اثرات واقعی مدیریت سود واقعی : شواهدی از نوآوری The Real Effects of Real Earnings Management: Evidence from Innovation

1. Introduction The reliance of external stakeholders on reported accounting information creates incentives for firms to report earnings that meet or exceed targets and forecasts.1 According to Graham et al. (2005), CFOs believe that reported earnings, rather than cash flows, are the primary metric used by external stakeholders to assess the value of the firm. These beliefs may lead to the management of reported results in order to enhance the firm’s reputation with external parties. The tools to do so involve engaging in potentially value-reducing actions to boost shortterm reported performance. The presence of such behavior is corroborated by a wealth of evidence indicating that firms take actions such as selling assets and reducing discretionary expenditures in order to manage reported earnings (Dechow and Sloan 1991; Bartov 1993; Bushee 1998; Roychowdhury 2006; Mizik and Jacobson 2007; Gunny 2010). In Graham et al. (2005), the most frequently cited mechanism to achieve reported earnings benchmarks is the reduction of discretionary spending. An alternative to managing reported earnings via changes in real activities is to do so by altering accounting accruals. There is well-documented evidence of substitution between these two types of earnings management strategies (e.g., Cohen et al. 2008; Cohen and Zarowin 2010; Badertscher 2011; Zang 2012). Accrual-based earnings management is short-lived and has subsequent reversals. In contrast, changes in real activities to manage short-term reported performance by definition involve suboptimal managerial decisions with potentially adverse consequences. In the surveys of Bruns and Merchant (1990) and Graham et al. (2005), managers indicate a greater willingness to use real activities rather than accruals to manipulate reported earnings. Roychowdhury (2006) argues that this preference for real activities manipulation, despite the potentially greater long-term costs to the company, may arise because managers expect to bear greater private costs if they engage in accruals manipulation. Real changes made to manage earnings upwards involve such actions as increasing inventory production to reduce reported cost of goods sold, timing asset sales opportunistically, and reducing discretionary spending on items such as R&D, advertising, and maintenance.2 Real earnings management (hereafter REM) is considered to be more difficult to control than accrualbased earnings management (Dichev et al. 2013), as it is an operational decision rather than an accrual estimate that is subject to auditor scrutiny. This notion is demonstrated analytically in the Ewert and Wagenhofer (2005) model in which tighter accounting standards make accruals management more difficult, resulting in greater REM. Consistent with this framework, Cohen et al. (2008) report an increase in the use of REM following the Sarbanes-Oxley Act, as this regulation curbed accruals management.