The Influence of The President Taipei and Kaoshiung Mayor Elections on Stock Returns

[[abstract]]中文摘要 本研究利用事件研究法的基本概念，選擇了三次總統選舉及二次北高市長選 舉做為事件樣本，以股市報酬率做為分析及解讀的標的，並將事件日前後各區分 了一個事件期及估計期，並採用不同的檢定方法以了解是否有選舉行情的存在。 早期的研究中多數肯定國內的選舉時期容易產生選舉行情及慶祝行情，在本 研究中不論是以事件研究法中的無母數檢定法或有明確事件日的普通橫剖面法來 分析，或者是另外採用變異數分析法及時間序列迴歸分析法來分析選舉前後之市 場報酬率，均得不到實證之支持，原因之一在於金融商品的多樣化提供了投資人 另外獲取超額報酬的機會，原因之二則是國內股市投資環境的改變，原因之三則 是股市是經濟的櫥窗，而構建經濟的成份不單只有選舉事件。 參考目前台灣交易所公佈的數據顯示，台灣目前股市的投資人比重，由散戶 型轉變為法人居多，外資、投信及自營商等三大法人的投資比重已逐年提升， （由民國八十年的3.08％提升至九十四年時的31.16％），目前上市公司家數更高 達693 家，眾多推陳出新的投資商品提供無數避險及投機管道，加上民主自治的 環境讓政黨有了輪替的機會，要在台灣股市做價已不如民國七、八十年時來得容 易而且低成本了。因此未來若有客戶在以後的選舉事件中期待選舉行情或慶祝行 情的產生，藉由本研究的結果可以提出有力的建議，應該要有較為保守的心態。 在本研究中國際股市的部份，可看出在政黨輪替之前，台股亦有少陴瑽t 出的經驗，但在政黨輪替之後，選舉事件似乎不再是構成脫序的主因，反而台股 與國際行情有更為共進退的現象，相關性逐年提高。而較為特別的是一般在參考 國際行情與台灣的互動性時，多數的投資機構及建議均是以美國股市為指標，但 在本研究中察覺與台灣較有相關性的是日本指數。這是整個研究中相當有趣的地 方，推測日本和台灣不論是國情、民風、經貿、氣候、出口內需、政治生態等 等，均有較高的相似度，因此在股市方面的表現亦有較高的相關度。 研究最終之目的在於了解選舉的結果是否會對股市投資造成長遠的影響，在 政黨輪替的前題下，發現選舉並不如早期所認同的具有選舉行情及慶祝行情的發 展，因此股市投資人在選舉事件上應以平常心看待。[[abstract]]ABSTRACT The basic concept of event study is introduced into this study choosing three times of President elections and two times of Taipei and Kaoshiung mayor elections from domestic elections as sample events and taking stock market returns to be the target of analysis and realization We divide days around event day into two parts each of them contains event period and estimate period four time intervals in all Each time interval is testing by different statistical method to see if there is election bull run In the studies of earlier period the most affirms in domestic election periods that the election bull run is easily to be happened While using nonparametric methods or Ordinary Cross-Sectional Method ANOVA analysis Time Series regression analysis we can not get sufficient evidences by tis study since there are 3 reasons The first reason is the diversification of the financial merchandise provides the investors to obtain another chance of excess profits The second reason is the environmental change of investment in domestic stock market The third reason is that stock market is an economic window and economy contains more than elections The reference of recent data announced by Taiwan Stock Exchange Corporation manifests that the proportion of investors in Taiwan Stock Market which consists of foreign investors Securities Investment Trust Companies and Dealers is increasing (rising from 3 08% in 1991 to 31 16% in 2005) although the proportion of individual investors is decreasing The total of listed companies is up to 693 currently Numerous merchandises that weed through the old to bring forth the new provide innumerable ways of hedge and speculation And in this democratic country party alternation is likely to happen again and again Synthesize the above one can hardly manipulate the stock market more than in 1980’s and 1990’s If customers expect that election bull run will take place in the next elections we should strongly recommend them to be conservative by the conclusion of this study In this study we see that some times Taiwan stock market’s performance is divorced from the other international stock markets before party alternation occurs However after the party alternation elections seem not to be the main factor to influence the stock market performance of Taiwan stock market itself binds the other markets and the connection goes higher and higher There is something special in this study we find that Taiwan stock market is relative to Nikkei 225 rather than Dow Jones although most of the investment institutions take U S stock market as a standard when we survey the interaction between Taiwan stock market and the international stock markets This is a very interesting view in the whole study we assumed that the national conditions folkway economy climate exportation domestic demand politics etc are similar between Taiwan and Japan So the performances of the two stock markets connect closely The final purpose of this study is to realize that the results of elections if there are profound influence on stock investment After the party alternation we find that the election bull run does not singnificantty exist which differs from the early studies As a stock market investor one should treat this result for calmnes