本研究建立在無形資產與資訊不對稱呈現正相關的假說上,主要為探討股票購回與無形資產之間的關聯性,研究對象為2003年至2007年台灣上市公司。本研究以研發費用率、員工每人營業利益與廣告費用率作為無形資產代理變數,並且加入一般資訊不對稱變數、閒置資金變數、淨值市價比、債益比與股利發放率等控制變數,以Logistic迴歸模型驗證股票購回機率與無形資產的關聯性,研究結果發現研發費用率與員工每人營業利益越高之公司,越有可能購回股票,與預期相同,但是廣告費用率與購回股票機率則呈現負相關,與預期不同。 本研究以事件研究法估算購回股票宣告之異常報酬,研究結果發現宣告前12天之異常報酬皆為負值,表示宣告前股價有受到低估的情形;而宣告後11天之異常報酬皆為正值,表示股票購回宣告有拉抬股價、抑制下跌的效果。此外,本研究將樣本依產業別分為電子產業、傳統產業與其他產業三類,事件窗口(-5, -1)標準化累計異常報酬以電子產業最低,表示宣告購回股票前5天,電子產業股價受低估的程度最為嚴重;事件窗口(+1, +5)標準化累計異常報酬則以其他產業最高,表示宣告購回股票後5天,其他產業股價上升的幅度最大。
The main prediction of this research is that intangible assets and information asymmetry are positive relation. For this reason, the primary objective of this study is to investigate the relation between share repurchase and intangible assets. We select the rate of research and development expense, the rate of advertisement expense and the operating profit per employee for intangible assets proxy and also include idle cash proxy and general information asymmetry proxy as control proxy. We build logistic regression model to study the relation between repurchase likehood and intangible assets, and we find that firms with more research and development expense and higher operating profit per employee are more likely to repurchase share, as predicted. Contrary to prediction, the relation of repurchase likehood and advertisement expense is negative. We use event study to estimate abnormal return for announce repurchase share and find abnormal return are negative for 12 days before announcement and positive for 11 days after announcement. The result means repurchase announce has power to stop falling and raise stock price. Besides we classify sample to electric industry, traditional industry and others, and find that electric industry are undervalued most before announcement and other industry’s stock price raise most after announcement.