機構投資人交易對報酬率動態行為的影響

Abstract

本論文以台灣股票市場交易資料為樣本，探討機構投資人交易對股價報酬率動態行為的影響。論文分為兩部份；第一個部份探討機構投資人交易對報酬率序列相關性的影響，第二個部份則探討機構投資人交易對報酬率波動性的影響。 Llorente, Michaely, Saar, and Wang (2002，以下簡稱LMSW) 使用橫斷面資料檢驗他們的理論預測：訊息交易及非訊息交易分別會造成正向及負向的報酬率序列相關。然而，由於橫斷面變數彼此間往往具有很強的關聯性，如何解讀實證結果不無爭議。有鑒於此，本論文的第一個部份從時間序列的角度出發，重新檢驗LMSW的理論預測。根據過去研究，台灣股市的機構投資人當中，投信及外資的交易較可能是訊息交易，自營商的訊息交易證據並不充分；我們發現，投信及外資的交易造成了正向的報酬率序列相關，而自營商的交易則是造成負向的報酬率序列相關。在台灣，機構投資人的借券交易受到很強的限制；我們則發現，相對於賣出行為，投信的買入行為造成了較強的正向報酬率序列相關。根據以上發現所建構的投資組合，在樣本期間獲得了顯著的正報酬。 當借券交易的成本過高時，私有訊息交易者不易透過賣出交易來遂行訊息交易，這可能會導致買進交易與賣出交易有不同的訊息涵量，進而導致它們對報酬率波動性的影響有所差異。本論文的第二個部份提出一些假說來說明，為何不同訊息涵量的交易會對報酬率波動性有不同的影響。根據過去研究，台灣股市的機構投資人很可能是私有訊息交易者，同時，法規並不允許機構投資人進行借券交易；據此，我們使用台灣股市的機構投資人買賣資料來檢驗這些假說。我們得到了與假說一致的實證發現：「預期中的機構買賣」會降低報酬率波動性，「非預期中的機構買賣」則會提高報酬率波動性；且無論是預期交易或非預期交易，買進交易對於報酬率波動性的影響程度都弱於賣出交易。

This doctoral dissertation comprises two essays regarding the effect of institutional trading on return dynamics in the Taiwan stock markets. Essay I focuses on the effect of institutional trading on return autocorrelation while Essay II focuses on the effect of institutional trading on return volatility. Essay I proposes new tests for the prediction of Llorente, Michaely, Saar, and Wang (2002) that information trading drives positive autocorrelation. Data from the Taiwan Stock Exchange is used to exploit the differences in the trading motivations of three groups of institutional investors. Consistent with the predictions, we find that heavy trading by foreigners and mutual funds will increase the autocorrelation particularly for large firms, and that heavy trading by dealers will not. We also find that the sell volume of mutual funds – short sales are disallowed by regulation – has significantly smaller effect on the autocorrelation of returns than buy volume. A portfolio strategy that exploits the observed autocorrelation pattern can generate a significantly positive daily return. When short selling is costly, sales tend to convey less information than buys. In Essay II, we propose hypotheses on how different information content changes the volatility-volume relationship. To test these hypotheses, we use a sample of institutional trading in the Taiwan stock market because these institutions cannot sell short owing to the regulations. Consistent with our hypotheses, the empirical findings show that expected institutional purchases have a less negative effect on volatility than expected institutional sales, and unexpected institutional purchases have a less positive effect on volatility than unexpected institutional sales.