others’. Sometimes you have to explain to them as their lawyer: it can actually be the best scenario for you. (R6)

Fourthly, the burden of going through a leniency process, in terms of financial costs, the (management) time and other strains it puts on firms’ operations is another reason to refrain from leniency. The high costs of legal consultation and of internal data collection to prepare a leniency report and the management time of highly paid executives puts a brake on business activities for months or years. More importantly, these are certain and short-term costs the firm is confronted with, in contrast to the highly uncertain and long-term consequences of an administrative fine. In that sense, it can be distorted cost–benefit analyses when firms do end up with a financial penalty. However, it is their perception at the time of deciding on leniency that drives their decision. There is also a high degree of uncertainty in terms of the legal position of the company during the procedure and the eventual outcome. Before the investigations by the authorities and the court decisions are done, this uncertainty can easily last years. Two competition lawyers state:

Applying for leniency requires a ton of work, which is often underestimated. It demands lots of statements, documents and paperwork and is therefore a costly procedure. (R25)

Sometimes you are in a leniency procedure and your gut already tells you: ‘this is not working out’. When the authorities keep asking for additional information again and again and there is no real progress. It is way less structured and comes with a lot more uncertainty, more uncertainty in fact than regular cartel investigations [ex officio investigations]! (R5)

Table 2 presents a summary of the most commented on perceived incentives and disincentives in applying for leniency in this study. The fifth disincentive – the risk of follow-on damage claims – involves the relationship between firms involved in cartel conduct violations and damaged parties in the market seeking compensation. These private stakeholders are consumers, anywhere down the line in the supply chain of a particular market, that can make a convincing claim for losses as a result of cartel agreements. Exposure to private cartel enforcement was frequently commented on by respondents.

Table 2.

Summary of perceived incentives and disincentives regarding leniency.

Incentives to apply for leniency	Disincentives to apply for leniency
Full disclosure in light of the Mergers & Acquisition process	Insufficient evidence
Opportunistic response	Social: response in the market
	Cultural objections
	Costly and time-consuming
	Follow-on damage claims

Deterrence: Interaction between private and public enforcement

Although private enforcement is relatively recent in the Netherlands, respondents explain how the risk of being confronted with follow-on civil damages proceedings induces more caution towards applying for leniency. It also causes most of the leniency testimony to be done verbally and not in writing. Lastly, it also makes parties litigate longer and makes them less inclined to accept settlements from the Commission or national competition authorities, because they can be seen as an admission of guilt. A competition lawyer comments:

I will not be quick to advise leniency anymore, especially because of these follow-on damages claims. That is a classic: ‘the operation was successful, but the patient died’. So leniency no longer means closure, today it is only the mere beginning. (R11)

In line with Cauffman (2011), these results indicate how leniency and private enforcement interact: the threat of follow-on damage claims is an important disincentive to come forward as a leniency applicant. Respondents indicate that civil litigation can lead to corporate liability for damages up to four or five times the amount in (public) financial penalties imposed by the competition authority.

Conclusions

The aim of this article was to compare the main assumptions underpinning leniency policy with the practice of leniency in the Netherlands. This study examined three main assumptions (see Stephan and Nikpay, 2015: 142): (1) firms act as rational and unified entities with centralized decision-making; (2) cartelists have accurate and predictable information about the expected costs and benefits of the cartel; (3) deterrent penalties and a credible threat of detection make leniency an attractive option for cartelists. These assumptions were compared with the existing empirical literature on leniency and a qualitative interview study on leniency in the Netherlands. Relating to these assumptions, this study comprises four main results. Firstly, the awareness of management in the case of cartel infringements diverges between small and large firms. The level of social responsiveness and professionalism in terms of compliance and the relationship with the law differs between smaller and larger companies. This influences decision-making procedures towards leniency, making leniency applications less likely for smaller firms. Secondly, there is a paradox between cartel criminalization – enhanced enforcement, significantly high financial penalties – on the one hand, and fixed collectivistic sentiments and values in business cultures on the other. This paradox results in firms gaming enforcement efforts and the strategic use of regulation. Thirdly, firms use leniency strategically, although in limited ways. Support for both selective framing and the opportunistic use of leniency was found in the interviews but not for the more sophisticated strategies suggested by some authors (Harding et al., 2015; Marx et al., 2015). Fourthly, several disincentives make leniency an unattractive option for cartelists. Disincentives concerning leniency include: insufficient evidence for an application, social control (reactions in the market), cultural (principle-based) objections such as ‘telling on your peers’, the costly and time-consuming internal investigations and preparation of leniency applications, and the influence of follow-on civil damages proceedings after investigations and fines from public competition authorities. Of the 11 cartelists interviewed for this study, 7 decided against leniency. The fear of the reactions of others in the industry is an important reason to refrain from applying for leniency. Respondents pursued leniency in four cases. In two of these cases cartelists indicated strategic use to improve their position by hurting their co-cartelists. However, strategic use in these cases can be characterized as an opportunistic use of leniency; the cartel had already ended for exogenous reasons.

Discussion and implications

The balance and interaction between compensations and leniency is one of the issues the European Commission and national competition authorities (NCAs) struggle with. Several suggestions for legal provision have been made, ranging from preventing disclosure of leniency applications and reducing the risks of damages claims, to introducing additional financial incentives to leniency recipients. However, NCAs can also direct attention towards efforts to better involve other stakeholders, such as professional litigation funders and specialized lawyers, in designing more attractive leniency programmes that include private interests and reduce obstacles and uncertainties in the current programmes. Private enforcement seems to decrease the number of leniency applications and thereby slows down the public enforcement of cartels. Private and public enforcement of cartels interact (Cauffman, 2011). For example, the financial penalty decision that the Commission and NCAs publish is the starting point for litigation funders to collect claims and start follow-on damage procedures. This is where NCAs have leverage to start negotiations with several stakeholders in the field of private enforcement and damages.

In short, this study has demonstrated the major disincentives to applying for leniency. This leaves room to speculate how many cartels fail to be revealed by leniency. Moreover, it gives food for thought regarding the dependency of competition authorities on those cases that did result in leniency applicants. In addition, this influences thinking and the dominant perceptions of what a cartel is, in both enforcement practice and academic research. Leniency applicants largely provide the mythology around what cartels are and what they look like (see Harding and Edwards, 2015) and this might feed collective blind spots. The issues with strategic gaming and framed and false confessions only begin to shed light on how biased is the regulatory and empirical understanding of cartels. Most insights arise from detected cases, which are largely induced by leniency. In that regard, can leniency been seen as regulatory success, in terms of its goal to destabilize active cartels? Or is it a form of regulatory deadweight loss, an inefficient allocation of resources? Moreover, this study has illustrated how the social context influences decision-making on leniency applications and that whistle-blower incentives, such as leniency, do not operate in a social vacuum. For example, in bid-rigging cartels there is interaction with legislation and enforcement around corruption (Luz and Spagnolo, 2017). In other words, besides the public enforcement there are other stakeholders in the process that can greatly influence the ultimate outcome of these confessions. These stakeholders are often insufficiently taken into account in the design of regulatory policies, as leniency proves.

Supplemental Material

Online_Appendix_-_Theoretical_scenarios_for_strategic_use_of_leniency – Supplemental material for Leniency in exchange for cartel confessions

Supplemental material, Online_Appendix_-_Theoretical_scenarios_for_strategic_use_of_leniency for Leniency in exchange for cartel confessions by J.D. Jaspers in European Journal of Criminology

Footnotes

I thank Judith van Erp,Henk van de Bunt and the anonymous reviewers for their comments on an earlier version of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research,authorship,and/or publication of this article: This study was funded by the Netherlands Organization for Scientific Research (grant number 406-14-027).

ORCID iD

J.D. Jaspers

Supplemental material

Supplemental material for this article is available online.

References

Aubert

Rey

Kovacic

(2006) The impact of leniency and whistle-blowing programs on cartels. International Journal of Industrial Organization 24(6): 1241–1266.

Beaton-Wells

(2015) Leniency policies: Revolution of religion? In: Beaton-Wells

Tran

(eds) Anti-Cartel Enforcement in a Contemporary Age: Leniency Religion. Oxford: Hart Publishing.

Beaton-Wells

Ezrachi

(eds) (2011) Criminalising Cartels. Critical Studies of an International Regulatory Movement. Oxford: Hart Publishing.

Becker

(1968) Crime and punishment: An economic approach. Journal of Political Economy 76: 169.

Bergman

Sokol

(2015) The air cargo cartel: Lessons for compliance. In: Beaton-Wells

Tran

(eds) Anti-Cartel Enforcement in a Contemporary Age: Leniency Religion. Oxford: Hart Publishing.

Blum

Steinat

Veltins

(2008) On the rationale of leniency programs: A game-theoretical analysis. European Journal of Law and Economics 25(3): 209–229.

Blumenthal

Christian

Slemrod

Smith

(2001) Do normative appeals affect tax compliance? Evidence from a controlled experiment in Minnesota. National Tax Journal 125–138.

Braithwaite

(1995) Games of engagement: Postures within the regulatory community. Law & Policy 17: 225–255.

Braithwaite

Makkai

(1991) Testing an expected utility model of corporate deterrence. Law and Society Review 25(1): 7–40.

10.

Brenner

(2009) An empirical study of the European corporate leniency program. International Journal of Industrial Organization 27: 639–645.

11.

Buccirossi

Marvão

CMP

Spagnolo

(2015) Leniency and damages. CEPR Discussion Paper No. DP106 82, June Available at SSRN, URL (accessed 22 August 2019): https://ssrn.com/abstract=2624637.

12.

Canenbley

Steinvorth

(2011) Effective enforcement of competition law: Is there a solution to the conflict between leniency programmes and private damages actions? Journal of European Competition Law & Practice 2(4).

13.

Carmeliet

(2012) How lenient is the European leniency system? An overview of current (dis)incentives to blow the whistle. Jura Falconis 48(3): 463–512.

14.

Cauffman

(2011) The interaction of leniency programmes and actions for damages. Interaction 7(2): 181–220.

15.

Choi

Hahn

(2014) How does a corporate leniency program affect cartel stability? Empirical evidence from Korea. Journal of Competition Law and Economics 10(4): 883–907.

16.

Crofts

Nylen

(2015) MLex interview with Romano Pisciotti. MLex December.

17.

Geis

(1977) White collar crime: The heavy electrical equipment antitrust cases of 1961. In: Ermann

Lundman

(eds) Corporate and Governmental Deviance. New York: Oxford University Press.

18.

Gneezy

Rustichini

(2000) A fine is a price. Journal of Legal Studies 29(1): 1–17.

19.

Guttuso

(2015) Leniency and the two faces of Janus: Where public and private enforcement merge and converge. In: Beaton-Wells

Tran

(eds) Anti-Cartel Enforcement in a Contemporary Age: Leniency Religion. Oxford: Hart Publishing.

20.

Haines

Beaton-Wells

(2012) Ambiguities in criminalizing cartels: A political economy. British Journal of Criminology 52(5): 953–973.

21.

Harding

(2013) Criminal Enterprise. Individuals, organisations and criminal responsibility. Oxfordshire: Routledge.

22.

Harding

Beaton-Wells

Edwards

(2015) Leniency and criminal sanctions in anti-cartel enforcement: Happily married or uneasy bedfellows? In: Beaton-Wells

Tran

(eds) Anti-Cartel Enforcement in a Contemporary Age: Leniency Religion. Oxford: Hart Publishing.

23.

Harding

Edwards

(2015) Cartel Criminality: The Mythology and Pathology of Business Collusion. London: Ashgate Publishing.

24.

Harrington

JE (2008) Optimal corporate leniency programs. Journal of Industrial Economics 56(2): 215–246.

25.

Hertogh

(2010) Crime and custom in the Dutch construction industry. Legisprudence 4(3): 307–326.

26.

Hinloopen

(2003) An economic analysis of leniency programs in antitrust law. De Economist 151(4): 415–432.

27.

Hinloopen

Soetevent

(2008) Laboratory evidence on the effectiveness of corporate leniency programs. RAND Journal of Economics 39(2): 607–616.

28.

ICN (International Competition Network) (2014) Drafting and implementing and effective leniency policy. In: Anti-Cartel Enforcement Manual (2014). URL (accessed 22 August 2019): https://www.internationalcompetitionnetwork.org/wp-content/uploads/2018/05/CWG_ACEMLeniency.pdf.

29.

Jamieson

(1994) The Organization of Corporate Crime: Dynamics of Antitrust Violation. London: Sage.

30.

Jaspers

(2017) Managing cartels: How cartel participants create stability in the absence of law. European Journal on Criminal Policy and Research 23(3): 319–335.

31.

Kagan

(2006) American and European ways of law: Six entrenched differences. UC Berkeley, Center for the Study of Law and Society Jurisprudence and Social Policy Program. URL (accessed 22 August 2019): https://escholarship.org/uc/item/0d83p6qk.

32.

Leslie

(2006) Antitrust amnesty, game theory, and cartel stability. Journal of Corporation Law 31: 453–488.

33.

Levenstein

Marvão

Suslow

(2016) Preventing cartel recidivism. Antitrust 30: 81.

34.

Levenstein

Suslow

(2006) What determines cartel success? Journal of Economic Literature 44(1): 43–95.

35.

Levenstein

Suslow

(2011) Breaking up is hard to do: Determinants of cartel duration. Journal of Law and Economics 54(2): 455–492.

36.

Luz

Spagnolo

(2017) Leniency, collusion, corruption, and whistleblowing. Journal of Competition Law & Economics 13(4): 729–766.

37.

Marvão

(2016) The EU leniency programme and recidivism. Review of Industrial Organization 48(1): 1–27.

38.

Marx

Mezzetti

Marshall

(2015) Antitrust leniency with multiproduct colluders. American Economic Journal: Microeconomics 7(3): 205–240.

39.

Maton

Poopalasingam

Kuijper

Angerbauer

(2011) The effectiveness of national fora in Europe for the practice of antitrust litigation. Journal of European Competition Law & Practice 2(5).

40.

Miller

(2009) Strategic leniency and cartel enforcement. American Economic Review 99(3): 750–768.

41.

OECD (Organisation for Economic Co-operation and Development) (2012) Policy Roundtables: Leniency for Subsequent Applicants. URL (accessed 22 August 2019): http://www.oecd.org/competition/Leniencyforsubsequentapplicants2012.pdf.

42.

Parker

(2012) Economic rationalities of governance and ambiguity in the criminalization of cartels. British Journal of Criminology 52(5): 974–996.

43.

Parker

(2013) The war on cartels and the social meaning of deterrence. Regulation & Governance 7(2): 174–194.

44.

Sauvagnat

(2015) Prosecution and leniency programs: The role of bluffing in opening investigations. Journal of Industrial Economics 63(2): 313–338.

45.

Snoep

(2009) To enforce and comply: Inside the Antitrust Agency – Two perverse effects of leniency programmes. Amsterdam Center for Law & Economics, 5 March.

46.

Sokol

(2012) Cartels, corporate compliance, and what practitioners really think about enforcement. Antitrust Law Journal 78(1): 201–240.

47.

Spagnolo

(2000) Optimal Leniency Programs. FEEM Working Paper No. 42.2000.

48.

Stephan

(2008) An empirical assessment of the European Leniency Notice. Journal of Competition Law & Economics 5(3): 537–561.

49.

Stephan

(2010) Cartel laws undermined: Corruption, social norms, and collectivist business cultures. Journal of Law and Society 37(2): 345–367.

50.

Stephan

Nikpay

(2015) Leniency decision-making from a corporate perspective: Complex realities. In: Beaton-Wells

Tran

(eds) Anti-Cartel Enforcement in a Contemporary Age: Leniency Religion. Oxford: Hart Publishing.

51.

Stucke

(2015) Leniency, whistle-blowing and the individual: Should we create another race to the competition agency? In: Beaton-Wells

Tran

(eds) Anti-Cartel Enforcement in a Contemporary Age: Leniency Religion. Oxford: Hart Publishing.

52.

Swaak

CRA

Wesseling

(2015) Reconsidering the leniency option: If not first in, good reasons to stay out. ECLR: European Competition Law Review 36(8): 346–354.

53.

Thomas

De Stefano

Hellings

Shulak

Imoto

Rozent

Agranat

Kay

(2017) Extradition and antitrust law: Businessmen involved in global cartels extradited to foreign countries. Concurrences No. 3-2017.

54.

Van Erp

(2013) De clementiethriller als nieuw filmgenre. Het gebruik van gedramatiseerde voorlichtingsfilms in het toezicht. Tijdschrift voor Toezicht 2013(3): 43–66.

55.

Van Erp

(2018) Anti-cartel thrillers as a new film genre: How regulator-produced films portray and problematize cartels and communicate deterrence. Crime, Media, Culture 14(2): 229–246.

56.

Van Wingerde

(2012) De afschrikking voorbij: Een empirische studie naar afschrikking, generale preventie en regelnaleving in de Nederlandse afvalbranche. Nijmegen: Wolf Legal Publishers.

57.

Van de Bunt

Huisman

(2007) Organizational crime in the Netherlands. Crime and Justice 35(1): 217–260.

58.

Whelan

(2013) Cartel criminalization and the challenge of ‘moral wrongfulness’. Oxford Journal of Legal Studies 33(3): 535–561.

59.

Wils

WPJ

(2007) Leniency in antitrust enforcement: Theory and practice. World Competition 30: 25–64.

60.

Wils

(2009) The relationship between public antitrust enforcement and private actions for damages. World Competition 32: 3.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.09 MB

0.00 MB