We argue that conflicting estimates of auditors’ industry leadership premia documented in prior research reflect model misspecification. We show that leadership measures can be expected to identify different auditors as leaders depending on the basis (assets, fees or auditee counts) used to measure leadership. To correctly estimate premia to any of these multiple types of leaders, estimation models must include a full set of indicators for leaders by all admissible bases at each admissible level (national, local or joint national-and-local), which prior studies fail to do. Using such an estimation model and including controls for local market factors ignored in prior research, we find that premia are primarily associated with local market leadership. Local fee-only leaders (leaders by audit fees but not by assets audited or auditee count) charge substantial premia, local asset-only and local count-only leaders offer significant discounts, while local leaders by various combinations of these three bases price somewhere in between. Our study provides a framework for future research to correctly estimate leadership premia when industry leadership by different bases and levels vests in different auditors.

We present the first large-sample empirical evidence on U.S. auditors' responses to changes in entity-level audit risk during 2006-2007, the period leading-up to the financial crisis of 2008-2009. Treating fiscal year 2005 engagements as a precrisis benchmark, we find that audit attention during fiscal year 2006 and 2007 bank audit engagements shifted in line with the shifting audit risks. One implication of these findings is that auditors were able to recognize and respond to financial statement impacts of the macroeconomic shocks that unfolded during the lead-up to the crisis. Another implication is that auditors' failure to issue advance warnings of increasing auditee riskiness during the time leading up to the financial crisis more likely reflects limitations of extant accounting and auditing rules rather than a lack of auditor awareness or attention to those risks.

Audit fee residuals (the error term from audit fee models) are widely used in the accounting research literature. Researchers, however, have adopted contrasting views of these fee residuals. One view is that fee residuals are a combination of noise and auditor rents (i.e., abnormal profits), while the other view is that they are a combination of noise and unobserved audit costs (including any risk premium and a normal rate of return on all factors of production). As a result, identical research findings are presently given conflicting policy interpretations. We use differences in fee residual persistence across continuing and new audit engagements to elucidate the extent to which fee residuals consist of unobserved audit costs, auditor rents, and noise elements. In a large sample of U.S. public company audit engagements, we find evidence suggesting that fee residuals largely consist of researcher-unobserved audit production costs common to all auditors. We discuss the implications of this finding for policy setters and for future auditing research.

The duration and magnitude of the private economic benefits obtained from firm-level R&D outlays are important to managers, investors and policy-makers. We present a modeling and estimation approach that facilitates reliable estimation of these benefits under mild a priori assumptions. Our estimates of R&D useful life and realized returns shed light on three important debates in the literature: Do firms invest optimally in R&D? Do investors entertain rational expectations about firm-level R&D benefits? Has the cost of R&D capital declined over time? We answer each of these questions in the affirmative. Further, our evidence indicates that conventional measures of R&D capital stock are grossly overstated.

ABSTRACTThe replacement of Auditing Standard No. 2 (AS2) by Auditing Standard No. 5 (AS5) creates a natural experiment that sheds light on (1) potential inefficiencies caused by regulatory responses to a political crisis and (2) audit efficiency and effectiveness improvements resulting from the risk‐based approach embodied in AS5. We study these effects by examining the impact of AS5 on audit fees. We find that AS5 audit fees are aligned with auditee fraud risk, but not AS2 audit fees. Second, relative to AS2 benchmark levels, AS5 audit fees are, on average, lower for all auditees. Third, relative to AS2 benchmarks, AS5 fees are lower for lower‐fraud‐risk auditees but greater for higher‐fraud‐risk auditees. Overall, the evidence is consistent with (1) initial overregulation (via AS2) followed by reform (via AS5) and (2) auditors deploying a risk‐based audit approach to obtain both efficiency and potential effectiveness gains in audit production.

Backdating is the deliberate falsification of accounting records to obscure the magnitude of employee stock option compensation expense. By backdating, management arguably violates its stewardship obligation (duty to report truthfully on the use of corporate resources). As a result, the propriety of backdating as a compensation scheme is also questionable. Using a sample of 3,892 firms with executive stock option grants outstanding, we investigate investor reactions to two early news reports of stock option backdating. Our key findings are that the first news report significantly impairs firm value for over a thousand firms (the third of the sample firms most at risk of having backdated) while the second report generates additional value impairment only for about 200 firms subsequently confirmed to have backdated. We also find no differential reaction across Big Four and Non-Big-Four auditees. Our findings suggest that investors perceived the net impact of backdating schemes as value-reducing and processed the implications of the reports in a timely and rational way. More broadly the results support the proposition that investors value the stewardship (truth-telling) role of accounting and react negatively when that function is seen to be impaired. Finally, our results suggest investors do not expect perceived audit quality differences between Big Four and Non-Big-Four auditors to significantly affect the likelihood of backdating.

ABSTRACTThe adoption of business risk audit (BRA) approaches during the 1990s by several leading audit firms has been the subject of considerable scrutiny and commentary. Under BRA, the auditor responds to the increasing complexity of auditee financial reports by acquiring a deep and comprehensive understanding of the auditee's industry, strategy, business models, and processes—tasks best accomplished by higher‐ranked labor—and by employing this understanding to make audit labor allocations. Using proprietary data for 165 audits conducted in 2002, we investigate three propositions about audit labor use under BRA. First, relative to pre‐BRA benchmarks for the same auditor, we expect BRA audits to use a greater proportion of higher‐ranked labor. Second, we expect engagements with high assessed auditor business risk (ABR), a summary risk assessment that reflects the BRA auditor's rich understanding of the auditee, to be allocated more labor and more higher‐ranked labor than pre‐BRA benchmarks. Third, at all ranks of labor, we expect a positive association between assessed ABR and levels of labor use. We find empirical evidence consistent with these propositions. We also find that total labor use in our sample is only modestly lower than pre‐BRA norms. Analysis of fee data from these engagements suggests that audit fees in 2002 are substantially less than would be expected under pre‐BRA benchmarks. After controlling for audit labor use, both total fees and fees per hour increase with assessed ABR for first‐year auditees but not for continuing auditees. Overall, our results provide evidence on the impact of the BRA audit regime and speak to the likely impact of BRA on audit effectiveness and efficiency.

In recent years, a substantial percentage of managers voluntarily choose to provide disaggregated earnings forecasts, i.e., earnings forecasts supplemented with additional numerical forecasts for other line items on the income statement. We examine whether such disaggregation yields higher-quality (i.e., more accurate and less biased) management earnings forecasts, and whether market reactions to disaggregated vs. aggregated forecasts are consistent with the quality of these forecasts. We find that: (1) for good news forecasts, earnings forecasts with disaggregated information are no different from aggregated earnings forecasts in either bias or accuracy; (2) for bad news forecasts, earnings forecasts with disaggregated information are significantly less accurate and, on average, more optimistic than aggregated earnings forecasts; 3) stock market reactions to disaggregated good news forecasts are no different from stock market reactions to aggregated good news forecasts, but stock market reactions to disaggregated bad news forecasts are more negative than stock market reactions to aggregated bad news forecasts. Taken together, our results suggest that disaggregated earnings forecasts are no better than and sometimes even worse than aggregated earnings forecasts and that investors anticipate and adjust for the biases associated with disaggregation in management earnings forecast.

Current auditor liability rules require the auditor to pay investors damages in the event of an audit failure. I show, in a simple model of investment under uncertainty, that no such payment mechanism will, in general, be efficient, i.e., implement first-best audit effort or investment levels. By contrast, a system of decoupled damages in which the damages paid by auditors vary with investment while damages received by investors do not implements first best. In contrast to most extant proposals for auditor liability reform (financial statement insurance, damages as a multiple of audit fees or auditee market value, auditor liability caps) which seek to modify aspects of the current system (of direct transfers), I show that efficiency calls for a radically different approach, namely moving to decoupled liability. The approach also highlights a role for non-transferable penalties upon auditors: such penalties can motivate auditor effort without distorting investor incentives but have been largely ignored in the literature which focuses for the most part on direct financial transfers from auditors to investors. Thus my approach also provides a framework to unify the study of different types of penalties to motivate auditors.

The implosion of Enron and the demise of Arthur Andersen in 2002 significantly altered the structure of the US audit market and, it has been argued, enhanced the appeal of smaller auditors relative to the Big N auditors that historically have dominated the US audit industry. We present empirical evidence on the ability of Non-Big-N auditors to compete with the Big N auditors in the post-disruption period (2003-2006) relative to pre-disruption benchmarks (1989-2001). We find that relative to the earlier period, the rate of auditee flow from Big N auditors to Non-Big-N auditors has increased in the post-2002 period, and that the flow from Non-Big-N auditors to Big N auditors has decreased. The average size of auditees switching to Non-Big-N has also increased. In the pre-2002 period, auditees that choose a Big N successor are larger, more profitable and less likely to receive a modified opinion than auditees that choose a Non-Big-N successor. We find this gap persists or even grows in the post-2002 period. For example, in the post-2002 era, the gap between Non-Big-N auditees that switch to Big N in comparison to those that switch to Non-Big-N has increased relative to the pre-2002 era, with larger differences in pre-switch stock returns, likelihood of a clean opinion, leverage and size. Overall, we find little evidence of a post-2002 increase in the relative appeal of Non-Big-N auditors to stronger auditees. AUDIT MARKET CONTESTABILITY IN THE POST-ENRON ERA.

Does bad news about one auditor’s conduct affect the credibility of other auditors? We address this question by investigating the existence, timing and magnitude of auditor credibility impairment spillovers around 25 bad news events involving Arthur Andersen LLP’s Waste Management, Sunbeam and Enron audits. We document strong evidence of market-wide negative abnormal returns (spillovers) around SEC sanctions and criticism of Andersen’s Waste Management and Sunbeam audits and around select events as Enron was failing. Thus, bad news about a leading auditor does, in our setting, impair other auditors’ credibility as well. In addition, we find that the timing and magnitude of reputation impairment are somewhat different for Andersen itself and for other Big Five auditors. This suggests that leading auditors’ reputations are highly but not perfectly correlated. ARE AUDITOR REPUTATIONS CORRELATED? EVIDENCE FROM THE ANDERSEN EXPERIENCE

Abstract Wepresent,evidence on the,determinants of Big Five and predecessor audit firm clienteles. We develop a framework based on a portfolio view of the firm’s audit practice and use it to investigate four important questions on the determinants of audit firm clienteles and audit firm mergers. We find that firms’ practice diversification strategies reflect the hypothesized ,trade-offs between ,risk- diversification and incentives to exploit scale economies. We also find systematic differences among large audit firms in the degree of practice concentration driven, in part, by differences in their internal organization. Further, all firms that were party to mergers (Big Eight to Big Six, Big Six to Big Five) were characterized by higher than average degrees of practice concentration, while the resulting entities are characterized by lower than average degrees of practice concentration. This suggests that a possible incentive for large audit firm mergers may have been the desire to diversify the practice. Finally, we examine and find evidence that the Ernst and Young merger was not followed by a change in the merged firm’s diversification strategy, while the Deloitte and Touche merger was followed by an increase in practice concentration. Keywords: Practice Concentration, Auditor Industry Specialization, Audit Clienteles, Audit Firm

Visiting Assistant Professor, University of Illinois (1998-2002). Assistant Professor, University of Notre Dame (1994-1998). Instructor, University of Notre Dame (1992-1994). Teaching Assistant, The Pennsylvania State University (1988-92). Research Assistant, Harvard Business School (summers, 1986-88). Partner, Doogar and Sethia, Chartered Accountants (1982-1985). Management Trainee, Majestic Packaging Co., (1981-2) Articled Clerk, S. S. Kothari & Co., Chartered Accountants, Calcutta, India (1978-81).

I examine calls to add teaching training to accounting doctoral program content using the lens of economic analysis. My analysis of key determinants of demand and supply in the market for accounting Ph.D.s offers little support for calls for additional emphasis on developing teaching skills. I suggest a two-part approach to the perceived lack of teaching skills among new faculty First, in keeping with standard economic theory, employers should re-examine the incentives they offer for good teaching before demanding changes in accounting doctoral program curriculum. Second, that the optimal locus of development of teaching skills be left to private contract between employers and faculty (as opposed to being bundled with the technical education currently imparted in doctoral programs). One way to do this might be additional, post-doctoral, teaching certification and continuing professional education.

We study what a simulated market of pure price competition reveals about market shares and per-partner profits in Big Six audit firm mergers for 1997. In the model of audit production we use, leverage (staff-to-partner ratio) reflects the productivity of an audit partner and is key to analyzing productivity changes that arise from merger. We model price competition among 270 firms and find that post-merger leverage improvement leads to increases both in market share and in per-partner profits. The rank of a merger on market-share increase is sensitive to alternative estimates of leverage, but the rank on per-partner profit increase is not. A merger involving Arthur Andersen generally ranks highest. Absent Arthur Andersen, we find that the 1998 merger between Price Waterhouse and Coopers & Lybrand produces higher per-partner profit increases than the failed merger between Ernst & Young and KPMG Peat Marwick.

The International Accounting Standards Committee (IASC) was organized in 1973 with the primary goal of harmonizing international accounting standards. In 1988, the IASC conducted a survey to determine the extent of conformity of national accounting standards and practices to IASC standards. While prior research has attempted to group countries into clusters based on conformity with IASC standards, little is known about the determinants of conformity. IASC standards vary both in the degree of flexibility they permit in reporting as well as in the type of provision covered by the standard. We use ordinal logistic regression analysis to examine the effect of extent of uniformity and topic of provisions contained in a particular IASC standard on conformity scores for that standard. The extent of national standard setters' adoption of IASC standards is negatively related to the extent of uniformity mandated by a standard and also to disclosure mandates. Mandates for absolute uniformity in disclosure are most strongly associated with lower conformity while flexibility in accounting principles provisions is most strongly associated with higher conformity. In the aggregate the data are consistent with the view that the nature of the standards provisions significantly influence the extent of conformity with IASC standards.

We show that a model of undifferentiated price competition closely predicts US audit market concentration. Contracting practices, client size distributions and differences in auditor productivity jointly determine audit firms' market shares. In contrast to prior literature, neither quality differences nor economies of scale for larger firms are necessary in our model to explain audit market concentration.

Observers report that, of late, US audit clients have come to view audits as commodities rather than as quality-differentiated products (Elliott 1994, AICPA 1995a), and to view audit providers as being undifferentiated except on price (Stevens, 1991). How might such a shift in the basis of competition affect the structure of the audit market? To examine this issue, we construct a model of pure price competition and predict audit industry structure for a sample of 5320 publicly traded US industrial companies and the 306 US audit firms that audit them. We find that under a variety of market clearing rules, price competition among auditors leaves current industry structure essentially unaltered. We also find that given current industry configuration, pure price competition generates industry structures similar to what we observe in our sample. Our results show that product differentiation is not necessary either to generate or to sustain observed industry structure.