Can auditors in this case being dependent objective? Why? and How?

• OPINION •

The Impossibility of Auditor Independence

Max H. Bazerman • Kimberly P. Morgan • George F. Loewenstein

Audit failures rarely result jrom the

deliberate collusion of auditors

with clients. Insteady auditors may

find it psychologically impossible to

remain impartial and objective.

I n 1992, Phar-Mor, Inc., the largest discount drug- store chain in the United States, filed for bankrupt- cy court protection following discovery of one of

the largest business fraud and embezzlement schemes in U.S. history. Coopers & Lybrand, Phar-Mor's for- mer auditors, failed to detect inventory inflation and other financial manipulations that resulted in $985 million of earnings overstatement during a three-year period. A federal jury unanimously found Coopers & Lybrand liable to a group of investors on fraud charges. The attorney for one investor argued that "this sends a strong signal to the accounting community that in- vestors take very seriously the role of audited financial statements and rely on them for their integrity."'

The investors who successfully sued Coopers & Lybrand contended that Gregory Finerty, the Coopers & Lybrand partner in charge of the Phar-Mor audit, was "hungry for business because he had been passed over for additional profit-sharing in 1988 for failing to sell enough of the firm's services."' In 1989, Finerty began selling services to relatives and to associates of Phar-Mor's president and CEO (who has been sen- tenced to prison and fined for his part in the fraud). Critics claim that Finerty may have become too close

to client management to maintain the professional skepticism necessary to conduct an independent audit.

The Phar-Mor case is one of many in which audi- tors have been held accountable for certifying faulty financial statements. Investors in the Miniscribe Cor- poration maintained that auditors were at least par- tially responsible for the now-defunct company's falsi- fied financial statements; at least one jury agreed, holding the auditors liable to investors for $200 mil- lion. In the wake of the U.S. savings and loan crisis, audit firms faced a barrage of lawsuits, paying hun- dreds of millions of dollars in judgments and out-of- court settlements for their involvement in the finan- cial reporting process of savings and loan clients that eventually failed.

The accounting profession maintains that it is be- ing unfairly assaulted by plaintiffs looking for a conve- nient "deep pocket" from which to recover losses that may result from their own poor investment decisions. The investing and lending public, on the other hand, has become cynical about the accounting profession and its role in the financial reporting process. How could auditors not see that so many of their savings and loan clients were about to fail? How could a prominent auditing firm with a reputation for in- tegrity overlook such large misstatements in Phar-

Max H. Bazerman is the J. Jay Cerber Distinguished Professor of Dispute

Resolution and Organizations at the J. L. Kellogg Graduate School of

Management, Northwestern University. Kimberly P. Morgan is a certified

public accountant and a Ph. D. candidate at the Katz School of Business,

University of Pittsburgh. Ceorge F. Loewenstein is professor of economics,

department of social and decision sciences, Carnegie Mellon University.

SLOAN MANAGEMENT REVIEW/SUMMER 1997 BAZERMAN ET AL. 89

Mor's financial records? Critics of the profession sug- gest that auditor neglect and corruption may be re- sponsible. We argue, however, that only very rarely are audit failures the result of deliberate collusion of audi- tors with clients in issuing faulty financial statements. Instead, we maintain that audit failures are the natural product of the auditor-client relationship. Under cur- rent institutional arrangements, it is psychologically impossible for auditors to maintain their objectivity; cases of audit failure are inevitable, even with the most honest auditors.

Many professional roles call for impartial judg- ments. We expect judges to pass sentences that are free from racial prejudice, doctors to recommend treat- ments that are best for their patients rather than for their wallets, and teachers to put aside their personal feelings toward their students when grading their pa- pers and exams. However, research in each area has shown that judges, doctors, and teachers are biased by their own interests and prejudices. There is no claim that these professionals are corrupt, only that biased judgment prevents them from making purely impar- tial, objective decisions.

In no profession is impartiality more important than in auditing. Auditors provide information to share- holders and to other stakeholders that is vital to firms' public ownership. An auditor's failure to detect signifi- cant misrepresentations in a company's financial state- ments can lead not only to losses by individual in- vestors, but also to an overall decline of trust in capi- talist institutions. Like members of other professions, however, auditors often face challenges to their inde- pendence. Many challenges arise because auditors are hired, paid, and even fired by the organizations that they audit rather than by the people they ostensibly represent.

The accounting profession is sensitive to the poten- tial for bias in audits. The American Institute of Certi- fied Public Accountants (AICPA) states in its Code of Professional Ethics:

"In the performance of any professional service, a member shall maintain integrity, shall be free of con- fiicts of interest, and shall not knowingly misrepresent facts or subordinate his or her judgment to others. . . . Members should accept the obligation to act in a way that will serve the public interest, honor the public trust, and demonstrate commitment to professionalism."'

The AICPA thus acknowledges the pressures on the integrity and objectivity of the auditor but con- tends that auditors can achieve a level of indepen- dence such that users can rely on audited financial statements as unbiased assessments of the reporting companies' positions.

The courts share the view that auditors must act in the interests of external users of financial statements and also assume implicitly that it is possible for them to do so. Former Chief Justice of the U.S. Supreme Court Warren Burger described the role of the audi- tor, in a 1984 opinion:

"The independent auditor assumes a public respon- sibility transcending any employment relationship with the client. The independent public accountant performing this special function owes ultimate alle- giance to the company's creditors and stockholders, as well as to [the] investing public. This 'public watchdog' function demands that the accountant maintain total independence from the client at all times and requires complete fidelity to the public trust."''

How realistic is the assumption that auditors — even those of high integrity — can provide impartial judgments that respond to the interests of creditors, stockholders, and the general public, rather than to the interests of the companies that hire them? Psychological research points to an inescapable conclusion: such im- partiality is impossible under current institutional ar- rangements.

In this paper, we review the structure of the audit- ing system in the United States. We then discuss psy- chological research that points to the impossibility of auditor independence. Next, we describe contempo- rary aspects of the auditing profession that exacerbate independence issues. Finally, we enumerate some po- tential solutions.

The Structure of the Auditing Relationship

Stockholders, potential stockholders, financial advis- ers, underwriters, regulators, lending institutions, and businesses that extend credit are among the users of au- dited financial statements. These users make decisions based on the information in the statements, which are prepared and issued by a company's management. Man- agement typically has incentives to present the compa- ny's financial position in the best possible light. Com-

90 BAZERMAN ET AL. SLOAN MANAGEMENT REVIEW/SUMMER 1997

pensation plans ofien tie management's pay to report- ed financial results. Management might consider fi- nancial statements to be public relations documents, instrumental as a means to infiuence external users. Thus management may be motivated to present fi- nancial information that is overly optimistic, mislead- ing, or false.'

To obtain some assurance that financial statements presented by management are valid, reliable, and com- plete, external users look to the report of the company's independent auditor. An audit is an examination of a company's financial statements to give an opinion on whether the information in those statements is reliable and is prepared and presented in accordance with gen- erally accepted accounting principles (GAAP). The company presents the auditor's opinion of its financial statement in a report along with the financial state- ment. An auditor's unqualified opinion states that, based on an examination in accordance with profes- sional auditing standards, the accompanying financial statement "fairly presents" the company's financial position and results of operation in accordance with CAAE

An unqualified audit report lends credence to the company's financial statement, providing external users with reasonable assurance that the information is reliable, consistent, and comparable across all peri- ods covered by the report.'̂ If the audit opinion is to provide the desired degree of assurance, the auditor

f the audit opinion is to provide the desired degree of assurance, the auditor must be able to form and

express an opinion without bias.

must be able to form and express an opinion with- out bias. Accountants traditionally use the term "in- dependence" to refer to an auditor's ability to make audit judgments objectively, "free and clear of any infiuence that other parties or factors might bring to bear."^

Although the auditor examines a company's fi- nancial statement on behalf of external users, the management of the company that prepared and is-

sued the statement under examination hires and pays the auditor. The company under audit and the indi- viduals who manage that company are the "client." Clients may hire and fire auditors at will. In addition to economic incentives that may bias an auditor's judg- ment in favor of the client who pays the fees, the rela- tionship that auditing firms strive to develop with the clients may add to the auditors' psychological diffi- culty to make truly independent judgments.

The Psychology of the Impossibility of Independence

Calls for auditor independence, such as the AICPA's (quoted above), implicidy adopt a naive, unrealistic model of auditor psychology. This model assumes that auditors form unbiased judgments but that the potential for bias arises at the point of reporting those judgments. Expressed differently, auditor bias, to the extent that it occurs, is viewed as a form of deliberate misrepresentation. The assumption of deliberative- ness is important because it implies that any tendency toward bias can potentially be rectified by moral sua- sion and/or the threat of sanctions.

Psychological research shows that this model is un- realistic. Bias typically enters unconsciously and unin- tentionally at the stage of making judgments, not of reporting on them, although there may be some de- liberate misreporting as well. When people are called on to make impartial judgments, those judgments are likely to be unconsciously and powerfully biased in a manner that is commensurate with the judge's self- interest. Psychologists call this the self-serving bias.** When presented with identical information, individ- ual perceptions of a situation differ dramatically de- pending on one's role in the situation. People first de- termine their preference for a certain outcome on the basis of self-interest and then justify this preference on the basis of fairness by changing the importance of attributes affecting what is fair.'̂ Thus the problem lies not in our desire to be unfair, but in our inability to interpret information in an unbiased manner. Self- serving biases exist because humans are imperfect in- formation processors. One of the most important non- objective infiuences on information processing is self- interest. People tend to confuse what is personally ben- eficial with what is fair or moral.'"

SLOAN MANAGEMENT REVIEW/SUMMER 1997 BAZERMAN ETAL. 91

In a series of experiments examining the self-serving bias, which we think represent a close analogy to the sit- uation in auditing, Loewenstein et al. presented partici- pants with diverse materials (depositions, police reports, doctors' reports, and so on) from a lawstiit that resulted after a collision between a car and a motorcycle." Par- ticipants were assigned the role of plaintifF or defendant and attempted to negotiate a settlement. If unable to do so, they paid substantial penalties and were told that an impartial judge, who had earlier read the same case ma- terials and reached a judgment, would determine the amount paid by the plaintifF to the defendant. Before they negotiated, participants were asked to predict the judge's ruling. They were told that the estimate would not be commtinicated to the other party and would not affect the judge's decision (which had already been made). Nevertheless, plaintiffs' predictions oFthe judge's award amount were substantially higher than those of defendants, and the degree of discrepancy between plaintifF and deFendant strongly predicted whether they setded the case (as opposed to relying on the judge's de- cision).

In follow-up experiments, the same researchers at- tempted to reduce the magnitude of the bias. They paid participants for accurately predicting the judge's ruling and had them write an essay arguing the other side's viewpoint. Neither intervention had a measur- able effect. Participants consistently believed that the judge would perceive judgments that were in their own material interest as fair. The researchers also at- tempted to reduce the magnitude of the self-serving bias by describing it to participants in detail and hav- ing them take a test to ensure that they understood the description. The experimental intervention was successful insofar as participants became convinced that their negotiating opponent would be highly bi- ased, but participants believed that they themselves would not succumb to the bias. The fact that partici- pants were unable to rid themselves of the bias when rewarded for doing so and their belief that they were not subject to bias both demonstrate clearly that the self-serving bias is unconscious and not deliberate.

Other findings from the same experiments point to a likely psychological mechanism underlying the self-serving bias. Researchers gave participants eight arguments favoring the side they had been assigned (plaintiff or defendant) and eight arguments favoring

the other side. They asked them to rate the impor- tance of the arguments as perceived by "a neutral third party." Participants tended to view arguments supporting their own position as more convincing than those supporting the other side, suggesting that the bias operates by distorting interpretation of evi- dence. Consistent with this interpretation, when the

W hen presented with identical information, individual perceptions

of a situation differ dramatically depending on one's role in the

situation.

parties were assigned their roles (plaintiff or defen- dant) only after they read the case materials, the mag- nitude of the bias was substantially reduced and al- most all the plaintifF-defendant pairs reached rapid agreement on damages.

In the studies we just reviewed, participants received no actual pecuniary benefit by reaching biased judg- ments; the only incentive For misrepresentation came firom the subjects' identification with their roles. More- over, as we mentioned, in many experiments, there were explicit monetary incentives For arriving at unbi- ased judgments. Nevertheless, in approximately six studies with hundreds oF subjects, the bias was consis- tently large.

The selF-serving bias is exacerbated by a number oF characteristics of the auditing relationship. First, the people who will be hurt by any misrepresentation are "statistical" — an auditor cannot identify them at the time the decision is made.'- People tend to be far less concerned about imposing harm on statistical victims than on known victims. Many people might lose a small amount of money, but it isn't clear who will. In contrast, the auditor is likely to be well acquainted with the people within the client firm who would be hurt by a negative opinion on the audit. Second, the negative consequences of a negative opinion are likely to be immediate — loss of a client's friendship, po- tential loss of the contract, and possible unemploy-

92 BAZERMAN ET AL. SLOAN MANAGEMENT REVIEW/SUMMER 1997

ment — whereas the effects of a positive report when a negative report was appropriate are likely to be down- played because they are delayed.'̂ Third, auditors form an ongoing relationship with the organizations they audit, and any deterioration in the audited company is likely to unfold gradually."* Auditors may unknowingly adapt to small imperfections in the company's financial practices. Fourth, financial reporting standards are often fiexible or ambiguous, so it may be easy for an auditor to rationalize a judgment that is consistent with self- interest rather than the interests of external users. Fifth, people possess a remarkable ability to mislead them- selves about the nature of trade-offs, to rationalize to themselves and to others the accuracy of their biased judgments.

In sum, auditors' judgments are likely to be biased in favor of their own and their client's interests. This bias occurs indirecdy as a result of selective sifting and integrating audit information. As a result, the bias is likely to be unintentional and impervious to moral suasion or the threat of delayed and probabilistic sanctions, which are likely to seem quite remote.

Issues That Exacerbate the Problem of Independence

The tensions regarding independence have existed for decades. Recent developments in auditing, however, threaten to exacerbate the problem. First, the auditing environment has become far more competitive in the 1990s than in earlier decades, increasing the conse- quences of losing a client and the incentives for main- taining good client relations. Previously, junior auditors were typically billed at a ratio of four times the cost of the employee. Now it is common for this ratio to fall below two and even below one when another firm tries to "steal" an account. In highly competitive markets, accounting firms ofien engage in lowballing — accept- ing unprofitable audit fees in the initial year or two in order to "buy" the business. When auditors accept drastically discounted fees, they are likely to be highly motivated to retain the client For several years. In the past, approaching another auditor's client was consid- ered inappropriate. Today, slow economic growth has made it more difficult to "grow the btjsiness," so luring accounts away From competitors is a mark oFsuccess.

Intensified competition has occurred not only be-

tween accounting firms but also within them. In con- trast to the gentlemanly nature oF the auditing busi- ness twenty-five years ago, contemporary auditing firms expect partners to generate significant revenue, and Failure to meet these expectations Frequently leads to "retirement." Being a partner in this industry has financial implications for profit sharing but does not ensure employment. Both dimensions of in- creased competitiveness are likely to focus auditor at- tention on immediate profits and intensify the conse- quences of losing a client due to a negative audit.

Second, the leading auditors are entities within larger partnerships that include tax and (rapidly grow- ing) consulting practices. Auditing is becoming less important to the overall profitability of the leading ac- counting firms. In many cases, a Firm's audit client is also a consulting client, with the consulting compo- nent oFthe relationship being Far more profitable than the audit. So an uiiFavorable opinion risks not only the audit but, potentially, the consulting relationship as well. In the past, firms have emphasized the inde- pendence oF their three components (audit, tax, and constdting). Firms have been changing their structural Form, however, to better integrate services within in- dustries and For specific clients. InFormation From one part oF the relationship with a client can help the ac- counting firm with another component. But the risk to independence is also increased. Imagine how diffi- cult it is For an auditor whose firm has been providing consulting services to a company to submit a quali- fied report. Simultaneously playing consultant and watchdog further conflises the issue oF whom the au- ditors are accountable to and working For.

Conclusion

What explains the current wave oF lawsuits against auditors? Past critics oF the auditing proFession have Focused on the obvious conflict oF fulfilling responsi- bility to external users versus the financial benefits oF pleasing the client. This confiict is typically viewed as a moral trade-ofF that auditors Face. The larger prob- lem, however, is not with the auditors' morality, but with limitations in the way that they process inFor- mation. Thus independence remains a problem For even the most moral, honest auditor. Despite the au- ditors' best efForts to place the external users' interests

SLOAN MANAGEMENT REVIEW/SUMMER 1997 BAZERMAN ET AL. 93

above the client's and to maintain objectivity, they may be unable to overcome cognitive or psychologi- cal biases that make them arrive at marginal decisions in the client's favor. As we cited earlier, the AICPA states that the auditor "shall not knowingly misrepre- sent facts or subordinate his or her judgment." The larger problem facing society is that there is good rea- son to believe that auditors will unknowingly misrep- resent facts and will unknowingly subordinate their judgment due to cognitive limitations.

While audits are done for external tisers, the negoti- ated relationship between the auditor and the client creates them. Both the auditor and the client benefit From auditors' selF-serving bias. We believe that the au- diting proFession and external users oF Financial state- ments should actively seek flmdamental changes in the current structure oFthe auditing relationship. Observers oF the proFession have suggested various possibilities, such as prohibiting a firm that conducts a company's audit from simultaneously providing other services for that client, prohibiting audit Firms From providing any related services, having external (perhaps governmen- tal) bodies appoint auditors or set Fee structures, requir- ing companies to periodically change auditors, increas- ing oversight oF auditing practices, or, the most drastic, having governmental agencies rather than the private sector conduct audits.

While we do not know that any oF these sugges- tions would be optimal, we believe we have made a convincing case For reForm oF the current auditing re- lationship. External users pay a huge price For the flaws in the current structure oFthe audit flinction, as do the accounting Firms devoting huge resources to deFending themselves against what they see as an "epidemic" oF litigation. Much like the Federal deficit, these problems are mounting and will get worse if not addressed. Ideally, the interested parties will deal with these problems before the government does. •

References 1. Adapted from M. Murray, "Coopers & Lybrand Is Found Liable

by Jury to Investors," Wall Street Journal, 15 February 1996, p. A-8.

2. Adapted from M. Pitz, "J'-'O' Finds Phar-Mor s Auditors Negli-

gent," Pittsburgh Post-Cazette, 15 February 1996, pp. A1-A6.

3. American Institute of Certified Public Accountants Code of Profes-

sional Ethics, 1988.

4. W. Burger, U.S. Supreme Court: 1984, United States v. Arthur

Young & Co., US Supreme Court Reports, IG April 1984, 79 L Ed

2d, 826-838.

5. J.C. Robertson, /!W/>/>/g-(Homewood, Illinois: Irwin, 1990).

6. E. Waples and M.K. Shaub, "Establishing an Ethic of Accounting,"

Joumalof Business Ethics, volume 10, 1991, pp. 385-393.

7. C.E. Jordan and J.G. Johnston, "Auditor s Independence: A Pro-

posal to the Profession and the Public," The Woman CPA, volume 49,

July 1987, pp. 3-9.

8. D.M. Messick and K.P. Sentis, "Fairness and Preference," Journal

of Experimental Social Psychologf, volume 15, 1979, pp. AMi-A'iA.

9. K.A. Diekmann, S.M. Samuels, L. Ross, and M . H . Bazerman,

"Self-interest and Fairness in Problems of Resource Allocation,"/O;»7M/

of Personality and Social Psychology (in press).

10. D.M. Messick, "Equality, Fairness, and Social Conflict," Social

Justice Research, vo\un\e 8, 1995, pp. 153-173; and

D.M. Messick and A.E. Tenbrunsel, eds.. Codes of Conduct {New

York: Russell Sage Foundation, 1996).

11. L. Thompson and C. Loewenstein, "Egocentric Interpretations of

Fairness and Interpersonal Conflict," Organizational Behavior and

Human Decision Processes, volume 51,1992, pp. 176-197;

C. Loewenstein, S. IssacharofF, C. Camerer, and L. Babcock, "Self-

Serving Assessments of Fairness and Pretrial Bargaining," Journal of

Legal Studies, \oV\nvtll, 1993, pp. 135-159;

L. Babcock, G. Loewenstein, S. Issacharoff, and C. Camerer, "Biased

Judgments of Fairness in Bargaining," American Economic Review, vol-

ume 85, December 1995, pp. 1337-1342.

12. K. Jenni and G. Loewenstein, "Explaining the Identifiable Victim

Effect," Journal of Risk and Uncertainty (forthcoming, 1997);

D.M. Messick, and M.H. Bazerman, "Ethical Leadership and the

Psychology of Decision Making," Sloan Management Review, volume

37, Winter 1996, pp. 9-22; and

L. Babcock and G. Loewenstein, "Explaining Bargaining Impasse:

The Role of Self-Serving Biases," Journal of Economic Perspectives (in

press).

13. SeeG. Loewenstein andj. Elster, Choiceove>- 7/>H(?(New York: Russell

Sage Foundation Press, 1992);

G. Loewenstein, "Behavioral Decision Theory and Business Ethics:

Skewed Ttade-offs between Self and Other," in Messick and Tenbrunsel

(1996).

14. See J.C. Corless, R.W. Bardett, and R.J. Seglund, "Psychological

Factors Affecting Auditor Independence," The Ohio CPA Journal, vol-

ume 49, Spring 1990, pp. 5-9.

Reprint 3848

94 BAZEHMAN ET AI,. SLOAN MANAGEMENT REVIKW/SUMMER 1997