Accrual-Based Accounting Case Study
To capture accrual-based earnings management we follow prior studies that use the cross-sectional Jones (1991) model
(e.g., DeFond and Jiambalvo, 1994; Subramanyam, 1996). To capture real earnings management, we follow Roychowdhury
(2006) and estimate abnormal levels of cash ﬂows from operations, discretionary expenses (advertising, R&D, and SG&A), and production costs. In addition, we combine these three measures into two comprehensive aggregate metrics of real earnings management.
Consistent with prior studies (e.g., Teoh et al., 1998, Rangan, 1998, among others) we ﬁnd that ﬁrms use accrual-based earnings management around SEOs. In addition, also consistent with these studies, we ﬁnd that SEO ﬁrms tend to both outperform their industry peers in the period preceding the SEO and underperform their peers following the SEO, as evidenced by their returns on assets …show more content…
Cohen, P. Zarowin / Journal of Accounting and Economics 50 (2010) 2–19
for SEO ﬁrms. Section 6 provides evidence on accrual and real earnings management activities around the time of the SEO and the post-SEO operating performance associated with both earnings management methods. Section 7 concludes.
2. Related literature and hypotheses development
Our paper unites two streams of research that have previously been disparate: one on real earnings management and the other on earnings management around SEOs.3 We ﬁrst discuss related research, and then we build on the existing body of evidence to state our hypotheses.
2.1. Accrual-based earnings management around SEOs
Beginning with Rangan (1998) and Teoh et al. (1998), researchers have been concerned with whether ﬁrms manage reported earnings during SEOs, and the accounting and stock market consequences associated with such activities. This literature has been motivated by the empirical ﬁndings that SEOs are followed by both poor stock returns and poor earnings performance, leading researchers to suspect that earnings have been managed upward in anticipation of the SEO, and subsequently reverse (e.g., Teoh et al., 1998; Rangan, 1998; Shivakumar,