Knowledge that Transforms

Abstract The Ohlson (1995) and Feltham and Ohlson (1995) valuation model provides a rigorous framework for summarizing the information in expected future earnings and book values. However, the model provides little guidance on selecting an empirical proxy for expected future earnings. We examine whether and under what circumstances historical earnings and analyst earnings forecasts offer comparable explanation of security prices. This issue is of particular interest because analyst forecasts are less readily available than historical data. Under appropriate circumstances, historical data may allow wider use of the Feltham‐Ohlson valuation model by researchers and investors. A related issue is the incremental explanatory power of historical earnings and realized future earnings (perfect‐foresight forecasts) for security prices beyond analyst forecasts. If historical earnings are incrementally informative, that would suggest that analyst forecasts do not fully reflect price‐relevant information in past earnings. If future earnings are incrementally informative, that would suggest that security prices reflect investors' implicit earnings forecasts beyond analyst forecasts. We examine these issues using a historical model (based on past earnings), a perfect‐foresight model (based on realized future earnings), and a forecast model (based on Value Line earnings forecasts). All three models provide significant explanatory power for security prices, and each set of earnings data provides incremental explanatory power for prices when used with the other sets of earnings data. We estimate the models separately for firms with moderate and extreme earnings‐to‐price (E/P) ratios, a proxy for earnings permanence. For moderate‐E/P firms, the historical model's explanatory power exceeds that of the perfect foresight model, and is indistinguishable from that of the analyst forecast model. In contrast, for extreme‐E/P firms, the perfect‐foresight model offers greater explanatory power than the historical model, but lower explanatory power than analyst forecasts. Our results suggest that financial analysts' forecasting efforts are best focused on firms whose earnings contain large temporary components (extreme E/P firms). However, in general, both historical data and analyst forecasts are complementary information sources for security valuation.

Abstract We investigate capital investment problems when a manager knows the costs of a set of available projects, while the owner only holds probabilistic beliefs about these costs. With mutually exclusive projects, an optimal policy can be defined by a series of cost targets, one for each of the possible projects. The project with the lowest reported cost relative to the target is chosen, and funded as if the cost were equal to the target. The optimal investment policy can deviate from a traditional policy of selecting the project with the highest, positive net present value (NPV) in a number of ways. First, under‐investment arises to limit the manager's ability to capture the economic rents. Second, when investment takes place, it is not always the project with the highest NPV that is implemented. Third, projects with lower cost variability can be favored. We extend the analysis to non‐mutually exclusive projects. With two independent projects, batch processing is superior to individual appraisal whenever both optimal individual appraisal cost targets are interior. Individual appraisal ignores the impact of individual targets on incentives to report the costs of other potential projects. Batch processing can improve individual assessment by cost effective switching of investment away from the individual projects and into the batch as a whole. The results suggest that the common practice of analyzing batches of capital requests in an annual capital budgeting cycle provides advantages in the organization's attempt to deal with asymmetric information and incentive problems.

Abstract This paper investigates the relation between industry‐wide information disclosures by the trade association for the semiconductor industry and both share prices and analyst forecasts. Such disclosures may have little impact on investors and analysts, since prior theoretical research suggests that trade associations may be unable to secure reliable data from firms in an industry. At the same time, such disclosures may be important, since prior empirical research suggests that share prices and analyst forecasts reflect industry‐wide earnings effects earlier than firm‐specific effects. We document significant stock price movements on release dates of industry Flash Reports by the Semiconductor Industry Association (SIA) each month that contain aggregate industry data on new orders and shipments. The magnitude of the price revisions on Flash Report disclosure dates is positively associated with changes in the numbers disclosed and varies across sample firms in a manner associated with identifiable characteristics of the firms. Further tests indicate that the Flash Report provides mainly forward‐looking information on new orders that is linked to firm‐specific sales changes and has explanatory power for quarterly stock prices beyond firm‐specific earnings. This information is used by security analysts mainly in assessing the persistence of firm‐specific quarterly sales changes. Our findings support the hypothesis that the SIA is able to obtain data from firms, compile it into reliable aggregate statistics, and then distribute these statistics in a timely fashion.

Abstract Previous empirical research on the informativeness of earnings has focused on stockholders, and has not examined differences in earnings' informativeness for stockholders and bondholders. Because stockholders are residual claimants and bondholders are fixed claimants, the informativeness of earnings should differ for these two types of investors. When a firm's default risk is low, changes in its financial condition should be of limited relevance to bondholders, but should be relevant to stockholders. In contrast, as the likelihood of financial distress increases, stockholders' limited liability allows them to abandon the firm to the bondholders (Fischer and Verrecchia 1997). Accordingly, as a firm's default risk increases, changes in its financial condition should be increasingly important to bondholders and less important to shareholders. Because earnings provide information on firm value, the stock return‐earnings association should decrease as the firm's financial strength declines, while the bond return‐earnings association should increase. We use two measures of a firm's financial strength: the firm's bond rating and its reporting of a loss. Consistent with our hypotheses, we find that the association between stock returns and changes in annual earnings decreases as bond ratings decline, while the association between bond returns and changes in annual earnings increases. These results suggest that as the company's financial condition deteriorates, earnings become less relevant for stock valuation and more relevant for bond valuation. When we partition firms based on their loss status, we find a stronger association between stock returns and annual earnings changes for firms with positive earnings (profit firms) than for firms with losses, consistent with earlier studies. In contrast, we find that the association between bond returns and earnings changes is greater for loss firms than for profit firms. These results suggest that losses reduce the informativeness of earnings for stockholders but increase informativeness for bondholders, suggesting that investors view losses as indicating increased credit risk.

Abstract This study investigates the extent to which property‐casualty insurers select levels of loss reserves, net capital gains, and net stock transactions to meet solvency and tax reporting goals. Insurer solvency is reflected in financial measures known as IRIS (Insurance Regulatory Information System) ratios. IRIS ratios are generally enhanced by underestimating loss reserves, accelerating the realization of capital gains, postponing the realization of capital losses, issuing stock, and cutting dividends. Taxable income is reduced by reporting higher reserves and lower net capital gains on investments. We use simultaneous equations to model the three discretionary choices individually, while controlling for potential tradeoffs among the decisions. During the sample period of the study (1990‐95), there is a shift in the regulatory environment that we argue tends to reduce incentives to meet IRIS goals. Specifically, risk‐based capital (RBC) requirements were adopted in 1994. Although IRIS ratios continued to be used for solvency screening, their effect is expected to be diluted in the post‐RBC period. Our results provide qualified support for this claim. Evidence of the phenomenon is stronger when the choice variables are net capital gains and stock transactions, and weaker when loss reserves are considered. Two of the three discretionary choices affect taxable income: loss reserves and capital gains. We find that tax incentives are significantly associated with the loss reserve estimate throughout the sample period. In contrast, our results are only weakly consistent with the view that capital gains are timed to achieve tax relief.

Abstract This paper examines the relation between earnings and operating cash flow to derive and test an indicator of financial statement fraud. Accrual measurement concepts indicate that financial statement fraud should be associated with high levels of earnings relative to operating cash flow. We demonstrate that the excess of earnings over operating cash flow is extreme in most fraud cases in years immediately prior to the fraud discovery based on a sample of 56 fraud cases from 1978 to 1991. We compare the distribution of the earnings minus operating cash flow variable for fraud firms with that for a sample of 60,453 firm‐years for firms listed on COMPUSTAT. We test a logistic regression model in which the discovery/nondiscovery of fraud is the dependent variable, and earnings minus operating cash flow is the explanatory variable. Other control variables are included in the model based on prior studies. Results are consistent with expectations derived from accrual measurement theory. We then examine the predictive ability of the model using our sample of fraud firms and a sample of nonfraud firms in the same four‐digit SIC code industries. Observations for the fraud firms are for the fiscal year prior to the discovery of fraud. Observations for the nonfraud firms are for the same fiscal years as the fraud firms in the same industries. The predictive ability of the model, including the excess of earnings over operating cash flow, is substantially higher than the predictive ability of the model omitting this variable. We conclude that the earnings‐operating cash flow relation provides important information for those interested in identifying financial statement fraud, especially when considered in conjunction with other factors associated with fraud risk.

Abstract This paper examines how market prices, volume, and traders' dividend expectations respond to public information releases in laboratory markets for a long‐lived financial asset. The objective is to study deviations from the symmetric information risk‐neutral rational expectations (RE) benchmark, which predicts no trade in such settings. The results of a series of double‐auction and call markets are reported in which traders manage a portfolio of cash and asset shares over 15 rounds of trading. A public signal regarding the value of the liquidating dividend is released every third round, and traders' subjective expectations of the liquidating dividend are elicited each round as cash‐motivated forecasts. We find that, despite the public dividend signal, traders' dividend forecasts are heterogeneous. Forecasts and prices both underreact to the public signals, with prices under‐reacting more than forecasts. In general, price changes are not closely associated with public signals, and there is greater excess price volatility in double auctions than in call markets. Forty‐three percent of trades are inconsistent with the trader's forecasts, and inconsistent trades occur more frequently in the double‐auction markets. On average, approximately 10 percent of the outstanding shares are traded in each round, and trading volume is increasing in the mean absolute forecast revision and decreasing in the contemporaneous dispersion in forecasts. These results suggest that differential processing of the public signal and/or speculative trading for short‐term gain may help to explain why symmetric information RE predictions are often not supported in empirical and experimental settings. They also suggest that market reactions to public information releases may be influenced by market microstructure.

Abstract This study examines the predictive value of Management Discussion and Analysis (MD&A) information. More specifically, this study tests the association between properties of analysts' earnings forecasts and MD&A quality, where MD&A quality is measured by the Securities Exchange Commission (SEC). We find that high MD&A ratings are associated with less error and less dispersion in analysts' earnings forecasts after controlling for many other expected influences on analysts' forecasts. We also find that estimated regression coefficients are consistent with MD&A information having a substantial effect on earnings forecasts. Finally, we find our results are driven by forward‐looking disclosures about capital expenditures and operations, and also by historical disclosures about capital expenditures. These findings are consistent with the suggestion by many constituencies (including the SEC) that the type of information found in high quality MD&A is particularly relevant for predicting earnings.

Abstract Accounting accruals are managers' subjective estimates of future outcomes and cannot, by definition, be objectively verified by auditors prior to occurrence. This causes audits of high‐accrual firms to pose more uncertainty than audits of low‐accrual firms because of potential estimation error and a greater chance that high‐accrual firms have undetected asset realization and/or going concern problems that are related to the high level of accruals. One way that auditors can compensate for this risk exposure is to lower their threshold for issuing modified audit reports, an action that will increase modified reports and, therefore, lessen the likelihood of failing to issue a modified report when appropriate. We call this auditor reporting conservatism and test if high‐accrual firms in the United States, are more likely to receive modified audit reports for asset realization uncertainties and going concern problems. Empirical results for a large sample of U.S. publicly listed companies support the hypothesis that auditors are more conservative, that is, more likely to issue both types of modified audit reports for high‐accrual firms. Further analyses show that income‐increasing accruals are somewhat more likely to result in reporting conservatism than income‐decreasing accruals, and that only the Big Six group of auditors show evidence of reporting conservatism. These findings add to our understanding of the audit report formation process and the potentially important role played by accounting accruals in that process.