Earnings Management and Institutional Ownership in Turkey

2. Literature Review

Earlier researchers focusing on the link between institutional ownership and earnings management mostly employed variables based on research and development expenditures (R&D) as proxies of managerial discretion (Graves, 1988; Hill & Hansen, 1989; Hansen & Hill, 1991; Baysinger et al. 1991; Eng, 1995; Bushee, 1997; Majumdar & Nagarajan, 1997; Bushee, 1998). It is presumed that increases or decreases in R&D expenses are utilized as a tool in managing earnings. However, recently new variables besides R&D related proxies are introduced.

Rajgopal and Venkatachalam (1997) using institutional investor ownership percentage and the number of institutional owners as explanatory variables together with a set of control variables found a negative relationship between institutional ownership and earnings management. Using a similar model, Rajgopal, et al. (1999) validated the significant and negative impact of institutional ownership on the selected proxy of managerial discretion. Some other researchers also confirmed the negative relationship between institutional investors and the intensity of earnings management. Chung et al. (2002) revealed the substantial role of institutional owners in discouraging managers to engage in earnings management while incentives to increase or decrease company profits were existed. Koh (2003) claimed that the relationship between earnings management and institutional ownership was non-linear. Liu (2005) showed that managers manipulated earnings upwards as the level of transient institutional ownership increased. With a concurrent study, Mitra and Cready (2005) verified the monitoring role of institutional owners in constraining accounting flexibility using a dataset of 824 firms listed on NYSE (New York Stock Exchange). A similar study on S&P 100 Index also verified that institutional ownership reduced earnings management by means of discretionary accruals (Cornett et al. 2008).