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

The Benefits of Disclosing Internal Control Weaknesses: Evidence from Taiwanese Banks 178 first stage, the dependent variable takes a value of 0 if the bank discloses that it has no ICW, or it takes a value of 1 if the bank discloses that it has one or more ICWs. Following previous studies (e.g., Doyle, Ge, and McVay, 2007), ICW is explained by several determinants, including size, age, proxies for financial health, variables capturing the complexity of the operating environment, proxies reflecting the soundness of corporate governance, and a dummy indicating whether the bank’s shares are owned by the government. Second, the application of the Ohlson model is used to examine the value effect of ICW disclosures. Specifically, the market value of equity is explained by the dummy variable equaling 1 if firms disclose non-zero ICWs (denoted as ICW ), along with the book value, earnings, size, degree of diversification, Z-score, and the inverse Mills ratio from the first stage. This study finds that the coefficient of ICW is 0.012, and is statistically significant ( p = 0.018). Thus, the results indicate that reporting non-zero ICWs is related to positive value effects, consistent with the hypothesis that reporting ICWs reveals the effort of firms in assessing their internal control effectiveness. Several additional analyses are conducted. First, the study discerns whether the positive value effect of ICW disclosures is influenced when the quality of internal auditing is considered. Specifically, three measures are used as proxies for the quality of internal auditing: whether the internal audit unit directly reports to the board of directors, the log of the number of employees working in the internal audit unit, and the number of meetings in which the internal audit unit holds discussions with the independent directors (or supervisors). The results show that ICW remains positively significant when the internal audit quality is controlled. Second, the effects of mergers and acquisitions are considered. Third, the materiality or severity of ICWs is also considered. Fourth, this study follows Cheng, Dhaliwal, and Zhang (2013) in applying propensity score matching as an alternative way to address the endogeneity problem. Lastly, this study also follows Larcker, Richardson, and Tuna (2007) and uses the positive/negative residuals method to mitigate the endogeneity issue. The adjustments above do not affect the main finding that banks disclosing non-zero ICWs are rewarded by the market. The contributions of this study are described as follows. First, this study is likely the first to investigate the value effect of ICW disclosures by using data from banks in Taiwan. Although prior results regarding the effects of ICW disclosures on the cost of