Knowledge that Transforms

We examine the implications of GAAP earnings forecast quality for accounting research. Using a tax law change with an estimable and material GAAP earnings impact, we find that analysts’ GAAP forecasts generally fail to incorporate this impact, whereas investors respond promptly, suggesting that GAAP forecasts omit earnings information deemed relevant by investors and are of low quality. Analyzing quarterly GAAP forecasts from 2004–2019 and classifying GAAP forecasts that equal their street counterparts when GAAP and street actuals differ as low quality, we again find widespread low GAAP forecast quality. Low quality GAAP forecasts affect research inferences: they dampen GAAP earnings response coefficient (ERC) estimates, reduce the explanatory power of GAAP surprises for returns, affect inferences regarding market rewards for meeting-or-beating via exclusions, and understate the extent that GAAP forecasts incorporate exclusion components. We propose two strategies to mitigate the adverse effects of low quality GAAP forecasts on research inferences. Data Availability: Data are from publicly available sources as identified within the manuscript.

We examine whether and how a widely established finding—employees respond negatively to managerial monitoring—generalizes to different performance-evaluation systems. We predict that relative performance evaluations (RPEs), a common evaluation feature in multiagent settings, attenuate employees’ negative responses to managerial monitoring. The results of our experiment support our theory. Consistent with prior literature, we find that without RPE, employees respond significantly more negatively to managerial monitoring compared to no monitoring. However, we find that the effect of managerial monitoring on employee responses is moderated in the presence of RPE—employees respond similarly regardless of whether their manager implements a monitoring control. We provide robust analyses to support our theory-derived mechanisms, demonstrating that our results are driven by employees’ fairness perceptions of managers’ monitoring decisions. Collectively, this study aids in the understanding of how and under what circumstances managerial monitoring may be more or less beneficial. Data Availability: Available upon request.

Seasonal fluctuations in a firm’s business activities can affect its balance sheet and give rise to seasonally predictable accruals. We find that seasonal patterns in accruals are associated with future stock returns. Specifically, we find that firms with historically lower (higher) accruals in a given fiscal quarter have higher (lower) stock returns in the months when those accruals are expected to be announced. Our results suggest that investors do not fully understand and price historical information on accruals seasonality. Additional analyses suggest that the emergence of this accruals seasonality anomaly is concentrated in the post-2001 period and driven by the effects of unsophisticated arbitrage against the accruals anomaly.

We examine year-to-year adjustments of performance measure weights in managers’ incentive contracts. Using survey panel data on financial middle managers over four years, we document that higher target achievement on a given measure is associated with an increase in the measure’s relative weight for the subsequent year. We explore three potential explanations for this pattern: retention considerations, managerial influence, and gradual strategic shifts. Consistent with retention considerations, we find that shifts toward better performing measures are stronger for managers who outperform their peers and in firms facing greater labor market competition. Consistent with managerial influence, we find that shifts are also stronger for managers with greater influence on their incentives and in firms with lower incentive design transparency. We find, however, no support for the gradual strategy change explanation, as results do not show that shifts vary with managers’ involvement in strategic decision-making or subsequently observed strategy changes. Data Availability: The survey data used in this project are protected by a nondisclosure agreement.

I examine whether contracts that require customers to privately share with suppliers forecasts of their future demand for the supplier’s products (“demand forecast contracts,” or “DF contracts”) affect the supplier’s reliance on an alternative information source—stock prices—when making investment decisions. If suppliers find these forecasts a more direct signal of future demand than stock prices, they may reduce their reliance on stock prices to guide investments. Using hand-collected data, I find that suppliers’ investments become significantly less sensitive to stock prices after entering a DF contract for the first time. This effect is stronger when forecasts are more credible, demand more uncertain, and investments more irreversible. Supplier performance, measured by return on assets and cash flow from operations, improves post-DF. Overall, these findings suggest that when a relatively direct information source about future demand becomes available, managers reduce their reliance on stock prices in making real decisions. Data Availability: Data are available from public sources cited in the text.

Why do seemingly similar firms show such different productivity? We argue that unresolved measurement problems, i.e., the persistent incongruity between economic transactions and their accounting representation, affect resource allocation. Our metric quantifies technical accounting terminology in firm disclosures to capture these unresolved measurement problems, UMPs, using over 90,000 10-Ks. A one standard deviation increase in UMP is associated with lower capital investment (6 percent), R&D (5 percent), and hiring growth (30 percent). We also find a reduction in total factor productivity (5 percent) and Tobin’s q (4 percent). Further, CEO compensation sensitivity to accounting performance decreases with UMP, whereas stock-based sensitivity remains unaffected. Our inferences continue to hold when we use a Bartik instrument, which exploits differential exposure to GAAP changes to isolate accounting-driven variation from the firm’s underlying economics. The results suggest unresolved measurement problems are a significant friction in resource allocation. Data Availability: The UMP dataset is available upon request. Please contact the first author.

We develop an empirical proxy for companies’ differential communication to local and foreign investors using translation differences in public disclosure. We validate our proxy using a field experiment and then use this proxy to document that differential communication is associated with increases in information asymmetry between local and foreign investors. It is also linked to decreases in the relative information quality of foreign analysts, even when foreign demand for information is high and communication costs are low. These and a variety of supporting tests, including those using an alternative artificial intelligence (AI)-based measure of translation differences, suggest that firms engage in differential communication because of a preference for local investors and when responding to incentives to maximize stock price. This study highlights the role of differential communication as one driver of local information advantage in our setting. Data Availability: Most data are available from publicly available sources, as described in the paper. The field experiment data are available upon request.

We examine how the prospect of research and development (R&D) disclosure affects a firm’s institutional investor base. Difference-in-differences (DiD) regressions around the enactment of the American Inventors Protection Act (AIPA), which mandated the public disclosure of patent applications within 18 months of filing, show that short-term institutional investors increase their holdings before public information is released, whereas long-term investors do not adjust their positions. This anticipatory shift is consistent with theoretical predictions that expected disclosure strengthens short-horizon investors’ incentives to acquire and trade on private information. We further document that stock prices reflect more firm-specific information leading up to disclosure and that improved liquidity at disclosure enables short-term investors to partially unwind their positions. Our paper offers novel evidence on how increases in the expected likelihood of disclosure shape investor behavior and the composition of firms’ investor bases.

Over three-quarters of U.S. taxpayers receive federal tax refunds, largely due to income tax over-withholding. This study explores how over-withholding impacts investments by comparing individuals’ investment behavior following wage receipts and tax refunds. We find a significantly higher marginal propensity to invest out of wages than refunds, suggesting that over-withholding, which alters the labeling and timing of income, meaningfully influences financial decision-making. Our cross-sectional analysis indicates that the differential investment rates are more pronounced for individuals with automatic investment setups and lower financial sophistication. The difference cannot be fully explained by alternative explanations such as lack of awareness, fixed dollar investment goals, timing differences between wages and refunds, uncertainty of refunds, transaction costs, or the perception of refunds as additional windfall income. Our findings underscore the importance of considering behavioral factors in the formulation of tax policies and contribute to the accounting literature that examines taxes and investment behavior.

The FASB recently issued ASU 2024-03, which requires disaggregation of significant expenses, like cost of goods sold (COGS) and selling, general, and administrative (SG&A) expenses. Proponents argue disaggregation will improve decision usefulness, whereas opponents suggest the information will be costly and provide little value. We provide large-sample evidence on the pre-ASU state of expense disaggregation, analyze whether it appears to provide decision-useful information, and explore differences across disaggregation components. Our findings suggest that disaggregation is relatively common, increasing over time, and correlated with demand for disclosure, disclosure incentives, and firm economics. Further, our evidence is consistent with COGS, but not SG&A, disaggregation providing decision-useful information for investors and analysts, and these benefits accrue via improved processing of expense-related news. Overall, our evidence suggests that not all disaggregation is equal. We also identify novel, large-sample expense disaggregation measures for U.S. firms, which are likely useful for evaluating other implications of disaggregation. Data Availability: Data are available from the public sources cited in the text.