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

The Effects of Relaxing the Reconciliation Requirement in Foreign Private Issuers’ SEC Filings on Earnings Management Strategies: IFRS Adopters versus U.S. GAAPAdopters 102 6.2 Correcting for Self-Selection Bias We estimate the probit regression on two samples, of which the first (second) comprises 78 (83) U.S. GAAP (IFRS) observations and 78 (83) comparison firm observations that applied their domestic GAAP. 18 Table 4 reports estimation results for the probit model. We find that the coefficient on LEV it ( ROA it and CS it ) is (are) significantly positive (negative) at better than p = 10% for the U.S. GAAP/non-U.S. domestic standards comparison, consistent with the notion that firms with poor financial performance, higher leverage, and a lower propensity to access equity markets are more likely to adopt U.S. GAAP. We also document that the coefficients on ROA it and LEV it are significantly positive at better than p = 1% for IFRS adopters relative to domestic GAAP filers, suggesting that firms with better financial performance and higher leverage have a greater propensity to adopt IFRS. Table 4 Estimation of the Binomial Probit Model of the Adoption Choice SELECT it = α 0 + α 1 ROA it + α 2 LEV it + α 3 MV it + α 4 NMKT it + α 5 CS it + α 7 LT_DEBT it + ε it (5) U.S. GAAP vs. Domestic Standards IFRS vs. Domestic Standards Variable a Parameter Estimate χ 2 Value Parameter Estimate χ 2 Value Intercept 0.158 0.037 1.144 2.312 ROA it -0.273 3.211* 6.427 10.458*** LEV it 2.939 11.451*** 3.588 11.977*** MV it -0.019 0.043 -0.091 0.819 NMKT it -0.106 0.184 0.150 0.348 CS it -1.152 5.902** 0.040 0.011 LT_DEBT it 0.288 0.520 -0.374 0.938 No. of firm-years 156 166 Pseudo- R 2 0.209 0.206 Note: ***,**,* denote significance at the 0.01, 0.05, and 0.10 level, respectively. Note a : Variable definitions: SELECT it = an indicator variable equal to 1 for the sample firms, and 0 otherwise. All other variables are defined in Table 2. 18 Our test sample comprises companies that did not domicile in countries that adopt U.S. GAAP/IFRS but changed their filing choice to either of the standards voluntarily after the elimination of the reconciliation requirement. We match each sample firm, based on industry, size, and year, with a comparison firm from the same country that reported using domestic GAAP. Once a local GAAP firm is selected as a match, it is not considered a potential match for other sample U.S. GAAP/IFRS firms. However, a breakdown of sample firms by country shows that Bermuda, Canada, and Israel are the only countries with both U.S. GAAP and IFRS adopters (see Panel A of Table 1). Accordingly, we use another matching procedure permitting matching a comparison firm to more than one test sample firm and find that four domestic GAAP adopters (three Canadian firms and one Bermudan firm) can be matched to more than one test sample firm (i.e., one U.S. GAAP and one IFRS firms), respectively. Untabulated findings reveal that using this alternative procedure has no effect on our results and inferences.