This paper examines the monitoring role of institutional investors on managers’ real earnings management for Korean firms. More specifically, we examine the association between the institutional investors’ ownership and the proxies for the magnitude of real earnings management,such as abnormal cash flow from operation, abnormal production cost and abnormal discretionary expenses. Real earnings management (hereafter REM) is defined as “management actions that deviate from normal business practices, undertaken with the primary objective of meeting certain earnings thresholds” (Roychowdhury 2006). Prior research on REM suggests that managers actually use the method of REM to manipulate earnings (e.g., Roychowdhury 2006;Kim, Goh and Koh 2008) and that REM has negative effects on firms’ long-term profitability or cost of equity capital (e.g., Ewert and Wagenhofer 2005; Kim and Sohn 2009). In particular,Cohen, Dey and Lys (2008) suggest that managers have began to substitute earnings management based on discretionary accruals by REM since accounting regulations were reinforced. Since traditional mechanism such as external audit or accounting regulation that prevents managers’earnings management based on discretionary accruals cannot restrain managers’ REM, it will be important to explore an alternative mechanism to constrain managers’ REM. In this study,we focus on the monitoring role of institutional investors as an alternative mechanism to constrain managers’ REM. While previous finance and accounting research has explored the monitoring role of institutional investors, which leads to mixed results regarding to its effectiveness, no prior study examines the monitoring role of institutional investors on managers’ REM. Our study fills this void in the literature. By using 2,292 firm/year observations between 1999 and 2003 in Korean stock market, we find that institutional investors’ ownership is negatively associated with the proxy for the magnitude of REM. Furthermore, we find that this association is maintained for both of income-increasing and income–decreasing REM. Thus, we conclude that institutional investors in Korean stock market may be able to effectively constrain managers’ REM through their monitoring role. This paper is the first study to document the negative association between institutional investors’ ownership and the magnitude of REM. Given that in the future managers may prefer REM to earnings management based on discretionary accruals to manipulate earnings and that external audit and/or accounting regulation cannot constrain managers’ REM, our empirical evidence will be of interest to accounting regulators as well as academia.