臺大管理論叢 NTU Management Review VOL.30 NO.2

81 NTU Management Review Vol. 30 No. 2 Aug. 2020 Zarowin, 2010; Zang, 2012). Consistent with prior U.S.-based research, we therefore predict that managers use accrual-based earnings management and real activities manipulation jointly to achieve desired earnings targets for firms that switched to U.S. GAAP/IFRS voluntarily after the relaxation of the reconciliation requirement (i.e., firms whose filing choice is most likely attributable to the regulatory change). Given the lack of clear evidence in the extant literature regarding the earnings management potential of permitting IFRS reporting in the United States, we do not specify a sign of the relation. That is, firms may use these two earnings management strategies as complements or as substitutes to achieve the desired earnings targets. This leads to the following hypothesis: H 1 : The magnitude of accrual-based earnings management is associated with the magnitude of real activities manipulation for firms voluntarily switching to U.S. GAAP/IFRS from domestic standards after the SEC’s removal of the reconciliation requirement. H 1 predicts that the magnitudes of discretionary accruals and real activities management are determined jointly. However, a joint decision can imply either a simultaneous or a sequential decision. On one hand, managers can determine the magnitudes of discretionary accruals and real activities manipulation simultaneously during the fiscal year, and the extent to which managers use the two earnings management strategies is likely endogenous (Barton, 2001). On the other hand, because real activities manipulation has to take place during the fiscal year but accrual management can occur between fiscal year-end and the earnings announcement date, managers can adjust the magnitude of the latter based on the realized outcomes of the former (Zang, 2012). That is, unexpectedly high (low) level of real activities manipulation can be offset by a lower (higher) level of accrual management. Accordingly, we also assess whether our results are sensitive to the simultaneous versus sequential assumption. Extant earnings management literature generally considers three executives’ reporting incentives: big bath, income smoothing, and debt covenants. Although a few studies explore the relation between accrual-based earnings management and real activities manipulation (e.g., Barton, 2001; Pincus and Rajgopal, 2002; Zang, 2012), the evidence on this issue should be contingent upon managerial incentives, firm characteristics, and regulatory and litigation environment (see SEC, 2010). However, relevant studies fail to take into account the “big bath” incentive.