Knowledge that Transforms

Abstract We explore the turnover intentions of creative R&D employees and the role of performance management practices in shaping these considerations. Since the success of a firm's R&D efforts hinges on the innovative ideas of its employees, it is crucial to retain particularly creative individuals. At the same time, however, we argue that this is especially difficult because both the higher outside options of creative employees and their specific individual characteristics make them, on average, more likely to leave their company. Most importantly, we suggest that two widely studied performance management design choices (target difficulty and career tournaments) typically used to motivate effort may influence the loss of creative talent. Using survey data from our unique access to R&D employees of a large manufacturing firm and a complementary experiment among business students, we find evidence that creative employees are, on average, more likely to leave their firm. Consistent with creative employees possessing a stronger learning orientation, we also predict and find that this tendency to leave is mitigated by target difficulty (as difficult targets speak to creative individuals' learning orientation) and exacerbated by the intensity of career tournaments (as they reduce team cohesion and, ultimately, undermine learning opportunities).

Abstract We examine the relative importance of audit firm versus partner decision styles in key audit matter (KAM) reporting. Standard setters intended KAMs to increase the usefulness of the audit report by requiring the partner‐led engagement team to disclose engagement‐specific information about the most significant judgments they made during the audit. However, stakeholders expressed widespread concern that audit firms' longstanding efforts toward standardization would result in generic KAMs at the audit firm level and provide partners little opportunity or incentive for engagement‐specific reporting. We evaluate this high‐stakes tension between standard setters' goals for audit reporting and auditors' deep‐rooted practices by leveraging data from the United Kingdom, which has required partner identification since 2009 and expanded audit reports since 2013. We find that clients sharing the same partner receive KAMs that are 10% more textually similar than clients with different partners. In contrast, clients sharing the same audit firm receive KAMs that are just 2% more textually similar than clients with different audit firms. This implies that partner decision styles are more important in influencing KAM outcomes than audit firm styles. Collectively, our results suggest that partners make unique KAM reporting judgments, countering concerns that audit firms' efforts toward standardization will yield boilerplate KAMs. This evidence extends the literature on expanded audit reporting and partner decision styles and provides valuable insights into a contemporary issue in audit regulation with broader implications for understanding dynamics within the profession.

Abstract We examine the information asymmetry between local and nonlocal investors with a large dataset of stock message board postings. We document that abnormal relative postings of a firm, that is, unusual changes in the volume of postings from local versus nonlocal investors, capture locals' information advantage. This measure positively predicts firms' short‐term stock returns as well as those of peer firms in the same city. Sentiment analysis shows that posting activities primarily reflect good news, potentially due to social transmission bias and short‐sales constraints. We identify the information driving return predictability through content‐based analysis. Abnormal relative postings also lead analysts' forecast revisions. Overall, investors' interactions on social media contain valuable geography‐based private information.

Abstract We examine how disclosure delivery mode—oral versus written—influences investors' reactions to managers' tone language. We hypothesize that listening to disclosures, relative to reading them, causes managers' qualitative word choices to have a greater impact on investors' judgments. We theorize that this effect occurs because oral delivery mode promotes heuristic processing and qualitative tone language is an easy‐to‐process disclosure element. The results from an experiment in a conference call setting are consistent with our hypothesis and suggest a boundary condition. Specifically, the interaction of mode and tone language is significant in a setting where heuristic processing is likely (good earnings news) but not in a setting where investors are likely to scrutinize the disclosure (bad earnings news). Our results inform investors about the potential consequences of how they consume disclosures. Specifically, we show that investors are more susceptible to managers' tone language when listening to disclosures containing good news than when reading them.

Abstract We explore how securities litigation affects corporate reputation. Experts remain concerned that nonmeritorious securities class actions—those that will be dismissed or settled for nuisance amounts—cause reputational damage. Although several prior studies show reputational costs for nonmeritorious cases, they generally use indirect measures based on returns or total market losses, which are mechanically associated with securities litigation elements. In contrast, we use a relatively direct reputation measure from Fortune 's “Most Admired Companies” list. We find significant reputational damage after meritorious litigation, with the strongest cases having the largest effects. However, we find no evidence of reputational damage after nonmeritorious litigation. We also find that Fortune 's reputational damage measure is associated with more negative returns around the litigation filing date. We show possible mechanisms for our results, as initial legal filings contain information allowing market participants to assess case merits. Our results imply that reputational damage is primarily due to fraud, which securities litigation helps reveal to the market, rather than litigation itself. Thus, reputational damage is not an issue in over 70% of securities class actions due to the high frequency of nonmeritorious cases.

Abstract We investigate the impact of firm‐level political risk on loan contracting. We find that firm‐level political risk is positively associated with bank loan cost and that this effect is stronger for firms experiencing increased operational uncertainty and higher default risks. Firm‐level political risk also leads to more unfavorable non‐pricing loan terms. To alleviate endogeneity concerns, we use an instrumental variable approach and placebo tests. We further find that political connections and relationship‐based borrowing can attenuate the adverse effect of firm‐level political risk on loan contracting.

Abstract Stock prices often provide firm managers with new information that can be used in real decisions. Studies generally focus on the ex ante disclosure policy and show that the presence of market feedback crowds out firms' disclosure. We instead examine a manager's ex post biasing incentives and find that market feedback amplifies overreporting bias, but not necessarily underreporting bias, due to three interacting effects. First, the manager biases the report more with feedback since decreased information quality crowds in the speculator's private information acquisition and improves investment efficiency, regardless of the reporting scenario (the information rationing effect). Second, the manager biases more in the overreporting scenario and less in the underreporting scenario, because reporting more favorable information crowds in private information acquisition, as the speculator expects a higher subsequent investment and therefore higher trading profits (the investment scale effect). Third, market feedback influences reporting bias not only through the speculator's information acquisition but also directly through the market maker's pricing function. Specifically, the market maker decreases the price discount, as he expects that the manager may learn correct information from the price and may invest more efficiently. Expecting a lower price penalty in the presence of feedback, the manager biases more in the overreporting scenario and less in the underreporting scenario (the investment correction effect). Overall, our results suggest that granting firms reporting discretion could improve investment efficiency and firm value when managers can learn through price.

Abstract This paper investigates how banks utilize soft information to provide contractual flexibility in loan covenant enforcement. We find that relationship lenders are significantly less likely than non‐relationship lenders to enforce covenants for material violations when borrowers are potentially in breach of financial covenants. The mitigation of information asymmetry by relationship lending, as opposed to alternative explanations, serves as the underlying mechanism. Furthermore, relationship borrowers with potential covenant breaches are less likely to experience increases in loan interest rates after renegotiation, to adopt conservative financial or investment policies, or to file for bankruptcy. Following potential covenant breaches of borrowers, relationship banks are better able to preserve regulatory capital and secure future lending business. Our findings suggest that soft information accumulated during lending relationships is vital for banks to provide contractual flexibility.

Abstract The study provides large‐scale descriptive evidence on the timing and nature of corporate financial tweeting. Using an unsupervised machine learning approach to analyze 24 million tweets posted by S&P 1500 firms from 2012 to 2020, we find that firms are more likely to tweet financial information around significantly negative or positive news events, such as earnings announcements and the filing of financial statements. This convex U‐shaped relation between the likelihood of posting financial tweets and the materiality of accounting events becomes stronger over time. Whereas research based on early samples concludes that firms are less likely to disseminate financial information on Twitter when the news is bad and material, the symmetric dissemination behavior we find suggests that these conclusions should be revised. We also show that a machine learning algorithm (Twitter‐Latent Dirichlet Allocation) is superior to a dictionary approach in classifying short messages like tweets.

Abstract Distributed, rather than co‐located, teams increasingly perform audit work, raising regulator concerns that distributed team communication issues may affect audit quality. We investigate upward communication (i.e., raising issues to supervisors), a key communication dimension related to audit quality. In Study 1, we survey 69 senior auditors to establish that distributed team upward communication suffers. Furthermore, distributed team auditors identify less with their teams and struggle to know when and how to speak up. Because both are linked to inhibited upward communication, we next experimentally test firm practices with the potential to attenuate these problems. Study 2, an experiment with 128 staff auditors and interns, reveals that making auditors' professional identities salient improves upward communication on co‐located teams with strong team identities but not on distributed teams with weak team identities. Study 3, an experiment with 58 staff auditors, replicates a key Study 2 finding with a design that makes no reference to team identity strength. Specifically, in the presence of a salient professional identity, upward communication is significantly higher for co‐located relative to distributed team auditors. Study 4, an experiment with 69 staff auditors, focuses solely on distributed teams. It indicates that several distributed team management practices, including one‐on‐one and standing daily meetings, encourage upward communication. Finally, Study 5, informal interviews with eight audit seniors, corroborates key survey and experimental findings. Combined, our results provide insights into why distributed team communication suffers, refine audit voice theory, and provide regulators, practitioners, and researchers with multiple paths to improve audit quality.