| 摘要: | 本文就中央銀行的貨幣政策對臺灣股票市場報酬率及波動性的影響力是否穩健(robust)，進行實證分析。本文探討的目標為二元：一為縱斷面的時間序列(time series)的分析；二為橫斷面跨部門(cross-section)的分析。在時間序列分析方面，我們將自1981年1月5日至2004年12月31日臺灣股市的日報酬率資料，分為四個次期間，以分析在這四個次期間內，中央銀行貨幣政策的改變對股票報酬率的平均數與波動性的影響力是否維持一致，或者有加深或減弱的現象。
跨部門分析方面，我們除了分析臺灣大盤加權股價指數外，我們亦採用French的分類方法，將股票依投資策略分為八種類型加以分析其受貨幣政策之影響：依市值分為大型股與小型股；依帳面價值與市值之比分為價值股與成長股，並繼續加以細分為大型成長股、大型價值股、小型成長股與小型價值股共八種投資策略型股。此外，由於不同的產業對貨幣政策的寬鬆與緊縮的敏感度不同，所受的影響也不同，故本文採用臺灣證券交易所所公佈的產業類別指數，也對18種產業類股指數進行實證分析。
在分析報酬率的平均數的變動，本文使用年平均報酬率；在報酬率的波動方面，除了使用報酬率離散的情況來作分析外，也利用GARCH-M模型，來探討股票報酬率的波動，即股票的風險，是否會影響到股票的報酬率，並探討臺灣股票市場的風險與報酬之關係，是否股價的行為合乎傳統的資本資產定價理論，在市場波動程度大時，投資人會要求較高的報酬，作為承擔風險的補償，還是合乎波動回饋理論(volatility feedback)或是杠杆效果(leverage effect)，其宣稱高風險反而只有低報酬的現象，或臺灣的股票市場對風險與報酬率並無太大相關。
我們的實證發現股市大盤指數在7個緊縮貨幣政策期間，有6個貨幣政策期間的報酬率均為負，而在6個寬鬆貨幣政策期間，有5個貨幣政策期間的報酬率為正，在統計上有顯著的差異。在股票報酬率的波動方面，發現在緊縮與寬鬆期間也有顯著的差異，在緊縮貨幣政策期間，股市波動程度比較大。實證結果並且顯示貨幣政策的影響力沒有減弱的現象。至於風險的大小與報酬率是否有關係，我們的實證顯示，大盤指數在緊縮貨幣政策期間，風險與報酬沒有關係，但是在貨幣政策寬鬆期間，風險與報酬有顯著的正相關，表示在貨幣政策寬鬆期間、高風險時，投資人會要求較高的報酬率。
至於貨幣政策對投資策略型股之影響，我們的實證發現，貨幣政策的寬鬆與緊縮對所有的投資策略型股均有顯著的影響，且影響力並沒有減弱的現象，寬鬆貨幣政策期間的報酬率顯著大於緊縮貨幣政策期間的報酬率。但是在波動率上，以全期間來看，貨幣政策的改變對小型股和價值股的波動率較有影響，但是影響力到了後段期間似乎減弱了。至於在風險與報酬之間的關係，實證顯示，在貨幣政策緊縮期間，所有八種投資策略型股的報酬率都不受風險大小的影響，但是在貨幣政策寬鬆期間，大型類股與成長類股的報酬率與風險則呈顯著的正相關，表示這些類股對高的風險會要要求高的報酬率。其餘投資策略股型則有不顯著的正相關。
在十八種產業類股方面，我們的實證發現貨幣政策的寬鬆與緊縮對其中十三種產業類股的報酬率有顯著的影響，其餘五種則不論在貨幣政策寬鬆或緊縮時期，報酬率均無太大差異。貨幣政策影響加深的產業有十項，大多是傳統產業，而受貨幣政策影響程度下降的有八項類股，大部份是服務業或新興產業。至於報酬率的波動，貨幣政策對各產業類股報酬率的影響有不同的結果，但是我們看不出貨幣政策對報酬率的波動的影響力有減弱的現象。在風險與報酬率之間的關係顯示出，大多數的類股指數報酬率，不論是在貨幣政策緊縮或寬鬆期間，股票風險的大小，對報酬率並沒有影響。
本論文分為六章，第一章為緒論，敍述臺灣經濟與股市發展概況與中央銀行的貨幣政策工具，第二章為理論基礎與文獻探討，分別討論貨幣政策傳遞機制，再貼現貨幣政策，資產定價與股市異常效應，貨幣政策與跨產業股價報酬率，風險與報酬。第三章為貨幣政策對大盤指數報酬率與波動的影響分析，以及風險與報酬率的關係。第四章為貨幣政策對投資策略型股的報酬率與波動的影響分析，以及風險與報酬率的關係。第五章為貨幣政策對產業類指之報酬率與波動的影響分析，以及風險與報酬率的關係分析。第六章為結論。
This thesis examines the effect of the monetary policy on the returns of the Taiwan stock market as well as changes of volatility through empirical tests. The purpose of this thesis is two-fold: to analyze the vertical time series and examine a cross-section of industries. For purposes of the time series analysis, the data for stock market returns during the period from January 5, 1981 to December 31, 2004 is divided into four sub-periods. In analyzing each of the four sub-periods, one can determine whether changes in the monetary policy are consistent with the effects of mean returns of stocks as well as volatility or if there are signs of either weakening or strengthening. In terms of the cross section analysis, we used French’s classification system to classify stocks according to investment strategy into 8 categories: big stocks and small stocks according to market value, value stocks and growth stocks according to book-to-market ratio, and further classified into big growth stocks, big value stocks, small growth and small value stocks. In addition, due to the fact that different sectors have varying degrees of sensitivity to changes in monetary policy, the degrees in which they are affected will be variable as well. For this reason, this thesis uses the sector indices published by the Taiwan Stock Exchange in order to conduct a meaningful analysis of eighteen sector indices.
In analyzing changes in the mean return, this thesis uses annualized mean returns. In regards to the volatility in returns, in addition to using the return deviations for analysis, the GARCH-M model developed by McCurdy and Morgan was used to explore such changes in order to determine if they were affected by risks. In doing so, this will determine the behavior of stock prices conforms to one of several possible models. The traditional capital asset pricing model assumes that during periods of market volatility, investors will seek higher required returns in order to justify higher risks. Other models include Pindyck’s volatility feedback theory and Black’s leverage effect theory, which posits that high risks only result in low returns. Another possibility would be that there is no causal relationship between risks and returns in the Taiwan stock market.
The results of the analyses have shown that of the seven restrictive monetary policy periods in the Taiwan weighted index, there are six of them the average rate of return was negative, while there were five monetary policy periods in which the average rate of return was positive in six expansionary periods where there was a statistical significance of the difference in return between restrictive and expansive monetary policy periods. In using return deviations to measure stock volatility, it has been found that there have been significant differences in the expansionary and restrictive periods, while the statistics indicate that the effects of monetary policy have had no signs of weakening. In regards to the possible relationship between the risk and the rate of return, analysis of the data has shown that in the restrictive period, there has been no correlation with risks and returns in the Taiwan weighted index. However, in the expansionary period, there is a significant positive relationship between the two. This indicates that in the expansionary monetary policy period, investors will seek greater returns during periods of higher risk.
In terms of the effects of monetary policy on investment strategy performance, it was also found that changes in monetary policy had a significant impact on all investment strategy performances and show no signs of weakening. However, the volatility rate in an full period has shown that changes in monetary policy seems to have a greater impact on small stocks and value stocks, although the effect seems to weaken in later stages. Empirical evidence also shows that during restrictive periods in monetary policy, the returns rates of all eight investment strategy performances were not affected by the degree of risk. Conversely, during expansionary monetary policy periods, there is a significant positive relationship between degree of risk and the rate of return of big and growth stocks. This indicates that these stocks require high rates of return when faced with high degrees of risk. Other forms of investment strategy performance do not seem to have a significant positive relationship.
Of the eighteen industry sector stocks, the empirical evidence revealed that expansionary and restrictive periods had a significant impact on the return rates of thirteen of them, whereas the remaining five remained unaffected. There is a deepening relationship between monetary policy and ten of the industry sectors, while it is a weakening relationship for the other eight sectors, mostly concentrating in the service and emerging sectors. In terms of the return volatility, although the impact of the monetary policy on the return varies depending on the stock, there do not seem to be signs of weakening effect of monetary policy on the volatility. In terms of relationship between risk and return, in most industry sectors, there are no signs of such relations, meaning the degree of risk would not affect the size of returns.
This thesis is divided into six chapters. Chapter one is an introduction which details the development of the Taiwan economy and stock market as well as the monetary policy tools. Chapter two is the theoretical basis and related research, discussing transmission mechanisms of monetary policy, rediscount rate policy, asset pricing and stock anomalies, monetary policy and cross section return rates, and risks and returns. Chapter three describes the affects of monetary policy on the Taiwan weighted index. Chapter four describes the affects of monetary policy on investment strategy performance. Chapter five describes the affects of monetary policy on the returns of industry indices. Chapter six contains conclusions and suggestions. |
