Toxic Corporate Culture: Assessing Organizational Processes of Deviancy

Abstract

There is widespread recognition that organizational culture matters in corporations involved in systemic crime and wrongdoing. However, we know far less about how to assess and alter toxic elements within a corporate culture. The present paper draws on management science, anthropology, sociology of law, criminology, and social psychology to explain what organizational culture is and how it can sustain illegal and harmful corporate behavior. Through analyzing the corporate cultures at BP, Volkswagen, and Wells Fargo, this paper demonstrates that organizational toxicity does not just exist when corporate norms are directly opposed to legal norms, but also when: (a) it condones, neutralizes, or enables rule breaking; (b) it disables and obstructs compliance; and (c) actual practices contrast expressed compliant values. The paper concludes that detoxing corporate culture requires more than changing leadership or incentive structures. In particular, it requires addressing the structures, values, and practices that enable violations and obstruct compliance within an organization, as well as moving away from a singular focus on liability management (i.e., assigning blame and punishment) to an approach that prioritizes promoting transparency, honesty, and a responsibility to initiate and sustain actual cultural change.

1. Introduction

Following major corporate scandals, most of the attention focuses on assigning individual liability to the highest possible executive. Certainly, this practice is important as impunity should not be permitted to continue. However, to reduce corporate crime and misconduct, we must look beyond the role of high-level executives or a few rogue employees to focus on the organizational traits that stimulate rule-breaking and harmful behavior. This paper outlines how we can assess and address such toxic elements in corporate cultures. It does so by analyzing three major cases of corporate wrongdoing: BP, Volkswagen, and Wells Fargo. These recent scandals demonstrate that addressing organizational traits is vital.

On 20 April 2010, an explosion and fire at the Deepwater Horizon drilling rig resulted in a massive crude oil leak in the Gulf of Mexico. On that day, BP, the British petroleum company that operated the rig, became responsible for the largest oil spill in U.S. history (PBS 2010). In addition to causing tremendous economic, ecological, and health effects around the Gulf’s coast, eleven employees died.

This was not the first time BP operations had caused a major disaster in the U.S. On 23 March 2005, an explosion at a BP refinery in Texas City killed 15 people and injured 170 more. And in March 2006, BP caused the largest spill on Alaska’s North Slope, leaking 267,000 gallons of crude oil on the freezing tundra. That particular spill went undetected for five days (Barringer 2006).

In all three disasters, the accidents occurred because BP had failed to perform sufficient maintenance, had constantly cut costs, had reduced its engineering capacity, and had not responded to employee complaints about safety hazards (Steffy 2010). These incidents demonstrate that well before the massive Deepwater Horizon spill in 2010, BP had continuously suffered from safety issues and had repeatedly received information about hazards. Yet incident after incident, complaint after complaint, report after report, and promise after promise, it continued to seek ways to cut costs instead of prioritizing safety and compliance. The issues at BP were not just the product of individual decisions, but rather of broader patterns within the company. As such, these problems were endemic to the culture at BP. Indeed, William Reilly, the co-chair of the presidential investigation commission on the Gulf of Mexico oil disaster concluded that BP had been operating under a “culture of complacency” (Goldenberg 2010).

BP is not unique. In 2014, American researchers started to suspect that German carmaker Volkswagen (VW) was using a device that would lower its vehicles’ emissions specifically during laboratory testing, while the nitrogen oxides (NOx) emissions under actual driving conditions would be about 40 times higher (Ewing 2017). Investigations uncovered that VW had actually installed this cheat device in over 11 million vehicles. After a West Virginia University research report first discovered the discrepancy between laboratory and real driving emissions, and California regulators later confirmed this, VW stalled the investigations, questioned the investigative methods, and even modified the cheat device to make it even more effective. Under pressure from the California Air Resource Board (CARB), which threatened to block VW from selling cars in California in 2016, the company admitted its cheating (Ewing 2017).

Just like at BP, the corporate wrongdoing in the VW case was not just the work of a single—or even a few—“bad apples.” Rather, it appears to have been embedded within the company for quite some time. As Eric Schneiderman, the Attorney General of New York at the time, concluded, “Hundreds of very high-level executives and engineers knew about this. We did not find one email saying that maybe we should not be doing this, or this is against the law or put the breaks on this system. So this was a corporate culture permeated by fraud.”1 VW had first been caught using a cheating device in 1973, when it settled a case for $1,200,000 with the U.S. Environmental Protection Agency (EPA) (Ewing 2017).

Moreover, diesel emission violations have been widespread in the industry, with 97% of cars failing emissions tests according to a 2016 report (Carrington et al. 2016). And there had been major fines against car companies for using such defeat devices earlier as well. In 1998, major diesel truck manufacturers including Caterpillar Renault and Volvo came to a $1 billion settlement with the DOJ and EPA for similar violations (Department of Justice 1998).

These types of corporate misconduct are not even limited to the automotive and oil industries. Between 2009 and 2016, Wells Fargo, the California-based U.S. bank, had fraudulently opened 3.5 million accounts without authorization from their customers (Cowley 2017). Moreover, the bank had enrolled 528,000 customers for online bill payment services without their authorization. Recent evaluations suggest that the fraudulent practices had actually been occurring for about 15 years (Independent Directors of the Board of Wells Fargo 2017). However, the offenses had for a long time been treated as individual offenses of local bank employees, who if found to have created unauthorized or fake accounts would simply be fired (Colvin 2017). Yet, these sales practices had become endemic in the bank, occurring across its local branches and even in its San Francisco headquarters.

When the scandal broke in 2016, Wells Fargo blamed and fired 5300 local employees found to be involved in the practices. However, soon it became clear that these were not simply the actions of bad individuals in a particular locale, but rather the result of corporate incentives within the firm, pushing employees to continually increase their sales, and sell customers as many products as possible. In a 2017 report of the Independent Directors of the Board of Wells Fargo, the bank itself came to admit that what happened was an organizational, not just an individual problem: “The root cause of sales practice failures was the distortion of the Community Bank’s sales culture and performance management system, which when combined with aggressive sales management, created pressure on employees to sell unwanted or unneeded products to customers and, in some cases, to open unauthorized accounts” (Independent Directors of the Board of Wells Fargo 2017).

These three cases are intended to illustrate major, recent corporate wrongdoing. In all three cases, the conduct was not merely attributable to particular executives or particular employees. In fact, broader organizational traits initiated, stimulated, and sustained the wrongdoing and offending behaviors. White collar criminologists have long known that corporate crime is not just the product of a few bad individual corporate criminals, but is also embedded in so-called “criminogenic” traits of the corporate organization (Sutherland 1940; Needleman and Needleman 1979; Apel and Paternoster 2009; Clinard 1983; Clinard and Yeager 1980).

Recently, company boards have expressed that they must address toxic elements in their cultures. As the Wall Street Journal reported in October 2017, companies, including Whirlpool, Citigroup, and CACI International, recently formed board culture committees (Lubin 2017). The U.S. National Association of Corporate Directors Blue Ribbon Commission on Culture as a Corporate Asset concluded that boards should create formal oversight of company culture (National Association of Corporate Directors 2017). Companies are also hiring culture experts who have provided them with survey tools to evaluate negative elements in their corporate culture (Lubin 2017). Wells Fargo, for instance, now regularly assesses the so-called “happy to grumpy ratio” of its employees, based on the idea that when more people are happy, unethical behavior is less likely to occur (Kellaway 2015).

Consequently, there appears to be a cultural moment in addressing corporate crime and wrongdoing (Glazer and Rexrode 2017). With all the recent scandals, major corporations and banks now at least have started to discuss culture. It seems that they can no longer get away with just deflecting blame. It is no longer sufficient to just focus on bad apples, but bad barrels need to be addressed (cf. Scholten and Ellemers 2016; Pertiwi 2018). Of course, it is not entirely clear whether there is a true commitment to address the culture, or whether firms just discuss this to move beyond yet another crisis and reduce their future liability. However, for the sake of reducing wrongdoing, it is clear that culture matters, and it is clear that it must be addressed. What is less well understood is exactly what organizational culture is or what makes a culture toxic or come to support wrongdoing and rule breaking. Corporations do not clearly know how to measure or assess their own culture and toxic elements within it, let alone how to change and detox it (Glazer and Rexrode 2017).

Presently, it is widely agreed that changing corporate wrongdoing through external enforcement is challenging. Indeed, a recent systemic review of all available studies on corporate deterrence found that: “The evidence fails to show a consistent deterrent effect of punitive sanctions on individual offending, company level of fending, geographic-level offending, or offending among studies” (Simpson et al. 2014). It must be acknowledged that this is not just another study in the literature. This study is the most comprehensive, rigorous, and up-to-date review of all available scientific evidence about deterrence for corporate crime and misconduct across a range of corporate crimes, and it failed to find a deterrent effect.

There are several explanations for why corporate deterrence is difficult. A first reason is that just like for individual crime (Nagin 2013), certainty of punishment likely matters more for a company than the purported severity of the punishment. Further, stronger sanctions only have an effect if a tipping point of certainty is achieved (Brown 1978; Chamlin 1991). However, because of the complexity of corporate organizations and processes, detecting corporate violations in the first place is quite challenging (Gray and Silbey 2014; Pontell et al. 1994; Gray and Mendeloff 2005; Gray and Scholz 1991; Plambeck and Taylor 2015; Henriques 2011). In fact, stronger punishment threats can actually lead to more investment to prevent getting caught, resulting in a cat and mouse game (cf. Plambeck and Taylor 2015). For instance, when VW discovered that California and US regulators knew about their cheating, their initial response was to improve the software and hide their cheating through a recall (Ewing 2017).

A second reason is that deterrence is subjective (cf. Apel 2013). Oftentimes, corporate executives who are supposed to be deterred simply are not aware of the certainty and severity of punishment (Thornton et al. 2005). Or for some companies, the expected penalties are seen as the price of business and made part of the budget. VW high level engineers and executives had since 2006 looked at penalties in other cases. But as New York Attorney General Eric Schneiderman concludes: “They had concluded we can survive this type of penalty.”2

Further, when wrongdoing is actually detected, it remains difficult to prosecute. For example, presenting complex evidence of corporate operations to lay-juries (in the U.S. context at least) is very challenging (Pontell et al. 1994). And even when it is successfully prosecuted, penalty fines often remain uncollected (Ross and Pritikin 2010). Of course, even when law enforcement would be successful to deter corporations from breaking the law, such corporations would still have to address organizational traits that had been responsible for rule-breaking and harmful behavior. Thus, corporate culture is key.

The chief aim of this paper is to show how existing scholarship can aid corporations that are willing to assess and detox their cultures. To do so, the present paper reviews existing social, behavioral, and management science literature on these questions. In addition, it illustrates their insights by looking deeper into the three cases of corporate misconduct at BP, VW, and Wells Fargo. Accordingly, this paper should be viewed not as an empirical investigation of theories through analyzing quantitative data, but rather as an illustration of how existing theories about offending and wrongdoing can be used in tandem to both assess and change toxic elements in corporate cultures.

We should note that this paper takes a broad approach to examining toxic corporate cultures. It does not merely view toxicity as violating the laws or regulations. Very often, the law provides vague and conflicting direction to organizations, and when companies and their compliance managers and lawyers interact with legal norms and regulatory agencies, these legal rules get re-interpreted (Talesh 2015, 2009; Edelman et al. 1991; Edelman and Talesh 2011; Lange 1999). In addition, with new forms of regulation, so-called process-oriented rules (Gilad 2010) and management-based regulation (Coglianese and Lazer 2003), the law often does not provide direct behavioral guidance. Especially in industries with complex technical procedures, industries themselves develop the actual standards for behavior (Mills and Koliba 2015). Thus, in this paper, toxic behavior includes both behavior that is clearly illegal, as well as corporate behavior that a reasonable person would see as clearly damaging or promotive of misbehavior.

The paper focuses on organizations with toxic cultures. These are organizations with long-term and systemic rule breaking and damaging behavior. The focus on toxic culture is different from existing studies. Toxic culture is not the same as a criminogenic organizational culture or a criminogenic organization (Sutherland 1940; Clinard 1983; Clinard and Yeager 1980). Organizations with toxic cultures do not operate to engage in crime and break the law. Toxic culture is negative and therefore different from studies that have focused on safety culture (Silbey 2009) or compliance culture (Interligi 2010).

The paper first outlines what organizational culture is, how it can become deviant, and how it can create and sustain rule-breaking behavior. It highlights how to assess elements of a toxic corporate culture through a framework for forensic ethnography. Then, it uses the forensic ethnography framework to analyze deviant elements of corporate culture at BP, VW, and Wells Fargo. Based on these analyses of the root cultural causes for corporate misconduct, the paper concludes by discussing their implications for addressing corporate wrongdoing and changing toxic corporate cultures.

2. What Is Corporate Culture and How Does It Become Toxic?

2.1. Tangible and Intangible Corporate Culture

Cultural anthropology has long been the most important field of study to seriously ponder what culture is. Whereas earlier anthropologists defined culture in observable expressions such as art, architecture, lineage structures, and customs, since the 1970s, the leading definitions stress immaterial and less tangible aspects that are harder to capture and observe (Eriksen 2001). Most influential has been Clifford Geertz’ approach (Geertz 1973). For Geertz, culture is not some essential, observable, material fact, or something that can be captured in an essential static way, as an unchanging trait of a certain society. Instead, culture is all about interpretation. Culture consists of shared interpretations within a social group (Geertz 1973).

Insights about organizational culture show a similar understanding of the complexity of both tangible and intangible interpretative elements of such organizational culture. While some commentators of the corporate scandals talk about cultural change solely by pointing to tangible aspects (e.g., changing the incentive structures; changing the leadership), there is now widespread recognition amongst management and organizational scientists that organizational culture exists at both tangible and intangible levels. Parker and Nielsen, in their review of compliance management programs, explain that organizational culture can refer to “shared values and beliefs, myths, interpretations and meanings within an organization, and actions and behaviors, including customs, practices, norms, rituals, and implementation of control systems” (Schein 2010). Certainly, this approach to culture is thus very much in line with cultural anthropological explanations (e.g., Geertz 1973).

The most widely-cited approach to organizational culture has been developed by Edgar Schein (2010). According to his approach, organizational culture exists at three levels. The first consists of what he calls “artifacts.” These are the tangible aspects of an organization’s culture and consist both of the “visible and feelable structures and processes” as well as of “observed behavior” (Schein 2010, p. 23). Artifacts are at the surface of an organization’s culture, they are how the organization manifests its culture. They are the visible products of an organization that may include its physical environment and architecture, its technology, its creations, its style, and its stories and myths. Moreover, artifacts also include the published documents that cover the values, operations, rituals, and organizational charts (Schein 2010). The second level are what Schein calls “the espoused beliefs and values,” which encompass the shared ideals, goals, values, and aspirations between individuals within an organization. These can come in the form of shared ideologies and also rationalizations for what the organization does. These espoused beliefs and values are deeper in the organization and cannot be directly observed, but must be learned by talking to organizational members (Schein 2010). The third and deepest level of an organizational culture are what Schein calls the “basic underlying assumptions.” These are most deeply embedded within the organization and operate unconsciously through organization members. They concern taken-for-granted beliefs and values that can “determine behavior perception, thought and feeling” (Schein 2010, p. 24).

There is a vital interaction between these three levels of Schein’s approach to organizational culture. The meaning and day-to-day influence of the surface level artifacts, including written rules, procedures, and evaluation standards, is embedded and constructed within the deeper organizational cultural fabric of the beliefs, values, and underlying assumptions (Schein 2010, p. 25).

2.2. Analyzing Toxicity in Corporate Culture

As Schein’s influential model of organizational culture suggests, any assessment of organizational culture includes both tangible and intangible elements. Schein, and most management science organizational culture specialists, do not provide much information about a toxic organizational culture that induces legal rule breaking. To understand how organizational culture can induce and sustain offending and harmful behavior and to trace the toxic elements in the corporate culture that are responsible for the wrongdoing, we shall draw on insights from sociology, anthropology, psychology, and criminology.

A first insight from sociological and anthropological accounts of law is that organizations and communities may come to form their own norms and these norms can come to oppose and resist those of the law, thus sustaining rule breaking. Sally Falk Moore showed in her seminal 1973 paper that very different communities (e.g., the garment industry in New York City; Tanzanian coffee farmers) both formed what she called semi-autonomous social fields. These are social fields (communities) that are open to the influence of the law, yet also form their own norms and rules that can come to resist those of the law (Moore 1973). As she explained: “A court or legislature can make custom law. A semi-autonomous social field can make law its custom” (Moore 1973, p. 744).

Similarly, Carol Heimer has found that legal norms can come to compete with norms of different communities they have to interact with. She found this in her study of neonatal intensive care units, where legally mandated institutions in hospitals tasked to implement the law had to compete with medical norms of the hospital staff, and familial norms of the families of children treated (Heimer 1999). She found that hospital staff would adopt in earnest only legal norms that were directly useful to them, and would ignore or resist other legally mandated procedures and practices. Instead of the law shaping hospital practices, she found the hospital routine shaped the law, as legal procedures had to adapt to the practices, interests, and even time-table of the hospital staff (Heimer 1999).

Heimer’s findings demonstrate how organizations can develop norms that can successfully resist and compete with the law (for this point see also (Clinard and Yeager 1980, p. 58)). However, that raises the question of how exactly this occurs. Vaughan’s study of the fatal NASA decision to launch the Space Shuttle Challenger in 1986 can shed light on this question. Her study focused on why NASA had gone ahead with the launch of the Challenger shuttle, even though there had been repeated evidence going back to 1977 of a particular technical risk with the so-called “O-Rings” that were “designed to seal a tiny gap created by pressure at ignition in the joints of the Solid Rocket Booster” (Vaughan 1997, p. xi). Even on the eve of the launch, several concerned NASA engineers had argued against the launch as they feared that these O-Rings might threaten flight safety (Vaughan 1989, p. 331). Vaughan has argued that NASA’s Solid Rocket Booster group had “normalized deviance” (Vaughan 1997, pp. 62–63). In other words, the Solid Rocket Booster Group had developed an organizational norm that could successfully oppose safety norms.

What is vital about her study is that it shows how such organizational normalization of deviance developed over a process. As Vaughan has shown, the normalization occurred in several stages. First, came the warning signals, starting all the way back in 1977, that there may be a technical deviation. These signals were then reinterpreted and finally labeled “an acceptable risk” (Vaughan 1997, p. 65). Thus, deviance became the standard. And this in turn became “a collectively constructed cultural reality, incorporated into the worldview of the group.” (Vaughan 1997, pp. 65–66). The culturally embedded norm in the group became that “risk had to be renegotiated“ (Vaughan 1997, p. 66). And this did not just play out on the eve of the launch, it was a repeated process that had before already become deeply engrained within the Solid Rocket Booster Group.

Vaughan’s research illuminates processes within an organization that can help to develop norms that run against those of the law and can become embedded in the organization’s values. Social psychology offers further insights about such social norms and how they can come to sustain illegal behavior. And although these psychological insights have been studied in the interactions and behaviors of individuals and in rather simple situations, they provide important insights relevant for understanding the interactive processes between social and legal-type norms. It should be noted that the insights discussed here come from field experiments in which real behavior outside of the lab was observed and analyzed.

The psychological study of social norms has demonstrated that social norms are very strong influences on individual behavior. Psychologists distinguish two kinds of social norms. First are injunctive social norms—these concern what people think others think they should do. Second are descriptive social norms, which consist of the behavior of others (Cialdini et al. 2006). A first insight from the psychological study of social norms is that human responses to others do not just flow through values, as Schein and the socio-legal approaches previously discussed focus on, but also through behavioral practices (Cialdini 2003, 2007; Cialdini and Goldstein 2004). A second insight is that social norms are not necessarily conscious: people respond to the behavior of others even when they are not aware that they are doing so (Cialdini 2003, 2007; Cialdini and Goldstein 2004). Social norms can become embedded in the social environment, creating automated situational norms where people subconsciously respond to cues in their environment that activate social norms (Aarts and Dijksterhuis 2003; Aarts et al. 2003).

Social psychologists have identified the processes through which social norms and legal norms interact, offering insights into how toxic norms may spread in organizational environments. First, when a negative social norm exists that is opposed to the legal norm—for instance, if most people steal valuable wood in a forest where this is not allowed—people will be more likely to break such legal rule (Cialdini and Goldstein 2004). In corporate settings, Baucus and Neal have found that firms with a longer history of violating the law (and thus firms who have developed social norms that violate legal norms) are more likely to violate again (Baucus and Near 1991). Simpson and Koper, similarly found that past illegal involvement predicted future offending (Simpson and Koper 1997). It is vital to prevent such negative social norms from being triggered or strengthened. Here, public messages—even messages by law enforcement—matter. Any message that stresses the scale of rule violation (even if it is to show how many people are punished) must be avoided lest more rule breaking will result (Cialdini 2007). And when there is a negative social norm with regard to one type of law (for instance when there is illegal graffiti) there is a danger of contagion to also create more rule violation for another legal norm (enhancing theft or littering) (Keizer et al. 2008).

A second situation occurs when a social norm supports compliance with the law (Cialdini 2007). When most people wear their seatbelt, others are more likely to comply with seatbelt laws, even when there is little enforcement. In this situation, it is vital to prevent legal interventions that erode these positive social norms. When punishment is introduced in an environment with strong positive social norms, this can send a signal that compliance is only necessary because of the punishment, not because of its intrinsic value. Consequently, in this way, punishment can erode positive social norms (Gneezy et al. 2011; Gneezy and Rustichini 2000).

A third situation occurs when a discrepancy exists between an injunctive social norm and a descriptive social norm. In other words, when what people perceive others to be telling them they should do is different from what others are actually doing. Keizer and colleagues have carried out a series of field experiments, where they looked at what would happen when they placed a sign clearly indicating what people should do (i.e., not litter, or not park their bike) in an environment where this behavior was common and where it was not common, also comparing this to when no sign had been there in both situations. Their findings indicate that compliance is best when there are no existing violations (when there is a positive descriptive social norm) and when there is a prohibition sign. They are second best when there is no sign and no existing violations. They are worse when there are existing violations, but compliance is the worst when there are existing violations and a sign forbidding such behavior. In other words, when there are actually negative descriptive social norms, the practice of emphasizing positive injunctive social norms can backfire (Keizer et al. 2011).

Organizational psychologists and management scientists have looked at what fosters an unethical climate that can stimulate rule breaking and immoral conduct. As Scholten and Ellemers (2016) have summarized the literature, three aspects are key here. First is the way organizations deal with errors and what sort of error management culture they have (cf. Van Dyck et al. 2005; Homsma et al. 2009). Organizations that promote ethical behavior recognize that errors are part of normal work and foster employees to acknowledge errors and learn from them. Organizations will foster unethical behavior if they respond in an ineffective way to errors. Such ineffective responses include denial that errors can and do occur, failing to act when they do occur and failing to assess what had caused the errors and make necessary changes to prevent future errors, or blame and punish employees, which may hamper successful learning and increase anxiety and stress and willingness to comply with organizational rules (Scholten and Ellemers 2016; Van Dyck et al. 2005; Homsma et al. 2009). Second, organizational psychologists and management scientists also have shown that organizations where small differences in performance can cause high differences in how employees are treated and what opportunities they get can breed more unethical behavior as employees feel treated unfairly and develop envy of colleagues (Cohen-Charash and Mueller 2007; Cohen-Charash and Spector 2001; Zoghbi-Manrique-de-Lara and Suárez-Acosta 2014; Scholten and Ellemers 2016). And third, finally, these scholars point to particular organizational traits that create so-called dysfunctional moral climates. These include organizations that do not focus on the moral implications of organizational activities and decisions (moral neglect), organizations where there is awareness of moral problems but no ability and opportunity to act against them (moral inaction), and organizations that reframe immoral actions to distort a proper understanding of moral content (moral justification) (Moore and Gino 2013; Scholten and Ellemers 2016).

There are also important criminological insights that apply to how organizational culture can come to stimulate corporate rule breaking and wrongdoing. Corporate crime criminologists have for a long time recognized that corporate culture matters. As Clinard and Yeager state: “ethical behavior […] is also the product of cultural norms operating within a given corporation or even industry that may be conducive to produce violations” (Clinard and Yeager 1980, p. 58). One strand of criminological theory finds that crime originates in distress, in so-called “strain.” Breaking rules or committing crime is a way for some people to cope with such strain. Two of the three core sources—strain originally identified in Agnew’s (1992) General Strain Theory—are the presentation of negative stimuli (physical or verbal assaults) and the inability to reach a desired result (Agnew 1992).

While strain originally pointed to the response to the strain of low socio-economic conditions, it can be applied to an organizational environment. If the environment puts stress on employees, employees may respond to the strain by violating laws (Agnew et al. 2009). Clinard and Yeager’s (1980) analysis of corporate crime points to the importance of economic performance, showing that violating firms are on average “less financially successful, [and] experience poorer growth (Clinard and Yeager 1980, p. 132). Baucus and Near’s model of illegal corporate behavior shows that corporations with a scarcity of resources are likely to break the law, but corporations with plentiful resources are even more likely to offend, while poor performance does not predict corporate offending (Baucus and Near 1991).

Criminologists have also shown how criminals are better able to break the law if they are able to “neutralize” the shame and guilt that comes with offending behavior. Sykes and Matza demonstrate several such neutralization techniques. These include: denying responsibility, denying injury, or reframing the victim into someone that deserves the harm (Sykes and Matza 1957). While these insights were originally developed for individual street crime, corporations can just as well develop organizational values, practices, and even structures that neutralize offending behavior (Stadler and Benson 2012; Gottschalk and Smith 2011; Benson 1985). Here, criminologists have recognized that other neutralization techniques like the defense of necessity (Minor 1981), the claim of normality (Benson 1985; see also Maruna and Copes 2005), and the metaphor of the ledger (where having done good acts balances out bad acts) (Klockars 1974; see also Piquero et al. 2005) play important roles in explaining corporate offending.

Criminologists have further focused on how the context in which a potential offender exists can create more or less opportunities for crime (Belknap 1987; Cohen and Felson 1979; Felson 1987; Osgood et al. 1996). These theories (including routine activity, situational crime prevention, and ecological criminology) all point to the fact that crime needs more than a motivated offender: it also requires such an offender to have access to an unguarded target, lacking a capable guardian. This is highly relevant for corporate offending and corporate culture. Some corporate cultures may produce organizational structures (rules and incentives), values, and practices that provide members with easier opportunities to break the law, for instance when external oversight is difficult (as the crimes are hidden inside the organization), when there is less internal auditing, when employees are left large discretion, or when rule breaking is normally condoned. An important context is the amount of dynamism and change that a corporation operates under, with the more changes in the market a corporation must adapt to the more likely the company will offend (Baucus and Near 1991). Here Baucus notes that a key issue is that corporations respond to dynamism by spreading out responsibility for decision making (Baucus and Near 1991). For similar points on the complexity of structures and corporate offending see (Clinard and Yeager 1980). Spreading decision making power reduces the ability of successful oversight while giving more people in the firm the power to make decisions that can result in offending and harmful behavior.

A vital insight from these three bodies of criminological literature is that organizational deviancy does not solely come from organizational structures, values, and practices that produce social norms that are opposed to the law, but can also originate from social norms that obstruct compliance (through strain), or that enable rule breaking (through neutralization and opportunity creation).

2.3. Levels, Aspects, Types of Toxicity and Processes

In sum, an assessment of toxic corporate cultures must take into account an organization’s tangible and intangible manifestations (Schein 2010), and in particular, the organizational structures (rules, authority lines, and incentives) (Schein 2010), organizational values (injunctive social norms), and organizational practices (descriptive social norms) (Cialdini 2007; Cialdini et al. 2006; Cialdini and Goldstein 2004). These form the core three organizational levels at which organizational culture can be assessed. They are entry points to do a cultural assessment, but are not wholly distinct, as structures can derive from and shape values and practices, and values can originate in structures and practices, and, of course, practices can follow the structures and values of an organization.

Table 1. Levels of Organizational Cultural Analysis.

Level		Aspects
1. Structures	Rules	Targets and Incentives	Hierarchy
2. Values	Explicit Shared Values	Injunctive Social Norms	Hidden Assumptions
3. Practices	Visible common behavior	Unaware Common Behavior	Situational Norms

outlines the three levels of organizational culture that a forensic ethnographer can assess to identify toxic elements.

To do a full forensic ethnography, one must then look at toxic elements at these three levels of an organizational culture. As the sociological, anthropological, psychological, management science, and criminological studies discussed above show, this process requires looking for norms at all three levels that are either directly opposed to the law, and thus have values or practices that undermine the law, or other law-type norms protecting against corporate harm (Moore 1973; Heimer 1999; Cialdini et al. 2006; Cialdini and Goldstein 2004). Thus, when common practices can come to normalize deviancy and damaging behavior (Vaughan 1989; Keizer et al. 2008; Scholten and Ellemers 2016), enable rule breaking through neutralization (Sykes and Matza 1957; Gottschalk and Smith 2011; Maruna and Copes 2005; Minor 1981; Siponen et al. 2012) or opportunity creation (Belknap 1987; Cohen and Felson 1979; Felson 1987; Osgood et al. 1996), obstruct rule following—through not providing sufficient informational or technical support to learn from errors (Scholten and Ellemers 2016; Homsma et al. 2009)—by creating strain and inducing negative rule breaking or harmful responses to such strain (Agnew 1992, 2001; Agnew et al. 2009; Simpson and Koper 1997; Scholten and Ellemers 2016; Homsma et al. 2009), toxic organizational norms and processes are achieved Finally, we must look at how the stated rules and values of the corporation are at odds with the practices and implied values, as this will stimulate extra rule offending as it delegitimizes compliance (when organizations stress values in line with the law that are commonly broken in practice) (Keizer et al. 2011).

Table 2. Toxic Organizational Norms and Processes.

Type of Toxic Norms	Processes
Directly Opposed	Resist and Compete with Legal Norms	Normalize Deviancy
Enabling Rule Breaking	Create Opportunity to Violate	Neutralize Offending
Obstructing Compliance	Lack of Support to Follow the Law and Learn from Errors	Strain Employees away from Compliance
Practice runs against Values	Delegitimize Positive Social and Legal Norms

below outlines these types of toxic norms and their processes.

3. Assessing Toxic Culture at BP, VW, and Wells Fargo

The goal of this paper is not to test the theories discussed above, nor to offer new empirical insights about which of these theories best predicts toxic corporate behavior. Rather, this paper illustrates how these ideas can be used to assess what aspects in organizations constitute a negative culture and how they sustain misbehavior. The remainder of this paper will analyze how the cultures in BP, VW, and Wells Fargo have come to support the deviant practices that culminated in the massive scandals introduced above. It will draw on extensive investigative journalism, court records, published internal reports, and academic literature available on all three cases to outline the core elements of deviancy in the organizational cultures that were at play. It will discuss key cultural elements these reports have found stimulate wrongdoing and offending behavior, and will analyze them in light of the social and behavioral science about toxic corporate culture. All three cases have been extensively reported, providing a wealth of information to draw from to analyze the toxic elements in their corporate cultures. For VW this paper relies heavily on the excellent reporting by New York Times journalist Jack Ewing published in his 2017 monograph, as it provides the most comprehensive discussion of all sources as well as balanced and critical reporting. Where necessary we have added other sources as updates or to provide a different perspective.3

The remainder of this section will assess the toxicity in the three corporate cultures. To structure the analysis of toxic elements in the cultures of these three organizations, the paper will focus on different relevant aspects of the business operations that are directly relevant for the illegal and harmful behavior these companies were engaging in. In each of these different aspects, it will offer a short sub-conclusion that highlights what toxic elements, referring back to Table 2, were at play at which levels of the culture (as outlined in Table 1 above).

First, this paper looks at how all three organizations developed their overall goals and strategies, and how this has been implemented in structures, values, and practices. Second, it explores how people in the corporations could respond to these goals and strategies and to what extent there was room for dissent and adaptation. Third, it analyses the extent to which there was illegal behavior and whether this behavior was condoned or disciplined internally. Fourth, it recounts what the companies did after the illegal behavior was discovered and made public, and to what extent they took responsibility or tried to deflect blame. And finally, it looks at how the companies’ messages over the years have compared to their practices in order to find out whether there was cognitive dissonance. After discussing these aspects separately, the final section examines the broader patterns of toxic cultural elements and explores what these patterns mean for attempts at cultural change.

3.1. Goals and Strategies

BP, VW, and Wells Fargo all started as underdogs with high ambitions. In the early 1990s, BP was struggling as oil prices had declined, many oil reserves in the Middle East had been nationalized by local countries, and the production costs of onshore drilling had hampered profits (Lustgarten 2012; Steffy 2010). At that time, Volkswagen had major difficulties in the U.S. market, where it had been unable to find a successor to the counterculture successes of the Beetle and the Transporter Van in the 1960s. Further, it struggled with the image of being unreliable and requiring frequent repair, had high labor and production costs, and barely broke even financially in 1992 (Ewing 2017). Wells Fargo was a local bank operating largely in California (Colvin 2017).

Yet, all three developed grand ambitions as new CEOs came to power. When John Browne became CEO of BP in 1995, he saw that BP could close its gap with the largest petroleum giants by rigorously focusing on divesting on-shore operations with low profits, by cutting costs, and by doing new oil exploration for major off-shore fields (so-called “Elephants”) in areas with higher risk but also higher profits (Lustgarten 2012). When Ferdinand Piëch, grandson of Ferdinand Porsche who had designed the original Beetle for Hitler, took over as CEO of VW, he soon declared that by 2018 the German carmaker should become the largest in the world (Ewing 2017). And when Richard Kovacevich became CEO of Wells Fargo, after it had been acquired by Minneapolis-based Norwest in 1998 that opted to drop its old name, he adopted the “Go for Gr-Eight” motto, seeking to guide the bank to the national and global top by outselling competitors and selling eight products per customer—four times the average rate for banks (Colvin 2017).

In each of these cases, high ambitions focused on growth and profit increase, and with risk-prone means to achieve the goals. Wells Fargo had to go where no bank had gone before—to achieve fast growth, and reach the highly ambitious sales target, it had to somehow convince its clients not to buy two of its products (e.g., credit cards, insurance, special accounts), but eight.

At BP, the risk came as the new strategy required two opposed interventions: cutting costs and laying off a large part of the engineering expertise on the one hand, and developing new oil exploration in high risk areas that required extra engineering resources (Steffy 2010). At VW, the mission to become number one could only be achieved by becoming competitive in the number one car market, the U.S. When Martin Winterkorn became CEO, he wanted to increase car production from six million to ten million in the next ten years, surpassing both GM and Toyota, which would mean that VW would finally have to succeed in the difficult U.S. car market.

To do so, Winterkorn bet on clean diesel and a new diesel engine: the EA 189. The problem was that the company had been unable to develop an engine that was actually economical, practical, and clean and could compete in the U.S. market. VW’s option—to cut the toxic NOx emissions that the higher temperature burning diesel engines produce—came with its own problems of higher price, occupying precious storage space, and requiring more frequent maintenance, which would all prohibit a successful strategy (Ewing 2017). Thus, VW set itself a target that may well have been impossible to achieve.

It was not long before the new ambitions and values they represented turned into practices and structures that produced risks of wrongdoing and damages. At BP, this was most evident in cost-saving reforms. After John Browne took the helm, he started to cut costs aggressively, with across-the-board cuts of 25%, first in 1999 and then again in 2004. This affected not just material costs, but also personnel, resulting in forced lay-offs, thus shrinking BP’s pool of engineering talent (Steffy 2010). It was not long until these general cuts started to affect the safety at BP operations. In the years prior to the 2005 Texas City explosion, BP had decided not to replace outdated equipment that later caused the massive accident. BP simply wanted to save on the $150,000 investment that was needed. One employee, as reported by PBS, wrote: “We need to decide if we want to invest $150,000 now to save money later on” (PBS 2010). And a senior manager wrote “that capital expenditure is ‘very tight. Bank the $150,000 in savings right now’” (PBS 2010). It is no wonder that former Secretary of State James Baker, who headed one of the two investigations into the Texas City explosion, concluded that “BP has not adequately embraced safety as a core value” (Baker et al. 2007). In the 2006 Alaskan spill, the leak occurred in a corroded pipe, “which hadn’t been cleaned in over a decade” (PBS 2010). In Alaska, BP had for years, as a subsequent investigation found, sacrificed proper maintenance for profit. Investigators found that “‘unacceptable’ maintenance backlogs ballooned as BP tried to sustain profits…even though production was declining” (Lustgarten and Knutson 2010). The investigation clearly concluded that these issues are at the heart of BP’s value system: “There is a disconnect between…management’s stated commitment to safety and the perception of that commitment” (Lustgarten and Knutson 2010). And preceding the Gulf spill in 2010, BP managers “were shaving maintenance costs with the practice of ‘run to failure,’ under which aging equipment was used as long as possible” (Lustgarten and Knutson 2010). In sum, BP in its day-to-day operations valued profits over anything else, even at the repeated sacrifice of safety and the environment.

The overall targets of cost reduction played a crucial role in BP safety issues, as the Chemical Safety and Hazard Investigation Board (CSB) concluded in its 341 page report after the 2005 report in response to the Texas explosion that had killed 15 (U.S. Chemical Safety and Hazard Investigation Board 2007). As a supervisor told lower-level managers who had questioned the 2004 budget cuts and their implications for ensuring operations: “Which bit of 25 percent do you not understand?” (Steffy 2010, p. 117). Another manager explained how this forced them towards risky and illegal practices: “The focus on controlling costs was acute at BP, to the point it became a distraction. They just go after it with a ferocity that’s mind-numbing and terrifying. No one’s ever asked to cut corners or take a risk, but often it ends up like that” (Steffy 2010, p. 58).

At VW, Winterkorn’s ambitious target, to sell ten million cars and focus on clean diesel, put engineers in a bind. They could develop a clean diesel engine. To do so, they could use technology from Daimler, called BlueTec, that sprayed a chemical substance called urea into the exhaust to help breakdown NOx emissions (Ewing 2017). However, it came with an extra cost of about $350 dollars for each car, and also required installing an extra tank that would take up cargo space and require owners to do frequent refills.

The other option was a so-called “lean NOx trap” that separated nitrogen oxide molecules into harmless oxygen and diatomic nitrogen. This technique was cheaper and did not need an extra tank, but it required an exhaust gas recirculation system that produced more carcinogenic fine particle emissions, and also caused the soot filter to wear out faster (Ewing 2017). Considering U.S. law obliged car makers to have emissions control systems effective for the entire lifespan of the vehicle, the soot filter wear was a core problem.

Nonetheless, VW’s highly ambitious targets played such a strong role that by late 2007, engine specialists had never seriously considered adjusting them to the engineering realties they faced. Rather than give up on the targets, VW engineers were forced to look for alternative solutions. They not only adopted the lean NOx trap, but also installed the software “defeat device” that would only switch on the emissions control system during emissions testing, so that during normal driving the soot filter would remain intact longer (Ewing 2017). Because of VW’s cost reduction efforts to make as many models as possible share the same parts, the decision to install cheating devices on the EA 189 engine came to affect over ten million cars VW produced and sold.

At Wells Fargo, the push for growth through the extremely high sales targets was to be achieved through bank employee incentive systems. The bank’s branches started to set product sales goals for employees to meet. This put tremendous pressure on Wells Fargo bankers to sell more products to their clients. Employees would be under constant scrutiny with daily and monthly “Motivator” reports tracking their sales volumes (Frost 2017). If they failed to meet targets, employees would undergo “coaching sessions.” One employee explained later to National Public Radio (NPR) journalists that these sessions were not there to support the workers but just to pressure them to sell more (Arnold 2016). Another employee explained how, when she failed to meet her target, two managers would lecture her at her desk and then perp-walk her while colleagues were watching. As she explained: “It’s like being called into the principal’s office. Sit down at the large conference table, no windows in this room, they shut the door, lock the door” (Arnold 2016). After that she was forced to sign a “formal warning” and was warned: “If you don’t meet your solutions you’re not a team player. If you’re bringing down the team then you will be fired and it will be on your permanent record” (Arnold 2016). She was simply afraid to lose her job and fearful not to get another one with the bad state of the economy. This employee said that things got so bad that she vomited under her desk.

Another employee compared his Wells Fargo job to “being in an abusive relationship” (Arnold 2016). Things were worst during special sales campaigns, such as the “Jump into January” campaign that sought to start the year with strong sales. During such campaigns, employees were to reach even higher targets than usual, sometimes even up to twenty per customer. In one local branch office, employees were forced to “run the gauntlet” by running past costumed district managers to write their sales numbers on a white board (Frost 2017). It became increasingly clear that employees had started to resort to creating unauthorized or fake accounts in response to these extreme pressures to boost their sales targets and achieve the overall goals set in the company. Yet, for a long time no change was made to the targets themselves or the pressures through which they were implemented. As one report found, Wells Fargo “was hesitant to end the program because (Carrie) Tolstedt (the head of community banking at the time) was ‘scared to death’ that it could hurt sales figures for the entire year” (Frost 2017).

And the strategy worked. In 1999, BP, after cutting costs and taking over several competitors, quadrupled in value, and finally caught up with the top oil companies. Its stocks soared, and its CEO, Browne, was dubbed “Sun King” in British newspapers. VW finally became successful in the U.S. market with its clean diesel, and eventually even after the emission scandal was discovered became the largest car company in the world in 2015, three years ahead of the planned schedule (Ewing 2017, p. 187). And Wells Fargo jumped from the ninth most valuable bank in America, to the most valuable one in the world in 2015. Wells Fargo achieved 18 consecutive quarters of over $5 billion in profits—a feat it shares only with Apple (Colvin 2017).

In sum, this analysis of how these companies set their targets and responded to challenges in the business environment demonstrates how toxic elements came to exist in these three cases. The most important type of toxic norm, from those discussed in Table 2 earlier, is strain (Agnew 1992, 2001; Agnew et al. 2009; Simpson and Koper 1997). What we see here is that the companies themselves were under strain of the shareholder’s expectations of growth and revenue formation. New leaders responded to these pressures by setting highly ambitious goals that because of their high risk and low feasibility nature brought the external strain into the company’s operations and ultimately to the employees, which in crucial instances forced them to make or go along with decisions that were damaging, and at worst illegal. These negative responses to strain and its resultant negative influences came at the cultural level of structures (see Table 1), as it was laid down in targets and incentives (cf. Schein 2010), but soon moved deeper into explicit shared values and visible common behavior (see Table 1).

3.2. Management and Employee Responses to Goals and Strategies

So why did BP employees, VW employees, and Wells Fargo employees go along with these goals? Why did they not successfully resist or change these goals or the practices they produced, even though they must have seen they were not realistic or risk free? To answer these questions, we must look at the corporate structures in the companies and how they shaped internal communication as well as the way responsibility for targets and work was shared.

BP and Wells Fargo have hierarchies that are strikingly different from Volkswagen. VW had a highly centralized structure with strong power vested in the CEO at the top. There was a centralized authority in decision-making that forced decisions to go up the chain of command. As one former manager trainee described it: “VW was like North Korea without the labor camps. You have to obey everyone” (Ewing 2017, p. 93). VW’s CEO played a central and direct role in day-to-day decisions, with first Piëch, and later his successor Winterkorn, as engineers who micro-managed engineering decisions all the way to the interior color of cars promoted at auto shows.

By contrast, Wells Fargo and BP had highly decentralized structures. Under Browne’s leadership, BP established business units that were to operate at autonomous companies within the overall BP structure. Each had its own targets, its own profit reporting, and its own leadership. At some point BP had over 300 such decentralized units. As Steffy argues, this created an unwieldy structure, with each unit leader caring for his own goals without focusing on how these affected the whole (Steffy 2010). BP also often used sub-contractors and outsourced key aspects of its work, further delegating responsibilities, but now outside of its own employees (Bozeman 2011). Wells Fargo was very similar to BP in that its operation was split amongst business units. Wells Fargo’s CEO Kovacevich nicely illustrated the structure when he called himself a “CEO of CEOs” (Colvin 2017).

While the three companies had such different hierarchical structures, in all three there was limited possibility to resist top-down targets and have effective dissent. Comparing the cases reveals that the overall hierarchy itself was not the core issue here, but rather how information flowed across the structure from bottom to top and top to bottom. Clearly, VW’s highly centralized hierarchy meant lower level employees had trouble getting their information heard at higher levels. Thus, at VW it was hard for lower level employees to correct faulty central level decision making and targets.

However, BP and Wells Fargo’s decentralized structures did not allow for much better information flow that might have corrected unrealistic budget cuts and sales targets. At BP for instance, in 1999, a group of 77 workers at the Alaskan operations—where later a major spill would occur—wrote a desperate letter to CEO John Browne trying to sway him from the intended cuts as they feared it would further undermine the already appalling safety conditions (Lustgarten 2012). The letter points to the difficulty of communicating critical information upwards: “Anything we say either stays at this level or gets filtered on the way up to a version of ‘can do sir’…Our feedback is ignored because it doesn’t support the preordained agenda…Your frontline management and supervision will continue to cut as long as you direct and sanction it, right up to the precipice of disaster and over” (Lustgarten 2012). John Browne did not reply and instead, a month later, announced another $4 billion budget cut.

Five years later, in 2004, the new BP Texas City Refinery Plant manager also tried to get attention from higher executives at the London office. He presented a detailed report entitled “Texas City is Not a Safe Place to Work,” about the horrible safety record at the plant, where over three decades 23 workers had died—one of the worst records in the industry. He also conducted a survey amongst workers that unearthed widespread safety concerns that had long gone unheard and unaddressed. The new manager used the report to ask for a budget increase to upgrade the safety at the plant. The main office denied his request and asked him to focus on the 25% budget cuts he had to meet for 2004, all the while the refinery was making $100 million a month for BP (Steffy 2010, p. 67).

Similarly, Wells Fargo senior executives refused to respond to challenges by regional leaders of the bank who had come forward to complain that the sales goals were too high and had become “increasingly untenable” (Frost 2017). What was at play here was not so much the structure itself, but the failure of higher level employees to adequately allow lower level employees to provide input and be heard. Targets were formulated in a top-down fashion and implemented while disregarding critique, regardless of the decentralized structure at Wells Fargo and VW.

In all three companies, the composition of the labor force and executive management practices made dissent difficult. At Wells Fargo the high strain of the job resulted in massive staff turnover, reaching up to 41% in one year (Colvin 2017). This left the bank with highly inexperienced employees, less likely to successfully raise concerns over the targets they had to meet. Through its mass lay-offs, BP similarly got rid of a large swatch of its senior engineers, who would have been in the best position to speak out against the safety hazards that were becoming increasingly apparent (Steffy 2010). BP CEO John Browne wanted a bench of followers, which he dubbed his “turtles,” referring to the Ninja Turtle cartoon (Steffy 2010, p. 58). And at VW, CEOs Piëch and later Winterkorn often fired executives they did not like, keeping only those who would agree with them and support their positions (Ewing 2017). At BP, meanwhile, frequent executive job rotations, which were a standard management practice, incentivized these lower level leaders to focus on short term targets while disregarding the longer term consequences, and thus made them more concerned with meeting the targets set by the company headquarters than enhancing long term safety (Lyall 2010).

Dissenting opinions were also suppressed through intimidating management practices. An investigative report following the 2010 Deep Water Horizon Spill found at BP “a pattern of intimidating workers who raised safety or environmental concerns” (Lustgarten and Knutson 2010). At Wells Fargo, as discussed above, daily intimidation practices were used to cajole and publicly shame employees to keep focused on their targets and the overall growth of the bank. Employees at all levels worked on the pressure of constant, sometimes hourly ranking of their sales rates in comparison with peers. When found to be lagging, they risked demotion or dismissal. As one employee recalled: “We were constantly told we would end up working for McDonald’s if we did not make the sales quotas…we had to stay for what felt like after-school detention, or report to a call session on Saturdays” (Reckard 2013). This clearly did not produce an atmosphere conducive to voicing critical opinions.

At VW, intimidation occurred at the highest levels. CEO Winterkorn was known for his Tuesday top executive meetings that included the highest-level officials in charge of major brands like Audi or Seat. At the meetings, with all present, Winterkorn would mercilessly criticize any executive that had failed to meet set targets. Ewing explains the humiliating tactics that Winterkorn used: “Managers who were favorites one week could suddenly fall from grace the next. Sometimes they learned they had been demoted or dismissed not from Winterkorn or a colleague but from reading about it in a German business publication, like Manager Magazine, that had somehow been tipped off” (Ewing 2017, p. 157). Similarly, at Wells Fargo the head of community banking, Tolstedt, an internal board review found, was “insular and defensive and did not like to be challenged or hear negative information. Even senior leaders within the Community Bank were frequently afraid of or discouraged from airing contrary views” (Colvin 2017).

Let us here also look at what we can draw out from these three cases about how toxic cultures form.

The most important insight here is that in all three cases a strong social norm (injunctive but also in the form of visible common (and probably unaware common) behavior) developed that dissent was not appreciated and that targets had to be met. A strong norm developed in all three companies not to resist or disagree with higher level targets and commands. This norm in and of itself is not directly opposed to the legal norms at play here. It does not support breaking safety standards, creating defeat devices, or fraudulently opening false or unauthorized bank accounts. This social norm here rather obstructs behavior that supports compliance. It makes it harder for employees or executives to come to speak out and resist practices that break the law. As such, it also undermines checks and balances within the company, especially over policies and practices of higher level leaders that come to break the law. In turn, this creates a larger opportunity to break the law (Belknap 1987; Cohen and Felson 1979; Felson 1987; Osgood et al. 1996), and when such rule breaking is allowed to occur without critique, it thus becomes condoned and even normalized (Vaughan 1989, 1997). As such employees here operated in an inactive moral climate, where there may have been recognition of unethical and immoral problems, but a very limited space to act on them (cf. Moore and Gino 2013; Scholten and Ellemers 2016).

In this context, the social norm against critique and dissent can work in tandem with generating further strain (Agnew 1992, 2001; Agnew et al. 2009; Simpson and Koper 1997). Specifically, employees and managers were under pressure due to the strong coercive processes as well as the fierce competition and job insecurity some had. As such, in light of Table 2’s framework to understand toxic organizational norms and processes, our analysis of employee and management participation in goals and targets shows that four toxic processes were at play in the organizational culture: obstruct behavior that supports compliance, strain away from compliance, normalization of deviancy, and creating opportunity to violate the law. Here, in light of Table 1’s outline of the levels and aspects of organizational cultural analysis, we see that these toxic processes developed first at the level of values, they became reinforced by the structures of hierarchy and incentives, and became deeply embedded as they developed into common practices (cf. Schein 2010).

3.3. Illegal Behavior and Internal Responses

By the time the scandals became public, all three companies had already engaged in the damaging and rule breaking behavior for years or even decades. BP had, from the 1990s onwards, developed an appalling safety record at its operations. Its operations were so bad that it had one of the worst safety records in the industry, paying one multimillion dollar fine and settlement after another to the EPA and OSHA, only to be found breaking the same safety standards again (Lustgarten 2012; Mattera 2016; Steffy 2010). Even after the major 2005 Texas Refinery explosion that had killed 15 workers and wounded 170, BP’s major safety problems continued. In the three years following the explosion, and after paying $20 million in fines to OSHA, and after being forced to do a $1 billion upgrade to the facilities, another four people were killed at the refinery. And another two were killed in another BP refinery in Washington State. BP thus had five fatalities in two facilities, while there had been a total of nine fatalities in all other 146 non-BP refineries in the U.S. (Steffy 2010, p. 139).

For comparison purposes, consider how BP had had 700 OSHA safety violations in three years, whereas Exxon, which after the Valdez disaster completely improved its safety record, had only one (Steffy 2010, p. 150). BP never seriously responded to the concerns of its employees, its lower level managers, or even regulators. Each time, BP would negotiate a settlement or simply pay the fine, and do what was demanded in paying for upgrades or installing safety management. But it would not end its relentless pursuit for higher profits by cutting costs and pursuing high risk high reward exploration and refinery. The norm in BP thus became that safety hazards were part of the job, that deviating from safety norms was normal, and that redress rather than prevention was the way to address them (Lustgarten 2012; Steffy 2010).

Volkswagen and Wells Fargo had similarly normalized deviancy. As we saw already, Volkswagen had used defeat devices all the way back to 1973, when it was first caught and ordered to pay a $120,000 fine to the EPA. Then, in 2005, VW had to pay a $1.1 million fine to the EPA for emissions cheating in Mexico. And starting in 1999, the company had installed a device in the software controlling the highly polluting noise control system in its Audi engines that would switch off this system and reduce pollution when it recognized the car was being tested (Ewing 2017). So, when engineers frantically sought to find a solution to make the new VW diesel clean, but also economical and practical, they had models to turn to. In fact, the Audi device served as a direct example for the much more widespread cheating that VW would do with the EA 189 engine. During a meeting where 15 engineers, including the head of VW engine development, met to discuss how to create an economical engine that would pass stringent U.S. emissions tests, the idea of this defeat device was presented and debated. While some pointed out of the risks breaking the law by adopting this device, others stated that this was normal and that many carmakers did so and VW had to do so as well if it were to keep up with competition (Ewing 2017, p. 122). According to Ewing’s analysis of the meeting, most engineers would not see this as “a grave violation of Volkswagen standards. There was plenty of precedent for using shortcuts to cope with inconvenient regulations” (Ewing 2017, p. 123). In all the earlier instances of cheating, VW as a company had turned a blind eye and condoned the behavior that had occurred, paying the fines should they come, without creating clear boundaries that this was unacceptable behavior. And once they had started using the cheat in their diesel engine there was no more stopping. As Ewing explains, “defeat devices which may have begun as a stopgap had become a habit” (Ewing 2017, p. 178).

Since adopting the ambitious Going for Gr-eight targets, Wells Fargo learned about more and more instances where its employees had opened fake and unauthorized accounts, from 63 in 2000, to 680 in 2004, to 288 in a single quarter in 2007, to 1469 in a single quarter in 2013 (Colvin 2017). In 2002, it was discovered that a whole Colorado branch had been opening unauthorized and fake accounts simply to reach their sales targets. The bank responded simply by firing individual employees involved, but not by addressing this as a systemic problem caused by their own targets. Bank leaders were actually positive about the numbers, as it showed them that only 1% of the work force had to be fired for cheating, while they assumed the other 99% were in compliance. As then CEO Stumpf said in an email to another bank leader: “Do you know only around 1% of our people lose their jobs [for] gaming the system, and about 2/3 of those are for gaming the monitoring of the system, i.e., changing phone numbers, etc. Nothing could be further from the truth on forcing products on customers. In any case, right will win and we are right. Did some do things wrong—you bet and that is called life. This is not systemic” (Colvin 2017).

While the company would fire employees caught red-handed in defrauding clients or the bank, it turned a blind eye to ongoing practices, never seeking to proactively find out how widespread they were and end them. Neither did Wells Fargo fully make clear that reaching sales targets was less important than compliance. As increasingly more employees started to cheat, a norm developed. As Colvin explains: “The message was clear to everyone in the retail bank: Everyone knew the goals were sheer fantasy for many branches and employees. At some branches not enough customers walked in the door, or area residents were too poor to need more than a few banking products. Bank leaders called overall quotas ‘50/50 plans’ because they figured only half the regions could meet them. Yet no excuses were tolerated. You met the quotas or paid a price” (Colvin 2017).

In all three companies, norms thus developed that normalized risky, rule breaking, and damaging practices that would come to shape the values and assumptions of corporate employees and executives. This was not merely a passive process in which the corporations allowed the practices to develop in response to the structures of budget cuts and highly ambitious targets. At times corporations would directly condone these practices. Volkswagen never strongly responded to any of the earlier cases where its engineers had installed defeat devices. At BP for instance, an independent investigation found that the oil company allowed “pencil whipping” and the fabrication of inspection data. One employee said that “BP workers felt pressure to skip key diagnostics, including pressure testing, cleaning of pipelines, and checking for corrosion, in order to cut costs” (Lustgarten and Knutson 2010). A former Wells Fargo assistant vice president and regional private banker has sued the bank claiming that she was fired when she refused “to participate in a scheme to manipulate accounts and sell products that weren’t in customers’ best interest.” She alleged that her superiors were running the scheme (Associated Press 2017). Other employees have come forth complaining that they were fired after trying to report the illegal practices to the ethic’s hotline (Egan 2017). In another case, a former branch manager had found out that bankers had swayed a homeless person to open six bank accounts getting her to pay $39 per month. She explained: “It’s all manipulation. We are taught exactly how to sell multiple accounts” (Reckard 2013). She reported the situation to higher executives but never received any answer (Reckard 2013). Or as another former employee explained: “Training in questionable sales practices was required or you were to be fired” (Colvin 2017). Clearly, at Wells Fargo this was not simply a matter of lower employees breaking Wells Fargo rules. And it was not simply a common practice: It had become something that was endorsed and for which no internal complaints were accepted, let alone seriously acted upon to change the root causes that sustained it.

The ongoing illegal behavior and internal responses to it in all three cases offer us further insight into toxic elements in their culture. Again, we will first discuss the types of toxic norms and their processes (drawing on Table 2 above) and then look at what level of culture we find these in (drawing on Table 1). A first clear toxic process was that in all three cases the existence of illegal practices formed visible common behavior (a descriptive social norm (cf. Cialdini et al. 2006)) that was directly against the law and thus came to compete and resist with the law (cf. Moore 1973; Heimer 1999). The lack of company responses to such illegal behavior spurred two further negative cultural processes. As illegal behavior could continue unaddressed, the companies normalized deviance (Vaughan 1989, 1997) and failed to foster learning processes that could prevent future misconduct (Homsma et al. 2009). As the companies turned a blind eye to the ongoing illegal practices, and never sought to proactively detect and stop them, they also created an opportunity for employees to cut corners at very little risk. As long as it did not create a major scandal with outside complaints or regulatory investigations, the companies were not proactively seeking to find violators or hold them accountable. And thus, the companies left the henhouse open to the foxes. Or in the terms of the Routine Activities Theory, “committed offenders” had access to “suitable targets” as there were no “capable guardians” (Cohen and Felson 1979). The negative cultural processes here started at the practice level, with visible common behavior, that probably became unaware common behavior, and then moved into the values, especially the injunctive social norms and hidden assumptions, as what people saw converged with what they came to think.

3.4. Responses to Exposure of Scandals

So, what happened when the three companies came to face strong public and legal scrutiny when each had its scandals exposed? Organizational responses to crisis offer a clear window into what values a company communicates both outwards and inwards to its own employees at the most critical moments. Crucially, responses to scandals are major organizational moments that can have a strong impact on the organizational culture and the extent to which it is toxic. In all three, the companies deflected blame, defended themselves from liability, and some, most notably VW, even tried to downplay the damages of their actions. By doing so, the companies neutralized their own culpability, thus enabling further rule breaking.

The Wells Fargo response to over a decade of fraudulent practices was to blame the individual employees, and not the unrealistic sales targets, lack of response to complaints, and threatening practices that had stimulated these lower level bankers to start cheating. As Frost explains: “It was convenient instead to blame the problem of low quality and unauthorized accounts and other employee misconduct on individual wrongdoers” (Frost 2017). At Wells Fargo, the fragmented corporate structure enabled blame shifting and obstructed taking responsibility for malpractice elsewhere in the company. As Colvin analyses: “For example, the corporate chief risk officer had no authority over the retail bank’s risk officer, who reported only to Tolstedt (who headed community banking). The HR department regarded employee misbehavior as an issue of training, incentive compensation, and performance management. The law department’s employment section focused mainly on litigation risks from firing employees. Each concerned itself with its assigned slice of the issue; no one looked for the root cause or envisioned big-picture consequences” (Colvin 2017).

Wells Fargo tried to persist in this strategy even in 2016, after the true scale of the fraudulent behavior became public and the company had to respond to news that it had defrauded millions of U.S. customers. Its first response was to blame individual employees and fire 5300 lower level bankers, without taking responsibility at the top. In a hearing before the Senate Banking Committee on 20 September 2016, CEO Stumpf apologized for not ending the illegal practices earlier and promised that the bank would undergo reform (Corkery 2016). Senators were angry that he did not offer any concrete steps against executives, including himself, and had just shifted blame and punishment to those at the bank’s lower levels. As Senator Elizabeth Warren asked him: “Have you returned one nickel of the money that you earned while this scandal was going on? Have you fired any senior management, the people who actually oversaw this fraud?” After Mr. Stumpf, answered that he had not, Warren retorted: “Your definition of accountability is to push this on your low-level employees. This is gutless leadership.” But Mr. Stumpf persisted stating: “The 5300 (fired employees) were dishonest, and that is not part of our culture. That is not scapegoating” (Corkery 2016). In a statement to National Public Radio inquiries following the hearings, Mr. Stumpf said: “Although the vast majority of our team members do the right thing, every day, on behalf of our customers, these allegations and accusations are very serious. And if any of these things transpired, it's distressing and it’s not who Wells Fargo is” (Arnold 2016).

Senators also pushed the CEO about whether the bank would seek to claw back the millions of compensations of top executives who had failed to fulfil their duty to stop the scandal. Here they especially focused on Carrie Tolstedt, who had been in charge of community banking where all the issues had happened. She retired at age 56 with a package of tens of millions of dollars, just three months before the hearing. Mr. Stumpf explained that although Ms. Tolstedt had let the illegal practices go on for three years after they were first discovered in 2013, he did not want to fire her because she performed so well in her other duties (Corkery 2016).

BP’s first response to major scandals was to deflect blame. One tactic was to try to place on blame on individual workers, at worst the ones who had been directly hurt in accidents. These were the same workers who had been concerned over the safety of the operations, as years and years of budget cuts had created a very hazardous working environment. And when things then did go wrong, as was likely to happen with the budget cuts, their company would blame them. Workers at the Texas Refinery, which was the site of the deadly 2005 explosion, had been interviewed previously about safety issues. One explained: “Yes I have been hurt and had management punish me and made a fool out of me. Need I say more?” (Steffy 2010, p. 66). Another, who had been hurt because of a mechanical failure: “I was blamed in the end. I was not the root cause” (Steffy 2010, p. 66).

And after the 2005 explosion at the Texas refinery facility BP followed a strategy of stonewalling and blame shifting (Smithson and Venette 2013). It first put the refinery on lockdown for eight days, not letting anybody in, claiming that it was too hazardous. Then two months later, it issued its internal investigation report and placed blame squarely on the low level employees who were alleged to have overfilled and overheated the raffinate splitter (Steffy 2010, p. 89). Steffy summarized BP’s response: “Human error-or workers not following rules-meant that BP itself wasn’t to blame” (Steffy 2010, pp. 89–90). Ironically enough, a BP executive had chaired the development of safety guidelines by the Center for Chemical Process Safety that concluded that “errant employees aren’t the root cause of an accident but rather its symptom” (Steffy 2010, p. 90).

Five years later, following the Deepwater Horizon spill, BP also sought to deflect blame. Its first tactic was to try to steer out of the scandal, when media during the initial 12 days focused on Transocean, the owner of the rig. Transocean could act as a good shield (Steffy 2010, p. 182). In one of the early statements BP CEO Hayward stated: “We are responsible, not for the accident, but we are responsible for the oil” (Smithson and Venette 2013, p. 402). He said this in spite of Transocean’s reliance for most of its business on BP for most of its business and the big oil company’s major influence in day-to-day operational decisions—decisions that enhanced risks and neglected industry standards, all to expedite the drilling and save costs (Smithson and Venette 2013, p. 402). Therefore at first, BP tried to use its decentralized structure with sub-contractors to defect blame. It had done so earlier. In the 1990s for instance, BP had blamed a sub-contractor, Doyon, when it was found that BP Alaska had been illegally injecting its toxic waste into the ground, even though BP was again directly in charge and Doyon relied for 80% of its income on BP (Lustgarten 2012, p. 61).

BP leadership also tried to downplay the role they themselves played in all this. This was most apparent in CEO Hayward’s testimony during the U.S. congressional hearings. He kept on deflecting critical questions about how BP had managed risk. Hayward instead focused on how much money the company had spent on safety (Smithson and Venette 2013, p. 402). Whenever he was pressed about problems, Hayward would insist that the investigation was ongoing and no firm conclusions about the role BP had played in all this could be made yet. When Representative Bart Stupak asked him, “Are you trying to tell me you have not reached a conclusion that BP really cut corners here?” Hayward answered simply: “I think it’s too early to reach conclusions, with respect, Mister Chairman. The investigations are ongoing” (Smithson and Venette 2013, p. 403). When Hayward was asked about BP’s decisions about particular aspects of the operation that had caused the risk, he would explain that he was not involved in the decision making. When pressed on details, he would claim ignorance. For instance, when Representative Michael Burges asked a question about why BP had installed fewer than the recommended number of centralizers that were to ensure the proper flow of cement, Hayward said: “I can’t answer that question, I’m not a cement engineer I’m afraid” (Smithson and Venette 2013, p. 404). But Hayward does have a PhD in geology, 28 years of experience in oil and gas exploration, and had been CEO of BP America for the three years prior. As Smithson and Venette conclude: “The suggestion that someone with Hayward’s experience and knowledge had insufficient information to determine whether BP made risky decisions was ridiculous” (Smithson and Venette 2013, p. 404).

BP did not stop there. It even tried to downplay the damage of the spill. At first, BP had stated that that oil was leaking at 1000 barrels a day, a relatively modest spill. That would indicate that the spill would only reach Exxon Valdez levels after one year. A little later, BP raised that figure from 1000 to 5000 barrels a day (Steffy 2010, p. 184). In fact, oceanographers from Florida had used satellite imaging data and found that the size of the leak was much larger at 30,000 barrels a day, meaning the leak could surpass the Valdez spill in only two weeks. CEO Hayward dismissed these findings saying that their own information was the most accurate: “A guestimate is a guestimate and the guestimate remains at 5000 barrels a day” (Steffy 2010, p. 184). When BP was forced to release its live video feed from the wellhead or face a congressional subpoena, experts found that the leak was even larger—about 60,000–70,000 barrels per day (Steffy 2010, p. 185). Hayward later even went as far as to deny that such a massive spill caused much damage. In an interview with the Guardian he stated: “The Gulf of Mexico is a very big ocean…the amount of oil and dispersant we are putting into [it] is tiny in relation to the total water volume” (Kollewe 2010).

But no one tried to deflect blame like Volkswagen. Soon after Volkswagen learned about the West Virginia University study that had demonstrated that the on-road emissions were many times higher than lab tested emissions, the head of product safety, Bern Gottweis, sent a memo to CEO Winterkorn. The memo concluded that “A thorough explanation for the dramatic increase in NOx emissions cannot be given to the authorities” (Ewing 2017, p. 177). He concluded that in further testing, the authorities would find out that there was a cheat device. And VW could revise the software to decrease emissions during road testing, but not to a compliant level (Ewing 2017). So very early on in the development of the scandal, the highest level executives at VW knew that regulators would find the defeat device and there was no way to salvage the situation. But rather than coming clean with the Californian and federal environmental regulators, VW opted to stall, to cheat even more, to deflect blame, and even to try to argue that the harm was limited. Soon after VW learned of the tests, there was a presentation that discussed the costs and benefits of different response options: refuse to acknowledge the problem and continue to stonewall and lie, offer an update to the engine software that would decrease emissions but not to the compliant level, or admit to the problem and buy back diesel cars in the US. The last option, Ewing concludes, “does not appear to have been seriously discussed at the time” (Ewing 2017, p. 179).

By this time, in May 2014, Volkswagen came to adopt a new NOx control system by installing urea tanks that catalyzed NOx into harmless oxygen and nitrogen. However, Volkswagen never installed a tank big enough to truly control the emissions at a sufficient level. So, the cars continued to have cheat devices that would only allow a sufficient level of urea to be used during lab testing, and not during road driving so that owners would not have to fill up their tanks with an extra chemical (Ewing 2017). In light of the investigation, Volkswagen tweaked its EPA application, indicating that owners would have to fill their tank “approximately” every ten thousand miles. This was a major change, as Ewing explains Volkswagen thus no longer promised that the system could work long enough through the full circle between regular oil changes. Volkswagen also updated the software so that the cars would use more urea to better catalyze NOx, and reduce the difference between road emissions and laboratory emissions, which would still not bring the true road emissions within the standards. In this way, Volkswagen tinkered and improved its cheating device, even when they already knew that in time it would be discovered. It continued to sell large volumes of cars that were not as clean as they claimed (Ewing 2017, p. 181).

Meanwhile, Volkswagen stalled and obstructed the CARB investigation into why there was a difference between the lab and road emissions. CARB at that time did not expect deliberate wrongdoing; they simply just tried to understand the cause and fix it. Volkswagen, however, was not really cooperative. As Ewing details: “The Volkswagen executives responsible for dealing with regulators gave answers that the regulators regarded as evasive, non-sensical, or dismissive. CARB’s testing was wrong, Volkswagen complained. The outside air temperature threw off the results. The routes followed [during the road tests] were inconsistent” (Ewing 2017, p. 182). As this process went on and started to consume more and more CARB time, Volkswagen informed the regulators that they would do a recall to update the software and “optimize” the emission control equipment in all clean diesels it had sold (Ewing 2017, p. 182). This was no admission of what actually had been happening, and instead gave the false promise that this would bring emissions within standards. Volkswagen also lied about the recall to customers and dealers, claiming that the recall was necessary to deal with a malfunction light defect (Ewing 2017, p. 183).

Most shocking, Volkswagen in fact used the recall to improve the effectiveness of the cheating software. It improved the car’s ability to detect when it was being tested in the lab, switching on its fully effective emissions controls only when the steering wheel was stationary for a longer time while driving, as it would only be at the lab (Ewing 2017, p. 183). The regulators responded with further and more stringent tests, and asked questions about other models with larger engines, which by then had also had defeat devices. And Volkswagen, even though by then it knew for sure that the game would soon be up, just continued its stonewalling and deceit.

Volkswagen executives became especially worried when they learned CARB was going to test an older clean diesel model. One internal email stated: “If the Gen I goes on to the roller at CARB then we’ll have nothing to laugh about” (Ewing 2017, p. 193). CARB also demanded that Volkswagen show them the software that controlled the urea injections in the new 2016 cars.

Volkswagen continued its cover-up all the way until it could no longer do so. It was forced to come clean and admit the existence of the defeat device only after CARB threatened that if Volkswagen failed to show them the software, it would refuse to approve the 2016 models onto the Californian market, which would keep them from the whole U.S. market (Ewing 2017, pp. 192–93). At first Volkswagen still would not admit to using the cheating software. A legal memo had estimated that the risk Volkswagen was running was still manageable. By now, Volkswagen had stalled and obstructed the CARB investigation for over a year. The company provided CARB a thick binder with the latest technical information, which seemed to indicate that VW had finally solved the problem. When CARB looked deeper into the information provided it found it was “all nonsense” (Ewing 2017, p. 197). The only explanation, CARB also now saw was that VW had been using a defeat device all along. It was August 2015, and CARB had still not approved the 2016 VW models, which were waiting in port to enter the market. And CARB was still waiting for the software information it had requested, and further it asked VW for a 2016 model car for new testing. This proved to be the final straw, as Volkswagen eager to get its cars on the US market ready for 2016, finally confessed that its cars had had defeat devices (Ewing 2017, pp. 197–98).

Volkswagen later claimed that it had failed to disclose the issue earlier because executives had not known about this. According to VW, it was only a small group of technicians who knew about the device. Top executives, Volkswagen claimed, had only learned about “conclusive proof” for the defeat device just before its confession to CARB (Ewing 2017, p. 200). And thus, Volkswagen moved from stonewalling, deceit and denial, to shifting blame downwards in the company. VW’s new CEO, Müller, has maintained this discourse since then. Stating, in an interview with a German newspaper: “Based on what I know today only a few employees were involved.” Defending his former CEO Winterkorn, he said: “Do you really think that a chief executive had time for the inner functioning of engine software?” (Ewing 2017, p. 216). Volkswagen leadership maintained this line, even when former CEO and grand architect of the Volkswagen growth strategy in the 1990s and early 2000s, Piëch came forward to claim that he had learned of the emissions problems in February 2015 while still chairing the board and that at the time Winterkorn had told them that there was nothing to worry about (Ewing 2017, p. 271).

Volkswagen ended up suspending several dozens of its midlevel engineers and executives, including the head of quality control and a member of the Audi management board (Ewing 2017, pp. 223, 256). Yet VW never addressed the more than a year period VW had tried to stall, obstruct, and deceive the ongoing investigations. Nor did VW seriously explore that the scandal involved a larger plot involving the highest level executives (Ewing 2017, p. 223). The company’s supervisory board never took any disciplinary action against the company’s top level executives who served on the management board (Ewing 2017, p. 256). And the car maker still paid out $33.9 million dollar in top executive bonuses, even when it reported a record $1.6 billion loss. This meant that even Winterkorn, who had been CEO during most of the saga, received a total compensation package of $8 million in 2015, and that was for ten months only, as he had retired in October (Ewing 2017, pp. 242–43). Volkswagen employees ended up paying the brunt of the costs, as in November 2016 the company announced it would cut 14,000 jobs (Ewing 2017, p. 258).

Volkswagen also tried to downplay that it had broken the law. VW CEO Müller tried to paint what had happened in a much more positive light. In an interview, he stated: “It was a technical problem…An ethical problem? I cannot understand why you [the reporter] say that.” As he explained, they did not have: “the right interpretation of the American law…We didn’t lie. We didn’t understand the question first. And then we worked since 2014 to solve the problem” (Glinton 2016). Meanwhile in Europe, Volkswagen took a directly confrontational legal approach, claiming that what had happened was not against the law there. A Volkswagen representative called to share the company’s response to the scandal to the UK House of Commons Transport Select Committee, called the software a “drive trace” and said that it “was not defined as a defeat device in Europe.” When members pressed that this was incorrect, the company’s representative simply stated that “in the understanding of the Volkswagen Group it is not a defeat device” (Ewing 2017, pp. 232–33).

The crassest deflection attempt of all was when Volkswagen tried to deny that it had damaged public health. It simply tried to refute that NOx was harmful. In a statement issued in late 2016 VW said: “A reliable determination of morbidity or even fatalities for certain demographic groups based on our level of knowledge is not possible from a scientific point of view” (Reuters 2016). In 2018, a German newspaper reported that Volkswagen had tried to back up its claims by exposing monkeys for hours to exhausts from the “clean diesel” engine of a 2016 Beetle, and compare them with monkeys exposed to the fumes from a 1997 heavy duty gasoline Ford F250 pick-up truck. Volkswagen had kept the study quiet, not in the least because the results had shown that the old Ford was less damaging to the monkeys than the state of the art Beetle. Several studies have now proven that the health effects are real. One of the most recent studies by MIT scientists, published in Environmental Research Letters, estimates that the extra NOx emissions emitted because of VW’s cheating will cost 1200 premature deaths in Europe, each dying a decade early (Chossière et al. 2017).

In conclusion, all three companies clearly tried to deflect blame, doing so each time they had a scandal and when the most major scandals erupted over the last years. Blame deflection had become a regular practice in all three companies, and with it came values that were harmful for compliance. The blame deflection resulted in several toxic cultural norms and processes. Again, we shall discuss these here referring back to Table 2 for the norms and processes and Table 1 for the levels these played out in within the cultures.

The first toxic process we see here is neutralization. By shifting away blame from the company and its executives to sub-contractors and lower level workers, the company neutralized the culpability of the corporation and its leaders, and it failed to take responsibility itself that could foster organizational learning from the wrongdoing that would help to prevent it (Homsma et al. 2009). This “denial of responsibility” can enable further rule breaking, as corporate executives rationalize and legitimize illegal practices in their firm as they reiterate time and again that it was not the corporation, that this is not who we are, or that it was just a few bad apples (cf. Maruna and Copes 2005; Minor 1981; Sykes and Matza 1957). This mentality prevents the corporation from developing normative values, from taking responsibility for mistakes, and from acknowledging that what happened was unacceptable and must be prevented at all cost. BP and Volkswagen also denied the damaging impact of the rule violations, claiming that the oil spill was but a drop in the ocean and that the NOx emissions were not damaging to health. This “denial of injury” is another classic neutralization technique that directly enables continued offending behavior, as it allows future rule breakers to tell themselves that this is not as bad as what people make it out to be (Maruna and Copes 2005; Minor 1981; Sykes and Matza 1957). In VW’s case, a different neutralization technique not discussed in the original criminological literature can be identified—a legal neutralization that occurred when the company claimed that its actions were not against the law. Again, this claim enabled offending and damaging behavior by appealing to the letter of the law to justify behavior was so clearly against its spirit.

The deflection of blame further strengthened the normalization of deviancy (cf. Vaughan 1989, 1997). As blame is pushed downward and outward, discipline for higher level leaders remains lagging. In all three cases, stronger internal action, to the extent that it did happen, only came after outside pressure. Wells Fargo, for instance, sought to claw back bonuses after relentless critique that it had not done so. And through this, the deflection of blame practices further enabled rule breaking and normalized deviancy, as it failed to establish a clear norm that offending behavior is not tolerated.

The deflection of blame does not play out in a vacuum, but exists as a response to the broader economic and legal forces companies operate in and thus interacts with the responses to strain these companies are under (Agnew 1992, 2001; Agnew et al. 2009; Simpson and Koper 1997). Blame deflection is not just part of the toxic culture at these three companies, but a much more common practice for firms that are trying to repair their image in the aftermath of scandals (Benoit 2014). Blame deflection is also a direct response against legal forms of strain that come with criminal and civil liability. With the push for punishment and compensation—that is highly justified and also necessary to end impunity—also comes the risk of steering companies towards blame deflection. Blame deflection undermines the value of taking true responsibility and setting true internal norms and, as is evident in all three cases, acknowledging that such behavior is not acceptable. And, in turn, the push to punish the highest levels of corporate leadership may instead lead to a shifting of blame onto those individual leaders, without truly addressing the toxic corporate culture.

Here we see that these toxic processes started most clearly at the level of values in the form of explicit shared values that came with statements from the companies. But when companies respond to a longer series of scandals, as BP so clearly did, employees will begin to expect that their company deflects blame and does not take responsibility, and at some point this is can become a hidden assumption, and thus become more deeply embedded in the corporate culture, reaching the levels of hidden assumptions.

3.5. Mixed Messages and Corporate Dissonance

None of the companies openly claimed that their actions of defrauding customers, cheating on emissions, or chafing of safety standards were good or intentional. All three companies ostensibly had ethical standards, positive corporate messages, and even branding and commercials that were highly aligned with compliance and the goals of the law. The expressed values all three companies promoted formally, however, were in stark contrast with their actual practices and the values the public observed through these practices. As New York Federal Reserve President William Dudley said about Wells Fargo: “There was a serious mismatch between the values Wells Fargo espoused and the incentives that Wells Fargo employed” (Puzzanghera 2017).

BP and Volkswagen provide the most detailed information about such “corporate dissonance” between preached values and practiced norms (Ewing 2017). During the 1990s, both companies had made environmental protection part of their core image. Volkswagen started with the development of the turbo charged direct injection diesel engine. The new technology that VW brought to market in 1989 made the fuel-efficient diesel cleaner and less noisy, thus making diesel a palatable option for small and midsized cars. The company dubbed this technology—associated with power and fuel efficiency—“TDI”, and made it a core part of its brand (Ewing 2017). In the 2000s, VW went a step further and developed “clean diesel,” which was promoted as an equally clean, fuel efficient, but more powerful and economical alternative to the hybrid models Toyota and other companies started to advertise. To promote the new technology, VW commissioned several new commercials. One series of commercial featured three old ladies driving in a VW clean diesel car and discussing “old wives tales” about diesel. Each commercial ended by suggesting that Volkswagen’s new clean diesel led to less noise, more power, and, most importantly, clean exhaust emissions . In one commercial, one of the elderly women placed her white handkerchief behind the exhaust just to show how clean the diesel was (Ewing 2017, p. 147). Another commercial, played during the break of the American Super Bowl, depicted environmental police busting people for minor issues, from not separating their trash to using the wrong lightbulbs. In the final scene, a police checkpoint verified if cars were environmentally friendly enough. A man driving a Volkswagen is waved through by a friendly cop who says: “Volkswagen Clean Diesel. Sir, you are good to go.”4

Volkswagen’s green branding and advertising stood in stark contrast to its actual practices, which were anything but clean. For those who had been in some way involved in the defeat device, the contrast between the public image of Volkswagen as environmentally friendly and the practice they knew was absolute.

This corporate dissonance between what was stated and what was practiced became even larger the moment the California regulators began to look into the discrepancy in VW lab and road test results. For over a year, VW’s official stance was that the testing had problems and that it was not accurate, but not that the company had been cheating. As the pressure escalated, VW executives must have come to know of the defeat device, and the more they strategized how to continue to deflect blame, the more the discrepancy grew between what VW said and what happened in the company. This is especially clear with the recall VW organized, formally letting regulators know it was to fix the emissions problem, and telling dealers that it was to fix a safety light, while in fact it was a deliberate attempt to make the cheating even better (Ewing 2017, pp. 182–83). What employees and executives learned at VW was that what the company said, whether in advertisements, against regulators, and against the public, had little to no relation to what it practiced. This continued after even VW admitted it had a defeat device, as it tried to argue that it had not broken EU law, that the emissions were not toxic, that leadership had not been involved, and that this was not a problem in the culture, but just the work of a small group of bad apples. All public statements, and all statements VW executives knew or came to know, were not true.

BP had a similar disconnect between its corporate communication and its actual practices. While BP cut costs and focused on risky exploration—resulting in a long stream of spills, hazards, and accidents—it sought to be seen as an environmentally responsible company. In 1997, BP CEO Browne announced that the link between greenhouse gases and global climate change was real and could no longer be ignored. He announced that BP would invest in alternative energy operations and research. A few years after, the company dropped its longer name, “British Petroleum”, and simply became BP, but with a new slogan “Beyond Petrol.” Also it launched a new shield logo of an upbeat yellow sun, signifying solar energy, surrounded by “what looked like leaves” (Steffy 2010). In the following years, and even after the Gulf disaster, the company repeatedly issued statements proclaiming that it was strongly committed to the environment. Yet this was in stark contrast not only to the company’s continued exploration operations, which were trying to get as much fossil fuels as possible out of the ground and sold to consumers to burn into the air (Frey 2002), but also to its appalling disregard for safety and the oil spills it could, and indeed would, cause. BP’s new imagine was not welcomed everywhere. Greenpeace said that a more fitting logo for the company would be “a miserable polar bear on an icecap shrinking because of global warming” (Frey 2002). And the environmental NGO honored CEO John Browne for the “Best Impression of an Environmentalist” (Frey 2002).

BP also claimed that it cared about safety—especially at times following major incidents. In 2000, for instance, BP’s Grangemouth refinery in Scotland had three separate accidents, all in one week. As a result, BP received a criminal fine of £750,000. The investigation found that BP’s quest for cost reductions had created the safety hazards, and that its fractured management structure had undermined a safety prevention strategy (Steffy 2010, p. 62). BP’s response was to state that it had gotten the message and that it had “shared the lessons it learned with its 11 other refineries in the world” (Steffy 2010, p. 62). While in fact, BP did not change its practices of cost cutting, ignoring safety concerns, and deflecting blame, even on those who had been hurt themselves in accidents BP had caused. With each new incident, BP claimed improvement and change, yet real change did not happen. Instead, BP’s internal communication on safety focused on minor, low-cost matters. As one former exploration engineer recalls, the company and its executives “focused so heavily on the easy part of safety, holding the hand rails, spending hours discussing the merits of reverse parking and the dangers of not having a lid on a coffee cup, but were less enthusiastic about the hard stuff, investing in and maintaining their complex facilities” (Steffy 2010, p. 57).

Real change did not even come when BP’s new CEO, Hayward, took over from Browne. Hayward promised to make safety a priority; he promised a “new BP.” Hayward promised a less complex organization with more transparency and accountability. He promised to hire 1000 engineers and improve safety in all of BP’s global operations (Steffy 2010, p. 151). However, Hayward never took full accountability for how BP’s cost cutting had led to the major accidents that had occurred before his tenure. He even resumed cost reductions, cutting $4 billion in 2009 in response to the 2008 crisis (Steffy 2010, p. 162). He was also unable to convince his executives that safety should become the priority, in part because English executives saw this as an American problem, and in part because he was not as popular as his predecessor Browne (Steffy 2010, p. 152).

Under Hayward’s tenure, BP would not change its core safety problems. As Steffy summarizes: “Its management structure was still convoluted, accountability was hard to find, decisions were made by committee, and cost cutting and financial performance continued to overshadow operations” (Steffy 2010, p. 160). A good illustration of this is a 2009 OSHA inspection at the Texas City refinery. The inspection resulted in one of the highest fines in OSHA history—$87 million. OSHA fined BP especially for hazards that had been identified before, but that BP had failed to fix. During another inspection in the BP refinery in Toledo, Ohio, OSHA found that BP had only done the repairs it had specifically mentioned, while leaving similar problems in other part of its operation unaddressed. As Steffy summarizes: “Hayward’s ‘laser’ was so precise that it was able to separate the letter of the rules from the intent” (Steffy 2010, p. 162). The more Hayward promised, as new CEO in the aftermath of the Texas refinery, to change BP, the more he undermined his own credibility and thus his ability to institute reform when practices on the ground did not actually change.

Here, again, we can analyze what this means for the corporate culture. Drawing on Table 2, we see that the most important type of toxic norm at play here is the disconnect between what the companies express as their values and what their actual day to day practice has been. A disconnect between the expressed values in support of compliance (safety, environmental protection, consumer protection) and practices that run directly counter to it will damage corporate compliance. When employees and executives hear one thing in corporate messaging, but see the complete opposite in everyday practices, they will not be convinced that their company and leaders are truly committed. This will either undermine the authority and credibility of corporate leadership within the company, or it will mean that lower level executives and employees learn that these messages are just for show, just to demonstrate commitment, but that what is truly expected is the opposite. This is very similar to findings in psychology that placing prohibition signs in environments where they are clearly being violated will create more offending (Keizer et al. 2011).

The damage in this case, however, can get worse. Once corporate employees and executives begin to doubt their leaders, achieving reform in corporate values and practices becomes very difficult. When CEOs preach improvement and preach that what has happened is not in line with the corporate culture year after year (or incident after incident), how will employees know when they truly mean it and when they should truly change what they do and think is right? Consequently, the corporate cognitive dissonance derived from conflicting corporate messages and practices can be viewed as the ultimate toxic element in a corporate culture—one that can obstruct any attempt to detoxify the culture.

4. Toxic Culture, Some Lessons

We have now discussed toxic norms and processes (drawing on Table 1 and Table 2) in these three cases by looking separately at five relevant business aspects: how goals and targets were established, how employees and managers could participate and dissent from these, what illegal behavior existed and how it was responded to, how the corporations responded to scandals, and what disconnect there was between public messaging and day to day practices. In this section, we analyze what we can learn when we combine these five aspects.

A first insight is that much of the toxic culture at BP, VW, and Wells Fargo did not come in the form of norms that were directly opposed to the law. We see that norms that enable rule breaking (by creating opportunity or neutralizing offending), norms that obstruct compliance (through strain and obstructing support), and norms that undermine the authority of either positive social norms or legal norms are highly important. This provides a very different view of what toxic culture is (cf. Moore 1973; Heimer 1999; Cialdini et al. 2006; Cialdini and Goldstein 2004; Vaughan 1989, 1997; Scholten and Ellemers 2016). It means that assessing toxicity requires identifying the norms that enable rule breaking, obstruct compliance, and delegitimize both the law and the social norms that support it.

A second insight is that toxic culture in these three cases was a matter of converging toxic processes. Strain in the market resulted in highly ambitious and risky targets that resulted in strain in the company (cf. Agnew 1992, 2001; Agnew et al. 2009; Simpson and Koper 1997). The same strain restricted management and employee input and dissent, which obstructed successful critique of high risk targets and reporting on illegal practices and resulted in both normalizing illegal behavior (cf. Vaughan 1989, 1997) and creating more opportunity for rule breaking (cf. Belknap 1987; Cohen and Felson 1979; Felson 1987; Osgood et al. 1996). Considering companies did not respond strongly to rule breaking when it did occur, such rule breaking itself became normalized, and organizations never fostered the learning from error that could help to prevent it (Homsma et al. 2009; Scholten and Ellemers 2016).

When the corporate wrongdoing ultimately ended in public scandals, political hearings, and legal investigations, the companies deflected blame towards lower employees, sub-contractors, and some executives. However, they rarely took full responsibility by admitting that this was a pervasive corporate issue. With the deflection, the corporations neutralized the company’s culpability (cf. Sykes and Matza 1957; Gottschalk and Smith 2011; Maruna and Copes 2005; Minor 1981; Siponen et al. 2012). It was not their fault, they could not help it, they did not know, it was not illegal, and there were no damages. All of these techniques of neutralization: (1) reinforced that norm violation was not an endemic problem that needed to be addressed, and (2) furthered the normalization of deviancy.

Ultimately, when companies were forced to state their guilt, address their culture, and change the tone at the top, it was too late. After years of dissonance between the ethical values the corporation had preached and the rule violating practices in their everyday operations, the company had undermined its own norms towards compliance (cf. Keizer et al. 2011). Consequently, employees and executives became less inclined to believe the leadership’s lofty promises to change the culture at the company and to reduce risk and rule breaking. Importantly, this is the ultimate ingredient in a toxic corporate culture, as the cognitive dissonance between values and practices undermines the authority of new leadership and values that support compliance. Figure 1 below outlines the convergence of these elements, and where they have occurred in the business processes: in normal business strategy and operations, in responses to rule breaking, and in corporate communications and advertising.

A third insight, closely related to the first two, is that the toxic cultural processes occurred at all three levels of the corporate culture: structures, values, and practices. Moreover, just as Schein (2010) argued, the three levels of corporate culture interact. For instance, explicit values (e.g., the public deflection of blame) become embedded in practices and structures, that structures (e.g., the high risk and high growth targets) shape values and practices, and that practices (e.g., the condoning of ongoing violations) become embedded in values and hidden assumptions. Although beyond the scope of this paper, it is very likely that what started as conscious and deliberative values and practices became more deeply embedded in hidden assumptions, unaware common practices, and situational norms, and thus more deeply engrained in the culture.

A fourth insight is that the culture in these three cases did not become toxic by design or by any singular influence. Toxic culture was not just a matter of one bad CEO, one bad set of incentives, or the tone at the top. Certainly, the fish can rot from the head, but that certainly is not the only way it rots. Indeed, most toxic processes within these companies were not directly against the law. Moreover, the processes interacted, and no one had full knowledge and control over this interaction and what it would cause. And finally, the toxic processes did not just exist in designed structures and formal policies, but also in values and practices that were shared throughout the company. Consequently, while individual blame and responsibility for elements and aspects of the toxic cultural processes can be found and should be assigned (such as the high-risk targets, or the organization of coercive personnel practices that stymied dissent), these are not responsible for the whole culture.

Of course, not all toxic cultures are the same. But these three specific cases illustrate how we can assess toxic culture at the level of structures, values, and practices, by identifying norms that run against the law, norms that enable rule-breaking, norms that obstruct compliance, and conflicting messages that undermine the authority of positive and legal norms.

5. Detoxing Corporate Culture

Once we learn what toxic corporate cultures are, the next question is what to do about them. In the case of VW, BP, and Wells Fargo, investigators and reporters have concluded that toxic culture was to blame and must be addressed. Naturally, changing a toxic corporate culture is remarkably difficult. Nonetheless, deriving from the analysis of the three cases, this paper offers some observations about detoxing corporate culture and suggests several possible steps.

For any toxic corporate culture, the first step begins with the proper assessment of the toxicity. Such assessment requires a deep analysis of structures, values, and practices that run against the law, that enable rule breaking, that obstruct compliance, and that delegitimize the law and norms supporting it. Unfortunately, common responses to assess corporate culture tend to latch on to singular elements that are thought to signify the culture. Most discussion of cultural change at Wells Fargo focuses on the incentive structures, while forgetting the strained environment and lack of proper venues for dissent they operate in, let alone the condoning, blame shifting, and neutralization of guilt that has taken place in response to public scandal.

To assess the toxicity, one’s initial reaction may be to turn to a survey. Certainly, there are several survey tools in academia designed to measure organizational and corporate culture. A widely-cited example of an academic survey measuring corporate culture is the one by Denisen that analyzes four cultural traits: mission, consistency, adaptability, and involvement (Denison 1990). Further, we can turn to Hofstede’s influential work that used surveys worldwide to identify organizational cultures along four dimensions: power distance, uncertainty avoidance, individualism vs collectivism, and masculinity vs femininity (Hofstede 1980, 1998; Hofstede et al. 1991, 1990). These surveys allow us to understand general traits of organizations that have often been linked to economic performance.

However, existing academic surveys or organizational culture have not focused on studying toxic elements in particular. The problem with surveys is that they can only measure what people know and what they are willing to share. As such, surveys only provide access to the conscious values and practices, and not to the hidden assumptions and automatic, unconscious habits. Moreover, once surveys focus on toxic cultural traits, they would necessarily touch on aspects of illegal or wrongful behavior. It is extremely difficult to get people to admit to rule-breaking or immoral values and practices without protection from prosecution. In addition, surveys that ask closed questions do not allow for learning about elements in a culture that the survey designers had not thought of, and thus may overlook vitally important toxic elements in corporate cultures. That is, our insights are limited to what we thought to measure.

To assess a company’s toxic culture, then, assessment cannot come solely through a survey. What is needed is a forensic ethnography, as outlined previously in this paper. Here, we have just done a first-level, desk-analysis using available resources from investigative reporting, company reports, and legal investigations. A true forensic ethnography would ask the same questions, but would do so through in-depth, open-response communication with current and former employees and executives involved in both operations and strategy of the company. Ideally, this should be done by investigators with training in qualitative social science research methods and ethnography, such as anthropologists and sociologists.

Step 2 is to change toxic structures of the corporation. This is the most straightforward way to start the corporate detox. Structures are the tangible, man-made, and changeable aspects of corporations. Should toxicity be found in such structures, as for instance the Wells Fargo incentives or the BP budget cut targets, then all that needs to be done is change such incentives and targets. This is simpler than it may look. First of all, the targets and incentives exist for a reason. Wells Fargo developed its targets under the strain of its overall performance in the market, and once it achieved growth with the incentives, it became very hard for leadership to let them go. Reformers may need to overcome resistance to change toxic structures, but that may well mean that first, they must address the strain that has caused them.

The second issue is that changing the structures will not eradicate negative behavior in and of itself, if such structures have become embedded in the deeper values and practices of the corporation. For instance, in the aftermath of the crisis, Wells Fargo ended its sales targets. Nonetheless, bank employees continued to feel the pressure to grow their sales and employees were still asked how many products they sold to each client and informally pushed to hit a certain target, even though such targets were no longer part of the official policy (Roberts 2017). Once toxic elements have moved from the structures into the values and practices, eradicating them through policy reform will not be enough. This has vital ramifications for how assessing toxic culture may undermine successful interventions. Most assessments focus on the structures because they are most visible, stimulating responses that sound good, but do not get to the root of the problem.

Step 3 is to address top executives. We see that CEOs played an important role in all three cases. They were the ones that had developed the risky targets. They had put strain on their executives to avoid critiquing these targets and to follow them compliantly, which the executives translated downwards. They turned a blind eye to the rule-breaking effects the targets were having. They failed to take disciplinary action against rule breakers. And they failed to take responsibility when the rule breaking became public. By deflecting blame, they neutralized culpability, and when they issued statements of support for the law that were in reality disconnected from the corporate practices and its actual values, they undermined the legitimacy of the law as well as their own legitimacy as the highest, internal leaders for compliance. Consequently, even when the CEOs and top executives were not directly involved in the rule breaking or had adopted policies that directly promoted rule breaking, they played a vital role in the toxic cultural elements that enabled rule breaking and obstructed compliance.

It seems natural to call for the replacement and punishment of these highest leaders—this may well be what is needed immediately. Replacement can start, or at least signal, the beginning of a shift in the culture of the firm. Punishment can show that what had happened in the past is wrong and should not happen again. However, while these may be necessary steps, they are not sufficient to change toxic culture and may even backfire. Discharging one CEO and hiring the next one, who enters an organization operating in a strained environment with toxic elements embedded in the values and practices, may inadvertently force the next CEO to adapt to the culture and continue the same problems. A good example is BP. Hayward vowed a turnaround when he succeeded Browne, promising an end to budget cuts and a true pledge to safety. Within a couple of years, however, it became clear that economic concerns forced him back to cost cutting, just like his predecessor.

Another challenge is knowing if the potential new CEO has the right values and is truly committed to compliance before taking the position. In this realm, management science may offer some insight on the characteristics of ethical leadership, which provides surveys to help discern the ethical style of the CEO candidates (Brown and Treviño 2006; Brown et al. 2005; Treviño et al. 2003). In practice, however, it is difficult to predict which leaders will be truly committed to compliance and will help solidify a compliance culture in a turnaround process.

Of course, addressing CEOs also means assigning accountability and punishment. Doing so could theoretically deter CEOs from allowing their companies to violate the law and cause harm. As we outlined at the outset of this paper, achieving deterrence in practice has proven difficult. But our analysis here shows that one of the unintended consequences of trying to prosecute executives is that it may harm a cultural detox. Trying to make CEOs liable for corporate misconduct, as seems to be so very justified with the impunity following the financial crisis, may result in more blame shifting, legal deflection, and a neutralization of guilt and culpability that only further strengthens the toxic culture. Also, it forces CEOs to continually come forth with public messages that run counter to the values and practices, and thus undermine credibility. Ideally, CEOs that come to clean up toxic cultures would be able to speak honestly and take responsibility. Unfortunately, this works best if they are under less pressure of potential liability and punishment.

This has insights for changing the tone at the top (Baggett 2003; Hansen et al. 2009; Schwartz et al. 2005). Analyses of toxic corporate cultures often point to the importance of a good, ethical, and compliant tone at the top. Indeed, such tone is vital, but it also needs to match the values and practices within the organization. When they are opposed, such as when leadership supports compliance while the organization does not actually practice it, the net result will be even more negative because it undermines the authority of the law and the CEO. Change in tone, therefore, must coincide with a true change in values and practices.

Further, the tone is not just about supporting ethics and compliance, but also, and maybe more importantly, about changing the norms that enable rule breaking and obstruct compliance. Here we can think of CEOs that express and initiate real support for worker empowerment and voice, for realistic targets, for honesty and transparency, for long-term over short-term goals. Detoxing the tone at the top must be honest and realistic for it to stimulate any meaningful change. This is especially important for a CEO unfortunate enough to face a long history of cognitive dissonance. Their first order of business is to convince the corporation that they stand by their word and that their expressed values will be matched by action.

Step 4 is to address managers and employees. Even when the structures are successfully altered, and there is new leadership with the right matching values and practices, corporations still face lower-level managers and employees whose values and habits may stimulate, enable, or sustain wrongdoing, or even simply obstruct compliance. One way to address this problem is to replace employees with toxic values and practices. However, how does the corporation exactly know who to fire, and who to hire? Of course, those caught breaking the law are clearly candidates for termination. But what about people who have simply not voiced dissent towards unrealistic targets, managers who have pressed their employees to meet these targets, or workers who have grown used to seeing violations around them? None of them may have broken the law, but their values and habits may well continue a toxic culture, and new hires may adopt similar practices and values. In some sectors, for instance in the automobile industry and its diesel emissions violations, rule breaking is more pervasive, as are norms and practices that sustain it (Ogbonna and Harris 2002). In some cases, everyone is involved in rule breaking to some degree. In 2002, Wells Fargo found that in one of its Colorado branches, all employees had been involved in defrauding clients. At first, Wells Fargo decided to fire all employees, but later had to rehire a number of them as it could no longer run its branch (Colvin 2017). A further problem with firing employees and managers is that it creates blame-shifting downwards, deflects responsibility from the company, and neutralizes its overall culpability—precisely the issues that need to be addressed within the toxic culture.

Consequently, replacing staff, while clearly necessary for the worst offenders, is not a silver bullet to achieve a cultural turnaround. The key challenge is to change the values and practices of existing employees. This is by far the most challenging step. Management science offers some ideas on how to do so, although it does not specifically focus on toxic cultures that promote rule breaking (Alvesson and Sveningsson 2015; Bass and Avolio 1993; Cameron and Quinn 2006; Harrison and Stokes 1992; Hatch 1993; Hofstede et al. 1990; Jones et al. 2005; Scalzi et al. 2006; Scott et al. 2003; Warren et al. 2014; Schein 2010).

One idea is to “unfreeze” organizations first (Schein 2010). This process prepares organizational staff and management to unlearn deeply-engrained practices, values, beliefs, and assumptions. The goal of unfreezing is to help employees overcome their natural resistance to letting go. Unfreezing starts by embedding so-called survival anxiety in the staff by shocking them into realizing that the organization, as well as their own position, are threatened unless there is a fundamental change.

Next, Schein (2010) stresses that to change an organizational culture requires creating so-called “psychological safety.” By this, he means that organizations must show their staff that changes are feasible and learnable. Once these first two steps have been successfully deployed, staff are ready to relearn new values and practices. These are major learning processes, as they run counter to what employees have experienced, both consciously and subconsciously. Schein explains that such learning involves a cognitive restructuring of the basic values and assumptions that underlie their behavior. Companies can initiate this cognitive restructuring when they relabel all core organizational concepts, tasks, and job descriptions. To do so, they have to redevelop core evaluation processes and standards, to stress the values of an open culture for worker complaints, and to formulate a long-term vision on sustainable development. This vision should include the long-term costs of unsafe and risky operations, and strive to eliminate any internal inconsistency between the tone at the top and the operations on the ground.

A core aspect of this relearning process is to “de-neutralize” the values that have come to legitimize rule breaking. After years of blame deflection, employees may have learned to see regulations as unfair, or the issues at their company as originating from a few bad apples. They may even deny that the rule breaking resulted in real injury. One option to de-neutralize the company’s employees and managers is through sessions where they meet with victims of the company’s actions. These sessions can be an opportunity for employees to come face-to-face with the injury caused, see the role the company played in the injury, and consider whose interests the law is actually protecting. Wells Fargo bankers, for example, would meet defrauded clients and mistreated former employees; VW engineers and executives would meet affected car owners and people suffering from pollution-related illnesses; and BP staff would meet the victims of its spills and accidents (Braithwaite 2010).

In these sessions, honesty, combined with realistic goals, is essential. Anyone who tries to initiate these change-processes may face a lack of credibility and inertia from staff who have heard it all before and who do not believe their leaders after years of corporate cognitive dissonance. Consequently, the final key element in addressing staff is empowerment.

Empowerment has two functions. First of all, staff need to be allowed to participate actively in the detoxing process. Only when they have a stake in the cultural overhaul will they come to support and stimulate a successful cultural change (Parker and Gilad 2011; Treviño et al. 2008; Weaver et al. 1999; Parker 2002; Hutter 2001). Second, worker empowerment is vital to creating a critical mass in the company that can resist unfeasible targets and overcome intimidation, in order to freely report wrongdoing and offending behavior. Empowerment entails much more than giving workers rights to speak out. As Garry Gray has shown, giving workers rights without giving them the ability to actually speak out may shift more blame on such employees when accidents and rule breaking happen (Gray 2009; Gray and Silbey 2014). Actual empowerment, therefore, must allow for workers to organize independently, to participate in decision making on issues that directly concern their work, and to have strong protections in disputes against superiors.

6. Conclusions

Toxic culture is the ultimate challenge when trying to prevent corporate crime and wrongdoing. Talking about toxic culture is easy, but actually analyzing its processes in order to change them is much harder. This paper has shown how toxic culture exists at the levels of structures, values, and practices in corporate organizations. Most crucially, it has shown that toxicity does not exist only when organizations develop norms that go against the law and promote rule breaking. As the cases at BP, Wells Fargo, and Volkswagen demonstrate, toxicity also exists when organizations have norms that enable rule breaking, obstruct compliance, and delegitimize the authority of the law or norms that support it.

Toxic cultures at BP, Wells Fargo, and Volkswagen resulted from a convergence of several core processes. Among these processes, this paper identified: the strain on the company and employees; unfeasible and risky targets; obstruction of dissent; the opportunity for rule breaking based on condoning ongoing violations; the normalization of deviance as increasingly more people break rules; deflecting blame away from the company; neutralizing company culpability and legitimizing offending behavior; and delegitimizing the law because of a dissonance between preached values and illegal practices. Many of these processes in of themselves are not illegal, nor do they directly promoting rule breaking, but rather they enable violations and make compliance harder.

In these cases, the toxic culture did not develop by some grand design or because of a singular act or actor. While CEOs and high-level executives played a major role in these processes, they did not plan them or fully control them. They instigated some (e.g., targets, blame deflection), while simply letting others develop unaddressed.

All this has important implications for how we address deeply embedded corporate wrongdoing. The normal impulse is to seek blame and punishment as high up as possible. The idea is that ending impunity will cure further wrongdoing. Of course, ending impunity is a necessary condition for successful behavioral change. However, when wrongdoing is endemic in the culture, it indicates that it is not the fault of a singular leader, but part and parcel of the broader organization.

The core task becomes to challenge the culture. To detox corporate culture means addressing all of the converging elements. This does include ending impunity of leaders. In fact, simply placing more pressure on top leaders may result in more blame shifting, neutralization, and corporate dissonance. Rather than solely focusing on holding the highest executives liable, detoxing corporate culture must also come with a full overhaul.

Detoxing culture begins with a full assessment of toxicity in the structures, values, and practices, as developed in this paper. Subsequently, detoxing must take into account both changes made in the structures along with addressing the values and practices of executives and employees. As a first step to creating meaningful change in the culture, it is important to create a balance between ensuring accountability for those found to have done wrong and creating an open and safe space for all organizational members to truly discuss what happened, why it happened, and what harm it caused. The legal responses to mass corporate wrongdoing have much to learn from the field of transitional justice, particularly its analysis of legal responses to war and mass atrocity and its ideas on how to balance justice (holding perpetrators accountable) and peace (ensuring that warring parties refrain from future conflict) (Bell 2009; Lambourne 2009; Laplante 2008). Detoxing corporate culture, then, may well require the creation of Truth and Reconciliation Commissions. These commissions would allow corporate wrongdoers to confront victims in a safe and open exchange and would focus on unearthing truth and giving insights rather than on punishment and retribution, paralleling post-apartheid South Africa (Wilson 2001).

This paper hopes to aid practitioners tasked with detoxing corporate cultures. A better understanding of toxic corporate processes may lead to the development of better ways to reform them; however, it is important to caution against promises of an easy fix. Cultures are notoriously difficult to change, and lofty promises of changed cultural values and practices can generate more corporate dissonance when real change does not occur (Ogbonna and Wilkinson 2003). Therefore, any successful corporate detox must begin with resisting the pressure for a quick fix or a quick turn-around. Successful cultural overhaul starts and ends with honesty and patience.