Despite a relatively small number of high profile corporate failures and accounting scandals such as Enron and WorldCom, the number of demonstrated audit failures as evidenced by successful litigation or SEC sanctions approaches an annual failure rate of close to zero. In addition, our interpretation of the academic research suggests that many of the “solutions” embodied in SOX are not only unlikely to solve the profession’s alleged “problems,” they may well have serious unintended negative consequences.
So the disconnect is large between the scientific evidence on audit quality and institutional changes premised on the assumption that auditing is broken. This paper attempts to stimulate research into some of the important questions implicitly raised by SOX regarding the audit profession’s potential failings. An outline of our primary observations and suggestions are presented in the paper’s Introduction. Does Auditing Matter? INTRODUCTION
Accounting scandals Essay Example
While the beginning of the 21st century has been marked by accounting scandals, a major stock market crash, and the most sweeping securities market reforms since the 1930’s, one unexpected consequence of these events is an increased awareness that auditing “matters. ” In particular, regulators, market participants, and the public all seem to have a greater appreciation for the critical role auditing plays in the successful functioning of the U. S. financial markets. Along with this greater appreciation, however, is also widespread criticism that many aspects of auditing are “broken” and in need of fundamental reform.
This combination of a greater appreciation for auditing and a call for auditing reforms provides an opportunity for auditing researchers to help shape the future of auditing. The purpose of this paper is to identify some of the key areas in which auditing reforms have been made, discuss some of the research issues that have been identified in these areas, and provide suggestions for researchers in addressing the critical research questions in these areas. The outline below summarizes our primary observations along with some suggestions for future research in each of the areas we discuss. . While historically the auditing profession is often intensely criticized following boom-bust economic cycles, the criticism embodied in SOX is unusually intense, and appears to be partially motivated by political expediency and often based on anecdotes. Thus, we strongly encourage researchers to use their training in the scientific method to bring rigor and discipline to address the criticism implicit in SOX. We caution, however, that auditing researchers must take great care to maintain their “independence” from the auditing profession.
If we become apologists for the auditing profession we not only lose our credibility, we also do great harm to those who put their trust in us (including our students and the profession itself). 2. SOX makes dramatic changes to some of the fundamental institutions that define auditing in the U. S. In particular, SOX transforms auditing from a self-regulated industry that is overseen by a government agency (the SEC), to an industry that is now directly controlled by a quasi-governmental agency (the PCAOB).
We encourage researchers to investigate issues related to this fundamental change in institutional arrangements and provide the following suggestions: a. Looking back, is there any direct evidence that lack of self-oversight or poor standard-setting contributed to the recent accounting and auditing failures? Going forward, should we expect the institutional changes mandated by SOX to improve audit quality? b. The legal system is an important institution that has a major impact on the auditing profession.
While SOX does not directly make changes to this institution, it is an open question whether we need the kind of extreme litigation exposure we currently have in the U. S. in order to achieve an appropriate level of audit quality, and we casually note that Canada and Australia appear to have credible auditing without imposing such a brutally litigious environment. Thus, we encourage researchers to address issues regarding the role and importance of litigation in maintaining high audit quality. 3.
SOX made changes to several engagement-specific characteristics with the ultimate aim of improving auditor independence. The most contentious change is the banning of most nonaudit services performed by the incumbent auditor. Other changes include mandatory audit partner rotation and a one-year waiting period before auditor employees can accept executive positions with their former clients. We have the following suggestions for investigating the propriety of these changes: a. We believe that the SOX provision that bans most nonaudit services is at best misguided, and at worst politically-motivated.
We also believe, however, that there are many important questions not yet addressed by researchers in this area, including the following: Do personal relationships created by nonaudit services threaten independence? Are contextual issues (such as the firm’s overall governance environment) important in explaining whether auditor independence is impaired? Should nonaudit fees be any different from total fees in terms of creating financial dependencies? b. Because there is a realistic concern that mandatory audit firm rotation may yet be proposed by the SEC, we encourage more research in this area.
Since there is little research on the effects of the “revolving door” we encourage more research in this area as well. 4. While some researchers were examining audit committee issues prior to SOX (e. g. Klein 1998), the emphasis that SOX puts on audit committees suggests that there is much more to be done. Luckily, we have a large body of corporate governance literature in finance and economics to help us leverage our differential advantages in this research area, which include a deep institutional understanding of auditing issues.
Below are our suggestions for future research in this area. a. There are no theories that explain why boards of directors exist, much less why audit committees exist; and thorny empirical issues (such as endogeneity) are a major obstacle to governance research. Thus, auditing researchers embarking on this area face major hurdles in motivating and designing their studies. Nevertheless, given the critical importance of this research, and the fact that we have guidance from a large body of prior research in other fields, we encourage research in this area. b.
While SOX mandates that all audit committee directors be independent, the majority of prior studies fail to find evidence that governance is improved by having all of the audit committee directors be independent (although there is evidence that a higher proportion of independent directors on the audit committee improves governance). This raises questions about the potential benefits of having insiders on the board and we encourage future researchers to investigate this question. c. SOX requires that firms disclose whether they have a financial expert on their audit committee, and if not, why not.
While several studies find evidence that financial expertise on the audit committee improves governance, the evidence also suggests that expertise only accrues to firms that already have a relatively strong governance environment. Thus, we encourage researchers to further address issues related to the contextual nature of the benefits from audit committee expertise. d. While issues of committee size and activity are not directly addressed in SOX, there is some research in this area which suggests larger and more active audit committees improve governance.
Given the paucity of existing research, we encourage more work in this area. e. We are not aware of any auditing related research regarding at least four SOX promulgations that affect audit committees. While most of these new rules might seem reasonable at first blush – such as setting up procedures to investigate employee complaints — it is important to keep in mind that they are not costless, and that one cost can be significant unintended negative consequences. Therefore, we encourage research that addresses the expected or realized efficacy of these new rules. 5.
There are many issues of audit quality that are not directly addressed in SOX. Several of these areas are important and we have several suggestions for pushing research in this area further. a. The existing research on auditor size may be interpreted as either larger auditors supplying higher levels of audit quality, or as smaller auditors supplying sub-optimal levels of audit quality. Given the SEC’s move towards improving audit quality, we suggest researchers consider the ramifications of future regulation that may seek to mandate a uniform level of audit quality across auditors of all sizes.
For example, what would be the incremental cost of mandating uniformly higher quality, and what would be the effect on market concentration and efficiency? Would such a move increase market concentration, reduce competition, and create monopoly pricing power by accounting firms? b. Regulators and researchers generally seem to assume (at least implicitly) that the auditor’s bias towards conservatively-reported financials improves earnings quality. We observe that while this may be true for creditors, equity holders may not benefit from the auditors conservative bias.
We therefore encourage researchers to consider issues related to the link between auditing and earnings quality. For example, is the current conservative bias in auditing optimal? What is the “cost” to equity markets (if any) of the conservative bias in auditing? What would be the effects of reducing the conservative bias in auditing? c. Researchers have used several outcome measures to infer audit quality, including abnormal accruals, earnings restatements, and going concern opinions. Despite inherent weaknesses in all of these measures we believe they are all justifiable and valid measures.
We also encourage researchers to explore additional outcome measures in future research. For example, while researchers typically look to earnings-related measures to capture earnings management, Barton and Simko (2002) present compelling evidence that balance sheet measures may also be useful in this regard. d. Given the FASB’s seemingly inevitable move towards “fair value-based” and “principles-based” accounting, we strongly encourage auditing researchers to address how these changes are likely to affect auditor behavior, and ultimately how the changes in auditor behavior are likely to impact financial reporting.
Given that several countries outside the U. S. allegedly use more fair value-based and principles-based standards, non-U. S. data may be useful route to address these questions. e. While cross-country research presents major empirical challenges, such as controlling for potentially confounding effects, its considerable strength is its ability to allow tests for differences in institutions. Thus, because the potential rewards from this research are so great, we strongly encourage researchers to continue to pursue research using cross-country data. f.
A slippery but important unanswered question is “what is the optimal level of auditing? ” For example, section 404 of SOX is anticipated to increase audit fees by 50-100%. Are there measurable benefits in audit quality that would justify these fee increases? WHY DOES AUDITING MATTER NOW? Recent events have lead to scrutiny of the auditing profession Economic cycles are a hallmark of capitalist economies and the more “free market” oriented the economy, the greater the risk of booms and busts (Rajan and Zingales, 2003). While stock market crashes can exact a terrible toll in terms of human suffering (e. g. nemployment and lost retirement savings), they can also provide benefits by triggering investigations that lead to improvements in the financial system. Like prior boom-bust cycles, the recent U. S. stock market crash resulted in numerous inquiries into massive corporate misconduct, a public exercise that can help reduce the pain associated with future crashes. While understanding the underlying factors that cause financial failures is the healthy part of post-mortem investigations into economic downturns, these investigations can also degenerate into politically-motivated “witch hunts” that have unhealthy side-effects.
For example, the large number of recent accounting scandals doubtlessly created political pressure for legislatures to act quickly in implementing reforms, a setting in which policy-makers are prone to “rush to judgment” and in the process pass legislation that does not necessarily address the real problems. Such legislation can have unintended consequences that result in more harm than good and this may well be the case with the Sarbanes Oxley Act of 2002 (SOX).
SOX was constructed in an environment where the stock market was plunging and nearly every day seemed to bring the announcement of a new accounting scandal on the front page. These events came to a head with the revelations about Enron, just a few months prior to mid-term congressional elections, giving politicians an added incentive to act quickly. The sense of urgency surrounding the passage of SOX contributes to the likelihood that the provisions are not well-reasoned (Romano, 2004). Public criticism of the auditing profession following economic downturns and revelations of accounting scandals has a long tradition in the U.
S. For example, following the economic malaise of the early 1970’s and the stunning collapse of Equity Funding in 1973, the Committee on Auditor Responsibility (a. k. a. the Cohen Commission), called for sweeping reforms in the auditing industry and explicitly referred to “… the mounting disclosures of corporate scandals… ” to motivate its call for reform.  Further, the increasing globalization of the U. S. economy is likely to cause this type of intense scrutiny of the auditing profession to become even more frequent in the future. This is because globalization has caused product and capital markets in the U.
S. to become progressively more “free-market” oriented and if this trend continues, then boom-bust cycles and the public inquiries that follow such cycles are likely to increase (Rajan and Zingales, 2003). Thus, one effect of globalization of the U. S. economy is that it is likely to intensify the public pressure that is currently focused on criticizing the auditing profession. The good news and the bad news One consequence of the recent scandals is a renewed respect for the importance of auditing in maintaining credible financial markets, and its role in the process of corporate governance.
But if the good news is that there is a greater awareness of the importance of auditing, the bad news is that auditing is under attack. In spite of the fact that actual audit failure rates are close to zero (Francis 2004), popular opinion seems almost unanimous in concluding that the auditing profession is broken, to the extent that regulators have even taken away the profession’s ability to set its own standards by setting up the Public Company Accounting Oversight Board (PCAOB).
However, many of the accusations that have been leveled against the auditing profession are based on anecdotes or shallow simplifications. As researchers our differential advantage is that we are trained in the scientific method and as such should be able to bring more rigor to inquiries about what is wrong with the system. Audit “researcher” independence In calling on academics to help “fix” the auditing profession we feel it is important to recognize that there is an inherent threat to our independence when we investigate the auditing profession.
This threat arises because the auditing industry hires our students, makes donations to our departments and schools, funds Professorships and chairs, gives us subjects for experiments and proprietary data, and hires us as expert witnesses. All of these factors create a temptation for auditing researchers, referees, and journal editors to adopt a sympathetic view to the profession, and while such sympathy might be driven by a rational fear of “biting the hand that feeds us” succumbing to this sympathy would seriously erode our intellectual integrity.
If auditing researchers become apologists for the auditing profession then we are doing the profession, our students, society and ourselves, a huge disservice. The remainder of the paper primarily addresses some of the issues about auditing that have been questioned as result of the recent events, either directly or indirectly. While some of these issues and questions have been around for a long time, others seem to reflect a recently evolving understanding of the role and responsibilities of auditing and the profession.
We attempt to provide our insights regarding these issues along with a brief discussion of what we already know from the extant literature about these issues. We then provide our thoughts on what we believe to be the unanswered questions in the area and potential avenues to pursue to address those questions. INSTITUTIONS THAT AFFECT AUDITING The recent criticism of the auditing profession essentially boils down to issues of audit quality, and audit quality is affected by a number of basic economic institutions that define the nature of auditing practices.
The primary institutional features of accounting pertain to the authority for standard-setting and for the enforcement of auditing standards. However, incentives created by the legal system are also important in shaping auditing practices, as are regulatory agencies such as the Securities and Exchange Commission. Is the auditing profession unable to regulate itself? Two important institutional features of auditing are the authority to write auditing standards and the policing or enforcement mechanism to ensure that auditing standards are being properly implemented.
Since the late 1940s when the first auditing standards were issued, the accounting profession has controlled both the standard-setting process and the enforcement process. Generally accepted auditing standards (GAAS) define the objective of an audit and, by implication, the auditor’s responsibilities and standard of due care. GAAS also provides general guidance for planning, implementing and reporting on the outcome of audits. Given the broad nature of GAAS, there is a fair degree of latitude in how accounting firms implement GAAS.
That is, there is no single correct audit approach to the planning, gathering and interpretation of audit evidence that culminates in the audit report. The enforcement process has also been controlled by the accounting profession through professional self-regulation. Prior to the late 1970s, enforcement was mainly focused on individual CPAs and was closely linked to licensing issues (State Boards of Accountancy). Under increasing pressure to improve quality control, the AIPCA created the Division of Firms in 1978 to provide stronger monitoring mechanisms, and the AICPA began issuing specific standards on quality control practices.
While membership was initially voluntary, all of the large national firms belonged to the SEC Practice Division, and many smaller accounting firms elected to join the Private Company Practice Division. A number of important control mechanisms were implemented including outside peer reviews and engagement partner rotation. In the late 1980s, the SEC made membership in the SEC Practice Section a mandatory requirement for accounting firms practicing before the SEC. In effect, the “voluntary” self-regulation program was made mandatory but it was still operated by the AICPA as professional self-regulation.
SOX has undone over 50 years of professional self-regulation by removing from the accounting profession the authority to set standards and the authority to monitor the application of standards (enforcement). Now, both standard setting authority and the monitoring and enforcement of standards are the province of a quasi-governmental agency, the PCAOB. The PCAOB is funded independently of the SEC, but the SEC has regulatory oversight of the PCAOB including the appointment of board members. Why did this happen and how is it related to Enron’s collapse and the post-Enron criticism of auditing?
While speculative, we believe one critical trigger occurred when Deloitte Touche issued a “clean” peer review report on Arthur Andersen in December 2001, just a few weeks before Andersen publicly announcing that it had shredded documents related to the Enron audit. The credibility of the entire accounting profession was immediately in doubt along with the credibility and integrity of the profession’s self-regulation program. One also needs to remember that the profession lobbied hard against various SEC initiatives and won a major victory against the SEC’s 2000 proposal to proscribe nearly all nonaudit services.
Even though Arthur Andersen was the main villain in the Enron affair, the SEC – lead by Arthur Levitt — was at war with the Big Five accounting firms during much of the late 1990s. Indeed, the SEC has long voiced serious objections to the provision of nonaudit services (formerly known as management advisory services, or simply MAS), with the SEC’s first public objections dating back to at least 1957 (Panel on Audit Effectiveness, 2000). Congress itself has also been a long-time critic of nonaudit services, beginning at least with the Metcalf Subcommittee released in 1977 that effectively called for curtailment of most nonaudit services.
A concerted effort at self-regulation by the profession staved off a definitive proscription of nonaudit services and after the deregulatory environment of the Reagan administration the issue faded into the background until Levitt’s recent unsuccessful attempt in the late 1990’s to ban non-audit services. Thus, given this long history of bitter conflict between the profession and both the SEC and Congress over nonaudit services, it should come as little surprise that the profession was collectively “punished” by Congress following Enron’s failure.
If we assume that SOX was drafted with significant input from SEC staffers, it strikes us that at least some portion of its harshness towards the auditing profession may well contain an element of payback for the profession’s prior campaigns against the SEC and Congress. The profession’s success in 2000 on nonaudit services in particular may well have partially sowed the seeds for its destruction in 2002.  Over the past 60 years, periodic crises have pushed the accounting profession with respect to both standard-setting and en orcement but the basic self-regulatory structure has remained intact. A case can be made that these changes are mainly positive and cost-effective in improving audit quality. However, SOX is radically different because it has created a fundamentally new institutional structure for setting and enforcing auditing standards. We have moved from a self-regulatory institutional arrangement by the profession under SEC oversight, to explicit control over the accounting profession by a quasi-governmental agency.  There are two obvious research directions here, one looking back, and the other forward.
Looking back, the question is whether the “old regime” was really broken and in need of fixing? While there is little research evidence suggesting that this is the case, there is also no smoking gun that the self-regulation approach to standard-setting and enforcement had failed. The final report of the Panel on Audit Effectiveness (2000) provides some evidence that things were decidedly not broken, although this report is viewed by some with skepticism as it was undertaken for the AICPA’s Public Oversight Board and thus potentially biased in its conclusions.
On the enforcement side, a recent working paper by Hillary and Lennox (2004) reports evidence that peer reviews reports may have been effective, although there were very few “non-clean” reports, which raise questions about possible under-detecting or under-reporting of quality control problems. In contrast, the first round of PCAOB inspection reports for the Big 4 accounting firms were recently issued, and it appears that these new inspections are going to be more rigorous and critical than the old peer reviews performed under self-regulation.
So, looking forward, the important overriding research question is whether SOX improves audit quality, and if so, at what cost? Is the legal environment broken? The legal system is another institution that affects auditor behavior and, by implication, audit quality, through the standard of care auditors must meet to legally satisfy audit responsibilities and avoid litigation. Since the legal system explicitly defines the auditor’s legal liability and standard of care, it also influences auditing standards, defining the objectives of auditing, and the auditor’s professional responsibilities.
The auditor’s liability is affected by common law and statutory law. The auditor’s common law liability to third parties has changed over time, with periods of expanding and contracting exposure based on court cases and legal precedents. Statutory responsibilities were initially defined by the Securities Act of 1933 and the Securities Exchange Act of 1934, and were largely unchanged until The Private Securities Litigation Reform Act of 1995, which reduced the auditor’s liability exposure under federal securities law.
There are essentially two ways to study the affect of legal systems on auditor behavior and audit quality. One way is to use a cross-country comparison of alternative legal systems, and this is discussed later in the paper. The other approach is a single-country study, which undertakes a comparative analysis at different points in time to examine the effects of different legal regimes. For example, consider two studies examining auditor behavior before and after the 1995 change in statutory law which reduced auditors’ legal liability.
Lee and Mande (2003) document that abnormal accruals of Big 4 clients increased after 1995, which is consistent with clients having greater discretion to manage earnings; and Francis and Krishnan (2002) find that Big 4 auditors were less likely to issue going concern audit reports after 1995. Both studies are consistent with auditors being “less tough” on clients following the 1995 liability changes due to the reduced likelihood of litigation for poor quality audits. The benefit of going soft on clients is a greater likelihood of client retention and possibly higher fees.
If litigation was significantly reduced after 1995, then the cost of “going soft” on clients is also reduced because it reduces the likelihood that the auditor will be sued for failing to detect or report material misstatements. Another study which indirectly examines auditor behavior is Basu (1997). He finds that ‘earnings conservatism’ increases in periods of expanding auditor liability and decreases in periods of contracting auditor liability, which is consistent with the idea that auditor incentives change under different legal regimes.
Auditor incentives are also affected by punishment for misconduct by regulatory agencies, particularly the SEC. The SEC’s enforcement actions have been examined in several studies (e. g. , Feroz, Park and Pastena 1991, Dechow, Sloan and Sweeney 1996; Beneish 1999) which are largely descriptive. However, Wilson and Grimlund (1990) examine the impact of enforcement actions on the auditor’s retention of clients and the ability to attract new clients.
They find evidence that accounting firms’ reputations were tarnished by enforcement actions and in the short term suffered in terms of their ability to retain and attract clients. While the above studies provide evidence that legal institutions “matter”, they provide no guidance on whether the legal system is likely to result in the “optimal” level of audit quality. SOX left untouched legal liability issues and the potential deleterious effects of unlimited legal liability exposure on private sector auditing.
After the demise of Enron we are down to only four large accounting firms, and each one has majoring pending litigation. Can we lose any more firms and have auditing remain a viable private sector activity? Of course this begs the larger question of whether private sector auditing (with regulatory oversight) is the optimal institutional arrangement in the first place. Other countries such as Canada and Australia appear to have credible auditing without imposing the very real risk of accounting firm failure which can probably only occur in the United States.
If auditing is to remain a private sector activity, there is a real need to better understand the role of legal liability in achieving audit quality and whether the kind of extreme litigation exposure we have in the U. S. is really necessary to achieve an appropriate level of audit quality. While the questions of optimal legal systems may never be answered, we can still investigate these issues and comparative cross-country research seems a particular attractive research approach in trying to answer these questions (as discussed later).
In sum, given the importance of the legal and regulatory environment in shaping auditor incentives, we believe there is a real need for future research to build on the above studies to better document the role of changing institutional environments on auditor incentives and behavior. These studies may also give us insight to social welfare questions concerning the optimal amount (and quality) of auditing, discussed later in the paper. ENGAGEMENT-SPECIFIC CHARACTERISTICS THAT AFFECT QUALITY
Engagement-specific characteristics can potentially affect audit quality in ways that are separate and distinct from the institutional features discussed above, although if audit quality is adversely affected it is possible it will ultimately be discovered through monitoring/enforcement programs, or litigation against the auditor. Do auditors lack independence? The engagement characteristic addressed in SOX that garnered the most attention is the auditor’s provision of nonaudit services and the potential for these services to compromise the auditor’s objectivity and independence.
Again it is instructive to note that there was no smoking gun driving public policy on this point. While it is true that Arthur Andersen earned about the same amount from nonaudit services as from audit services in its dealing with Enron, it turns out that this is the rule, not the exception (e. g. , Frankel, Johnson and Nelson 2002). While the dollar amount was large ($26 million), there is no evidence whatsoever that nonaudit fees and services are a source of Andersen’s problems in the audit of Enron. So why did SOX place so much emphasis on this particular aspect of the auditor-client relationship?
As previously noted, one partial explanation is the profession’s history of conflict between Congress and the SEC over nonaudit services, most recently in 2000, and a kind of payback by Congress and the SEC in drafting the SOX legislation in 2002. The evidence in Frankel et al. (2002) indicates that companies paying higher levels of nonaudit fees are more likely to have abnormal accruals and more likely to meet or beat analysts’ earnings forecasts, which implies more discretion to manage earnings, and this study was cited and relied upon in drafting the SOX legislation. However, numerous studies following Frankel et al. 2002) find their results are extremely sensitive to their research design, sample selection, and model specification (Chung and Kallapur, 2003; Ashbaugh, LaFond and Mayhew, 2003; Francis and Ke, 2003; Larcker and Richardson 2004; Reynolds, Deis and Francis 2004). Thus, the vast majority of research on nonaudit fees fails to find any evidence that they impair auditor independence. Even though this topic remains important, given the flurry of research in the past few years, it may be hard to design a study which adds important new evidence to what we already know about nonaudit services.
However, at a more conceptual level, regulators and critics have never been clear about the exact nature of the concern over nonaudit services. One possibility is a concern that the magnitude of nonaudit fees causes the auditor to become financially dependent on the client. If this is the concern, however, then “total fees” paid to the auditor is a more relevant measure of the threat to auditor independence than the nonaudit component of total fees.
Interestingly, the SEC did not raise the issue of financial dependence when it proposed a ban on nonaudit services in 2000. Some studies have examined total fees (Ashbaugh et al. 2003; Chung and Kallapur 2003; DeFond, Raghunandan and Subramanyam, 2002) or a surrogate for total fees based on clients’ sales revenue (Reynolds and Francis 2000) and find no evidence that relatively larger clients get favorable treatment by the auditor in their financial statements or audit report.
On the contrary, there are even studies that find evidence that large audit fees are associated with improved audit quality and a greater propensity for auditors to issue qualified reports (Reynolds and Francis 2000). We believe that disentangling the effects of audit from nonaudit fees, and examination of total fees is an important but underdeveloped area of research that has been overshadowed by the heavy emphasis on nonaudit services and fees.
Another issue that is important to keep in mind is the appropriate level of analysis in evaluating the economic bond created by fee dependence. Francis et al. (1999) and Reynolds and Francis (2000) argue that incentive effects are better analyzed at the local office level because the economic importance of (large) clients is more salient at the office level. Audits are administered by an engagement partner based in a specific practice office typically in the same geographical locale as the client’s headquarters.
The economic impact of losing an important client, or gaining a new client, is far more important to the engagement partner and the office’s well-being than to the firm as a whole. To date there have only been a few studies using the office-level of analysis. Reynolds and Francis (2000) examine client accruals and auditor going concern reports at the office-level of analysis and conclude that auditors treated larger clients more conservatively (the opposite of what independence impairment would predict). Craswell, et al. 2002) examine office-level audit reports and find no evidence that client size affects auditor reporting decisions. Chung and Kallapur (2003) use the office level framework as one of their tests and find no evidence that fees paid to auditors affect client accruals. Finally, Krishnan (2004) examines Arthur Andersen’s Houston clients and finds that the earnings of these clients are less conservative using the Basu (1997) framework than the earnings of other Andersen clients, suggesting that there were systemic problems in the Houston office but not the rest of the firm.
We believe the office level of analysis has great potential for future research to expand our basic understanding of auditor behavior and audit quality. In addition, it is possible in some countries to push the “local analysis” down to individual engagement partners. In Australia and Taiwan, the name of the engagement partner appears in the audit report and so it is possible to construct individual partner “clienteles” and measure the relative importance of individual clients to partners.
Rather than financial dependence, another possibility is that the “nature” of nonaudit services impairs auditor independence. While we believe this is a very important issue, there is very little research in this area and therefore we know little about this potential threat to independence. The idea here is that certain kinds of nonaudit services may subtly change the auditor-client relationship to more of a partnership or alliance that could undermine the more skeptical role expected of the auditor.
However, we currently do not know if this is more likely to be true for some kinds of nonaudit services than others. SOX assumes that information system design and implementation is an example of this kind of compromising nonaudit service which is why it was prohibited. However, we know of only one working paper that addresses this issue. Abbott, Parker, Peters and Rama (2004) find that audit committees with governance characteristics that are likely to make them more effective are less likely to outsource routine internal audit services to their incumbent auditor.
More effective audit committees, however, are more likely to outsource non-routine services to their external auditors where the incumbent auditor is expected to have an economic advantage over alternative sources. They conclude that there results suggest that effective audit committees are likely to appropriately monitor the delegation of nonaudit services to the incumbent auditor, and are consistent with SOX being overly heavy-handed in placing such a comprehensive ban on non-audit services.
At this particular time tax services are emerging as the largest source of nonaudit service revenues and there are legitimate concerns that aggressive tax planning may undermine the integrity of the financial statements. The only published paper to date on this topic that we are aware of is Kinney, Palmrose and Scholz (2004) who look at the association between various categories of nonaudit fees and the likelihood of an earnings restatement. Earnings restatements imply the possibility that past audits were of low uality and the question is whether the amount of fees paid for nonaudit services, or the type of nonaudit services, is systematically related to restatements. They find little evidence of a consistently positive relation between audit fees or most nonaudit services and restatements, but they do find a consistently negative association between tax service fees and restatements. Thus, their findings suggest that particular nonaudit services such as taxation may actually reduce the incidence of earnings management and improve earnings quality.
In summary, we believe that the SOX provision that bans most nonaudit services is at best misguided, and at worst politically-motivated. Having said this, we also believe that while the existing research finds little evidence that the magnitude of nonaudit service fees impair independence on average, there are many important questions not yet addressed by researchers in this area. In addition, most studies use linear models, and fee dependence may occur only when fees reach some large threshold in which case some kind of non-linear formulation may be more appropriate.
Thus, going forward, we believe researchers should look more carefully at issues such as total fee dependency, whether the personal relationships created by nonaudit services are likely to threaten independence, and whether there are contextual issues (such as the firm’s overall governance environment) that are important in explaining whether auditor independence is impaired. Does auditor tenure or the “revolving door” impair auditor independence?
Two other engagement-specific characteristics have been examined in the research literature that are also addressed in SOX: auditor tenure and the revolving door. The arguments regarding long auditor tenure have been expressed off and on for many years, and underlie the call for mandatory rotation. This argument suggests that long auditor tenure may create entrenchment and lead to what Bazerman, Loewenstein and Moore (2002) calls self-serving bias, both of which can reduce auditor objectivity and result in lower audit quality.
However, short tenure may also lower audit quality due to a learning curve effect, and presents an argument against mandatory rotation. Indeed, if audit quality improves as a function of auditor tenure then it argues for mandatory “auditor retention,” and is perhaps consistent with the assertions that auditor-switching is opportunistic. Despite the concerns expressed in SOX, there is little or no evidence that long tenure impairs audit quality (Myers, Myers and Omer 2003).
On the contrary, several studies find evidence consistent with short auditor tenure impairing audit quality (Johnson, Khurana and Reynolds 2002), and one study even finds evidence consistent with longer auditor tenure actually reducing earnings management via so-called “cookie jar” reserves (Myers et al. 2003). While SOX stops short of mandating auditor rotation, it does require that the “lead or coordinating partner and reviewing partner” must rotate every five years. The SEC also explicitly instructed the GAO to study this issue further, and the GAO (2003) has subsequently concluded that there is not a good case for mandatory rotation.
Given the SEC’s concerns, and the very real possibility that mandatory rotation might again be proposed, we believe further research investigating the effects of short- and long-tenure on audit quality is important. SOX also attempts to put the brakes on the so-called “revolving door” policy in which accounting firm alumni take senior management positions in clients. Specifically, SOX imposes a one-year waiting period for audit firm employees who leave an accounting firm to become an executive for a former client.
As with nonaudit fees, the Enron-Andersen affair provided anecdotal rationale for this policy because many of accounting personnel at Enron were Andersen alumni, including Ben Glisan, an employee credited with instigating some of Enron’s most egregious deal-structuring schemes. In a note of irony, Andersen alum Sharon Watkins is the whistle-blower responsible for bringing Glisan’s unscrupulous deals to light. While there is very little research in this area, two recent studies do indicate that auditors may be less skeptical towards their clients that have highly placed alumni.
Menon and Williams (2004) report that abnormal accruals are larger for such clients, and Lennox (2004) finds that auditors are less likely to issue going concern opinions for such clients. We believe this continues to be an important issue and further research is desirable because of the chilling effect it may have on accounting firms. Specifically, one of the most appealing aspects of working in large accounting firms, and which may attract talented individuals, is the high-level outplacement opportunities to clients. With this fringe benefit curtailed, accounting firms may ttract less capable individuals which in the long term may reduce audit quality. THE ROLE OF AUDITING IN CORPORATE GOVERNANCE A major portion of the blame for the recent alleged accounting and auditing failures was not leveled at the auditing profession, but rather at audit committees of boards of directors. Targeting audit committees explicitly recognizes that auditing is a key component of firm’s overall corporate governance environment and a few auditing researchers were examining audit committee issues even prior to SOX (e. . Klein 1998; Carcello and Neal 2000). The emphasis that SOX puts on audit committees, however, suggests that auditing researchers can do a great deal more to understand the role of auditing in governance. Luckily, we have a large body of literature in the finance and economics areas to build upon and to help us sort out some of the fundamental problems that arise when empirically investigating governance issues related to boards of directors.
We believe that as researchers we can use this existing research to leverage our strengths, which include our deep understanding of auditing institutions. Are audit committees effective? A potential stumbling block in understanding how audit committees impact governance is that there are no theories that explain why boards of directors exist in the first place (Hermalin and Weisbach, 2003), much less why audit committees exist.
The absence of theoretical direction presents obvious problems in investigating questions about what makes audit committees effective. Never-the-less, since we observe that boards of directors and audit committees do exist, and that they potentially have significant influence over the audit process, we believe that lack of theory should not stand in the way of empirical studies that examine the links between audit committees and auditing-related outcomes.
While theory does not help, there are many practice-related sources that describe the role and objectives of audit committees (e. g. Blue Ribbon Committee 1999; Coopers and Lybrand 1999). These sources suggest that the primary role of the audit committee is to oversee the financial reporting process with the objective of ensuring high quality financial reporting (SEC 2003). We can infer from the significant changes mandated in SOX, however, that Congress does not believe that audit committees have been fulfilling this role.
The new audit committee rules in SOX are: (1) all firms must have an audit committee that is composed entirely independent of management; (2) the firm must state whether the audit committee contains a so-called “financial expert” – and if not why not; (3) the audit committee is responsible for appointing the outside auditor; (4) the firm must provide outside counsel and other advisors that the audit committee deems necessary to fulfill its duties; (5) the audit committee must implement procedures to receive and investigate employee complaints about accounting policies and practices; and (6) the audit committee must approve the purchase of nonaudit services that are not specifically prohibited by SOX. Will 100% independent members make audit committees more effective? The belief that 100% independent audit committees will improve audit committee effectiveness does not originate with SOX. At the recommendation of the Blue Ribbon Committee (1999), all of the major U. S. tock exchanges in 1999 began encouraging their registrants to have 100% independent audit committee members, but gave the board the discretion to appoint inside directors when appropriate, and also exempted small issuers. SOX, however, provides no exceptions and mandates that companies with less than 100% audit committee members be de-listed. This raises several obvious questions, such as: is governance improved by having 100% independent audit committees, and should exceptions be allowed? These questions can be empirically addressed, and we already have quite a bit of empirical evidence on the first question. The decision to require all of the audit committee members to be independent seems to follow from the conventional wisdom that independent directors are better monitors of management behavior than inside directors.
Contrary to this wisdom, however, several “meta-analysis” studies in the governance literature conclude that there is very little empirical evidence that the proportion of independent directors on the board (as a whole) is associated with outcomes that we would expect to find in better governed firms, such as better accounting performance or higher share values (Bhagat and Black 1999; Dalton, Daily, Ellstrand and Johnson 1998). Some studies even find evidence that in some settings firm value increases when inside managers are appointed to the board. The value of insiders on boards may be explained by the fact that insiders bring firm-specific knowledge to the board, and/or because some insiders have greater incentives to improve firm performance than outsiders (Rosenstein and Wyatt 1997). If outside directors have less to gain from the firm’s success, they will be more likely to shirk.
Finding benefits from having inside directors at the board level, however, does not necessarily mean that there are benefits from having inside directors on the audit committee, and we are fortunate to have a nascent but growing body of literature that provides evidence on audit committee composition with respect to independence. Many of these studies provide nice tests of the SOX mandate by specifically testing whether 100% independent audit committee membership is associated with good outcomes, such as higher market values, lower abnormal accruals (to capture earnings quality), and fraud or other financial misconduct. In a meticulously detailed review of 16 of these papers, Romano (2004) reports that these studies generally fail to find any significant association between 100% independent audit committees and the outcomes we would expect to be related to good governance. 4] While this research provides some evidence that governance is improved in the presence of a higher proportion of independent directors on the audit committee, the question raised by SOX is whether 100% is the desired proportion, and the majority of the empirical evidence resoundingly suggests that it is not.  If the evidence thus far is correct, and 100% independent directors on the audit committee do not improve governance, it is useful to contemplate why. As previously mentioned, it is possible that outside directors are more likely to shirk or that they lack the firm-specific expertise to choose the best auditor for the firm.
While some authors argue that outside directors have incentives to build reputations for being tough monitors (Fama and Jenson 1983), others argue that they may have even stronger incentives to build a reputation for not upsetting the CEO (Hermalin and Weisbach 2003). Future researchers (who will now have to rely on pre-SOX data for U. S. studies) can address these issues by investigating whether there are settings in which is it optimal to have inside directors on the audit committee, and what type of insiders are best. Before leaving this topic it is important to acknowledge that there are several problems with drawing conclusions from the “lack” of associations found in the audit committee literature (and much of the literature discussed in this paper). The obvious problems include the possibility of lack of statistical power, omitted correlated variables, and poor construct validity.
Lack of power may result in the studies where the samples are necessarily small, such as the fraud data used in Beasley (1996) and the restatement data used in Agrawal and Chadha (2005). Omitted correlated variables and construct validity are particularly problematic in papers that rely on abnormal accruals to surrogate for earnings management, since this measure is likely to capture the construct of earnings management with noise. Having said this, there are a large number of these studies that look at a broad array of variables across a large number of different samples and research designs. Thus, while we can never technically “accept the null” of no association, the consistent lack of evidence at some point must revise our priors on whether 100% independent audit committees are likely to be the optimal arrangement.
In addition to the above research design issues, Hermalin and Weisbach (2003) point out that endogeneity problems and assumptions regarding equilibrium plague all of the research examining director composition. Endogeneity is a problem because all of the variables of interest in these studies are potentially jointly determined. For example, even if we find that higher stock values are associated with firms that have 100% independent audit committees, this could either result from independent audit committees taking actions that improve performance, or it could result from the fact that better performing managers choose more independent audit committee members, perhaps because they have less to hide.
Equilibrium assumptions are important because if we assume that all firms are in equilibrium, then even if we find an association between 100% audit committee members and improved performance, we cannot say that mandating 100% audit committees would improve performance in all firms. Thus, auditing researchers that look at audit committee issues face some important research design issues that are inherent to the governance literature. Does financial expertise improve audit committee effectiveness? As with the SOX rule on audit committee independence, the requirement for audit committee expertise is also included among the recommendations made in the Blue Ribbon Committee (1999), and beginning around 1999 all of the major U. S. stock exchanges implemented the requirement that their member firms have “financially literate” audit committee members.
As with the audit independence rules just discussed, this raises the question of why Congress decided to make rules that are redundant to the rules already implemented by the exchanges. One possible answer is that legislators are under pressure from their constituents to take quick action, and implementing new rules (even redundant rules) gives the appearance of action. If constituents do not have incentives or the ability to discover that the rules are redundant, then the legislators will get credit for signing into law what is already required by the exchanges. Another possibility, which is at least as compelling, is that federal criminal statutes are more effective deterrents than exchange rules.
This explanation is consistent with Rajan and Zingales (2003), who argue that criminal statutes impose larger penalties on offenders, and that criminal prosecutors have greater incentives than exchanges to pursue offenders. Empirically addressing whether financial expertise improves corporate governance presents several difficulties, including the fact that the concept of “financial expertise” is not well defined. Complicating this issue, the initial SOX promulgations recommended a fairly narrow definition of financial expertise, the final rules had a much broader definition, and neither of these definitions quite capture the idea of “financial literacy” that is required by the major stock exchanges.
Further compounding this complication, proxy statements and press releases reveal limited information on director attributes that are likely to tell us about financial expertise, and the final SOX rules give a great deal of discretion to the directors in deciding whether a given audit committee member qualifies as an expert. This means that it is empirically difficult for researchers to use public information to accurately classify audit committee members as financial experts. Despite the above problems, the issue of audit committee expertise has been addressed in the auditing literature, although not to the extent that audit committee independence has been.
Unlike the evidence on 100% audit committee independence, however, most of the research in this area seems to suggest that financial expertise on the audit committee (where expertise is measured in a variety of ways) does improve corporate governance. Specifically, research in this area generally finds that audit committee expertise is associated with less earnings management, lower cost of debt, greater disclosure, fewer SEC enforcement actions, fewer restatements, and higher firm value. (e. g. , Andersen, Mansi and Reeb, 2004); Agrawal and Chadha, 2005; Bedard et al. 2004; Felo, Krishnamurthy and Soliere, 2003; McMullen and Raghunandan, 1996; Xie et al, 2003; DeFond, Hann and Hu, 2004).  It is also important to note that SOX stops short of actually requiring financial expertise on audit committees.
Rather, firms must report whether they have a financial expert on their audit committee and explain why if the answer is no. While this disclosure requirement obviously exerts pressure on firms to retain a financial expert, it also presumably means that the SEC is willing to consider good reasons why some firms should not have experts, and raises the obvious empirical question of whether some audit committees do not require a financial expert. Romano (2004) points out that allowing an “out” for firms without financial expertise also presents an irony, because it suggests the SEC got it completely backwards with respect to the independence and financial expert rules.
That is, while the SEC mandates 100% audit committee independence, the research generally finds no benefit from having 100% independent audit committee members; and while the SEC does NOT strictly mandate financial expertise, the research generally finds there are benefits from having financial expertise on the audit committee. Does audit committee governance complement or substitute for other governance mechanisms? An important issue raised in this literature is whether audit committee governance complements the existing governance mechanisms within the firm, or whether factors such as audit committee expertise and independence substitute for deficiencies in other governance areas. The report of the Blue Ribbon Committee (1999, p. 6, 7) argues that “audit committee performance relies on the practices and attitudes of the entire board” and that the audit committee is unlikely to be effective when the rest of the board is “dysfunctional. Thus, the BRC seems to believe that firms that already have strong governance on the overall board are more likely to benefit from good governance on the audit committee. This complementarity view is also consistent with empirical evidence, which seems to suggest that boards with stronger overall governance attributes are more likely to value the services of high quality audit committees (Klein 2002b, Beasley and Salterio 2001, and DeFond et al. 2004). While the research generally supports a complementary role for audit committee governance, we know relatively little about the particulars. For example, are some governance mechanisms (or sets of mechanisms) better at complementing audit committee governance than others? While DeFond et al. 2004) look at complementarity between financial expertise and a composite measure that captures multiple governance attributes, it is possible that some subset of these governance attributes are more important complementing audit committee expertise than others. Another question is whether complementarity and substitutability among governance factors are at least a partial function of non-governance factors, such the nature of the firm’s business or its financial condition? While understanding how governance mechanisms interact is important, prior governance research outside of the accounting area has done very little to address this issue. Thus, we believe that the dearth of research in this area provides an opportunity for auditing researchers to make a potentially important contribution to the broader governance literature. Are larger or more active audit committees more effective?
Two audit committee characteristics that are not addressed in SOX, but that have been addressed in a few academic papers, is audit committee size and activity. The number of studies examining these issues is fairly small and the results are mixed regarding whether audit committee size or activity (typically captured by the number of annual meetings) impacts governance. The motivation for looking at audit committee size is that relatively larger audit committees are consistent with boards committing more resources to improve financial reporting quality. Following the recommendations of the Blue Ribbon Committee (1999], the NYSE and NASDAQ now require their registrants to have a minimum of three directors on their audit committees. Anderson et al. 2004) find that bond yield spreads are negatively related to audit committee size, consistent with larger audit committees improving financial reporting quality, while Bedard, et al. fail to find that audit committee size reduces earnings management. The BRC also suggests that audit committees should meet at least four times per year in order to provide the type of interaction and deliberation necessary to fulfill their duties. While few studies address this issue, there is some evidence that more frequent meetings are indeed associated with better-governed firms. For example, McMullen and Raghunandan (1996) find that the audit committees of firms with SEC enforcement actions or earnings restatements are less likely to have frequent meetings; and with Abbott et al. 2004) find that when frequent meetings are combined with independence they are associated with a lower incidence of fraud. Because audit committee size and activity are not directly addressed in SOX, there is some concern that researchers will neglect these issues going forward. Given the paucity of research in these areas, however, and the suggestion from the existing research that these factors may indeed be important, we think this would be a mistake. Therefore we encourage researchers to continue investigating these topics. What else impacts audit committee effectiveness? We are not aware of any auditing related research regarding four of the six new audit-committee SOX promulgations listed earlier.
These are: (1) the audit committee is responsible for appointing the outside auditor; (2) the firm must provide outside counsel and other advisors that the audit committee deems necessary to fulfill its duties; (3) the audit committee must implement procedures to receive and investigate employee complaints about accounting policies and practices; and (4) the audit committee must approve the purchase of nonaudit services that are not specifically prohibited by SOX. While these new rules might seem reasonable at first blush, it is important to keep in mind that they are not costless, and that one cost can be significant unintended negative consequences. Therefore, it is important to assess whether the regulatory solution is likely to solve the problem.
For example, in the case of the rule requiring the independent audit committee to hire the auditor, it is possible that inside management may possess important information about the company that makes them better at assessing the auditor’s ability to efficiently service the firm. This is particularly likely to be the case when the firm is very innovative or in a fast changing industry. Indeed, deriving benefits from having insiders participate in selecting the auditor is consistent with the research that suggests that insiders on the board, at least in some settings, are associated with better governed firms (Bhagat and Black 1999; Dalton et al. 1998).
Similarly, while the rule that requires establishing a system for employees to voice their concerns about accounting treatments seems appropriate, especially in light of the role that whistle-blowing has played in uncovering many recent accounting scandals, the potential unintended consequences of such a system in a large organization are unclear. For example, such a system might degenerate into a (costly) mechanism for seeking retribution against management by disgruntled employees. Given the potential costs of the above new rules, and the fact that we know virtually nothing about their potential effects, we encourage research that addresses there expected or realized efficacy. AUDIT QUALITY AND OTHER ACCOUNTING FIRM CHARACTERISTICS Does size matter?
As unpopular as it might be to say right now, the overwhelming evidence points to higher quality audits by larger accounting firms (see Francis 2004 for a review). Relative to the size of their clienteles, smaller accounting firms have higher litigation rates and are the subject of more SEC enforcement actions. Smaller accounting firms report less conservatively (issue fewer nonclean audit reports), and their clients are more likely to have abnormal accruals which is suggestive of more aggressive earnings management. This research has traditionally been caste in terms of auditor differentiation, and was not pejorative with respect to smaller accounting firms. The assumption is that all accounting firms, big or small, can potentially perform a competent audit in accordance with GAAS.
So the argument is that some accounting firms have voluntarily elected to be of higher quality and there are clienteles that demand such audits. However, it is also possible to interpret this literature as indicating the smaller firms systematically produce lower quality audits. While the dominant Big Four accounting firms audit most companies, there are still a surprisingly large number of smaller companies audited by local and regional accounting firms, and there is some recent evidence the Big Four are more of their dropping riskier clients (post-SOX) which are mainly being picked up by second-tier the national firms (Public Accounting Report, September 30, 2004).
Further, there is also evidence that a non-trivial number of client firms are delisting because they can no longer afford the regulatory burden imposed by the Section 404 provisions of SOX. So the research question of importance here is whether differential audit quality means larger firms supply higher quality audits or alternatively if smaller firms produce unacceptably low quality audits. If the answer is the latter, then one public policy implication is that smaller accounting firms should not be allowed to practice before the SEC. Pushing the same line of reasoning, the same questions pertain to the quality of audits performed by industry experts.
If audits by industry experts are of higher quality, as evidenced in recent studies (Balsam, Krishnan and Yang 2003; Krishnan 2003), should the SEC mandate industry specialization as an audit requirement? The broader issue here is the whether differential audit quality should be allowed, and whether there should be lower-cost providers available in the audits of publicly-listed companies. As we know from a long stream of research, audit fees of large accounting firms are higher than the fees of smaller accounting firms, and recent evidence also indicates that the fees of Big Four industry experts are relatively higher than the fees of other (non-expert) Big Four firms (Craswell, Francis and Taylor 1995; Francis, Reichelt and Wang 2005).
The above discussion raises a multitude of intriguing questions that might be addressed by future researchers. For example, what would be the incremental cost of mandating uniformly higher quality, and what would be the effect on market concentration and efficiency? Would such a move increase market concentration, reduce competition, and create monopoly pricing power by accounting firms? These are not easy issues to address, but they seem central to any discussions related to improving audit quality. Does higher quality auditing lead to higher quality earnings? Arthur Levitt’s famous “numbers game” speech and the rhetoric preceding SOX make it clear that Congress and the SEC believes that higher quality auditing will lead to higher quality earnings.
The academic research that attempts to link audit quality with earnings quality, however, makes it equally clear that whether this is true or not depends upon your definition of both audit quality and earnings quality. Several studies have attempted to directly link audit firm characteristics with earnings quality measures. The first papers in this area, Becker, DeFond, Jiambalvo and Subramanyam (1998), and Francis, Maydew and Sparks (1999) look at the association between auditor size and the behavior of abnormal accruals. Both papers find that the clients of larger auditors tend to report less income-increasing abnormal accruals, and less variability in their abnormal accruals, consistent with larger auditors being more conservatism and allowing less accounting flexibility. Subsequent to Becker et al. (1998) and Francis et al. 1999), numerous studies confirm that larger auditors are associated with clients that report smaller and lower variance abnormal accruals (Chung and Kallapur 2003; Frankel et al. 2002; and many others). In addition, studies that examine other outcomes, such as restatements, qualified opinions, and how earnings are impounded in stock prices, corroborate these results by finding consistent evidence that the clients of larger auditors report more conservative earnings. While these findings in the prior literature have generally been interpreted to mean that larger auditors are associated with higher audit quality, and hence higher earnings quality, this conclusion is not necessarily warranted.
While creditors are likely to prefer conservative earnings because it may lower default risk, stock market participants are expected to prefer unbiased earnings. In fact, allowing managers greater income-decreasing discretion may even exacerbate earnings management by giving managers permission to create “cookie-jar” reserves. Similarly, it is not clear that constraining management’s flexibility in reporting discretionary accruals improves earnings quality from the perspective of either creditors or stock market participants. For example, Subramanyam (1996) finds that discretionary accruals convey information about future cash flows, suggesting that constraining management discretion will reduce earnings informativeness.
While difficult to empirically address, we encourage future researchers to investigate the questions that are raised by the above discussion. For example, is the current conservative bias in auditing optimal? What is the “cost” to equity markets (if any) of the conservative bias in auditing? What would be the effects of reducing the conservative bias in auditing? What can we infer about auditor behavior from reported financial statements? As discussed above, much of the prior literature uses abnormal accruals to make inferences about auditor behavior. We know, however, that our traditional measures of abnormal accruals are noisy and potentially performance-biased.
In addition, another limitation of these studies is that their evidence is necessarily indirect, making it difficult to convincingly rule out alternative explanations for the findings. Researchers have attempted to improve on the literature that traditionally uses abnormal accruals by using alternative measures designed to capture earnings management, such as earnings restatements (e. g. , Palmrose and Scholz 2004; Myers, Myers, Palmrose and Scholz 2004). The biggest advantage of using restatements are that they provide more direct evidence that the auditor failed to either detect or report an accounting treatment that is inconsistent with GAAP. Limitations with using restatements, however, are that they are relatively rare, and that researchers are unable to identify the management manipulations that were not restated.
That is, researchers asymmetrically identify only the misstatements that are eventually restated, and to the extent that undiscovered misstatements occur, it limits the ability to draw definitive conclusions. Another alternative to using abnormal accruals to infer auditor behavior is to use the propensity of the auditor to issue a qualified opinion (DeFond, Raghunandan and Subramanyam 2002; Reynolds and Francis 2000). An advantage of this approach is that the audit opinion is the only channel through which auditors are able to directly communicate with shareholders. Thus, unlike financial statement-related measures, this construct is unambiguously captures an outcome that is explained by auditor behavior.
Also, relative to abnormal accruals and restatements, we have a better understanding of how to model the auditor’s propensity to issue a qualified audit report, and have the advantage of drawing on an extensive body of research that goes back nearly 25 years (e. g. Simunic 1980; Mutchler 1984). A limitation with using qualified reports is that they are relatively unusual events. While all of the above measures have both strengths and weaknesses, we believe the strengths far outweigh the weaknesses, and we encourage researchers to continue using these measures as a means for helping to understand the influence of auditors on financial reporting. One appealing method for mitigating the weaknesses of the individual methods is to use multiple measures in a given study and look for consistency across the measures.
We also encourage researchers to explore additional measures of audit quality in future research, such as management’s propensity to meet or beat earnings benchmarks (Burgstahler and Dichev 1997), and measures that do not focus exclusively on earnings, such as the “bloated balance sheet” measure used in Barton and Simko (2002) How would “fair value-based” and “principles-based” accounting impact auditors? The FASB recently issued proposals suggesting that U. S. GAAP adopt more fair value-based measurement of assets and liabilities, and a greater emphasis on so-called principles-based accounting standards (FASB 2002 and 2004). The FASB believes that both proposed changes will result in managers providing investors with information that is more decision-useful. Aside from the obvious questions of whether fair-value and principles-based accounting will result in managers producing more relevant information, it is critically important to consider how these changes are likely to impact auditing and audit quality.
Many scenarios seem possible and we find it difficult to make firm predictions about whether the net effect will be improved reporting and auditing, impaired reporting and auditing, or no change. While speculative, it seems that both of the proposals are likely to allow greater discretion to both management and the auditor (although the term “principles-based” is not yet clearly defined, so it is premature to make definitive judgments at this point). If managers have incentives to overstate their financial performance, and auditors have incentives to encourage conservative reporting, then one outcome might be more frequent disagreements between management and the auditor.
On the other hand, if greater discretion allows the auditor to more easily “defend” management’s choices in court, then auditors may lose some (or all? ) of their conservatism. Conversely, if greater discretion allows the auditor to more easily “defend” their conservative bias, then financial reports may become even more conservative. In any event, given the seeming inevitability of these important newly proposed changes, we strongly encourage auditing researchers to address questions related to their expected effects on auditor behavior and how these changes impact the consequences for financial reporting. Given that GAAP in some countries tends to be more fair value-based and principles-based that in the U. S. non-U. S. data may be useful in addressing these questions. What can we learn from cross-country comparisons? Most auditing research studies are single-country studies. These are important and tell us a great deal. However, a very promising area is comparative cross-country research. The appeal of this research approach is that it allows us to test the effects of different institutions, legal systems and auditor liability, and regulatory systems on auditor behavior and audit quality. With the advent of global data bases such as Global Vantage and Worldscope it has become much easier to do comparative empirical research. The appeal is obvious.
We can “test” the effects of alternative institutional arrangements on auditing, by using differences across countries to create a natural experiment (e. g. , Choi and Wong 2002; Francis, Khurana and Pereira 2003). The downside, as Bushman and Smith (2001) point out is that there are country-level factors that need to be controlled in order mitigate correlated omitted variables problems. While this research stream presents major challenges, such as controlling for potentially confounding effects, its considerable strength is its ability to allow us to test for differences in institutions. Thus, because the potential rewards from this research are so great, we strongly encourage researchers to continue to pursue research in this small but growing area. What is the “Optimal” level of Auditing?
A critical and unaddressed question that underlies all research in audit quality is “what is the optimal level of audit quality? ” While comparative cross-country research can describe the effects of differences in auditing practices across different institutional settings, it cannot tell us the optimal amount of auditing or audit quality in terms of social welfare. Even so, we may be able to make evaluative judgments that more (less) auditing, at the margin, in a given case, is more (less) desirable because of the associated benefits (costs). For example, section 404 of SOX is anticipated to increase audit fees by 50-100%. Are there measurable benefits in audit quality that would justify these fee increases?
A real challenge for policy-makers, as well as researchers, is getting a better handle on the economic value of auditing to the economy relative to the cost it imposes. CONCLUSION While it’s highly unlikely we will ever return to the pre-SOX world of self-regulation by the accounting profession with SEC oversight, there is a greater need than ever for objective scientific evidence to guide public policy-making in auditing. To this end there are encouraging signs that the Public Company Accounting Oversight Board (PCAOB) is reaching out to the academic community. However, it remains to be seen if policy-makers will genuinely seek scholarly input or if these overtures are little more than an exercise in legitimization.
For our part, the exciting news is that the presumptions underlying SOX, along with the new policies being implemented by PCAOB, provide a rich set of questions and research opportunities for accounting scholars. There is at least the possibility that academic research can have a greater impact on policy-making and auditing practice than has been the case in the past. At the same time, it is incumbent on scholars to be staunchly independent of the accounting profession if they expect to be taken seriously by policy-makers. This means asking hard questions, and publishing papers with results that may not necessarily be flattering to the profession. References Abbott, L. J. , S. Parker, and G. F. Peters. 2004a. Audit committee characteristics and restatements. Auditing: A Journal of Practice & Theory 23 (1): 69-87.
Abbott, L. J. , S. Parker, G. F. Peters, and D. V. Rama. 2004b. Audit, non-audit and information technology fees: Some empirical evidence. Working paper, University of Memphis. Agrawal, A. , and S. Chadha. 2005. Corporate governance and accounting scandals. Journal of Law and Economics : Forthcoming. Anderson, R. C. , S. A. Mansi, and D. M. Reeb. 2004. Board characteristics, accounting report integrity, and the cost of debt. Journal of Accounting and Economics 37 (3): 315-342. Ashbaugh, H. , R. LaFond, and B. W. Mayhew. 2003. Do nonaudit services compromise auditor independence? Further evidence. The Accounting Review 78 (3): 611-639. Balsam, S. , J.
Krishnan, and J. S. Yang. 2003. Auditor industry specialization and earnings quality. Auditing: A Journal of Practice & Theory 22 (2): 71-97. Barton, J. , and P. J. Simko. 2002. The balance sheet as an earnings management constraint. The Accounting Review 77 (Supplement): 1-27. Basu, S. 1997. The conservatism principle and the asymmetric timeliness of earnings. Journal of Accounting and Economics 24 (1): 3-37. Basu, S. , L. Hwang, and C. -L. Jan. 2001. Differences in conservatism between big eight and non-big eight auditors. Working paper, Emory University. Bazerman, M. H. , G. Loewenstein, and D. A. Moore. 2002. Ambiguity in accounting and auditing.
Harvard Business Review 80 (11): 97-102. Beasley, M. S. 1996. An empirical analysis of the relation between the board of director composition and financial statement fraud. The Accounting Review 71 (4): 443-465. Beasley, M. S. , and S. E. Salterio. 2001. The relationship between board characteristics and voluntary improvements in audit committee composition and experience. Contemporary Accounting Research 18 (4): 539-570. Becker, C. L. , M. L. DeFond, J. Jiambalvo, and K. R. Subramanyam. 1998. The effect of audit quality on earnings management. Contemporary Accounting Research 15 (1): 1-24. Bedard, J. , S. M. Chtourou, and L. Courteau. 2004. The effect of audit ommittee expertise, independence, and activity on aggressive earnings management. Auditing: A Journal of Practice & Theory 23 (2): 13-35. Beneish, M. D. 1999. Incentives and penalties related to earnings overstatements that violate GAAP. The Accounting Review 74 (4): 425-457. Bhagat, S. , and B. Black. 1999. The uncertain relationship between board composition and firm performance. Business Lawyer 54 (3): 921-963. Blue Ribbon Committee. 1999. Report and Recommendations of the Blue Ribbon Committee on Improving the Effectiveness of Corporate Audit Committees. New York Stock Exchange and National Association of Securities Dealers. Burgstahler, D. , and I.
Dichev. 1997. Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics 24 (1): 99-126. Bushman, R. , and A. Smith. 2001. Financial accounting information and corporate governance. Journal of Accounting and Economics 32 (1-3): 237-333. Carcello, J. V. , and T. L. Neal. 2000. Audit committee composition and auditor reporting. The Accounting Review 75 (4): 453-467. ——. 2003. Audit committee characteristics and auditor dismissals following “new” going-concern reports. The Accounting Review 78 (1): 95-117. Choi, J. , and T. Wong. 2002. Auditor choice and legal environments: An international investigation.
Working paper, The Hong Kong University of Science and Technology. Chung, H. S. , and S. Kallapur. 2003. Client importance, nonaudit services, and abnormal accruals. The Accounting Review 78 (4): 931-955. Cotter, J. , and M. Silvester. 2003. Board and monitoring committee independence. Abacus 39 (2): 211-232. Craswell, A. , D. J. Stokes, and J. Laughton. 2002. Auditor independence and fee dependence. Journal of Accounting and Economics 33 (2): 253-275. Craswell, A. T. , J. R. Francis, and S. L. Taylor. 1995. Auditor brand name reputations and industry specializations. Journal of Accounting and Economics 20 (3): 297-322. Dalton, D. R. , C. M. Daily, A. E. Ellstrand, and J. L. Johnson. 1998.
Meta-analytic reviews of board composition, leadership structure, and financial performance. Strategic Management Journal 19 (3): 269-290. Dechow, P. M. , R. G. Sloan, and A. P. Sweeney. 1996. Causes and consequences of earnings manipulation: An analysis of firms subject to enforcement actions by the SEC. Contemporary Accounting Research 13 (1): 1-36. DeFond, M. L. 1992. The association between changes in client firm agency costs and auditor switching. Auditing: A Journal of Practice & Theory 11 (1): 16-31. DeFond, M. L. , R. N. Hann, and X. Hu. 2004. Does the market value financial expertise on audit committees of boards of directors? Journal of Accounting Research : Forthcoming. DeFond, M. L. , K.
Raghunandan, and K. R. Subramanyam. 2002. Do non-audit service fees impair auditor independence? Evidence from going concern audit opinions. Journal of Accounting Research 40 (4): 1247-1274. Fama, E. F. , and M. C. Jensen. 1983. Separation of ownership and control. Journal of Law and Economics 26 (2): 301-325. Felo, A. J. , S. Krishnamurthy, and S. A. Solieri. 2003. Audit committee characteristics and the perceived quality of financial reporting: An empirical analysis. Working paper, Penn State Great Valley. Feroz, E. H. , K. Park, and V. S. Pastena. 1991. The financial and market effects of the sec’s accounting and auditing enforcement releases.
Journal of Accounting Research 29 (Supplement): 107-142. Francis, J. 2004. What do we know about audit quality? The British Accounting Review 36 (4): 345-368. Francis, J. , and B. Ke. 2003. Do fees paid to auditors increase a company’s likelihood of meeting analysts’ forecasts? Working paper, University of Missouri. Francis, J. , and J. Krishnan. 2002. Evidence on auditor risk-management strategies before and after the Private Securities Litigation Reform Act of 1995. Asia-Pacific Journal of Accounting and Economics 9 (2): 135-157. Francis, J. , L. E. Maydew, and H. C. Sparks. 1999. The role of Big 6 auditors in the credible reporting of accruals.
Auditing: A Journal of Practice & Theory 18 (2): 17-34. Francis, J. , K. Reichelt, and D. Wang. 2005. The pricing of national and city-specific reputation for industry expertise in the U. S. audit market. The Accounting Review: Forthcoming. Francis, J. , and E. Wilson. 1988. Auditor changes: A joint test of theories relating to agency costs and auditor differentiation. The Accounting Review 63 (4): 663-682. Francis, J. R. , I. K. Khurana, and R. Pereira. 2003. The role of accounting and auditing in corporate governance, and the development of financial markets around the world. Asia-Pacific Journal of Accounting and Ecoomics 10: 1-31. Frankel, R. , M.
Johnson, and K. Nelson. 2002. The relation between auditors’ fees for nonaudit services and earnings management. The Accounting Review 77 (Supplement): 71-105. GAO. 2003. Public Accounting Firms: Required Study on the Potential Effects of Mandatory Audit Firm Rotation. United States General Accounting Office. Hay, D. , and D. Davis. 2002. The voluntary choice of an audit of any level of quality. Working paper, University of Auckland. Hermalin, B. E. , and M. S. Weisbach. 2003. Boards of directors as an endogenously determined institution: A survey of the economic literature. Economic Policy Reivew 9 (1): 7-26. Hilary, G. , and C. Lennox. 2004.
Self-regulation of the accounting profession: evidence on the AIPCA peer review porgram. Working Paper (Hong Kong University of Science and Technology). Jensen, M. C. , and W. H. Meckling. 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3 (4): 305-360. Johnson, V. , I. Khurana, and J. K. Reynolds. 2002. Audit-firm tenure and the quality of financial reports. Contemporary Accounting Research 19 (4): 637-660. Kinney, W. R. , Z. V. Palmrose, and S. Scholz. 2004. Auditor independence, non-audit services, and restatements: Was the US government right? Journal of Accounting Research 42 (3): 561-588. Klein, A. 1998.
Firm performance and board committee structures. Journal of Law and Economics 41 (1): 275-303. ——. 2002a. Audit committee, board of director characteristics, and earnings management. Journal of Accounting and Economics 33 (3): 375-400. ——. 2002b. Economic determinants of audit committee independence. The Accounting Review 77 (2): 435-452. Krishnan, G. 2003. Does big 6 auditor industry expertise constrain earnings management? Accounting Horizons 17 (Supplement): 1-16. ——. 2004. Did Houston clients of Arthur Andersen recognize publicly available bad news in a timely manner? Contemporary Accounting Research : Forthcoming. Larcker, D.
F. , and S. Richardson. 2004. Fees paid to audit firms, accrual choices, and corporate governance. Journal of Accounting Research 42 (3): 625-658. Lee, H. , and V. Mande. 2003. The effect of the Private Securities Litigation Reform Act of 1995 on accounting discretion of client managers of big 6 and non-big 6 auditors. Auditing: A Journal of Practice & Theory 22 (1): 93-108. Lennox, C. 2004. Audit quality and executive officers’ affiliations with CPA firms. Journal of Accounting and Economics : Forthcoming. Levitt, A. 1998. The “Numbers Game”. U. S. Securities and Exchange Commission. September, 28. McMullen, D. A. , and K. Raghunandan. 1996.
Enhancing audit committee effectiveness. Journal of Accountancy 182 (2): 79-81. Menon, K. , and J. D. Williams. 2004. Audit partners and abnormal accruals. The Accounting Review : Forthcoming. Mutchler, J. F. 1984. Auditors perceptions of the going-concern opinion decision. Auditing: A Journal of Practice & Theory 3: 17-30. Myers, J. , L. Myers, and T. Omer. 2003. Exploring the term of auditor-client relationship and the quality of earnings: A case for mandatory auditor rotation? The Accounting Review 78 (3): 779-799. Myers, J. , L. Myers, Z. Palmrose, and S. Scholz. 2004. Mandatory auditor rotation: Evidence from restatements. Working paper, University of Southern California.
Palmrose, Z. V. , and S. Scholz. 2004. The circumstances and legal consequences of non-GAAP reporting: Evidence from restatements. Contemporary Accounting Research 21 (1): 139-180. Panel on Audit Effectiveness. 2000. The Panel on Audit Effectiveness Report and Recommendations. Public Oversight Board, Stamford, CT. PricewaterhouseCoopers. 1999. Audit Committees: Good Practices for Meeting Market Expectations. PricewaterhouseCoopers. Rajan, R. G. , and L. Zingales. 2003. The great reversals: the politics of financial development in the twentieth century. Journal of Financial Economics 69 (1): 5-50. Reynolds, J. K. , D. Deis, and J. Francis. 2004.
Professional service fees and auditor objectivity. Auditing: A Journal of Practice & Theory 23 (1): 29-52. Reynolds, J. K. , and J. Francis. 2000. Does size matter? The influence of large clients on office-level auditor reporting decisions. Journal of Accounting and Economics 30 (3): 375-400. Romano, R. 2004. The Sarbanes-Oxley Act and the making of quack corporate governance. Working paper, Yale University. Rosenstein, S. , and J. G. Wyatt. 1997. Inside directors, board effectiveness, and shareholder wealth. Journal of Financial Economics 44 (2): 229-250. SEC. 2000. Proposed rule: Revision of the Commission’s Auditor Independence Requirements. U. S.
Securities and Exchange Commission. ——. 2003. Final Rule: Standards Relating to Listed Company Audit Committees. U. S. Securities and Exchange Commission. Simunic, D. 1980. The pricing of audit services: Theory and evidence. Journal of Accounting Research 18 (1): 161-190. Subramanyam, K. R. 1996. The pricing of discretionary accruals. Journal of Accounting and Economics 22 (1-3): 249-281. Vafeas, N. , and E. Theodorou. 1998. The relationship between board structure and firm performance in the UK. British Accounting Review 30 (4): 383-407. Watts, R. L. , and J. L. Zimmerman. 1983. Agency problems, auditing, and the theory of the firm – Some evidence.
Journal of Law and Economics 26 (3): 613-633. Wilson, T. E. , and R. A. Grimlund. 1990. An examination of the importance of an auditor’s reputation. Auditing: A Journal of Practice & Theory 9 (2): 43-59. Xie, B. , W. N. Davidson, and P. J. DaDalt. 2003. Earnings management and corporate governance: The role of the board and the audit committee. Journal of Corporate Governance 9 (3): 295-316. Zeff, S. 2003. How the U. S. accounting profession got where it is today: Part II. Accounting Horizons 17 (4): 267-286. ———————– Public criticism of the auditing profession around this time was also voiced by the Trueblood Committee, the Moss Report, the Metcalf Report and others.
See Zeff (2003) for a detailed and insightful account of the recent history of the U. S. Accounting profession.  One retired Big Four audit partner we spoke with who began his career in 1957 told us that one explanation of the speed of the SOX legislation is that Congress already had most of the rules drafted in prior failed legislation.  SOX also represents a radical change in the way in which the SEC traditionally regulates companies (Romano, 2004). While prior SEC regulation focuses on the disclosures required of public registrants, SOX focuses on mandating firm behavior, such as the composition of the audit committee and the delegation of duties between the board and management. 4] For example, see Klein (1998, 2002a), Bedard, Chtourou and Courteau (2004); Xie, Davidson and DaDalt (2003); and Beasley (1996); Cotter and Silvester (2003(; Vafeas and Theodorou (1998).  While a few studies do find evidence that 100% audit committee independence is associated with positive firm outcomes, Romano (2004) reports that these studies are either univariate in nature, or use multivariate models that have potentially omitted correlated variables problems.  In contrast, however, at least two papers find no association between measures of financial expertise on the audit committee and the “good governance” outcomes examined in those papers (Carcello and Neal, 2003 and Abbott, Parker and Peters, 2004).