This study examines whether the term of the auditor–client relationship (i.e., auditor tenure) is associated with future stock price crash risk measured both ex ante and ex post. Using a large sample of U.S. public firms with Big 4 auditors, we find robust evidence that auditor tenure is negatively related to one†year†ahead stock price crash risk. The evidence is consistent with monitoring†by†learning where development of client†specific knowledge over the term of the auditor–client relationship enhances auditors’ ability to detect and deter bad news hoarding activities by clients, thereby reducing future crash risk. This result holds even after controlling for endogeneity of the tenure/crash risk relation. We further provide evidence indicating that option market investors do not fully incorporate the information contained in the term of auditor–client relationship in predicting future stock price crash risk. Our empirical results have important policy implications for regulators concerned with ensuring auditor independence.Les auteurs se demandent si la durée de la relation auditeur†client (soit la durée du mandat de l'auditeur) est en relation avec le risque d'effondrement futur du cours des actions, évalué ex ante et ex post. En étudiant un vaste échantillon de sociétés des États†Unis faisant appel public à l’épargne qui ont recours aux services des Quatre Grands cabinets d'audit, ils recueillent des preuves convaincantes que la durée du mandat de l'auditeur est en relation négative avec le risque d'effondrement du cours des actions une année à l'avance. Ces preuves sont conformes à la pratique de la « surveillance par l'apprentissage », le perfectionnement des connaissances relatives au client pendant la durée de la relation auditeur†client améliorant l'aptitude des auditeurs à déceler et prévenir chez les clients le comportement de thésaurisation des mauvaises nouvelles, ce qui a pour effet de réduire le risque d'effondrement futur. Ces résultats persistent même une fois contrôlée l'endogénéité de la relation entre la durée du mandat et le risque d'effondrement. Les auteurs produisent d'autres preuves que les investisseurs sur le marché des options n'incorporent pas entièrement l'information que recèle la durée de la relation auditeur†client dans la prévision du risque d'effondrement futur du cours des actions. Les résultats empiriques de l’étude ont d'importantes conséquences au chapitre des politiques pour les autorités de réglementation soucieuses de l'indépendance des auditeurs.

Prior research (e.g., Dechow, Huson, and Sloan ) documents that, on average, compensation practices appear to shield CEO pay from income‐decreasing special items. In some circumstances, compensation shielding can be efficient. For example, it may encourage CEOs with earnings‐sensitive pay to take an action that reduces current earnings but nevertheless enhances value. Compensation shielding can be inefficient in other circumstances, such as when a board of directors is captured by an overly powerful CEO or the magnitude of negative special items has been overstated (e.g., by shifting core expenses into special items). This paper explores whether strong governance can explain cross‐sectional variation in compensation shielding, and whether stronger governance and auditing are associated with less shifting of expenses. We find that strong corporate governance mechanisms, as captured by board (and committee) independence, the Sarbanes‐Oxley (2002) Act (SOX) and its related governance reforms, and switches to Big 4 auditors, are all associated with less compensation shielding. While our evidence suggests that strong overall governance is associated with a reduction in manipulation of core earnings through classification shifting in the cross‐section, we find inconclusive evidence to suggest that board independence or SOX influence classification shifting.

We examine the influence that geographic proximity to bank branches and nonbank financial providers has on use of financial transaction services among U.S. households. We specify a bivariate probit model of bank account ownership and nonbank transaction product use to reflect the joint nature of these choices, and estimate the model on a large, nationally representative dataset. Our results indicate that households with reasonable geographic access to bank branches are more likely to have a bank account and less likely to use nonbank transaction products. The influence of bank and nonbank provider locations is fairly modest overall, although effects are bigger for households that are more likely to be on the margin of bank account ownership. Even among such households, however, the effects of bank and nonbank provider locations on financial transaction services use are not as large as those associated with key household-level attributes, such as income, education, or race.

We examine the association between employee quality and financial reporting outcomes. Using the average workforce education level in MSA(s) where the firm operates as a proxy for employee quality, we find that firms with a high-quality workforce exhibit higher accruals quality, fewer internal control violations, and fewer restatements. These firms also issue superior management forecasts, in terms of frequency, timeliness, accuracy, precision, and bias. Employees located at the firm's headquarters primarily drive our findings. Our evidence suggests employee quality, particularly at a firm's headquarters, is associated with both mandatory and voluntary disclosure quality.

Spending on big-time college athletics is often justified on the grounds that athletic success attracts students and raises donations. We exploit data on bookmaker spreads to estimate the probability of winning each game for college football teams. We then condition on these probabilities using a propensity score design to estimate the effects of winning on donations, applications, and enrollment. The resulting estimates represent causal effects under the assumption that, conditional on bookmaker spreads, winning is uncorrelated with potential outcomes. We find that winning reduces acceptance rates and increases donations, applications, academic reputation, in-state enrollment, and incoming SAT scores.

We investigate the economic effects of capitalizing development costs during a race between two firms to discover and develop a new technology. Winning the race requires success in the research stage and success in the development stage. Development costs are expensed in some settings, but capitalized in others. Capitalization of development costs provides a credible signal regarding progress in the race, allowing the rival to make a more informed decision regarding whether to proceed with development. We study the effects of this signal on the firms’ investment decisions and social welfare. We show that if both firms capitalize instead of expense development costs, aggregate investment in research weakly increases but aggregate investment in development weakly decreases. We also characterize the accounting policies that the two rival firms would adopt if they could freely choose either an expensing policy or a capitalization policy.