Jennifer Francis, Per Olsson, Dennis R. Oswald, Comparing the Accuracy and Explainability of Dividend, Free Cash Flow, and Abnormal Earnings Equity Value Estimates, Journal of Accounting Research, Vol. 38, No. 1 (Spring, 2000), pp. 45-70

This paper investigates how data requirements often encountered in archival accounting research can produce a data-restricted sample that is a non-random selection of observations from the reference sample to which the researcher wishes to generalize results. We illustrate the effects of non-random sampling on results of association tests in a setting with data on one variable of interest for all observations and frequently-missing data on another variable of interest. We develop and validate a resampling approach that uses only observations from the data-restricted sample to construct distribution-matched samples that approximate randomly-drawn samples from the reference sample. Our simulation tests provide evidence that distribution-matched samples yield generalizable results. We demonstrate the effects of non-random sampling in tests of the association between realized returns and five implied cost of equity metrics. In this setting, the reference sample has full information on realized returns, while on average only 16% of reference sample observations have data on cost of equity metrics. Consistent with prior research (e.g., Easton and Monahan The Accounting Review 80, 501–538, 2005), analysis using the unadjusted (non-random) cost of equity sample reveals weak or negative associations between realized returns and cost of equity metrics. In contrast, using distribution-matched samples, we find reliable evidence of the theoretically-predicted positive association. We also conceptually and empirically compare distribution-matching with multiple imputation and selection models, two other approaches to dealing with non-random samples.

We examine how the criteria for choosing estimation samples affect the ability to detect discretionary accruals, using several variants of the Jones (1991) model. Researchers commonly estimate accruals models in cross-section, and define the estimation sample as all firms in the same industry. We examine whether firm size performs at least as well as industry membership as the criterion for selecting estimation samples. For U.S. data, we find estimation samples based on similarity in lagged assets perform at least as well as estimation samples based on industry membership at detecting discretionary accruals, both in simulations with seeded accruals between 2% and 100% of total assets and in tests examining restatement data and AAER data. For non-U.S. data, we find industry-based estimation samples result in significant sample attrition and estimation samples based on lagged assets perform at least as well as estimation samples based on industry membership, both in simulations and in tests examining German restatement data, with substantially less sample attrition.

We examine the association between chief executive officer (CEO) reputation (proxied by the extent of press coverage) and the quality of the firm's earnings (proxied by two accruals-based measures). We test three explanations for an association between these constructs: the efficient contracting hypothesis suggests that reputed CEOs are associated with good earnings quality, while the rent extraction and matching explanations argue that reputed CEOs are associated with poor earnings quality. Using a simultaneous equations system to capture the endogeneity of the constructs, we find (consistent with the rent extraction and matching arguments) that more reputed CEOs are associated with poorer earnings quality than are less reputed CEOs. Further tests find little support for the rent extraction hypothesis. We conclude that the reason more reputed CEOs are associated with poor earnings quality firms is that such firms require more talented managers and, therefore, employ more reputed CEOs. © CAAA.

We examine the relation between the cost of equity capital and seven attributes of earnings: accrual quality, persistence, predictability, smoothness, value relevance, timeliness, and conservatism. We characterize the first four attributes as accounting-based because they are typically measured using accounting information only. We characterize the last three attributes as market-based because proxies for these constructs are typically based on relations between market data and accounting data. Based on theoretical models predicting a positive association between information quality and cost of equity, we test for and find that firms with the least favorable values of each attribute, considered individually, generally experience larger costs of equity than firms with the most favorable values. The largest cost of equity effects are observed for the accounting-based attributes, in particular, accrual quality. These findings are robust to controls for innate determinants of the earnings attributes (firm size, cash flow and sales volatility, incidence of loss, operating cycle, intangibles use/intensity, and capital intensity), as well as to alternative proxies for the cost of equity capital.

For a broad sample of firms, we use structural equations modeling to construct latent variables for real-action aggressiveness and reporting policy aggressiveness. We estimate the association between the latent variables and the associations of each latent variable with shareholder payoffs (returns) and CEO payoffs (annual compensation to the CEO position). Results show the two types of aggressiveness are positively correlated but have different associations with the payoffs we consider. Greater policy-choice aggressiveness is associated with higher returns and compensation; the opposite is true for greater real-action aggressiveness. We find a positive association between policy-choice aggressiveness and restatement likelihood. Compared with nonrestatement firms, abnormal returns of restatement firms with aggressive policy choices are larger in the pre-restatement period and lower in the post-restatement period. Negative returns at the restatement announcement do not, on average, eliminate long-run (multi-year) positive returns of the pre-restatement period or of the period whose results are restated.

We examine the properties of a returns‐based representation of earnings quality, estimated from firm‐specific asset‐pricing regressions augmented by an earnings quality mimicking factor. The coefficient on the earnings quality factor (the “e‐loading”) captures the sensitivity of the firm's returns to earnings quality in a given year or quarter, analogous to beta as a measure of the sensitivity of returns to market movements. Relative to other proxies for earnings quality, e‐loadings can be calculated for larger samples of firms and can be estimated for shorter intervals at any point in time. Along all dimensions examined, we find that e‐loadings perform well in capturing notions of earnings quality.

We examine the properties of a returns-based representation of earnings quality, estimated from firm-specific asset-pricing regressions augmented by an earnings quality mimicking factor. The coefficient on the earnings quality factor (the “e-loading”) captures the sensitivity of the firm's returns to earnings quality in a given year or quarter, analogous to beta as a measure of the sensitivity of returns to market movements. Relative to other proxies for earnings quality, e-loadings can be calculated for larger samples of firms and can be estimated for shorter intervals at any point in time. Along all dimensions examined, we find that e-loadings perform well in capturing notions of earnings quality.