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

261 NTU Management Review Vol. 30 No. 2 Aug. 2020 which contain 18 categories, including approximately 250 indicators. The rating of each category is weighted equally by sub-indicators, and the overall environmental, social, and governance (ESG) performance is weighted equally by four dimensions. The KLD STATS database tracks the community, corporate governance, diversity, employee relations, environment, human rights, and product performance of companies, which include approximately 80 qualitative indicators. The database provides each dimension’s indicators, including both strengths and concerns, to provide comprehensive measures for overall CSR performance. As mentioned by Bendell (2010), who identifies nine key weaknesses of ESG analysis, different ESG rating providers have different understandings of the concept of CSR. Consistent with the findings of prior US studies, our study supports the robustness of the negative association between CSR performance and the cost of capital. Third, our results have implications for firms which are going to participate in corporate responsibility activities, as they indicate that the benefit of superior CSR performance outweighs the increased costs of CSR. Although shareholders may not agree with a high CSR cost (Barnea and Rubin, 2010), if a firm acts in a socially irresponsible manner, for example mismanaging stakeholders, such behavior can result in lost markets and revenues, and even a decrease in shareholders’ wealth (Downing, 1997; Frooman, 1997). 2. Related Research and Hypothesis Development 2.1 The Effect of CSR on Corporate Behavior So far, there are no theoretical models which directly link CSR performance to market risk (systematic risk), liquidity risk, or the cost of capital. Heinkel, Kraus, and Zechner (2001) analytically explore the effect of green investment on a firm’s cost of capital. In their model, there are three categories of firms: acceptable firms which satisfy the investment criteria of green investors; unacceptable firms which do not satisfy the investment criteria of green investors; and reformed firms which previously do not satisfy the investment criteria of green investors but are now accepted by green investors after having incurred the reform cost required to satisfy those criteria. According to their comparative statics, a reformed firm’s cost of capital decreases (i.e., its stock price increases) as the reform cost increases. In addition, under the assumption of a fixed number of investors, more green investors mean fewer neutral investors; since unacceptable firms have a lower price. Because the price of reformed firms equals the