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

75 NTU Management Review Vol. 30 No. 2 Aug. 2020 on examining only two manifestations of earnings management—income smoothing and managing toward positive earnings. To our best knowledge, no prior studies have explicitly investigated other competing theories of managerial opportunistic behavior (e.g., taking a “big bath”). Accordingly, we also examine whether management preferences and incentives affect the trade-off between accrual-based earnings management and real activities manipulation for non-U.S. cross-listed firms electing to adopt U.S. GAAP/IFRS after the elimination of the reconciliation requirement. Our empirical results suggest that discretionary accruals and real activities manipulation are partial substitutes for earnings management, and that their magnitudes are determined simultaneously for non-U.S. firms switching either to U.S. GAAP or IFRS voluntarily after the SEC waived the reconciliation requirement (i.e., the two test samples in this study). We also provide evidence of a trade-off decision between these two earnings management strategies as a function of executives’ reporting incentives, including “big bath,” income smoothing, and debt covenant restrictions, although the existence and tightness of accounting-based debt covenants seem to play a minor role in this decision for firms in the IFRS test sample. We also conduct several additional analyses. First, to address the issue of small sample size, we apply the bootstrapping methods. The results are qualitatively similar and the inferences are unchanged. Second, we conduct a battery of sensitivity analyses to further validate the “income smoothing” proxy. The evidence suggests that the “income smoothing” variable is more likely picking up reporting incentive behaviors, as opposed to the underlying economics of the firm. Third, to control country effect and macroeconomic factors (i.e., GDP, foreign direct investment, or inflation rate), we repeat our analysis from two-stage least squares (2SLS) regression after including a country dummy and three macroeconomic regressors. The coefficients on most of the macroeconomic factors are insignificant. All other estimated coefficients remain virtually unchanged and all inferences remain unaffected. Fourth, for robustness check, we repeat the analysis after removing firms that domicile in countries that adopt IFRS mandatorily or have already made announcement of its adoption roadmap of the standards during the sample period. The results reveal essentially similar inferences between the samples of IFRS adopters in the primary analysis and those excluding the two types of firms, respectively. Fifth, we repeat the analyses after removing the sample firms from countries (i.e., Bermuda, Channel Islands, Marshall Islands, and British Virgin Islands) the Organization for Economic Co-operation and Development (OECD) has designated as tax havens and firms