Florida State University Libraries

Florida State University Libraries Electronic Theses, Treatises and Dissertations The Graduate School 2004 Partner Influence, Team Brainstorming, and Fraud Risk Assessment: Some Implications of SAS No. 99 Tina Daly Carpenter Follow this and additional works at the FSU Digital Library. For more information, please contact lib-ir@fsu.edu

THE FLORIDA STATE UNIVERSITY COLLEGE OF BUSINESS PARTNER INFLUENCE, TEAM BRAINSTORMING, AND FRAUD RISK ASSESSMENT: SOME IMPLICATIONS OF SAS NO. 99 By TINA DALY CARPENTER A Dissertation submitted to the Department of Accounting in partial fulfillment of the requirements for the degree of Doctor of Philosophy Degree Awarded: Spring Semester, 2004

The members of the Committee approve the dissertation of Tina Daly Carpenter defended on March 26, 2004. Jane L. Reimers Professor Co-Directing Dissertation M. G. Fennema Professor Co-Directing Dissertation Neil H. Charness Committee Member Gregory J. Gerard Committee Member Approved: William A. Hillison Committee Member Melvin T. Stith, Dean, College of Business The Office of Graduate Studies has verified and approved the above named committee members. ii

For Don and me iii

ACKNOWLEDGEMENTS I would like to thank the members of my dissertation committee, Jane Reimers (Chair), Bud Fennema, Greg Gerard, Bill Hillison, and Neil Charness. Each of you has given me invaluable guidance and support throughout this project and my Ph.D. program. Special thanks to Jane Reimers who has been an amazing role model, and who has believed in me from the very beginning. I am grateful for the financial assistance provided by the Florida State University Research Foundation and to Ernst &Young LLP, Deloitte & Touche LLP, KPMG LLP, and PricewaterhouseCoopers LLP for their participation in my research. I would like to thank faculty members, colleagues and my friends at Florida State University. I have learned so much from all of you. I especially thank Lisa Victoravich for her help and for choosing me as her mentor. I would like to thank Arianna Pinello for the experiences we have shared and the friendship that will last beyond our days in the doctoral program. I would also like to thank my family, my parents, Mary Lou and Ray, my brother Ryan and my grandparents, Ella and Dick. You have each helped me in your own special way. Most importantly, I would like to thank my husband, Don, whose love, kindness and continuous encouragement has pushed me to be the very best that I could be. I love you. iv

TABLE OF CONTENTS List of Tables... vii List of Figures... viii List of Appendices... ix Abstract...x 1. INTRODUCTION...1 1.1 Overview of Research Question...1 1.2 Contribution of Research...1 1.3 Overview of Method and Results...3 1.4 Organization of Dissertation...3 2. BACKGROUND AND HYPOTHESES DEVELOPMENT...4 2.1 SAS No. 99...5 2.2 Audit Teams...5 2.2.1 Individual Decision Making: The Auditor...6 2.2.2 Multi-Person Decision Making: The Audit Team...7 2.3 Fraud...10 2.4 Brainstorming...11 2.5 Accountability...12 2.6 Hypotheses Development...14 2.6.1 SAS No. 99, Brainstorming and Accountability...14 2.6.2 Brainstorming Teams versus Individuals: Quality of Judgment.15 2.6.3 Brainstorming Teams versus Individuals: Number of Ideas Generated...16 2.6.4 Improvements in an Individual s Quality of Judgment as a Result of Brainstorming...17 2.6.5 Professional Skepticism...18 2.7 Summary of Chapter 2...19 v

3. EXPERIMENT AND METHODOLOGY...20 3.1 Participants...20 3.2 Research Design...20 3.2.1 Phase I...22 3.2.2 Case Materials and Procedures...23 3.2.3 Phase II...24 3.2.4 Phase III...25 3.3 Summary of Chapter 3...25 4. RESULTS...26 4.1 Manipulation Check...26 4.2 Tests of Hypotheses...26 4.2.1 Hypothesis 1...26 4.2.2 Hypothesis 2...27 4.2.3 Hypothesis 3...29 4.2.4 Hypothesis 4...30 4.2.5 Hypothesis 5...33 4.3 Summary of Chapter 4...39 5. CONCLUSION...40 5.1 Discussion of Results...40 5.2 Contributions...41 5.3 Limitations and Suggestions for Future Research...42 APPENDICES...44 REFERENCES...115 BIOGRAPHICAL SKETCH...124 vi

LIST OF TABLES 1. Amount of Time Spent Brainstorming by the Audit Teams...27 2. Mean (Standard Deviation) Fraud Risk Assessments by Treatment for Individual Auditors, Nominal Group and Interacting Audit Team...29 3. Mean (Standard Deviation) Number of Unique Ideas Generated by Treatment for Individual Auditors, Nominal Group and Interacting Audit Team...30 4. Individual Fraud Risk Assessment Shifts...31 5. Individual Overall Professional Skepticism Shifts...35 6. Individual Professional Skepticism Subscale Shifts...36 vii

LIST OF FIGURES 1. Diagram of Various Literature Streams Reviewed...4 2. Overview of the Phases of the Experiment...21 3. Comparison of Individual Auditors Fraud Risk Assessments...32 viii

LIST OF APPENDICES A. Experimental Instrument...44 B. Financial Statement Case Materials...69 C. Example of Experiment Transcriptions and Coding...112 D. Human Subjects Committee Approval...113 ix

ABSTRACT Brainstorming sessions are now a requirement on each audit per Statement on Auditing Standards (SAS) No. 99, Consideration of Fraud in a Financial Statement Audit, but concerns have been raised about their effectiveness in helping auditors better detect fraud. Standard setters recognize the potential for partners of the audit team to dampen the effectiveness of brainstorming by their influence on the brainstorming session, particularly when they stress efficiency over effectiveness. Additionally, audit team interaction has been shown to improve auditors judgments, but interaction among individuals brainstorming has been shown to diminish the overall number of ideas generated. In an experiment, I test partner influence on a brainstorming session and the effects of this brainstorming session on auditor fraud judgments. I find that partners influence the amount of time spent brainstorming by the audit team. The results also indicate that, although brainstorming among the audit team does reduce the number of unique fraud ideas identified, it increases auditors fraud risk assessments and questioning mind, potentially improving their ability to assess the risk of fraud. Keywords: Fraud; brainstorming; audit teams; accountability; risk assessments; professional skepticism Data availability: Contact the author x

CHAPTER 1 INTRODUCTION 1.1 Overview of Research Question The Auditing Standards Board (ASB) has issued a new auditing standard, Statement on Auditing Standards (SAS) No. 99, Consideration of Fraud in a Financial Statement Audit, to supercede SAS No. 82 (AICPA 2002), highlighting the new requirement for audit teams to conduct a brainstorming session, in which auditors will exchange ideas about where an entity s financial statements may be susceptible to fraud. This standard suggests that brainstorming will aid auditors in detecting fraud. Failure to detect existing fraud in financial statements has proved quite costly to auditing firms (Bonner, Palmrose and Young 1998). The purpose of this dissertation is to examine whether an audit team brainstorming session, now required on every audit, influences auditors fraud judgments, including auditors fraud risk assessments, fraud ideas identified and professional skepticism. It also seeks to determine whether this brainstorming session will be influenced by accountability of the audit team members (i.e., staff, senior and manager) to the audit partner. 1.2 Contribution of Research Investigating the effects of brainstorming among the audit team on auditor fraud judgments and the effects of partners influences on the brainstorming session is important for several reasons. First, corporate scandals involving fraud at companies, including Enron, WorldCom, and HealthSouth, have raised questions concerning the quality of current audits. According to the results of a fraud survey conducted by KPMG, 75 percent of companies experienced an instance of fraud in 2003, an increase of 13 percent from 1998; and 36 percent of companies incurred $1 million or more in costs due to fraud, an increase of 15 percent from 1998 1

(KPMG 2003). 1 Financial reporting fraud, the most costly of all frauds discussed in the survey, cost firms on average more than $250 million in 2003. The incidence of financial reporting fraud increased from 3 percent in 1998 to 7 percent in 2003 (KPMG 2003). Second, standard setters recognize that the ultimate effectiveness of brainstorming in improving the auditors ability to detect fraud rests in the importance placed on it by engagement team leaders (AICPA 2002, 2003). Generally, the majority of audits are performed through the hierarchical structure of reviews (i.e., staff, senior, manager and partner) (Messier 2003; Rich, Solomon and Trotman 1997a). Prior auditing studies have shown that disclosure of a supervisor s views influence subordinates judgments (Peecher 1996; Gramling 1999; Brown, Peecher and Solomon 1999; Turner 2001; Bierstaker and Wright 2001; Wilks 2002). The psychology literature has shown that, in general, subordinates tend to make judgments that strategically agree with supervisor views, whether good or bad, in order to efficiently complete the task while exerting the minimum amount of cognitive effort (Tetlock 1985; Lerner and Tetlock 1999). Each of these accountability studies has evaluated individual judgments. However, the audit environment is complex and involves a hierarchy of individuals (Rich, Solomon and Trotman 1997a; Messier 2003). Effects of partner influences on an audit team s judgment (i.e., a group judgment) remain an uninvestigated area. Third, several accounting researchers have studied individual auditors, but not teams, in a fraud context, finding that auditors are generally subject to biases that impair their fraud risk assessments (Hackenbrack 1992; Hoffman and Patton 1997; Wilks and Zimbelman 2003). They do not appropriately use their risk assessments to modify the nature of their audit work (Zimbelman 1997; Glover et al. 2003); they rely unduly on decision aids (Asare and Wright 2004); and auditors ultimately struggle with detecting fraud (Jamal, Johnson and Berryman 1995; Braun 2000; Knapp and Knapp 2001). However, individual auditor studies may not generalize to multi-person settings (Solomon 1987). Group judgments are most often superior to individual judgments, but depending on the various tasks, the decisions are most often different from those of individuals (Hill 1982; Miner 1984; Solomon 1987; Rich, Solomon and Trotman 1997b). While the majority of all audits are conducted in teams, little audit team research exists despite calls for it (Solomon 1987; Libby and Luft 1993; Rich, Solomon and Trotman 1997b; Sutton and Hayne 1997). In summary, an examination of auditor s risk assessment of fraud, professional skepticism and of the efficacy of brainstorming, now required by SAS No. 99, is of interest to regulators, academicians, and the accounting profession. This study offers a contribution to the fraud and accounting literature by extending the evaluation of fraud judgments provided by the individual auditor to those of an interacting hierarchical team of auditors, and may provide answers to the profession and standard setters about the effectiveness of brainstorming in helping auditors better detect fraud. 1 These statistics include all types of fraud including: employee fraud, medical/insurance fraud, financial reporting fraud, check fraud, payroll fraud, kickbacks, inventory theft and misappropriation of assets. 2

1.3 Overview of Method and Results Each of the Big 4 accounting firms provided audit teams comprised of a staff, a senior and a manager to participate in a multi-phase experiment designed to investigate individual and team judgments. The case materials provided to each of the participants contained either fraudulent financial statements issued by a public company or the reissued and restated (i.e., fairly stated) financial statements of the same company for the same time period. Using these case materials, in Phase I auditors individually performed analytical procedures on these financial statements, considered where the financial statements were susceptible to fraud, made an assessment of the client s fraud risk and completed an assessment of professional skepticism. In Phase II of the experiment, the individual audit staff, seniors and managers formed a threeperson hierarchical audit team where they brainstormed about where the financial statements were susceptible to fraud and then assessed the risk of fraud for the client. Finally, in Phase III, the individual auditors again assessed the risk of financial statement fraud for the client presented in the case materials and completed a second assessment of professional skepticism. The results of this study suggest that audit teams with knowledge of a partner s concerns for effectiveness spend significantly more time brainstorming than those audit teams with knowledge of a partner s concern for efficiency. Additionally, although brainstorming among the audit team does diminish the overall number of unique fraud ideas identified, it increases auditors fraud risk assessments and questioning mind, thus improving auditors ability to assess the risk of fraud. 1.4 Organization of Dissertation The remainder of this dissertation is organized as follows. Chapter 2 provides background and hypotheses development. Chapter 3 outlines the experiment and methodology. Chapter 4 presents and discusses the results. Chapter 5 concludes the dissertation with an overall summary, and with a discussion of the contributions, limitations, and avenues for future research. 3

CHAPTER 2 BACKGROUND AND HYPOTHESES DEVELOPMENT This chapter first reviews relevant literature and then develops the formal hypotheses to be tested. This dissertation seeks to determine if (1) audit team brainstorming about ways a company can engage in fraud, and (2) biases that may be introduced by the audit partner affect the auditors fraud risk assessments and professional skepticism. To effectively evaluate this, it is important to review all of the components including the decision maker and the context of the decision. To understand these components and how they might interact, this literature review consists of prior literature from several different areas of research as illustrated in Figure 1. Section 2.1 explores SAS No. 99, Consideration of Fraud in a Financial Statement Audit (AICPA 2002). Sections 2.2, 2.3, 2.4 and 2.5 evaluate the literature on audit teams, fraud, brainstorming and accountability, respectively. Section 2.6 develops five hypotheses based on this literature. Section 2.7 closes Chapter 2 with a brief summary. Brainstorming Audit Teams SAS No. 99 Fraud Accountability FIGURE 1: Diagram of Various Literature Streams Reviewed 4

2.1 SAS No. 99 SAS No. 99 (AICPA 2002) distinguishes fraud from error on the basis of whether the underlying action that results in a misstatement is intentional or unintentional. While fraud is a broad legal concept, the external auditor s concern with fraud specifically relates to fraudulent acts that cause material misstatement of the financial statements. SAS No. 99 is a complex and detailed standard, requiring the exercise of considerable judgment. This standard provides fraud risk factors that the auditor should consider and provides guidance on how the auditor should respond to the risk assessment. The auditor has the responsibility to plan and perform the audit to obtain reasonable assurance about whether the financial statements are free of material misstatement, whether caused by error or fraud (AICPA 2002, p. 3). Based on this standard, the auditor must assess the risk of fraud based on interpretations of this standard and make judgments about the likelihood that there could be material misstatement caused by fraud. SAS No. 99 requires the audit team to discuss the potential for a material misstatement in the financial statements due to fraud before or during the information gathering process. The required brainstorming session is a new concept in the auditing literature. According to the guidance provided by the standard, this discussion should include brainstorming among the audit team members. This procedure was designed to improve the professional skepticism of the auditor in later gathering and evaluating evidence. More specifically, it was designed to lead auditors to thoroughly probe the issues, acquire additional evidence as necessary, and consult with other team members, and if appropriate, experts in the firm, rather than rationalize or dismiss information or other conditions that indicate a material misstatement due to fraud may have occurred (AICPA 2002, p. 11). As part of the required discussion, the standard directs auditors to include a consideration of known external and internal factors affecting the entity that might a) create incentives/pressures for management and others to commit fraud, b) provide the opportunity for fraud to be perpetrated, and c) indicate a culture or environment that enables management to rationalize committing fraud (AICPA 2002, p. 11). 2.2 Audit Teams SAS No. 99 allows for professional judgment in determining which audit team members should be included in the discussion, but it should involve key members of the audit team (AICPA 2002). The typical audit team is composed of, in order of authority, a partner, a manager, a senior and staff members. Staff members are typically responsible for conducting portions of the audit work assigned to them by the senior. The senior participates in the planning, conduct and supervision of the audit work. The senior also reviews the staff members work prior to reviews by the manager and partner. In addition to reviewing the staff and senior s work, 5

the manager and partner have various administrative responsibilities related to the audit engagement, such as scheduling the engagement and billing the client. The partner has the final authority and decision making responsibility for accounting and auditing matters, including the issuance of the audit report (Messier 2003). While the partner is ultimately responsible for the aggregate decisions made by each engagement team, there are many auditor judgments and decisions that occur in various tasks and domains by varying levels of individuals before this decision is reached. There have been several recent calls for team research in accounting (e.g., Solomon 1987; Rich, Solomon and Trotman 1997b; Peecher, Rich and Tubbs 2003) recognizing that team decisions are sometimes better or worse, but are often different than individuals. 2.2.1 Individual Decision Making: The Auditor Academic researchers, like investors, have often described the auditor as an expert in his or her domain. Much of the individual decision making literature in auditing has historically been regarded as auditor expertise research. Some researchers have defined this expertise as a performance-based notion (Davis and Solomon 1989) while others have defined expertise as task specific superior performance (Bonner and Lewis 1990). Even with these definitions however, much of the research has labeled the auditor an expert based on years of experience. In their review of the auditor expertise literature, Bedard and Chi (1993) proposed that auditor experts, typically an auditor with several years of experience, like a manager or partner, know more about their domain than do novices, who are usually accounting and auditing students or entry level staff auditors. The expert auditors not only know more, but their knowledge is better organized than the novices. Bedard and Chi (1993) also summarized areas where audit experts do not excel, namely when a correct solution procedure does not exist, the situation is not well understood, or when there is a mismatch between the expert and the task (Bedard and Chi 1992); detection of earnings manipulation or fraud might be one of these tasks (Bonner and Lewis 1990). Bedard and Chi (1993) also pointed out that, in order to perform an auditing task, the auditor needs both auditing knowledge (general and specialized) and knowledge about the domains of auditing. According to the standards, one of these domains should be the identification of fraud that causes material misstatement to the financial statements. Individual auditor decision making in fraud contexts has been evaluated by a number of accounting researchers. Jamal, Johnson and Berryman (1995) found that, when auditors were given cases in which management had created a misleading description of the company (a frame) and a financial statement fraud, half of the auditors were deceived by management s frame and thus did not detect the fraud. Hackenbrack (1992) investigated the influence of relevant and irrelevant information on auditors fraud risk assessments and found that auditors given irrelevant information incorrectly lowered their assessment of fraud risk. He attributed this to the cognitive limitation known as the dilution effect (Nisbett, Zukier and Lumley 1991). Hoffman and Patton (1997) extended this work finding that accountability did not exacerbate the dilution effect, and Glover (1997) found that time pressure reduced but did not eliminate this effect. 6

Zimbelman (1997) evaluated whether or not requiring auditors to separately assess the risk of fraud, as required by SAS No. 82, would lead auditors to design audit plans that were more sensitive to fraud risk. This study suggested that auditors did not develop budgets that were sensitive to changes in fraud risk and did not appropriately modify the nature of their audit plans in response to fraud risk. In a follow up study, conducted after the issuance of SAS No. 82, auditors judgments were found to be more sensitive to fraud risk factors, but these auditors still did not modify the nature of their planned tests in response to fraud risk (Glover et al. 2003). Braun (2000) evaluated the effects of time pressure imposed on auditors and its influence on their detection of fraudulent financial reporting. In Braun s study, less than 10 percent of the auditors detected the fraudulent misstatement. Additionally, time pressure was found to be a factor; those who detected the misstatement were all in a low time pressure condition. In light of the results of all of these studies, Braun (2000) suggested that research needs to be done to determine if specialized fraud training is needed to improve external auditors fraud risk judgments. Asare and Wright (2004) examined the effect of alternative risk assessments and program planning tools on the quality of audit procedures and auditors propensity to consult with fraud experts. They found that auditors, with a standard audit program, designed relatively less effective fraud programs than those without this template. However, the auditors were not more willing to seek consultation with fraud experts, calling into question the effectiveness of standard audit checklists for fraud commonly used in practice. Wilks and Zimbelman (2003) also studied fraud risk assessments, specifically evaluating the effect of fraud triangle decomposition on fraud risk assessments. The researchers defined fraud triangle decomposition as separately assessing attitude, incentive risks, and incentive opportunity, on auditors sensitivity to incentive and opportunity cues. In a low attitude risk setting, the researchers found that auditors increased sensitivity to a low risk attitude cue offset the benefit attained by decomposition. While these studies provide us with important information about individual auditor s decision making in fraud contexts (see Nieschwietz, Schultz and Zimbelman 2000 for review), drawing inferences from the individual auditor studies may be problematic because the results of single-person studies do not readily generalize to multi-person settings (Solomon 1987). That is, multi-person judgments and decisions, while often better and sometimes inferior, are almost always different from those of individuals (Hill 1982; Miner 1984; Solomon 1987), thus it is important to consider the literature on audit groups and teams. 2.2.2 Multi-Person Decision Making: The Audit Team There are numerous reviews of multi-person decision making in the psychology literature (e.g., Hill 1982; McGrath and Kravitz 1982; Hinsz, Tindale and Volrath 1997). In their literature review of group decision making, Hinsz, Tindale and Volrath (1997) highlighted several important features of group decision making that are different from that of the individual. First, they pointed out that, at the group level, information processing involves the degree to which information, cognitive processes and ideas are shared. Second, they suggested that researchers 7

have consistently found groups to be more reliable than individuals because there is less variability in their judgments (e.g., Einhorn, Hogarth and Klempner 1977). Third, group interaction has been shown to improve decision making because the interaction between the individuals resolves implicit differences of opinion, resulting in improved accuracy and consensus. Fourth, groups may be more influenced than individual decision makers by some of the biases that negatively affect decision making, like, neglecting base rates, overusing representativeness information, and being overly susceptible to persuasion. Finally, Hinsz, Tindale and Volrath (1997) suggested that groups may be better able to use feedback than individuals, which contributes to decreased variability in the way information is processed, thus leading to more useful solutions. There are only two reviews on multi-auditor decision making (Solomon 1987; Rich, Solomon and Trotman 1997b). The existing multi-auditor research, consistent with the psychology literature, has reported that multi-auditor decision making is different from that of individuals (Solomon 1987; Rich, Solomon and Trotman 1997b). Multi-auditor research can be classified into two areas. The first area is the review process and audit team judgments and decisions (e.g., Watson 1975; Bamber and Bylinski 1982; Bamber 1983; Trotman and Yetton 1985; Trotman 1985). These studies evaluated the hierarchical reporting structure of an audit team (i.e., staff, senior, manager and partner). This research seeks to test each individual in that hierarchy, for example whether the individual was a reviewer or a preparer of the work. More recently, we have seen the review process studied in a variety of tasks with different levels of auditors serving different roles in the tasks. Most of the more recent studies have suggested gains from the review in terms of accuracy and control for biases (e.g., Libby and Trotman 1993; Messier and Tubbs 1994; Asare and McDaniel 1996; Reimers and Fennema 1999). Some researchers have examined knowledge differences on review effectiveness (e.g., Ramsay 1994; Bamber and Ramsay 1997; Harding and Trotman 1999), in general, finding benefits to the hierarchical review due to knowledge and experience differences between staff, senior and manager auditors, and those with industry specialization (Owhoso, Messier and Lynch 2002). Other research has evaluated the efficiency of the review process (e.g., Ismail and Trotman 1995; Bamber and Ramsay 2000; Ballou 2001), suggesting overall that the review results in more hypotheses generated and greater efficiency; with the efficiency benefits coming primarily from greater levels of experience that manager auditors possess. Still others have enhanced our understanding by studying influences on the individual judgments from cognitive limitations, reputation effects between the decision makers and the auditing environment (e.g., Rich, Solomon and Trotman 1997a; Sprinkle and Tubbs 1998; Rau and Moser 1999; Yip-Ow 2000; Wilks 2002), finding that the review process may decrease cognitive limitations. 8

However, the review process has not been able to correct for all biases; the benefits depend on a variety of factors including the task (Peecher, Rich and Tubbs 2003) and the perceived competence and auditor expertise (Tan and Jamal 2001). In general, this research suggests that the value of an audit is enhanced by the review process. This review process incorporates individual judgments, refining them with more expert members of the hierarchical audit team to arrive at a final decision on any given task. While this research has incorporated various members of the audit team, generally it has not focused on groups per se. Instead it has evaluated individual s decisions as part of a collective decision of the audit team, which culminates in the review. In summary, this research moves closer to group decisions in auditing than that of the auditor expertise literature, but it does not incorporate interaction between the individuals in reaching their decision. Therefore, one can still not predict the behavior of an audit team on a particular task like evaluating the risk of fraud. The second area of group research focuses more specifically on group audit judgments and decisions (e.g., Schultz and Reckers 1981; Reckers and Schultz 1982; Solomon 1982; Abdelkhalik, Snowball and Wragge 1983; Trotman, Yetton and Zimmer 1983). These early studies make comparisons between groups of auditors and individual auditors and find, in general, that audit team judgments are less variable and more accurate than those of individual auditors. Additionally, from these studies we learn that consensus and accuracy are substantially enhanced and other judgment attributes, such as complexity and consistency, can be improved when the auditors make judgments in a group format. Some more recent research has been compiled based on the results of questionnaires given to audit teams and on field observations (Rudolph and Welker 1998). This research suggests that audit team decision making is valuable and the structures of these teams are important. While this study makes an important contribution, generalizations may be limited by the methodology, which does not include input from interacting audit teams. Other accounting researchers have recently evaluated group decision making using exploratory techniques with student subjects (e.g., Arnold et al. 2000). The results suggest that techniques like Shared Awareness Cognition Training and Group Support Systems can be employed to reduce group biases. Others have studied the effects of contracts on group productivity and reporting (e.g., Sprinkle, Peffer and Fisher 2003; Towry 2003), while still others have used analytical models to demonstrate that the self-serving bias often attributed to auditors can me mitigated by group affiliation (King 2002). Again, the generalizations that can be made to audit group decision making are limited because of the exploratory nature of the studies and the use of students as participants, rather than auditors. Each of these bodies of research discussed has contributed differently to individual and group judgments as they relate to auditing and fraud. However, in order to assess whether brainstorming among the audit team will improve the auditor s professional skepticism and fraud risk assessments and whether brainstorming is biased by a partner s influence, it is necessary to evaluate auditors judgments in a controlled setting. The experimental design employed in this dissertation has an individual and team phase in a potentially fraudulent environment. This 9

design thus attempts to capture the important features of the relationship between the individual auditor and the audit team in a complex environment. 2.3 Fraud To my knowledge, Reckers and Schultz (1993) is the only study to incorporate interacting individuals in groups with a fraud task. Their study was based on SAS No. 53, The Auditors Responsibility to Detect and Report Errors and Irregularities (AICPA 1988), which has since been twice superceded by SAS No. 82 (AICPA 1997) and the current standard SAS No. 99 (AICPA 2002). The authors evaluated the individual and group judgments of 99 Big 6 audit seniors in a potentially fraudulent environment. The auditors were given a description of a corporation with varying levels of fraud risk, and they were asked to make judgments about the likelihood that the financial statements contained fraud. Each of the subjects was asked to make an individual assessment prior to any group interaction, which was later compared to a second individual assessment after group interaction. Half of the individuals were randomly combined into groups of three and told that they were to act as if each person in the group was another senior auditor who had experience with this particular corporation. The researchers referred to this additional component as group counsel. Reckers and Schultz (1993) considered several variables, including experience, order effects, and tolerance for ambiguity as controls in their experiment with a primary interest of assessing whether auditors behaved in accordance with SAS No. 53. Overall, the results of their experiment revealed that the group assisted judgments, when compared to individual judgments, showed a greater adherence to the guidance presented by SAS No.53, resulting in better accuracy in their decisions. While the findings of Reckers and Schultz (1993) are very pertinent to the current study, there are several issues raised that have not yet been addressed. Specifically, Reckers and Schultz (1993) used audit seniors as their participants who were randomly assigned to an individual treatment or group counsel. This group counsel was created in the experiment, combining three seniors who would not typically work together and were not in the normal hierarchy of an audit team (i.e., staff, senior and manager) in which each individual differs in the level of experience. Reckers and Schultz (1993) evaluated the importance of group discussion among audit seniors under the guidance of SAS No. 53, essentially studying a peer group discussion. This can be contrasted to brainstorming, required by SAS No. 99, which will likely occur with individuals at differing levels in the audit team (i.e., staff, senior and manager). Psychology literature suggests that differences in group composition, for example, whether there is a leader or not and whether there are minority versus majority members, make a difference in the decisions reached by these teams (Hinsz, Tindale and Volrath 1997). Additionally, accounting researchers have found that each member of the audit team plays a different role and makes different individual judgments, whether this is due to motivation 10

differences (e.g., Ramsay 1994; Bamber and Ramsay 1997), knowledge differences (e.g., Libby 1985; Frederick 1991; Libby 1995) or both (Owhoso, Messier and Lynch 2002). Knapp and Knapp (2001) made comparisons between individual audit team members and found differences among individual members in a fraud context. They examined the effect of audit experience and explicit fraud risk assessment instructions on the effectiveness of analytical procedures in detecting financial statement fraud. They found that audit managers are more effective than audit seniors in assessing the risk of fraud with analytical procedures. However, they like Reckers and Schultz, did not evaluate the interactive effects of the manager and the senior in a fraud task. Additionally, the fraud standards used in these studies were SAS No. 82 for Knapp and Knapp (2001) and SAS No. 53 for Reckers and Schultz (1993), both of which have since been superceded by the current standard SAS No. 99. The importance of examining interactions between individuals in the hierarchical audit team to arrive at a judgment or decision is demonstrated by continuous calls for research in this area (Solomon 1987; Rich, Solomon and Trotman 1997b; Sutton and Hayne 1997). This demand has been strengthened by the recent auditing standard, SAS No. 99, requiring auditors to brainstorm. This dissertation makes a contribution to the literature by examining an interacting audit team making decisions about fraud. It also provides information to standard setters on the effectiveness of this new requirement for auditors to brainstorm about where the financial statements might be susceptible to fraud as required in SAS No. 99. 2.4 Brainstorming SAS No. 99 requires the audit team to brainstorm about how and where the auditors believe that the entity s financial statements might be susceptible to material misstatement due to fraud, and how management could perpetrate and conceal fraudulent financial reporting. This additional requirement is a new audit procedure and thus introduces a new task into the typical hierarchical team. Therefore, an understanding of the additional influence that brainstorming among the team brings beyond that provided by the group is necessary. In the seminal psychology work, Osborn (1957) suggested that interactive groups will experience an increase in productivity for idea generation tasks. This belief led to widespread acceptance of group brainstorming, but the many benefits described by Osborn have remained illusory (Dennis and Valacich 1993). The justification behind this hypothetical synergistic effect of brainstorming is that members of the group will be stimulated hearing other peoples ideas (see McGrath 1984 for review). Brainstorming is still widely used for decision making even though research has consistently shown that nominal groups (i.e., individuals working separately without communicating combined randomly into groups) outperform interactive groups (see Diehl and Stroebe 1987 for review). Most brainstorming studies in psychology have compared nominal groups (i.e., individuals working separately) to interacting groups, (i.e., face-to-face groups) to determine 11

whether communication among members improves or impairs performance (e.g., Diehl and Stroebe 1987; Gallupe, Bastianutti and Cooper 1991; Dennis and Valacich 1993). Studies most often evaluate the quantity of ideas produced by the group and find that nominal groups outperform interacting groups. Much of the brainstorming literature has focused on productivity losses within groups in an attempt to explain the inferiority of interacting groups. Several productivity losses have been identified throughout the literature (see Nunamaker et al. 1991), but three have received the most attention in verbal brainstorming: 1) production blocking; 2) evaluation apprehension; and 3) free riding or social loafing (Dennis and Valacich 1993). Production blocking occurs because only one member can communicate at once and because members are talking and listening, while waiting to contribute an idea that may be lost before they take their turn (Diehl and Stroebe 1987). Evaluation apprehension involves a group member s concern over the appraisal by other group members. The productivity of the group is reduced when members become personally concerned with how their fellow members will respond to their ideas (Gallupe, Bastianutti and Cooper 1991). Anonymity of the group members and their ideas has been suggested as a way to reduce this loss in brainstorming of ideas, but it may also increase free riding, as it is more difficult to determine then who is contributing and who is not (Dennis and Valacich 1993). Free riding or social loafing is the reduction of effort of some of the group s members even when they are qualified to contribute fully (Diehl and Stroebe 1987). Hackman (1987) has called for research on group brainstorming to discover ways to achieve productivity gains, possibly in reducing the evaluation apprehension, free riding and production blocking. Very little brainstorming research has been done in accounting, but there is a substantial amount of information systems research that has focused on improving brainstorming processes with system intervention (see Murthy 2002 for review). Small group psychology research suggests that small group decision making processes are considerably more complex than individual processes, due in large part to intricate social interactions that are affected by many factors including unequal power distribution, evaluation apprehension, and group member status (McGrath 1984). An audit team is designed as a hierarchy of individuals with increasing experience, each member with different skills, knowledge and status. Each member is accountable to the quality of the audit, to each other and to authorities generated in the hierarchical structure. Researchers have suggested that accountability may induce evaluation apprehension (Kroon, Kreveld and Rabbie 1992), thus leading to a reduction of productivity in brainstorming. 2.5 Accountability In their review of accountability research in psychology, Lerner and Tetlock (1999) defined accountability as implicit or explicit expectation that one may be called on to justify one s beliefs, feelings, and actions to others. They suggested that regardless of whether the views of one s audience are known or unknown, people often seek approval from their respective 12

audience, but when audience views are known prior to forming one s own opinion, conformity becomes a likely coping strategy in order to gain the approval of those to whom they are accountable. Several accounting researchers have evaluated the effects of accountability on auditors. Some have shown that accountability increases effort and improves performance (Johnson and Kaplan 1991; Kennedy 1993). Others have studied the interaction of numerous variables with accountability; in a fraud task environment (Hoffman and Patton 1997; Glover 1997); accounts receivable collectibility task (Turner 2001); and analytical procedures task (Asare, Wright and Trompeter 2000). In some instances the task has been varied in order to assess its interaction with accountability (Tan and Kao 1999). There is consistent experimental evidence that when the audience views are known to the participants, they shift their views toward that of their audience (see Lerner and Tetlock 1999 for review). In an audit context, when the individuals are brainstorming, the views of each member of the team will be known as they are discussed. Additionally, if the audit partner begins the brainstorming session for the audit team and provides his or her views prior to the team having a discussion, these views may influence the brainstorming session. These potential influences were of concern to the Fraud Task Force, and thus ways to guard against these influences are discussed in the SAS No. 99 Implementation Guide (AICPA 2002). Some accounting researchers have studied the influence of accountability when audience views are known (Rich, Solomon and Trotman 1997a; Turner 2001) in a reviewer/preparer setting (i.e., a senior may be reviewing a staff person s work, a manager reviewing the senior s work and ultimately, the partner reviewing the entire audit team s work). In some instances, knowledge of these known views may compromise audit effectiveness (Peecher 1996; Wilks 2002). Whether and how accountability may influence performance of individual auditors in a brainstorming task with other members of a hierarchical team is unclear (Tan and Kao 1999). Some accounting studies have shown that knowledge or experience differences among participants sometimes moderated the effects of accountability (e.g., Kennedy 1993; Cloyd 1997). Tan and Kao (1999) found that accountability exerts the strongest positive influence when the individual s requisite knowledge and abilities are matched with the characteristics of the task. The effects of accountability on auditors have been evaluated in the accounting literature at the individual auditor level. Auditing involves a hierarchical review process for some audit tasks and a consultation (group discussion) on other tasks (Messier and Quilliam 1992). Brainstorming among the audit team about fraud will involve a group discussion among individual members in the normal audit team hierarchy (i.e., staff, senior, manager and partner) and thus represents a unique and rich environment for the auditor and researchers. Whether or not the effects of accountability on individual auditors are the same as those for groups of auditors remains an empirical question. Even with consideration of the existing results in group decision making and accounting, there are unanswered questions about how these topics relate to fraud risk assessments. Will a team discussion (i.e., brainstorming) among auditors improve their judgments about fraud and will those judgments be influenced by the accountability to the audit partner as the leader of the 13

team? Research attempting to bridge these literature streams and address the limitations that have been discussed should make a contribution to auditing research. 2.6 Hypotheses Development 2.6.1 SAS No. 99, Brainstorming and Accountability SAS No. 99 defines fraud as an intentional act that results in a material misstatement in financial statements that are the subject of an audit (AICPA 2002, p. 6). According to guidance provided by the standard, brainstorming must occur among audit team members and should meet several objectives (AICPA 2003). First, auditors are encouraged to identify and discuss ideas about where an entity s financial statements might be susceptible to fraud and how management could perpetrate and conceal it. Second, brainstorming should allow auditors with less experience to learn from more experienced members of the audit team. Third, brainstorming, when combined with the results of preliminary analytical procedures, should allow the auditor to better assess the likelihood of material misstatement in the financial statements as a result of fraud. Finally, brainstorming should heighten the auditors professional skepticism, because a significant part of the brainstorming is geared towards reminding auditors of the importance of maintaining the proper state of mind regarding the potential for material misstatement due to fraud (AICPA 2002). All of these benefits of brainstorming described by the standard setters can be significantly diminished without the proper environment set by engagement team leaders. Standard setters recognize the potential influence of supervisors on the subordinates of the hierarchical audit team and warn audit team leaders to set the proper tone at the top for each engagement. Standard setters suggest that the ultimate effectiveness of brainstorming in improving the auditor s ability to detect material misstatements caused by fraud is dependent on the influences of engagement team supervisors on the other members of the team. SAS No. 99 is concerned with the effectiveness of the auditor in detecting fraud, but in the audit environment, there are associated costs that will come with these additional procedures. Partners preferences for effectiveness or efficiency are likely to vary for a variety of reasons (Brown, Peecher and Solomon 1999; Gramling 1999; Bierstaker and Wright 2001) and do not always have a positive influence on the decisions of their subordinates. Some have suggested that partners should carefully weigh communications with their subordinates about efficiency because they run the risk of jeopardizing effectiveness (Bierstaker and Wright 2001). Enhancing auditors effectiveness in assessing fraud (Knapp and Knapp 2001) and auditors efficiency (Zimbelman 1997; Glover et al. 2003) have recently been examined under the guidance set forth in SAS No. 82 (fraud standard superceded by SAS No. 99). For each audit, the engagement team is faced with efficiency and effectiveness tradeoffs that arise from the profitability of the engagement and partner concerns about litigation. When 14

evaluating the likelihood of fraud for a given client, the complexity of this tradeoff increases. It is unlikely that any member of the engagement team will encounter fraud in their career (Nieschwietz, Schultz and Zimbelman 2000); and competitive pressures among the firms, as well as pressures on each engagement to maintain a positive relationship with the client, create a tension to maintain an efficient audit (Rich, Solomon and Trotman 1997a). Creating an opposing pressure is the fact that the investing public relies on auditors to detect fraudulent reporting; and if an audit fails to detect the fraud when it exists, the firm could face significant litigation and reputation loss (e.g., Enron, WorldCom and HealthSouth). Accountability research shows that when individuals know the views of their audience prior to forming an opinion, they will likely adopt the position that is expected to gain favor with the person to whom they are accountable (see Lerner and Tetlock 1999 for review). Several accounting studies have shown that individuals tailor their responses to their audience. In auditing, partner views have been shown to influence subordinate auditors in analytical reviews (Peecher 1996), reliance on internal auditors (Gramling 1999), going concern judgments (Wilks 2002), audit planning decisions (Bierstaker and Wright 2001; Bierstaker and Wright 2003), and accounts receivable collectibility reviews (Turner 2001). In tax, litigation support, and financial reporting, the audience views that are known to the individual are most often those of the client (Cuccia, Hackenbrack and Nelson 1995; Ponemon 1995; Hackenbrack and Nelson 1996; Cloyd and Spilker 1999) or an advisor (Perkins 2003). Rich, Solomon and Trotman (1997a) warned of the potential negative effects that knowledge of a supervisor s views or preferences may have on the effectiveness and efficiency of audits. Several studies have found a compromise of effectiveness when partner preferences create incentives to be less than objective (Peecher 1996; Turner 2001; Wilks 2002), or provide pressure to enhance audit efficiency (Gramling 1999; Bierstaker and Wright 2001). This leads to my first hypothesis. H 1 : Teams of auditors with knowledge of a partner's effectiveness preference will spend more time brainstorming than will teams of auditors with knowledge of a partner s preference for efficiency. 2.6.2 Brainstorming Teams versus Individuals: Quality of Judgment Brainstorming, as suggested in SAS No. 99, is likely to include a hierarchical team structure of individual auditors (i.e., staff, senior and manager) interacting face-to-face. These teams are to discuss how and where they believe the entity s financial statements might be susceptible to fraud, how management could conceal it, and what audit procedures need to be performed to appropriately address fraud risks. There are several important features of group decision making that are different from that of individual decision making (e.g., Hill 1982; McGrath and Kravitz 1982; Hinsz, Tindale and Volrath 1997). In general, one cannot assume that the group will provide a decision the same as an individual, nor can one simply sum the parts (i.e., the individual decisions of all members of the group). Group interaction has been shown to improve decision making because the 15

interaction between the individuals resolves implicit differences of opinion. This results in improved accuracy and consensus (e.g., Sniezek 1992; Henry 1995), and provides less variability in judgments than in those of individuals (e.g., Einhorn, Hogarth and Klempner 1977). In the auditing literature, there are two comprehensive reviews on multi-auditor decision making (Solomon 1987; Rich, Solomon and Trotman 1997b). Consistent with the psychology literature, these reviews have reported superior team or group performance, suggesting gains from the group decision making when compared to individuals. Early audit group studies, where interacting auditors were the subject of analysis, provided evidence that individual judgment is enhanced by group processes (Schultz and Reckers 1981; Solomon 1982; Trotman, Yetton and Zimmer 1983). More recently, gains from groups have been recognized between individual judgments and group judgments in recognition accuracy and confidence (Johnson 1994), adherence to guidance (Reckers and Schultz 1993), and increases in consensus and more complete problemanalysis (Bamber, Watson and Hill 1996). Trotman and Yetton (1985) found that groups performed better than individuals in an auditing task. Trotman (1985), in a more complex task, found that both interacting groups and review process groups outperformed a composite and individuals, but he found no distinction between interacting groups and review groups. He suggested that interacting groups can reduce random error while the review process reduces systematic bias. In this dissertation, auditors either receive fraudulent financial statements or they receive non-fraudulent financial statements. The auditor s effectiveness (i.e., higher fraud risk assessment when fraud is present and lower fraud risk assessment when fraud is not present) can thus be measured for the individual auditor, for the interacting audit team, and for the composite or nominal group (i.e., average of the individual staff, senior and manager on each team). Based on the above discussion, the interacting audit team should be more effective than the nominal group (i.e., average of the individual staff, senior and manager on each team). Stated formally: H 2 : The fraud risk assessment of the brainstorming team will be more effective (i.e., higher when fraud is present and lower when fraud is not present) than that of the nominal group. 2.6.3 Brainstorming Teams versus Individuals: Number of Ideas Generated In contrast to the many studies in psychology and auditing showing groups outperforming individuals and composites, there are several reviews of research comparing interacting groups in the brainstorming literature that report composites (i.e., nominal groups) outperform interacting groups in the number of ideas that can be generated (e.g., Hill 1982; Diehl and Stroebe 1987; Gallupe, Bastianutti and Cooper 1991; Dennis and Valacich 1993). Diehl and Stroebe (1987), in their literature review of over fifty brainstorming studies, concluded that brainstorming groups do not outperform nominal groups (i.e., total unique ideas of collective individual judgments). This research has suggested that the productivity of the group is reduced 16

(i.e., results in a process loss) when members interact. In general, these studies have found that communication among members impairs performance, suggesting that nominal groups outperform interacting groups in the quantity of ideas produced. While SAS No. 99 implies that more ideas are better in the brainstorming session among the audit team members, the brainstorming literature suggests that interaction among the audit team will result in a loss of ideas generated. Therefore, the overall number of unique ideas identified by the individual auditors (i.e., the staff, senior and manager) should be greater than those identified by the audit team in the interactive brainstorming session. This leads to the third hypothesis, stated formally: H 3 : Brainstorming will influence the auditors judgments such that the number of ideas generated by the interacting group will be fewer than the number of ideas generated by the nominal group. 2.6.4 Improvements in an Individual s Quality of Judgment as a Result of Brainstorming Although brainstorming may not result in more ideas generated by the audit team than would have been generated by the individuals, other potential gains are still possible. SAS No. 99, in its implementation guidance, establishes rules for the engagement team to maximize the benefits of brainstorming on auditors fraud detection ability. For example, brainstorming among the audit team should be done in an environment that fosters sharing of ideas among both senior members and junior members of the engagement team (AICPA 2003). Several researchers have suggested that some of the benefits of the audit team and its hierarchical structure come from an opportunity for individual members to learn from other more experienced members of the team (Libby and Luft 1993; Rich, Solomon and Trotman 1997a, AICPA 2003). Standard setters suggest that many auditors do not encounter fraud in their careers in public accounting (AICPA 2003). However, they suggest that the mere act of the audit team engaging in a serious discussion of an entities susceptibility to fraud (i.e., brainstorming) reminds each auditor of the possibility of fraud on every audit. This is expected to contribute to auditors overall heightened awareness of fraud (AICPA 2003), likely increasing their fraud risk assessments. Additionally, group discussion has been found to magnify the inclinations of individual members of the group (see Plous 1993; Myers and Lamm 1976). For example, if a task directs individuals to be risky, they are more risky after group discussion, and likewise if a task directs individuals to be more cautious, they are more cautious after group discussion. This is known as the risky shift (Stoner 1968) or choice shift phenomenon. Brainstorming, which includes a discussion with team members about how and where management could commit fraud, is therefore likely to result in auditors increasing their fraud risk assessments after the team discussion. Reckers and Schultz (1993) evaluated the individual judgments of audit seniors in a potentially fraudulent environment and varied whether the individuals were allowed to interact in a group or not. Overall, their results showed that the judgments of those individuals that had 17

group assistance (i.e., were able to interact with two other seniors prior to their final judgment) adhered more closely to the guidance presented by SAS No. 53, the fraud standard in effect at the time of their study. This group assistance resulted in more accurate individual auditor decisions than those of the individual auditors who did not have the benefit of interaction with other auditors. Based on the above discussion, auditors should increase their fraud risk assessments as a result of the brainstorming session. Therefore, the individual s pre-brainstorming fraud risk assessment should be lower than the post-brainstorming fraud risk assessment as a result of brainstorming and audit team interaction. In addition, if auditors improve as a result of team interaction, like those auditors in Reckers and Schultz (1993), then the fraud risk assessments of auditors in the fraud condition will be higher than those of auditors in the no fraud condition. Stated formally: H 4 : Individual auditors final judgments on the risk of financial statement fraud will increase, and will increase more when fraud is present, as a result of the brainstorming session. 2.6.5 Professional Skepticism Professional skepticism is defined in SAS No. 99 as an attitude that includes a questioning mind and a critical assessment of audit evidence (AICPA 2002, p. 10). This standard discusses improvements to professional skepticism as an additional benefit of the brainstorming session among auditors. Studies have cited lack of professional skepticism as a leading cause of audit failure in fraud detection (Beasley, Carcello and Hermanson 1999). Brainstorming is supposed to emphasize the importance of maintaining the proper state of mind throughout the audit regarding the potential for material misstatement due to fraud (AICPA 2002). Because of the characteristics of fraud, it is important for the auditor to approach each audit recognizing the possibility that fraud may be present. In their review of professional skepticism, Hurtt, Eining and Plumlee (2003) suggested that, despite the prominence of professional skepticism in auditing pronouncements, there is insignificant guidance for auditors as to the definition or its implementation on audits. There is limited academic research in this area. Relevant to the current study, professional skepticism has been used in the language for manipulating partner preferences in several studies (Peecher 1996; Gramling 1999; and Turner 2001). Partner preferences for professional skepticism have been used as proxies for an effectiveness condition. This manipulation has been shown to influence auditors decisions, causing auditors in this condition to behave differently than those who were encouraged to be efficient by the partner. SAS No. 99 suggests that the brainstorming session will increase the professional skepticism of members of the engagement team. Additionally, prior research suggests that, when reminded about the importance of professional skepticism, auditors behave differently than when they are not reminded (Peecher 1996; Gramling 1999; Turner 2001). 18

Hurtt, Eining and Plumlee (2003) developed a model of professional skepticism, and Hurtt (2003) developed a scale that measures an individual s level of professional skepticism. Participating auditors professional skepticism was measured by the scale developed by Hurtt (2003) in various phases of the experiment. Based on the foregoing arguments, auditor s professional skepticism should increase as a result of brainstorming. Stated formally: H 5 : Brainstorming will increase individual auditors professional skepticism. 2.7 Summary of Chapter 2 This dissertation takes an interdisciplinary approach to evaluating the effectiveness of brainstorming as proposed by SAS No. 99. It draws on accounting and psychology paradigms specifically bridging audit team, fraud, accountability and brainstorming literatures to develop and test hypotheses related to auditors fraud judgments. Hypothesis 1 evaluates the partner s influence on the audit team during brainstorming. Hypotheses 2 and 3 compare the fraud judgments of individual auditors to composites of these judgments and to those judgments produced by the interacting team. Hypotheses 4 and 5 evaluate changes in individual auditors judgments as a result of the brainstorming session. Hypothesis 4 focuses on changes in fraud risk assessments and Hypothesis 5 evaluates changes in auditors professional skepticism. The research design employed to test these hypotheses is described in Chapter 3. 19

CHAPTER 3 EXPERIMENT AND METHODOLOGY Chapter 3 describes the research method used to gather data to test the hypotheses developed in Chapter 2. The participants are described in Section 3.1. The research design is outlined in 3.2. Section 3.3 concludes the chapter. 3.1 Participants Eighty three-person hierarchical audit teams from each of the Big 4 firms participated in the experiment. Each audit team consisted of an audit staff, senior and manager, resulting in 240 individual auditor participants. On average, the managers had 7.93 years of experience (standard deviation 2.98); the seniors had 3.12 years of experience (standard deviation 2.00) and the staff had 0.53 years of experience (standard deviation 0.58) as auditors of a Big 4 firm. 2 3.2 Research Design The experiment had three phases: pre-brainstorming, brainstorming and postbrainstorming as illustrated in Figure 2. The case materials were provided to each participant in each of the three phases. In Phase I, participants individually completed the experimental task. This ensured that the participants were familiar with the case before brainstorming with team members, and provided individual pre-brainstorming responses for measuring any change in position arising from audit team interaction and brainstorming. The second phase involved the brainstorming session among individuals assigned to teams of three auditors in hierarchies consisting of a staff, senior and manager. In Phase III, auditors again completed the experimental task individually and provided demographic information. 3 2 Neither firm nor location had an effect on the results in this study. 3 This design has been used in a number of studies including (Bamber, Watson and Hill 1996) who used auditor participants. Having three phases allows for a powerful test of individual and group responses and captures any change within the individual as a result of the group interaction (Hegedus and Rasmussen 1986; Gigone and Hastie 1997). 20

Partner Influence Phase 1- Individual Assessment of Fraud prior to Brainstorming Phase II- Group Interaction- Brainstorming about Fraud Phase III- Individual Assessment of Fraud after Brainstorming Individual Staff Staff Senior Manager Individual Staff Individual Senior Individual Senior Individual Manager Individual Manager Performance Measures Ideas of fraud Ideas of fraud Fraud risk assessment Fraud risk assessment Fraud risk assessment Professional skepticism Professional skepticism FIGURE 2: Overview of the Phases of the Experiment 21

3.2.1 Phase I Eighty audit managers, 80 seniors and 80 staff completed the initial phase. All participants were asked to put their names on all of the materials and were told that partners from their office would be reviewing the materials in the future. 4 At each firm office location, the individuals at each level (i.e., staff, seniors and managers) were randomly assigned to teams of three with each hierarchical level represented equally. Before the introduction of any case materials or treatments, the individual participants completed a warm up exercise where they were asked to brainstorm and list all of the possible uses of an ordinary paper clip in two minutes. After this warm up exercise, auditors were given instructions and the case materials. The initial task required auditors to use analytical procedures to analyze financial statements and the associated notes of an actual company identified by the Securities Exchange Commission (SEC) to have issued fraudulent financial statements. See Appendix A for experimental instrument. Phase I of the experiment employed a 2x2 design. The first independent variable, the presence or absence of fraud in the financial statements, and the second, partner influence (i.e., preference for efficiency or effectiveness), were varied between subjects. This second variable (partner influence) was included in the introduction of the case. Auditors were assigned to the efficient or effective condition and read the following: 5 Efficiency- While helping to construct this case, the engagement partner on this audit expressed numerous times his concern about the associated costs of implementing brainstorming on this engagement. Specifically, he is concerned about the audit team members being overly sensitive to unusual account balance fluctuations noted in the initial analytical procedures, as this sensitivity may lead to costly increases in unjustified investigations and efficiency losses on the audit. He would like the assessment of fraud risk and the associated brainstorming to be sufficient to comply with the standard, but he hopes that you will be aware of the costs and complete this phase of the audit as efficiently as possible. Effectiveness- While helping to construct this case, the engagement partner on this audit expressed numerous times his concern about implementing brainstorming on this engagement with a sufficient level of professional skepticism. Specifically, he is concerned about the audit team members not being sensitive enough to unusual account balance fluctuations noted in the initial analytical procedures, as this insensitivity may lead to costly litigation or losses to the reputation of the firm. He would like you to approach the assessment of fraud risk and the associated brainstorming with the appropriate level of professional skepticism as suggested by the standard and to complete this phase of the audit as effectively as possible. After reading this statement, participants were asked to list three reasons that auditors can sometimes be overly sensitive to 1) the cost of unjustified investigations; or 2) cost of potential litigation when evaluating a client s fraud risk in the planning stage of the engagement. Next, participants ranked these reasons in order of their impact on auditors judgments. These steps 4 Copies of samples of the completed instruments were mailed to the partners of the firms following the experiment so as to increase accountability and to not deceive auditor participants. 5 These partner preferences are adapted from several studies that have manipulated partner preferences in a similar fashion (Peecher 1996; Gramling 1999; Brown, Peecher and Solomon 1999; Bierstaker and Wright 2001; Turner 2001; Wilks 2002). 22

were included prior to auditors beginning the task to make salient the partners concern (e.g., Peecher 1996, Wilks 2002). 3.2.2 Case Materials and Procedures Two versions of the case were constructed, one containing fraud in the financial statements as issued and detected by the SEC and cited in an Accounting and Auditing Enforcement Release (AAER) and the other containing the restated financial statements containing no fraud. This case was created using annual and 10-K reports of an actual company. A narrative description of the company s business was provided at the beginning of the case, describing its competition, management, markets and products. A set of financial statements with related notes was provided after the narrative along with common size balance sheets, income statements and selected financial ratios to aid auditors in performing analytical procedures. Despite the presence of financial statement fraud, the company received an unqualified opinion. The SEC charged the company with using a variety of scheming devices to overstate earnings in its financial statements. 6 The SEC required this company to restate and reissue its financial statements. The originally issued financial statements were used in the fraud condition, and the restated financial statements were used in the no fraud condition. See Appendix B for fraud case materials. 7 Participants at each firm office location were randomly assigned to the experimental conditions. In Phase I of the experiment the auditors were provided the case materials 8 and were asked to perform preliminary analytical procedures on a new client s financial statements with instructions that they would be assessing the risk of fraud in this phase of the audit. 9 The auditors were asked to assume that they were in the initial planning stages of the audit. They were then asked to list unusual items noted in the financial statements and provide potential explanations for the unusual items noted. Participants were not allowed to use reference materials or confer with one another while completing this phase of the experiment, although they were told about Phase II of the experiment in which they would be sharing these ideas in a brainstorming session. Upon completion of the analytical procedures, the next section asked auditors to list how and where they believed the entity s financial statements might be susceptible to material misstatement due to fraud and how management could perpetrate and conceal fraudulent financial reporting. 10 Next, participants were asked to provide the likelihood 6 This company settled with the SEC and paid a large fine in excess of $5 million dollars. Additionally, members of top management were terminated and barred from serving as managers or accountants indefinitely. 7 In the post-experimental questionnaire, participants were asked if they could determine the identity of company used in the case materials. Only four out of 240 auditors correctly identified the company, with three of them from the same team. The staff and senior members of this team wrote in the comments that they were guessing what their manager thought. Overall, identification of the real company had no significant effect on any of the results in this dissertation. 8 All case and experimental materials were reviewed by two audit partners from a Big 4 firm to ensure realism of the materials and congruence with the brainstorming procedures and tasks auditors performed in this experiment with those in practice. In addition, all materials were pilot tested resulting in minor changes made to the wording of the initial materials. 9 Specific fraud risk instructions have been found to improve the auditors judgments (Knapp and Knapp 2001). Based on these findings and the standard setters urging of preparation using analytical procedures before the brainstorming phase, these instructions were given to all participants in Phase I of the experiment. 10 These idea units (i.e., lists from the paper clip, analytical procedures, and fraud ideas task) were transcribed from the written instruments. The transcriptions were then coded by the researcher and another independent coder with 23

of fraud in the case they had completed on an 11-point scale with the endpoints labeled Extremely Unlikely, and Extremely Likely, and the center point labeled Neutral. Finally, auditors provided answers to a questionnaire designed to evaluate professional skepticism. The scale used in the questionnaire was developed by Hurtt (2003) and was validated with both students and working auditors. 11 This scale provides a measurement for each individual as to the level of professional skepticism that he or she has, and the scale has been shown to have both adequate inter-item consistency between questions and adequate test-retest reliability between students and auditors. Each of the measures collected in Phase I was used to obtain auditors initial judgments prior to Phase II, the brainstorming phase. At the completion of Phase I, the completed materials were collected; however, the participants were able to retain a copy of their individual answers. 12 Phase I of the experiment took approximately 45 minutes to complete. 3.2.3 Phase II Phase II, the brainstorming phase, began after all participants had completed Phase I. Participants at each firm office location were randomly assigned to a team with one staff auditor, one senior auditor and one manager auditor; and they sat face-to-face around a table to discuss their ideas (this assignment took place in Phase I to ensure consistent materials and treatments for each member of the team). In Phase II, the participants used the same auditing case as the one used in Phase I. The second phase contained an additional statement (a reminder) of the partner s preferences in order to make these views salient prior to beginning the team phase. To ensure a team process with a single solution in Phase II of the experiment, individuals did not record their own ideas; rather a designated recorder was randomly selected in each team 13 to record the team s conclusions. Each team produced a single list of how and where they believed the entity s financial statements might be susceptible to material misstatement due to fraud and how management could perpetrate and conceal fraudulent financial reporting, just like the participants had done individually in Phase I. Because the time spent on the brainstorming session was predicted in Hypothesis 1 to vary with the type of partner influence (i.e., efficiency vs. effectiveness), the time spent in Phase II was not restricted, 14 and the team recorder recorded a prior Big 4 audit experience. Inter-rater consistency on the number of ideas across all of these tasks was 99% and any discrepancies were resolved. See Appendix C for an example of the transcriptions and coding. 11 Hurtt (2003) provides a questionnaire that measures professional skepticism using 30 questions. The answers are provided using a six point scale ranging from strongly disagree to strongly agree. The scale has undergone rigorous testing including a developmental pre-test, pilot, and re-test for approximately 250 students and 90 professional auditors. Factor analysis was performed showing that 30 questions loaded on six factors, (i.e., questioning mind, suspension of judgment, search for knowledge, interpersonal understanding, self-confidence and self-determination). These loadings match the theoretical constructs suggested in the model of professional skepticism developed by Hurtt, Eining and Plumlee (2003). 12 Keeping a copy of the individual answers (the ideas generated and the fraud risk assessments) is used in this study because memory is not a variable under consideration; and it is reasonable to assume that auditors, in practice, would have notes taken from the analytical procedures conducted likely to occur prior to brainstorming as suggested by SAS No. 99. This technique has been used in other studies evaluating group interaction (e.g., Libby, Trotman and Zimmer 1987). 13 Which individual on the engagement team served as the team recorder had no significant influence on any of the results reported in this study. 14 Time restrictions have been shown to interfere with brainstorming, and influence the results of most studies in this area. Several researchers have cited these restrictions as a confound contributing to the reasons for conflicting results among studies and have suggested that time restrictions be removed in future research (Miner 1984; Hegedus and 24

start time and completion time for this phase. On average audit teams spent 18 minutes discussing ideas and recording a team fraud risk assessment in this phase of the experiment. 3.2.4 Phase III After Phase II, a final individual fraud risk assessment and debriefing questionnaire comprised the final (i.e., post-brainstorming) phase of the experiment. When a team completed the tasks from Phase II and thus concluded their brainstorming session, the case materials were collected and each team member was given the final experimental questionnaire. Participants were asked to complete Phase III individually. This questionnaire requested a final assessment of the likelihood of fraud in the financial statements and a final assessment of the auditor s professional skepticism. Background information on the participants was also collected in this phase. Phase III took approximately 10 minutes to complete. 15 3.3 Summary of Chapter 3 This chapter described the research method used to gather data to test the hypotheses developed in Chapter 2. The research design employed three phases. The first phase, prebrainstorming, required individual auditors to evaluate financial statement case materials using analytical procedures. The auditors were asked to list unusual items noted during analytical procedures and provide a list of how and where they believed the financial statements were susceptible to fraud. Additionally, auditors assessed the risk of fraud and completed a questionnaire designed to measure their professional skepticism. The second phase, brainstorming, allowed the audit team consisting of a staff, senior and manager to interact and brainstorm, and again provide a list of how and where they believed the financial statements were susceptible to fraud. These audit teams were also asked to provide a team fraud risk assessment in this phase. Finally, in the third phase, post-brainstorming, individual auditors were asked to provide a final assessment of the risk of fraud and a final measure of professional skepticism. Chapter 4 summarizes the data collected in the experiment and describes the statistical tests of the hypotheses. Rasmussen 1986). Additionally, this variable was expected to vary based on the partner influence hypothesized in H 1. 15 In the post-experimental questionnaire, participants were asked if they had participated in a brainstorming session thus far on an audit, and if so, about how much time was spent. Of the 240 participants, 162 had participated and they suggested the time usually spent on the brainstorming session, including preliminary review of client information, analytical procedures and group discussion was on average 68 minutes. This amount aligns well with the average of 63 minutes spent by auditors participating in this experiment in Phases I and II. 25

CHAPTER 4 RESULTS This chapter describes the statistical techniques used to test the hypotheses developed in Chapter 2. The data used in these statistical tests were gathered from auditors from all of the Big 4 firms as outlined in Chapter 3. The chosen α level in this dissertation was.05. All numbers except the p-values were rounded to the nearest hundredth. 4.1 Manipulation Check Data from all individual auditors (i.e., 80 managers, 80 seniors and 80 staff) indicate that the manipulation of the partner s influence was successful. Each of these individuals remembered the preference of the partner, 120 accurately recalling an efficiency partner, and 120 accurately recalling an effectiveness partner when asked in the debriefing questionnaire in Phase III. 4.2 Tests of Hypotheses 4.2.1 Hypothesis 1 Hypothesis 1 predicts that teams of auditors with knowledge of a partner s effectiveness preference will spend more time brainstorming than will teams of auditors with knowledge of a partner s preference for efficiency. To test H 1, I analyzed the amount of time spent by the audit teams during the brainstorming session using a 2x2 ANOVA model with fraud condition (fraud, no fraud) and partner influence (efficiency, effectiveness) varied between subjects (i.e., audit teams). The analysis reported in Table 1, Panel A reveals a significant main effect for partner influence (p=0.033), supporting H 1. The means presented in Table 1, Panel B, indicate that audit teams in the efficiency condition spent less time on average brainstorming than those audit teams in the effectiveness condition. 26

TABLE 1 Amount of Time Spent Brainstorming by the Audit Teams Panel A: Results of a 2x2 ANOVA of Fraud, and Partner Influence on Amount of Time Spent Brainstorming Source of Variation Df SS MS F p-value Fraud (No Fraud) 1 27.37 27.37 0.65 0.423 Partner influence (efficiency, effectiveness) 1 200.81 200.81 4.77 0.033 Fraud*Partner influence 1 32.04 32.04 0.76 0.387 Error 62 2612.07 42.13 Total (Adjusted) 65 2866.26 Panel B: Mean (Standard Deviation) Amount of Time Spent in Minutes Spent by Audit Teams Brainstorming Fraud Condition Partner Influence Fraud No Fraud Efficiency 16.29 (8.06) 16.40 (6.78) Effectiveness 21.19 (5.14) 18.50 (5.60) 4.2.2 Hypothesis 2 Hypothesis 2 predicts that the fraud risk assessments of interacting audit teams will be more effective (i.e., higher when fraud is present and lower when fraud is not present) than the nominal group. To test H 2, the average of individual audit team risk assessments (i.e., the nominal group) was compared to the interacting team fraud risk assessment (i.e., the consensus team judgment). To test audit effectiveness (i.e., higher fraud risk assessments when fraud is present and lower fraud risk assessments when fraud is not present), I partitioned the individuals 27

and teams based on fraud condition (fraud or no fraud). 16 The mean fraud risk assessments for individual auditors (i.e., staff, senior and manager), the nominal group, and the interacting audit team are presented in Table 2. For the fraud condition, a paired sample t-test reveals a significant difference between the average fraud risk assessments of the individuals (i.e., the nominal group) and the team (p<0.001). The audit team performed significantly better (µ=7.15) than the nominal group (µ=6.46). This offers support for H 2 in the fraud condition. For the no fraud condition, the same test was performed revealing a significant difference between the team and the nominal group (p<0.001). The same pattern held in this condition, the team s fraud risk assessments were significantly higher (µ=6.95) than the nominal group (µ=6.23), but in this condition that is less effective because the fraud risk assessment should be lower when no fraud is present. Thus, H 2 is not supported in the no fraud condition. Brainstorming sessions, as suggested by the standard setters, encourage audit teams to brainstorm about all of the ways that managers could commit and conceal fraud and reminds auditors of the possibility that fraud exists on every audit. It thus seems reasonable that this group discussion (i.e., brainstorming session) about fraud will likely lead the team to assess the risk of fraud as higher in all conditions. 17 Additional analysis revealed that in the fraud condition, while the interacting team assessed the risk of fraud higher in this condition than the nominal group, thus outperforming the nominal group, the team likewise significantly outperformed the individual senior (p=0.002), and individual staff (p<0.001), with means of (µ=6.38 and 5.93), respectively for seniors and staff. However, the interacting team did not significantly outperform the individual audit managers (µ=7.08), likely the best member of the team (p=0.776). In general, these findings are consistent with interacting teams outperforming individual members of the team, but not always the best member (e.g., Hill 1982, Hastie 1986). In the no fraud condition, additional analysis revealed that the interacting team had significantly higher fraud risk assessments than managers (p<0.001), seniors (p=0.029) and staff (p=0.002). This reveals a tendency for the team to assess the risk of fraud as higher after brainstorming about fraud than the individuals did prior to the brainstorming session. 16 No predictions were made about partner influence (i.e., efficiency vs. effectiveness) on fraud risk assessments or number of ideas generated, and no significance was found for this variable when a 2x2 ANOVA analysis is performed. The means are presented by partner influence and fraud condition for consistency and comparative purposes. 17 In the post-experimental questionnaire, participants were asked to estimate the percentage of companies that fail due to fraudulent financial reporting; the mean percentage was 13.00% with a standard deviation of 15.32 revealing a large range in responses. While this serves as auditors overall baseline measure of fraud among companies, the answers to this question had no overall effect on the results in this dissertation. 28

TABLE 2 Mean (Standard Deviation) Fraud Risk Assessments by Treatment for Individual Auditors, Nominal Group and Interacting Audit Team Partner Influence Managers Seniors Staff Nominal Group Interacting Team Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Efficiency 6.80 (1.88) 5.85 (1.63) 6.35 (1.42) 6.55 (1.28) 5.75 (1.37) 6.05 (1.28) 6.30 (1.61) 6.15 (1.41) 6.85 (1.14) 7.00 (1.12) Effectiveness 7.35 (1.50) 6.40 (1.31) 6.40 (1.70) 6.25 (1.25) 6.10 (1.21) 6.25 (1.33) 6.62 (1.55) 6.30 (1.28) 7.45 (0.89) 6.90 (0.85) Totals 7.08 (1.70) 6.13 (1.49) 6.38 (1.55) 6.40 (1.26) 5.93 (1.29) 6.15 (1.29) 6.46 (1.58) 6.23 (1.34) 7.15 (1.05) 6.95 (0.99) Supplemental analysis was performed on the mean fraud risk assessments for individual auditors. This analysis revealed that managers assessed the likelihood of fraud as higher in the fraud case than in the no fraud case (p=0.010). However, managers were the only individuals on the audit team who demonstrated the expected difference in risk assessments between fraud and no fraud. These individual results are consistent with those found in Knapp and Knapp (2001). 4.2.3 Hypothesis 3 Hypothesis 3 predicts that the number of ideas generated by the team will be less than the number of ideas generated by the nominal group (i.e., the number of overall unique ideas of the three individual team members). To test H 3, the number of ideas is compared between the interacting audit teams and the nominal groups. A paired sample t-test reveals a significant difference in the number of ideas generated (p<0.001). The mean number of unique ideas generated for the individuals (i.e., staff, senior and manager), the nominal group, and the audit team are presented in Table 3. As predicted, the interacting team generated significantly fewer ideas (µ=7.44) than the nominal group (µ=12.46), thus supporting H 3. Additional analysis revealed that while the team did generate significantly fewer ideas than the nominal group, the team generated significantly more ideas than the individuals (i.e., average number of ideas) on the team (µ=4.54, p<0.001). This finding is consistent with the brainstorming literature (e.g., Diehl and Stroebe 1987; Dennis and Valacich 1993), where the interacting group generates more 29

unique ideas than the individuals on the team, but does not generate more unique ideas than the nominal group (i.e., the overall number of unique ideas combined from each team member). Supplemental analysis of the individual auditor number of ideas generated reveals that managers generate significantly more ideas than seniors (p=0.005) and staff (p<0.001), and that seniors also generate significantly more ideas than staff (p=0.032) with the mean number of ideas of (µ=5.49, 4.46, and 3.68), respectively for managers, seniors and staff. TABLE 3 Mean (Standard Deviation) Number of Unique Ideas Generated by Treatment for Individual Auditors, Nominal Group and Interacting Audit Team Partner Influence Managers Seniors Staff Nominal Group Interacting Team Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Efficiency 5.20 (2.35) 5.55 (2.11) 3.60 (2.04) 4.60 (3.27) 4.40 (2.21) 3.15 (1.87) 11.90 (4.31) 12.35 (3.80) 7.05 (2.15) 7.80 (2.72) Effectiveness 5.05 (2.42) 6.15 (3.07) 4.85 (1.95) 4.80 (2.48) 4.40 (2.06) 2.75 (1.25) 13.10 (3.33) 12.50 (3.74) 7.90 (2.28) 7.00 (2.72) Fraud Totals 5.13 (2.35) 5.85 (2.62) 4.23 (2.07) 4.70 (2.87) 4.40 (2.11) 2.95 (1.58) 12.50 (3.88) 12.43 (3.76) 7.48 (2.25) 7.40 (2.74) Overall Totals 5.49 (2.50) 4.46 (2.50) 3.68 (1.99) 12.46 (3.81) 7.44 (2.50) 4.2.4 Hypothesis 4 Hypothesis 4 predicts that individual auditors final judgments on the risk of financial statement fraud will increase and improve (i.e., increase more when fraud is present) as a result of the brainstorming session. This improvement can be seen by an increase in the difference between fraud risk assessments of those individuals in the fraud condition and those in the no fraud condition from pre-brainstorming to post-brainstorming. Hypothesis 4 was tested using a repeated-measures ANOVA run at the individual auditor level, with fraud risk assessment as the dependent variable, fraud condition (fraud or no fraud) as a between-subjects independent variable and a shift variable representing the change between pre-brainstorming and postbrainstorming judgments. 18 18 This analysis was performed with partner influence in the model, however, there were no predictions made for this variable, and likewise no significance was found, thus the means presented in Table 4, Panel B are not partitioned on this variable. 30

TABLE 4 Individual Fraud Risk Assessment Shifts Panel A: Results of a Repeated Measures ANOVA of Fraud Between Subjects on Individual Fraud Risk Assessments Pre-Brainstorming and Post-Brainstorming Source of Variation Df SS MS F p-value Fraud (No Fraud) 1 14.01 14.01 4.92 0.027 Partner(Efficiency, Effectiveness) 1 5.63 5.63 1.98 0.161 Shift (Post-Pre Fraud Risk) 1 67.50 67.50 85.01 0.000 Fraud*Shift 1 1.41 1.41 1.78 0.184 Partner*Shift 1 0.03 0.03 0.04 0.838 Fraud*Partner 1 3.01 3.01 1.06 0.305 Fraud*Partner*Shift 1 0.68 0.68 0.85 0.357 Error 236 187.38 0.79 Panel B: Means (Standard Deviations) of Individual Auditor Fraud Risk Assessments Pre-Brainstorming and Post- Brainstorming for Managers, Seniors and Staff Experience level Pre and post Managers Seniors Staff Overall Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Prebrainstorming Postbrainstorming 7.08 (1.70) 7.60 (1.41) 6.13 (1.49) 6.40 (1.50) 6.38 (1.55) 7.20 (0.99) 6.40 (1.26) 7.05 (0.96) 5.93 (1.29) 7.15 (1.08) 6.15 (1.29) 7.15 (1.17) 6.46 (1.58) 7.32 (1.18) 6.23 (1.34) 6.87 (1.26) The analysis, shown in Table 4, Panel A reveals that the shift (repeated measures) factor is significant (p<0.001). The means displayed in Table 4, Panel B show that the overall mean fraud risk assessment in the fraud condition increased significantly from 6.46 pre-brainstorming to 7.32 post-brainstorming (p<0.001) and the overall mean fraud risk assessment in the no fraud 31

condition increased significantly from 6.23 pre-brainstorming to 6.87 post-brainstorming (p<0.001). To assess the individual auditors improvement, the pre-brainstorming and postbrainstorming fraud risk assessments of the individuals assigned to the fraud condition were compared to those assigned to the no fraud condition. As Figure 3 illustrates, an improvement is made in the individual judgments as a result of the brainstorming session. Even though there is no significant fraud*shift interaction (p=0.184), because of the importance of this issue in practice and given the issue of power in this design, a more sensitive t-test was performed. Paired samples t-test reveal no significant difference between the pre-brainstorming judgments of those individuals in the fraud condition (µ=6.46) and the no fraud condition (µ=6.23). Therefore, prior to the brainstorming session, on average, individual auditors are not assessing the risk of fraud as higher when fraud is present and lower when fraud is not present (p=0.209). However, after the brainstorming session (i.e., post-brainstorming), a paired samples t-test reveals there is a significant difference between risk assessments of the individuals in the fraud condition and those of the participants in the no fraud condition (p=0.003). The means presented in Table 4, Panel B, show that individuals in the fraud condition have significantly higher fraud risk assessments (µ= 7.32) than those individual auditors in the no fraud condition after brainstorming (µ= 6.87). These results support H 4. Pre-brainstorming vs. Post-brainstorming 7.40 Mean fraud risk assessment 7.20 7.00 6.80 6.60 6.40 6.46 (1.58) 7.32 (1.18) 6.87 (1.26) Fraud No fraud 6.20 6.23 (1.34) Pre-brainstorming Post-brainstorming Individual auditor shift FIGURE 3: 4: Comparison of Individual Auditors Fraud Risk Assessments 32

Taken together, the results in both the fraud condition and no fraud condition reveal a significant increase in fraud risk assessments of auditors after brainstorming. 19 This supports the suggestions made by standard setters--that brainstorming will likely remind auditors of the possibility that fraud exists on every audit, thus heightening their awareness of fraud whether fraud is present or not (AICPA 2003). This result highlights the tension between the objectives of standard setters and those of audit firms. When fraud is present, both parties would agree that increased fraud risk assessments after brainstorming is a positive result. However, when fraud is not present, increased fraud risk assessments may lead to over-auditing, thus validating concerns that audit firms have about the efficiency of audits upon implementing the new requirements of SAS No. 99. 4.2.5 Hypothesis 5 Hypothesis 5 predicts that brainstorming will increase individual auditors professional skepticism. This increase can be measured by a positive change in the difference between the professional skepticism scores from pre-brainstorming to post-brainstorming. Hypothesis 5 was tested using a repeated-measures ANOVA run at the individual auditor level. Professional skepticism was measured as the dependent variable, with fraud condition (fraud or no fraud), partner influence (efficiency or effectiveness), and rank (staff, senior or manager) as betweensubjects independent variables and a shift variable representing the change between prebrainstorming and post-brainstorming judgments. The analysis, shown in Table 5, Panel A reveals that the shift (repeated measures) factor is significant (p=0.007). No predictions were made about the influence of any of the independent variables, and as seen in Table 5, Panel A, there are no significant influences by any of the independent variables. While the shift is significant, the means displayed in Table 5, Panel B, show that the overall mean professional skepticism scores decreased in all conditions. The mean overall professional skepticism scores decreased significantly from (µ=141.70) prebrainstorming to (µ=140.75) post-brainstorming. However, additional analysis reveals that this significant decrease overall is being driven by significant decreases in three subscales, selfdetermining, search for knowledge and interpersonal understanding. Even though there is a significant increase in questioning mind, the overall professional skepticism score is a sum of the six subscales. Therefore, this increase in the questioning mind subscale is masked by the significant decreases in the three other subscales mentioned above. As displayed in Table 6 by condition, self-determining decreased significantly with overall mean scores decreasing from (µ=26.03) pre-brainstorming to (µ=25.58) postbrainstorming (p=0.007). Likewise, search for knowledge decreased significantly with overall mean scores decreasing from (µ=29.93) pre-brainstorming to (µ=29.58) post-brainstorming (p=0.001). Finally, lending to the overall decrease in professional skepticism, interpersonal understanding decreased significantly with overall mean scores decreasing from (µ=24.00) prebrainstorming to (µ=23.52) post-brainstorming (p<0.001). These significant decreases when summed together cancel the significant increase in questioning mind which increased significantly from (µ=12.73) pre-brainstorming to (µ=12.97) post-brainstorming (p=0.008). 19 Additional analysis using a repeated measures ANOVA, revealed auditors client business risk assessments also increased significantly (p<0.001) with fraud a significant factor (p=0.018), thus showing a similar effect in an overall risk measure. 33

To my knowledge, there are no other scales that measure professional skepticism other than the one used in this dissertation, that created by Hurtt (2003). The Hurtt (2003) scale was validated using auditors from a major international accounting firm. The mean professional skepticism of these auditors was tested twice in order to provide test-retest reliability. Auditors professional skepticism as measured in this dissertation is slightly higher than that of the 88 auditors evaluated in Hurtt (2003). The overall mean scores of professional skepticism from the 240 auditors studied in this dissertation as reported above: (µ=141.70) pre-brainstorming and (µ=140.75) post-brainstorming are higher than both mean scores (µ=138.50), (µ=135.40) reported in Hurtt (2003). The pre-brainstorming and post-brainstorming measures of auditors in this dissertation are separated only by the audit team brainstorming phase of the experiment. The two scores reported in Hurtt (2003) are separated by a mean time of 22 days. She used this lapse of time between dates to measure the internal consistency of the scores and reported Cronbach s alpha of.91, and a correlation between the first and second score of.85, significant at the.01 level. It is interesting to note the similar decrease between the Hurtt (2003) first and second test score and the decrease between auditors pre-brainstorming and post-brainstorming professional skepticism scores reported in this dissertation. Hurtt (2003) did not build and test theoretical constructs in her study, but instead she designed a scale to be used in academic research to measure professional skepticism. She does not discuss the decrease in professional skepticism between the first and second professional skepticism measurements in her paper. Without predictions of a change based on theory, one can only speculate as to the cause of this overall decrease in measurements across time. SAS No. 99 defines professional skepticism for auditors as an attitude that includes a questioning mind and a critical assessment of audit evidence (AICPA 2002, p. 10). It thus seems very important that the questioning mind subscale measure increased significantly, and it was the only value to increase significantly of the six measured. Some of the statements scored in the questioning mind subscale are: I frequently question things that I see or hear, and I often reject statements unless I have proof that they are true. These statements are scored on a scale from 1, strongly disagree to 6, strongly agree. These statements seem to reflect the questioning mind that auditors are supposed to have according to SAS No. 99. An increase in the auditors score on this subscale shows a propensity for them to agree more with these statements after brainstorming than before. The significant increase in this score thus suggests that auditors questioning mind, which is by definition professional skepticism in SAS No. 99, is enhanced by the new audit requirement to brainstorm about fraud. These findings thus provide partial support for H 5. 34

TABLE 5 Individual Overall Professional Skepticism Shifts Panel A: Results of a Repeated Measures ANOVA of Fraud, Partner Influence and Rank Between Subjects on Individual Overall Professional Skepticism Pre-Brainstorming and Post-Brainstorming Source of Variation Df SS MS F p-value Fraud (No Fraud) 1 60.92 60.92 0.19 0.664 Partner Influence 1 539.75 539.75 1.68 0.197 Rank 2 569.65 284.83 0.88 0.414 Shift (Post-Pre Professional Skepticism) 1 109.25 109.25 7.51 0.007 Fraud*Shift 1 0.75 0.75 0.05 0.820 Partner Influence*Shift 1 1.30 1.30 0.09 0.765 Rank*Shift 2 18.90 9.45 0.65 0.523 Error 228 3315.03 14.54 Panel B: Means (Standard Deviations) of Individual Auditor Professional Skepticism Pre-Brainstorming and Post- Brainstorming for Managers, Seniors and Staff Experience level Pre and post Managers Seniors Staff Overall Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Pre-brainstorming 140.80 (12.00) 145.50 (12.89) 141.80 (11.88) 144.00 (12.57) 144.70 (14.14) 140.45 (17.35) 142.43 (12.61) 143.20 (14.33) Post-brainstorming 141.20 (11.98) 142.95 (15.82) 140.60 (11.90) 144.35 (13.24) 144.40 (12.96) 137.05 (16.04) 142.07 (12.19) 141.45 (15.17) Pre-brainstorming 140.05 (10.05) Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud 137.95 (11.99) 137.85 (11.86) 138.80 (11.82) 146.25 (12.08) 142.65 (12.42) 140.18 (11.46) 141.00 (12.35) Post-brainstorming 138.95 (10.97) 137.95 (13.69) 136.00 (12.81) 139.00 (12.96) 141.45 (12.99) 145.10 (12.65) 138.80 (12.29) 140.68 (13.27) Overall Totals Pre-brainstorming 140.99 (11.85) 140.61 (12.06) 143.51 (14.05) 141.70 (12.71) Post-brainstorming 140.26 (13.14) 139.99 (12.86) 142.00 (13.84) 140.75 (13.26) 35

TABLE 6 Individual Professional Skepticism Subscale Shifts Means (Standard Deviations) of Individual Auditor Professional Skepticism Subscale Measurements Pre-Brainstorming and Post-Brainstorming for Managers, Seniors and Staff Self-determining (overall mean shift decrease, p=0.007) Experience level Pre and post Managers Seniors Staff Overall Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Pre-brainstorming 25.55 (3.62) Post-brainstorming 24.95 (2.98) Pre-brainstorming 26.30 (3.85) Post-brainstorming 25.30 (3.53) 27.95 (3.83) 28.20 (5.04) 24.80 (2.50) 24.15 (3.94) 26.65 (4.02) 26.75 (4.31) 27.05 (4.06) 26.35 (3.51) 24.90 (5.00) 24.25 (3.71) 25.80 (3.52) 25.15 (3.56) 26.50 (4.42) 26.40 (4.62) Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud 26.30 (3.40) 26.60 (3.62) 25.25 (3.99) 24.00 (3.95) 25.95 (4.27) 25.35 (4.18) 25.05 (4.02) 25.20 (3.83) 26.65 (3.33) 25.90 (3.77) 25.53 (3.93) 24.83 (3.76) 26.30 (3.64) 25.95 (3.83) Overall Totals Pre-brainstorming 26.53 (3.72) 25.66 (3.75) 25.91 (4.18) 26.03 (3.89) Post-brainstorming 26.26 (4.01) 25.06 (4.17) 25.43 (3.72) 25.58 (3.99) Self-confidence (overall mean shift increase, p=0.143) Experience level Pre and post Managers Seniors Staff Overall Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Pre-brainstorming 25.70 (2.05) Post-brainstorming 26.35 (2.62) Pre-brainstorming 25.30 (2.70) Post-brainstorming 25.65 (2.80) Overall Totals Pre-brainstorming 25.66 (2.67) 25.95 (3.00) 25.60 (2.80) 24.80 (4.06) 25.30 (3.71) 25.55 (3.63) 26.10 (2.99) 26.05 (2.82) 26.45 (2.63) 25.00 (4.28) 24.10 (4.80) 25.52 (3.08) 26.03 (3.03) 25.50 (3.64) 25.27 (3.68) Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud 25.70 (2.98) 26.00 (2.71) 25.15 (2.62) 25.40 (2.62) 25.04 (3.17) 24.65 (2.18) 25.00 (2.29) 25.50 (3.25) 25.45 (3.25) 25.44 (3.38) 25.20 (3.16) 25.10 (3.42) 25.32 (2.83) 25.50 (2.86) 25.38 (3.08) 25.18 (2.79) 25.37 (2.83) Post-brainstorming 25.90 (2.70) 25.45 (2.92) 25.28 (3.64) 25.54 (3.11) 36

TABLE 6 - continued Individual Professional Skepticism Subscale Shifts Means (Standard Deviations) of Individual Auditor Professional Skepticism Subscale Measurements Pre- Brainstorming and Post-Brainstorming for Managers, Seniors and Staff Suspension of judgment (overall mean shift decrease, p=0.531) Experience level Pre and post Managers Seniors Staff Overall Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Pre-brainstorming 23.70 (3.66) Post-brainstorming 24.15 (3.53) Pre-brainstorming 23.25 (2.71) Post-brainstorming 22.75 (3.27) Overall Totals 24.40 (3.12) 23.65 (3.77) 24.80 (3.24) 24.35 (3.33) 23.70 (1.72) 23.70 (2.25) 23.60 (3.41) 23.60 (3.15) 24.20 (3.47) 24.10 (3.71) 24.03 (3.42) 24.03 (3.30) 24.10 (2.84) 23.82 (3.27) Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud 22.55 (2.76) 22.60 (2.50) Pre-brainstorming 23.48 (3. 10) 22.40 (3.30) 22.80 (3.11) 23.41 (3. 05) 22.75 (3.26) 23.30 (3.56) 23.55 (3.17) 23.20 (2.65) 24.65 (3.51) 24.40 (3.52) 24.00 (3. 36) 23.07 (3.06) 22.92 (2.98) 23.32 (3.28) 23.43 (3.26) 23.63 (3. 17) Post-brainstorming 23.29 (3. 30) 23.54 (3. 09) 23.83 (3. 25) 23.55 (3. 21) Search for knowledge (overall mean shift decrease, p=0.001) Experience level Pre and post Managers Seniors Staff Overall Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Pre-brainstorming 29.45 (3.72) Post-brainstorming 29.35 (3.67) Pre-brainstorming 28.60 (3.02) Post-brainstorming 28.55 (2.78) Overall Totals Pre-brainstorming 29.20 (3.98) 30.05 (5.42) 29.00 (6.11) 30.70 (3.34) 30.50 (3.43) 30.45 (3.63) 30.40 (3.70) 30.95 (3.50) 30.95 (3.61) 29.90 (4.25) 28.95 (4.45) 30.37 (3.53) 30.27 (3.57) 30.13 (4.43) 29.45 (4.83) Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud 28.70 (3.47) 27.65 (3.70) 29.15 (3.88) 28.55 (4.01) 29.88 (3.58) 29.20 (3.43) 29.10 (3.80) 30.50 (4.48) 30.45 (4.71) 30.73 (3.89) 31.55 (3.33) 31.55 (2.89) 29.42 (3.86) 29.18 (3.95) 29.93 (3.86) 29.82 (3.58) 29.43 (3.79) Post-brainstorming 28.64 (4.22) 29.64 (3.77) 30.48 (4.02) 29.58 (4.06) 37

TABLE 6 - continued Individual Professional Skepticism Subscale Shifts Means (Standard Deviations) of Individual Auditor Professional Skepticism Subscale Measurement Pre- Brainstorming and Post-Brainstorming for Managers, Seniors and Staff Interpersonal understanding (overall mean shift decrease, p<0.001) Experience level Pre and post Managers Seniors Staff Overall Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Pre-brainstorming 23.70 (3.37) Post-brainstorming 23.30 (3.81) Pre-brainstorming 24.00 (2.64) Post-brainstorming 23.55 (3.66) Overall Totals 24.30 (4.13) 23.75 (4.68) 23.35 (3.34) 23.55 (4.01) 24.25 (4.24) 23.80 (4.34) 24.65 (3.94) 24.20 (4.21) 23.85 (4.23) 22.85 (4.21) 23.90 (3.54) 23.68 (3.96) 24.13 (4.14) 23.47 (4.36) Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud 22.55 (3.61) 22.10 (3.82) Pre-brainstorming 23.64 (3. 48) 23.20 (3.97) 22.40 (4.33) 23.80 (3.25) 23.10 (3.51) 23.65 (3. 68) 25.30 (3.42) 24.60 (3.65) 25.10 (4.44) 25.05 (4.38) 24.73 (3. 99) 24.17 (3.44) 23.52 (3.93) 23.82 (3.88) 23.42 (4.05) 24.00 (3. 74) Post-brainstorming 23.18 (3. 99) 23.21 (4. 02) 24.18 (4. 13) 23.52 (4. 06) Questioning mind (overall mean shift increase, p=0.008) Experience level Pre and post Managers Seniors Staff Overall Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Efficiency Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud Pre-brainstorming 12.70 (2.27) Post-brainstorming 13.10 (2.15) Pre-brainstorming 12.60 (2.28) Post-brainstorming 13.15 (2.06) Overall Totals Pre-brainstorming 12.49 (2.23) 12.50 (1.61) 12.75 (1.68) 13.35 (1.95) 12.75 (1.55) 13.40 (2.56) 13.60 (2.21) 12.40 (2.60) 12.85 (2.37) 12.60 (2.80) 12.80 (2.55) 12.82 (2.29) 12.90 (2.02) 12.83 (2.37) 13.05 (2.17) Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Effectiveness Partner Influence Fraud No Fraud Fraud No Fraud Fraud No Fraud Fraud No Fraud 12.15 (2.74) 13.00 (2.38) 12.70 (2.23) 12.85 (2.11) 12.98 (2.29) 12.45 (2.42) 13.15 (2.45) 12.75 (2.55) 12.55 (2.44) 12.71 (2.55) 13.10 (2.38) 13.10 (2.55) 12.68 (2.32) 12.85 (2.18) 12.73 (2.36) 12.57 (2.51) 13.08 (2.42) Post-brainstorming 13.00 (2.05) 13.09 (2.09) 12.83 (2.44) 12.97 (2.19) 38

4.3 Summary of Chapter 4 This chapter described the statistical techniques used to test the hypotheses developed in Chapter 2 of this dissertation. Hypothesis 1 predicted that teams of auditors with knowledge of partner s effectiveness preference would spend more time brainstorming than would teams of auditors with knowledge of a partner s preference for efficiency. This hypothesis was supported. Hypothesis 2 predicted that the fraud risk assessments of the brainstorming team would be more effective (i.e., higher when fraud is present and lower when fraud is not present) than the nominal group. This hypothesis was supported in the fraud condition. Hypothesis 3 predicted that brainstorming would influence auditors fraud judgments such that the number of ideas generated by the interacting group would be fewer than the number of ideas generated by the nominal group. This hypothesis was supported. Hypothesis 4 predicted individual auditor judgments on the risk of financial statement fraud would increase, and would increase more when fraud was present as a result of brainstorming. This hypothesis was supported. Hypothesis 5 predicted that brainstorming would increase individual auditor s professional skepticism. This hypothesis was partially supported as additional analysis revealed the auditors questioning mind increased, which is defined as professional skepticism for auditors in SAS No. 99. Chapter 5 will conclude the dissertation with a summary and a discussion of the contributions and limitations. Chapter 5 will also provide suggestions for future research. 39

CHAPTER 5 CONCLUSION This chapter concludes the dissertation with a discussion of the results in Section 5.1, a summary of the contributions in Section 5.2, and an evaluation of the limitations and suggestions for future research in Section 5.3. 5.1 Discussion of Results The results suggest that the partner s preferences for efficiency or effectiveness influence the amount of time that audit teams spend brainstorming. Therefore, the concerns standard setters have about the tone at the top set by engagement team leaders and the potential negative effects these leaders could have on brainstorming sessions, especially those leaders who encourage efficiency over effectiveness, seem justified. It appears that partners should be aware of their influence on subordinates (i.e., less experienced members of audit teams) and should follow the guidance set forth in SAS No. 99 to maximize the benefits of brainstorming (AICPA 2003). The results also suggest that, while an interacting hierarchical team of auditors (i.e., staff, senior and manager) generate fewer ideas than the nominal group, the brainstorming session causes an increase in the risk assessments of the individual members of the team. This seems to support the conjecture made by the standard setters that the mere act of brainstorming with engagement team members reminds auditors of the possibility that fraud can exist on any audit, thus improving their overall ability to detect fraud. 20 The professional skepticism results seem to suggest that auditors questioning mind is elevated during brainstorming. The professional skepticism scale used (Hurtt 2003) may contain other measures of skepticism that the standard setters did not envision when proposing that brainstorming should increase auditors professional skepticism. Since professional skepticism is defined by standard setters as an attitude that includes a questioning mind (AICPA 2002), the results suggest that the questioning mind subscale of the overall professional skepticism scale does increase as a result of brainstorming. 20 Improvements in the judgments of junior members of the team from brainstorming and interacting with the manager while not significant, are to some extent supported from various written protocols provided by senior and staff participants. For example, written in the comments section of the experimental instrument by a senior, I enjoyed the study - learned a lot!, and from a staff, I knew very little in the individual section and learned a lot in the brainstorming. 40

This finding seems to suggest yet another benefit of the brainstorming session as prescribed in SAS No. 99. 5.2 Contributions This dissertation s results contribute to the understanding of brainstorming and provide insight to practicing auditors, standard setters and to academic research. With the implementation of brainstorming as required by SAS No. 99, standard setters and the Fraud Task Force were concerned about the tone at the top set by engagement partners over these brainstorming sessions (AICPA 2003). The results of this dissertation validate the concern that the partner does influence more junior members of the engagement team. Therefore standard setters warnings to create a proper environment in order to maximize the benefits of brainstorming seem warranted. Audit partners are currently receiving specific training among the Big 4 firms about how to create the appropriate environment, and this training is likely to improve the way they lead each session. Tension exists between the standard setters recommendations for SAS No. 99 implementation and audit firms concerns about the efficiency of audits. The results of this dissertation provide support to practicing auditors concerns about the costs of these new brainstorming procedures and whether the benefits will outweigh these costs. The results suggest that on average auditors fraud risk assessments increase overall after brainstorming. Standard setters and practicing auditors can both see value in these increased assessments when fraud is present, but when fraud is not, these increased fraud risk assessments will likely lead to additional costs on the audit where the benefits can not be seen. This struggle is likely to continue throughout the implementation of this standard especially with the recent litigation against audit firms for failure to detect fraud. The results from this dissertation provide support for audit firms efficiency concerns and provide an opportunity for discussion as to the benefits of brainstorming when fraud is not present. Some of the benefits of brainstorming, as prescribed by SAS No. 99, seem apparent in the improvements among individual auditors fraud risk assessments and their questioning mind (i.e., professional skepticism as defined by SAS No. 99) after brainstorming. While fraud risk assessments did increase overall after brainstorming, the increase was greater when fraud was present than when it was not. The significant difference in fraud risk assessments between the fraud and no fraud condition was only evident in audit managers judgments prior to brainstorming and was not significant overall. However, after brainstorming, the difference between the fraud and no fraud judgments are significantly different and the shift between prebrainstorming and post-brainstorming judgments is also significant. This offers evidence that the standard setter s objective for junior members to learn from more experienced members of the team is likely occurring allowing, on average, all members of the team to benefit from the discussion about fraud. Additionally, auditors questioning mind did increase as a result of brainstorming. Again, this objective of SAS No. 99 seems to have been met on average for all auditors, regardless of level in the firm (i.e., staff, senior and manager). This benefit is likely to provide more inexperienced auditors with a reminder that fraud can exist on any audit and heighten their awareness and curiosity when completing much of their audit work. In general, the 41

staff and senior auditors seemed to value the brainstorming experience as evidenced by their written comments. This dissertation contributes to the fraud and multi-auditor literature by extending the findings of fraud judgments provided by individual auditors with varying levels of experience (Knapp and Knapp 2001) to those of an interacting hierarchical team of auditors (staff, senior and manager). While experience of the individual auditor is a significant factor in the individual auditor s effectiveness at detecting fraud, the interaction with other members of the team with different experience levels provides an opportunity for individual improvements to be made at all levels within the team. The results offer a contribution to the accountability literature by extending the accountability effects on individuals to an interacting team. It is interesting that those teams influenced by a partner with preferences for efficiency spent less time brainstorming than those influenced by an effectiveness partner. The results of this dissertation also offer a contribution to the brainstorming literature by showing that, even though overall losses in the number of ideas generated by interacting teams of auditors seem to mimic those of other interacting teams in a majority of brainstorming studies (e.g., Diehl and Stroebe 1987, Nunamaker et al. 1991), there are other possible benefits to brainstorming that extend beyond recording as many ideas as possible. At least in this context, brainstorming seems to offer a platform for auditors to be reminded that the possibility of fraud exists and could exist on any audit, and it provides them with a unique opportunity to explore these possibilities. Overall, these results suggest that brainstorming sessions, now required on all audits, can improve auditors fraud judgments. The implications extend beyond answering questions about the effectiveness of brainstorming on auditor s fraud detection and provide hope for auditors, standard setters and the profession that, as these brainstorming sessions continue to be performed on audit engagements in the future, we may see an overall improvement in auditors ability to detect fraud. 5.3 Limitations and Suggestions for Future Research While this dissertation investigates differences in fraud risk assessments before and after brainstorming and the ideas generated during individual and group brainstorming, it does not address the other areas of planning for detection of fraud in an engagement and it only evaluates auditors judgments about financial statement fraud not misappropriation of assets. It would be interesting to determine if effectiveness in fraud judgments (i.e., higher fraud judgments when fraud is present and lower when fraud is not) is associated with the appropriate testing to match these assessments. Zimbelman (1997) and Glover et al. (2003) suggest that auditors adjust the extent of their testing but not the nature of their tests. Future research can examine whether or not auditors are able to generate more appropriate tests during the brainstorming discussion that lead to a better match between risk assessment and audit procedures. In order to control for variability across audit teams, this dissertation uses randomly assigned staff, senior and managers to teams of three for the brainstorming phase. While this dissertation makes a contribution in that it evaluates an interacting hierarchical team, to the extent that these hierarchical roles would have been different in a practiced (i.e., natural) team, the results may vary. The partner influence used in this dissertation is similar to other studies 42

(Peecher 1996; Turner 2001; Wilks 2002) allowing the influences to be controlled in the experimental design. If a practiced team is investigated, it may be useful to measure the partner s influence by having the partner participate in the brainstorming. However, this will be a challenging task due to the number of auditors needed to participate in a controlled design such as the one employed in this dissertation. The professional skepticism scale used in this dissertation (Hurtt 2003), while recognized in the academic literature, has not been used in published studies to date. The validity of the instrument, whether it actually measures the professional skepticism that it purports to measure, has thus not been sufficiently determined. As Hurtt (2003) suggests, additional research should be done by multiple researchers to allow for this type of validation and comparison. Several suggestions have been made to improve group interaction that would be useful to explore, including group strategies to select their best member (Libby, Trotman and Zimmer 1987; Henry 1995) and designing interactive sessions to encourage sharing of all pertinent information (Stasser, Stewart and Wittenbaum 1995). The current study is designed to test the benefits of brainstorming as prescribed by SAS No. 99, but evaluating group interaction techniques to create additional process gains could be very useful to auditors in practice. Each of these areas provides important avenues for future research to help auditors better detect fraud before financial statements are issued. Research in any of these areas is important to increase our understanding of this complex and critical issue that continues to plague our profession today. 43

APPENDIX A EXPERIMENTAL INSTRUMENT FRAUD RISK ASSESSMENT CASE PART I Please print your name and address of the practice office below. Name (printed): Practice Office Address: The results of this study will be published in the aggregate and your materials will remain confidential, to the extent allowed by law, only the property of the researcher, Tina Carpenter. The results of this research study may be published, but your name will not be used. The data will be locked in the researcher s office and stored electronically with password protection. If you would like to learn more about the results of this study, please contact Tina Carpenter, doctoral student at tld8218@cob.fsu.edu. You may also contact Jane Reimers, doctoral advisor at jreimer@cob.fsu.edu. If you have any questions about your rights as a subject/participant in this research, or if you feel you have been placed at risk, you can contact the Chair of the Human Subjects Committee, Institutional Review Board, through the Vice President for the Office of Research at (850) 644-8633. 44

GREETING AND INTRODUCTION (Read aloud by experimenter) Hello, my name is Tina Carpenter and I am a Ph.D. student at Florida State University. I once worked as a CPA and public accountant just like you and I have worked diligently will several partners from Big 4 firms to make this project as useful and interesting to you as possible. Thank you participating in my dissertation research today. I know how valuable your time is, and I therefore appreciate you taking part in my experiment. The partners in your office strongly support this project and all of the materials have been reviewed by partners and managers among the Big 4 firms and have been tested through other audit teams. The results of this experiment will be sent to and reviewed by partners in your office and you will thus be asked to fill out your name and practice office. Since your name will be on the materials, please take great care and thought in formulating your responses as you would on an audit in practice. At this time, please take a minute to print your name and practice office on the front page of your booklets. PAUSE Now that you have all put your name on the booklets, I would like you to turn to the first page where you will be doing a warm up exercise in order to help prepare you for the remainder of the materials that follow. You will be given 2 minutes to complete this exercise from when I say start. When time is up, I will ask you to stop and put down your pencils. What I am about to read is printed at the top of the page where you will be working, so it is not necessary to read it again. Your task is as follows: A common self-test of creative thinking is the Paper Clip Quiz. Here is how it goes: you will have two minutes to brainstorm and list all possible uses for an ordinary paper clip. Write down your answers as fast as you can. I ll call STOP at the end of two minutes. PAUSE Now that you have completed this exercise, you may turn the page again and begin. Please make sure to CAREFULLY read each set of instructions. You will be given 45 minutes to complete this next section. I will ask you to Stop at the end of those 45 minutes. You may begin. PAUSE Please take a 5 minute break while I get assistance copying portions of your materials so that you can use them during the remainder of the experiment. Please do not discuss any of the experiment with others in the room. When I return, you will be given copies and we will begin Part II in your teams. PAUSE 45

PART II INSTRUCTIONS (Read aloud by experimenter) Please turn over your case booklets. The member of the team with the sticker has been randomly chosen to be the group recorder. The group recorder will be recording the group response for this entire part. The group recorder should take great care to write down everything that is discussed in the group (as many unique ideas as possible). Please signal to the experimenter who your group recorder is by raising your hand so that you can be handed the materials to complete for the group. DISCUSSION AMONG ENGAGEMENT PERSONNEL REGARDING THE RISKS OF MATERIAL MISSTATEMENT DUE TO FRAUD SAS No. 99, Consideration of Fraud in a Financial Statement Audit, suggests the overall objective of brainstorming is to identify how and where the entity s financial statements might be susceptible to material misstatement due to fraud and how management could perpetrate and conceal it. It suggests that the discussion among the audit team members about the susceptibility of the entity s financial statements to material misstatement due to fraud should include a consideration of the known external and internal factors affecting the entity that might a) create incentives/pressures for management and others to commit fraud, b) provide the opportunity for fraud to be perpetrated, and c) indicate a culture or environment that enables management to rationalize committing fraud. The discussion should occur with an attitude that includes a questioning mind, setting aside any prior beliefs the audit team members have that management is honest and has integrity. In this regard, the discussion should include a consideration of the risk of management overriding controls. The discussion among the audit team members should emphasize the need to maintain a questioning mind and to exercise professional skepticism. It is VERY important for each group recorder to read aloud the instructions provided in each group task booklet for PART II. Please realize that there are no right or wrong answers and your case is most likely different than any other group in the room. Therefore, it is of no use to listen to other group conversations. Please focus on these tasks as your own group, as you did when performing the tasks earlier individually. Please signal to me when you have completed this session and I will provide you with Part III, the final materials, which you will complete individually. 46

Warm Up Exercise A common self-test of creative thinking is the Paper Clip Quiz. Here is how it goes: you will have two minutes to brainstorm and list all possible uses for an ordinary paper clip in the space provided below. Write down your answers as fast as you can. I ll call STOP at the end of two minutes. 47

INSTRUCTIONS Participation is voluntary and there is no penalty for non-participation. If you would like to participate, by continuing on at this point, you are providing consent. Thank you for agreeing to participate in this study. The purpose of this study is to investigate judgments auditors make during financial statement audits. More precisely, the study examines how auditors formulate judgments when performing analytical procedures and assessing the risk of fraud. In the following pages, you will be asked to complete several tasks. Please recognize that there are no right or wrong answers, as most of the questions deal with matters of professional judgment, so please provide your best judgments. Your individual responses will be coded with a corresponding participant number and all results will be reported in aggregate, not revealing your identity and remaining confidential to the extent allowed by law. SAS No. 99, Consideration of Fraud in a Financial Statement Audit, requires members of the audit team to discuss the potential for material misstatement due to fraud. The discussion should include: an exchange of ideas or brainstorming among the audit team members about how and where they believe that entity s financial statements might be susceptible to material misstatement due to fraud and how management could perpetrate and conceal fraudulent financial reporting. For purposes of this standard, fraud is an intentional act that results in a material misstatement in the financial statements that are the subject of an audit. The standard also suggests that the auditor should consider any unusual or unexpected relationships that have been identified in performing analytical procedures in planning the audit. Analytical procedures and analysis of background information about the client should be performed by each individual prior to the group brainstorming in order to provide a similar level of understanding about the client. Mimicking the guidance set forth in SAS No. 99, you will participate in a study today designed with three parts. Part I will ask that you perform analytical procedures on the financial statement information provided to you. You will then be asked to brainstorm individually, providing an initial list of unusual items, and identifying how and where the client s financial statements might be susceptible to material misstatement due to fraud. Additionally, you will list ways management could perpetrate and conceal this fraud, and you will be asked for your overall fraud risk assessment. When you have individually completed Part I, Part II will require that you work with two other auditors in a group for the brainstorming session described in SAS No. 99. Part II will ask you (as part of this group) to provide a list of how and where the client s financial statements might be susceptible to material misstatement due to fraud; how management could perpetrate and conceal fraud; and for your group s overall fraud risk assessment. When you have completed the group brainstorming and associated tasks, all of your materials will be collected. You will then be asked to complete the final set of tasks -Part III- individually. If you have any questions during this study, please ask me, but please don t talk to others about this study while you are taking part in it. In Part II, you will be asked to discuss items with other members of your group; however you should only confer with these group members at this specified time. Additionally, because others from your firm and other firms in this area will also be participating in this research at a later time, please do not talk to others about this study for at least two weeks after your participation. Thank you, Tina D. Carpenter Florida State University 48

PART I You have been provided a case booklet concerning an audit client. The case booklet is separated from this packet for easy reference and making notes. The booklet includes the following information: A Narrative Description of the Company Comparative Income Statements Comparative Balance Sheets Notes to the Financial Statements Common-sized Balance Sheets and Income Statements Financial Ratios Assume that you are in the preliminary planning stage of this client s audit for fiscal 2000. No audit procedures have been performed other than general information gathering. All of the information available to you about this Company is included in the case booklet. Please perform preliminary analytical procedures on the financial statements for 2000 with the specific objective of assessing the likelihood that financial statement fraud is or is not present. In Part I you will be asked to first indicate (there is paper available for this purpose) unusual items noted and to generate what you consider to be the most likely explanation for the unusual items identified. In developing your explanations, you may refer to the case information, perform calculations, and make assumptions, if necessary. Second, you will be asked to brainstorm individually listing how and where you believe the Company s financial statements may be susceptible to material misstatement due to fraud and how management could perpetrate and conceal fraudulent financial reporting. Third, you will be asked to provide a fraud risk assessment and complete an individual questionnaire. You will be given approximately 45 minutes to complete the three tasks described above. Feel free to refer to the case booklet at any time while completing these tasks. While helping to construct this case, the engagement partner on this audit expressed numerous times his concern about the associated costs of implementing brainstorming on this engagement. Specifically, he is concerned about the audit team members being overly sensitive to unusual account balance fluctuations noted in the initial analytical procedures, as this sensitivity may lead to costly increases in unjustified investigations and efficiency losses on the audit. He would like the assessment of fraud risk and the associated brainstorming to be sufficient to comply with the standard, but he hopes that you will be aware of the costs and complete this phase of the audit as efficiently as possible. This partner will review your responses in the individual and group phases, so please take great care and thought in formulating your responses as you would on an audit in practice. Please list three reasons that auditors can sometimes be overly sensitive to the cost of unjustified investigations, (i.e., efficiency), when evaluating a client s fraud risk in the planning stage of an engagement. 1. 2. 3. Please number these reasons in order of their impact on auditor s judgments (1 representing the highest impact and 3 the lowest). Please note the current time, this time will mark the beginning of the task on the next page. 49

TASK I Unusual Items Noted During Analytical Procedures Please indicate unusual items noted and generate what you consider to be the most likely explanation for the unusual items identified. In developing your explanations, you may refer to the case information, perform calculations, and make assumptions, if necessary. Feel free to refer to the case booklet at any time while completing this task. Additionally, the calculations provided in the common sized balance sheet, income statement and the financial ratios have been recalculated and are accurate. Unusual items noted Most likely explanation for this unusual item 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 50

TASK II How and Where the Financial Statements are Susceptible to Fraud and How Management Could Perpetrate and Conceal It Please brainstorm, listing how and where you believe the client s financial statements just reviewed might be susceptible to material misstatement due to fraud, and how management could perpetrate and conceal fraudulent financial reporting. In developing this list, you may refer to the case information and any prior work, if necessary. It may be helpful to identify events or conditions that indicate incentives/pressures to perpetrate fraud, opportunities to carry out the fraud, or attitudes/rationalizations to justify a fraudulent action. These events are referred to in SAS No. 99 as fraud risk factors. Fraud risk factors do not necessarily indicate the existence of fraud; however, they often are present in circumstances where fraud exists. Feel free to refer to the case booklet at any time while completing this task. You are to provide as many unique ideas as possible, as there are no dumb ideas. How and where you believe the client s financial statements might be susceptible to material misstatement due to fraud? How management could perpetrate and conceal this fraudulent financial reporting? 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 51

TASK III Individual Risk Assessments INSTRUCTIONS When responding to the following questions, please do not look back to the case materials. SAS No. 56, Analytical Procedures, states that planning analytical procedures should focus on identifying areas that may represent specific risks relevant to the audit. Based on your review of the financial statements, what is your judgment of the likelihood of the following risks relevant to this client? Circle one number on the scale below corresponding to your judgment. For example, if after reading a statement you believe the likelihood of risk is neutral, you might respond by circling 5, as shown below. Extremely [ ] Extremely Unlikely 0 1 2 3 4 5 6 7 8 9 10 Likely Neutral 52

TASK III When responding to the following questions, please do not refer back to the case materials. Based on your review of the financial statements, what is your judgment of the likelihood of the following risks relevant to this client? Circle one number on the scale below corresponding to your judgment. FRAUD RISK LIKELIHOOD OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Extremely [ ] Extremely Unlikely 0 1 2 3 4 5 6 7 8 9 10 Likely Neutral SIGNIFICANCE OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Highly [ ] Highly Insignificant 0 1 2 3 4 5 6 7 8 9 10 Significant Neutral PERVASIVENESS OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Extremely [ ] Extremely Non-Pervasive 0 1 2 3 4 5 6 7 8 9 10 Pervasive Neutral CLIENT BUSINESS RISK: Risk that this entity s business objectives will not be attained as a result of the external and internal factors, pressures, and forces brought to bear on the entity and, ultimately, the risk associated with the entity s survival and profitability. Extremely [ ] Extremely Low 0 1 2 3 4 5 6 7 8 9 10 High Neutral If you believe that the risk of material misstatement due to fraud is likely, can it be related to specific financial statement account balances or classes of transactions? Yes No Please indicate any overall trends in the financial statement information or any other information that signal a red flag to you that may heighten the risk of material misstatement due to fraud. (List all of them below). What is your response to the risk of material misstatement of the financial statements due to fraud (i.e., what audit procedures do you intend to perform if any)? 53

TASK IV Statements that people use to describe themselves are given below. Please select the response that indicates how you generally feel by circling the number that corresponds best to your feeling. There are no right or wrong answers. Do not spend too much time on any one statement. Strongly Strongly Disagree Agree 1. I often accept other peoples explanations without further thought.... 1 2 3 4 5 6 2. I feel good about myself... 1 2 3 4 5 6 3. I wait to decide on issues until I can get more information... 1 2 3 4 5 6 4. The prospect of learning excites me... 1 2 3 4 5 6 5. I am interested in what causes people to behave the way that they do... 1 2 3 4 5 6 6. I am confident of my abilities... 1 2 3 4 5 6 7. I often reject statements unless I have proof that they are true... 1 2 3 4 5 6 8. Discovering new information is fun... 1 2 3 4 5 6 9. I take my time when making decisions... 1 2 3 4 5 6 10. I tend to immediately accept what other people tell me... 1 2 3 4 5 6 11. Other peoples behavior doesn t interest me... 1 2 3 4 5 6 12. I am self-assured... 1 2 3 4 5 6 13. My friends tell me that I usually question things that I see or hear... 1 2 3 4 5 6 14. I like to understand the reason for other peoples behavior... 1 2 3 4 5 6 15. I think that learning is exciting... 1 2 3 4 5 6 16. I usually accept things I see, read or hear at face value... 1 2 3 4 5 6 17. I don t feel sure of myself... 1 2 3 4 5 6 18. I usually notice inconsistencies in explanations... 1 2 3 4 5 6 19. Most often I agree with what the others in my group think... 1 2 3 4 5 6 20. I dislike having to make decisions quickly... 1 2 3 4 5 6 21. I have confidence in myself... 1 2 3 4 5 6 22. I don t like to decide until I ve looked at all of the readily available information... 1 2 3 4 5 6 23. I like searching for knowledge... 1 2 3 4 5 6 24. I frequently question things that I see or hear... 1 2 3 4 5 6 Please turn over the page and continue answering questions. 54

Strongly Strongly Disagree Agree 25. It is easy for other people to convince me... 1 2 3 4 5 6 26. I seldom consider why people behave in a certain way... 1 2 3 4 5 6 27. I like to ensure that I ve considered most available information before making a decision... 1 2 3 4 5 6 28. I enjoy trying to determine if what I read or hear is true... 1 2 3 4 5 6 29. I relish learning... 1 2 3 4 5 6 30. The actions people take and the reasons for those actions are fascinating... 1 2 3 4 5 6 55

You have now completed Part I, the first individual portion of this exercise. Please again note the current time, your completion time for this task. You will turn in this portion and the contents will be copied. The original will be retained by the experimenter to ensure that there are no changes made to this section of materials and a copy of your responses will be returned to you for your use in other parts of this experiment. Please signal to the experimenter that you have completed this section at this time. 56

FRAUD RISK ASSESSMENT CASE PART II Please print the practice office followed by each individual participant of this group below to aid in keeping your responses in part one matched to your group responses in part two. Additionally, please designate the position of each individual (i.e., staff, senior, manager) and who was randomly selected to be the group recorder. Practice Office: Names and position (printed): Group Recorder Name Please introduce yourselves to each other at this time and write out a name tag to wear during this phase (a nametag and pen have been provided to you for this purpose). 57

PART II INSTRUCTIONS Please read the following information and begin brainstorming as a group. All activities in this part of the case will be done as a group. You are to provide as many unique ideas as possible, as there are no dumb ideas. Recall, that while helping to construct this case, the engagement partner on this audit expressed numerous times his concern about the associated costs of implementing brainstorming on this engagement. Specifically, he is concerned about the audit team members being overly sensitive to unusual account balance fluctuations noted in the initial analytical procedures, as this sensitivity may lead to costly increases in unjustified investigations and efficiency losses on the audit. He would like the assessment of fraud risk and the associated brainstorming to be sufficient to comply with the standard, but he hopes that you will be aware of the costs and complete this phase of the audit as efficiently as possible. Please note the current time, this time will mark the beginning of the task on the next page. 58

TASK I How and Where the Financial Statements are Susceptible to Fraud and How Management Could Perpetrate and Conceal It (Group Response) Please brainstorm, listing how and where you believe the client s financial statements just reviewed might be susceptible to material misstatement due to fraud, and how management could perpetrate and conceal fraudulent financial reporting. In developing this list, you may refer to the case information and any prior work, if necessary. It may be helpful to identify events or conditions that indicate incentives/pressures to perpetrate fraud, opportunities to carry out the fraud, or attitudes/rationalizations to justify a fraudulent action. These events are referred to in SAS No. 99 as fraud risk factors. Fraud risk factors do not necessarily indicate the existence of fraud; however, they often are present in circumstances where fraud exists. Feel free to refer to the case booklet at any time while completing this task. You are to provide as many unique ideas as possible, as there are no dumb ideas. How and where you believe the client s financial statements might be susceptible to material misstatement due to fraud? How management could perpetrate and conceal this fraudulent financial reporting? 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 59

TASK II- Group Risk Assessments When responding to the following questions, please do not look back to the case materials. Based on your review of the financial statements, what is your judgment of the likelihood of the following risks relevant to this client? Circle one number on the scale below corresponding to your judgment. Because this is a group judgment, it should reflect the consensus judgment of the group, not the individual response of the group recorder. FRAUD RISK LIKELIHOOD OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Extremely [ ] Extremely Unlikely 0 1 2 3 4 5 6 7 8 9 10 Likely Neutral SIGNIFICANCE OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Highly [ ] Highly Insignificant 0 1 2 3 4 5 6 7 8 9 10 Significant Neutral PERVASIVENESS OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Extremely [ ] Extremely Non-Pervasive 0 1 2 3 4 5 6 7 8 9 10 Pervasive Neutral CLIENT BUSINESS RISK: Risk that this entity s business objectives will not be attained as a result of the external and internal factors, pressures, and forces brought to bear on the entity and, ultimately, the risk associated with the entity s survival and profitability. Extremely [ ] Extremely Low 0 1 2 3 4 5 6 7 8 9 10 High Neutral If you believe that the risk of material misstatement due to fraud is likely, can it be related to specific financial statement account balances or classes of transactions? Yes No Please indicate any overall trends in the financial statement information or any other information that signal a red flag to you that may heighten the risk of material misstatement due to fraud. (List all of them below). What is your response to the risk of material misstatement of the financial statements due to fraud (i.e., what audit procedures do you intend to perform if any)? 60

Group recorder: Please read aloud. You have now completed Part II, the group portion of this exercise. Please again note the current time, the time that you completed this part. At this time you will be given Part III of this case where you are again asked to respond individually. Please do not discuss this section with your group members and please do not simply write down your prior individual or group responses unless these responses represent what you believe during this final individual phase. Please signal to the experimenter that you have completed this section at this time. 61

FRAUD RISK ASSESSMENT CASE PART III Please print the practice office followed by your name below to aid in keeping your responses in part one and your group responses in part two matched to your responses in this part, Part III. Practice Office: Name (printed): 62

PART III INSTRUCTIONS This is the final part of your participation. This part is to be completed individually without conferring with any other members of your group or anyone else in the room. Please do not feel like you must denote the group response from the last part or your initial individual responses from the first part unless these responses reflect your individual assessment at this time. 63

TASK I When responding to the following questions, please do not refer back to the case materials. Based on your review of the financial statements, what is your judgment of the likelihood of the following risks relevant to this client? Circle one number on the scale below corresponding to your judgment. FRAUD RISK LIKELIHOOD OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Extremely [ ] Extremely Unlikely 0 1 2 3 4 5 6 7 8 9 10 Likely Neutral SIGNIFICANCE OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Highly [ ] Highly Insignificant 0 1 2 3 4 5 6 7 8 9 10 Significant Neutral PERVASIVENESS OF RISK: Financial statement fraud (an intentional act that results in a material misstatement in the financial statements) is present. Extremely [ ] Extremely Non-Pervasive 0 1 2 3 4 5 6 7 8 9 10 Pervasive Neutral CLIENT BUSINESS RISK: Risk that this entity s business objectives will not be attained as a result of the external and internal factors, pressures, and forces brought to bear on the entity and, ultimately, the risk associated with the entity s survival and profitability. Extremely [ ] Extremely Low 0 1 2 3 4 5 6 7 8 9 10 High Neutral If you believe that the risk of material misstatement due to fraud is likely, can it be related to specific financial statement account balances or classes of transactions? Yes No Please indicate any overall trends in the financial statement information or any other information that signal a red flag to you that may heighten the risk of material misstatement due to fraud. (List all of them below). What is your response to the risk of material misstatement of the financial statements due to fraud (i.e., what audit procedures do you intend to perform if any)? 64

TASK II Statements that people use to describe themselves are given below. Please select the response that indicates how you generally feel. There are no right or wrong answers. Do not spend too much time on any one statement. Strongly Disagree Strongly Agree 1. I enjoy trying to determine if what I read or hear is true... 1 2 3 4 5 6 2. I relish learning... 1 2 3 4 5 6 3. The actions people take and the reasons for those actions are fascinating....1 2 3 4 5 6 4. I am interested in what causes people to behave the way that they do... 1 2 3 4 5 6 5. I am confident of my abilities... 1 2 3 4 5 6 6. I often reject statements unless I have proof that they are true... 1 2 3 4 5 6 7. Discovering new information is fun... 1 2 3 4 5 6 8. Other peoples behavior doesn t interest me... 1 2 3 4 5 6 9. I am self-assured... 1 2 3 4 5 6 10. My friends tell me that I usually question things that I see or hear... 1 2 3 4 5 6 11. I like to understand the reason for other peoples behavior... 1 2 3 4 5 6 12. I think that learning is exciting... 1 2 3 4 5 6 13. I take my time when making decisions... 1 2 3 4 5 6 14. I tend to immediately accept what other people tell me... 1 2 3 4 5 6 15. I usually accept things I see, read or hear at face value... 1 2 3 4 5 6 16. I don t feel sure of myself... 1 2 3 4 5 6 17. I usually notice inconsistencies in explanations... 1 2 3 4 5 6 18. I don t like to decide until I ve looked at all of the readily available information... 1 2 3 4 5 6 19. I like searching for knowledge... 1 2 3 4 5 6 20. I frequently question things that I see or hear... 1 2 3 4 5 6 Please turn over the page and continue answering questions. 65

Strongly Strongly Disagree Agree 21. I often accept other peoples explanations without further thought 1 2 3 4 5 6 22. I feel good about myself... 1 2 3 4 5 6 23. I wait to decide on issues until I can get more information... 1 2 3 4 5 6 24. The prospect of learning excites me... 1 2 3 4 5 6 25. It is easy for other people to convince me... 1 2 3 4 5 6 26. I seldom consider why people behave in a certain way... 1 2 3 4 5 6 27. I like to ensure that I ve considered most available information before making a decision... 1 2 3 4 5 6 28. Most often I agree with what the others in my group think... 1 2 3 4 5 6 29. I dislike having to make decisions quickly... 1 2 3 4 5 6 30. I have confidence in myself... 1 2 3 4 5 6 Please note the current time, this time will mark the end of this task. 66

GENERAL QUESTIONS Please answer the following questions. 1. How long have you worked for your current employer? years, months. 2. What is your current position or rank in the firm?. 3. How long have you been at your current position or rank? years, months. 4. How much audit experience do you have? years, months. 5. Approximately how many different audits (audit clients?) have you worked on?. 6. In your firm, at what position or rank does an auditor typically perform preliminary analytical procedures?. 7. Have you ever performed preliminary analytical procedures? Yes No If yes, on approximately how many audits?. 8. Have you ever worked on an audit where management fraud was detected? Yes No. 9. If you answered yes to the above question, was an adverse opinion actually issued on one or more of these engagements? Yes No. 10. What is your estimate of the percentage of all companies/entities that fail? %. 11. What is your estimate of the percentage of all companies/entities that have materially misstated financial statements due to fraud? %. 12. What is your estimate of the percentage of the companies/entities that fail due to fraudulent financial reporting? %. 13. What is your estimate of the percentage of fraud detected by the auditor prior to issuance of the financial statements? %. 14. Please indicate the extent of your education. Bachelors Degree Masters Degree Other 15. Are you a Certified Public Accountant (CPA)? Yes No. 16. Have you been involved in a brainstorming session to discuss the likelihood of fraud on any of your engagements thus far? Yes No, if so how many engagements?. 17. Do you think that brainstorming will help auditors to better assess the risk of fraud, to create unique tests for investigating its possibility and to lead to its ultimate detection better than previous techniques suggested in the standards? Yes No. 18. If you have participated in brainstorming on previous engagements, what is the average amount of time that the engagement team usually spends on brainstorming?. 67

19. If you have participated in brainstorming on previous engagements, what individuals have been involved? staff seniors managers partners. 20. Was the session conducted face to face, group support system over email conference call, other (please describe). 21. Have you worked previously with either of your group members before? Yes No. If so, please name them and describe the nature of your work, number of audit engagements, relationship. 22. In Part I of today s case, and held constant throughout Part II and III, the engagement partner on this audit expressed concern that team members implement brainstorming with which of these in mind. (Please check ONLY the one that applies). That audit team members should be aware of the costs of brainstorming and complete this phase of the audit as efficiently as possible. That audit team members should approach brainstorming with the appropriate level of professional skepticism, and complete this phase of the audit as effectively as possible. 23. In completing the tasks in this case today, what was the significance of the partner preference to you in your decisions? Highly Insignificant Highly Significant [ ] 0 1 2 3 4 5 6 7 8 9 10 24. To what extent do you agree that the engagement partner s concerns about brainstorming are realistic across audit engagements? (circle one number) Strongly Disagree Strongly Agree [ ] 0 1 2 3 4 5 6 7 8 9 10 25. Please provide the name of the best member of your group (i.e., the individual in your group who you believe was the most knowledgeable, provided the most useful responses and was most accurate in his or her assessments), even if this person is you. Name of best member:. 26. Were you able to determine the actual company that the case materials were based on? Yes No. If you think so, please identify the company. 27. Please list any comments that you may have on this study. Thank you again for participating in this study. Your time and effort are greatly appreciated. 68

APPENDIX B FINANCIAL STATEMENT CASE MATERIALS CASE BOOKLET FINANCIAL STATEMENT INFORMATION Please print your name and address of the practice office below. Name (printed): Practice Office Address: 69

Narrative Description of the Company Calico Corporation (Calico or Company) is engaged primarily in developing, manufacturing, and marketing traditional and digital imaging. Calico is a leader in selling equipment and providing solutions including hardware, services and software that enhance productivity and knowledge sharing. References herein to us or our refer to Calico and consolidated subsidiaries unless the context specifically requires otherwise. We distribute our products in the Western Hemisphere through divisions and whollyowned subsidiaries. In Europe, Africa, the Middle East, India and parts of Asia, we distribute Calico Limited and related companies (collectively Calico Limited). Calico had 92,500 employees at year-end 2000. Dao Calico Co., Limited, an unconsolidated entity jointly owned by Calico Limited and Dao Photo Film Company Limited, develops, manufactures and distributes film processing products in Japan and other areas of the Pacific Rim, Australia and New Zealand. Japan represents approximately 80 percent of Dao Calico revenues, and Australia, New Zealand, and Singapore, Malaysia, Korea, Thailand and the Philippines represent 10 percent. The remaining 10 percent of Dao Calico revenues are sales to Calico. Dao Calico conducts business in other Asian Pacific Rim countries through joint ventures and distributors. In December 2000, as part of the asset disposition element of our turnaround plan, we completed the sale of our China operations to Dao Calico for $550 million cash and their assumption of $118 million of debt. The sale included all of our manufacturing, sales and services functions in China and Hong Kong, including ownership of Calico (China) Limited and Calico (Hong Kong) Limited. The sale strengthened our liquidity and produced a $119 million after tax gain. In March 2001, we sold half our ownership interest in Dao Calico to Dao Photo Film for $1,283 million in cash. The company retains significant rights as a minority shareholder. All product and technology agreements between Calico and Dao Calico will continue, ensuring that the two companies retain uninterrupted access to each other s portfolio of patents, technology and products. Nature of the Business Our activities encompass developing, manufacturing, marketing, servicing and financing a complete range of film processing products, solutions and services designed to make organizations around the world more productive. Calico is a major participant in infoimaging, composed of devices (digital cameras and personal data assistants), infrastructure (online networks and delivery systems for images) and services and media (software, film and paper), enabling people to access, analyze and print images. Calico harnesses its technology, market reach and a host of industry partnerships to provide innovative products and services for customers who need the information-rich context that images contain. 70

Calico Corporation Consolidated Balance Sheets (dollars in millions) December 31, 2000 1999 (Unaudited) Assets Cash and cash equivalents 1,741 126 Accounts receivable, net 2,281 2,633 Finance receivable, net 5,097 4,961 Inventories, net 1,932 2,290 Equipment on operating leases, net 724 695 Deferred taxes and other current assets 1,247 1,230 Total Current Assets 13,022 11,935 Finance receivables due after one year, net 7,957 8,058 Land, buildings and equipment, net 2,495 2,456 Investments in affiliates, at equity 1,362 1,615 Intangible and other assets, net 3,061 2,810 Goodwill, net 1,578 1,657 Total Assets 29,475 28,531 Liabilities and Equity Short term debt and current portion of long term debt 2,693 3,957 Accounts Payable 1,033 1,016 Accrued compensation and benefits cost 662 715 Unearned income 250 186 Other current liabilities 1,630 2,176 Total Current Liabilities 6,268 8,050 Long Term Debt 15,404 11,044 Postretirement medical benefits 1,197 1,133 Deferred taxes and other liabilities 1,876 2,521 Deferred ESOP benefits (221) (299) Minorities interests in equity of subsidiaries 141 127 Obligation for equity put options 32 0 Company obligated, mandatorily redeemable preferred securities of subsidiary trust holding solely subordinated debentures of the Company 638 638 Preferred Stock 647 669 Common shareholder s equity 3,493 4,648 Total Liabilities and Equity 29,475 28,531 71

Calico Corporation Consolidated Income Statements (dollars in millions) 2000 (unaudited) December 31, 1999 1998 Revenues Sales 10,059 10,441 10,668 Service, outsourcing and rentals 7,718 8,045 7,783 Finance income 924 1,081 1,142 Total Revenues 18,701 19,567 19,593 Costs and Expenses Cost of Sales 6,197 5,944 5,880 Inventory charges 90 0 113 Cost of service, outsourcing and Rentals 4,813 4,599 4,323 Equipment Financing Interest 605 547 570 Research and Development Expenses 1,044 992 1,035 Selling, administrative and general Expenses 5,649 5,292 5,343 Restructuring charge and asset Impairment 540 0 1,531 Gain on affiliate s sale of stock (21) 0 0 Purchased in-process research and Development 27 0 0 Gain on sale of China operations (200) 0 0 Other, net 341 285 219 Total Costs and Expenses 19,085 17,659 19,014 Income (Loss) from Continuing Operations before Income Taxes (Benefits) Equity Income and Minorities Interest (384) 1,908 579 Income taxes (Benefits) (109) 588 145 Income (Loss) from Continuing Operations after Income Taxes (Benefits) before Equity Income and Minorities Interests (275) 1,320 434 Equity in net income of unconsolidated affiliates 61 68 74 Minorities interests in earnings and subsidiaries 43 49 45 Income (loss) from continuing Operations (257) 1,339 463 Discontinued Operations 0 0 (190) Net Income (Loss) (257) 1,339 273 Basic Earnings (Loss) per share (0.44) 1.96 0.34 Diluted Earnings (Loss) per share (0.44) 1.85 0.34 72

Calico Corporation Common-sized Balance Sheets (dollars in millions) December 31, 2000 1999 (Unaudited) Assets Cash and cash equivalents 5.91% 0.44% Accounts receivable, net 7.74% 9.23% Finance receivable, net 17.29% 17.39% Inventories, net 6.55% 8.03% Equipment on operating leases, net 2.46% 2.44% Deferred taxes and other current assets 4.23% 4.31% Total Current Assets 44.18% 41.83% Finance receivables due after one year, net 27.00% 28.24% Land, buildings and equipment, net 8.46% 8.61% Investments in affiliates, at equity 4.62% 5.66% Intangible and other assets, net 10.39% 9.85% Goodwill, net 5.35% 5.81% Total Assets 100.00% 100.00% Liabilities and Equity Short term debt and current portion of long term debt 9.14% 13.87% Accounts Payable 3.50% 3.56% Accrued compensation and benefits cost 2.25% 2.51% Unearned income 0.85% 0.65% Other current liabilities 5.53% 7.63% Total Current Liabilities 21.27% 28.21% Long Term Debt 52.26% 38.71% Postretirement medical benefits 4.06% 3.97% Deferred taxes and other liabilities 6.36% 8.84% Deferred ESOP benefits (0.75)% (1.05)% Minorities interests in equity of subsidiaries 0.48% 0.45% Obligation for equity put options 0.11% 0.00% Company obligated, mandatorily redeemable preferred securities of subsidiary trust holding solely subordinated debentures of the Company 2.16% 2.24% Preferred Stock 2.20% 2.34% Common shareholder s equity 11.85% 16.29% Total Liabilities and Equity 100.00% 100.00% 73

Calico Corporation Common-sized Income Statements (dollars in millions) 2000 (unaudited) December 31, 1999 1998 Revenues Sales 53.79% 53.36% 54.45% Service, outsourcing and rentals 41.27% 41.12% 39.72% Finance income 4.94% 5.52% 5.83% Total Revenues 100.00% 100.00% 100.00% Costs and Expenses Cost of Sales 33.14% 30.63% 30.01% Inventory charges 0.48% 0.00% 0.58% Cost of service, outsourcing and Rentals 25.74% 23.50% 22.06% Equipment Financing Interest 3.24% 2.80% 2.91% Research and Development Expenses 5.58% 5.07% 5.28% Selling, administrative and general Expenses 30.21% 27.05% 27.27% Restructuring charge and asset Impairment 2.89% 0.00% 7.81% Gain on affiliate s sale of stock (0.11)% 0.00% 0.00% Purchased in-process research and Development 0.14% 0.00% 0.00% Gain on sale of China operations (1.07)% 0.00% 0.00% Other, net 1.82% 1.46% 1.12% Total Costs and Expenses 102.05% 90.25% 97.04% Income (Loss) from Continuing Operations before Income Taxes (Benefits) Equity Income and Minorities Interest (2.05)% 9.75% 2.96% Income taxes (Benefits) (0.58)% 3.01% 0.74% Income (Loss) from Continuing Operations after Income Taxes (Benefits) before Equity Income and Minorities Interests (1.47)% 6.75% 2.22% Equity in net income of unconsolidated affiliates 0.33% 0.35% 0.38% Minorities interests in earnings and subsidiaries 0.23% 0.25% 0.23% Income (loss) from continuing Operations (1.37)% 6.84% 2.36% Discontinued Operations 0.00% 0.00% (0.97)% Net Income (Loss) (1.37)% 6.84% 1.39% 74

Calico Corporation Financial Ratios 2000 1999 Liquidity: Current 2.08 1.48 Quick 1.77 1.20 Solvency: Debt to Assets 0.83 0.79 Times Interest Earned 0.37 4.49 Long-Term Debt to Equity 3.11 1.82 Activity: Inventory Turnover 3.21 2.62 Age of Inventory 112.15 days 137.41 days Accounts Receivable Turnover 4.41 3.97 Age of Accounts Receivable 81.63 days 90.68 days Total Asset Turnover 0.63 0.69 Profitability: Gross Margin 38.39% 42.59% Profit Margin on Sales (1.37)% 6.84% Return on Total Assets 1.18% 6.61% Return on Equity (5.19)% 22.02% Equations: Current ratio: current assets / current liabilities Quick ratio: (current assets-inventory) / current liabilities Debt to assets: total debt / total assets Times interest earned: earnings before interest and taxes / interest charges Long term Debt to Equity: long term debt / stockholder s equity Inventory turnover: cost of sales / inventory Age of inventory: 360 days / inventory turnover A/R turnover: sales / accounts receivable Age of A/R: 360 days / accounts receivable turnover Total asset turnover: total revenues / total assets Gross Margin: total gross margin (sales cost of sales) / sales Profit Margin on Sales: net income/ total revenues Return on Total Assets: (net income + interest expense) / total assets Return on Equity: net income / stockholder s equity 75

Notes to Consolidated Financial Statements (Dollars in millions, except per-share data and unless otherwise indicated) 1. Summary of Significant Accounting Policies Description of Business. Calico Corporation is the leader in helping people take, share, enhance, preserve, print and enjoy images- for memories and for entertainment. Our activities encompass developing, manufacturing, marketing, servicing, and financing a complete range of film processing products and solutions. Basis of Consolidation. The Consolidated Financial Statements include the accounts of Calico Corporation and all majority owned subsidiaries (the Company). All significant intercompany accounts and transactions have been eliminated. References herein to "we" or "our" refer to Calico and consolidated subsidiaries unless the context specifically requires otherwise. Calico Limited, Calico Holding (Nederland) BV, Calico Investments (Bermuda) Limited, Calico Holdings (Bermuda) Limited and their respective subsidiaries are referred to as Calico Limited. Investments in which we have a 20 to 50 percent ownership interest are generally accounted for on the equity method. Upon the sale of stock by a subsidiary, we recognize a gain or loss in our consolidated statement of operations equal to our proportionate share of the increase or decrease in the subsidiary's equity. For acquisitions accounted for by the purchase method, operating results are included in the consolidated statements of operations from the date of acquisition. Earnings per Share. Basic earnings per share are based on net income less preferred stock dividend requirements divided by the average common shares outstanding during the period. Diluted earnings per share assume exercise of in-the-money stock options outstanding and full conversion of convertible debt and convertible preferred stock into common stock at the later of the beginning of the year or date of issuance, unless they are antidilutive. Income (loss) from Continuing Operations before Income Taxes (Benefits), Equity Income and Minorities' Interests. Throughout these notes to Consolidated Financial Statements, we refer to the effects of certain changes in estimates and other adjustments on Income (loss) from Continuing Operations before Income Taxes (Benefits), Equity Income and Minorities' Interests. For convenience and ease of reference, that financial statement caption is hereafter referred to as "pre-tax income (loss)." Use of Estimates. The preparation of financial statements in accordance with generally accepted accounting principles requires management to make estimates and assumptions that affect the reported amounts of assets and liabilities and the disclosure of contingent assets and liabilities at the date of the financial statements, and the reported amounts of revenues and expenses during the reporting period. Estimates are used for, but not limited to: accounting for residual values; allocation of revenues and fair values in multiple element arrangements; allowance for doubtful accounts; inventory valuation; merger, restructuring and other related charges; asset impairments; depreciable lives of assets; 76

useful lives of intangible assets and goodwill; pension assumptions; and tax valuation allowances. Future events and their effects can not be perceived with certainty. Accordingly our accounting estimates require the exercise of judgment. The accounting estimates used in the preparation of our Consolidated Financial Statements will change as new events occur, as more experience is acquired, as additional information is obtained and as the Company's operating environment changes. Actual results could differ from those estimates. Changes in Estimates. In the ordinary course of accounting for items such as revenue allocations and related estimated fair values in multiple element arrangements, allowances for doubtful accounts, inventory valuation, and residual values, among others, we make changes in estimates as appropriate in the circumstances. Such changes and refinements in estimation methodologies are reflected in reported results of operations and, if material, the approximate effects of changes in estimates are disclosed in the Notes to our Consolidated Financial Statements. Accounting Changes-Accounting for Derivative Instruments. In 1998, the Financial Accounting Standards Board (FASB) issued Statement of Financial Accounting Standards (SFAS) No. 133, "Accounting for Derivative Instruments and Hedging Activities." SFAS No. 133 requires companies to recognize all derivatives as assets or liabilities measured at their fair value. Gains or losses resulting from changes in the fair value of derivatives would be recorded each period in current earnings or other comprehensive income, depending on whether a derivative is designated as part of a hedge transaction and, if it is, depending on the type of hedge transaction. SFAS No. 133, as amended, is effective for the Company as of January 1, 2001. With the adoption of SFAS No. 133, we will record a net cumulative after-tax charge of $2 in our statements of operations and a net cumulative after-tax loss of $19 in Accumulated Other Comprehensive Income. Further, as a result of recognizing all derivatives at fair value, including the differences between the carrying values and fair values of related hedged assets, liabilities and firm commitments, we will recognize a $403 increase in Total Assets and a $424 increase in Total Liabilities. We expect that the adoption of SFAS 133 will increase the future volatility of reported earnings and other comprehensive income. In general, the amount of volatility will vary with the level of derivative and hedging activities and the market volatility during any period. Reclassifications. The FASB Emerging Issues Task Force (EITF) issued a pronouncement that requires a change in the way we classify shipping and handling costs billed to customers. Commencing in the fourth quarter of 2000, the EITF required that all amounts billed to a third party customer related to product shipping and handling must be classified as revenue, and costs incurred must be classified as an expense. Shipping and handling amounts billed to customers have historically been recorded as a reduction of cost of goods sold. Prior period financial statements have been reclassified to conform with the 2000 presentation. Revenue Recognition. In the normal course of business the Company generates revenue through the sale of equipment, services, and supplies and income associated with the financing of its equipment sales. Revenue is recognized when earned. More specifically revenue related to the Company's sales of its products and services are as follows: 77

Equipment: Revenues from the sale of equipment under installment arrangements, from sales-type leases or on credit are recognized at the time of sale or at the inception of the lease, respectively. For equipment sales which require the Company to install the product at the customer location, revenue is recognized when the equipment has been delivered to and installed at the customer location. Sales of customer installable and retail channels type products are recognized upon shipment. Revenues from equipment under other leases and similar arrangements are accounted for by the operating lease method and are recognized over the lease term. Sales of equipment subject to the Company's operating leases to third party finance companies (the counter-party) or through structured financings with third parties are recorded as sales at the time the equipment is accepted by the counter-party. The counter-party accepts the risks of ownership of the equipment. Remanufacturing and remarketing of off-lease equipment belonging to the counter-party is performed by the Company for a fee on a non-discriminatory basis. In North America these transactions are structured to provide cash proceeds up front from the counter-party versus collection over time. In Latin America the counter-party pays the Company a fixed amount each month, mitigating risk and variability from the cash flow stream. Services: Service revenues are derived primarily from maintenance contracts on our equipment sold to customers and are recognized over the term of the contracts. Supplies: Supplies revenue generally is recognized upon shipment. Financing: Finance income is earned on an accrual basis under an effective annual yield method. The Company sells its equipment and services on a stand-alone basis and also enters into bundled arrangements which contain multiple deliverable elements. These multiple element arrangements typically include equipment, services, supplies and financing components for which the customer pays a single defined price for all elements. When separate prices are listed in these multiple element arrangements with our customers they may not be representative of the fair values of those elements because the prices of the different components of the arrangement may be altered in customer negotiations, although the aggregate consideration may remain the same. Therefore, revenues under these arrangements are allocated based upon estimated fair values of each element, in accordance with Generally Accepted Accounting Principles (GAAP). The fair value of each element is estimated based on a review of a number of factors including average selling prices for the elements when they are sold on a stand-alone basis. The average selling prices are based on management's best estimates of market conditions and competitive pricing considerations. The principal change in estimate relating to such revenue allocations among multiple elements is made with respect to the estimated fair value of those elements and their related margins. This is a significant factor considered in our revenue allocation process along with other factors, such as pricing changes and customer discounts, which also affect the overall allocation process. The effect of such changes in estimates of fair values and related margins in the years 2000, 1999 and 1998 was $193, $202, and $141, respectively, which generally resulted in increases of sales revenues and decreases to deferred elements of those arrangements. The net effects of such allocations when offset by corresponding decreases in the amortization of deferred revenues was to increase pre-tax income in 2000, 1999 and 1998 by $44, $102, and $101, respectively. 78

In December 1999, the Securities and Exchange Commission issued Staff Accounting Bulletin No. 101 "Revenue Recognition in Financial Statements" (SAB 101). SAB 101 summarizes certain of the staff's views in applying generally accepted accounting principles to revenue recognition in financial statements. We conducted a review of our revenue recognition policies during the fourth quarter of our fiscal year ended December 31, 2000. We have determined that our policies are in conformance with SAB 101 in all material respects. Cash and Cash Equivalents. Cash and cash equivalents consist of cash on hand and investments with original maturities of three months or less. Provisions for Losses on Uncollectible Receivables. The provisions for losses on uncollectible trade and finance receivables are determined principally on the basis of past collection experience. Inventories. Inventories are carried at the lower of average cost or realizable values. Buildings and Equipment and Equipment on Operating Leases. Our fixed assets are depreciated over their estimated useful lives. Equipment on operating leases is depreciated to its estimated residual value. Depreciation is computed using principally the straight-line method. Significant improvements are capitalized; maintenance and repairs are expensed. Goodwill. Goodwill represents the cost of acquired businesses in excess of the fair value of identifiable net assets purchased, and is amortized on a straight-line basis over periods ranging from 15 to 40 years. Goodwill is reported net of accumulated amortization, and the recoverability of the carrying value is evaluated on a periodic basis by assessing current and future levels of income and cash flows as well as other factors. Accumulated amortization at December 31, 2000 and 1999 was $220 and $176, respectively. Classification of Commercial Paper and Bank Notes Payable. As of December 31, 2000, all indebtedness is classified as a short-term or long-term liability based upon its contractual maturity date. In prior years, it was our policy to classify as long-term debt that portion of commercial paper and notes payable that was intended to match fund finance receivables due after one year to the extent that we had the ability under our revolving credit agreement to refinance such commercial paper and notes payable on a longterm basis. Foreign Currency Translation. The functional currency for most foreign operations is the local currency. Net assets are translated at current rates of exchange, and income and expense items are translated at the average exchange rate for the year. The resulting translation adjustments are recorded in Accumulated Other Comprehensive Income. The U.S. dollar is used as the functional currency for certain subsidiaries that conduct their business in U.S. dollars or operate in hyperinflationary economies. A combination of current and historical exchange rates is used in remeasuring the local currency transactions of these subsidiaries, and the resulting exchange adjustments are included in income. Aggregate foreign currency gains/(losses) were $99, $(1) and $(29) in 2000, 1999 and 1998, respectively, and are included in Other, net in the consolidated statements of operations. Stock-Based Compensation. The Company follows the intrinsic value-based method of accounting for its stock-based compensation. 79

2. Restructuring March 2000 Restructuring. In March 2000, we announced details of a worldwide restructuring program. In connection with this program, we initially recorded a pre-tax provision of $596 ($423 after taxes, including our $18 share of a restructuring provision recorded by Dao Calico, an unconsolidated affiliate). The $596 pre-tax charge included severance costs related to the elimination of 5,200 positions worldwide. Approximately 65 percent of the positions to be eliminated are in the U.S., 20 percent are in Europe, and the remainder is predominantly in Latin America. The employment reductions primarily affected employees in manufacturing, logistics, customer service and back office support functions. For facility fixed assets to be disposed of, the impairment loss recognized is based on the fair value less cost to sell, with fair value based on estimates of existing market prices for similar assets. The inventory charges relate primarily to the consolidation of distribution centers and warehouses and the exit from certain product lines. Included in the original provision were reserves related to the incurrence of liabilities due to various third parties and several asset impairment charges. Liabilities recorded for lease cancellation and other costs originally aggregated $51 and included $32 for various contractual commitments, other than facility occupancy leases, that will be terminated early as a result of the restructuring. The commitments include cancellation of supply contracts and outsourced vendor contracts. Included in the asset impairment charge of $71 was: $44 for machinery and tooling for products that were discontinued or will be alternatively sourced; $7 for leasehold improvements at facilities that will be closed; and $20 of sundry surplus assets, individually insignificant, from various parts of our business. These impaired assets were primarily located in the U.S. and the related product lines generated an immaterial amount of revenue. Approximately $71 of the $90 of inventory charges related to excess inventory in many product lines created by the consolidation of distribution centers and warehouses. The remainder was primarily related to the transition to inkjet technology in our wide format printing business. Weakening business conditions and operating results during 2000 required a re-evaluation of the initiatives announced in March 2000. As a result, we were unable to, and do not expect to, complete certain actions originally contemplated at the time that the March 2000 restructuring provision was recorded. Accordingly, during the fourth quarter of 2000, and in connection with the turnaround program discussed below, $71 ($47 after taxes), $59 related to severance costs for 1,000 positions and $12 related to lease cancellation and other costs, of the original $596 provision, was reversed. The reversals primarily relate to delays in the consolidation and outsourcing of certain of our warehousing and logistics operations and the cancellation of certain European initiatives no longer necessary as a result of higher than expected attrition. As of December 31, 2000, approximately 2,400 employees have left the Company under the March 2000 restructuring program. Turnaround Program. During 2000, the significant business challenges that we began to experience in the second half of 1999 continued to adversely affect our financial performance. These challenges include: the ineffective execution of a major sales force realignment, the ineffective consolidation of our U.S. customer administrative centers, increased competition and adverse economic conditions. These operational challenges, exacerbated by significant technology and acquisition investments, have led to a net loss in 2000, credit rating agency downgrades, limited access to capital markets and market-place 80

concerns regarding our liquidity. In response to these challenges, in October 2000, we announced a turnaround program which includes a wide-ranging plan to sell assets, cut costs and strengthen core operations. Additionally, we are exploring alternatives to provide financing for customers in a manner that does not involve the Calico balance sheet, and over time will provide financing for customers using third parties. In December 2000, we sold our operations in China to Dao Calico for $550. We are engaged in other activities which will enhance our liquidity. These activities include asset sales, strategic alliances, and the sale or outsourcing of certain manufacturing operations. It is expected that in most cases asset sales will result in a gain. Regarding the cost reductions, we are in the process of finalizing plans designed to reduce costs by at least $1.0 billion annually. In connection therewith, during the fourth quarter of 2000, we recorded an additional pre-tax restructuring provision totaling $105 ($87 after taxes, including our $19 share of an additional provision recorded by Dao Calico) in connection with finalized initiatives under the turnaround program. This charge included estimated costs of $71 for severance costs associated with work force reductions related to the elimination of 2,300 positions worldwide and $34 of asset impairments associated with the disposition of a non-core business. The severance costs relate to further streamlining of existing work processes, elimination of redundant resources and the consolidation of existing activities into other existing operations. The following table summarizes the status of the March 2000 restructuring reserve and the turnaround program: Original Reserve Reversals Charges Against Reserve 12/31/00 Balance March 2000 Restructuring: Cash charges Severance and related costs 384 (59) (130) 195 Lease Cancellation and other costs 51 (12) (19) 20 Subtotal 435 (71) (149) 215 Non-cash Charges Asset Impairment 71 0 (71) 0 Inventory Charges 90 0 (90) 0 Subtotal 161 0 (161) 0 Currency changes 0 0 (6) (6) Subtotal 596 (71) (316) 209 Turnaround Program: Severance and related costs 71 0 0 71 Asset Impairment 34 0 (34) 0 Subtotal 105 0 (350) 71 Total $701 $ (71) $ (34) $280 81

With respect to the March 2000 restructuring program as of March 31, 2001, the remaining liability is $131. All remaining liabilities represent committed obligations of the Company to be paid primarily during 2001 and are included in the caption Other current liabilities in the consolidated balance sheet. 1998 Restructuring. In 1998, we announced a worldwide restructuring program. In connection with this program, we recorded a pre-tax provision of $1,644. As of December 31, 2000, this program has been substantially completed and the remaining liability balance is $107 after fourth quarter reversals of $11. The remaining liability is for salary continuance payments and the runoff of lease cancellation payments. There were no material changes to the program since its announcement in April 1998. The remaining liability is fully committed and the majority will be utilized throughout 2001. 3. Acquisitions In January 2000, we and Dao Calico completed the acquisition of the Color Printing and Imaging Division of Tektronix, Inc. (CPID). The aggregate consideration paid of $925 in cash, which includes $73 paid directly by Dao Calico, is subject to certain post-closing adjustments. CPID manufactures and sells color printers, ink and related products, and supplies. The acquisition was accounted for in accordance with the purchase method of accounting. The excess of cash paid over the fair value of net assets acquired has been allocated to identifiable intangible assets and goodwill using a discounted cash flow approach by an independent appraiser. The value of the identifiable intangible assets includes $27 for purchased in-process research and development which was written off in 2000. This charge represents the fair value of certain acquired research and development projects that were determined not to have reached technological feasibility as of the date of the acquisition and was determined based on a methodology that focused on the after-tax cash flows of the in-process products and the stage of completion of the individual research and development projects. Other identifiable intangible assets are exclusive of intangible assets acquired by Dao Calico, and include the installed customer base ($209), the distribution network ($123), the existing technology ($103), the workforce ($71), and trademarks ($23). These identifiable assets are included in Intangibles and other assets in the Consolidated Balance Sheets. The remaining excess has been assigned to Goodwill, however such amount may be affected by any post-closing adjustments which could potentially reduce the purchase price. Other identifiable intangible assets and Goodwill are being amortized on a straight-line basis over their estimated useful lives which range from 7 to 25 years. In connection with the CPID acquisition we recorded approximately $45 for anticipated costs associated with exiting certain activities of the acquired operations. These activities include: the consolidation of duplicate distribution facilities; the rationalization of the service organization; and the exiting of certain lines of the CPID business. The costs associated with these activities include inventory write-offs, severance charges, contract cancellation costs and fixed asset impairment charges. We expect these actions to be completed in 2001. In August 1999, we purchased the OmniFax division from Kilkenny Business Systems for $45 in cash. OmniFax is a supplier of business laser multifunction fax systems. The acquisition resulted in goodwill of approximately $22 (including transaction costs), which is being amortized over 15 years. 82

Also during 1999, we paid $62 to increase our ownership in our India operations from approximately 40 percent to 68 percent. This transaction resulted in additional goodwill of $48, which is being amortized over 40 years. In May 1998, we acquired Systems Solutions, Inc., an information technology services company, and its parent company, Intelligent Electronics, Inc., for $413 in cash. The transaction resulted in goodwill of $395, which is being amortized over 25 years. The Company is continuing to integrate Systems Solutions Inc. with its Industry Solutions business segment. This integration is designed to increase the revenue of our industry solutions operations, and to achieve cost savings and synergies. While this integration is taking longer than originally anticipated, the Company believes that events and changes in circumstances since the acquisition do not presently indicate an impairment of goodwill. However, the Company intends to continue the integration efforts and will perform an assessment of the recoverability of goodwill should circumstances change. 4. Divestitures In December 2000 we sold our China operations to Dao Calico for $550. In connection with the sale, Dao Calico assumed $118 of indebtedness. The pre-tax gain recorded in the fourth quarter of 2000, was $200. In June 2000, we sold the U.S. and Canadian commodity paper business, including an exclusive license for the Calico brand, to Papel Inc. and recorded a pre-tax gain of approximately $40 which is included in Other, net. In addition to the proceeds from the sale of the business, the Company will receive royalty payments on future sales of Calico branded commodity paper by Papel Inc. and will earn commissions on Calico originated sales of commodity paper as an agent for Papel Inc. In April 2000, we sold a 25 percent ownership interest in our wholly owned subsidiary, Security Inc., to Software Unlimited, Inc. for $50 and recognized a pre-tax gain of $23, which is included in Other, net. An additional pre-tax gain of $27 was deferred, pending the resolution of certain performance criteria, and is included in Unearned income in the Consolidated Balance Sheets. In connection with the sale, Security Inc. also received $40 from Software Unlimited for a non-exclusive license of its patents and other intellectual property and a $25 advance against future royalty income from Software Unlimited on sales of products incorporating Security Inc.'s technology. The license payment is being amortized over the life of the license agreement of 10 years and the royalty advance will be recognized in income as earned. 83

5. Receivables, Net Finance Receivables. Finance receivables result from installment arrangements and sales-type leases arising from the marketing of our business equipment products. These receivables generally mature over two to five years and are typically collateralized by a security interest in the underlying assets. The components of Finance receivables, net at December 31, 2000, 1999 and 1998 follow: 2000 1999 1998 Gross receivables $14,556 $14,478 $15,957 Unearned income (1,733) (1,733) (2,185) Unguaranteed residual values 681 697 617 Allowance for doubtful accounts (450) (423) (441) Finance receivables, net 13,054 13,019 13,948 Less current portion 5,097 4,961 5,055 Amounts due after one year, net $7,957 $ 8,058 $ 8,893 Contractual maturities of our gross finance receivables subsequent to December 31, 2000 follow: 2001 2002 2003 2004 2005 Thereafter $5,654 $3,980 $2,706 $1,580 $540 $96 Experience has shown that a portion of these finance receivables will be prepaid prior to maturity. Accordingly, the preceding schedule of contractual maturities should not be considered a forecast of future cash collections. Unguaranteed residual values are assigned primarily to our high volume copying, printing and production publishing products. The assigned values are generally established in order to result in a normal profit margin in the subsequent transaction. In September 2000, we transferred $457 of finance receivables to a special purpose entity for cash proceeds of $411 and a retained interest of $46. The transfer agreement includes a repurchase option; accordingly the proceeds were accounted for as a secured borrowing. At December 31, 2000 the balance of receivables transferred was $411 and is included in Finance receivables, net in the Consolidated Balance Sheets. The remaining secured borrowing balance of $325 is included in Debt. In 1999, we sold $1,495 of finance receivables and recorded a net increase in finance income of approximately $17 which includes the unfavorable flow-through impacts. The retained interests remaining from these sales were not material at December 31, 2000. Beginning in 1999 several Latin American affiliates entered into certain structured transactions involving contractual arrangements which transferred the risks of ownership of equipment subject to operating leases to third party finance companies, who are obligated to pay the Company a fixed amount each month. The Company accounts for these transactions similar to its sales-type leases. These transactions resulted in sales of $126 and $280 in 2000 and 1999, respectively. The contribution to Pre-tax income resulting from these transactions was $92 and $155 in 2000 and 1999, respectively. 84

Finance Interest Rates. Financing income is determined by the discount rate applied to minimum contract payments, excluding service and supplies, used in the estimation of the fair value of the equipment. Finance interest rates include the aforementioned discount rates in customer arrangements, as well as related sources of income. Over the years, the Company's finance interest rates have changed as a result of a number of factors including money market conditions; the economic environment; debt coverage; return on equity; debt to equity ratios and other external factors which are particularly relevant to our financing business. During the period of 1998 to 2000 such finance interest rates as a percentage of the average finance receivables portfolio and the Company's average cost of funds used in our customer financing activities were: 2000 1999 1998 Average finance interest rates 8.3% 9.2% 9.3% Average cost of funds 5.4% 4.7% 5.1% In line with market comparables, the Company's financing operations are targeted to achieve a 15 percent return on equity. The Company periodically reviews, and may change, the discount rates in order to be consistent with this objective and to reflect the estimated fair value of the financing component in its lease arrangements. Changes in the rate applied to a bundled arrangement may affect one or more elements of the arrangement. In general, the following changes in discount rates are reflected as reciprocal changes in equipment revenues, partially offset by the resulting change in customer finance income. Such changes in accounting estimate had the following approximate effects on pre-tax income (loss): Increase/decrease 2000 1999 1998 Effect of changes in discount rates /1/ $24 $101 $128 /1/ Represents the impact of changes in customer finance rate estimates net of amortization of the related cumulative unearned income effects. Accounts Receivable. In 2000, we entered into agreements to sell, on an ongoing basis, a defined pool of accounts receivable to special purpose entities. At December 31, 2000, the total pool of accounts receivable transferred was approximately $900. The special purpose entities, in turn, sell participating interests in such accounts receivable to investors up to a maximum amount of $330. Under the terms of the agreement, new receivables are added to the pool as collections reduce previously sold accounts receivable. Investors have a priority collection interest in the entire pool of receivables, and as a result, we have retained credit risk to the extent the pool exceeds the amount sold to investors. We continue to service the receivables on behalf of the special purpose entities and receive a servicing fee adequate to compensate for our responsibilities. At December 31, 2000, $328 in net proceeds were received from sales of participating interests to investors and were recorded as a reduction in Accounts receivable, net in the Consolidated Balance Sheets. The earnings impact related to the receivables sold under these agreements was not material. Our retained interests, which are included in Accounts receivable, net, are recorded at fair value using estimates of dilution based on historical experience. These estimates are adjusted regularly based on actual experience with the pool, including defaults and credit deterioration. 85

If historical dilution percentages were to increase one percentage point, the value of the Company's retained interest would be reduced by approximately $9. Allowances for doubtful accounts on our accounts receivable balances at December 31, 2000, 1999 and 1998 amounted to $282, $137 and $102, respectively. 6. Inventories and Equipment on Operating Leases, Net The components of inventories at December 31, 2000, 1999 and 1998 follow: 2000 1999 1998 Finished goods $1,439 $1,805 $1,929 Work in process 147 122 111 Raw materials 346 363 464 Inventories $1,932 $2,290 $2,504 Equipment on operating leases and similar arrangements consists of our business equipment products that are rented to customers and are depreciated to estimated residual value. Equipment on operating leases and the related accumulated depreciation at December 31, 2000, 1999 and 1998 follow: 2000 1999 1998 Equipment on operating leases $2,124 $1,777 $2,057 Less: Accumulated depreciation 1,400 1,082 1,260 Equipment on operating leases, net $724 $695 $797 We sold equipment subject to operating leases and similar arrangements to third party finance companies for cash in the amounts of $22, $120 and $74 in 2000, 1999 and 1998, respectively. The contribution to Pre-tax income resulting from these transactions was $9, $65 and $24 in 2000, 1999 and 1998, respectively. Depreciable lives vary from two to four years. Our business equipment operating lease terms vary, generally from 12 to 36 months. Scheduled minimum future rental revenues on operating leases with original terms of one year or longer are: 2001 2002 2003 Thereafter Scheduled future rental revenues $320 $151 $71 $43 Total contingent rentals, principally usage charges in excess of minimum allowances relating to operating leases, for the years ended December 31, 2000, 1999 and 1998 amounted to $120, $163 and $161, respectively. 86

7. Land, Buildings and Equipment, Net The components of land, buildings and equipment, net at December 31, 2000, 1999 and 1998 follow: Estimated Useful Lives (Years) 2000 1999 1998 Land $70 $66 $80 Buildings and building equipment 25 to 50 1,064 1,087 973 Leasehold improvements Lease term 426 434 425 Plant machinery 4 to 12 1,981 1,897 1,926 Office furniture and equipment 3 to 15 1,304 1,339 1,299 Other 3 to 20 199 235 260 Construction in progress 295 328 283 Subtotal 5,339 5,386 5,246 Less accumulated depreciation 2,844 2,930 2,880 Land, buildings, and equipment, net $2,495 $2,456 $2,366 We lease certain land, buildings and equipment, substantially all of which are accounted for as operating leases. Total rent expense under operating leases for the years ended December 31, 2000, 1999 and 1998 amounted to $344, $397 and $436, respectively. Future minimum operating lease commitments that have remaining non-cancelable lease terms in excess of one year at December 31, 2000 follow: 2001 2002 2003 2004 2005 Thereafter Future minimum operating lease commitments $290 $238 $193 $155 $132 $426 In certain circumstances, we sublease space not currently required in operations. Future minimum sublease income under leases with non-cancelable terms in excess of one year amounted to $50 at December 31, 2000. In 1994, we awarded a contract to Electronic Data Systems Corp. (EDS) to operate our worldwide data processing and telecommunications network through the year 2004. Calico has the right to terminate this agreement with six months' notice to EDS. Minimum payments due EDS under the contract follow: 2001 2002 2003 2004 Minimum payments $217 $198 $183 $95 8. Investments in Affiliates, at Equity Investments in corporate joint ventures and other companies in which we generally have a 20 to 50 percent ownership interest at December 31, 2000, 1999 and 1998 follow: 2000 1999 1998 Dao Calico $1,259 $1,513 $1,354 Other investments 103 102 102 Investments in affiliates, at equity $1,362 $1,615 $1,456 87

Calico Limited owned 50 percent of the outstanding stock of Dao Calico, a corporate joint venture with Dao Photo Film Co., Ltd. (Daofilm) at December 31, 2000. Dao Calico is headquartered in Tokyo and operates in Japan and other areas of the Pacific Rim, Australia and New Zealand. Condensed financial data of Dao Calico for its last three fiscal years follow: 2000 1999 1998 Summary of operations Revenues $8,401 $7,751 $6,809 Costs and expenses 8,115 7,440 6,506 Income before income taxes 286 311 303 Income taxes 146 201 195 Net income $140 $110 $108 Balance Sheet Data Assets Current assets $3,162 $3,521 $2,760 Non-current assets 3,851 3,521 3,519 Total assets $7,013 $7,042 $6,279 Liabilities and Shareholder s Equity Current liabilities $3,150 $2,951 $2,628 Long-term debt 445 297 234 Other non-current liabilities 852 951 895 Shareholders equity 2,566 2,843 2,522 Total liabilities and shareholders equity $7,013 $7,042 $6,279 9. Segment Reporting In the first quarter of 2000, we completed the realignment of our operations to better align the Company to serve its diverse customers/distribution channels. As a result of this realignment, our reportable segments have been revised accordingly and are as follows: Industry Solutions, General Markets, Developing Markets and Other businesses. The Industry Solutions operating segment (ISO) covers the direct sales and service organizations in North America and Europe. It is organized around key industries and focused on providing our largest customers with printing solutions consisting of hardware, software and services, including outsourcing, systems integration and consulting. The General Markets operating segment (GMO) includes sales agents in North America, concessionaires in Europe and our Channels Group which includes retailers and resellers. It is responsible for increasing penetration of the general market space, including small office solutions, products for networked work group environments and personal/ home office products. In addition, it has responsibility for product development and acquisition for its markets, providing customer and channel-ready products and solutions. The Developing Markets operating segment (DMO) includes operations in Latin America, Russia, India, the Middle East and Africa. It takes advantage of growth opportunities in emerging markets/countries around the world, building on the leadership Calico has already established in a number of those markets. 88

Other businesses include several units, none of which met the thresholds for separate segment reporting. The revenues included in this group are primarily from Calico Supplies Group (CSG) and Calico Engineering Systems (CES) and corporate inter-segment eliminations. All corporate and shared service unit expenses, including interest and depreciation, have been allocated to the operating segments. Other businesses' total assets include CES, CSG, deferred tax assets, which have not been allocated, the investment in Dao Calico and the remaining investments in discontinued operations. The accounting policies of the operating segments are the same as those described in the summary of significant accounting policies. It is not practicable to discern the segment information for 1998 for the above segments due to internal reorganizations. Accordingly, 1998 realigned segment amounts have not been presented. Operating segment profit or loss information for the years ended December 31, 2000 and 1999 for our realigned segments is as follows: Industry Solutions General Markets Developing Markets Other Businesses 2000 Information about profit or loss Revenues from external customers $ 8,619 $4,827 $2,533 $1,798 $17,777 Finance income 529 238 157 0 924 Intercompany revenues 38 215 0 (253) 0 Total segment revenues 9,186 5,280 2,690 1,545 18,701 Depreciation and amortization 584 227 132 5 948 Interest expense 584 272 165 10 1,031 Segment profit (loss) /1/ 110 (126) (116) 194 62 Earnings (loss) of non-consolidated affiliates /2/ (1) 0 0 99 98 Information about assets Investments in nonconsolidated Affiliates 16 0 7 1,339 1,362 Total assets 18,662 3,129 4,470 3,214 29,475 Capital expenditures 184 137 79 52 452 1999 Information about profit or loss Revenues from external customers $ 9,463 $4,489 $2,542 $1,992 $18,486 Finance income 622 249 210 0 1,081 Intercompany revenues 60 132 0 (192) 0 Total segment revenues 10,145 4,870 2,752 1,800 19,567 Depreciation and amortization 486 182 106 5 779 Interest expense 488 189 120 6 803 Segment profit 1,328 162 214 204 1,908 Earnings (loss) of non-consolidated Affiliates 4 0 (5) 69 68 Information about assets Investments in nonconsolidated Affiliates 25 0 8 1,582 1,615 Total assets 18,487 1,703 4,636 3,705 28,531 Capital expenditures 300 153 94 47 594 89 Total

/1/ Segment profit (loss) excludes the impact of the 2000 restructuring charge $(619), the purchased inprocess research and development $(27), and the gain on sale of our China operations $200. /2/ Excludes our $37 share of a restructuring charge recorded by Dao Calico. Products and services and geographic area data follow: Information about products and services Black and white office and small office/home office (SOHO) Revenues 2000 1999 1998 $ 7,410 $8,150 $8,384 Black and white production 4,940 5,904 5,954 Color copying and printing 2,897 1,851 1,726 Other products and services 3,454 3,662 3,529 Total $18,701 $19,567 $19,503 Revenues Long-Lived Assets 2000 1999 1998 2000 1999 1998 Information about Geographic areas United States $10,397 $10,585 $10,211 $2,282 $2,228 $2,085 Europe 4,870 5,414 5,237 968 616 503 Other Areas 3,434 3,568 4,145 600 751 804 Total $18,701 $19,567 $19,593 $3,850 $3,595 $3,392 Operating segment profit or loss information, using the prior years' basis of presentation, for the years ended December 31, 2000, 1999, and 1998 is as follows: Core Business Dao Calico Paper and Media 2000 Information about profit or loss Revenues from external customers $ 14,493 0 $1,156 $2,128 $17,777 Finance income 918 0 0 6 924 Intercompany revenues (294) 0 0 294 0 Total segment revenues 15,117 0 1,156 2,428 18,701 Depreciation and amortization 943 0 0 5 948 Interest expense 1,031 0 0 0 1,031 Segment profit (loss) /1/ 352 0 85 (375) 62 Earnings (loss) of non-consolidated affiliates /2/ 4 $107 0 (13) 98 Information about assets Investments in nonconsolidated affiliates 71 1,259 0 32 1,362 Total assets 26,224 1,259 69 1,923 29,475 Capital expenditures 430 0 0 22 452 Other Total 90

1999 Information about profit or loss Revenues from external customers $ 15,501 0 $1,148 $1,837 $18,486 Finance income 1,071 0 0 10 1,081 Intercompany revenues (206) 0 0 206 0 Total segment revenues 16,366 0 1,148 2,053 19,567 Depreciation and amortization 774 0 0 5 779 Interest expense 803 0 0 0 803 Segment profit 1,886 0 62 (40) 1,908 Earnings (loss) of non-consolidated affiliates 13 55 0 0 68 Information about assets Investments in nonconsolidated affiliates 102 1513 0 0 1,615 Total assets 25,036 1,513 86 1,896 28,531 Capital expenditures 580 0 0 14 594 1998 Information about profit or loss Revenues from external customers $ 15,623 0 $1,162 $1,666 $18,451 Finance income 1,133 0 0 9 1,142 Intercompany revenues (326) 0 0 326 0 Total segment revenues 16,430 0 1,162 2,001 19,593 Depreciation and amortization 709 0 0 18 727 Interest expense 749 0 0 0 749 Segment profit (loss) /3/ 2,240 0 58 (75) 2,223 Earnings (loss) of non-consolidated affiliates 19 72 0 1 92 Information about assets Investments in nonconsolidated affiliates 81 1,354 0 21 1,456 Total assets 25,842 1,354 84 2,348 29,628 Capital expenditures 539 0 0 27 566 /1/ Segment profit (loss) excludes the impact of the 2000 restructuring charge $(619), the purchased inprocess research and development $(27), and the gain on sale of our China operations $200. /2/ Excludes our $37 and $18 share of a restructuring charge recorded by Dao Calico in 2000 and 1998, respectively. /3/ Segment profit (loss) excludes the impact of the 1998 restructuring charge of $1,644. 91

10. Discontinued Operations Our remaining investment in our Insurance and Other Financial Services (IOFS) and Third-Party Financing and Real Estate discontinued businesses is included in the Consolidated Balance Sheets at December 31, 2000 and 1999 as follows: Balance Sheet Caption 2000 1999 Intangible and other assets $534 $1,130 Long-term debt 0 50 Deferred to CES and other liabilities 0 378 Net Investment $534 $702 The majority of the remaining investment relates to a $462 performance-based instrument received from the sale of one of the Inspirion Holdings, Inc. (Inspirion) insurance companies, The Daly Group, Inc. (TDG). The instrument is Class 2 preferred stock of TDG. TDG has two classes of stock outstanding. The Class 1 shares are 100 percent owned by Fairfax Financial Holdings Limited, one of the largest insurers in North America. We own substantially all of the Class 2 shares. The terms of the performance criteria relate to TDG's available cash flow as defined. Commencing in January 2001, the Class 2 shareholders are entitled to receive 72.5 percent of the available cash and the Class 1 holder receives the remaining 27.5 percent. An initial distribution of $4 was received by us in January 2001. Current projections indicate that we expect to fully recover the remaining $458 by 2018. Calico Financial Services, Inc. (CFSI), a wholly owned subsidiary, continues to provide aggregate excess of loss reinsurance coverage (the Reinsurance Agreements) to one of the former Inspirion units and TDG through Valley Reinsurance Limited (Valley Re), a wholly owned subsidiary of CFSI. The coverage limits for these two remaining Reinsurance Agreements total $578. Both the Company and CFSI have guaranteed that Valley Re will meet all of its financial obligations under the two remaining Reinsurance Agreements. In April 2001 we replaced $660 of letters of credit, which supported Valley Re ceded reinsurance obligations, with trusts which include the Valley Re investment portfolio of $405 plus approximately $255 in cash. These trusts are required to provide security with respect to aggregate excess of loss reinsurance obligations under the two remaining Reinsurance Agreements. 11. Debt Short-Term Debt. Short-term borrowings data at December 31, 2000 and 1999 follow: Weighted Average Interest Rates at 12/31/00 2000 1999 Notes payable 10.20 % $169 $0 Commercial paper 7.01 141 0 Total short-term debt 310 0 Current maturities of long-term debt 2,383 3,957 Total $2,693 $3,957 92

Debt classification. Prior to the year 2000 we had employed a match funding policy for customer financing assets and related liabilities. Under this policy, the interest and currency characteristics of the indebtedness were, in most cases, matched to the interest and currency characteristics of the finance receivables. At December 31, 1999, our debt was classified based on the expected date of repayment of such indebtedness in accordance with our match funding policy. Further, at December 31, 1999, certain other short-term obligations were classified as long-term based on management's intent to refinance certain of these obligations on a long term basis and the ability to do so with credit available under the Revolving Credit Agreement (Revolver). The full utilization of our Revolver and our recent credit downgrades significantly changed the nature of our indebtedness and impacted our ability to continue with our historical match funding policy. We no longer match fund our indebtedness with cash collections expected to be generated from finance receivables. We expect to pay down our outstanding obligations as they mature. Accordingly, at December 31, 2000, our debt has been classified in the Consolidated Balance Sheets, based on the contractual maturity dates of the underlying debt instruments. Prior years' balances have not been reclassified. The Company believes its liquidity is presently sufficient to meet current and anticipated needs going forward, subject to the timely implementation and execution of various business initiatives. Long-Term Debt. A summary of long-term debt by final maturity date at December 31, 2000 and 1999 follows: Weighted Average Interest Rates at 12/31/00 2000 1999 U.S. Operations Calico Corporation (parent company) Guaranteed ESOP notes due 2000-2003 7.53% $ 221 $ 299 Notes due 2000-0 2,041 Notes due 2001 6.50 737 721 Notes due 2002 7.59 330 230 Notes due 2003 5.61 1,313 1,398 Notes due 2004 5.01 483 502 Notes due 2006 7.25 25 0 Notes due 2007 7.38 25 0 Notes due 2011 7.01 50 0 Notes due 2016 7.20 250 250 Convertible notes due 2018 3.63 617 601 Notes due 2038 5.96 25 25 Revolving credit agreement, maturing in 2002 6.93 4,400 0 Capital leases and other debt due 2000-2018 8.17 91 120 Subtotal 8,567 6,187 93

Calico Credit Corporation Notes due 2000-0 2,026 Notes due 2001 6.66 326 401 Notes due 2002 2.80/1/ 666 668 Notes due 2003 6.61 460 200 Notes due 2005 1.50/1/ 904 0 Notes due 2007 2.00/1/ 270 0 Notes due 2008 6.30 25 0 Notes due 2012 7.09 125 0 Notes due 2013 6.50 60 0 Notes due 2014 6.06 50 0 Notes due 2018 7.00 25 0 Secured borrowings due 2001-2003 6.70 325 0 Revolving credit agreement, maturing in 2002 6.94 1,020 0 Floating rate notes due 2048 6.44 60 60 Subtotal 4,316 3,355 Total U.S. operations $ 12,883 $ 9,542 /1/ Weighted average interest rates include Japanese yen bonds of $1,174 and $488 issued by Calico Credit Corporation in 2000 and 1999, respectively, with interest rates ranging from 1.50-2.00% and 0.80%, respectively. International Operations Calico Capital (Europe) Weighted Average Interest Rates at 12/31/00 2000 1999 Various obligations, payable in: Euros due 2000-2008 5.50 % $ 698 $ 755 Japanese yen due 2001-2005 0.53 950 0 U.S. dollars due 2000-2008 6.10 1,025 1,991 Revolving credit agreement, maturing in 2002 (U.S. dollars) 6.96 1,080 0 Subtotal 3,753 2,746 Other International Operations Various obligations, payable in: Canadian dollars due 2000-2007 11.74 55 88 Pounds sterling due 2000-2003 9.00 187 202 94

Italian lire due 2000-2001 4.72 117 133 Euros due 2000-2008 7.90 159 194 U.S. dollars due 2000-2008 7.67 128 249 Revolving credit agreement, maturing in 2002 (U.S. dollars) 6.83 500 0 Capital leases and other debt due 2000-2004 6.23 5 20 Subtotal 1,151 886 Total international operations 4,904 3,632 Other borrowing deemed longterm 0 1,827 Subtotal 17,787 15,001 Less current maturities 2,383 3,957 Total long term debt $ 15,404 $ 11,044 Consolidated Long-Term Debt Maturities. Payments due on long-term debt for the next five years and thereafter follow: 2001 2002 2003 2004 2005 Thereafter $2,383 $8,994 $2,630 $1,718 $1,010 $1,052 Certain of our debt agreements allow us to redeem outstanding debt prior to scheduled maturity. Outstanding debt issues with call features are classified in the preceding five-year maturity table in accordance with management's current expectations. The actual decision as to early redemption will be made at the time the early redemption option becomes exercisable and will be based on liquidity, prevailing economic and business conditions, and the relative costs of new borrowing. Convertible Debt. In 1998, we issued convertible subordinated debentures for net proceeds of $575. The amount due upon maturity in April 2018 is $1,012, resulting in an effective interest rate of 3.625 percent per annum, including 1.003 percent payable in cash semiannually beginning in October 1998. These debentures are convertible at any time at the option of the holder into 7.808 shares of our stock per $1,000 principal amount at maturity of debentures. This debt contains a put option which requires us to purchase any debenture, at the option of the holder, on April 21, 2003, for a price of $649 per $1,000 principal. We may elect to settle the obligation in cash, shares of common stock, or any combination thereof. Lines of Credit. We have a $7 billion revolving credit agreement with a group of banks, which matures in October 2002. This revolver is also accessible by the following wholly owned subsidiaries: Calico Credit Corporation (up to a $7 billion limit) and Calico Canada Capital Ltd. and Calico Capital (Europe) plc (up to a $4 billion limit) with our guarantee. Amounts borrowed under this facility are at rates based, at the borrower's option, on spreads above certain reference rates such as LIBOR. This agreement contains certain covenants the most restrictive of which require that we maintain a minimum level of tangible net worth and limit the amounts of outstanding secured borrowings, as defined in the agreement. We are in compliance with these covenants at December 31, 2000. The balance outstanding under this line of credit was $7 billion at December 31, 2000. In addition, our foreign subsidiaries had unused committed long- 95

term lines of credit used to back short-term indebtedness that aggregate $43 in various currencies at prevailing interest rates. Guarantees. At December 31, 2000, we have guaranteed the borrowings of our ESOP and $4,710 of indebtedness of our foreign subsidiaries. Interest. Interest paid by us on our short- and long-term debt, amounted to $1,024, $787 and $859 for the years ended December 31, 2000, 1999 and 1998, respectively. A summary of the cash related changes in consolidated indebtedness for the three years ended December 31, 2000 follows: 2000 1999 1998 Cash proceeds from (payments of) short term debt, net $ (1,234) $ (4,140) $ 553 Cash proceeds from long-term debt 10,520 5,446 3,464 Principal payments on long-term debt (5,713) (1,489) (1,580) Total net cash changes in debt $ 3,573 /1/ $ (183) /2/ $ 2,437 /1/ Excludes debt of $118, which was assumed by Dao Calico in connection with the divestiture of our China operations, and accretion of $16 on convertible debt. /2/ Excludes debt of $51 assumed with the increased ownership in our India joint venture and accretion of $26 on convertible debt. 12. Financial Instruments Derivative Financial Instruments. Certain financial instruments with off-balance-sheet risk have been entered into by us to manage our interest rate and foreign currency exposures. These instruments are held solely for hedging purposes and include interest rate swap agreements, forward exchange contracts and foreign currency swap agreements. We do not enter into derivative instrument transactions for trading or other speculative purposes. We typically enter into simple, unleveraged derivative transactions which, by their nature, have low credit and market risk. Our policies on the use of derivative instruments prescribe an investment-grade counterparty credit floor and at least quarterly monitoring of market risk on a counterparty-bycounterparty basis. We utilize numerous counterparties to ensure that there are no significant concentrations of credit risk with any individual counterparty or groups of counterparties. Based upon our ongoing evaluation of the replacement cost of our derivative transactions and counterparty creditworthiness, we consider the risk of credit default significantly affecting our financial position or results of operations to be remote. We employ the use of hedges to reduce the risks that rapidly changing market conditions may have on the underlying transactions. Typically, our currency and interest rate hedging activities are not affected by changes in market conditions, as forward contracts and swaps are arranged and normally held to maturity in order to lock in currency rates and interest rate spreads related to underlying transactions. During 2000 the agencies that assign ratings to our debt downgraded our debt. These downgrades significantly reduced our access to capital markets. Furthermore, the specific downgrade of our debt on December 1, 2000 triggered the repurchases of a number of derivative contracts, which were in place at 96

that time, and further downgrades could require that we repurchase additional outstanding contracts. Therefore, our ability to continue to effectively manage the risks associated with interest rate and foreign currency fluctuations, including our ability to employ our match funding strategy, has been severely constrained. These derivative contract repurchases resulted in un-hedged foreign currency denominated assets and liabilities. We recorded mark-to-market gains during December 2000 of $69 as a direct result of these un-hedged exposures. None of our hedging activities involves exchange-traded instruments. Interest Rate Swaps. We enter into interest rate swap agreements to manage interest rate exposure, although the recent downgrades of our indebtedness have limited our ability to manage this exposure. An interest rate swap is an agreement to exchange interest rate payment streams based on a notional principal amount. We follow settlement accounting principles for interest rate swaps whereby the net interest rate differentials to be paid or received are recorded currently as adjustments to interest expense. Virtually all customer financing assets earn fixed rates of interest. Accordingly, through the use of interest rate swaps in conjunction with the contractual maturity terms of outstanding debt, we "lock in" an interest spread by arranging fixed-rate interest obligations with maturities similar to the underlying assets. Additionally, in industrialized countries customer financing assets are funded with liabilities denominated in the same currency. We refer to the effects of these conservative practices as "match funding" our customer financing assets. This practice effectively eliminates the risk of a major decline in interest margins resulting from adverse changes in the interest rate environment. Conversely, this practice does effectively eliminate the opportunity to materially increase margins when interest rates are declining. As previously disclosed, our credit ratings have been downgraded during 2000. These downgrades have severely limited our current ability to manage our exposure to interest rate changes which has historically been managed through the practice of match funding our finance receivables. More specifically, pay-fixed/receive-variable interest rate swaps are often used in place of more expensive fixed-rate debt for the purpose of match funding fixed-rate customer contracts. Pay-variable/receive-variable interest rate swaps (basis swaps) are used to transform variable rate, medium-term debt into commercial paper or local currency LIBOR rate obligations. Pay-variable/receivefixed interest rate swaps are used to transform term fixed-rate debt into variable rate obligations. The transactions performed within each of these three categories enable the cost-effective management of interest rate exposures. During 2000, the average notional amount of an interest rate swap agreement was $25. The total notional amounts of these transactions at December 31, 2000 and 1999, based on contract maturity, follow: 2000 1999 Commercial paper/bank borrowings $ 4,538 $ 5,352 Medium-term debt 8,666 10,493 Long-term debt 2,267 4,238 Total $15,471 $20,083 97

The aggregate notional amounts of interest rate swaps by maturity date and type at December 31, 2000 and 1999 follow: 2000 2001 2002-2005-2018 Total 2004 2000 Pay fixed/ receive variable $ 0 $ 1,321 $ 4,490 $ 1,319 $ 7,130 Pay variable/ receive variable 0 1,677 1 0 1,678 Pay variable/ receive fixed 0 1,540 4,175 948 6,663 Total $ 0 $ 4,538 $ 8,666 $ 2,267 $ 15,471 Memo: Interest rate paid - 5.74% 5.95% 7.01% 6.04% Interest rate received - 5.08% 5.85% 5.36% 5.55% 1999 Pay fixed/ receive variable $ 2,699 $ 2,202 $ 6,742 $ 340 $ 11,983 Pay variable/ receive variable 443 550 0 0 993 Pay variable/ receive fixed 2,210 718 3,544 635 7,107 Total $ 5,352 $ 3,470 $ 10,286 $ 975 $ 20,083 Memo: Interest rate paid 5.94% 4.39% 5.41% 6.25% 5.42% Interest rate received 5.48% 5.18% 5.38% 6.75% 5.44% Forward Exchange Contracts. We utilize forward exchange contracts to hedge against the potentially adverse impacts of foreign currency fluctuations on foreign currency-denominated receivables and payables; firm foreign currency commitments; and investments in foreign operations. Firm foreign currency commitments generally represent committed purchase orders for foreign-sourced inventory. These contracts generally mature in six months or less. At December 31, 2000 and 1999, we had outstanding forward exchange contracts of $1,788 and $3,838, respectively. Of the outstanding contracts at December 31, 2000, the largest single currency represented was the Euro. Contracts denominated in Euros, Canadian dollars, U.S. dollars, Brazilian reais and Japanese yen accounted for over 90 percent of our forward exchange contracts. On contracts that hedge foreign currency-denominated receivables and payables, gains or losses are reported currently in income, and premiums or discounts are amortized to income and included in Other, net in the Consolidated Statements of Operations. Gains or losses, as well as premiums or discounts, on contracts that hedge firm commitments are deferred and subsequently recognized as part of the underlying transaction. At December 31, 2000, we had a net deferred loss of $8. Gains or losses on contracts that hedge an investment in a foreign operation are reported currently in the balance sheet as a component of cumulative translation adjustments. The premium or discount on contracts that hedge an investment in a foreign operation are amortized to income and included in Other, net in the Consolidated Statements of Operations. During 2000, the average notional amount of a forward exchange contract amounted to $14. Foreign Currency Swap Agreements. We enter into cross-currency interest rate swap agreements, whereby we issue foreign currency-denominated debt and swap the proceeds with a counterparty. In return, we receive and effectively denominate the debt in local currencies. Currency swaps are utilized as hedges of the underlying foreign currency borrowings, and exchange gains or losses are recognized currently in Other, net in the Consolidated Statements of Operations. At December 31, 2000 and 1999, we had outstanding cross-currency interest rate swap agreements with aggregate notional amounts of $4,222 and $3,968, respectively. Of the outstanding agreements at December 31, 2000, the largest single 98

currency represented was the U.S. dollar. Contracts denominated in U.S. dollars, British pounds sterling, Japanese yen and French francs accounted for over 75 percent of our currency interest rate swap agreements. Fair Value of Financial Instruments. The estimated fair values of our financial instruments at December 31, 2000 and 1999 follow: 2000 1999 Carrying amount Fair value Carrying amount Fair value Cash and cash equivalents $ 1,741 $ 1,741 $ 126 $ 126 Accounts receivable, net 2,281 2,281 2,633 2,633 Short-term debt 2,693 2,356 3,957 3,957 Long-term debt 15,404 9,433 11,044 10,882 Interest rate and currency swap agreements 0 129 0 (40) Forward exchange contracts 0 (59) 0 131 The fair value amounts for Cash and cash equivalents and Accounts receivable, net approximate carrying amounts due to the short maturities of these instruments. The fair value of Short and Long-term debt was estimated based on quoted market prices for these or similar issues or on the current rates offered to us for debt of the same remaining maturities. The difference between the fair value and the carrying value represents the theoretical net premium or discount we would pay or receive to retire all debt at such date. We have no plans to retire significant portions of our debt prior to scheduled maturity. We are not required to determine the fair value of our finance receivables. The fair values for interest rate and cross-currency swap agreements and forward exchange contracts were calculated by us based on market conditions at year-end and supplemented with quotes from brokers. They represent amounts we would receive (pay) to terminate/replace these contracts. We have no present plans to terminate/replace significant portions of these contracts. 99

13. Employee Benefit Plans We sponsor numerous pension and other postretirement benefit plans in our U.S. and international operations. Pension Benefits Other Benefits Change in Benefit Obligation 2000 1999 2000 1999 Benefit obligation, January 1 $ 8,418 $ 8,040 $ 1,060 $ 1,095 Service cost 167 191 24 27 Interest cost 453 1,009 85 77 Plan participants contributions 19 14 0 0 Plan amendments 1 0 0 0 Actuarial (gain)/loss 48 (79) 218 (78) Currency exchange rate changes (197) (139) (2) 2 Divestitures (15) 0 0 0 Curtailments (10) (3) 0 0 Settlements 0 2 0 0 Special termination benefits 34 11 4 2 Benefits paid (663) (628) (75) (65) Benefit obligation, Dec. 31 8,255 8,418 1,314 1,060 Change in Plan Assets Fair value of plan assets, Jan. 1 8,771 7,958 0 0 Actual return on plan assets 651 1,422 0 0 Employer contribution 84 96 75 65 Plan participants contribution 19 14 0 0 Currency exchange rate changes (218) (91) 0 0 Divestitures (18) 0 0 0 Benefits paid (663) (628) (75) (65) Fair value of plan assets, December 31 8,626 8,771 0 0 Funded status (including under funded and non-funded) 371 353 (1,314) (1,060) Unamortized transition assets (15) (36) 0 0 Unrecognized prior service cost 17 21 (3) (4) Unrecognized net actuarial (gain) loss (433) (381) 120 (69) Net amount recognized $ (60) $ (43) $ (1,197) $ (1,133) Amounts recognized in the consolidated balance sheets consist of: Prepaid benefit cost $ 378 $ 377 $ 0 $0 Accrued benefit liability (468) (456) (1,197) (1,133) Intangible asset 3 4 0 0 Accumulated other comprehensive income 27 32 0 0 Net amount recognized $ (60) $ (43) $ (1,197) $ (1,133) Under-funded or non-funded plans Aggregate benefit obligation $ 348 $ 497 $ 1,314 $ 1,060 Aggregate fair value of plan assets $ 180 $ 174 $ 0 $ 0 Weighted average assumptions as of Dec. 31 Discount rate 7.0% 7.4% 7.5% 8.0% Expected return on plan assets 8.9 8.9 Rate of compensation increase 3.8 4.2 100

Pension Benefits Other Benefits 2000 1999 1998 2000 1999 1998 Components of Net Periodic Benefit Cost Defined benefit plans Service cost $ 167 $ 191 $ 172 $ 24 $ 27 $ 26 Interest cost 453 1,009 916 85 77 72 Expected return on plan assets (522) (1,090) (1,010) 0 0 0 Recognized actuarial (gain)/loss 4 11 10 0 1 0 Amortization of prior service cost 4 8 6 0 0 0 Recognized net transition asset (16) (18) (19) 0 2 0 Recognized curtailment/settlement (46) (9) (60) 0 0 0 gain Net periodic benefit cost 44 102 15 109 107 98 Defined contribution plans 14 28 32 0 0 0 Total $ 58 $ 130 $ 47 $ 109 $ 107 $ 98 Assumed health care cost trend rates have a significant effect on the amounts reported for the health care plans. For measurement purposes, an 8.5 percent annual rate of increase in the per capita cost of covered health care benefits was assumed for 2000. The rate was assumed to decrease to 5.25 percent in 2005 and thereafter. A one-percentage-point change in assumed health care cost trend rates would have the following effects: One-percentage- point One-percentage- point decrease increase Effect on total service and interest cost Components $ 4 $ (3) Effect on postretirement benefit obligation $ 75 $ (60) Employee Stock Ownership Plan (ESOP) Benefits. In 1989, we established an ESOP and sold to it 10 million shares of Series B Convertible Preferred Stock (Convertible Preferred) of the Company for a purchase price of $785. Each ESOP share is convertible into six common shares of the Company. The Convertible Preferred has a $1 par value and a guaranteed minimum value of $78.25 per share and accrues annual dividends of $6.25 per share. The ESOP borrowed the purchase price from a group of lenders. The ESOP debt is included in our consolidated balance sheets because we guarantee the ESOP borrowings. A corresponding amount classified as Deferred ESOP benefits represents our commitment to future compensation expense related to the ESOP benefits. The ESOP will repay its borrowings from dividends on the Convertible Preferred and from our contributions. The ESOP's debt service is structured such that our annual contributions (in excess of 101

dividends) essentially correspond to a specified level percentage of participant compensation. As the borrowings are repaid, the Convertible Preferred is allocated to ESOP participants and Deferred ESOP benefits are reduced by principal payments on the borrowings. Most of our domestic employees are eligible to participate in the ESOP. Information relating to the ESOP for the three years ended December 31, 2000 follows: 2000 1999 1998 Interest on ESOP Borrowings $ 24 $ 28 $ 33 Dividends declared on Convertible Preferred Stock $ 53 $ 54 $ 56 Cash contribution to the ESOP $ 49 $ 44 $ 41 Compensation expense $ 48 $ 46 $ 44 We recognize ESOP costs based on the amount committed to be contributed to the ESOP plus related trustee, finance and other charges. 14. Income and Other Taxes The parent company and its domestic subsidiaries file consolidated U.S. income tax returns. Generally, pursuant to tax allocation arrangements, domestic subsidiaries record their tax provisions and make payments to the parent company for taxes due or receive payments from the parent company for tax benefits utilized. Income (loss) before income taxes from continuing operations for the three years ended December 31, 2000 consists of the following: 2000 1999 1998 Domestic income $ 49 $ 1,176 $ 616 Foreign income (loss) (433) 732 (37) Income (loss) before income taxes $ (384) $ 1,908 $ 579 Provisions (benefits) for income taxes from continuing operations for the three years ended December 31, 2000 consist of the following: 2000 1999 1998 Federal income taxes Current $ 8 $ 168 $ 265 Deferred (131) 166 (152) Foreign income taxes Current 76 124 178 Deferred (77) 51 (201) State income taxes Current 23 52 70 Deferred (8) 27 (15) Income taxes $ (109) $ 588 $ 145 102

A reconciliation of the U.S. federal statutory income tax rate to the effective income tax rate for continuing operations for the three years ended December 31, 2000 follows: 2000 1999 1998 U.S. federal statutory income tax rate (35.0) % 35.0 % 35.0 % Foreign earnings and dividends taxed at different rates 40.7 (7.0) (9.0) Goodwill amortization 3.0.7 1.0 Tax-exempt income (4.1) (1.0) (3.0) State taxes 1.6 2.7 6.2 Audit resolutions (32.6) 0 0 Other (2.0).4 (5.2) Effective income tax rate (28.4) % 30.8 % 25.0 % The 2000 effective tax rate of (28.4) percent includes a tax benefit for the 2000 restructuring, the CPID in-process research and development write-off, and the tax provision for the gain on sale of the China operations. Excluding these items, the 2000 effective tax rate is 32.1 percent which is 1.3 percentage points higher than 1999. The increase in the effective tax rate is due primarily to losses in a low-tax rate jurisdiction offset by favorable resolution of tax audits. The 1999 effective tax rate of 30.8 percent is 0.7 percentage points lower than 1998 after excluding the 1998 worldwide restructuring program from the 1998 effective tax rate. On a consolidated basis, we paid a total of $354, $238 and $217 in income taxes to federal, foreign and state income-taxing authorities in 2000, 1999 and 1998, respectively. Total income tax expense (benefit) for the three years ended December 31, 2000 was allocated as follows: 2000 1999 1998 Income taxes (benefits) on income (loss) from continuing operations $ (109) $ 588 $ 145 Tax benefit included in minorities interests (20) (20) (20) Discontinued operations 0 (26) (54) Goodwill (42) 0 0 Common shareholders equity 39 (106) (140) Total $ (132) $ 436 $ (69) Deferred income taxes have not been provided on the undistributed earnings of foreign subsidiaries and other foreign investments carried at equity. The amount of such earnings included in consolidated retained earnings at December 31, 2000 was approximately $5.0 billion. These earnings have been substantially reinvested, and we do not plan to initiate any action that would precipitate the payment of income taxes thereon, except for any actions contemplated by the Company's turnaround program. It is not practicable to estimate the amount of additional tax that might be payable on the foreign earnings. 103

The tax effects of temporary differences that give rise to significant portions of the deferred taxes at December 31, 2000 and 1999 follow: 2000 1999 Tax effect of future tax deductions Depreciation $ 386 $ 385 Postretirement medical benefits 448 438 Restructuring reserves 143 175 Other operating reserves 162 199 Allowance for doubtful accounts 170 111 Deferred compensation 149 159 Tax credit carryforwards 159 116 Research and development 866 641 Other 270 283 $ 2,753 $ 2,507 Valuation allowance (51) (49) Total $ 2,702 $ 2,458 Tax effect of future taxable income Installment sales and leases $ (872) $ (962) Deferred income (1,017) (846) Other (298) (348) Total $ (2,187) $ (2,156) The valuation allowance for deferred tax assets as of January 1, 1999 was $53. The net change in the total valuation allowance for the years ended December 31, 2000 and 1999 was an increase of $2 and a decrease of $4, respectively. The valuation allowance relates to foreign credit carryforwards and foreign net operating loss carryforwards for which the Company has concluded it is more likely than not that these tax credits and net operating loss carryforwards will not be realized in the ordinary course of operations. The above amounts are classified as current or long-term in the Consolidated Balance Sheets in accordance with the asset or liability to which they relate. Current deferred tax assets at December 31, 2000 and 1999 amounted to $450 and $478, respectively. Although realization is not assured, we have concluded that it is more likely than not that the deferred tax assets for which a valuation allowance was determined to be unnecessary, will be realized in the ordinary course of operations based on scheduling of deferred tax liabilities and income from operating activities. The amount of the net deferred tax assets considered realizable, however, could be reduced in the near term if actual future income taxes are lower than estimated, or if there are differences in the timing or amount of future reversals of existing taxable temporary differences. A substantial portion of our net deferred tax assets are in jurisdictions where the net operating loss carryforward periods are either unlimited (net deferred tax asset of $64) or 20 years (net deferred tax asset of $1.2 billion). At December 31, 2000, we have tax credit carryforwards for income tax purposes of $159 available to offset future income taxes, of which $136 is available to carryforward indefinitely. We also have net operating loss carryforwards for income tax purposes of $157 that are available to offset future taxable income through 2007 and $1.0 billion available to offset future taxable income indefinitely. 104

The Company incurs indirect taxes such as property and payroll taxes in the various countries in which it operates. Changes in estimates for these taxes occur in the ordinary course of accounting for such items. Changes resulting from, but not limited to, refinements of tax computations, systems and other procedural changes as well as other factors amounted to an increase in pre-tax income (loss) of $17, $35 and $21 in the years 2000, 1999 and 1998, respectively. The Company is also subject to sales and consumption taxes in the various countries in which it operates. Changes in estimates for these taxes resulting from structural realignments or other factors amounted to an increase in pre-tax income (loss) of $11 and $51 in the years 2000 and 1998, respectively. Calico's Brazilian operations have received assessments for indirect taxes totaling approximately $400 million related principally to the internal transfer of inventory. We do not agree with these assessments and intend to vigorously defend our position. We, as supported by the opinion of legal counsel, do not believe that the ultimate resolution of these assessments will materially impact the consolidated financial statements. 15. Litigation On April 11, 1996, an action was commenced by Dalmation Corp. (Dalmation), in the United States District Court for the Southern District of New York, against the Company seeking unspecified damages for infringement of a patent of Dalmation which expired in 1993. The suit, as amended, was directed to facsimile and certain other products containing scanning functions and sought damages for sales between 1990 and 1993. On April 1, 1998, the jury entered a verdict in favor of Dalmation for $40. However, on September 14, 1998, the court granted the Company's motion for a new trial on damages. The trial ended on October 25, 1999 with a jury verdict of $10. The Company's motion to set aside the verdict or, in the alternative, to grant a new trial was denied by the court. The Company is appealing to the Court of Appeals for the Federal Circuit. Dalmation is appealing the new trial grant which reduced the verdict from $40 and seeking a reversal of the jury's finding of no willful infringement. Briefing at the Court of Appeals for the Federal Circuit is complete and oral argument took place on May 9, 2001. On June 24, 1999, the Company was served with a summons and complaint filed in the Superior Court of the State of California for the County of Los Angeles. The complaint was filed on behalf of 681 individual plaintiffs claiming damages as a result of the Company's alleged disposal and/or release of hazardous substances into the soil, air and groundwater. On July 22, 1999, April 12, 2000, November 30, 2000, and March 31, 2001 respectively, four additional complaints were filed in the same court on behalf of an additional 79, 141, 76, and 51 plaintiffs, respectively, with the same claims for damages as the June 1999 action. Three of the four additional cases have been served on the Company. Plaintiffs in all five cases further allege that they have been exposed to such hazardous substances by inhalation, ingestion and dermal contact, including but not limited to hazardous substances contained within the municipal drinking water supplied by the City of Salvia and the California Water Company. Plaintiffs' claims against Registrant include personal injury, wrongful death, property damage, negligence, trespass, nuisance, absolute liability for ultra-hazardous activities, civil conspiracy, battery and violation of the California Unfair Trade Practices Act. Damages are unspecified. The Company denies any liability for the plaintiffs' alleged damages and intends to vigorously defend these actions. The Company has not answered or appeared in any of the cases because of an agreement 105

among the parties and the court to stay these cases pending resolution of several similar cases currently pending before the California Supreme Court. However, the court recently directed that the five cases against the Company be coordinated with a number of other unrelated groundwater cases pending in California. 16. Preferred Securities As of December 31, 2000, we have four series of outstanding preferred securities. In total we are authorized to issue 22 million shares of cumulative preferred stock, $1 par value. Convertible Preferred Stock. We sold, for $785, 10 million shares of our Series B Convertible Preferred Stock (ESOP shares) in 1989 in connection with the establishment of our ESOP. As employees with vested ESOP shares leave the Company, these shares are redeemed by us. We have the option to settle such redemptions with either shares of common stock or cash. Outstanding preferred stock related to our ESOP at December 31, 2000 and 1999 follows (shares in thousands): 2000 1999 Shares Amount Shares Amount Convertible Preferred Stock 8,260 $ 647 8,551 $ 669 Preferred Stock Purchase Rights. We have a shareholder rights plan designed to deter coercive or unfair takeover tactics and to prevent a person or persons from gaining control of us without offering a fair price to all shareholders. Under the terms of the plan, one-half of one preferred stock purchase right (Right) accompanies each share of outstanding common stock. Each full Right entitles the holder to purchase from us one threehundredth of a new series of preferred stock at an exercise price of $250. Within the time limits and under the circumstances specified in the plan, the Rights entitle the holder to acquire either our common stock, the surviving company in a business combination, or the purchaser of our assets, having a value of two times the exercise price. The Rights may be redeemed prior to becoming exercisable by action of the Board of Directors at a redemption price of $.01 per Right. The Rights expire in April 2007. The Rights are non-voting and, until they become exercisable, have no dilutive effect on the earnings per share or book value per share of our common stock. Deferred Preferred Stock. In 1996, a subsidiary of ours issued 2 million deferred preferred shares for Canadian (Cdn.) $50 million. The U.S. dollar value was $37 and is included in Minorities' interests in equity of subsidiaries in the Consolidated Balance Sheets. These shares are mandatorily redeemable on February 28, 2006 for Cdn. $90 million. The difference between the redemption amount and the proceeds from the issue is being amortized, through the redemption date, to Minorities' interests in earnings of subsidiaries in the Consolidated Statements of Operations. We have guaranteed the redemption value. 106

Company-obligated, mandatorily redeemable preferred securities of subsidiary trust holding solely subordinated debentures of the Company. In 1997, a trust sponsored and wholly owned by the Company issued $650 aggregate liquidation amount preferred securities (the Original Preferred Securities) to investors and 20,103 shares of common securities to the Company, the proceeds of which were invested by the trust in $670 aggregate principal amount of the Company's newly issued 8 percent Junior Subordinated Debentures due 2027 (the Original Debentures). In June 1997, pursuant to a registration statement filed by the Company and the trust with the Securities and Exchange Commission, Original Preferred Securities with an aggregate liquidation preference amount of $644 and Original Debentures with a principal amount of $644 were exchanged for a like amount of preferred securities (together with the Original Preferred Securities, the Preferred Securities) and 8 percent Junior Subordinated Debentures due 2027 (together with the Original Debentures, the Debentures) which were registered under the Securities Act of 1933. The Debentures represent all of the assets of the trust. The proceeds from the issuance of the Original Debentures were used by the Company for general corporate purposes. The Debentures and related income statement effects are eliminated in the Company's consolidated financial statements. The Preferred Securities accrue and pay cash distributions semiannually at a rate of 8 percent per annum of the stated liquidation amount of $1,000 per Preferred Security. The Company has guaranteed (the Guarantee), on a subordinated basis, distributions and other payments due on the Preferred Securities. The Guarantee and the Company's obligations under the Debentures and in the indenture pursuant to which the Debentures were issued and the Company's obligations under the Amended and Restated Declaration of Trust governing the trust, taken together, provide a full and unconditional guarantee of amounts due on the Preferred Securities. The Preferred Securities are mandatorily redeemable upon the maturity of the Debentures on February 1, 2027, or earlier to the extent of any redemption by the Company of any Debentures. The redemption price in either such case will be $1,000 per share plus accrued and unpaid distributions to the date fixed for redemption. 17. Common Stock We have 1.05 billion authorized shares of common stock, $1 par value. At December 31, 2000 and 1999, 98.1 and 84.3 million shares, respectively, were reserved for issuance under our incentive compensation plans. In addition, at December 31, 2000, 13.2 million common shares were reserved for the conversion of $670 of convertible debt, and 48.9 million common shares were reserved for conversion of ESOPrelated Convertible Preferred Stock. Treasury Stock. The Board of Directors has authorized us to repurchase up to $1 billion of our common stock. The stock may be repurchased from time to time on the open market depending on market and other conditions. No shares were repurchased during 2000 or 1999. During 1998, we repurchased 3.7 million shares for $172. Since inception of the program we have repurchased 20.6 million shares for $594. Common shares issued for stock option exercises, conversion of convertible securities and other exchanges were partially satisfied by reissuances of treasury shares. Put Options. In connection with the share repurchase program, during 2000, 1999 and 1998, we sold 7.5 million, 0.8 million and 1.0 million put options, respectively, that entitle the holder to sell one share of 107

our common stock to us at maturity at a specified price. These put options can be settled in cash at our option. The put options had original maturities ranging from six months to two years. In 2000, we recorded the receipt of a premium of approximately $24 on the sale of equity put options. This premium was recorded as an addition to Common shareholders' equity. In October 2000, the holder of these equity put options exercised their option for early termination and settlement. The cost of this settlement to the Company was approximately $92 for 7.5 million shares with an average strike price of $18.98 per share. This transaction was recorded as a reduction of Common shareholders' equity. At December 31, 2000, 0.8 million put options remain outstanding with a strike price of $40.56 per share. Under the terms of this contract we had the option of physical or net cash settlement. Accordingly, this amount is classified as temporary equity in the consolidated balance sheets at December 31, 2000. In January 2001 these put options were net cash settled for $28. Funds for this net cash settlement were obtained by selling 5.9 million unregistered shares of our common stock for proceeds of $28. In 1999, put options on 1.0 million shares of common stock were exercised and settled for a net cash payment of $5. Stock Option and Long-Term Incentive Plans. We have a long-term incentive plan whereby eligible employees may be granted nonqualified stock options and performance unit rights. Beginning in 1998 and subject to vesting and other requirements, performance unit rights are typically paid in our common stock. The value of each performance unit is based on the growth in earnings per share during the year in which granted. Performance units ratably vest in the three years after the year awarded. Stock options and rights are settled with newly issued or, if available, treasury shares of our common stock. Stock options generally vest in three years and expire between eight and ten years from the date of grant. The exercise price of the options is equal to the market value of our common stock on the effective date of grant. At December 31, 2000 and 1999, 36.0 million and 36.2 million shares, respectively, were available for grant of options or rights. The following table provides information relating to the status of, and changes in, options granted: Employee Stock Options 2000 1999 1998 (Options in thousands) Stock Options Average Option Stock Options Average Option Price Stock Options Average Option Price Price Outstanding at January 1 43,388 $42 30,344 $33 27,134 $26 Granted 19,338 22 19,059 51 8,980 47 Cancelled (4,423) 38 (870) 47 (199) 37 Exercised (70) 22 (5,145) 23 (5,571) 20 Outstanding at December 31 58,233 35 43,388 42 30,344 33 Exercisable at end of year 23,346 13,467 9,622 108

Options outstanding and exercisable at December 31, 2000 are as follows: Thousands except per-share data Options Outstanding Options Exercisable Range of Exercise Prices Number Outstanding Number Exercisable Weighted Average Remaining Contractual Life Weighted Average Exercise Price Weighted Average Exercise Price $10.94 to $16.38 319 8.25 $14.59 0 $0 16.91 to 23.25 22,694 7.34 21.47 5,780 20.59 25.38 to 36.70 13,799 5.67 31.43 7,794 33.04 41.72 to 60.95 21,421 6.33 53.26 9,772 51.45 $10.94 to $60.95 58,233 6.58 $35.48 23,346 $37.66 We do not recognize compensation expense relating to employee stock options because the exercise price of the option equals the fair value of the stock on the effective date of grant. If we had determined the compensation based on the value as determined by the modified Black-Scholes option pricing model, the pro forma net income (loss) and earnings (loss) per share would be as follows: 2000 1999 1998 Net income (loss) as reported $ (257) $ 1,339 $ 273 Net income (loss) pro forma (359) 1,238 228 Basic EPS - as reported (0.44) 1.96 0.34 Basic EPS pro forma (0.59) 1.81 0.27 Diluted EPS as reported (0.44) 1.85 0.34 Diluted EPS pro forma (0.59) 1.71 0.27 These pro forma disclosures are not necessarily indicative of future amounts. As reflected in the pro forma amounts in the previous table, the fair value of each option granted in 2000, 1999 and 1998 was $7.50, $15.83 and $13.31, respectively. The fair value of each option granted was estimated on the date of grant using the following weighted average assumptions: 2000 1999 1998 Risk-free interest 6.7% 5.1% 5.2% Expected life in years 7.1 6.2 5.3 Expected volatility 37.0% 28.0% 24.9% Expected divided yield 3.7% 1.8% 1.4% 109

18. Earnings per Share A reconciliation of the numerators and denominators of the basic and diluted EPS calculation follows: Income (Numerator) Shares (Denominator) 2000 1999 1998 Pershare Income Shares Per- Income (Numer (Denom share (Numer- amount -ator) -inator) amount ator) In March 2001, we completed the sale of half of our ownership interest in Dao Calico to Daofilm for $1,283 in cash. The sale resulted in a pre-tax gain of $769 ($300 after taxes). Under the agreement, 110 Shares (Denominator) Pershare amount (Shares in thousands) Basic EPS Income (loss) from continuing operations $ (257) $ 1,339 $ 463 Accrued dividends on preferred stock, net (35) (38) (46) Basic EPS $ (292) 667,581 $(0.44) $1,301 663,493 $1.96 $417 658,956 $0.63 Diluted EPS Stock options and other incentives 8,727 9,811 ESOP adjustment, net of tax 43 51,989 Convertible debt, net of tax 17 13,191 3 5,287 Diluted EPS $ (292) 667,581 $(0.44) $1,361 737,400 $1.85 $420 674,054 $0.62 Note: Recalculation of per-share amounts may be off by $0.01 in certain instances due to rounding. 19. Subsequent Events In January 2001, we transferred $898 of finance receivables to a special purpose entity for cash proceeds of $435, received from an affiliate of Omni Power Capital Corporation (OP Capital), and a retained interest of $463. The proceeds were accounted for as a secured borrowing. At March 31, 2001 the balance of receivables transferred was $734 and is included in Finance receivables, net in the Consolidated Balance Sheets. The remaining secured borrowing balance of $340 is included in Debt. The total proceeds of $435 are included in the Net change in debt in the Consolidated Statements of Cash Flows. The borrowing will be repaid over 18 months and bears interest at the rate of 8.98 percent. In the first five months of 2001, we retired $136 of long-term debt through the exchange of 17 million shares of common stock valued at $107. The retirements resulted in a pre-tax extraordinary gain of $29 ($18 after taxes) for a net equity increase of approximately $125.

Daofilm's ownership interest in Dao Calico increased from 50 percent to 75 percent. While Calico's ownership interest decreased to 25 percent, we retain rights as a minority shareholder. All product and technology agreements between us and Dao Calico will continue, ensuring that the two companies retain uninterrupted access to each others portfolio of patents. We maintain a cash position of approximately $194 in a trust account representing the par value and one years interest relating to the bonds issued by our subsidiary Calico Finance (Nederland) BV. This cash is withdrawable upon 21 days written notice to the Trustee. During the first quarter of 2001, and in connection with the turnaround program, we recorded an additional pre-tax restructuring provision totaling $108 ($73 after taxes), in connection with finalized initiatives under the turnaround program. This charge includes estimated costs of $97 for severance costs associated with work force reductions related to the elimination of 1,000 positions worldwide and $11 of asset impairments. The severance costs relate to continued streamlining of existing work processes, elimination of redundant resources and the consolidation of existing activities into other existing operations. In April 2001, we announced the sale of our leasing businesses in four European countries to Sonia Leasing (Sonia) for approximately $370 in cash. The assets were sold for approximate book value and include the leasing portfolios in the respective countries, title to the underlying equipment included in the lease portfolios and certain employees and systems used in the operations of the businesses. Under the terms of the agreement Sonia will provide on-going exclusive equipment financing to Calico customers in those countries. 111

APPENDIX C EXAMPLE OF EXPERIMENT TRANSCRIPTIONS AND CODING 1. 2. 3. 4. How and where you believe the client s financial statements might be susceptible to material misstatement due to fraud? Cash is understated by $550 million- sale of assets (1) Equipment/Manufacturing is overstated-did not record sale (2) Finance receivable is understated by amount of note receivable (3) Minority interest is understated (4) How management could perpetrate and conceal this fraudulent financial reporting? In unconsolidated financial statements of subsidiaries- environment susceptible to fraud Opportunities exist when so much operations takes place in foreign countries 5. 6. 7. 8. 9. 10. CODER #1 (# of unique ideas): 4 112

APPENDIX D HUMAN SUBJECTS COMMITTEE APPROVAL Office of the Vice President for Research Tallahassee, Florida 32306-2763 (850) 644-8673 - FAX (850) 644-4392 APPROVAL MEMORANDUM Human Subjects Committee Date: 1/28/2004 Tina Carpenter 449 Teal Lane Tallahassee, FL 32308 Dept.: Accounting From: John Tomkowiak, Chair Re: Use of Human Subjects in Research Audit team brainstorming: Can we improve auditors' fraud risk assessments and professional skepticism? The forms that you submitted to this office in regard to the use of human subjects in the proposal referenced above have been reviewed by the Secretary, the Chair, and two members 113

of the Human Subjects Committee. Your project is determined to be exempt per 45 CFR 46.101(b) 2 and has been approved by an accelerated review process. The Human Subjects Committee has not evaluated your proposal for scientific merit, except to weigh the risk to the human participants and the aspects of the proposal related to potential risk and benefit. This approval does not replace any departmental or other approvals, which may be required. If the project has not been completed by 1/27/2005 you must request renewed approval for continuation of the project. You are advised that any change in protocol in this project must be approved by resubmission of the project to the Committee for approval. Also, the principal investigator must promptly report, in writing, any unexpected problems causing risks to research subjects or others. By copy of this memorandum, the chairman of your department and/or your major professor is reminded that he/she is responsible for being informed concerning research projects involving human subjects in the department, and should review protocols of such investigations as often as needed to insure that the project is being conducted in compliance with our institution and with DHHS regulations. This institution has an Assurance on file with the Office for Protection from Research Risks. The Assurance Number is IRB00000446. Cc: Jane Reimers HSC No. 2003.772 114

REFERENCES Abdel-khalik, A., D. Snowball, and J. Wragge. 1983. The Effects of Certain Internal Audit Variables on the Planning of External Audit Programs. Accounting Review 58 (2): 215-227. American Institute of Certified Public Accountants (AICPA) 1988. The Auditors Responsibility to Detect and Report Errors and Irregularities, Statement on Auditing Standards No. 53. New York, NY: AICPA..1997. Consideration of Fraud in a Financial Statement Audit, Statement on Auditing Standards No. 82. New York, NY: AICPA..2002. Consideration of Fraud in a Financial Statement Audit, Statement on Auditing Standards No. 99. New York, NY: AICPA..2003. AICPA Practice Aid Series, Fraud Detection in a GAAS Audit: SAS No. 99 Implementation Guide. New York, NY: AICPA. Arnold, V., S. Sutton, S. Hayne, and C. Smith. 2000. Group Decision Making: The Impact of Opportunity Cost Time Pressure and Group Support Systems. Behavioral Research in Accounting 12: 69-96. Asare, S. and L. McDaniel. 1996. The Effects of Familiarity with the Preparer and Task Complexity on the Effectiveness of the Audit Review Process. Accounting Review 71 (2): 139-160., G. Trompeter, and A. Wright. 2000. The Effect of Accountability and Time Budgets on Auditors Testing Strategies. Contemporary Accounting Research 17(1): 539-560. 115

, and A. Wright. 2004. The Effectiveness of Alternative Risk Assessment and Program Planning Tools in a Fraud Setting, forthcoming Contemporary Accounting Research. Ballou, B. 2001. The Relationship between Auditor Characteristics and the Nature of the Review Notes of Analytical Procedure Working Papers. Behavioral Research in Accounting 13: 25-48. Bamber, E. 1983. Expert Judgment in the Audit Team: A Source Reliability Approach. Journal of Accounting Research 21 (Autumn): 396-412., and J. Bylinski. 1982. The Audit Team and the Audit Review Process: An Organizational Approach. Journal of Accounting Literature (Spring): 35-58., R. Watson and M. C. Hill. 1996. The Effects of Group Support System Technology on Audit Group Decision Making. Auditing: A Journal of Practice and Theory 15 (1): 122-134., and R. Ramsay. 1997. An Investigation of the Effects of Specialization in Audit Workpaper Review. Contemporary Accounting Research 14 (3): 501-513., and R. Ramsay. 2000. The Effects of Specialization in Audit Workpaper Review on Review Efficiency and Reviewer s Confidence. Auditing: A Journal of Practice and Theory 19 (2):147-157. Beasley, M, J. Carcello and D. Hermanson. 1999. Fraudulent Financial Reporting: 1987-1997, An Analysis of U.S. Public Companies. Committee of Sponsoring Organizations of the Treadway Commission: Jersey City. Bedard, J. and M. Chi. 1992. Expertise. Current Directions in Psychology Science: 135-139.. 1993. Expertise in Auditing. Auditing: A Journal of Practice and Theory (Suppl.): 21-45. Bierstaker, J. and A. Wright. 2001. The Effects of Fee Pressure and Partner Pressure on Audit Planning Decisions. Advances in Accounting 18: 25-46.. 2003. The Effect of Partner Preferences on the Development of Risk-Adjusted Program Plans. Working Paper, Boston University, Boston, MA. Bonner, S. and B. Lewis. 1990. Determinants of Auditor Expertise. Journal of Accounting Research (Suppl.): 1-20. 116

, Z.V. Palmrose, and S. Young. 1998. Fraud Type and Auditor Litigation: An Analysis of SEC Accounting and Auditing Enforcement Releases. Accounting Review 73 (4): 503-532. Braun, R. 2000. The Effect of Time Pressure on Auditor Attention to Qualitative Aspects of Misstatements Indicative of Potential Fraudulent Financial Reporting. Accounting, Organizations and Society 25: 243-259. Brown, C., M. Peecher, and I. Solomon. 1999. Auditors' Hypothesis Testing in Diagnostic Inference Tasks. Journal of Accounting Research 37 (1): 1-26. Cloyd, C. 1997. Performance in Tax Research Tasks: The Joint Effects of Knowledge and Accountability. Accounting Review 72 (1): 111-131., and B. Spilker. 1999. The Influence of Client Preferences on Tax Professionals' Search for Judicial Precedents, Subsequent Judgments and Recommendations. Accounting Review 74 (3): 299-322. Cuccia, A., K. Hackenbrack and M. Nelson. 1995. The Ability of Professional Standards to Mitigate Aggressive Reporting. Accounting Review 70 (2): 227-248. Davis, J. and I. Solomon. 1989. Expertise and Experience in Behavioral Accounting Research. Journal of Accounting Literature: 150-164. Dennis, A. and J. Valacich. 1993. Computer Brainstorms: More Heads Are Better Than One. Journal of Applied Psychology 78 (4): 531-537. Diehl, M and W. Stroebe. 1987. Productivity loss in brainstorming groups: Toward the solution of the riddle. Journal of Personality and Social Psychology 53 (3): 497-509. Einhorn, H., R. Hogarth, and E. Klempner. 1977. Quality of Group Judgment. Psychological Bulletin: 517-539. Frederick, D. 1991. Auditors Representation and Retrieval of Internal Control Knowledge. Accounting Review: 240-258. Gallupe, R. L. Bastianutti and W. Cooper. 1991. Unblocking brainstorms. Journal of Applied Psychology 76 (1): 137-142. Gigone, D. and R. Hastie. 1997. Proper Analysis of the Accuracy of Group Judgments. Psychological Bulletin 121(1): 146-167. 117

Glover, S. 1997. The Influence of Time Pressure and Accountability on Auditors Processing of Nondiagnostic Information. Journal of Accounting Research 35 (2): 213-226., D. Prawitt, J. Schultz, and M. Zimbelman. 2003. A Test of Changes in Auditor s Fraud Related Planning Judgments since the Issuance of SAS No. 82. Auditing: A Journal of Practice and Theory 22 (2): 237-251. Gramling, A. 1999. External Auditors' Reliance on Work Performed by Internal Auditors: Influence of Fee Pressure on this Reliance Decision. Auditing: A Journal of Practice and Theory 18 (Suppl.): 117-135. Hackenbrack, K. 1992. Implications of Seemingly Irrelevant Evidence in Audit Judgment. Journal of Accounting Research 30 (1): 126-136., and M. Nelson. 1996. Auditor's Incentives and Their Application of Financial Accounting Standards. Accounting Review 71 (1): 43-59. Hackman, J. 1987. The Design of Work Teams. In J.W. Lorsch (Ed.), Handbook of Organizational Psychology (p. 315-342). New York: Prentice Hall. Harding, N. and K. Trotman. 1999. Hierarchical Differences in Audit Workpaper Review Performance. Contemporary Accounting Research 16 (4): 671-684. Hegedus, D., and R. Rasmussen. 1986. Task Effectiveness and Interaction Process of Modified Nominal Group Technique in Solving an Evaluation Problem. Journal of Management 12 (4): 545-560. Henry, R. 1995. Improving Group Judgment Accuracy: Information Sharing and Determining the Best Member. Organizational Behavior and Human Decision Processes 62 (2): 190-197. Hill, G. W. 1982. Group Versus Individual Performance: Are N+1 Heads Better Than One? Psychological Bulletin 91 (3): 517-539. Hinsz, V., R. Tindale and D. Volrath. 1997. The Emerging Conceptualization of Groups as Information Processors. Psychological Bulletin 121 (1): 43-64. Hoffman, V. and J. Patton. 1997. Accountability, the Dilution Effect, and Conservatism in Auditors Fraud Judgments. Journal of Accounting Research (Autumn): 227-237. Hurtt, K. 2003. Development of an Instrument to Measure Professional Skepticism. Working Paper. University of Wisconsin-Madison, Madison,WI. 118

., M. Eining, and D. Plumlee. 2003. Professional Skepticism: A Model with Implications for Research, Practice and Education. Working Paper, University of Wisconsin-Madison, Madison, WI. Ismail, Z. and K. Trotman. 1999. The Impact of the Review Process on Hypothesis Generation. Accounting, Organizations and Society 20 (5): 345-358. Jamal, K., P. Johnson and R. Berryman. 1995. Detecting Framing Effects in Financial Statements. Contemporary Accounting Research (Fall): 85-105. Johnson, E. 1994. Auditor Memory for Audit Evidence: Effects of Group Assistance, Time Delay, and Memory Task. Auditing: A Journal of Practice and Theory 13 (1): 36-56. Johnson, V. and S. Kaplan. 1991. Experimental Evidence on the Effects of Accountability on Auditor Judgments. Auditing: A Journal of Practice and Theory 10 (Suppl.): 96-107. Kennedy, J. 1993. Debiasing Audit Judgment with Accountability: A Framework and Experimental Results. Journal of Accounting Research 31 (2): 231-245. King, R. 2002. An Experimental Investigation of Self-Serving Biases in an Auditing Trust Game: The Effect of Group Affiliation. Accounting Review 77 (2): 265-284. Knapp, C. and M. Knapp. 2001. The Effects of Experience and Explicit Fraud Risk Assessment in Detecting Fraud with Analytical Procedures. Accounting, Organizations and Society 26: 25-37. KPMG. 2003. Fraud Survey 2003. www.kpmg.com. Kroon, M., D. Kreveld and J. Rabbie. 1992. Group versus Individual Decision Making. Small Group Research. 23 (4): 427-458. Lerner, J. and P. Tetlock. 1999. Accounting for the Effects of Accountability. Psychological Bulletin 125 (2): 255-275. Libby, R. 1985. Availability and the Generation of Hypotheses in Analytical Review. Journal of Accounting Research 23 (2): 648-667.. 1995. The Role of Knowledge and Memory in Audit Judgment. Judgment and Decisions Research in Accounting and Auditing, edited by A.H. Ashton and R. H. Ashton. New York: Cambridge University Press. 119

, and J. Luft. 1993. Determinants of Judgment Performance in Accounting Settings: Ability, Knowledge, Motivation, and Environment. Accounting, Organizations and Society 18 (5): 425-450., and K. Trotman. 1993. The Review Process as a Control for Differential Recall of Evidence in Auditor Judgments. Accounting, Organizations and Society 18 (5): 559-575., K. Trotman, and I. Zimmer. 1987. Member Variation, Recognition of Expertise, and Group Performance. Journal of Applied Psychology 72 (1): 81-87. McGrath, J. and D. Kravitz. 1982. Group Research. Annual Review of Psychology: 195-230. McGrath, J. 1984. Groups: Interaction and Performance. Englewood Cliffs, NJ: Prentice Hall. Messier, W., and W. Quilliam. 1992. The Effect of Accountability on Judgment: Development of Hypotheses for Auditing. Auditing: A Journal of Practice and Theory (Suppl.): 123-145., and R. Tubbs. 1994. Recency Effects in Belief Revision: The Impact of Audit Experience and the Review Process. Auditing: A Journal of Practice and Theory 13 (Spring): 57-72.. 2003. Auditing and Assurance Services: A Systematic Approach. Burr Ridge, IL. McGraw-Hill Companies, Inc. Miner, F. 1984. Group versus Individual Decision Making: An Investigation of Performance Measures, Decision Strategies and Process Losses/Gains. Organizational Behavior and Human Performance 33: 112-124. Murthy, U. 2002. Group Support Systems Research in Accounting: A Theory-Based Framework and Directions for Future Research. Researching Accounting as an Information Systems Discipline: 135-155. American Accounting Association, Sarasota, FL. Myers, D. and H. Lamm. 1976. The Group Polarization Phenomenon. Psychological Bulletin 83: 602-627. Nieschwietz, R., J. Schultz, and M. Zimbelman. 2000. Empirical Research on External Auditors Detection of Financial Statement Fraud. Journal of Accounting Literature: 190-246. Nisbett, R., H. Zukier and R. Lumley, 1981. The Dilution Effect: Nondiagnostic Information Weakens the Implications of Diagnostic Information. Cognitive Psychology 13: 248-277. Nunamaker, J., A. Dennis, J. Valacich, D. Vogel, and J. George. 1991. Electronic Meeting Systems to Support Group Work. Communications of the ACM 34 (7): 40-61. 120

Osborn, A. 1957. Applied Imagination. (rev. ed. ) New York: Scribner Owhoso, V., W. Messier, and J. Lynch. 2002. Error Detection by Industry-Specialized Teams during Sequential Audit Review. Journal of Accounting Research 40 (3): 883-900. Peecher, M. 1996. The Influence of Auditors' Justification Processes on Their Decisions: A Cognitive Model and Experimental Evidence. Journal of Accounting Research 34 (1):125-140. Peecher, M., J. Rich and R. Tubbs. 2003. Determinants of Judged Bias, Noise and Justifiability in Audit Teams Judgments. Working paper, University of Illinois, Champaign, IL. Perkins, J. 2003. Informal Consultation in Public Accounting: A Strategic View and an Experimental Investigation. Working Paper. University of Illinois, Champaign, IL. Plous, S. 1993. The Psychology of Judgment and Decision Making. Burr Ridge, IL. McGraw- Hill Companies, Inc. Ponemon, L. 1995. The Objectivity of Accountants' Litigation Support Judgments. Accounting Review 70 (3): 467-488. Ramsay, R. 1994. Senior/Manager Differences in Workpaper Review. Journal of Accounting Research 32 (Spring): 127-135. Rau, S. and D. Moser. 1999. Does performing Other Audit Tasks Affect Going-Concern Judgments? Accounting Review 74 (4): 493-508. Reckers, P. and J. Schultz. 1982. Individual Versus Group Assisted Audit Evaluations. Auditing: A Journal of Practice and Theory 2 (Fall): 64-74., and J. Schultz. 1993. The Effects of Fraud Signals, Evidence Order, and Group Assisted Counsel on Independent Auditor Judgment. Behavioral Research in Accounting 5: 125-144. Reimers, J. and M. Fennema. 1999. The Audit Review Process and Sensitivity to Information Source Objectivity. Auditing: A Journal of Practice and Theory 18 (1): 117-123. Rich, J., I. Solomon and K. Trotman. 1997a. The Audit Review Process: A Characterization from the Persuasion Perspective. Accounting, Organizations and Society 22 (5): 481-505..1997b. Multi-Auditor Judgment/Decision Making Research: A Decade Later. Journal of Accounting Literature 16: 86-126. 121

Rudolph, H. and R. Welker. 1998. The Effects of Organizational Structure on Communication within Audit Teams. Auditing: A Journal of Practice and Theory 17 (2): 1-14. Schultz, J. and P. Reckers. 1981. The Impact of Group Processing on Selected Audit Disclosure Decisions. Journal of Accounting Research 19 (Autumn): 482-501. Sniezek, J. 1992. Groups under Uncertainty: An Examination of Confidence in Group Decision Making. Organizational Behavior and Human Decision Processes 52: 124-155. Solomon, I. 1982. Probability Assessment by Individual Auditors and Audit Teams: An Empirical Investigation. Journal of Accounting Research 20 (Autumn): 689-710.. 1987. Multi-Auditor Judgment/Decision Making Research. Journal of Accounting Literature 6: 1-25. Sprinkle, G. and R. Tubbs. 1998. The Effects of Audit Risk and Information Importance on Auditor Memory during Working Paper Review. Accounting Review 73 (4): 475-502., S. Peffer, and J.Fisher. 2003. Budget-Based Contracts, Budget Levels, and Group Productivity. Journal of Management Accounting Research 15: 51-74. Stasser, G., D. Stewart and G. Wittenbaum. 1995. Expert Roles and Information Exchange during Discussion: The Importance of Knowing Who Knows What. Journal of Experimental Social Psychology 31: 244-265. Stoner, J. 1968. Risky and Cautious Shifts in Group Decisions: The Influence of Widely Held Values. Journal of Experimental Social Psychology 27: 442-459. Sutton, S. and S. Hayne. 1997. Judgment and Decision Making, Part III: Group Processes. In V. Arnold, and S. Sutton, Behavioral Accounting Research Foundations and Frontiers (p. 134-163), Sarasota, FL: American Accounting Association. Tan, H. and A. Kao. 1999. Accountability Effects on Auditors Performance: The Influence of Knowledge, Problem-Solving Ability, and Task Complexity. Journal of Accounting Research 37 (1): 209-223., and K. Jamal. 2001. Do Auditors Objectively Evaluate Their Subordinates Work? Accounting Review 76 (1): 99-110. Tetlock, P. E. 1985. Accountability: A Social Check on the Fundamental Attribution Error. Social Psychology Quarterly 48 (3): 227-236. 122

Towry, K. 2003. Control in a Teamwork Environment- The Impact of Social Ties on the Effectiveness of Mutual Monitoring Contracts. Accounting Review 78 (4): 1069-1095. Trotman, K. 1985. The Review Process and the Accuracy of Auditor Judgments. Journal of Accounting Research 23 (Autumn): 740-752., and P. Yetton. 1985. The Effect of the Review Process on Auditor Judgments. Journal of Accounting Research 23 (Spring): 256-267., and I. Zimmer. 1983. Individual and Group Judgments of Internal Control Systems. Journal of Accounting Research 21 (Spring): 286-292. Turner, C. 2001. Accountability Demands and the Auditor s Evidence Search Strategy: The Influence of the Reviewer Preferences and the Nature of the Response (Belief vs. Action). Journal of Accounting Research 39 (3): 683-706. Watson, D. 1975. The Structure of Project Teams Facing Differentiated Environments: An Exploratory Study in Public Accounting. Accounting Review 2: 259-273. Wilks, T. 2002. Predecisional Distortion of Evidence as a Consequence of Real-Time Audit Review. Accounting Review 71 (1): 51-71. Wilks, T., M. Zimbelman. 2003. The Effects of a Fraud-Triangle Decomposition of Fraud Risk Assessments on Auditors Sensitivity to Incentive and Opportunity Cues, Working paper, Brigham Young University, Provo, UT. Yip-Ow, J. 2000. Effects of the Preparer s Justification on the Reviewer s Hypothesis Generation and Judgment in Analytical Procedures. Accounting, Organizations and Society 25: 203-215. Zimbelman, M. 1997. The Effects of SAS No. 82 on Auditors Attention to Fraud Risk Factors and Audit Planning Decisions. Journal of Accounting Research 35 (Suppl.): 75-104. 123

BIOGRAPHICAL SKETCH Tina Daly Carpenter Education Florida State University Doctoral Candidate, Ph.D. in Accounting, May 2004 Master of Accounting, May 1996 Bachelor of Science in Accounting, May 1995 Research and Teaching Interests Research: Judgment and Decision Making in Financial Statement Fraud Teaching: Financial Accounting, Auditing, Forensic Accounting Working Papers and Work in Progress Partner Influence, Team Brainstorming, and Fraud Risk Assessment: Some Implications of SAS No. 99, Dissertation in progress, proposal defended July 2003. Dissertation Committee: Jane Reimers (Chair), Neil Charness, Bud Fennema, Greg Gerard, and Bill Hillison. Unethical and Fraudulent Financial Reporting: Applying the Theory of Planned Behavior, Working Paper 2003 (co-authored with Jane Reimers, Rollins College). The Role of Experience in Professional Skepticism, Knowledge Acquisition, and Fraud Detection, Working Paper 2003 (co-authored with Cindy Durtschi, Utah State University and Lisa M. Gaynor, Georgetown University). Corporate Change: The Effect of Sarbanes-Oxley, Working Paper 2003 (co-authored with Bud Fennema and Bill Hillison, Florida State University, and Phil Fretwell, Protiviti). Differences in Attitudes Toward the Uncertainty of Fraud: External Auditors versus Forensic Auditors, Working Paper 2002. Presentations The Role of Experience in Professional Skepticism, Knowledge Acquisition, and Fraud Detection, American Accounting Association (AAA) Auditing Midyear Meeting, Huntington Beach, California, January 2003; Florida State University workshop, October 2002. Unethical and Fraudulent Financial Reporting: Applying the Theory of Planned Behavior, Florida State University workshop, October 2001. Differences in Attitudes Toward the Uncertainty of Fraud: External Auditors versus Forensic Auditors, poster presentation at Judgment and Decision Making (JDM) Annual Conference, Orlando, Florida, November 2000. 124

Research Grants Received Partner Influence, Team Brainstorming, and Fraud Risk Assessment: Some Implications of SAS No. 99, Florida State University Dissertation Research Grant 2003. Internal Auditors Assessment and Communication of Risk: Can Improvements Be Made? (co-authored with Jane Reimers, Rollins College and Phil Fretwell, Protivti); The Institute of Internal Auditors Research Foundation Research Grant 2003. Teaching and Professional Experience Florida State University; Tallahassee, Florida Instructor, Auditing Theory and Application I (Spring 2003, Fall 2002), Senior level, Accounting Major Students Instructor, Auditing Theory and Application I (Summer 2003, Summer 2002), Senior level, Accounting Major, Non-traditional students, Panama City Campus Instructor, Introduction to Financial Accounting (Spring 2002, Fall 2001, Summer 2001, Summer 2000), Sophomore and Junior level, Business Students Mentor, Online Distance Learning Program, Introduction to Financial Accounting (2001-2002), Office for Distributed and Distance Learning under the direction of Jane Reimers to Internal Revenue Service Students Research and Teaching Assistant (1999-2000) Arthur Andersen LLP; Senior Auditor, Orlando, Florida (1996-1999) Florida Licensed Certified Public Accountant Teaching Honors Florida State University Outstanding Teacher Award 2003 (University Award) Florida State University College of Business Outstanding Teacher Award 2002 (College Award) Academic Awards and Honors American Accounting Association & Deloitte Foundation Doctoral Consortium Fellow 2002 Florida Institute of Certified Public Accountants Educational Foundation Scholarship 1994, 1995 E.C. and Tillie Allen, Department of Accounting Scholarship 1995 E. Ray Solomon, College of Business Scholarship 1995 Deloitte and Touche Outstanding Senior Award 1995 Leadership and Service Awards and Honors Florida State University, Panhellenic Greek Advisor of the Year, 2003 Florida State University, Senior Hall of Fame (awarded to 5 outstanding seniors from class of 7,500) Junior Achievement, Community Achievement Award, 1999 National Kappa Delta Sorority: Chapter Advisor Award, 2003; Snyder and Cavin Graduate Scholarship 1996; Most Outstanding Senior Award, 1995; Corre Anding Stegall National Leadership Award, 1994; 125

Marion Day Mullins Scholarship for Service to Community, 1994 National Panhellenic Conference: Business Professional Women s Organization (BPW) Scholarship 1995; Katherine Warren Scholarship, 1994; Daytona Panhellenic Association Scholarship, 1994 Conference Participation Business Measurement and Assurance Services (BMAS), University of Texas at Austin, 2003, 2001 American Accounting Association Auditing Meeting and Doctoral Consortium, 2004, 2003, 2002, 2001 American Accounting Association Annual Meeting, Honolulu, Hawaii, 2003 American Accounting Association Annual Meeting, Atlanta, Georgia, 2001 Judgment and Decision Making Annual Conference, Orlando, Florida, 2000 Journal of Accounting Literature Conference, University of Florida, 2000 Honor Societies and Leadership Honoraries Omicron Delta Kappa National Honor Society Mortar Board Senior Leadership and Scholarship Honorary Beta Alpha Psi Accounting Honor Society Order of Omega National Greek Leadership Honorary Gold Key Leadership Honorary Golden Key National Honor Society Phi Kappa Phi Honor Society Beta Gamma Sigma National Business Honor Society University and Community Service Florida State University Steering Committee, 2000-2002 Kappa Delta Sorority: Advisor, 2000- present; Vice President, 1995 Burning Spear Leadership Honor Organization: Alumni Advisor, 1999, 2000; Alumni President, 1998; President, 1996; Treasurer, 1995 Junior Achievement Teaching Volunteer, 1998 Omicron Delta Kappa Secretary, 1994 Gold Key Vice President, 1994 Student Alumni Association Treasurer, 1994 Academic and Professional Memberships American Accounting Association (Auditing, and Accounting, Behavior & Organizations Sections) American Institute of Certified Public Accountants Society for Judgment and Decision Making 126

References Jane Reimers Chairman, Dissertation Committee jreimers@rollins.edu, (407) 646-2499 Bud Fennema Ernst and Young Professor of Accounting Chairman, Accounting Department, Dissertation Committee bfennema@cob.fsu.edu, (850) 644-8231 Al Bathke Ph.D. Advisor abathke@cob.fsu.edu, (850) 644-7888 Rollins College Crummer Graduate School of Business 1000 Holt Avenue- 2722 Winter Park, FL 32789-4499 Florida State University Department of Accounting Rovetta Business Building Tallahassee, FL 32306-1110 Florida State University Department of Accounting Rovetta Business Building Tallahassee, FL 32306-1110 127