Corporate Transparency:Code of Ethics Disclosures

By Richard A. Bernardi and Catherine C. LaCross

E-mail Story Print Story

The following research examines the responses by 97 large U.S. companies from the global Fortune 500 in making their codes of ethics readily available after the passage of SOA.

Reform Responses

After the demise of Andersen and the collapse of Enron shook corporate America, stockholders adjusted their decision-making processes to reflect their concerns over unethical business practices. Recent corporate scandals have increased the demands of stakeholders that financial reporting, as a 2003 Pricewater-houseCoopers white paper put it, go “beyond current requirements in order to achieve a deeper level of transparency in corporate reporting.”

In the mid-1980s, defense contractor Sundstrand was fined $227 million for ethical violations; Sundstrand’s stock fell 25% and its image was tarnished by the incident. Sundstrand was successful in changing its ethical culture through internal initiatives, many of which parallel the ethical growth process (or ethical spectrum) outlined by Rossouw and van Vuuren in “Modes of Managing Morality: A Descriptive Model of Strategies for Managing Ethics” [Journal of Business Ethics, 46 (4), 2003]. This spectrum includes an immoral mode, a reactive mode, a compliance mode, an integrity mode, and a totally aligned organization mode. In the immoral mode, the focus of a company is the “bottom line,” which overlooks and alienates the company’s stakeholders. As a result, stakeholders convey their frustrated expectations, and corporations try to protect themselves against the dangers of unethical behavior (the reactive mode).

A company’s desire to have a good ethical reputation characterizes the compliance mode; the focus is a rule-based approach to ethics. In the integrity mode, the firm becomes proactive in the promotion of ethical behavior and engages all of its stakeholders; the company begins to “walk the ethics talk.” The integrity mode mirrors SEC Chairman Donaldson’s conviction that corporate America faces an environment with responsibilities that:

[G]o beyond mere adherence or conformity to new dictates and laws of Sarbanes-Oxley: responsibilities that rest at the very heart of their obligation to create a corporate culture of transparency and accountability [emphasis added].

Legitimacy Theory

During the 1960s and 1970s, the public became increasingly aware of the “adverse consequences of corporate growth.” Companies introduced codes to attest to their ethical awareness and behavior, amidst an array of business scandals during the 1970s and 1980s that raised concerns about corporations acting in the best interests of society. Legitimacy theory arose from these concerns, and it implied a social contract that:

[A]n institution must constantly meet the twin tests of legitimacy and relevance by demonstrating that society requires its services and that the groups benefiting from its rewards have society’s approval [A.D. Shocker and S.P. Sethi, “Approach to Incorporating Social Preferences in Developing Corporate Action Strategies,” in S.P. Sethi (ed.), The Unstable Ground: Corporate Social Policy in a Dynamic Society, 1974].

Patten [“Intra-Industry Environmental Disclosures in Response to the Alaskan Oil Spill: A Note on Legitimacy Theory,” Accounting Organizations and Society, 17 (5), 1992] examined the petroleum industry’s reaction to the Exxon Valdez incident. If the incident had ramifications only for Exxon, then Exxon alone would have had to publicly justify its continued legitimacy by demonstrating its increased awareness for caution in the future. Exxon included substantial coverage (3.5 pages) of the incident and its cleanup efforts in its 1989 annual report. Exxon also included 2.5 pages of environmental disclosures in the same report. The six pages of coverage in its 1989 annual report was 10 times the coverage of environmental issues provided in its 1988 annual report.

On the other hand, if the Valdez spill incident threatened the legitimacy of the seven oil companies in the Alyeska consortium, then there would have been a pattern of similar disclosures across the industry. Patten found exactly such an increased level of transparency about environmental issues in the 1989 annual reports throughout the oil industry, and especially for those in the Alyeska consortium.

Accounting Firms and Legitimacy Theory

It was not just Enron’s and WorldCom’s legitimacy that was challenged by the public’s growing concern, but also that of public company auditors. Although the public’s focus was mostly on Andersen, because of its connection with Enron and WorldCom, the other large auditors and the AICPA did not escape questions about the legitimacy of the accounting and auditing process. Both the Senate and the House held hearings about the scandals, newspapers followed them closely, and Congress passed groundbreaking legislation.

In the aftermath of SOA, did the clients of the large accounting firms respond uniformly to challenges to accounting and auditing legitimacy? Or, more specifically, did the former clients of Andersen make extra efforts to address their legitimacy, similar to the oil companies in the Valdez case?

Government Regulation and Ethics Disclosures

Ruhnka and Boerstler [“Governmental Incentives for Corporate Self-regulation,” Journal of Business Ethics, 17 (3), 1998] provide evidence that government regulations can affect a change in codes of ethics. The activity in enacting or updating codes of conduct by the Fortune 500 industrial and service companies from 1960 through 1994 (Exhibit 1) shows very limited activity before 1975, followed by a significant increase in the level of activity in 1976—coinciding with the amnesty offered for voluntary disclosure of illegal actions—and a resulting enactment of codes of ethics as a result of the Foreign Corrupt Practices Act (FCPA) of 1977. Afterward, the activity level decreased to its earlier rate.

Even though SOA does not require a complex code of ethics for corporations, it is the first piece of legislation that requires some kind of ethics code. Nonetheless, legitimacy theory would suggest that companies and their auditors would adopt new efforts to demonstrate their commitment to ethical codes in the wake of the legislation. To determine the status of corporate codes of ethics, Deloitte & Touche sent a questionnaire to the directors of the top 4,000 publicly traded companies in late July 2003, published as “Business Ethics and Compliance in the Sarbanes-Oxley Era.” Deloitte & Touche reported that 90% of the responding companies mentioned outside stakeholders (i.e., shareholders, suppliers, customers) in their codes, but only 52% distribute their codes to these stakeholders. The report also notes that electronic distribution of codes of ethics—for example, posting on a company website—is a cost-effective mechanism for distribution.

Industry Specialization and Ethics Disclosures

Audit firms have indicated that industry specialization is a goal they pursue. For example, in the Price Waterhouse and Cooper & Lybrand merger, the parties claimed that their specializations were complementary. On the other hand, the proposed merger of KPMG and Ernst & Young was called off because of what the firms described as incompatible industry mixes.

Hogan and Jeter found that auditor industry specialization significantly increased between 1975 and 1993 [C.E. Hogan and D.C. Jeter, “Industry Specialization by Auditors,” Auditing: A Journal of Practice & Theory (Spring): 1–17, 1999]. Specialization complicates a comparison of disclosures by auditors’ clients, because certain industries do not make disclosures that are common in other industries. On the other hand, the outcry after Enron and WorldCom and the enactment of the Sarbanes-Oxley Act was aimed at corporate America rather than a specific industry.

Results

The authors visited the websites of the 97 sample companies and searched for disclosure of their corporate code of ethics. Emphasizing ethics was defined as disclosure of its corporate code of ethics within two levels of the website homepage (e.g., an “About Our Company” link). For commercial-oriented sites, the homepage was considered the primary corporate information page.

To be classified as a “code of ethics,” a document did not need to specifically contain that phrase. Titles such as “code of conduct” or “company values” were equally acceptable, as long as each contained specific information concerning unacceptable behavior (e.g., conflict of interest or bribery). Documents that claimed to be a code of ethics, but listed general values instead of specific actions, were not included as an actual code. The analysis collected data at four dates: July 2002, January 2003, March 2003, and July 2003.

Analysis

The data in Panel A of Exhibit 2 depict the increases in the emphasis that companies gave to ethics codes, presented by auditor. While legitimacy theory would predict that Andersen’s former clients would be more likely to have made their codes of ethics readily available, the data actually indicate the opposite for two of the dates and no difference for the other two dates. None of the former clients of Andersen disclosed ethics policies on their websites in July 2002.

An analysis of the other three observation dates indicates that the former clients of Andersen were less likely to include data about their codes of ethics on their corporate websites than were clients audited by PricewaterhouseCoopers (13.3% versus 45.2%; p = .046) for the March 2003 observation. The differences were not significant for either the January 2003 or the July 2003 observations. The data also indicate that Ernst & Young’s clients’ transparency rate is closest to Andersen’s for the last three observations.

The Big Four and Disclosure

With regard to disclosure after SOA, Ernst & Young’s clients had the lowest rate of ethics transparency. Exhibit 3, Panel A, shows the disclosure rate by auditor. The data in Exhibit 3, Panel B, indicate no statistical significant differences for the first three dates between the data for Ernst & Young and the other three firms. At the July 2003 date, however, Deloitte & Touche (51.9%; p = 0.053) and Pricewaterhouse-Coopers (51.4%; p = 0.047) are significantly above Ernst & Young’s clients’ transparency rate (22.2%). Additionally, KPMG’s clients’ transparency rate (47.1%) is close to significance (p = 0.129).

An interesting point about the data in Exhibit 3, Panel B, is the progression of levels of significance over the four observations, which is exactly opposite to what was observed in the Valdez case. For the July 2002 observation, the probability of there being a difference between the transparency rate for Ernst & Young’s clients and the rate of the other three firms’ clients is not significant (p = 0.857), but it declines over time to p = 0.036 in July 2003. This trend is attributable to the significant increase in KPMG’s clients’ disclosure rate after January 2003.

Industry Comparison

The apparent differences between auditors’ client disclosure rates could be the result of industry concentrations. To examine this potential, the sample was stratified by the two-digit standard industrial code (SIC) for those industries with four or more representatives.

The data for the clients in industries in which Ernst & Young specializes were used to develop weighted average expected disclosure rates. The industry data in Exhibit 4 agree with Hogan and Jeter’s data on industry concentrations. Additionally, except for those involving Andersen, there were no auditor switches during the period of the research.

The data do not support differences caused by industry specialization, but they do show differences between Ernst & Young’s clients’ expected and observed frequencies for the observations of March 2003 (expected 37.5%; observed 20%) and July 2003 (expected 44%; observed 30%). The observed rate of 30% for July 2003 in the selected industries is higher than the overall rate of 22.2%. The difference between auditors’ clients’ rates of ethics transparency in this sample of 97 companies cannot be attributed to industry specialization. There could be any number of other reasons for the data; it is beyond the scope of this article to speculate as to the cause of the clients’ different disclosure rates.

Conclusions

In the post-SOA environment, it is not surprising that corporations are more sensitive to the public’s expectation that they provide transparency about their values. One might expect that Andersen’s former clients would be extremely concerned with their image and would make disclosures supporting their commitment to ethics in reporting and social responsibility in general. However, those former Andersen clients in this sample of 97 large U.S. companies have less aggressive website disclosures over the same time than do clients of other auditors.