An Analysis of Corporate Ethical Code Studies: "Where Do We Go From Here?" Bets)/Stevens

ABSTRACT. The dramatic increase in the number of these days. Companies without them are scrambling corporate ethical codes over the past 20 years has been to commit corporate values to paper." The prolifera- attributed to the Watergate scandal and the !aoreign Corrupt tion of corporate ethical codes over the past 20 years Practices Act. Ethical codes differ somewhat from profes- has attracted the interest of scholars who have sional codes and mission statements; yet the terms are primarily used content analysis to discover subjects frequendy interchanged and often confused in the literature. in these codes. The studies have been most useful in Ethical code studies are reviewed in terms of how codes are determining the factual subject matter found in communicated to employees and whether implications for violating codes are discussed. Most studies use content codes and the extent to which the same subjects analysis to determine subjects in codes. Little information is occur in different codes. available about how codes are communicated, whether they At the same time we might ask "Have we learned are accepted and used by employees, and whether they affect what we need to know about corporate codes?" employee/corporate behavior. More research on ethical Certainly the content analyses have yielded rich codes is rmeded m answer some of these questions. information about subject matter, but can they answer our quesdons about corporate codes in terms of how- they are communicated to employees and Incidents surrounding Watergate and passage of the whether they are truly effective? This paper first Foreign Corrupt Practices Act in 1977 have been attempts to present some definitions of ethical codes, credited to the tremendous growth and popularity of professional codes and mission statements, then corporate ethical codes. Following the 1975 Water- reviews the existing ethical code literature, and gate scandal, a number of corporations were impli- finally suggests some new directions for ethical code cated in illegal or questionable acts and calls for research. government regulation received increased support. Smarting from criticism and low public opinion, a number of corporations wrote codes of ethics to Definitions betoken a new, more moral image and to redefine their values. Corporate ethical codes differ somewhat from pro- This increase in codes, noted by Cressey and fessional codes and mission statements, yet the terms Moore (1983), Becket and Fritszche (1987), Beren- are frequently interchanged by corporations and beim (1987), Brooks (1989) and Rogers and Swales occasionally confused in the literature (Pearce and (1990) also received attention in major business David, 1987). Lines between these genres are not newspapers. In a Wall &reetJournal article, Bennett really clear cut. Moreover, few code studies have (t988, p. 17) remarked, "Formal ethical codes are hot undertaken the task of defining corporate codes. Cressey and Moore (t983) moved toward defining Betsy &evens is a lecturer at the University of Michigan &hoot of ethical codes when they exctuded documents from Business" Administration. She received her Ph.D. in 1992 in their study which did not "address themselves to organizational communication. Her research interests are corpo- activities which reach beyond the firm itself" (p. 55). rate ethical codes and human resource issues such as employment- A somewhat basic definition is offered by Berenbeim at-wilt. (1987) - "A code of ethics is a major vehicle for

Journal of Business Ethics 13: 63--69, 1994. © ! 994 KlmverAca&mic Pa~.b!ishers.Printed in the Netherlands, 64 Betsy Stevens stating ethical principles" (p. vii). Pitt and Grosk- mission statements frequently resemble ethical codes aufmanis (1990) take it further by using "corporate when they incorporate more metaphysical concepts codes of conduct" and "corporate codes" as terms to such as values and ideals. Distinctions between designate "any written statement of ethics, law or ethical codes, professional codes, and mission state- policy.., delineating the obligations of one or more ments often become blurred and sometimes overlap. classes of corporate employees" (1559, # 1). For example, Rogers and Swales (1990) describe the ,We might first attempt to define corporate ethical Dana Corporation code as "a hybrid code for it is codes through description. They are written docu- both a credo or mission statement and an outline of ments, ranging in length from one paragraph to fundamental principles and policies for Dana Cor- more than fifty pages, which are intended to impact poration and Dana employees" (p. 297). employee behavior. Firms frequently attempt to Corporate ethical codes have roots in early pro- manage and articulate ethics through these messages fessional codes. While a number of ethical codes which are designed for both internal and external were written in the early"part of this century (i.e., the audiences. Codes, then, are managerial tools for J. c. Penney code in 1913), most came much later. shaping change. They often demand from employees Professional codes, however, were common in the higher standards of behavior than required by law. A 1920's. An early collection of professional ethical code may be part of a personnel policies manual, codes (Heermance, 1924) includes codes of the which many courts interpret as a legal contract National Association of Straw Hat Manufacturers, between employee and employer, or it may be a the National Association of Peanut Butter Manufac- separately issued document which stands alone. turers, the National Confectioners Association of the Some have affidavits of understanding or oaths United States, a collective of candy makers. These which require the employee's signature and others associations reflect the pre-corporate status of many do not. Most importantly, they are messages through small companies which existed to produce single which corporations hope to shape employee behavior products. and effect change through explicit statements of The morals and values articulated in these codes desired behavior. embody the prevailing ethic of the time. For exam- Professional codes articulate the goals and beliefs ple, The National Confectioners Association of the for groups of professionals by which individual United States disparaged games of chance in candy practitioners can be guided. These frequently require packaging by arguing "they encourage and develop higher standards than are legally mandated and have the gambling habit in children, are injurious to their evolved in response to social, environmental and morals and . .. are a menace to the future develop- economic demands.1 Frankel (1988, p. 110) noted ment of the industry" (Heermance, p. 119). that a profession's code of ethics is "the most visible Ethical codes exist, in part, because corporations and explicit enunciation of its professional norms... are held legally accountable for the actions of their and embodies the collective conscience of a profes- employees. While corporate criminal liability is a sion." Additionally, Frankel argued that professional 20th century phenomena, it is based on the notion of codes could be classified as aspirational, educational, respondeat superior: "Let the master answer" (Black, or regulatory. The first classification states ideals to 1983, p. 681). This principle from common law which practioners should strive, the second enhances holds the head (master) liable for illegal or wrongful understanding through commentary and the third acts of his servants. From this has emerged the provides a detailed set of rules. doctrine holding the corporation responsible for the Mission statements attempt to address strategic actions of its employees. In addition to responsibility, management issues. They define the "fundamental punitive damage assessment has a long history dating unique purpose that sets a business apart from other back to the Middle Ages (Pitt and Groskaufmanis, firms of its type and identify the scope of businesses' 1990). Corporate codes have been constructed partly operations in product and market terms" (Pearch as a defense against respondeat superior; however they and David, 1987, p. 109). While they differ from do not always get the company off the hook. Courts corporate ethical statements by more directly ad- have ruled that instructions to employees do not dressing strategic planning and bottom line issues, necessarily relieve the organization of responsibility Corporate Ethical Codes 65

for employee wrongdoing (Pitt and Groskaufmams, with federal laws (67%) and misuse of corporate p. 1560). A corporation can find itself held doubly assets (67%). They found an overall lack of con- responsible for employees misconduct and also for sistency of subject matter which they attributed to demonstrating it has successfully communicated inconsistency among the corporations. Most firms appropriate behavioral standards to employees. Thus, required officers and key employees to periodically the transmissionat aspects of corporate ethical codes sign the code, but only about 25% required other - how they are communicated to employees - are employees to sign. significant both in terms of legal expectations and in Their study does not address the manner in which communicating corporate cultural expectations. codes were communicated or include information Some ethical codes are little more than legal about tone or format, except to describe s@es as barriers and self-defense mechanisms; others are ranging "from format and legalistic to nearly conver- intended to influence and shape employee behavior. sationai" (p. 86). Additionally they concluded that In either event the impact of these codes as messages some codes of conduct are "carefully constructed, is an important concern. Whether a code can influ- thoughtful, and packaged in a manner suggesting ence organizational change or successfully articulate they are intended to be read and understood." elements of a corporate culture depends upon its Others, they argued, were "haphazard in design, effectiveness as a message. Thus, it is important to superficial in content, and are presented in a ~shion examine these, codes in light of how their messages which clearly does not encourage serious reading or are communicated to employees. facilitate understanam~,1 • 0~, (p. 86). They attributed Ethical codes studies were reviewed using the these differences to managerial attitudes toward the following two questions: (1) In what manner are the codes. Regarding enforcement, they found that 83% codes communicated to employees? This involves of the organizations had administrative procedures examining language choice and tone, particularly to enforce the codes, but provided no further infor- regarding requests or demands for compliance. (2) mation about how enforcement might occur or To what extent is the code enforced and implica- implications for violating the codes. tions for violating the code discussed? This addresses Chatov (1980) categorized the content of 281 whether the code spells out what can happen if the corporate codes into 12 broad ethical dimensions. employee violates the code. What procedures will be Extortion, gifts, and kickbacks led the list of pro- invoked if the code is not followed? Will the hibited employee behaviors followed by conflict of employee be terminated? Answers to these questions interest, illegal political payments, violation of laws may reveal whether the intent behind the code is in general, use of insider information, and bribe U. self-protection, an attempt at promoting ethical Chatov concluded that U£. corporations agree upon leadership or management of a public image. which areas of employee misconduct cause the greatest concern. He does not discuss how codes are communicated except to note that in fbrbidding Review of ethical code studies extortion, gifts and conflict of interest, they used "strong language" to deter these behaviors and Corporate ethicat code studies to date have either observed that some codes "express moral standards focused on individual codes (i.e. Rogers and Swales, in very personal terms" exhorting employees not to 1990) or primarily used content analysis to address do anything they would not tell their spouse or not subject matter in the codes. White and Montgomery want to see in a newspaper (p. 24). (1980) performed a content analysis on 30 codes Cressey and Moore (1983) performed a content obtained from CEO's in the Fortune 1000 and a analysis on 119 corporate conduct codes on file ,with random samp!e of 1000 members of the Financial the Conference Board in New York City and found Executives Institute. They listed 36 possible subjects that conduct against the firms was discussed more ranging from a general statement of ethics and frequently than conduct on behalf of the firm. philosophy to insider trading. Conflict of interest IntegritT of financial books and records was dis- was the most commonly addressed subject (73%) cussed less than half the rime while about 50% of followed by employees' responsibility to comply codes mentioned relations with competitors. Again, 66 Bets)/Stevens conflict of interest was the most frequently men- how many employees would actually read an ethical tioned subject. Cressey and Moore found that the code which was hard going unless a signature was codes were mostly concerned with unethical be- required. Sanderson and Varner offer the only com- havior which might decrease profits and concluded ments on format. They observed that most codes that the codes exhibited a weak commitment to were presented" as small booklets ranging from 10 to social responsibility. Like White and Montgomery, 25 pages in length. they found that executives held diverse views on the Regarding the second question of code violations, ethical importance of most policy areas. Sanderson and Varner indicated that only about 50% Nearly all of the codes were grounded in some contained information as to how the code would be type of authority - "precepts, tenets, or principles" enforced. They observed that the language of en- (p. 59) seemingly to legitimize the code. There was forcement was general and nondescriptive in codes no discussion of how the codes were communicated, which discussed enforcement and seemed to con- or if a CEO personally endorsed the code; however, clude that these were not effective. they do focus extensively on employee sanctions for A Conference Board survey of 300 international violating the codes and conclude that the codes are senior executives revealed that 76% of the respond- weak in following through with reprisals. They ents indicated their firm had an ethical code (Beren- found that two thirds of the codes referred questions beim, 1987). While language and tone for the codes about the code's provisions to other sources which were not summarized in any way, comments from indicates the code was not seen as the final word. executives regarding distribution and communica- Regarding enforcement and implications for vio- tion are included in this study. lators, they tell us only that 75% of codes contained One ethics program director noted, "The person information on compliance and most codes were the individual works for is in the best position to enforced by upper management as opposed to convey this (ethical code) information. If they cannot management by personal integrity or by regulation do it, who can or will? It is the same situation as a through an outside agency. About 50% had an parent confronts with sex education" (p. 17). SLxty- affidavit of compliance or understanding which six percent of respondents said everyone in their required a signature. organization receives the code and 33% reported that Sanderson and Varner (1984) analyzed 39 codes middle and top managers receive it. from the top tabrtune 500 companies for content, Penalties of violating the code included termina- organization, purpose, enforceability, and format. tion (52%), suspension (30%), demotion (19%), proba- They found that 75% of content was related to tion (21%), appraisal comments (14%), and other complying with federal laws. Seventy percent of the penalties (11%). They do not discuss the manner in codes addressed conflict of interest, 72% dealt with which these codes are communicated to employees political contributions, and 59% addressed customers, or comment on language and tone. accuracy of accounting records and antitrust issues. Matthews (1987) performed the most extensive The authors noted that "most of the codes consist content analysis using 64 topics in 202 Fortune 500 mainly of rules based on laws" (p. 29). codes. She, like Cressey and Moore, categorized each Sanderson and Varner addressed the question of topic as "not discussed", "discussed", "discussed in how the codes are communicated by assessing the detail", or "emphasized." Results showed firms em- readability of codes and determining the intended phasized employee misconduct against the firm and reader. They concluded that most codes were ad- illegal activities and spent little time talking about dressed generically to "all employees" and suggested quality, product safety, and the environment. Mat- that firms have separate codes for management and thews attempted to correlate ethical codes with nonmanagement personnel. The wisdom of such a corporate misconduct (i.e. corporations accused of move is questionable as it invokes the concept of a illegal activity by federal agencies) and found only a double standard. weak link between ethical codes and corporate Using the Flesch test, 2 to measure reading diffi- behavior. She found that codes did not provide self- culty, they found that 22 codes were "veW difficult" regt~adon, demonstrate social responsibility', and and 16 were" "difficult." It seems fair to question promote a culture of ethical behavior. There is no Corporate Ethical Codes 67

discussion in this study about how codes are commu- codes3 This underscores some of the complexity- nicated. Regarding the second question of enforce- encountered in comparing these studies. "Employee ment, the vast majority of codes did not mention conduct" was not chosen as an umbrella category in specific reprisals such as reprimands, fines, demo- the studies cited although Chatov's "extortion, gifts, tion, or dismissal as penalties for illegaI behavior. and kickbacks" or White and Montgomery's "misuse About 25% discussed legal prosecution. Matthews of corporate assets" or Pitt and Groskaufmanis's does not focus on linguistical dimensions in the "misuse of confidential information" certainly refer codes. to employee conduct. Thus, when we try to compare Pitt and Groskaufmanis (1990) report the results results across studies, it becomes a difficult task and of two ethical code surveys conducted by their law some confusion is inevitable. firm) Like other studies cited in this paper, they do Except for reporting that 26% of European and not address whether the code has the personal British firms restrict the codes to specific categories endorsement of a CEO. However, 90% of their of employees, managers, or boards of directors, respondents in their 1987 survey 4 acknowledged Schlegelmilch and Langois do not discuss the manner having a code and 87% responded affirmatively to in which codes are communicated to emp?.oyees. the question asking if the code was "widely known" Code enforcement and implications for vioIations among employees. Interestingly, none of the other are also not discussed. studies asked this question. It addresses the extent to which the code has been communicated throughout the organization. If an ethical code is buried inside a Discussion policy book, it may not be read with great care by the employee. Also, if an organization has a code, yet Several rather interesting trends emerge from review- few employees are aware it exists, the responsibitiD~ ing these ethicai code studies. First, it is apparent that of the organization to fully and effectively com- firms are strongly concerned with self-protection; municate ethical policies may not be fulfilled. that is, conflict of interest is a common theme in all With regard to the second question of enforce- nearly all the studies. Firms seem primarily con- ment and violation implications, 68% of respondents cerned with employee misconduct wkich might reported they had terminated an employee for a damage the firm. Second, the codes seem preoccu- code violation and 11% noted that the code had pied with follovdng laws. While ethical codes should become an issue in litigation with the firm. Thirty- promote law-abiding behavior, it appears that they two percent asserted they had formal enforcement are preoccupied with law enforcement and self- measures for their code and 79% stated that the code defense and often do not rise above this plateau to was effective :in either ensuring or enforcing appro- successfully articulate the -values, beliefs, and pre- priate behavior. cepts of a desirable corporate culture. Ethical codes are not unique to the United States. White and Montgomery, Chatov, Cressey and Schlegelmilch and Langlois (1990) report that num- Moore, Sanderson and Varner, Chatov, and Siege- bers of European codes grew in the mid-1980's, but milch and Langlois all used content analysis to not to the extent found in the US. Their survey, of examine groups of codes. From White and Mont- 200 of the largest German, British, and French gomery we learn that the most frequently men- companies showed that the Germans have the tioned subjects were acts against the corporation. greatest number of corporate codes (51%) followed Little information is given about how- codes were by" the British (41%), while orfly 30% of French firms communicated to employees, save that a high per- surveyed have: codes. centage of key officials were required to sign them. In comparing these with American codes, they Language and tone are not discussed. Similarly, Cha- found that American codes discuss both government tov's content analysis provides information almost and customer relations more extensively than do exclusively on code subjects and little on how the European and British codes. Additionally they report codes are communicated. His comments on %trong almost all codes in their study address employee language" used in codes are particularly interesting, conduct in contrast with only 55% of American yet no additional details are given. Cressey and 68 Betsy Stevens

Moore focused primarily on code content rather Notes than addressing effectiveness or communication efficacy; Matthews performed an extensive content For a historical discussion of professional codes in the analysis on codes and provided a rich body of legal, medical and accounting professions, see Backof,J and information, yet language, tone, and how the codes Martin, C (1991). Historical Perspectives: Development of are communicated was not discussed. the Codes of Ethics in the Legal, Medical and Accounting Are corporate codes effective? We lack solid Professions.Journal ofBusiness Ethics 10, 99-110. ~- The Flesch Reading Ease Scale involves (1) averaging the evidence to support this notion. If, by effective, we number of words per sentence and multiplying that number mean corporations with codes behave more ethically, by 1.105 2) figuring the average number of syllables per only Matthews addressed this question writing ... word and multiplying by, 0.846 3) adding the two numbers "it cannot be concluded that codes of ethics demon- together, then subtracting the total from 206.835. This yields strate either (1) social responsibility, (2) a corporate a number between zero and one hundred. Zero is considered culture which promotes anti-criminal behavior pat- very difficult and 100 is consideredvery easy. terns or (3) self regulation" (p. 128). Are corporate 3 The law firm of Fried, Frank, Harris, Shriver and Jacob- codes transformational messages? Do they effect son, located in Washington, D. C., conducted the studies change in significant ways providing managers with referred to as the Fried, Frank sm-veys in the Pitt and support and guidance for negotiating sticky wickets? Groskaufmanis (1990) article. Are they communicated to employees in ways which '~ The Fried, Frank 1984 survey received 121 responses from the 500 companies they contacted (24.7%). In 1987, 150 of underscore the importance of ethical conduct and 414 companies responded (36.2%). encourage altruistic behavior. Evidence exists, at least 5 Slegemilch and Langlois compared their data with the with teenagers, that codes may alter behavior. A Foundation of the Southwestern Graduate School of Bank- recent USA Weekend survey reported that 29% of ing's 1980 "Study of Corporate Ethical Policy Statements." 13-18 years olds said they would cheat on a math test given the opportunity to view an "A" student's paper, however only 13% said the}, woMd do so if they had signed a school pledge not to cheat (Ansley References and McCleary, 1992). Rather than focusing so extensively on code Ansley, L. and K. McCleary: 1992, August 21-23, 'Ethics at content, perhaps we need to spend more time Work: Help Wanted', USA Weekend, 5. Becket, H. and D. Fritzche: t987, 'Business Ethics: A Cross- evaluating corporate codes as messages. Most of the Ctdtural Comparison of Manager's Attitudes', Journal of questions about how codes function as messages Business Ethics 6,289-295. remain unanswered. If ethical codes are to be studied Bennett, A.: 1988, July 15, 'Ethics Codes Spread Despite as messages intended to direct and motivate strong Skepticism', The Wall StreetJournal, 17. ethical behavior, we need to explore the more Berenbeim, tL E.: 1987, Corporate Ethics (Conference Board, exegetic aspects of these texts and analyze thdr New York). messages more rhetorically" rather than relying so Black, H. C.: 1983, Black's Law Dicth)nar), (abridged 5th heavily on content analysis. Studies focusing on the edition) (West, St. Paul). effectiveness of codes are needed. Do they work? Brooks, L. J.: 1989, 'Corporate Codes of Ethics', Journal of Also, are the codes communicated in meaningful Business Ethics 8, 117--119. ways? Are employees aware of their organ~ation's Chatov, R.: 1980, 'What Corporate Ethics Statements Say', ethical code and accepting of its guiding principles. California Management Review 22 (4), 20--29. Cressey, D. and C. A. Moore: 1983, 'Managerial Values and Studies done thus far have shed some light on the Corporate Codes of Ethics', California Management Review subjects corporations consider important and laid an 25, 53-77. important foundation for further work. Codes, as Frankel, M.: 1989, 'Professional Codes: Why, How, and expressions of values, are an integral part of cor- With What Input?',JournaI ofBusiness Ethics 8, 109--115. porate ethics. We need to focus more intensely on Heermance, L Edgar: 1924, Codes of Ethics (Free Press, the effectiveness of these messages and analyze how Burlington, Vt). codes use language to express values and beliefs. Langlois, C. C. and B. B. Schlegelmilch:1990, 'Do Corporate Codes of Ethics Reflect National Character? Evidence Corporate Ethical Codes 69

from Europe and the United States',Journal ofInternational Rogers, P. and J. Swales: i990, 'We the People? A~ Analysis Business Studies, fourth quarter, 519-539. of the Dana Corporation Policies Document', TheJournal Matthews M. C.: 1987, 'Codes of Ethics: Organizational ofBusiness Communication 27 (3), 293-313. Behavior and Misbehavior', Research in Corporate Social Sanderson R. and I. I. Varner: 1984, 'What's Wrong with Performance, vol. 9 (JAI Press Inc, Greenwich, CT), pp. Corporate Codes of Conduct?', Management Accounting 107-130. 66, 28--35. Pearce, J. A. and F. David: 1987, 'Corporate Mission State- White, B. J. and R. Montgomery: 1980, 'Corporate Codes of merits: The Bottom Line', Academy of Management Execu- Conduct', California Management Review, 13 (2). tive 1(2), 109-116. Pitt, H. L. and K. A. Groskaufmanis, t990, 'Minimizing &hool ~fBusiness Administl~tio< Corporate Civil and Criminal Liability: A Second Look at U~iversity ofMichigan, Corporate Codes of Conduct', The Georgetown Law Jour- Ann A&or ~kII, 48t08-!234, nai 78, 1559. U.S.A.