Capital Market Consequences of Earnings Quality

The empirical accounting literature has developed a variety of proxies for earnings quality (see, e.g., the surveys by Schipper and Vincent, 2003, Dechow and Schrand, 2004, Francis, Olsson, and Schipper, 2006, and Dechow, Ge, and Schrand, 2010). Most of them are based on intuitive and plausible conceptions about desirable characteristics of an accounting system. Despite the wide body of empirical research on earnings quality, there is no consensus regarding the interpretation of the information conveyed by the measures; thus, there is little guidance on what measure is appropriate to address specific research questions. This project aims at contributing to this debate by exploring different aspects of the capital market consequences of the earnings quality measures. The results of this work should provide a better understanding of what the earnings quality measures capture and, hence, on how they can be used. The findings should also have relevant implications for regulators, both in the standard setting and in the post-implementation review process. In this study we plan to consider the commonly used earnings quality measures to address the three following research questions: (1) How does the stock market value earnings quality attributes? We examine the association between earnings quality measures and future excess (or abnormal) returns. Excess returns are computed as the difference between actual returns and expected returns (which correspond to the ex- ante cost of equity capital). Excess returns can result from errors in computing expected returns and from market mispricing. The analysis help us to disentangle these effects and thus sheds light on how the information implied by the earnings quality attributes is incorporated into prices. (2) Do earnings quality attributes affect stock price informational efficiency? Our tests of informational efficiency (stock return autocorrelations and variance ratios) are based on the assumption that informational efficient prices follow a random walk. This part of the project helps us to understand the extent and the speed with which the market incorporates information conveyed by earnings quality attributes. (3) How does the bond market value earnings quality attributes? We study the association between earnings quality measures and future excess (or abnormal) bond returns. The research design is analogous to the one used in the stock market (research question 1) and the results allow us to examine the differential response of bond versus equity investors to earnings quality information.

This project examines several important aspects of the capital market consequences of earnings quality using archival empirical methods and large U.S. firms samples, as the U.S. has the most efficient capital markets and is therefore appropriate for studies of earnings quality. The results of this research provide a better understanding of what the earnings quality measures capture and, hence, for what analyses they can be used appropriately. Besides academics, the findings have implications for firms that want to improve the quality of their financial reports, for auditors, investors and financial analysts, and for standard setters, both in the standards development and in post-implementation reviews. The project comprises three main parts that are interrelated.1. Assessment how well empirical measures of earnings quality measure decision usefulness for investors, based on how much of the market mispricing each measure can explain. The measures are persistence, predictability, two measures of smoothness, abnormal accruals, accruals quality, the earnings response coefficient, and value relevance. Accrual-based measures perform best, followed by smoothness and market-based measures.2. Analysis of the relation between managerial discretion in accounting accruals and future informational efficiency. Under the presumption that efficient prices follow a random walk, we measure informational efficiency by using stock return variance ratios. We find that managerial discretion in accruals is positively associated with informational efficiency, which is consistent with the view that discretionary accruals convey useful information to investors and enhance the price convergence to the fundamental value.3. Analysis of the relation between future bond returns and a large number of accounting-based fundamental variables, including earnings quality measures, that are related to future stock returns. We find that the frequency of significant returns to trading strategies based on these anomalies in the bond market is similar to that found in the stock market. While the significance and Sharpe ratios of these returns are comparable to stocks, the magnitude of returns is generally lower for bonds.