Nicholas J.
Hallman

Both popular media and academic research have long asserted that the Big 4 audit firms ("the firms") are "boys clubs" that fail to equitably retain and promote female auditors. Recently mandated Form-AP disclosures have reignited interest in gender disparities at the firms by revealing that, despite their explicit commitments to gender equity, only a small proportion of their public company engagements are led by female audit partners. We use newly available data on nearly 150 thousand rank-and-file auditors to show that these assertions about gender disparities at the firms were historically accurate. Throughout the 1990s and 2000s, female auditors faced lower retention and promotion rates than their male counterparts. These disparities were even more pronounced in auditing than in most other financial services professions. However, we also show that these disparities narrowed among auditors during the mid 2010s and reversed by the late 2010s. Indeed, we present evidence which suggests that from the late 2010s through the end of our sample period in 2023, it was male auditors who were at higher risk of leaving the firms and faced lower probabilities of promotion conditional on staying. The absence of a similar reversal in retention rates for other financial services professions during the same period bolsters our conclusion that the trends observed are not driven by broader industry or labor market factors. All the code used to create this paper is publicly available at https://github.com/njhallman/Gender-public.

We study the effects of analyst coverage on syndicated lending. We hypothesize that analyst research alleviates information asymmetries between lead arrangers and participant lenders within a syndicate, increasing the participants' credit supply and reducing the required loan interest spread. Using exogenous shocks to firms' analyst coverage, we find that firms pay higher loan interest spreads and that participant lenders fund smaller fractions of the loans after firms experience a reduction in analyst coverage. Participants are more likely to be nonbank institutional investors and to transact with familiar lead arrangers after the coverage shocks.

We investigate whether non-GAAP reporting affects the audit process and thereby the quality of the related financial statements. First, we provide evidence that auditors in numerous countries, including the United States and the United Kingdom, rely to varying degrees on non-GAAP profit before tax as a benchmark for determining quantitative materiality. Then, using Premium Listed companies on the London Stock Exchange, we document that U.K. auditor reliance on non-GAAP materiality benchmarks often results in a higher quantitative materiality amount and can lower audit quality. Although U.K. auditors appear skeptical of managers' more aggressive non-GAAP adjustments, auditors adopt more of management's low-quality adjustments when auditor independence is weaker. In sum, our results suggest that non-GAAP reporting can indirectly affect investors by reducing the rigor of the financial statement audit.

Prior research provides mixed evidence about whether competition among auditors impairs or improves audit quality. An impediment to this stream of research is the inability of researchers to observe the audit engagement bidding process. We develop a method of detecting bidding by applying a machine learning algorithm to non-incumbent (i.e., competitor) auditor views of public companies' SEC filings. We validate our method using a proprietary sample where all instances of bidding are known. We then examine the associations between bidding, audit quality, and audit pricing. Contrary to concerns that competitive pressure may cause auditors to compromise their independence, we find that incumbent auditors perform higher quality audits during bidding years. This improvement in audit quality occurs regardless of whether the bidding ultimately results in an auditor change and persists for several years when the incumbent auditor wins reappointment. We also find that bidding is associated with modest audit fee concessions.

We examine how audit partners' geographic proximity to clients affects audit quality. We use hand-collected data to show that approximately half of audit partners are assigned to clients headquartered more than 100 km away from the partners' home locations. Few of these partners relocate after receiving their assignments and, as a result, more than one-third of clients are audited by partners who must commute long distances to visit the client in person. We explore this phenomenon by first modeling how distance affects partner-client matching. We find that partners' geographic proximity to a prospective client is an important matching criterion, but also that trade-offs are made when other partner characteristics such as industry specialization are more likely to be important. Next, consistent with our prediction, we show that audit quality is lower when partners reside farther from their clients. We corroborate our primary findings by showing that the association between partner distance and audit quality is mitigated when partners have access to direct flights to their clients' headquarters and when clients are geographically dispersed. Our paper should be informative for regulators, practicing auditors, and academics interested in how partner-client matching affects audit outcomes.

Literature to date reveals relatively little about the role of expertise in auditor selection beyond basic preferences for Big 4 and industry specialist auditors. We hypothesize that audit committees whose members have no Big 4 auditing experience are likely to struggle when interviewing prospective Big 4 partners, leading such committees to draw on superficial, heuristic cues in lieu of conducting more substantive evaluations. To test this prediction, we obtain independent ratings of the facial attractiveness of audit partners identified from Form AP filings recently mandated by the US PCAOB. Our primary finding is that audit committees with no Big 4–experienced members are more likely to favor partners whose photographs raters view to be highly attractive. We characterize attractiveness as a superficial attribute for auditor selection because we detect no relation between attractiveness and accruals- or restatement-based measures of financial reporting quality for audit committees with one or more Big 4–experienced members. We do find an inverse association between attractiveness and financial reporting quality for committees without this experience, likely reflecting the statistical implication of a selection bias. We conclude that auditing expertise mitigates the influence of superficial considerations in auditor selection, enabling audit committees to fulfill their stewardship role more effectively.

Recent research suggests that insiders of distressed firms, fearing legal jeopardy, pressure auditors not to issue going concern opinions (GCOs) for periods in which they undertake abnormally large sales of their shares. We propose and evaluate an alternative explanation that managers anticipate GCOs and time their trades to avoid insider sales in the GCO year (hereafter, the timing hypothesis). Consistent with the timing hypothesis, we find that insider sales increase two to four years prior to the issuance of a GCO and then decline in the year of GCO. Additional analysis suggests that insiders' anticipatory trading is enabled, at least in part, by early communication between auditors and their most important clients regarding the likelihood of a GCO. These early communications appear to reduce the likelihood of dismissal when auditors do eventually issue a GCO.

Audit firms are organized as collections of geographically decentralized offices. Decentralization allows for increased proximity between offices and clients, improving the efficiency of auditors' interactions with client personnel. Yet decentralization also decreases the proximity between offices within each firm, potentially impeding auditors' interactions with each other. We show that decreased proximity between offices reduces inter-office audit quality "spillovers" and that this effect is driven primarily by reduced monitoring and knowledge sharing. Our findings expand the "within office" view of audit production by demonstrating the importance of interactions between offices and the role of geographic proximity in facilitating them.

We find that non-Big 4 audit offices with greater awareness of SEC enforcement are more likely to issue first-time going-concern reports to distressed clients; where SEC "awareness" is measured using (i) audit office proximity to SEC regional offices, and (ii) proximity to specific SEC enforcement actions against auditors. We also show that these non-Big 4 audit offices issue more going-concern opinions to clients who do not subsequently fail, indicating a conservative bias that reduces the informativeness of audit reports. This conservative reporting bias is also associated with higher audit fees and higher auditor switching rates. These findings are important because non-Big 4 firms now audit 39 percent of SEC registrants and issue 88 percent of going-concern audit reports. For Big 4 offices, we find some evidence that awareness of SEC enforcement may improve reporting accuracy by reducing Type II errors (failing to issue a going-concern report to a company that fails), although the number of cases is small.

We examine the decision relevance of a commonly suggested adjustment to how state governments report governmental pension liabilities by recalculating such pension liabilities using the return on a portfolio of high-quality municipal bonds as the discount rate. Calculated as the difference between the state's expected rate of return and the municipal bond return, we find that the discount rate adjustment associates with lower credit ratings and higher interest costs. We also find that credit rating agencies are more likely to issue conflicting ratings when the calculation of the discount rate adjustment involves greater uncertainty. Overall, while financial statement users agree about the need for and the direction of a pension liability rate adjustment, there is less consensus about the proper magnitude of this adjustment, suggesting that current accounting treatment of pensions in the public sector leads to costly uncertainty among financial statement users.

Foreign listings in the United States (U.S.) have fallen sharply over the past two decades. Using the staggered introduction of Public Company Accounting Oversight Board (PCAOB) inspection access in 55 countries, we test whether the regulator's expanding global oversight contributed to this decline. Our stacked difference-in-differences estimates show that foreign firms are significantly less likely to pursue U.S. listings when doing so subjects their audits to PCAOB inspection. This deterrent effect comes from a drop in new listings on major exchanges, not from over-the-counter listings or deregistrations. The effect is strongest for (1) firms facing higher compliance costs of inspections, (2) industries likely viewed as strategically sensitive by foreign governments, and (3) countries with greater political tension with the U.S. Placebo tests show no similar decline in non-U.S. listings after PCAOB access; in fact, listings in some foreign markets rise, consistent with substitution away from U.S. exchanges.

Due to concerns about the quality of non-GAAP reporting, standard setters worldwide have considered permitting or requiring companies to report non-GAAP measures in the audited financial statements. Using a within-firm-year research design for a sample of United Kingdom Main Market companies, we compare non-GAAP earnings reported exclusively in the unaudited strategic report to those that are also reported on the face of the income statement or in footnotes. Consistent with assurance improving information quality and curbing non-GAAP aggressiveness, we find that non-GAAP earnings are higher quality when they are audited. To understand how assurance provides this benefit, we examine the engagement-level procedures auditors apply to non-GAAP reporting. Procedures that evaluate adjustments against existing criteria such as the firm's own policies, standards, or historical practices are more effective at curbing non-GAAP aggressiveness than procedures that rely on auditor judgment alone. Finally, non-GAAP earnings are more value relevant when they are audited than when unaudited. Our findings offer timely evidence for the FASB and IASB as they evaluate non-GAAP reporting requirements.

Due to an audit fee measurement issue in years when companies change auditors, Barua, Lennox, and Raghunandan (2020) (hereafter, BLR) dispute prior findings that successor auditors discount audit fees. In this study, we re-examine whether audit fee discounting occurs using the same dataset as BLR and arrive at a different conclusion. We test for audit fee discounting by examining fees in the years surrounding an auditor change, thus side-stepping the BLR measurement issue. We show that successor auditors discount audit fees for new clients and that these discounts persist for several subsequent years. We find that discounting occurs regardless of whether the predecessor auditor charged abnormally high or low fees prior to being replaced, suggesting that the reduction in fees after auditors changes is not simply an elimination of economic rents. We also find that discounting is most (least) likely when clients change from Big 4 to non-Big 4 (non-Big 4 to Big 4) auditors, although discounting also occurs for within-type (i.e., within Big 4 and within non-Big 4) auditor changes. Finally, consistent with a "winner's curse" effect, we find that large fee discounts are associated with lower quality initial-year audit engagements, but that quality quickly improves as newly appointed auditors become more familiar with their clients.