Although rapidly evolving, quality control (QC) systems are a poorly understood determinant of audit quality. We interview 27 QC system leaders to understand how they navigate the challenges they face in changing QC systems. We find that many challenges—including obtaining buy-in, evaluating costs and benefits, and advancing proactive over reactive changes—are caused by conflicting demands arising internally. Consistent with institutional theory, our data reveal that leaders respond to conflicting demands by seeking partial conformity (e.g., negotiating among stakeholders) or by altering the framing of the demands (e.g., using scientific logic to legitimize a QC change). Interviews of eight QC system users complement and corroborate our main results and suggest opportunities for future research. Our study sheds light on how firms update QC systems and informs practitioners, regulators, and academics of the forces that shape the evolution of QC systems. Data Availability: Our data are not publicly available to preserve participant anonymity

In examining the loan contracting implications of SEC investigations, we document that banks charge higher loan spreads when borrowers are under investigation, with the rise in interest rates varying predictably with lender characteristics. Further, our evidence implies that the debt pricing impact of SEC investigations is amplified for borrowers suffering worse credit quality and information asymmetry as well as those relying more on bank loans. These findings suggest that banks perceive increased risk for borrowers under SEC scrutiny while also leveraging their knowledge of the investigations to extract rents. Supplemental analyses reveal tighter nonspread loan terms and a higher likelihood of amending existing loan contracts during SEC investigations. Additionally, the tightening of loan terms reverses for investigations that conclude without enforcement actions. Overall, our research identifies an economic cost of SEC investigations and alerts regulators to these costs when deciding whether to launch an investigation. Data Availability: All data used are available from the sources indicated in the paper

We examine the consequences of the 2020 amendment to Exchange Act Rule 12b-2, which exempted low-revenue issuers from the ICFR audit requirement and reignited the debate among regulators, issuers, auditors, and academics about the costs and benefits of ICFR audits. We find that exempt issuers rarely obtain ICFR audits voluntarily, indicating they generally do not view such audits as cost beneficial. We also find that exempt issuers report fewer MWs than nonexempt issuers, although this effect is driven by a recent increase in MWs for nonexempt issuers rather than a decline in MW for exempt issuers. We find no evidence that exempt issuers misstate more frequently or that they have less informative control effectiveness disclosures. Overall, our findings suggest that the exemption was welcomed by affected issuers and did not materially impair reporting quality, informing the literature and ongoing policy discussions regarding the appropriate scope of ICFR audit requirements. Data Availability: Data are publicly available from the sources identified in the text

I examine whether mandating the disclosure of investments influences firms’ strategic interactions. I exploit an SEC regulation requiring firms to report off-balance sheet purchase obligations, such as commitments to inventory purchases, CAPEX, R&D, and advertising. Motivated by theory on strategic investments, I predict and find that firms increase investments if they have substitutive product market strategies with competitors and decrease investments if they have complementary strategies. This two-way finding is consistent with firms strategically using investments to influence competitors’ behavior. I show that changes in investments are concentrated among dominant firms (i.e., oligopolistic firms with large market shares), especially those with more irreversible investments, which have a greater ability to influence competitors’ actions. I also show that such changes in investments have real effects on firms’ sales growth, profit margins, and market share. Collectively, my results illustrate a novel channel through which financial reporting shapes firms’ investments and competition

Companies across industries face increasing pressure to assess the costs of decarbonizing their operations. This paper develops a generic model for constructing abatement cost curves in connection with carbon dioxide emissions. The resulting abatement cost curves provide a planning tool for companies seeking to project their decarbonization pathways and to determine optimal abatement levels in response to environmental regulations such as carbon pricing. We calibrate our model in the context of European cement producers that are required to obtain emission permits under the European Emissions Trading System. We find that a price of €85 per ton of carbon dioxide, as observed on average in 2023, incentivizes firms to reduce their annual direct emissions by about one-third relative to the status quo. Yet, this incentive increases sharply when prices rise above the benchmark of €100 per ton of carbon dioxide. Data Availability: Data used in this study are referenced in the paper and the Appendix. Data underlying the plots are provided in a supplemental material file Constructing Abatement Cost Curves - Supplementary Data.xlsx. Additional information is available upon request to the first and third authors

Algorithm designers increasingly optimize not only for accuracy, but also for the fairness of the algorithm across pre-defined groups. We study the tradeoff between fairness and accuracy for any given set of inputs to the algorithm. We propose and characterize a fairness-accuracy frontier, which consists of the optimal points across a broad range of preferences over fairness and accuracy. Our results identify a simple property of the inputs, group-balance, which qualitatively determines the shape of the frontier. We further study an information-design problem where the designer flexibly regulates the inputs (e.g., by coarsening an input or banning its use) but the algorithm is chosen by another agent. Whether it is optimal to ban an input generally depends on the designer's preferences. But when inputs are group-balanced, then excluding group identity is strictly suboptimal for all designers, and when the designer has access to group identity, then it is strictly suboptimal to exclude any informative input

Although audit regulators worldwide concur that the rigor of audit firms’ internal engagement quality reviews (EQRs) is paramount to an audit firm’s audit quality, they are concerned that relationships between audit engagement partner (AEP) and engagement review partner (ERP) working in the same firm can undermine these reviews. We utilize confidential proprietary ERP and audit adjustment data to investigate how school alma mater ties between AEPs and ERPs affect audit adjustments in China. Our results reveal that audit adjustments are significantly lower when an AEP and ERP share school ties. This effect is more pronounced when the ERP is less competent than and lacks authority over the AEP, when the ERP and AEP perform mutual reviews or previously coaudited a client, and when the AEP’s work is flawed. Finally, semistructured partner interviews corroborate our empirical findings and strengthen our practical and policy implications. Data Availability: These proprietary data were provided to us with the understanding that it would be used only for independent academic research. Our research team is obligated under nondisclosure and confidentiality conditions to not disclose or share the confidential information which includes all of the data received from the data providers of the Chinese government and not to report information in any article or presentation that would inadvertently reveal (including any potential link) the identity of the partners, firms, their characteristics, and the audit adjustment data. No other restrictions were imposed by the data providers regarding the use of the data purely for academic research purposes and in communicating the research findings. Other data not subject to this confidentiality nondisclosure are available from sources identified in the text