The normalization of corruption in organizations (2003) [pdf]

Number of Pages: 52 # THE NORMALIZATION OF CORRUPTION IN ORGANIZATIONS Blake E. Ashforth and Vikas Anand ## ABSTRACT Organizational corruption imposes a steep cost on society, easily dwarfing that of...

Number of Pages: 52 # THE NORMALIZATION OF CORRUPTION IN ORGANIZATIONS Blake E. Ashforth and Vikas Anand ## ABSTRACT Organizational corruption imposes a steep cost on society, easily dwarfing that of street crime. We examine how corruption becomes normalized, that is, embedded in the organization such that it is more or less taken for granted and perpetuated. We argue that three mutually reinforcing processes underlie normalization: (1)institutionalization, where an initial corrupt decision or act becomes embedded in structures and processes and thereby routinized; (2) rationalization, where self-serving ideologies develop to justify and perhaps even valorize corruption; and (3)socialization, where na¨ıve newcomers are induced to view corruption as permissible if not desirable. The model helps ex-plain how otherwise morally upright individuals can routinely engage in cor-ruption without experiencing conflict, how corruption can persist despite the turnover of its initial practitioners, how seemingly rational organizations can engage in suicidal corruption and how an emphasis on the individual as evil-doer misses the point that systems and individuals are mutually reinforcing. I will never believe I have done anything criminally wrong. I did what is business. If I bent any rules, who doesn't? If you are going to punish me, sweep away the system. If I am guilty, there are many others who should be by my side in the dock (on trial). - an architect, convicted of corrupt practices (Chibnall & Saunders, 1977, p. 142). Research in Organizational Behavior Research in Organizational Behavior, Volume 25, 1-52 Copyright ©2003 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0191-3085/doi:10.1016/S0191-3085(03)25001-2 12 BLAKE E. ASHFORTH AND VIKAS ANAND ## INTRODUCTION What comes to mind when you think about criminals and crime? Odds are that you picture a burglar breaking into a home, a pusher dealing drugs, or some related image. Odds are, in short, that you picture a lone individual engaged in street crime (Collins, 1998; Ermann & Lundman, 1996). In fact, Clinard (1979, p. 16) concluded that "Far more persons are killed through corporate criminal activities than by individual criminal homicides," and the U.S. Department of Justice estimates that the economic costs of corporate crime are seven to 25 times greater than that of street crime (Donziger, 1996). As Darley (1996, p. 13) argues, "the typical evil action is inflicted . . . by individuals acting within an organizational context" rather than by "evil actors carrying out solitary actions." Moreover, Edwin Sutherland (1949), who coined the term white collar crime and pioneered early research, argued that corporate misdeeds tear the social fabric more so than street crime because they corrode trust in authorities and institutions. An even more troubling trend in white-collar crime is that these crimes appear increasingly to be perpetrated through the actions of numerous employees in the organization as opposed to being the actions of a single misguided individual. A glance at today's headlines illustrates the point. Enron's senior executives are accused of hiding the company's precarious financial position while cashing out their stocks and shredding incriminating documents (Eichenwald & Henriques, 2002), U.S. Catholic Church officials admit that several key administrators systematically covered up the predatory behaviors of pedophile priests (Miller & France, 2002) and Merrill Lynch agreed to pay the State of New York $100 million to settle accusations that its employees gave its investors misleading and over-optimistic research reports about the stock of its investment banking clients (Gasparino & Smith, 2002). In this paper, we develop a model that explains how corruption becomes normalized in an organization. We define corrupt acts as the misuse of authority for personal, subunit and/or organizational gain (cf. Sherman, 1980). Because we will argue that corruption often becomes institutionalized in organizations, it is important to note that "misuse" refers to societal norms. We focus on relatively severe or "morally intense" (Jones, 1991) forms of corruption (e.g. self-dealing versus sleeping on the job) because these are more difficult to normalize. We in-clude corruption on behalf of the organization (e.g. price-fixing, bribing outsiders to win contracts) and against the organization (e.g. theft, nepotism) (Coleman, 1987), often referred to as corporate/organizational crime and occupational crime, respectively (Clinard & Quinney, 1973). Our model examines collective corruption (Brief, Buttram & Dukerich, 2001) - acts that require cooperation among two or more individuals. Our analysis is The Normalization of Corruption in Organizations 3confined to the group level (throughout the paper, we use the term group to refer to a collective, such as a workgroup, department, or organization): we do not examine individual differences in susceptibility to corruption. Further, we focus largely on groups that are housed within an organization rather than distributed across organizations (as in, for example, bid rigging). Additionally, our interest is not in the antecedents of corrupt acts - the primary focus of much previous research (e.g. Baucus, 1994; Brass, Butterfield & Skaggs, 1998) - but in how such acts become normalized , that is, become embedded in organizational structures and processes, internalized by organizational members as permissible and even desirable behavior, and passed on to successive generations of members (cf. Brief et al., 2001; Zucker, 1977). We propose that there are three pillars that contribute to the normalization of corruption in an organization: (1) institutionalization , the process by which corrupt practices are enacted as a matter of routine, often without conscious thought about their propriety; (2) rationalization , the process by which individuals who engage in corrupt acts use socially constructed accounts to legitimate the acts in their own eyes; and (3) socialization , the process by which newcomers are taught to perform and accept the corrupt practices. As shown in Fig. 1, the three pillars are mutually reinforcing and reciprocally interdependent; once established in an organization, the pillars create a situation where corruption is practiced collectively by employees and may endure indefinitely. Our analysis suggests an answer to the intriguing question of how a person who is a loving parent, thoughtful neighbor and devout churchgoer is able to engage in workplace corruption. The first three sections of the paper describe each of the pillars in detail. The fourth section focuses briefly on how institutionalization, rationalization and socialization interact to cement normalization. Finally, we discuss the implications of our model for research and practice. > Fig. 1. The Three Pillars of Normalization. 4 BLAKE E. ASHFORTH AND VIKAS ANAND ## INSTITUTIONALIZING CORRUPTION > Fig. 2. Institutionalizing Corruption. Phase 1: The Initial Decision or Act Much of the literature on organizational corruption has focused on the genesis of corruption (e.g. Baucus, 1994; Brass et al., 1998; Cohen, 1995; Coleman, 1998; Finney & Lesieur, 1982; Geis & Salinger, 1998; Poveda, 1994; Shover & Bryant, 1993; Szwajkowski, 1985; Yeager, 1986); thus, we will keep our overview brief. This literature argues that the motivation and opportunity to engage in corruption are the product of various environmental (e.g. strong competition, lax legal and regulatory enforcement), organizational (e.g. poor performance, structural complexity) and, to a far lesser extent, personal factors (e.g. cognitive moral development, fear of failing). An implicit assumption seems to be that corruption is usually the product of strong situations that override individual differences or that ordinary people populate the ranks of the corrupt; indeed, Coleman (1998, p. 178) states that "Almost all the studies have agreed on one point: White-collar offenders are psychologically 'normal.' " In particular, many scholars have linked corruption to a permissive ethical cli-mate at the societal, industry, or organizational levels. Major organizational goals are often tied to resource procurement and financial success and a constellation of values extolled in business is conducive to an amoral and even immoral pursuit of those goals: free enterprise (minimal regulation), individualism, competitive achievement, profitability, efficiency, pragmatism and so on (e.g. Jackall, 1988; Mills, 1956). Brief et al. (2001) argue that belief in the sanctity of the corporation 6 BLAKE E. ASHFORTH AND VIKAS ANAND is so strong that managers tend to believe that by serving the corporation's interests they are also serving the public's interests: what's good for Microsoft is good for the country. Thus, ethical issues often are not perceived or are subordinated to or reframed as economic, legal, public relations, or other kinds of "business" issues, leaving managers free to engage in amoral reasoning (Jones & Ryan, 1998). For example, Brenner and Molander (1977, p. 62) surveyed Harvard Business Re-view (HBR ) readers and found that nearly half agreed that "the American business executive tends not to apply the great ethical laws immediately to work. He is preoccupied chiefly with gain." According to the rational choice perspective in criminology (Cornish & Clarke, 1986), individuals are thought to engage in "an assessment, however crude or incomplete, of options and the potential risks and payoffs of each" (Shover & Bryant, 1993, p. 153), along with the skills, planning, time needed and so on. Given a permissive ethical climate, an emphasis on financial goals (with com-mensurate rewards for success), and an opportunity to act amorally or immorally, the ends may soon come to justify the means, leading to a decision to engage in corruption. Further, the leniency and low frequency of formal sanctioning by governments and professional associations often makes corruption economically rational (Braithwaite, 1989; Leaf, 2002): it appears that crime often does pay. Finally, as we argue later under "Rationalizing Corruption," groups often provide self-serving accounts to neutralize the countervailing force of morals and ethics. To be sure, there are other routes to corruption besides amoral and immoral calculation, including principled disagreement with public policy and laws, managerial incompetence, and the unintended consequences of myriad actors and actions ricocheting within complex systems and contexts. However, we believe, following Vaughan (1992) and others, that amoral or immoral calculation is the most common route - in part because such a calculation is often necessary for a given act of corruption to continue and become institutionalized, regardless of its genesis. Leadership Leadership plays a potentially huge role in the institutionalization process. Leaders not only control many of the levers of institutionalization, but are very potent role models for organizational members. Baumhart (1961) and Brenner and Molander (1977) asked HBR readers to rank five factors according to their influence on unethical decisions (the 1977 survey included a sixth factor, "society's moral climate," which ranked fifth). In both surveys, the "behavior of superiors" was ranked as the most influential factor, followed by a cluster of factors - "formal policy or lack thereof," "industry ethical climate," and "behavior of one's equals in the company"; tellingly, "one's personal financial needs" came last. The Normalization of Corruption in Organizations 7Leaders do not have to actually engage in corruption to serve as role models: rewarding, condoning, ignoring, or otherwise facilitating corruption - whether intentionally or not, or explicitly or not - often sends a clear signal to employees (Baucus, 1994; Ermann & Lundman, 1996). Brief et al. (2001) argue that an emphasis on ends rather than means, supported by high standards and strong rewards (punishments) for attaining (not attaining) them, creates a permissive ethical climate. For example, Sims and Brinkmann (2002) discuss the role of CEO John Gutfreund in helping to foster an unethical climate at Salomon Brothers. Gutfreund focused on and rewarded short-term results, modeled aggressive and Machiavellian behavior, failed to punish employees who broke laws in the pursuit of those results, lied about and covered up ethical and legal lapses, promoted like-minded employees, and fired others capriciously, encouraging conformity. In addition to serving as role models, leaders - as the legitimate agents of the organization - authorize corruption (Kelman, 1973; cf. displacement of responsibility, Bandura, 1999; sanctioning, Brief et al., 2001). As with role modeling, authorizing need not be formal and explicit: a manager who informally encourages or tacitly condones corruption can also be said to be authorizing it. The subordinate, as a designated role occupant, is expected to execute the authorized acts, not to second-guess them: the normative duty to obey is expected to trump personal preferences, particularly in rigidly hierarchical organizations (Hamilton & Sanders, 1992). As illustrated by Milgram's (1974) obedience experiments, the reflexive impulse to obey authority figures - the "habit of obedience" (Hamilton & Sanders, 1992, p. 72) - is so strong and pervasive that most people have a difficult time actively defying orders they do not condone. In any event, because the individuals who perform the corrupt acts "are not the actual agent of their actions, they are spared self-condemning reactions" (Bandura, 1999, p. 196). Further, the relative powerlessness often associated with subordination may induce individuals to abdicate responsibility for moral issues: "'Let the people making the high salaries tackle the difficult ethical decisions' seems to be a widely held view among occupants of lower echelon corporate positions" (Jones & Ryan, 1998, p. 440). Two final notes regarding leadership and corruption are in order. First, the situ-ational power of office may be complemented (or even supplanted) by the personal power of charisma. The more charismatic the authority figure(s), the greater the identification, trust and reflexive obedience that he or she is likely to engender among subordinates (Conger & Kanungo, 1998). Charisma speaks to a quality of the person, not the mission, and a charismatic leader can use the allegiance of others for benign ends or corrupt ends (Howell, 1988). The cult of personality that arose around Michael Milken enabled him to assemble a force of like-minded disciples at Drexel Burnham Lambert (Stone, 1990). 8 BLAKE E. ASHFORTH AND VIKAS ANAND Second, organizational structures and processes are often contrived to insulate senior managers from blame, thereby further encouraging corruption. For example, a focus on performance goals rather than on means or the use of verbal and general (i.e. ambiguous) orders regarding means, coupled with minimal oversight and documentation, afford the managers "strategic ignorance" (Katz, 1979, p. 297) and therefore "plausible deniability" (Braithwaite, 1989; Browning, 1989, p. 1). Baker and Faulkner (1993) found that top executives involved in two decentralized price fixing networks were less likely to be found guilty than top executives in a centralized price fixing network, and Braithwaite (1984) reports that pharmaceutical firms instituted a "vice-president responsible for going to jail," namely, the only executive who needed to know about corrupt activities. Similarly, isolating subunits allows senior managers to be "willfully blind" (Braithwaite, 1989, p. 351) to corrupt practices (e.g. the independent cells of the Mafia). Thus, the managers can claim that they neither knew of nor approved the corrupt action. Phase 2: Embedding Corruption in Organizational Structures and Processes Organizational memory is the metaphor used to describe the process through which an organization acquires, stores and uses the knowledge that is applied to its activities (Anand, Manz & Glick, 1998). In its quest for efficiency, an organization tends to commit a promising activity to its memory (and thus institutionalize it) when it has been performed repetitively in the immediate past (Yates, 1990). Activities stored in organizational memories are often performed through the execution of a series of inter-related routines. The outputs of one routine trigger the initiation of the next routine in the sequence. Thus, entire activities can be performed without any one individual knowing them in their entirety (Levitt & March, 1988; Nelson & Winter, 1982; Weick & Roberts, 1993). Once a corrupt decision or act produces a positive outcome and is included in organizational memory, it is likely to be used again in the future. When similar issues confront other managers, and if solutions are not readily obvious, answers are sought from the memory because: (1) past decisions and acts are assumed to have been made for rational reasons; and (2) following a precedent helps legitimate the decision and act (Henisz & Delios, 2001; Ocasio, 1999; Simon, 1976). For instance, Ermann and Lundman (1996, pp. 23-24) describe how Gulf Oil laundered and distributed corporate funds illegally to American politicians: After the comptroller who helped develop the system left, three other people sequentially occupied his position. None of them had to make any difficult decisions, much less consciously involve themselves in criminal activities. They merely were told that they would receive requests for money . . . Easy to do and easy to live with. The Normalization of Corruption in Organizations 9The use of previous unethical decisions is especially significant because individ-uals often do not consider the contextual issues (including ethics) surrounding the decision or act; the past success of such practices is assumed to validate the process through which it was determined (cf. Miller, 1993). And if the renewed corrupt practice results in a positive outcome an even stronger precedent is set for the future. The organization comes to expect and then depend on the payoffs from corruption. In time, goals, budgets, information flows, rewards and punishments and so on may be skewed to support the practices. Eichenwald (2000), for instance, discusses the bureaucratic machinery and processes that evolved to support price-fixing by Archer Daniels Midland and its competitors. The result is institutionalized corrup-tion: personal behaviors become impersonal norms, emergent practices become tacit understandings and idiosyncratic acts become shared procedures. Moreover, the increasingly casual practice of corruption tends to further degrade the ethical climate of the organization. In short, micro and macro practices are mutually reinforcing, encouraging further corruption (Cialdini, 1996; Vaughan, 1992). Culture As the repertoire of corrupt practices becomes embedded in the ongoing routines of the organization, a deviant culture (or subculture, in the case of localized corruption) tends to emerge to normalize the corruption. Assumptions, values and beliefs, perhaps drawing on the business values noted above, evolve to rationalize the corrupt practices in ways that neutralize the stigma of corruption. (Later, under "Rationalizing Corruption," we discuss the types of accounts through which this is accomplished.) Indeed, the culture may come to valorize the corruption and promulgate recipes for corruption. For instance, Kappeler, Sluder and Alpert (1994) describe subcultural norms that support police corruption: don't give up another cop; if you get caught, don't implicate anybody else; don't get involved in another cop's affairs; don't trust new cops until they've been checked out; and don't tell anybody more than they need to know. Deviant (sub)cultures insulate actors from the wider culture with