A review of the literature in audit litigation

A review of the literature in audit litigation

Latham, Claire Kamm

1.0 INTRODUCTION

The auditor plays an important role in the economy by acting as the public’s “watchdog.” The auditor attests to the assertions made by another, thereby lending credibility to those assertions and increasing the reliability of information used in economic decision making. Participants in the economy evaluate each new disclosure or revelation and use this information to take actions consistent with their self-interests.1 Because the actions of economic agents are influenced by information, reliable information is critical to maintain order within the economy and to achieve economic efficiency by encouraging appropriate resource allocation.

Unfortunately, since wealth transfers may be obtained through the use of inaccurate information, there is a potential for abuse by the release of faulty information. Society relies on several mechanisms to prevent this kind of abuse. One such mechanism is the auditor who is expected to detect and report material inaccuracies in the release of financial information, and whose presence may even prevent the inappropriate actions from occurring in the first place.

Because of the important role of the auditor in the release of reliable financial information, the public needs assurance that the auditor is performing appropriately. Standards have been created to ensure this. In addition, the legal system can be invoked to impose civil liability on an auditor who causes harm through a failure to perform his or her duty.2 Several statutory and common law remedies are available to those allegedly harmed by the auditor’s actions. For example, to help prevent the release of inaccurate or incomplete information associated with securities, the Securities Act of 1933 and the Securities Exchange Act of 1934 both require the use of an independent auditor and provide for civil and criminal penalties to those involved with the release of inaccurate information.

Currently, a debate is raging in the U.S. about the role of the civil liability system in settling economic disputes. Calls for tort reform have been issued from several sources and are based primarily on purported “abusive” lawsuits which have been filed despite a lack of merit in the plaintiff’s claims. Litigation is alleged to impose a significant cost on the accounting profession and, consequently, the profession has been actively seeking reform in audit litigation since the late 19,70s.3,4

Research in audit litigation can provide valuable insight into the on-going debate by identifying the magnitude of the problem (and whether a problem even exists), areas where the civil liability system fails to meet societal objectives, and the effects of past reform efforts. In his call for more research in audit litigation, Kinney [1994] notes that by performing research using a positive framework, academic researchers may provide valuable information to the profession as it seeks, in a normative framework, reforms to the legal system.

The primary objective of our paper is to examine the current state of audit litigation research. Although a large number of studies have been conducted in this area, no overall synthesis of the extant literature has been performed.5 As discussed in the next section, the societal functions of a civil liability system comprise the framework for analysis. Audit litigation serves certain purposes in society, and this framework permits us to evaluate research by determining what is known about the satisfaction of these societal functions and to identify additional lines of inquiry in order to eliminate gaps in our knowledge. The myriad of research methodologies used to explore this research area is also presented, along with suggestions for refinements in future research. The key finding of our analysis is that the complexity of the existing audit litigation environment and its continuing evolution provide many future research opportunities.

The scope of this paper is limited to domestic audit litigation. We examine the literature on any type of audit litigation the CPA may face, but not litigation associated with other services. Studies on litigation outside the U.S. are not examined, owing to significant differences in institutional details. Both of these omitted areas possess future research potential, but are beyond the scope of this project. Literature on litigation against parties other than the auditor, such as officers and directors, is included when it can help to understand audit litigation.

The paper is organized as follows. The next section details the framework which is based on the functions of a legal system. This framework allows analysis of pertinent research issues. Research issues are grouped into three areas and each area is examined in terms of what has been learned, what still needs to be learned, and data issues that may exist. Section three examines the literature that views litigation as the dependent variable (the existence of litigation is a function of other factors). Section four explores the literature that views litigation more as an independent variable that causes changes in other factors. Section five discusses the literature on the legal system itself, including other means of dispute resolution. A table is provided for each of these three sections, summarizing, as applicable, samples) used, variables) examined, methodology employed, and key findings. Finally, section six provides concluding remarks.

2.0 FRAMEWORK

2.1 Purposes for Civil Liability

Vago [1988] poses and answers this question: “More specifically, what functions does law perform? As with the definition of law, there is no agreement among scholars of law and society on the precise functions, nor is there consensus on their relative weight and importance. A variety of functions are highlighted in the literature, depending on the conditions under which the law operates at a particular time and place” [p.11]. Vago goes on to note that social control, dispute settlement, and social engineering are commonly stated functions. Providing a different emphasis and classification, Simpson [1988] identifies five functions of a legal system that we have modified to better fit the civil liability system for audit litigation. These five functions are (Simpson’s original description is given in parentheses, if modified): resolution of conflict; communication of expectations (reconciliation of stability and change); damage recovery (distribution of wealth and entitlements); deterrence (regulation of behavior); and restraint (distribution of power). Figure 1 diagrams the relationship between the functions that are described in more detail in the following sections.

2.1.1 Resolution of Conflict

Legal systems exist to provide society with an orderly means to resolve disputes. Society is defined as a body of individuals living as members of a community. Yet, as people interact within this community, conflicts are inevitable. Society requires a way to efficiently and effectively settle these disputes without major dysfunctional effects to the community; otherwise such conflicts could pose a viable threat to the order of society.

The possibility for conflict in commerce is high, since commerce is composed of numerous economic agents seeking their own best interests in the pursuit of scarce resources. Because the audit plays an important role in commerce, disputes may arise about deficiencies in the audit. Society has a developed a legal system to address, and hopefully resolve, such disputes with a minimum of disruption.6

Underlying this objective is the concept of equity (justice). Participants in a dispute generally have the right to argue their side of the dispute and have it ruled upon by an impartial court. Equity dictates that rights are to be protected and society has given the courts authority to do so. In this way, society obtains a means of resolving disputes in a manner perceived as equitable.

2.1.2 Communications of Expectations

If conflicts are to be minimized, appropriate conduct must be defined. Simpson [1988] describes one function of the law as the reconciliation of stability and change. The law serves as an agent of stability by establishing the current “rules” of interaction. Yet at the same time, the law can be a mechanism of change by modifying rules as necessary.

In the context of audit litigation, the legal system serves as an agent of stability by establishing expectations about the roles and responsibilities of auditors. Stability is established in statutory law by the codification of legislative intent, and in common law by the legal concept of precedent. Unless there is a compelling reason to change an existing principle, precedent means that once a court has ruled on a controversy and issued an opinion, that ruling principle will be followed by the court in subsequent cases [Corley and Robert, 1975]. This stability allows the auditor to generally understand how various legal provisions will be applied in disputes regarding his or her services.

Stability does not guarantee that the law will be the same in all jurisdictions, since each jurisdiction may have a different court and legislative history. Even though the same societal objectives exist whether the civil action is brought against the auditor under the federal securities laws, or under state securities laws, or as a tort action under common law, the legal features of these three systems can differ markedly. The reader is referred to Miller and Young [1997] for detailed descriptions of current legal provisions important to audit litigation, and to Siliciano [1997] for a discussion of the unique legal nature of the independent audit.

Although stability is desirable, the legal system must be a dynamic system allowing evolution of the law to meet changing societal needs. The courts strive to obtain an appropriate balance between the competing interests of litigants; what once may have been appropriate may not be appropriate now. Statutory law may be overturned by courts on constitutional grounds or by the legislature itself through the passage of a new statute. Common law may be changed by a court itself or a reviewing court if a previous court decision is found to be “palpably wrong” [Corley and Robert, 1975]. A legislature can also overturn common law by statute.7 Attempts to change the audit litigation system are likely to continue into the future, since large wealth transfers may accompany a court decision and because the court system is adversarial. Each side to the dispute has an economic incentive to present innovative arguments to capture or defend wealth.8

Change poses both a challenge and an opportunity to the researcher. Although the researcher must obtain a high level of institutional knowledge, several reference sources are available.9 The researcher must also be aware of how changes to the law may affect a study. Fortunately, several articles exist that provide both an analysis of the evolution of legal statutes and the necessary background to understand important institutional details.10 This evolution presents an opportunity to the researcher, for as the legal environment changes, new questions arise and knowledge may become outdated.

2.1.3 Damage Recovery

The courts have the authority to distribute wealth by compelling parties to pay damages [Simpson, 1988).11 In the American system of jurisprudence, a party that breaches a duty owed to another and, as a result, causes injury, is responsible to compensate the injured party. Theoretically, in such a fashion, parties injured by the wrongful actions of another can be made whole once again. Because the payments are being made by those who are adjudicated as having caused the injury due to a failure to perform a required duty in an appropriate manner, arguably, fairness is achieved. Injured parties are made whole at the expense of those who wrongfully caused the injury, rather than forcing the victims to bear the cost of the injury or having society bear the cost.12

This concept of equity is based on the assumption that each member of society has certain rights that must be protected [Corley and Robert, 1975]. In the area of audit litigation, the allegedly injured party has a right to receive professional services consistent with the standards of the profession. If the auditor unintentionally fails to provide services consistent with those that would be provided by a reasonably prudent auditor in the same situation, the auditor has acted negligently. If another party is injured by reliance on the negligently performed audit, the concept of equity (justice) requires the auditor be held responsible to compensate for those injuries.

As discussed in the next section, auditor liability generally involves failure by the auditor to detect a material misstatement or omission created by management. If the material misstatement or omission causes damage to another, the auditor and management may be held jointly liable. Although management’s involvement in the creation of damages may, in some cases, reduce the amount of damages that the auditor must individually pay, the responsibility of management in the creation of damages, by itself, does not eliminate the responsibility of the auditor with respect to third parties.

2.1.4 Deterrence

The legal system has the ability to influence behavior [Simpson, 1988]. Whereas damage recovery and restraint are reactive, since the initial damage has already occurred, the deterrence objective is proactive, seeking to prevent injury from occurring. Deterrence is achieved if the threat of damage payments or other negative consequences is sufficient to prevent the breach of a duty. Thus, society may obtain the appropriate actions from the parties by making inappropriate actions costly (in other words, provide an inducement to be better auditors). For this to occur, lawsuits must be associated with improper actions by auditors. The reader is directed to Palmrose [1997] for an expanded discussion on the role of merits in audit litigation.

In the context of audit litigation, the notion of deterrence is generally applied to the auditor. However, deterrence should also consider desirable actions by the users to avoid damages. Optimal deterrence is obtained when the sum of the cost of care (auditors’ actions and users’ actions to limit their risk or mitigate their damages) and the expected costs of accidents (costs to users) are minimized [Shavell, 1987].13 With this definition, inadequate or excessive deterrence is possible. Inadequate deterrence occurs when the auditor performs substandard audits (low cost of care) with little fear of the consequences. In such a case, the auditor incurs low costs but the users are likely to incur more costs (both in their own care and in the cost of resulting accidents). Excess deterrence imposes high costs of care on the auditor without offsetting benefits in the form of reduced costs of care by the users or expected costs of accidents.

Deterrence may also be accomplished through client screening. An auditor attests to the assertions made by management in their financial statements. Most commonly, for litigation to arise, the auditor must have provided an unqualified audit opinion to a set of financial statements that are subsequently discovered to have contained material misstatements or material omissions. Absent the initial management-generated material misstatement or omission, the auditor’s legal liability is decreased substantially. Thus, if auditors are concerned about litigation, they should be selective in their acceptance of audit clients to avoid clients who may have material misstatements or omissions in their unaudited financial statements.

Identifiable client characteristics associated with litigation provide the auditor with criteria for a screening process. If a company with such characteristics cannot find an auditor, the company may be pressured to change this characteristic (if it is under its control). To the degree that these characteristics pose the potential for damage to the investing public, their elimination is a socially desirable outcome.

Finally, deterrence can also be accomplished by allowing auditors to correctly price audit services. Client characteristics associated with litigation indicate risk factors that should be incorporated into the auditor’s cost function. As expected litigation costs rise, the auditor must increase fees to maintain a normal profit. This added cost is borne by the client who has incentive to eliminate this cause of additional risk. In such a fashion, socially desirable outcomes are achieved. Either the client eliminates the problem or, if it is allowed to remain, the cost is appropriately borne by the responsible party.

2.1.5 Restraint

The legal system has the ability to compel participants to its will in what Simpson [1988] refers to as the distribution of power. By the authority placed in them, the courts can force a party to act against its own interests (e.g., pay monetary damages to another). The courts or other authoritative bodies may also use this authority in a reactive manner to restrain a party that has acted in a particularly egregious fashion. As a result of the findings in a civil lawsuit, an auditor may be subject to sanctions or other restraints imposed by the Securities and Exchange Commission, state Boards of Accountancy, or other regulatory bodies, and/or face criminal prosecution. These sanctions can include suspension or termination of the professional license or a prohibition against practicing in front of the regulatory body. Such sanctions are imposed because of unacceptable practices.14

2.2 Application of Objectives to Issues

The societal functions for a civil liability system described above comprise the framework which is used to analyze the contributions of the literature. Once the purposes for the civil liability system are understood, research can be more easily interpreted and organized. This research serves as evidence in the on-going debate about the role of the civil liability system for audit litigation.

As noted in the previous sections, the societal functions of the civil liability system overlap. For this reason, and also because research studies seldom identify their research topics in terms of societal functions, the research is not classified using these functions. Instead, the literature is grouped into the three categories discussed below. The literature within each category is then discussed in terms of what can learned about the societal functions.

The first area of research comprises the litigation-risk environment. These studies generally present the existence of litigation as the dependent variable and attempt to understand the factors that lead to litigation. The second area of research views litigation more in the role of an independent variable. Consistent with the deterrence aspect of civil liability, the influence of litigation on audit quality is frequently explored. The third general area of research involves papers that analyze the mechanism of dispute resolution as a means of providing insight into the institutional processes affecting audit litigation. Included among these studies are those exploring the settlement process, the adjudication process, and strategies of the plaintiff and defendant. Throughout these categories, we attempt discrete classifications with a minimum of overlap. However, a few studies address several issues and these papers are discussed in more than one section.

3.0 LITIGATION AS THE DEPENDENT VARIABLE

To some extent, litigation risk exists in all audit situations. Yet, research shows that this risk is not constant across all audits or audit firms. The studies in this section examine the characteristics of clients and auditors who are involved in audit litigation.15 These factors are important to researchers to develop and test theories that explain the existence of litigation risk, and should also provide important insight into several of the societal objectives detailed previously.

These studies identify samples by the occurrence of litigation. Several of the studies have a sample comprised entirely of litigation companies, while some use a matched-pairs approach to compare litigation companies to non-litigation companies. Since many of the studies use the same sources to identify litigation, the samples often overlap. Table 1 provides a summary of the studies conducted in this area, with particular emphasis on which variables have been examined and the type of cases used in the analysis. As noted in the following discussion, several results of the studies are sensitive to experimental design (most commonly, the type of econometric analysis) and some results are not replicated in other studies.

3.1 Client Characteristics Associated with Audit Litigation

Several studies examine the client characteristics associated with audit litigation. For example, two characteristics, industry affiliation and publicly-traded status, are consistently associated with the existence of audit litigation.16 St. Pierre and Anderson [1984] and Palmrose [1988] find that public companies and companies in finance, insurance, real estate, and manufacturing are those most commonly involved in audit litigation. Stice [1991] likewise finds that audit litigation most commonly involves companies in manufacturing.17 Audits in high technology industries are seen as being associated with higher rates of litigation [Palmrose, 1988; Francis et al., 1994a, 1994b, 1998].18

As discussed in Sections 2.1.3 and 2.1.4, an audit failure can occur only if the company first issues financial statements with material misstatements or omissions, and then the auditor fails to detect the misstatements or omissions. Thus, this examination of the characteristics of the client whose audit results in litigation can provide evidence about the factors coincidental with or causing the financial reporting difficulties. Because most audit work is unobservable, some situational indicator must alert interested parties of a potential audit failure. Consistent with the model of St. Pierre [1983], the signal initiates a further investigation. Based on the results of the investigation, which may be part of a legal discovery process, the determination of specific wrongdoing can be refined and a decision about whether or not to proceed with the lawsuit can be made. These client characteristics may serve as the signal of an underlying problem not detected in the audit process.

Several studies have examined financial statement variables to determine their association with audit litigation. Stice [1991] controls for industry effect and general economic conditions by using a matched-pairs design to investigate the association between pre-audit engagement characteristics of the client and subsequent filings of a lawsuit against the auditor. Because certain assets require more subjective judgments in establishing their valuation, Stice considers asset structure. He finds that the ratios of accounts receivable to total assets and inventory to total assets, proxies for asset structure, are significantly higher for clients associated with litigation. Lys and Watts [1994] examine 153 lawsuits against auditors using a matched-pairs design and find that total accruals and total assets are associated with litigation.19,20 Francis et al. [1994a; 1994b; 1998] and Carcello and Palmrose [1994) also find a significant association between size and the probability of litigation. Because the market value of stock is likely a proxy for size, the finding of Stice [1991], that clients with higher market value tend to have a higher incidence of litigation, is consistent with the other findings.21

Actual client bankruptcy occurrence is hypothesized to affect the probability of audit litigation, and several studies explore this issue. St. Pierre and Anderson [1984) find that bankruptcy or a significant client loss occurred in almost half of the lawsuits. They assert that a “surprise” element to the bankruptcy or client loss was present in many cases. Palmrose [1987] also finds an association between economic conditions and auditor litigation. As the general business climate deteriorates, she finds that the rate of auditor litigation increases, with close to half of her sample involving business failure or clients with severe financial difficulties. Palmrose also notes that many of the cases allege management fraud, and it is in these cases that payments by auditors are more common. Making use of additional information on lawsuit settlements, she finds that cases involving only bankruptcy are generally settled without auditor payment. Lys and Watts [1994], Stice [1991] and Carcello and Palmrose [1994] also find a significant association between the probability of bankruptcy and litigation.

Carcello and Palmrose [1994] extend this issue of financial distress by examining the role of a qualified audit opinion. They investigate public companies that declared bankruptcy between 1972 and 1992. They use auditor litigation as the dependent variable and the presence of a modified report and persistence of modified reporting as test variables. Based on several analyses, the authors conclude that the low litigation rate for modified audit opinions suggests that a modified audit opinion reduces, but does not eliminate, the likelihood of litigation.22 In examining the resolution of the lawsuits, the highest dismissal rate and the lowest payment rate were achieved when modified reports were issued for several years prior to the bankruptcy, rather than in just the year prior. In an industry-specific study of the same issue, Blacconiere and DeFond [1997] examine the failure of 24 publicly-traded savings and loans. They find a going concern report in the year prior to failure does not prevent audit litigation.

Besides the probability of bankruptcy and actual bankruptcy occurrences, several studies have shown the association between other financial distress measures and litigation. Francis et al. [1994b] find that lawsuits are associated with higher payment of dividends and higher systematic risk.23 Kellogg [1984] finds the most commonly litigated issue was a decline in the realized value of assets that can be interpreted as an indication of financial distress. Summers and Sweeney [1998] find that a combination of financial distress measures and insider trading activity may differentiate companies that issue fraudulent financial statements from those that do not.24

Several studies support the existence of fraudulent financial reporting as a significant factor in auditor litigation [Carcello and Palmrose, 1994; Palmrose, 1987; St. Pierre and Anderson, 1984]. Bonner et al. [1998] extend the issue of fraud by examining whether the type of fraud matters in the occurrence of litigation against auditors. The authors investigate 261 companies that were the subject of SEC enforcement actions between 1982 and 1995, and further divide their sample into those companies who have no disclosure or reporting litigation, those with auditor litigation, and those with litigation that does not involve an auditor. They find support for a higher incidence of auditor litigation when a company’s financial statements contain a fraud that involves fictitious transactions and events or a fraud that is designated as more common. Based on these findings, the authors hypothesize that fraud type affects juries’ and judges’ perceptions about auditors’ responsibility for detecting fraud. In a related fashion, Lys and Watts [1994] suggest that managers whose companies are threatened with acquisition have higher incentive to manipulate financial results in an attempt to fight off acquisition or to obtain a higher purchase price. Consistent with this analysis, they find an association between the probability of acquisition and litigation.

The results concerning stock performance as a predictor of litigation are inconsistent. Kellogg [1984] examines class action lawsuits by stockholders against companies where an accounting issue is disputed and finds that the lawsuits were initiated after a stock price decline.25 Stice [1991] finds that clients whose audits resulted in litigation have a higher variance of returns than clients whose audit did not result in litigation. Francis et al. [1994b] do not support the finding of returns volatility. Lys and Watts [1994] and Francis et al. [1994a] do not find a significant association between stock returns and litigation.26 Using a comparison of the disclosure profiles and stock return patterns of firms sued over adverse earnings, Francis et al. [1994a] generally find a lack of support for the plaintiff allegation that defective disclosures lead to inflated share prices.

Francis et al. [1994a] also find that companies who make more frequent disclosures appear to have greater litigation risk. They explore two explanations for this increased litigation risk [Francis et al. 1998]. First, consistent with the model of St. Pierre [1983], “surprise” adverse announcements are hypothesized to cause the initiation of an error search that could ultimately result in a lawsuit. However, inconsistent with the St. Pierre model, the authors find that companies with unpredictable sales have a higher incidence of litigation. Second, they consider the operating environment of the company. Companies with high operating leverage, a proxy for an inflexible cost structure, have higher litigation rates.

3.2 Auditor Characteristics Associated with Audit Litigation

Palmrose [1988] examines litigation activity among audit firms as one means of assessing auditor quality, using the assumption that a higher/lower quality auditor is involved in less/more audit litigation. Using 472 cases of litigation involving alleged audit failure and a frequency rate scaled by either the number of clients or U.S. auditing fees, she finds that Big 8 auditors are significantly less likely to be defendants in litigation than the eight largest non-Big 8 firms.27 Based on the assumption that high quality audit work results in less litigation, her results are consistent with the Big 8 firms being the high-quality audit suppliers. Because the sample is limited to litigation against the 16 largest CPA firms in the U.S., the ability to pay damages is controlled.

Lys and Watts [1994] consider auditor characteristics hypothesized to be associated with audit litigation: existence of a qualified opinion; audit structure; industry specialization; variance of client size; auditor size; audit tenure; and the client’s proportion of the audit firm’s total revenues (a proxy for independence). In the cross-sectional multivariate tests, they find that only the proportion of total audit fees represented by the audit fees of the client is associated with litigation. Stice [1991] did not find this variable to be significant once the effects of time period and industry are controlled.28

3.3 Conclusions Concerning Litigation as a Dependent Variable

Although this research is still relatively new, some consistent results are beginning to emerge. The studies suggest that several client characteristics are associated with lawsuits: size; public status; financial distress (actual bankruptcy or a high probability of bankruptcy); fraudulent financial reporting, and membership in certain industries (financial services and high technology). The association between bankruptcy and litigation is affected by two factors. The existence of fraud along with the bankruptcy makes litigation more likely, whereas the existence of a timely modified opinion may decrease the probability of litigation. Additionally, information environments that can lead to surprise announcements or increase the frequency of announcements are associated with litigation.

Identification of auditor characteristics associated with litigation is less clear. The only consistent finding suggests that non-Big 5 status is associated with a higher incidence of litigation.

The results of the studies provide evidence about the deterrence aspect of the civil liability system. The auditor characteristic associated with litigation, Big 5 membership, suggests some deterrence effect. If Big 5 membership is correlated with audit quality, the higher incidence of litigation for non-Big 5 auditors (the assumed lower quality auditors) suggests that auditors can avoid litigation by doing higher quality work, a socially desirable outcome. The association of specific client characteristics with litigation that allows for client screening or price adjustments for additional risk to occur further enhances deterrence. (This topic is discussed in more detail in Section 4.1.)

A civil liability system with deterrence as an objective must also be concerned about unintended consequences. In results that merit more examination, Francis et al. [1994a] find an association between litigation and voluntary information release.29 Clearly, society should be encouraging information release to allow for the efficient allocation of resources, but if this release leads to increased litigation, a perverse incentive is in place. Recognizing that this disincentive may exist, the Private Securities Litigation Reform Act of 1995 creates a safe harbor for the release of certain information.

These papers also provide some insight into the damage recovery objective. The association of financial difficulties with litigation suggests that, consistent with societal objectives, litigation is being used to seek recovery of damages created by the financial difficulties. The inconsistent results related to negative stock performance may suggest that the civil liability system is not being used regularly to make auditors provide “insurance” for investments. The association between litigation and fraudulent financial reporting also has a damage recovery element. Following the arguments of the fraud on the market theory, because securities are priced in part on reported financial results, fraudulently reported results should have lead to an inflated price. Once the proper information is impounded in the securities price, the price should return to a lower (not manipulated) price. The difference between the two prices represents recoverable damages.

Finally, these papers provide some insight into the type of conflicts resolved and how societal expectations have or have not altered over time. If auditors are successfully sued in litigation which has little to do with an accounting or auditing issue, society, through the courts, is imposing an additional responsibility on the auditors. There is some evidence that suggests auditors are not being successfully sued in great numbers for non-accounting or non-auditing issues [Bonner et al., 1998; Palmrose, 1987. St. Pierre and Anderson, 1984]. Taken together, these studies indicate that courts to date seem to recognize that auditors should be held accountable for financial reporting failures.

As might be anticipated, in many of these studies the research relies on actual court cases and is often drawn from a variety of sources (such as Mead Data’s LEXIS/NEXIS and NAARS, Federal Securities Law Reporter, Securities Class Action Alerts, The Wall Street Journal, the Public Accounting Report, the International Accounting Bulletin, and cases identified in prior litigation research). Some of the research in this area has been hampered by small sample sizes. To assist in future research efforts, the Accountants’ Coalition and the American Accounting Association recently sponsored a monograph by Palmrose [forthcoming]. The database included in this monograph contains audit litigation brought against the Big 5 firms from 1960 to 1995 (over 1,000 instances). Additional information on research design and data-gathering issues is also included.

The main methodological concern, which has only been recently discussed in the literature, and which can be addressed with the new database, involves the lack of homogeneity of the sample lawsuits. In the majority of the studies, analyses on combined heterogeneous samples of lawsuits are performed without regard to legal bases. In other words, cases brought under federal securities law are combined with those brought under common law [Francis, 1994]. Legal features differ between common law and statutory law and, depending upon the research question, these differences may be influential.

In her discussion of Lys and Watts [1994], Francis [1994] notes that if researchers wish to learn of the auditor’s role in securities litigation, they should continue to compare lawsuits that name the auditor as a defendant with those that do not. Bonner et al. [19981 is an example of this type of research. Since more is known about client characteristics associated with litigation than auditor characteristics, more research of the latter type is needed.

Finally, the relationship between independence and litigation needs to be more fully explored. The ratio of client fees to total firm audit fees is a crude proxy for independence and this crudeness may account for the inconsistent results noted above. Though a significant challenge to researchers, more carefully crafted proxies may allow for a more refined examination of the role of independence in litigation. Because independence is the cornerstone of auditing, a diminishment of independence decreases the audit’s credibility and increases the auditor’s litigation risk. The role that the new Independence Standards Board plays in the perception of independence and its effect on litigation needs to be examined.30

4.0 LITIGATION AS AN INDEPENDENT VARIABLE

The studies in this section view litigation more in the role of an independent variable and explore its effects on other variables. Much of the research conducted here focuses on the auditor; however, litigation’s effect can be analyzed more narrowly, such as studying the influence of various liability regimes on individual behaviors (e.g., including all lawsuit participants or potential lawsuit participants), or more broadly, with examinations of firm behavior, to arguably the broadest perspective, studying its impact on the allocation of economic resources. As noted previously, on a normative level, the properly operating civil liability system, communicates societal expectations and is structured to encourage desired effects. Research in this area can provide insight into existing and potential problems in the various legal regimes, including unintended consequences stemming from opportunistic behavior. The studies discussed in this section are presented in Table 2 which summarizes the approaches used and key findings of this body of research.

For litigation to have a deterring influence (or, conversely, for it to motivate socially desirable behavior), the presence of litigation should affect the actions and performance of audit participants and reinforce the role of the audit in society. For example, auditors faced with increasing litigation risk may mitigate this risk by screening clients, charging higher fees for risk if competitively feasible, and/or auditing better.31

4.1 Evidence Concerning Client Screening and Risk Premium

Simunic and Stein [1990) examine the effects of various audit firm portfolio characteristics on an audit firm’s bid and the possible impact of a portfolio view of audit risk upon the structure of the auditing industry. They note that audits are a form of investment and that auditors are investors. The choice of which companies to accept as audit clients is similar to the choice of investments to add to an investment portfolio. The auditors’ portfolio is constructed indirectly through a process of bidding (deciding on a minimum acceptable fee) against competitors. The mathematical analysis supports the hypothesis that this portfolio view of audit risk is appropriate. This conclusion, i.e., clients more at risk for litigation impact the auditor’s portfolio decision, is consistent with the papers previously discussed in Section 3.

Building on this portfolio view of audit risk, Pratt and Stice 1994 examine the variables used in the screening process to assess client litigation risk and the impact of risk assessment on the audit plan and fees. They conduct a prospective client case field experiment with 243 audit partners and managers of four Big 6 firms. Their results indicate that the client’s overall financial condition is the primary consideration in the litigation risk assessment and audit plan recommendations.32 This finding is consistent with the findings of Section 3.1 where the client’s overall financial condition is associated with litigation, and is also consistent with the predictive model of Stice [1991]. Interestingly, the client’s stock price is generally ignored by partners and managers or is perceived as being negatively related to litigation risk, which is consistent with the mixed findings related to stock price as a determinant of litigation.

The structure associated with the client screening process is also a significant component in the managed portfolio. Huss and Jacobs [1991] focus on this aspect in their field experiment involving Big 6 firms in Atlanta. They find wide variation in client acceptance and retention policies and determine that there are significant differences in the amount of autonomy granted local offices to accept or reject clients. Ayers and Kaplan [1998] also identify differences in client acceptance input and output across the two types of partners (engagement and risk review) who participate in the client acceptance decision process. Risk review partners make more conservative assessments of client’s financial condition and perceive the consequences of accepting riskier clients as more severe. These tendencies appear to lead them to make more conservative client acceptance decisions.

Krishnan and Krishnan [1997] note that auditors can adjust their client portfolios by becoming more selective in the acceptance of new audit engagements or by resigning from engagements in which there is a perceived high potential for litigation.33 Using two approaches to capture litigation risk, they test the hypothesis that litigation risk motivates auditor resignations. First, using factors identified as influencing auditor litigation in multivariate models of three prior studies [Carcello and Palmrose, 1994; Lys and Watts, 1994; Stice, 1991], they compare resignation companies with client companies that dismissed their auditor. The authors find resignations more often associated with engagements that involve financial distress, high variability of stock returns, low auditor independence, long auditor tenure, and the receipt of a modified opinion.34 Second, they construct a litigation proxy based on the prediction model of Stice [1991] and find that it is positively associated with the probability that auditor will resign rather than be dismissed from the engagement. The authors suggest that the current litigious environment appears to be reducing important audit services for clients perceived to be litigation risks.

The field study of Pratt and Stice [1994] finds that an audit fee premium is added to cover perceived litigation risk. Simunic and Stein [1996] provide an extensive review of the literature on audit pricing and risk premium, as well as additional analyses concerning the audit hours, audit fees, and litigation risk measures associated with the audits of one Big 6 firm in one year. The authors’ review also provides support for audit fees reflecting variations in litigation risk; however, their extension suggests that while fees are positively associated with litigation risk, the upward adjustment in fees reflects higher levels of auditor efforts as opposed to a risk premium.

4.2 Evidence Concerning Impact on the Audit

A significant group of analytical studies examine whether changes in the legal environment can motivate an auditor to produce a higher quality audit. The civil liability environment surrounding an audit failure is comprised of differing contractual settings (i.e., those to whom a duty of performance is owed), liability sharing provisions, and assessments of various parties’ negligence and/or malfeasance. This research investigates the impact of these factors on the auditor’s actions.

Several papers extend Simon [1981; 1982] to an auditing environment by modeling the effects of alternative liability regimes, litigation incentives, or privileges on audit quality. For example, DeJong [1985] determines that a higher average audit quality is provided in the marketplace under contingent legal fees and class-action privileges. Consistent with Simon [1982], Balachandran and Nagarajan [1987] suggest that a system incorporating negligence with supplemental insurance is preferable to a strict liability regime. The authors find that the quality of the signal from the internal audit tests, combined with the auditor’s perception of the probability of his client’s financial state being good, determine the level of liability required to induce the auditor to adopt a socially optimal due care level. Implementing strict liability may require more information than negligence thus; if information is costly, strict liability may be less desirable.

Narayanan [1994] models auditor litigation brought under federal securities law only, an important refinement, and then examines audit quality under joint and several and under proportionate liability regimes. His model also incorporates comparative negligence theory in the liability regimes and, interestingly, shows a difference between the regimes only when one defendant is insolvent. Contrary to recent expert testimony presented to the U.S. Senate during deliberations on the Private Securities Litigation Reform Act of 1995, which links the level of damages imposed on the auditor to audit quality, the results suggest that there is a potential increase in audit quality under proportionate liability, at least in this setting.

Under the proportionate liability regime, the auditor pays only for the damages created by his or her actions; thus, the auditor’s litigation cost is more sensitive to his or her effort, i.e., the auditor has a greater incentive to minimize litigation cost by working harder. In contrast, under joint and several liability when the auditee is insol vent, the auditor must pay all of the damages, regardless of the effort expended. Thus, proportionate liability encourages higher audit quality than joint and several liability whenever the incentive provided by comparative negligence is stronger than the incentive provided by the higher liability of the joint and several regime.

In his discussion of Narayanan, Gigler [1994] notes that it is the marginal cost of litigation, not the absolute cost, that determines the cost-minimizing amount of auditor effort. Thus, it is under the proportionate liability regime that the auditor’s litigation cost appears more sensitive to his or her effort. Gigler further asserts that Narayanan’s addition of comparative negligence captures an institutional richness not previously included when comparing liability regimes. However, Palmrose [1987] found that in approximately 70% of her bankrupt sample, the auditor made no payment, suggesting that the situation modeled here may be relatively uncommon.35

Unlike the previous audit quality studies, Moore and Scott [1989] conclude that auditors will underaudit relative to a socially desirable level of audit when limitations on exposure exist. This underaudit discrepancy is less for a larger, wealthier audit firm. When the authors include auditor collusion with management, the conclusions about audit firm size and business risk still hold. They conclude that since regulatory authorities lack knowledge of an auditor’s cost structure, it may be prohibitively expensive to force an “optimal” audit. However, in a more recent paper which continues to enhance institutional richness by considering more than just audit effort, Schwartz [1997] finds that the socially optimal audit effort level, as well as the socially optimal level of investment, are induced in a regime consisting of a strict liability rule. Even in a high audit quality situation, a measure of damages based on actual investment may lead to an over-investment in risky assets, as the liability payments are perceived by the investor as a form of insurance.

The studies reported above focus on audit effort from an analytical perspective. Two other studies examine litigation’s impact on the audit more broadly and utilize different methodology. Dopuch and King [1992] investigate 15 negligence liability markets in a laboratory setting to assess the effect of varying liability on the demand for and supply of auditing services. They devise a game imitating the roles of auditor, firm management, and users of audit information (who represent possible litigants) with three defined levels of responsibility (no liability, negligence, and strict liability). The prices paid for audit services in the no-liability markets were lowest, medium in negligent liability, and highest in strict liability. Interestingly, buyers seemed to treat strict liability almost like insurance. Their findings suggest negligence liability markets induce more economic efficiency. The authors note that their findings suggest no systematic benefits from imposing a strict liability rule, and conclude that expanding the scope of auditor liability may not achieve the desired effect. Wallin [1992] also devises a game scenario involving 32 experimental markets, which tests whether the availability of auditing and/or legal recourse affects the efficient allocation of resources in the market His results show a demand for auditing with or without the presence of legal recourse.

4.3 Evidence Concerning Unintended Consequences

The previous section examines literature that provides evidence concerning the ability of the civil liability system to encourage desired behavior (i.e., increased audit quality) and appropriate resource allocation (i.e., higher economic efficiency). This section emphasizes studies that provide insight regarding potential problems in the system caused by economic agents pursuing their own self-interests. As noted by Palmrose [1997], many of the questions concerning the role of merits in lawsuits against auditors are still unresolved, suggesting that research in the strategic considerations of lawsuit participants is still in its infancy. However, three foundational studies are emphasized here to present preliminary evidence of opportunistic behavior that could undermine societal objectives and represent an impediment to the system.

Alexander [1991] examines a group of securities class actions involving claims of fraud in initial public offerings of 17 companies. This study is often cited as providing evidence of a non-merits-related settlement regime (i.e., a strong case is not worth more in settlement than a weak one). She notes that a combination of factors, such as risk-averse defendants including auditors, extremely high potential damages, the plaintiff attorney’s contingency compensation system, agency problems, and insurance and indemnification rules, give some litigation participants strong incentives to avoid going to trial. In her explanation of the results, she suggests opportunistic behavior on the part of plaintiffs who target those, such as auditors, who have a large stake in avoiding lawsuits.

Palmrose [1994] also finds potential evidence of opportunistic behavior by plaintiffs in her examination of accountants as perceived “deep-pocket” defendants. In an examination of 227 allegations of audit failure against the largest audit firms from 1960 to 1994, she finds that auditors are sued most often jointly with other defendants. Her comparison of what auditors pay compared to these other defendants indicates that auditors typically pay only small amounts or make no payments at all to the plaintiffs, suggesting that a significant percentage of the claims examined were “weak.” Finally, reinforcing the findings of both Palmrose and Alexander, Cloyd, Frederickson and Hill [1996] study the role of proximate cause in a plaintiff attorney’s decision to recommend third-party lawsuits against auditors. The authors determine that proximate cause is less important in the initiation of lawsuits than anticipated given the principles of tort law.36

4.4 Conclusions Concerning the Effects of Litigation

Referring back to our framework of societal objectives, the results of the studies in this section seem to provide evidence concerning the ability of the civil liability system to encourage desired behavior. The predominant finding of the research on litigation’s effect on audit quality appears to be a link between higher quality and restrictions on auditor liability in certain settings (e.g., the negligence standard versus the strict liability standard, proportionate liability versus joint and several with a bankrupt audit client). By implication, it is in these settings that the societal objective of deterrence is met. However, as noted in Schwartz [1997], a high quality audit does not ensure a socially optimal investment level, raising questions about capital market efficiency. Thus, continued research efforts which vary contractual arrangements, liability sharing provisions, and assessments of parties’ negligence in various settings should provide important insights. There naturally exists a tradeoff between capturing the institutional complexity and modeling a situation so unique that generalizations are extremely limited.

As noted previously, the societal objective of deterrence may be accomplished through client screening as well as correctly pricing audit services. The initial evidence concerning client screening and risk adjustments supports their existence; however, individual audit firm strategies in response to litigation risk merit further study. Building on the work of Huss and Jacobs [1991], several areas of research into client screening seem promising: the determinants of client acceptance/retention; the intensity of the client acceptance/retention process; the role of insurance in the client acceptance/retention process; the autonomy of local offices in client acceptance/retention decisions; and the effect of client acceptance/retention decision on individual compensation. Again, an objective of the civil liability is communication of expectations; thus, empirical studies measuring before and after differences with the new liability regimes recently enacted, as well as the new Statements on Auditing Standards 82 [AICPA, 1997] on fraud detection could provide interesting comparisons. Likewise, building on the research of Simunic and Stein [1996], several areas of research into the risk premium seem promising: determinants of a risk premium; client actions to reduce risk premiums; and the effect of the marketplace on the risk premium.

Pratt and Stice [1994] note that their own experimental study is limited by not having competitive pressures and no economic incentives for the auditor to make accurate judgments and judicious decisions. They call for studies to further investigate how auditors respond to highly risky clients and alternative economic incentives. Consistent with this call, Dye [1995] models the effect of allowing auditors to limit their liability by incorporating. In what can clearly be interpreted as a strategic action, his analysis finds that auditors would attempt to shelter wealth through limited liability provisions if available.

Gigler [1994] suggests incorporating management incentives, efforts, and behaviors in future research. Building on management choice and the work of Alexander [1991] and Palmrose [1994], further examination of the strategic consequences of all parties to a lawsuit would be insightful.37 Focusing on the potential for unintended consequences stemming from the civil liability system, investigation into management incentive mechanisms, as well as board structure and stock ownership, and other aspects of corporate governance could aid our understanding of agency relationships which were not successful in protecting stockholder interests. For example, Dechow et al. [1996] capture several interesting incentive variables in their investigation of firms subject to accounting enforcement actions by the Securities and Exchange Commission for alleged earnings manipulation. Individual analyses of the cases may also provide interesting behavioral insights. Further study could investigate the interaction of client and audit firm characteristics, i.e., whether a more or less at-risk environment is defined by the inappropriate or appropriate match of the client and audit firm.

Finally, although several analytical pieces examine the effect of litigation on audit quality, additional empirical verification of the models’ results would be helpful. Since state common law provisions vary considerably, cross-sectional analysis of the effects of differing legal standards is possible.

5.0 THE MECHANISM OF DISPUTE RESOLUTION

When differences of opinion arise between parties about their duties and rights, society needs to have orderly means to resolve these disputes. Two major forms of dispute resolution exist: adjudication or settlement [Vago 1988]. Several studies examine different attributes of these dispute mechanisms. The first set of papers examines the issues associated with adjudication in a courtroom trial. The second set examines the issues associated with the settlements of disputes outside of a courtroom, including settlements of cases in which the formal litigation process may have already started. The studies discussed in this section are presented in Table 3, which summarizes the approaches used and key findings of this body of research.

Examination of the dispute mechanism can provide insights into whether societal functions are being met in a satisfactory method. Primarily, this research provides information about whether systematic biases exist or if unintended consequences result. Existence of either suggests that the civil liability system could be improved. The existence of biases also influences the perception of equity. As mentioned in Section 2.1.3, each member in society has certain rights that must be protected. However, if biases are present, someone’s rights are diminished and an impartial decision is less likely in a trial, violating the concept of equity.

5.1 Issues Related to Adjudication

The resolution of conflict through adjudication is achieved because the courts have the “authority to intervene and decide what is to be done without the joint consent of the parties” [Simpson, 1988, p. 11]. The perceptions of those who render judgments in audit litigation, jurors and judges, are of particular interest in the study of the trial mechanism. If these perceptions are in some way biased, trial outcomes may be unrelated to the merits of the plaintiff’s case and the civil liability system may lose some of its deterrence ability. If the trial is determined primarily without merits, the auditor has little protection in superior auditing abilities, and therefore little incentive to improve skills. Since the evidence examined is primarily about the auditor’s performance, the court’s evaluation will have great impact on the auditor attempting to limit litigation risk. Modifications to the design of work papers and audit procedures are likely to ensue only if court decisions are based on an unbiased evaluation of an auditor’s performance.

Although few cases in audit litigation are tried to verdict, the examination of trial decision making is still important. Clearly, although the numbers may be small, some cases do go to verdict and are subject to the decision making process. Although most cases are settled before a verdict, these cases are also affected by the possibility of a trial and the decision making process that may be exercised in court. One of the factors that influences the settlement process is the recognition that if the dispute is not settled, it will ultimately go to court.

5.1.1 Court Perceptions about Audit Evidence and Audit Standards

In the typical audit litigation trial, a key issue of dispute is the quality of the audit. The plaintiff typically alleges that the audit was substandard, while the auditor defends the quality of the audit.38 The court, to decide this issue, evaluates the audit evidence collected by the auditor. Auditing standards should provide the criteria for evaluation yet those making this evaluation, judges and jurors, are not obligated to use these standards.

Three papers examine the perceptions of the quality of audit evidence in a trial. In a pair of papers, Jennings, Kneer, and Reckers [1991; 1993] examine how changes within the auditing profession affect the perceptions of judges. The earlier paper finds that the disclosure of materiality amounts serves to reduce the potential liability of auditors. The second study finds that the use of a decision aid (in this case, for materiality limits) reduces the judge’s assessment of the auditor’s culpability. Although the auditor is asked to apply professional judgment in the audit process, as shown by these two papers, legal liability appears to be reduced as audit subjectivity is reduced. Yet, the elimination of subjectivity through the use of more sophisticated techniques may introduce other problems, as shown by Wallace [1983]. In this early piece examining audit evidence issues, she investigates five court cases to determine if regression analysis is viewed by the courts as a valid method of statistical analysis. Although she finds that regression analysis is accepted by the courts, its use requires considerable explanation to facilitate understanding by trial participants and to justify its conclusions. Thus, the auditor is confronted with a dilemma: although the method can result in statistically valid associations between variables, its complexity may create problems. Sophistication is needed to know when to apply regression analysis, to identify problems in the data, to interpret the results, and to explain the method to those less experienced in its use.

Audit standards are a common defense used by the auditor against charges of inadequate performance. Yet, this defense has two potential weaknesses. First, the court may choose to consider the auditor’s compliance with these standards as only one factor among many in the determination of the adequacy of the auditor’s performance. Second, audit standards are set by the profession primarily through the Auditing Standards Board of the AICPA. Arguably, this represents a conflict of interest, as the standard setting bodies are responsible to the public, but possibly still influenced by their profession. If this perception of a conflict of interest is held by jurors or judges, standards developed by the profession could be inherently suspect. Buckless and Peace [1993] examine these issues. Using members of jury pools, they present simplified legal cases asking for a decision in favor of the plaintiff (client) or the defendant (auditor). The facts of the cases and the auditing standards used as a defense were held constant, but the standard setting body was either the auditing profession or the government, and the jury was instructed to use compliance with the audit standard as either one of many or the sole criterion in evaluating the quality of the auditor’s performance. When the standard setting body is the profession itself, the jurors are more likely to find against the defendant (auditor).

5.1.2 Other Biases in the Trial Process

A bias in the court process can result in a decision that is not warranted based on the facts. Such a result violates the concept of equity. Several biases in the trial process have been alleged and the studies reported below find evidence for the existence of three commonly alleged biases in audit litigation.

Defendant auditors have frequently claimed to be victims of a hindsight bias in court decisions. Auditors must make critical decisions based on the information known at the time but those evaluating the auditor’s decisions have additional information unknown to the auditor at the time of the decision. If the evaluation is influenced by this additional information, hindsight bias is present. In Anderson, Jennings, Lowe, and Reckers [1997], 157 state court judges participated in an experiment designed to examine the hindsight bias. The results indicate the presence of a hindsight bias that is mitigated if the judges are asked to consider the additional stakeholders who would have been affected by alternative auditor decisions.

In audit litigation, like other litigation involving complex issues, expert witnesses are frequently used. The use of expert witnesses raises several issues, including the objectivity of the testimony. The court is interested in establishing fact, but the adversarial parties are interested in obtaining a favorable decision. Expert witnesses who are hired by the parties may tailor testimony, either consciously or subconsciously, to favor their employers. Ponemon [1995] explores this issue and finds that auditors and litigation specialists make estimates consistent with the positions of their hypothesized employers. To some degree, experience and ethical reasoning ability mitigate this apparent bias; however, his findings suggest that decision makers in a trial may have difficulty determining facts due to a lack of objectivity by expert witnesses. The accounting profession must also be concerned, since most experts in audit litigation are drawn from the ranks of the profession. Litigation support has become a growing source of fees in several accounting firms and the profession may face pressure to formulate a code of conduct tailored for litigation support services.

Finally, in a study of outcomes (e.g., auditor success rates, judgment amounts, appeals and settlements) and utilizing the model of Priest and Klein [1984] (see Section 5.2), Palmrose [1991b] presents a comprehensive examination of the use of trials for resolving auditor legal conflicts.39 She finds a higher success rate for auditors in a judge trial (as opposed to jury trial), a higher trial rate for audits of nonpublic companies (as opposed to the audits of public companies), and some limited success by auditors in settlements and appeals after the initial judgment. Her results suggest the common wisdom that auditors have a better chance of success at trial if the decision is made by a “sophisticated” judge rather than a jury comprised of “common” citizens may have merit.

5.2 Issues Related to Alternatives to Courtroom Adjudication

Certainly, the number of parties supporting tort reform suggests that many are dissatisfied with the current system. If, as has been alleged, the civil liability system is not meeting societal objectives, alternatives to courtroom adjudication should be available. The reform process, however, attempts to fix the basic trial mechanism, and with the exception of steps designed to encourage settlements, seldom deals with alternative mechanisms.40

The most common alternative to a dispute resolution by trial is a private negotiated settlement. The settlement may be reached at any time during the dispute process and may be heavily influenced by the trial itself or the threat of a trial or appeal. The important feature of a settlement is that the private negotiation process is conducted by the parties themselves, although the court may encourage and loosely supervise the negotiation.41

Priest and Klein [1984] analyze the economic factors that influence the decision to take a case to trial rather than accept a settlement. Constructing a model which defines the plaintiff’s minimum settlement price (asking price) and the defendant’s maximum settlement price (bidding price), they find that settlements will generally occur when the expectations about the outcome become more certain. With equal stakes to the parties, as the probability of plaintiff success moves away from 50%, one of the parties has an incentive to settle. In their model, the disputes that will go to trial are those with expectations of success of approximately 50%. Yet, when the stakes are higher for the defendant, such as a trial involving a defendant accountant, the incentives for settlement change. In such a situation, Priest and Klein predict more defendant victories at trial, since the plaintiff is more likely to press a weaker case.

Expanding upon Priest and Klein’s analysis, Alexander [1991] finds that settlements in securities class action lawsuits are systematically different than the cases that go to trial.42 Unlike Priest and Klein’s analysis that focuses primarily on the probability of success at trial, Alexander primarily examines the cost imposed on the defendants. She finds that certain defendants are targeted for class action lawsuits because of their strong motivations to avoid a trial. Even though these defendants may have done nothing inappropriate, they are willing to pay settlements to either avoid other costs (e.g., a damaged reputation or the loss of insurance coverage) or a catastrophic judgment in a lawsuit.

5.3 Conclusions about the Mechanism of Dispute Resolutions

The findings in this study area raise questions about whether the deterrence aspect of the civil liability system is being met. The research suggests that a trial against an auditor may not be decided solely on the quality of the auditor’s work. Both judges and juries exhibit tendencies that would diminish the importance of merits. Juries may not be able to understand sophisticated audit procedures and may view standards developed by the profession with skepticism. Judges seem to exhibit a hindsight bias and a bias against auditors induced by the subjectivity of the audit process. Even if the trial does not go to judgment, the settlement process may be influenced by factors other than the merit of the case against the auditor. Court decisions not based on the merits of the case can mean that an auditor who does high quality work may lose the trial, while an auditor who does low quality work may win the trial. Clearly, this is not a desirable outcome for society.

Research into the mechanism of dispute resolutions is in its infancy. Building on the results of earlier studies, particularly the findings in Palmrose [1991b], the prescription for future direction suggests a need for efforts aimed at exploring why an audit trial is different than other trials. For example, why do auditors appear to have a higher success rate in a trial by a judge than a trial by jury? Why is the trial rate for auditor litigation in excess of the rate for general securities litigation? Why are certain participants in securities fraud (e.g., underwriters) not sued as frequently as auditors? Such research would be fruitful in determining whether such differences are an indication of a unique civil liability setting (audit litigation) where the system is failing to meet its objectives and where reform may be needed.

In discussing Palmrose [1991b], Ettredge [1991] raises several interesting questions for further research. Of particular interest, he points to the need for research to determine the influence of litigation announcements on auditor reputation. Given the assumed importance of reputation effects on the settlement process, research in this area could be particularly useful.

Another area for future research involves further exploration of the perceptions of key players in adjudication such as attorneys, judges, and jurors. Behavioral studies need to further identify biases and investigate the influences of uncertainty and complexity in a trial setting. Such research could improve our understanding of the risk factors related to the adjudication mechanism and help develop means to counter the risks. For example, expansion of the work of Jennings, Kneer and Reckers [1993] could result in the development of decision aids to counter the effects of ambiguity.

The settlement process needs to be better understood. The findings of Alexander [1991] about the motivations of settlement participants need to be expanded. Of particular note, she finds that malpractice insurance plays a major role in the settlement process. Research into the use of malpractice insurance in audit litigation would be enlightening. Alexander’s paper is rich in institutional insight, however, a replication of the basic research study would be helpful. Although her results are fascinating and have been widely cited, they are based on a sample of 17 settlements. All the settlements involved companies from computer-related industries who issued initial public offerings in the first half of 1983 and whose cases were filed primarily in one federal court district. An analysis of a larger, more diverse sample would lend more validity to the results.

Finally, we have concentrated in this section primarily on trials and settlements. However, the applicability of alternative dispute resolution (ADR; see footnote 40) as a means of mitigating risk with audit clients could provide a useful direction for future research. As noted previously, its use in other disciplines is increasing and its acceptance becoming more widespread.

6.0 CONCLUDING REMARKS

Audit litigation is a topic of interest in the academic, practitioner, and regulatory communities. We introduced this paper by noting that audit litigation serves certain purposes in society. By research area, we used a framework to evaluate what is known about the satisfaction of these societal objectives with the purpose of identifying gaps in our knowledge. We believe this examination shows that, although significant progress has been made, much remains to be learned about such areas as the role of independence, client screening and the use of a risk premium in litigation prevention, behavior of participants in lawsuits and strategic decision analysis, the cost structure of litigation, and the impact of evolving legal factors. Given the potential wealth transfers and the competitive nature of a courtroom setting, changes to the legal provisions are likely to continue. Thus, this constantly-changing environment presents opportunities for research as our current understandings become dated. There are arguably few arenas in auditing research which present a greater opportunity to influence policy and the future direction of the accounting profession.

I Nowhere is this better seen than in the securities markets where substantial shifts in aggregate market indices may result from the release of macroeconomic data, and where large price changes in a specific security may accompany a new company disclosure.

2 Criminal liability may also be imposed in cases where the auditor knowingly attests to a false set of financial statements. Such cases are relatively rare [Davies, 1983; Arens and Loebbecke, 1997].

Several recent events are commonly cited as proof of litigation “excess.” In May 1992 Price Waterhouse lost a $338 million judgment (later, ordered back to trial). In the same year, Ernst and Young agreed to pay a $400 million settlement, and in March 1994 Deloitte and Touche agreed to a $312 million settlement both with the Resolution Trust Corporation. Negligence lawsuits were filed against Price Waterhouse and Ernst and Young for work on the bankrupt Bank of Credit and Commerce International (BCCI), with eight billion dollars in damages being sought. Two large CPA firms (Laventhol and Horwath and Spicer and Oppenheim) were forced into bankruptcy due, in part, to litigation.

4 Not all agree with those proposing legal reforms. Kaplan [1987] proposes that the solution to the litigation crisis lies within the accounting profession and not with external reforms. He suggests closing the expectations gap by providing an audit more in line with users’ expectations. This suggestion contrasts with proposals from the profession that say users should be educated about the attest function so that their expectations align with the profession’s. Kaplan plausibly concludes that the failure to provide a service consistent with users’ expectations will result in continual expansion of auditor liability, regardless of tort reform.

5 Palmrose [1997] examines the literature on the role of merits in audit litigation, one critical component in the ongoing debate.

6 The legal system includes several laws or legal theories that may be used to initiate a civil action against an auditor for an allegedly deficient audit. The most common are federal securities acts, state “Blue Sky” securities acts, and tort action under common law.

7 The use of different approaches to reach a common legal standard is illustrated by California and New Jersey, which have both replaced the foreseeable standard of third-party liability with a less inclusive third-party liability standard. The change in California was accomplished in 1992 by a state supreme court decision, Bily v. Arthur Young & Co. and the change in New Jersey by legislation in 1995.

8 Two recent examples of attempts to tailor the legal system to the preferences of competing litigants are the Private Securities Litigation Reform Act of 1995 (Public Law 104-67) and California Initiative 211. Supported by several tort reform organizations, the Private Securities Litigation Reform Act of 1995, which was enacted over a presidential veto, is generally seen to be pro-defendant. California Initiative 211, which was rejected by the electorate in 1996, was widely perceived to be pro-plaintiff.

9 Among the references available are Causey [1982], Davies [1983], Epstein and Spalding [1993], Miller and Brady [1986], Murray [1992], and St. Pierre [1983] along with the papers cited in later sections. to The reader is referred to Allen [1990], Cave [1985], Goldstein and Dixon [1989], Herskovitz [1990].

Hill, Metzger, and Schatzberg [1993], Johnson and Terando [1990], Martin [1989], and Siliciano [1988; 1997].

II This ability is dependent on the court’s authority to compel participants to its will (see Section 2.1.5 for a further discussion of this authority).

12 The preceding discussion assumes complete recovery. The basic tenet of the tort system is that a party wrongfully injured has the right to be compensated by the injuring party for the damage inflicted (CoRey and Robert, 1975]. While complete recovery may be the ideal, it is not likely to occur in reality, nor is it necessarily always desirable from a societal perspective.

13 Defining this as a cost minimization problem implies a utilitarian framework which attempts to obtain the greatest good for the greatest number of people. Emmanuel Kant and other philosophers of the deontological school would suggest that retribution is an important part of deterrence. They feel that justice demands that certain actions receive punishment regardless of their future effect.

14 This paper does not directly address the issue of sanctions or other actions designed to restrain a wrongdoer. Some sanctioned firms are included in the samples of research discussed.

15 Besides examining characteristics of those involved in litigation, Stice [ 19911 also develops a model useful in predicting which audit clients may result in litigation.

16 Exploring the possibility that initial public offerings (IPOs) represent high litigation risk engagements, Bonner et al. [1998] include an indicator variable for companies with IPOs within three years of the first year of a fraud. The variable is not significant in any of the models tested although IPO status is frequently cited in the practitioner literature as a major risk factor for audit litigation. The reader is referred to Simunic and Stein [1996] for a review of the evidence concerning the effect of IPO status on litigation risk. 17 The sample in Stice [1991] does not include firms in the financial and service industries.

18 Jones and Weingram [1996c] also suggest that certain types of litigation risk are higher for technology and financial services firms. In three working papers, Jones and Weingram [1996a; 1996b; and 1996c] explore Rule lOb-5 (Securities Exchange Act of 1934) litigation risk. Their focus is not auditor litigation, however, their findings provide useful additional evidence concerning potential client characteristics associated with audit litigation.

19 Lys and Watts [1994] test the variables hypothesized to be associated with litigation in four ways: a cross-sectional univariate test; a cross-sectional multivariate test; a temporal univariate test; and a temporal multivariate test. Unless otherwise noted, the reported results are drawn from the multivariate tests since the inclusion of controls should produce more reliable results.

20 In their cross-sectional univariate tests, Lys and Watts [1994] find the association between accruals and litigation weakly significant. Unlike Stice [1991], Lys and Watts [1994] find no significant association with litigation for either receivables (scaled by total assets) or inventory (scaled by total assets).

21 Likewise, industry affiliation and public status are correlated with size; thus, size is possibly driving these results.

22 In their multivariate analysis, the absence of a modified audit opinion is not significantly associated with litigation.

23 The authors suggest that their findings pertaining to beta and size may proxy for firm-specific characteristics that create a special vulnerability to securities litigation.

24 Insider trading activity is not supported as a significant factor in the Dechow et al. [1996] investigation of SEC enforcement actions or in Jones and Weingram [1996b].

25 Although the external auditor may not have been named as a co-defendant, the fact that an accounting issue is a basis for the conflict suggests that he or she could have been; thus, the findings are of interest to an accounting audience.

26 Lys and Watts [1994] did find a significant association between stock returns and litigation in the cross-sectional univariate tests.

27 The largest national CPA firms have been called Big 8, Big 6 and Big 5 depending on their composition at the time. We adopt whatever label was used by the author of any cited study and in current references use the term Big 5.

28 Because fee information is generally unavailable, both studies use client sales instead. The client’s sales are divided by the sales of all the auditoris clients to obtain a measure of the relative importance of the client to the auditor. The relative importance of the client can give a rough measure of the audit firm’s ability to resist a client’s attempt to extract concessions from the auditor (a component of auditor independence). To the extent that audit fees positively correlate with client sales, this is a valid proxy for the proportion of fees. A more fundamental concern is whether the proportion of fees is a valid proxy for audit independence. 29 Skinner [1997] has a similar finding.

30 For background reading as well as a current focus on the ongoing debate concerning the nature and extent of threats to auditor independence, the reader is directed to Allen [ 1997] and the February 18, 1998 policy statement issued by the SEC (The Establishment and Improvement of Standards Related to Auditor Independence).

31 As noted previously, audit litigation, in its various forms, may impose costs on the auditor. Since a competitive marketplace may limit the auditor’s ability to pass the cost on to clients, the auditor should be motivated to avoid these costs. Supporting this statement, Wilson and Grimlund [1990] find that firms censured by the SEC had difficulty attracting and retaining clients, particularly in the states mentioned in the SEC enforcement releases.

32 Nelson, Ronen, and White [1988], however, suggest that there exists a set of audit clients for which optimal auditing procedures will not reduce the auditor’s expected losses.

33 The reader is referred to Krishnan and Krishnan [1997) for a comprehensive summary of the literature concerning auditor change, auditor switching, and auditor resignations.

34 The authors provide an important discussion of the sample design differences between their study and these three studies as well as among the three studies.

335 From a different perspective, Thoman [1996] notes that in a competitive market setting where clients shop for opinions, regardless of the current legal environment, auditors are motivated to reduce their exposure by reporting more conservatively, not by improving the audit

36 The single-period equilibrium model of DeJong [1985], discussed previously, includes, along with the audit firm, the representative investor and lawyer. The primary focus is on the investor and the attorney and their respective incentives to litigate, The results find that both the investor and the lawyer will litigate more often with class-action privileges and contingent legal fees, which suggests opportunistic behavior.

37 For background reading on litigants’ behavior in securities class actions in general, the reader is referred to Dunbar et al. [ 1995].

38 Other defenses are commonly available to the auditor depending on the allegations made but the quality of the audit work must usually be defended.

39 Excluding Alexander [1991] and Palmrose [1991b; 1994], most studies do not consider the effects of settlements nor the outcome of the trials. Audit litigation studies primarily use only filed lawsuits, with little consideration of settlements or resolutions. In other words, a lawsuit won by the auditor is not considered differently than a lawsuit lost by the auditor. The reason for these exclusions is the lack of available data on most lawsuits. Palmrose [1991a] determines that only about 20% of auditor litigation cases have widely disclosed information. She also finds that a small number of cases (52) account for nearly half of the total Wall Street Journal disclosures.

40 Alternative dispute resolution (ADR) has developed as a means of avoiding court litigation. The process, which involves arbitration or mediation, often takes much less time than a trial and can use expedi ted rules of evidence and discovery. ADR has somewhat limited applicability in audit litigation. Courts are reluctant to force contract provisions onto those who were not involved with the contract negotiations; thus, ADR would be most appropriate in client-initiated lawsuits or when a small third-party class exists. ADR has received little attention in the academic accounting literature, though recent publications in accounting practitioner journals, as well as in academic journals of other disciplines suggest that its use may become more prevalent as a means to limit litigation.

41 An insurance carrier may force the settlement process upon the defendant but otherwise it is a process freely entered into by the participants. Strong arguments can be made that the settlement process is in society’s best interest by leading to better resource allocation decisions. Each party enters the agreement without coercion (unlike a trial decision) and the trial costs, borne heavily by society, are avoided.

42 The focus of Alexander [1991] is on securities class action lawsuits in general, not just those lawsuits that name the external auditor as a defendant.

ANNOTATED BIBLIOGRAPHY

1. Alexander, J. 1991. Do the merits matter? A study of settlements in security class actions. Stanford Law Review (February): 497-598.

The author studies whether settlement outcomes in a group of securities class actions reflect case merits. She examines 17 computer-related companies who issued initial public offerings in the first six months of 1983 and determines that there appears to exist a non-merits based settlement rate. Though the conclusion is based on a small sample and the article appears outside of mainstream accounting journals, it has been influential in providing a theoretical framework for accounting researchers investigating the incentives of litigation participants and, in particular, continues to provide support for the need to study the existence of opportunistic behavior.

2. Bonner, S., Z. Palmrose, and S. Young. 1998. Fraud type and auditor litigation: An analysis of SEC Accounting and Auditing Enforcement Releases. The Accounting Review (October): 503-532.

The authors investigate 261 companies that were the subject of SEC enforcement actions between 1982 and 1995. They divide their sample into those companies who have no disclosure or reporting litigation, those with auditor litigation, and those with litigation that does not involve an auditor. The authors develop a fraud taxonomy which includes 12 general categories of fraud and use a multivariate model which includes control variables for client, auditor, and case characteristics. They find support for a higher incidence of auditor litigation when a company’s financial statements contain a fraud that involves fictitious transactions and events or a fraud that is designated as more common.

3. Carcello, J., and Z. Palmrose. 1994. Auditor litigation and modified reporting on bankrupt clients. Journal of Accounting Research (Supplement): 1-30.

The authors investigate 655 public companies that declared bankruptcy between 1972 and 1992 as a means of assessing whether modified audit reports issued prior to bankruptcy offer protection to the auditor from litigation. Of this sample, 118 had auditor litigation and 41 more had financial or reporting litigation that did not involve the auditor. A multivariate, logistic regression is tested with occurrence of auditor litigation as the dependent variable, the presence of a modified report and persistence of modified reporting as test variables, and the control variables of client financial condition, unrevealed irregularities, client size and client industry. Neither test variable was significant and the authors note that multicollinearity could be an issue. In the univariate tests, non-litigation clients were significantly more likely to have a modified opinion, as well as persistent modified opinions, than litigation clients. The low litigation rate for modified audit opinions suggests that a modified audit opinion reduces, but does not eliminate, the likelihood of litigation.

4. Dopuch, N., and R. King. 1992. Negligence versus strict liability regimes in auditing: An experimental investigation. The Accounting Review (January): 97-120.

This paper uses a laboratory study involving 120 undergraduate business students to assess the demand for and supply of auditing services in 15 experimental markets. Six of the markets involve a negligence liability regime similar to what auditors currently face, six involve a strict liability regime and three involve a no-liability regime. Each of the markets involve two sellers, two verifiers (auditors), and four buyers who transact for 20 periods. The authors test hypotheses concerning size of offers, hiring of verifiers, frequency of costly investment, verifier’s testing decision, pricing by buyers, and market efficiency. The findings suggest that negligence liability markets induce more economic efficiency.

5. Francis, J., D. Philbrick, and K. Schipper. 1994a. Shareholder litigation and corporate disclosure. Journal of Accounting Research 32: 137-164.

The authors examine corporate disclosures of companies involved in shareholder litigation brought under Rule lOb-5 of the Securities Exchange Act of 1934 during 1988-1992. The companies are in the chemical, computing, electronic, and retailing industries. They identify companies “at risk” for litigation due to their disclosure of adverse earnings news and companies that experienced earnings-based lOb-5 litigation. They did not find support for the hypothesis that severe earnings declines lead to a shareholder lawsuit. Lawsuit companies had approximately three times the number of disclosures as the “at risk” companies in the year preceding the adverse earnings news. An analysis of forecast revisions also shows no evidence of greater anticipation of bad news by the analysts following the “at risk” companies.

6. Lys, T., and R. Watts. 1994. Lawsuits against auditors. Journal of Accounting Research 32 (Supplement): 65-93.

The study examines 12 client characteristics and 9 audit firm characteristics using a pair-matched sample of 153 firms whose auditors are and are not sued during 19551994. The authors measure the variables in both the wrongdoing year and the year of filing. They conduct a univariate analysis (both parametric and nonparametric tests) and a multivariate analysis across subperiods hypothesized to reflect different levels of auditor exposure. Their findings indicate that lawsuits against auditors are associated with client’s financial difficulty (probability of bankruptcy and probability of acquisition), larger size, and the presence of income-increasing accruals. Lawsuits naming auditors as defendants are also more likely as the audit firm’s fraction of audit revenues derived from the client increases. The authors attribute this last finding to potential impairment of independence.

7. Narayanan, V. 1994. An analysis of auditor liability rules. Journal of Accounting Research 32 (Supplement): 39-59.

The paper examines audit quality under a joint and several liability regime and a proportionate liability regime through modeling of auditor litigation brought under Rule lOb-5 of the Securities Exchange Act of 1934. The author uses a single-period

game in which all players are risk-neutral and auditing is mandatory with fixed audit fees. The model incorporates comparative negligence and finds a potential increase in audit quality under proportionate liability. Under proportionate liability, the auditor pays only his/her share of the damages, and thus he/she has greater incentive to minimize litigation cost by working harder. Under joint and several liability, the auditor must pay all of the damages when the auditee is insolvent, regardless of the audit effort expended.

8. Palmrose, Z. 1988. An analysis of auditor litigation and audit service quality. The Accounting Review (January): 55-73.

The study examines litigation activity among audit firms as a means of assessing auditor quality. The author uses 472 cases of litigation involving alleged audit failure during the period 1960 to 1985. The 16 largest audit firms were classified as Big 8 and non-Big 8. Using a frequency rate scaled by either the number of clients or U.S. auditing fees, she finds that Big 8 auditors are significantly less likely to be defendants in litigation than the eight largest non-Big 8 firms. She notes no significant difference between Big 8 and non-Big 8 auditors in resolution rates of lawsuits.

9. Pratt, J., and J. Stice. 1994. The effects of client characteristics on auditor litigation risk judgments, required audit evidence, and recommended audit fees. The Accounting Review (October): 639-656.

A field experiment involving 243 audit partners and managers of four Big 6 firms is used to examine client screening. Participants assess litigation risk, make planning decisions, and set audit fees for potential clients. Poor financial condition is associated with higher levels of litigation risk, collection of more audit evidence, and higher fees. Stock price was not found to be a significant determinant.

10. St. Pierre, K., and J. A. Anderson. 1984. An analysis of the factors associated with lawsuits against public accountants. The Accounting Review (April): 242-263. The authors examine 129 cases filed against public accountants. The cases are ana

lyzed as to: 1) type of engagement; 2) type of alleged error; 3) the event which initiated the error search; and 4) situational characteristics. Their results indicate that procedural matters in the conduct of an engagement are less important, and that lawsuits most frequently contest the interpretation of an auditing or accounting standard (GAAS or GAAP interpretation). The lawsuit is usually initiated after the occurrence of a client-specific or situational factor that suggests an audit failure may have occurred. Thirty of the 129 cases analyzed involve public accountants with three years or less tenure with the client. Public companies and companies in the finance, insurance, real estate and manufacturing industries have the highest litigation rate.

REFERENCES

Alexander, J. 1991. Do the merits matter? A study of settlements in security class actions. Stanford Law Review (February): 497-598.

Allen, J. 1990. Damage apportionment in accounting malpractice actions: The role of comparative fault. Brigham Young University Law Review 3: 949-85.

Allen, W. 1997. Auditor independence and the promotion of the public trust. The Second Annual Seymour Jones Distinguished Lecture. Ross Institute of Accounting: November 24.

American Institute of Certified Public Accountants (AICPA). 1997. Consideration of fraud in a financial statement audit. Statement on Auditing Standards No. 82. New York, New York: AICPA.

Anderson, J., M. Jennings. D. Lowe, and P. Rockers. 1997. The mitigation of hindsight bias in judges’ evaluation of auditor decisions. Auditing: A Journal of Practice and Theory (Fall): 20-31.

Arena, A., and J. Loebbecke. 1997. Auditing: An Integrated Approach. Upper Saddle River, New Jersey: Prentice Hall.

Ayers, S., and S. Kaplan. 1998. Potential differences between engagement and risk review partners and their effect on client acceptance judgments. Accounting Horizons (June): 139-153.

Balachandran, B., and N. Nagarajan. 1987. Imperfect information, insurance, and auditors’ legal liability. Contemporary Accounting Research (Spring): 281-301.

Bily v. Arthur Young do Co. 92 D.A.R. 11971 (1992).

Blacconiere, W., and M. DeFond. 1997. An investigation of independent audit opinions and subsequent independent auditor litigation of publicly-traded failed savings and loans. Journal of Accounting and Public Policy 16:415-454.

Bonner, S.E., Z. Palmrose, and S. Young. 1998. Fraud type and auditor litigation: An analysis of SEC Accounting and Auditing Enforcement Releases. The Accounting Review (October): 503-532. Buckless, E, and R. Peace. 1993. The influence of the source of professional standards on juror decision

making. The Accounting Review (January): 164-175.

Carcello, J., and 7- Palmrose. 1994. Auditor litigation and modified reporting on bankrupt clients. Journal of Accounting Research 32 (Supplement): 1-30.

Causey, D. 1982. Duties and Liabilities of Public Accountants. Hamate, Illinois: Scott, Foresmann and Company.

Cave, G. 1985. Extension of accountants’ liability for negligence: One step closer to a new implied warranty of results. University of Colorado Law Review 56 (Winter): 265-287.

Cloyd, C., J. Frederickson, and J. Hill. 1996. Motivating factors against independent auditors: Experimental evidence on the importance of causality. Journal of Accounting and Public Policy (Fall): 185-218. Corley, R. and W. Robert 1975. Principles of Business Law. Englewood Cliffs, New Jersey: Prentice Hall. Davies, J. 1983. CPA Liability: A Manual for Practitioners. New York, New York: Wiley.

Dechow, P, R. Sloan, and A. Sweeney. 1996. Causes and consequences of earnings manipulations: An analysis of firms subject to enforcement actions by the SEC. Contemporary Accounting Research (Spring): 136.

DeJong, D. 1985. Class-action privileges and contingent legal fees: Investor and lawyer incentives to litigate and the effect on audit quality. Journal of Accounting and Public Policy 4: 175-200.

Dopuch, N., and RL King. 1992. Negligence versus strict liability regimes in auditing: An experimental investigation. The Accounting Review (January): 97-120.

Dunbar, F, V. Juneja, and D. Martin. 1995. Shareholder litigation: Deterrent value, merit and litigants’ options. Business, Law and Economics Center, John M. Olin School of Business, Law and Economics Center, John M. Olin School of Business. Washington University, St. Louis, Missouri.

Dye, R. A. 1995. Incorporation and the audit market. Journal of Accounting and Economics (February): 75114.

Epstein, M., and A. Spalding. 1993. The Accountant’s Guide to Legal Liability and Ethics. Homewood, Illinois: Irwin.

Eta-edge, M. 1991. Discussion of trials of legal disputes involving independent auditors: Some empirical evidence. Journal of Accounting Research 29: 186-193.

Francis, J. 1994. Discussion of lawsuits against auditors. Journal of Accounting Research 32 (Supplement): 95-102.

Francis J., D. Philbrick, and K. Schipper. 1994a. Shareholder litigation and corporate disclosure, Journal of Accounting Research 32: 137-164.

. 19946. Determinants and outcomes in class action securities litigation. University of Chicago working paper.

. 1998. Earnings surprises and litigation risk. Journal of Financial Statement Analysis (Winter): 15-27. Gigler, F. 1994. Discussion of an analysis of auditor liability rules. Journal of Accounting Research 32 (Supplement): 61-64.

Goldstein, J., and C. Dixon. 1989. New teeth for the public’s watchdog: The expanded role of the independent accountant in detecting, preventing, and reporting financial fraud. The Business Lawyer (February): 439-502.

Herskovitz, P. 1990. Auditors and third party negligence suits: Judicial approaches and legislative reforms. The Ohio CPA Journal (Winter): 21-28.

Hill, J., M. Metzger, and J. Schatzberg. 1993. Auditing’s emerging legal peril under the national surety doctrine. Accounting Horizons 7: 12-28.

Huss, H. and F. Jacobs. 1991. Risk containment: Exploring auditor decisions in the engagement process. Auditing: A Journal of Practice and Theory (Fall): 16-32.

Jennings, M., D. Kneer, and P. Reckers. 1991. Selected auditor communication and perceptions of legal liability. Contemporary Accounting Research 7: 449-465.

– 1993. The significance of audit decision aids and precase jurists’ attitudes on perceptions of audit firm culpability and liability. Contemporary Accounting Research 9:. 489-507.

Johnson, O., and W. Terando. 1990. From contract to tort: The evolution of accountants’ legal liability for negligence. Research in Accounting Regulation 4: 77-98.

Jones, C. and S. Weingram. 1996a. The determinants of lOb-5 litigation risk. Stanford University working paper.

. 19966. The effects of insider trading, seasoned equity offerings, corporate announcements, accounting restatements, and SEC Enforcement Actions on IOb-5 litigation risk. Stanford University working paper.

. 1996c. Why IOb-5 litigation risk is higher for technology and financial services firms. Stanford University working paper.

Kaplan, R. 1987. Accountants’ liability and audit failures: When the umpire strikes out. Journal of Accounting and Public Policy 6: 1-8.

Kellogg, R. 1984. Accounting activities, security prices, and class action lawsuits. Journal of Accounting and Economics (December): 185-204.

Kinney, W. 1994. Audit litigation research: Professional help is needed. Accounting Horizons 8: 80-86. Krishnan, J., and J. Krishnan. 1997. Litigation risk and auditor resignations. The Accounting Review (April): 539-560.

Lys, T., and R. Watts. 1994. Lawsuits against auditors. Journal of Accounting Research 32 (Supplement): 6593.

Martin, J. 1989. SEC case law: A summary for accountants. Research in Accounting Regulation: 227-238. Miller, R., and G. Brady. 1986. CPA Liability: Meeting the Challenge. New York, New York: Wiley.

Miller, R. and M. Young. 1997. Financial reporting and risk management in the 21st century. Fordham Law Review 65 (April): 1987-2064.

Moore, G., and W. Scott. 1989. Auditors’ legal liability, collusion with management, and investors’ loss. Contemporary Accounting Research 5: 754-774.

Murray, M. 1992. Managing the Malpractice Maze. New York, New York: American Institute of Certified Public Accountants.

Narayanan, V. 1994. An analysis of auditor liability rules. Journal of Accounting Research 32 (Supplement): 39-59.

Nelson, J., J. Ronen, and L. White. 1988. Legal liabilities and the market for auditing services. Journal of Accounting, Auditing, and Finance 3: 255-295.

Palmrose, Z. 1987. Litigation and independent auditors: The role of business failures and management fraud. Auditing: A Journal of Practice and Theory (Spring): 90-103.

– .1988. An analysis of auditor litigation and audit service quality. The Accounting Review (January): 55-73.

. 1991a. An analysis of auditor litigation disclosures. Auditing: A Journal of Practice and Theory 10 (Supplement): 54-71.

. 1991b. Trials of legal disputes involving independent auditors: Some empirical evidence. Journal of Accounting Research 29 (Supplement): 149-193.

– .1994. The joint and several liability debate: An empirical investigation of audit-related litigation. Stanford Journal of Law, Business & Finance (Fall): 53-74.

. 1997. Audit litigation research: Do the merits matter? An assessment and directions for future research. Journal of Accounting and Public Policy 16: 355-378.

– .Forthcoming. Empirical research on auditor litigation: Considerations and data. American Accounting Association.

Ponemon. L 1995. The objectivity of accountants’ litigation support judgments. The Accounting Review (July): 467-488.

Pratt, J., and J. Stice. 1994. The effects of client characteristics on auditor litigation risk judgments, required audit evidence and recommended audit fees. The Accounting Review (October): 639-656.

Priest, G., and B. Klein. 1984. The selection of disputes for litigation. The Journal of Legal Studies 13 (January): 1-55.

Private Securities Litigation Reform Act of 1995. Public Law No. 104-167.

Schwartz, R. 1997. Legal regimes, audit quality and investment. The Accounting Review (July): 38S

Shavell, S. 1987. Economic Analysis of Accident Law. Cambridge, Massachusetts: Harvard University Press. Siliciano, J. 1988. Negligent accounting and the limits of instrumental tort reform. Michigan Law Review (August): 1929-1980.

. 1997. Trends in independent auditor liability: The emergence of a sane consensus? Journal of Accounting and Public Policy 16: 339-353.

Simon, M. 1981. Impact of information, costly litigation, and product quality. Bell Journal of Economics (Spring): 171-184.

. 1982. Diagnosis and medical malpractice: A comparison of negligence and strict liability systems. Bell Journal of Economics (Spring): 170-180.

Simpson, A. 1988. Invitation to Law. Now York, New York: Basil Blackwell.

Simunic, D., and M. Stein. 1990. Audit risk in a client portfolio context. Contemporary Accounting Research 6: 329-343.

1996. The impact of litigation risk on audit pricing: A review of the economics and the evidence. Auditing: A Journal of Practice and Theory 15 (Supplement): 119-134

Skinner, D. 1997. Earnings disclosures and stockholder lawsuits. Journal of Accounting and Economics 23: 249-282.

St. Pierre, K. 1983. Auditor Risk and Legal Liability. Ann Arbor, Michigan: UMI Research Press.

St. Pierre, IC, and J. Anderson. 1984. An analysis of the factors associated with lawsuits against public accountants. The Accounting Review (April): 242-263.

Stice, J. 1991. Using financial and market information to identify pre-engagement factors associated with lawsuits against auditors. The Accounting Review (July): 516-533.

Summers, S. and J. Sweeney. 1998. Fraudulently misstated financial statements and insider trading: An empirical analysis. The Accounting Review (January); 131-146.

Thomas, L. 1996. Legal damages and auditor efforts. Contemporary Accounting Research 13: 275-306. Vago, S. 1988. Law Qc Society. Englewood Cliffs, New Jersey: Prentice Hall.

Wallace, W 1983. The acceptability of regression analysis as evidence in a courtroom: Implications for the auditor. Auditing: A Journal of Practice and Theory 2: 66-90.

Wallin, D. 1992. Legal recourse and the demand for auditing. The Accounting Review (January): 121-147. Wilson, T., and R. Grimlund. 1990. An examination of the importance of an auditor’s reputation. Auditing: A Journal of Practice and Theory (Spring): 43-59.

Claire Kamm Latham

Washington State University

Mark Linville

Washington State University

The authors would like to thank Bill Messier and two anonymous referees for their helpful comments on previous versions of this paper. We would also like to thank Joe Carcello, Jane Cote, David Gilbertson, Fred Jacobs, Donna Philbrick, John Thornton, and the participants at the American Accounting Association Western Regional Meeting for their insightful suggestions related to this review. This review was supported in part by a research grant from Washington State University.

Copyright University of Florida, Accounting Research Center 1998

Provided by ProQuest Information and Learning Company. All rights Reserved