We investigate whether income smoothing affects the usefulness of earnings for contracting through the monitoring role of earnings‐based debt covenants. First, we examine initial contract design and predict that income smoothing will increase (decrease) the use of earnings‐based covenants if income smoothing improves (reduces) the usefulness of earnings to monitor borrowers. We find that private debt contracts to borrowers with greater income smoothing are more likely to include earnings‐based covenants. A structural model that explores the cause of this relationship provides evidence that smoothing improves the ability of earnings to reflect credit risk. Second, we examine technical default following contract inception. We find that income smoothing is associated with a lower likelihood of spurious technical default (when the borrower's economic performance has not declined but the loan nevertheless enters technical default). In contrast, we find no association between income smoothing and performance technical default (when the borrower's economic performance has declined). Collectively, this evidence is consistent with income smoothing improving the effectiveness of earnings‐based information in monitoring borrowers.

We examine how fair value accounting affects debt contract design, specifically the use and definition of financial covenants in private loan contracts. Using SFAS 159 adoption as our setting, we find that a small but significant proportion of loans (14.5%) modify covenant definitions to exclude the effects of SFAS 159 fair values. Only a limited number of these modifications exclude assets elected at fair value (less than 7%), while all exclude liabilities elected at fair value. Notably, we document that covenant definition modification is unassociated with ex ante fair value elections. We find that covenant definition modification positively varies with common incentive problems attributed to fair value accounting and negatively varies with benefits attributed to fair value accounting. Our results suggest that fair value accounting is not uniformly detrimental for debt contracting and fair value adjustments are included when they are most likely to improve performance measurement.

We provide the first evidence that prudential principles shape bankers' executive compensation, a phenomenon we call “pay for prudence” (PfP). We conjecture that PfP incentivizes bankers to balance shareholders' preference for risk with regulators' preference for prudence. Although PfP terms are often used in bank compensation contracts, we find that the use of detailed and concrete PfP terms are positively associated with equity incentives for risk-taking. Furthermore, detailed and concrete PfP terms are associated with lower tail risk, fewer bad loans, and lower likelihood of regulatory downgrades. While we do not find evidence that PfP is associated with lower profitability, PfP is associated with more diversified loan portfolios and reduced exposure to real estate. Our results shed light on a new dimension of bankers' pay and suggest that PfP-based incentives complement widely studied equity-based incentives for risk-taking by acting as guard rails that guide managers’ pursuit of investment opportunities.

We document a novel trend in syndicated lending where some participants voluntarily waive their rights to borrowers' private information. We posit that “public-side” lending emerged to facilitate broad lender participation in the syndicated loan market by mitigating concerns about the leakage of borrowers' private information into public securities markets. In line with this proposition, we find that public-side lending facilitates the loan market participation of lenders for which maintaining robust information barriers is particularly costly. Furthermore, while public-side lending increases within-syndicate information asymmetry, our findings indicate that it does not materially increase interest spreads and is associated with lower coordination costs among syndicate participants. Collectively, we document how debt contracting practices evolved to address frictions associated with the protection of borrowers’ private information and the related changes in loan contracting equilibria.

We examine the relation between accounting conservatism and creditor recovery rates for firms in default. We also test the link between conservatism and the length of bankruptcy resolutions. We find that creditors of firms with more conservative accounting before default have significantly higher recovery rates, and that this positive relation is more pronounced for default firms that violated covenants before the default. We also find that conservative firms have higher asset productivity, shorter bankruptcy resolution, and a significantly higher probability of emerging from bankruptcy. These results suggest that accounting conservatism preserves firm value, leading to higher creditor recovery upon borrower default.

We view audit-quality choice as one among many that managers make to maximize firm value. We question whether audit-quality differences among publicly traded companies are of significant interest to investors, clients, and auditors and ask for research on this topic. Relatedly, we ask for research on whether auditors and their clients show behavior consistent with regulated audit quality exceeding the audit quality level demanded absent regulation. We propose that researchers incorporate the competitive advantages of auditors and the institutional features of the audit process into the definition of audit quality. We propose that audit quality research test for externalities and inefficiencies to understand whether auditors and their clients are choosing the efficient level of audit quality. We note the legislative, judicial, and executive powers residing in the PCAOB.

We examine whether lenders use analyst forecasts of the borrower's earnings as inputs when establishing covenant thresholds in private debt contracts. We find that, among debt contracts that include an earnings covenant, earnings thresholds are set closer to analyst forecasts when analysts have historically issued more accurate earnings forecasts. These results are robust to firm fixed effects and an instrumental variable approach. Further, we find that, following a plausibly exogenous decline in the availability of analyst earnings forecasts, debt contracts are less likely to include earnings covenants. Our evidence is consistent with lenders using analyst earnings forecasts as an input when establishing debt covenant thresholds and suggests sell-side analysts play a role in debt contracting.

We study whether recognition versus disclosure affects debt contracting using SFAS 158 as a setting. Upon recognition of previously disclosed pension liabilities, we find that debt contracts of firms that have high pension underfunding contain relatively more covenants based on the balance sheet and a lower cost of debt. Additional analysis suggests that improved reliability of recognized information is a likely mechanism for these findings. We also document that recognition did not affect the credit risk of the borrower or the use of income statement–based covenants that were unaffected by the accounting standard change. Collectively, our evidence suggests that recognition of previously disclosed accounting information can facilitate the incorporation of that information into lenders’ screening and monitoring. Data Availability: Data used in this study are available from public sources identified in the study.

Despite private access to managers, issuer-paid credit rating agencies (CRAs) are often criticized for failing to promptly reflect material negative private information in their ratings and being ineffective corporate watchdogs. We utilize a novel dataset of private SEC investigations to examine the timeliness and informativeness of CRAs’ rating adjustments in response to material negative private information. Our evidence suggests that CRAs adjust ratings downward within a quarter following the opening of an SEC investigation. Moreover, these adjustments are over three times larger for those investigations that ultimately yield an enforcement action than for those that do not, suggesting that CRAs quickly form sophisticated expectations about the materiality of the private information. Additionally, rating downgrades during the investigations are more informative to the stock market than other rating downgrades. Overall, our evidence suggests that CRAs reduce information asymmetry in the capital markets by timely incorporating material private information in their ratings. Data Availability: All data are available from the sources cited in the text.