Management of reported earnings and the quality of financial statements have been of interest to academics and regulators for several decades. Ball and Brown (1968) and Beaver (1968) were two of the first to make a connection between accounting information and market value of firms. If accounting revelations have an economic impact, then there is a potential for agency conflict and an incentive to manipulate and falsely report the accounting information of the firm.
Earnings management (EM) has been defined as financial reporting which has been affected by managers’ deliberate attempt to mislead shareholders, influence contractual outcomes, or for some other private gain (Healey and Wahlen 1999; Schipper 1989). This definition restricts EM to the behavior that is willful, misleading, and aims to achieve a gain for managers at the expense of other stakeholders. This definition, thus, leaves out the acceptable practice of income smoothing which Zarowin (2002) and Tucker and Zarowin (2006) demonstrate to increase the information content of accounting information.
Hayn (1995) and Burgstahler and Dichev (1997) report a unique pattern in reported earnings. They find an unusually low concentration of companies that reported slightly negative earnings and an equally unusually high concentration of companies that reported earnings just above zero. Assuming a normal distribution, such frequencies deviate significantly from expected frequencies. They conclude that their findings are consistent with the hypothesis that managers whose earnings are expected to fall just below zero or decline, engage in activities to manipulate their earnings and push them across the line by just a small margin. Between 30-44% of firms whose earnings in the absence of manipulation would be slightly negative use discretion to report positive earnings (Burgstahler and Dichev 1997).
Download
Patterns in Accounting Data and Detection of Earnings Management
