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

Earnings Informativeness of Long-Lived Assets Impairment Recognized and Reversals 228 firm without recognized impairment whose predicted probability of being a recognized impairment firm is closest to that of the firms with impairment. This process enables us to select a set of benchmark firms that are characteristically similar to the impairment firms but without such impairments. We obtain a final sample consisting of 933 recognized impairment firms matched with an equal number without recognized impairment. The results from the matched sample testing are presented in Table 6. From the IM model in Table 6, the coefficients of IM t *X t and IM t *X t3 are respectively -9.518 ( t = -1.68) and 3.708 ( t = 1.81), both statistically significant at the 10% level. The coefficients of IM t *X t and IM t *X t3 are respectively -9.803 ( t = -1.81) and 4.205 ( t = 2.97) in the REV model, both statistically significant. The results indicate the informativeness of current earnings (future earnings) decrease (increase) when firms recognize asset impairment. We find that the coefficient of REV t *IM t *X t3 is -18.613 ( t = -1.46) and is negative and statistically insignificant. The combined coefficient of IM t *X t3 and REV t *IM t *X t3 ( β 8 + β 13 ) is -14.407 ( t = -1.07) and is negative and statistically insignificant. These diagnostic checks again demonstrate that the results do not qualitatively change the empirical findings in the propensity-score matching specification.