비상장기업의 적자회피 및 이익감소회피를 위한 이익조정 활동

This paper investigates whether private firms in Korea use accrual-based earnings management or real earnings management to avoid losses and earnings decreases. Managers have incentives to manipulate accruals as well as real activities during the year to meet certain earnings targets. Unlike accrual management, real activities manipulation affects cash flows and in some cases, accruals. Specially, real activities manipulation can reduce firm value because actions taken in the current period to increase earnings can have a negative effect on cash flows in future periods (Roychowdhury 2006; Cohen and Zarowin 2010 etc). Burgstahler and Dichev (1997) classify firms that show small positive earnings (small earnings increases) on the basis whether they avoid losses (earnings decreases) by a narrow margin because such small beaters are suspected of managerial intervention by accrual management or real earnings management. After Burgstahler and Dichev (1997), most studies observe firms that meet or slightly beat zero earnings threshold to indirectly examine whether managers manage their earnings through accrual-based or real earnings management. However, much of the current research on earnings management focuses on “detecting” abnormal accruals or abnormal real activities. Choi and Kwak (2010) report that suspected non-listed firms that slightly achieve earnings benchmarks tend to manage earnings using discretionary accruals as well as real activities devices, such as abnormal cash flow from operations, abnormal production costs, and abnormal selling and administrative expenses. They provide some limited evidence on earnings management behavior of non-listed firms. However, it is difficult to conclude that non-listed firms with small positive earnings or small positive earnings increases always use both accrual-based and real manage earnings to achieve earnings benchmark. Therefore, our study examines whether managers’ behavior to use accrual-based earnings management or real earnings management affect the probability to include the small positive earnings and small earnings increases interval. Following Burgstahler and Dichev (1997), we find unusually high frequencies of small positive earnings (small earnings increases) in cross-sectional distribution of earnings level (earnings changes) and define as suspected firms whose managers might manage their earning. Burgstahler and Dichev (1997) document that firms manage their reported earnings mostly to avoid losses or earnings decreases, thus increasing the incidence of producing the small positive earnings and the small earnings increases. In order to measure accrual-based earnings management, we estimate discretionary accruals by using the performance matched model by Kothari et al. (2005). Similar to Roychowdhury (2006), we estimate abnormal cash flow from operations, abnormal production costs, and abnormal selling and administrative expenses to measure real earnings management. Our observations are based on 40,362 firm-year data, selected from non-listed firms in Korea, for six years (2003 to 2008). The empirical findings of this paper are following. First, we find that the suspected firms whose managers might manage their earnings to avoid losses prefer to use real earnings management rather than to use accrual-based earnings management. Our results are same even after we control a certain variable that might affect earnings management. We also report the suspected firms that manage earnings to avoid losses use real earnings managements such as abnormal cash flow from operations, abnormal production costs, and abnormal selling and administrative expenses. Second, we find that the suspected firms whose managers might manage their earnings to avoid earnings decreases prefer to use real earnings management, usually abnormal production costs, and abnormal selling and administrative expenses, rather than to use accrual-based earnings management. We find the consistent results after we redefine the suspected interval by a narrow or wide section. Finally, the results of our study show that non-listed firms in Korea tend to avoid losses and earnings decreases by using usually real earnings management rather than using accrual-based earnings management even though the real earnings management might cause a negative effect on their business performance in future periods. Thus, our results are different from previous studies such as Choi and Kwak (2010), showing that non-listed firms use both real earnings management and accrual-based earnings management to avoid losses and earnings decreases. Our paper contributes to a literature on earnings management for non-listed firms, the majority of which concentrate mainly on real earnings management to meet earnings goals (e.g. to avoid a loss or to avoid an earnings decrease). Our findings that managers for non-listed firms in Korea prefer to use real earnings management to avoid losses and earnings decreases, imply that managers focus on their short-term achievement rather than on activities to increase firm value in a long-term perspective. Therefore, our findings provide an important implications to not only investors and creditors but also regulators who are interested in managers’ behavior in non-listed firms. Researchers who are interested in this area can also apply the discussion in this paper for the related studies.