This study presents empirical evidence on the ex post costs of employee stock option (ESO) grants to issuing firms and examines whether the Black–Scholes [1973] model provides reasonable estimates of these values. Because there are no market prices for ESOs, the traditional avenues for testing option–pricing models are unavailable. This research relies instead on techniques from the economic forecasting literature, viewing model values as forecasts of the options’ payoff. The theoretically appropriate rate at which to discount ESO payoffs is derived under the maintained hypothesis that the Black–Scholes model is valid. This rate is used in estimating ex post ESO costs at the time of grant, which are then compared with Black–Scholes estimates using Theil’s [1966] tests of forecast rationality. Based on a sample of 966 ESO grants over 1963–1984, the results suggest that the Black–Scholes model, adjusted for concavity in the time to exercise using the Hemmer, Matsunaga, and Shevlin [1994] procedure, appears to provide reasonable estimates of ex post ESO costs for the average ESO grant. However, there is significant variability in the amount of model error on an individual grant basis.

This paper provides evidence on the characteristics of firms that include “pro forma” earnings information in their press releases, whether the usefulness of pro forma earnings to investors varies systematically with these characteristics, and whether the investor response to pro forma earnings is consistent with market efficiency or mispricing. Using a sample of 249 press releases from 1997–99, we find that firms with low GAAP earnings informativeness are more likely to disclose pro forma earnings than other firms. We also find that strategic considerations, measured using the direction of GAAP earnings surprises, are an important determinant of pro forma reporting. In addition, our examination of the relative and incremental information content of pro forma earnings shows that investors find pro forma earnings to be more useful when GAAP earnings informativeness is low or when strategic considerations are absent. Tests of the predictive ability of pro forma earnings for future profitability and returns are mixed, and we therefore cannot conclusively determine whether the investor reaction to pro forma earnings at the time of the press release is consistent with market efficiency or mispricing. The paper contributes to the growing literature on pro forma earnings and more generally to the literature on voluntary and strategic disclosure.

ABSTRACT In this article we examine whether firms structure their convertible bond transactions to manage diluted earnings per share (EPS). We find that the likelihood of firms issuing contingent convertible bonds (COCOs), which are often excluded from diluted EPS calculations under Statement of Financial Accounting Standard (SFAS) 128, is significantly associated with the reduction that would occur in diluted EPS if the bonds were traditionally structured. We also document that firms' use of EPS‐based compensation contracts significantly affects the likelihood of COCO issuance and find weak evidence that reputation costs, measured using earnings restatement data, play a role in the structuring decision. These results are robust to controlling for alternative motivations for issuing COCOs, including tax and dilution arguments. In addition, an examination of announcement returns reveals that investors view the net benefits and costs of COCOs as offsetting one another. Our results contribute to the literature on earnings management, diluted EPS, financial reporting costs, and financial innovation.

We examine the effect of voluntary adoption of clawback provisions on non-GAAP earnings disclosures. Prior literature documents that voluntary clawback adoption improves financial reporting quality by increasing the costs of misstating GAAP earnings. However, managers may respond to perceptions of reduced discretion over GAAP reporting by increasing their reliance on non-GAAP earnings disclosures. Using a propensity score matched sample, we find that non-GAAP earnings disclosure frequency increases and non-GAAP exclusion quality decreases after clawback adoption, consistent with a more opportunistic use of non-GAAP reporting. Additional cross-sectional tests help support this interpretation.

Abstract There is relatively little evidence on the specific accruals used to manage earnings. This paper examines this issue by considering the use of specific accruals in three earnings‐management contexts: equity offerings, management buyouts, and firms avoiding earnings decreases. We argue that the costs of managing earnings through different income statement items vary and that the benefits of earnings management through each of these items depend on the context. We thus make differential predictions regarding which specific accrual will be used to manage earnings in each of the three contexts we consider. To measure earnings management for specific accruals, we develop performance‐matched measures to capture the unexpected component of accounts receivable, inventory, accounts payable, accrued liabilities, depreciation expense, and special items. Consistent with our predictions, we find that firms issuing equity appear to prefer managing earnings upward by accelerating revenue recognition. Specifically, we find that accounts receivable for these firms are unexpectedly high. Conversely, for the management buyout context, we predict and find unexpected accounts receivable to be negative. For firms trying to avoid reporting an earnings decrease, we expect firms to be less concerned with earnings persistence and therefore more likely to use more transitory, and less costly, items to achieve their goal. We find that special items are significantly more positive for this group. This paper provides a further step toward understanding how the incentives behind earnings management affect the method used to achieve earnings goals, and it illustrates the usefulness of examining individual accruals in specific contexts.

Abstract This paper examines the relation of voluntary disclosure of management earnings forecasts and information asymmetry to insider selling through secondary equity offerings. We hypothesize that the pattern of voluntary disclosure and level of information asymmetry prior to secondary equity offerings differs systematically based on the identity of the seller. Specifically, we predict a greater frequency of voluntary disclosure and decreased level of information asymmetry when managers sell their stock through a secondary offering. We examine this hypothesis in a cross‐sectional analysis of 210 secondary equity offerings from 1984‐91, using a two‐stage conditional maximum likelihood simultaneous equations estimation procedure, which allows for possible endogeneity in the manager's decision to sell stock. Consistent with our predictions, we document a significantly positive association between managerial participation and voluntary disclosure of earnings forecasts in the nine‐month period prior to registration of the offering. We also document a significantly negative association between managerial participation and two proxies for information asymmetry. The findings provide evidence that managers act as if reduced information asymmetry correlates with a reduced cost of capital.

This paper examines accounting and nonaccounting factors behind accounting losses over a 50-year period. Using multivariate time-series analysis, we report evidence that the annual percentage of losses for U.S. firms is significantly related to accounting conservatism, Compustat coverage of small firms, real firm performance as measured by cash flows from operations, and business cycle factors. We further find that nonaccounting factors tend to play the dominant role in explaining accounting losses over our sample period. Our results are robust to alternative definitions of macroeconomic productivity, as well as to varying model specifications. Our findings contribute to the literature on accounting losses and accounting conservatism and have implications for the use of accounting loss information in numerous settings.