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

181 NTU Management Review Vol. 30 No. 2 Aug. 2020 adequacy and training of personnel). These studies therefore argue that there are at least two reasons to expect a higher cost of equity for ICW-disclosing firms. First, ICWs can result in poor accounting quality, thereby increasing information risk, which has been theoretically and empirically linked to a higher cost of equity. Second, ICWs may be a symptom of poor management controls in general, which can increase business risk and the cost of equity to the extent that the risk is systematic. Empirically, Ogneva et al. (2007) shows that the higher cost of equity associated with ICWs disappears when primitive firm characteristics and analyst forecast bias are controlled. Ashbaugh-Skaife et al. (2009) find that, when other risk factors are controlled, firms with ICWs exhibit significantly higher idiosyncratic risk, systematic risk, and cost of equity. Dhaliwal et al. (2011) test the relationship between the changes in a firm’s cost of debt and the disclosure of a material weakness in an initial SOX 404 report. The results indicate that, on average, a firm’s credit spread on its publicly traded debt marginally increases when it discloses a material weakness. Similarly, Kim, Song, and Zhang (2011) show that the loan spread is approximately 28 basis points higher for ICW firms than for non-ICW firms. In summary, prior studies provide well-documented evidence regarding the adverse effect of disclosing ICWs on the cost of capital. Some other studies examine the association between reporting ICWs and financial reporting quality. Feng, Li, and McVay (2009) argue that the quality of internal controls affects the accuracy of internal reports used by managers to form earnings guidance. Hence, they find less accurate guidance among firms reporting ineffective internal controls. Cheng, Yu, and Wang (2012) investigate the relationship between reporting ICWs and discretionary accruals. Their results reveal that, when sample selection bias is controlled, firms that report ICWs have greater absolute discretionary accruals (i.e., lower earnings quality). Tseng, Wang, Wang, and Shiue (2015) note that reporting ICWs is positively related to both real earnings management and discretionary accruals. Chiang, Shiue, and Lo (2015) use the number of sanctions as a proxy for deficiencies in banks’ internal control systems and show that banks with more sanctions are associated with poorer earnings quality, as measured by discretionary loan loss provisions. Relatively few studies evaluate whether disclosing ICWs is related to firm valuation. Beneish et al. (2008) propose that investors who already recognize the information risk is higher within ICW-disclosing firms, and the disclosures are unlikely to provide them with incremental information. They find that the negative abnormal returns around the announcement of ICWs is significant for non-accelerated filers but not for accelerated