透過您的圖書館登入
IP:54.205.116.187
  • 學位論文

台灣共同基金績效-流量與偏袒策略之研究

A Study of Fund Performance-Flow Relationship and Favoritism Strategy of Taiwan Mutual Funds

指導教授 : 古永嘉
共同指導教授 : 王祝三(C. Edward Wang)
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


共同基金已成為台灣投資人重要的理財工具之一,過去許多財務文獻著重於研究共同基金經理人的行為,例如投資組合持股的從眾與窗飾等。2003年下半年以來美國陸續發生的基金醜聞引起國內專家學者的關注。然而,過去卻顯少有文獻探討台灣投信公司之公司層級的行為,因此,本研究試圖探討台灣投信公司是否因為公司與投資人的利益衝突而採取某些公司策略進而造成投資人的利益受損,同時並研究投信公司採取偏袒策略的動機。本研究可分為兩個部分:第一部分探討基金流量與基金績效之關係,以瞭解基金投資人的行為;第二部分則探討基金公司之行為,是否有採取某些公司策略以獲取公司本身的最大利益卻犧牲投資人利益的情形。本研究之主要目的有五:(一)研究共同基金市場之基金流量與基金績效之關係;(二)探討投信公司是否採取風險分擔策略以平滑旗下基金之績效表現;(三)檢驗投信公司是否存在偏袒策略,亦即偏袒對於公司有較高貢獻度的基金(高價值基金)而犧牲貢獻度低的基金(低價值基金);(四)假若投信公司存在風險分擔策略,則公司執行該策略時是否與當時的同類型基金的績效表現有關;(五)假若投信公司有偏袒策略存在,則公司執行該策略時是否與當時的同類型基金的績效表現有關。 本研究以2001年1月至2010年6月之台灣開放式國內股票型共同基金為研究對象,分為一般股票型、中小型、高科技型、價值型、特殊型、中概股型與上櫃股票型共同基金等七大類。樣本為172支基金與38家投信公司的基金月資料,包括基金月報酬率、費用率、基金淨資產、基金年齡、投信公司之國內股票型基金的總資產規模、台灣加權股價指數之月報酬率與投信公司年齡等。以橫斷面迴歸模型與Panel迴歸模型研究基金流量與基金績效之間的關係;並以複迴歸模型(Multivariate Regression Model)研究投信公司是否存在偏袒策略。 本研究結果發現:(一)對於基金流量與基金績效之關係的實證部分,發現(1)前一個月的基金績效排名,不論是與同類型基金比較或是與投信公司內旗下基金比較,皆顯著地正向影響基金流量。基金流量與基金績效之關係呈現類似U型,基金的高績效正向影響基金流量,而基金的低績效負向影響基金流量。此與美國的實證結果顯示投資人追逐高績效基金,而低績效基金的基金流量卻無顯著減少的情況略不相同。(2)前一個月排名前三名的基金之績效亦顯著地正向影響基金流量。(3)假若一家投信公司有至少一支明星基金,則該公司旗下基金的基金流量成長率將會減少(排擠效果)。(4)當期的大盤指數報酬率與基金流量成長率則呈現負相關。(二)對於投信公司策略的實證部分,發現(1)投信公司並未採取風險分擔策略,而有偏袒策略,亦即偏袒績效高的基金(高價值基金)而犧牲績效低的基金(低價值基金),且這種情形以高科技股票型基金較為普遍。(2)管理資產規模屬於中型,公司成立年數較久之投信公司,或者成立年數久但屬於中型規模的投信公司有偏袒策略的情形較為普遍。(3)研究結果亦顯示投信公司也有偏袒高費用基金之現象,但並不以高/低價值基金為同類型基金來執行偏袒策略。

並列摘要


Mutual funds have become an indispensable financial instrument for Taiwanese investors in asset allocation. Previous studies have been focused on the behavior of fund managers, such as herding behavior and window dressing of the fund holdings. Despite that the American mutual fund scandals happening in September 2003 have drawn our attentions on fund governance, there are few literature studies on fund company-level strategies. Therefore, this study intends to investigate whether fund companies in Taiwan actively pursue company-level strategies to maximize their overall profits at the expense of fund investors under the conflict of interest between fund companies and fund investors. The motivation for these company-level strategies is also examined in this research. The purposes of this research are: (1) to examine the fund performance—flow relationship in the Taiwan fund market. (2) to investigate whether fund companies pursue a risk-sharing strategy to support the fund whose performance is lagging behind. (3) to examine whether there exist favoritism strategies in fund companies—that is, fund companies give favoritism to high-contribution-value funds at the expense of the low-contribution-value funds for their companies. (4) to find whether a fund company’s risk-sharing strategy depends on the performances of the investment category for high-value funds or low-value funds. (5) to find whether a fund company’s favoritism strategy depends on the performances of the investment category for high-value funds or low-value funds. This study uses the Taiwan open-end domestic equity funds from Jan. 2001 to June 2010 as the research sample. The whole sample data can be divided into 7 categories: Common Equity Fund, Medium-Small Capital Equity Fund, High-Tech Fund, Value-Stock Fund, Theme Fund, Taiwan Enterprise Fund and OTC Equity Fund. The resulting base sample consists of 172 funds and 38 fund companies, including fund monthly return, monthly fund fee ratios, fund monthly total net assets, fund age, and company age, and the monthly rate of return of TAIEX. The cross-sectional regression and panel regression model are employed to examine the relationship between fund flows and fund performances. And the multivariate regression model is used to examine the fund company-level strategies. The results of this research are summarized as follows: (1) For the empirical results on fund performance—flow relationship, we find (a) the previous fund monthly ranking, no matter compared with the peer funds or the funds with the same company, is positively related to the growth rate of fund flows. There exists a similar U-shaped fund performance—flow relationship. In other words, fund investors chase past winners but also flee from past losers at nearly the same rate, which is a little different from the asymmetric fund performance—flow relationship of the U.S. fund data. (b) The top performers in the previous month have significantly positive effect on the growth rate of fund flows. (c) A crowding out effect is found when there exists at least one top performer within the same fund company. (d) The concurrent market rate of return is negatively related to the fund flows. (2) For the empirical results on fund company-level strategies, we find (a) there exist favoritism strategies in the fund companies—that is, the fund companies give favoritism to high-value funds at the expense of low-value funds. Besides, the favoritism strategies are more prevalent in the High-Tech Funds. (b) After considering the characteristics of fund companies, we find that favoritism strategies are more prevalent in the fund companies that have medium-sized management assets, set up earlier or belong to medium-sized and old companies. (c) The results also show fund companies give favoritism to high-fee funds.

參考文獻


Chih H. H., Y. E. Lin and P. H. Chou (2007). Performance Persistence and Smart Money Effect: Evidence from Taiwan. Journal of Management, 24(3), 307-330.
Li, K.L. (2005). A Study on the Relationship Between Fund Performance Persistence and Fund Flow of Taiwan Mutual Funds. Unpublished Ph.D. Thesis, Department of Finance, National Cheng-Chi University.
Lin M. C. (2007). Test of Coordinating Strategies across Funds: Conflicts of Interest between Fund Companies and Investors”, Journal of Financial Studies 15(1), 65-102
Yu C. H. and W. T. Tzeng (2003). The Spillover and Crowding-out Effects of Mutual Funds. Journal of Financial Studies 11(2), 99-123
1. References in Chinese:

被引用紀錄


李思遠(2012)。共同基金流量的周期性效果:以台灣開放式股票型基金為例〔碩士論文,國立臺北大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0023-2507201211391600

延伸閱讀