Evidence of the Abnormal Accrual Anomaly Incremental to Operating Cash Flows Evidence of the Abnormal Accrual Anomaly Incremental to Operating Cash Flows Presented by Agnes CHENG Professor University of Houston #2005/06-13 The views and opinions expressed in this working paper are those of the author(s) and not necessarily those of the School of Accountancy, Singapore Management University. Evidence of the Abnormal Accrual Anomaly Incremental to Operating Cash Flows C.S. Agnes Cheng* University of Houston Securities and Exchange Commission chengasec.gov Wayne Thomas School of Accounting Michael F. Price College of Business University of Oklahoma Norman, OK 73019-4004 wthomasou.edu, ph: 405-325-5789 Revised November, 2005 *The Securities and Exchange Commission, as a matter of policy, disclaims responsibility for any private publication or statement by any of its employees. The views expressed herein are those of the author and do not necessarily reflect the views of the Commission or of the authors colleagues upon the staff of the Commission. We are thankful for comments received from two anonymous reviewers. We also thank Kriengkrai Boonlert-U-Thai, Ilia Dichev, Yeyun Sejati, Terry Shevlin, Glyn Winterbotham, Hong Xie, and workshop participants at American University, Baruch College, and Oklahoma State University for useful suggestions.Evidence of the Abnormal Accrual Anomaly Incremental to Operating Cash flows Abstract The purpose of our paper is to determine the extent to which operating cash flows may subsume the relation between future returns and various estimates of abnormal accruals. We are motivated by recent research which provides evidence that, after controlling for the operating cash flows-to-price ratio, accruals no longer relate to future returns. If accruals relate to future returns in a univariate model but not in a multivariate model that includes operating cash flows, then it must be the portion of accruals related to operating cash flows that drives the relation between accruals and future returns. If we label the portion of accruals that relates to operating cash flows as normal, then it is normal accruals, rather than abnormal accruals, that lead to evidence of an accrual anomaly. This supports the conclusion that the accrual anomaly is part of the overall value-glamour anomaly. We examine this issue in detail by testing the relation between future returns and residuals from various accrual expectation models (i.e., abnormal accruals). We employ several accrual expectation models, including those that specifically consider operating cash flows in the estimation. Moreover, we apply different accrual measures, including total and working capital accruals, and various estimation procedures, including industry-specific and firm-specific regressions. We find that for most of our accrual expectation models, abnormal accruals relate to future returns after controlling for the operating cash flow-to-price ratio. We also document that the operating cash flows-to-price ratio is incremental to abnormal accruals in explaining future returns. I. INTRODUCTION This study investigates the extent to which abnormal accruals provide evidence consistent 1with mispriced securities. Several papers in the accounting literature document the relation between accruals and future returns, as initially identified by Sloan (1996). In addition, several papers in the finance literature demonstrate the relation between cash flows and future returns and refer to this as the value-glamour anomaly. Desai, Rajgopal, and Venkatachalam (2004) are the first to show that the explanatory power of accruals for future returns is subsumed by the operating cash flows-to-price ratio, suggesting that the accrual anomaly is also part of the overall 2value-glamour anomaly. We are interested in whether the abnormal component of accruals derived from various accrual expectation models may also be subsumed by operating cash flows in explaining future returns. Of particular relevance to our study is the accrual expectation model introduced by Dechow and Dichev (2002). They suggest that a substantial portion of current accruals normally relates to operating cash flows. Only the portion of current accruals which does not relate to past, current, and future operating cash flows should be considered abnormal. Therefore, since 1) normal accruals relate to operating cash flows and 2) operating cash flows relate to future 1 Prior research uses the terms “abnormal” or “discretionary” accruals to label the difference between reported and expected accruals. In this paper, we use the term “abnormal” since it is less suggestive as to whether unusual accruals arise from intentional versus unintentional actions of managers. Moreover, since there is no widely accepted accrual expectation model for identifying true normal or a