ABSTRACT We investigate whether reviews of transactional filings by the SEC unexpectedly constrain SEC resources, leading to lower quality comment letters for periodic reports. The Sarbanes‐Oxley Act requires the SEC to review periodic reports (e.g., 10‐Ks) at least once every three years. However, the SEC also reviews transactional filings (e.g., initial public offerings and acquisitions), which are unpredictable and often occur in waves. We find comment letters for periodic reports are of lower quality (in terms of outputs, inputs, and firm responses) during periods of abnormally high transactional filings. We also find that comment letters issued during periods of abnormally high versus low transactional filings are associated with increased information asymmetry and lower earnings response coefficients in the quarter after the resolution of the comment letter. Overall, our results suggest that unexpected resource constraints affect the quality of SEC oversight of periodic reports.

ABSTRACT Regulators and researchers have expressed concerns that social interaction leads auditors to unjustifiably trust managers, constituting a lack of sufficient professional skepticism. Using both an abstract laboratory experiment and a contextually rich experiment with practicing auditors we predict and find that higher Dark Triad auditors (those with higher levels of the shared core between psychopathy, narcissism, and Machiavellianism) are relatively more resistant to lapses in professional skepticism due to the effects of social interaction. This is likely driven by higher Dark Triad auditors' callousness, lack of empathy, and lack of response to social stimuli. In contrast, while higher social interaction initially increases lower Dark Triad auditors' unjustified trust in managers, this effect reverses in subsequent interactions when lower Dark Triad auditors observe evidence suggesting managers have reported aggressively. These findings add to research on the effect of auditor personality traits, audit‐client social interaction, and the interaction of these two variables, and suggest that practitioners and researchers account for the interplay of Dark Triad traits and social interaction and their effect on professional skepticism.

ABSTRACT Bagnoli and Watts (2005) proposed that a manager could reduce information asymmetry by choosing an income‐decreasing accounting choice that signals the firm's relatively good future prospects. A limitation in testing this theory is that most income‐decreasing accounting choices over time reverse such that aggregated earnings would be the same, regardless of the choice. One income‐decreasing accounting choice that never reverses is the choice of upward asset revaluation, where the resulting gains are recognized through other comprehensive income and reduce future earnings by increasing future depreciation expense. In the United Kingdom, prior to FRS15, firms had the option to upwardly revalue on a one‐time basis. FRS15, and subsequently International Financial Reporting Standards, however, require those firms that upwardly revalue precommit to revalue on a consistent basis. This precommitment sacrifices future reporting discretion, which, according to the aforementioned study, serves as a costly signal of a firm's relatively good future prospects that reduces information asymmetry. The choice not to upwardly revalue, therefore, serves as a signal of a firm's relatively poor future prospects and also reduces information asymmetry, but this choice does not require precommitment such that the reduction in information asymmetry would be less than the choice to precommit to upward revaluations. Using a propensity‐score matched‐pair design on a sample of United Kingdom firms to test our predictions during the period requiring precommitment, we find lower forecast dispersion, lower return volatility, and a lower cost of capital for firms that precommit to upward asset revaluations, relative to those firms that choose not to upwardly revalue their operating assets. Keywords: upward asset revaluations, income‐decreasing accounting choice, information asymmetry, precommitment

ABSTRACT We argue that volatility in a manager's disclosure tone across time should be a function of two components: (i) the firm's innate operating risk and (ii) the extent to which the manager's disclosure transparently reflects that risk. Consistent with this argument, we find that both operating risk and disclosure transparency are important determinants of disclosure tone volatility. We then examine whether investors incorporate the incremental information provided by disclosure tone volatility into their assessments of firm risk. If disclosure tone volatility primarily provides investors with incremental information about a firm's operating risk, we should find a positive association between tone volatility and market‐based assessments of risk. On the other hand, if disclosure tone volatility primarily provides investors with incremental information about a manager's disclosure transparency, we should find a negative association between tone volatility and market‐based assessments of risk. Consistent with an operating risk explanation, we find a positive association between disclosure tone volatility and market‐based assessments of firm risk after controlling for a comprehensive set of proxies for operating risk and transparency. We find little support for an information risk explanation, even when we examine multiple measures specifically designed to capture information risk. Taken together, our results suggest that although disclosure tone volatility is a function of both a firm's operating risk and a manager's disclosure transparency, investors appear to respond as if disclosure tone volatility only provides incremental information about a firm's operating risk.

ABSTRACT Technological advances are creating a shift in the information disclosure environment allowing more investors to interact with management. We examine three key levels of trader‐management interaction to assess the accuracy of traders' market‐tested value estimates and resulting market price. These data require an engaging experiment and a complex, contextually rich asset, which we create by playing a popular gaming app before the experiment. Participants view financial information, ask management questions, estimate value, and trade. We find that receiving non‐personalized question responses improves trader estimates of value and market price efficiency relative to when traders ask questions but do not expect a response. This occurs because traders exert more effort estimating value and trading. However, receiving personalized versus non‐personalized responses harms value estimates and market efficiency. This occurs because traders receiving personalized responses fixate on the interaction with management, dividing their attention and diverting it away from valuing and trading the asset.

ABSTRACT Inadequate testing of fair value accounting estimates, including goodwill, is often cited as an audit deficiency in PCAOB inspection reports, and, in some cases, these deficiencies have led to enforcement actions against the auditor. As a result of these issues, the PCAOB recently proposed a new auditing standard for fair value accounting. While these regulatory actions suggest that auditors are challenged by the fair value regime of accounting for goodwill, they also highlight an area where the auditor could be influenced by their financial ties to a client. In this study, we test whether nonaudit fees are associated with goodwill impairment decision outcomes. Our results indicate that the nonaudit fees a client pays are inversely related to the likelihood of impairment in settings where goodwill is likely to be impaired. Additional examinations suggest that the negative relation between nonaudit fees and auditor independence is driven by clients who are most incentivized to exert their influence over the auditor.