研發密集公司超額報酬為風險溢酬之額外證據

Abstract

范宏書、林彥廷（民99）以實證方式測試研發密集公司之超額報酬究竟為風險溢酬或是市場錯誤評價之結果，其證據直接支持風險溢酬假說，意謂我國證券市場對於研發支出之評價具有效率性，而Fama（1991）主張市場效率性需要眾多證據支持，因此本研究旨在提出新的實證證據佐證范宏書、林彥廷（民99）之結論。Lev and Sougiannis（1999）主張若風險溢酬假說成立，預期較高風險的研發支出對於後續期間股票報酬率應有較高之解釋力；然而，企業所投入的研發支出哪些為高風險支出無法從現行財報得知，因此本研究引入企業生命週期、公司治理、產業別、盈餘變異程度等企業風險相關因子，來劃分研發支出的風險程度，並測試不同風險程度之研發支出與風險溢酬的關聯程度。本研究發現高風險樣本之研發資產變數對於後續期間報酬率之解釋力顯著高於低風險樣本，支持風險溢酬假說。有鑑於過去國內相關文獻之闕如，本研究之潛在貢獻在於豐富我國股票市場對研發支出反應效率性的證據。

Fan and Lin (2010) find that the excess returns of R&D intensive firms are more likely to result from the failure to control adequately for risk. That is, Taiwan stock market is efficient in reflecting R&D information on stock prices. However, we need more evidence to validate whether the stock market in Taiwan is efficient (Fama, 1991). This study aims to provide additional evidence to support the finding of Fan and Lin (2010). Lev and Sougiannis (1999) argue that if the excess returns of R&D intensive stocks are risk premium, then the R&D expenditures with higher risk should have higher association with subsequent stock returns. Inspired by Lev and Sougiannis (1999) in research design, but we differentiate our research to divide R&D intensive firms into two subgroups: high-risk firms and low-risk firms, by several different risk measures, such as corporate life-cycle, corporate governance, electronics industry, and standard deviation of return on assets. Then, we test the risk premium hypothesis that if the R&D related excess returns are risk premium of R&D assets, the excess returns of R&D assets will be significantly higher for high-risk firms than for low-risk firms. The empirical results of this study provide additional supportive evidence for the risk premium hypothesis.