We examine whether the complexity of mandatory accounting disclosures prompts managers to voluntarily disclose adjusted measures of actual earnings performance, and whether this practice reflects attempts to obfuscate or mitigate the informational opacity accounting complexity creates for investors. Using the metadata in XBRL filings, we construct measures of accounting complexity that map directly to the mandated standards applied in financial statement filings. We find a positive and economically significant association between accounting complexity and managers’ propensity to disclose non-GAAP earnings information. This relation is robust and incremental to common measures of business and linguistic complexity, and the transitory nature of firms’ economic activities. We also find that the quality and informativeness of adjusted earnings information increases with accounting complexity, consistent with motives to better inform investors when accounting disclosures are complex. Overall, our results suggest that managers use non-GAAP earnings disclosure to mitigate the adverse informational effects of accounting complexity. Data Availability: All data are available from sources identified in the paper.

We examine how firms’ contractual relationships with their employees affect the design of their debt contracts, and the use of financial covenants in particular. Viewing the firm as the nexus of both explicit and implicit contractual relationships, we argue that managers cater to their employees’ preferences when negotiating contractual terms with creditors. We argue that an increase in unemployment-insurance benefits reduces employees’ cost of job loss, which, in turn, allows managers to take more risk. First, we show that more generous benefits are associated with a higher operating leverage, operating cash flow volatility, and product-development frequency. We then find that loans initiated following an increase in unemployment-insurance benefits include a higher proportion of performance, rather than capital covenants. Overall, our study demonstrates how the design of debt contracts changes in response to arguably exogenous changes in employees’ collective tolerance—and, in turn, managers’ preferences—for risk.