Scott A. Richardson

Why are Earnings Kinky? An Examination of the Earnings Management Explanation

Prior research has documented a “kink” in the earnings distribution: too few firms report small losses, too many firms report small profits. We investigate whether boosting of discretionary accruals to report a small profit is a reasonable explanation for this “kink.” Overall, we are unable to confirm that boosting of discretionary accruals is the key driver of the kink. We caution the use of the ratio of small profit firms to small loss firms as a measure of earnings management. We investigate and discuss a number of alternative explanations for the kink.

Credit Markets and Financial Information

The last decade has seen rapid growth in trading of credit instruments on secondary markets. The ensuing availability of a rich set of credit market data has created a novel environment for testing a variety of financial economic theories. In this discussion, we provide a simple framework for linking asset pricing research using equity and credit market data, and offer some suggestions for future archival empirical research aiming to establish relations between financial information and credit markets. Credit instruments are intrinsically linked to equity instruments. The strength of this link varies temporally and cross-sectionally in measurable ways that can, and should be, used to guide future empirical research linking information to credit markets.

An Accounting-Based Characteristic Model for Asset Pricing

We develop a framework to identify firm characteristics that forecast stock returns. We show that forecasting returns is equivalent to forecasting earnings and earnings growth. Thus a characteristic indicates expected returns if it indicates expected earnings and earnings growth that the market prices as being at risk. The model identifies two important characteristics that explain cross-sectional variation in equity returns: (1) earnings-to-price (E/P) and (2) book-toprice (B/P). Because E/P is a yield, it is a valid characteristic to indicate expected returns, though it has not been emphasized in most prior research. B/P is a valid characteristic because it is positively associated with future (risky) earnings growth. This positive correlation is surprising since most prior research labels low B/P stocks as growth stocks. As a validation of our model, we revisit the puzzling negative relation that has been observed between leverage and realized returns. We find evidence of a positive relation between leverage and returns only when returns are conditioned on the set of characteristics identified by our model.The paper presents an accounting framework for identifying characteristics that indicate expected returns. A model links expected returns to expected earnings and earnings growth, so a characteristic indicates expected returns if it indicates expected earnings and earnings growth that the market prices as being at risk. In applying the framework, the paper confirms book-to-price (B/P) as a valid characteristic in asset pricing: B/P is associated with higher expected earnings growth and also captures the risk of that growth not being realized. However, the framework also points to the forward earning-to-price (E/P) as a risk characteristic. Indeed, E/P, rather than B/P, is the relevant characteristic when there is no expected earnings growth, but the weight shifts to B/P with growth. The framework also enables the separation of the expected return for operating risk from that due to financing risk. With this separation, the paper revisits the puzzling negative relation that has been observed between leverage and realized returns, a finding that has been attributed to failure to control for operating risk. We find a positive relation between leverage and returns when operating risk characteristics identified by our model are recognized.

Corporate Governance, Fees for Non-audit Services and Accrual Choices

We examine the relation between the relative amount of fees paid to auditors for non-audit services and the behavior of accrual measures. We extend prior research in two important directions. First, using a pooled sample of 2,295 firms for the fiscal year 2000, we find very little evidence of a relation between the provision of non-audit services and measure of accruals. However, there appears to be three distinct clusters of firms where only one cluster (consisting of only about 20 percent of the sample) exhibits a statistically positive association between non-audit fees and accrual behavior. Second, we examine the corporate governance characteristics of firms in the cluster with a positive association between non-audit fees and accrual behavior relative to the remaining firms. We find that this subset of firms have a smaller market capitalization, lower institutional holdings, higher insider holdings, smaller board of directors (and audit committee), and lower percentage of independent board (and audit committee) members. These results suggest that the provision of non-audit services is potentially problematic only for a small subset of firms that appear to be de facto controlled by management.

Asset Measurement Uncertainty and Credit Term Structure

In this paper, we test the impact of uncertainty in asset measurement on credit term-structure. The theory of Duffie and Lando (2001) suggests that the inability of creditors to assess asset values precisely will support the existence of non-zero short term credit spreads. Developments in recent years enable us to directly test this theory. First, the development of the credit default swap market allows for examination of more precise and cross-sectionally comparable short term credit spreads. Second, FAS 157, which requires detailed disclosures of the fair values of financial assets to be included in financial statements, provides us a proxy for asset measurement uncertainty. Third, the financial crisis over the last two years presents a natural setting where asset measurement uncertainty was in focus, allowing for powerful tests of the theory. For a sample of U.S. financial institutions over the period August 2007 to March 2009, we find strong support for the Duffie and Lando (2001) theory. Specifically, we find that asset measurement uncertainty, attributable to Level 2 and Level 3 financial assets, is a significant determinant of short term credit spreads. Our findings are robust to a variety of control variables and research design choices.

Discussion of Consequences of Financial Reporting Failure for Outside Directors: Evidence from Accounting Restatements and Audit Committee Members

Srinivasan examines turnover of outside directors in firms experiencing a restatement. He finds that outside directors (in particular outside members of the audit committee) are more likely to leave the board of a company that completes a restatement and to subsequently lose directorships at other firms. These results are interpreted as evidence of outside directors bearing reputation costs from financial reporting failures. Understanding the reputation consequences for outside directors following a financial reporting failure is an important research question. This type of analysis offers insights into the design of governance structures and possibly the managerial labor market serving an ex post role in disciplining managerial activity. Srinivasan extends the existing research examining whether firms experiencing restatements were characterized by weak governance at the time of the alleged accounting manipulation (e.g., Beasley [1996], Dechow, Sloan, and Sweeney [1996], Farber [2005], Desai, Hogan, and Wilkins [2004], Agarwal and Chadha [2005]). Collectively, this research

Accruals and Future Performance: Can It be Attributed to Risk?

We decompose broad based measures of accruals into firm specific and related firm components. We find that the negative relation between accruals and future firm performance is almost entirely attributable to the firm specific component. Standard risk based explanations are hard to reconcile with this fact. To the extent expected returns have a common component spanning related firms, a risk based explanation would suggest a stronger negative relation between accruals and future firm performance when related firms are also growing. Instead, the attenuation we document is more likely attributable to sub-optimal investment decisions, which the stock market and analysts do not incorporate in a timely manner.

Discussion—A League of Their Own? Financial Analysts' Responses to Restatements and Corrective Disclosures

Griffin (2003) examines the decisions made by a set of relatively informed capital market participants around the announcement of a corrective disclosure that gave rise to a class action lawsuit. The capital market participants include financial analysts, institutions, short sellers, and insiders. He finds no evidence that financial analysts adjust their forecasts, nor do they drop coverage in anticipation of the corrective disclosure announcement. He does, however, find strong evidence of a revision in forecasts and termination of coverage after the corrective disclosure is announced. In contrast, other informed capital market participants do appear to alter their behavior in the period leading up to the announcement of the corrective disclosure. For example, Griffin finds evidence of increased short selling in the period leading up to the announcement (see also Dechow, Sloan, and Sweeney [1996] for a similar finding for the announcement of SEC enforcement actions). Examining the behaviors of informed capital market participants around a significant “bad news” event is a useful exercise. It allows an assessment of the timeliness with which information is impounded into the decisions of these market participants. Griffin’s findings suggest that financial analysts do not provide particularly timely information to the investment community. Analysts appear to wait until the “bad news” is announced before altering their behavior. The findings for short sellers (and to a lesser extent insiders and institutions) suggest that some informed capital market participants are trading in a manner consistent with the impending “bad news” of the corrective disclosure. These findings corroborate with an extensive literature in finance that has examined the ability for short interest positions to predict future stock price performance (e.g., Figlewski [ 19811; Dechow, Hutton, Meulbroek, and Sloan [2001]; Desai, Thiagarajan, Ramesh, and Balachandran [2002]). The remainder of the discussion is organized as follows. First, I address the philosophical question: “Does the fact that we observe a large stock price reaction to the announcement of a corrective disclosure mean that no one was trading on the basis of this information?” Second, I discuss in detail the results on how in-

Prepayment Risk and Expected MBS Returns

We present a simple, linear asset pricing model of the cross section of Mortgage-Backed Security (MBS) returns. We measure prepayment risk and estimate security risk loadings using real data on prepayment forecasts vs. realizations. Estimated loadings are monotonic in securities’ coupons relative to the par coupon, as predicted by the model. Prepayment risks appear to be priced by specialized MBS investors. In particular, we find convincing evidence that prepayment risk prices change sign over time with the sign of a representative MBS investor’s exposure to prepayment risk.

Does Fundamental Volatility Help Explain Credit Risk

We examine whether fundamental measures of volatility are incremental to market based measures of volatility in (i) predicting bankruptcies (out of sample), (ii) explaining cross-sectional variation in credit spreads, and (iii) explaining future credit excess returns. Our fundamental measures of volatility include (i) historical volatility in profitability, margins, turnover, operating income growth, and sales growth, (ii) dispersion in analyst forecasts of future earnings, and (iii) quantile regression forecasts of the interquartile range of the distribution of profitability. We find robust evidence that these fundamental measures of volatility improve out of sample forecasts of bankruptcy and are useful in explaining cross-sectional variation in credit spreads. This suggests that an analysis of credit risk can be enhanced with a detailed analysis of fundamental information. As a test case of the benefit of volatility forecasting, we document an improved ability to forecast future credit excess returns, particularly when using fundamental measures of volatility.