臺大管理論叢 NTU Management Review VOL.29 NO.1

231 NTU Management Review Vol. 29 No. 1 Apr. 2019 5.2 Alternative Earnings Informativeness Model Testing To test the first hypothesis, we follow Fan and Wong (2002), Yeo, Tan, Ho, and Chen (2002) and Francis, Schipper, and Vincent (2005) in measuring the informativeness of current earnings by examining a regression of cumulative abnormal stock returns (CAR) on net income. Naturally, we also use the unbalanced-panel fixed effect with year dummy model to run the regressions. 15 Naturally, we also use the unbalanced-panel fixed effect with year dummies model to run the regressions. The models are presented as follows: CAR t = β 0 + β 1 X t + β 2 IM t + β 3 IM t *X t + β 4 LEV t + β 5 MB t + β 6 SIZE t + ε t (4) CAR t = β 0 + β 1 X t + β 2 IM t + β 3 IM t *X t + β 4 REV t + β 5 REV t *IM t *X t + β 6 LEV t + β 7 MB t + β 8 SIZE t + ε t (5) CAR t is the firm’s market-adjusted annual stock returns for the 12-month period ending four months following the end of the fiscal year. 16 The definitions of the remaining variables are the same as Reg. (1) and Reg. (2). The results are presented in Table 7. From Table 7, we find that the coefficients of IM t *X t are -9.282 ( t = -2.77) and -10.258 (t = -3.14), which are both statistically significant at the 1% level in the IM and REV models. The coefficient of REV t *IM t *X t is 18.903 ( t = 1.58) in the REV model and is statistically insignificant. The combined coefficient of IM t *X t and REV t *IM t *X t ( β 3 + β 5 ) is 8.645 ( t = 0.65), which is positive and statistically insignificant. These results again show the informativeness of current earnings decreases when firms recognize the impairment of long-lived assets. Thus, the initial empirical result for the informativeness of current earnings in the present study are not qualitatively different from the empirical findings of informativeness model which is suggested by Fan and Wong (2002), Yeo et al. (2002) and Francis et al. (2005). 15 Note that most of assets write-off information was included in a firm’s financial reports and was announced together with the filed earnings numbers to the Market Observation Post System (MOPS). This feature limits this study to identify a clearly separating write-off date and uses the short-term event study to examine the information content of assets impairment announcement. 16 Taiwan Securities Exchange Law §36 requires listed firms to issue an annual financial report during four months after the end of the calendar year before 2012. As a consequence, the abnormal stock return is measured as May 1 of the current calendar year to April 30 of the next calendar year.