CEO－Executive Connections and Informativeness of Discretionary Accruals

[Purpose] The purpose of this paper is to examine the impact of the network connections between the CEO and the top management team on discretionary accruals and the informativeness of accrual component of earnings. [Methodology] This paper uses regression analysis of 6,253 S&P 1500 firm－year observations over a thirteen－year period between 2000 and 2012. Four different models are applied to estimate discretionary accruals. (1) Jones model (1991), (2) Modified Jones model (Dechow, Sloan, and Sweeney 1995), (3) Jones model after controlling for performance (Kothari, Leone, and Wasley 2005), and (4) Modified Jones model after controlling for performance (Kothari et al. 2005). To test the informativeness of the accrual component of earnings, this paper adopts the model used in Subramanyam (1996). [Findings] This paper finds a positive association between CEO’s network connections with the top management team and discretionary accruals, and finds that these connections increase the informativeness of discretionary accruals in predicting future earnings. Moreover, this paper also documents that such association is more prominent in the sample with low incentives to manipulate earnings. This is suggestive of connections between the CEO and the top management team increasing the relevance of the reported earnings through discretionary accruals. [Implications] This paper complements previous studies that document negative effect of network connections between the CEO and the directors on financial reporting quality by documenting a positive effect of CEO’s pre－existing network connections with the top management team on the informativeness of accruals in predicting future earnings. The findings of this paper may be important to legislators, regulators, and investors who have generally been concerned about the negative effects of connections on corporate governance.