景氣循環、總體經濟變數與台灣股價指數的關係性研究

Abstract

景氣循環（business cycle）與企業的生產行為與個人的消費行為息息相關，而股市成份企業的營運狀況與策略亦會反向回饋至景氣的榮枯，因而推論景氣循環、總體經濟變數與股價指數之間應存在極為密切的關係。惟基於股價指數在不同的總體環境背景下與總體經濟變數不易存在長期的線性關係，本研究因而僅以皮爾森相關係數做為研究工具，探討每一景氣循環內股價指數與總體變數之間的關係性，又股市一向有經濟櫥窗之說，因此研究方法的設計將針對不同的景氣循環，探討股價指數對總體經濟變數的領先性。研究結果顯示，股價指數對景氣循環後期之GDP、就業、進出口貿易總值、製造業銷售值存在領先性，對景氣循環前期之GDP、票據交換金額、進出口物價指數存在同時性，而對工業生產指數僅於第十循環存在領先性；其次，本研究發現近期股市對GDP的領先性有增加之勢，推論此與台灣資本市場的成熟度增加有關；而早期股市與票據交換金額雖具同時性，但於股票交易改以券款劃撥且近年支付工具多元化後，該指標之代表性降低，因而與股市的同時關係性亦不復存；另股市雖對就業、進出口貿易總值、製造業銷售值與工業生產指數存在程度不等的領先性，但由於此部分變數與經濟體的長期結構變化或政策有關，因此股市對此部分變數的領先性在部分景氣循環階段並不明顯；而在實務上，由於經建會編制景氣指標時將股價指數納入領先指標之構成項目，而將本研究擷取之多項總體經濟變數列為同期指標之構成項目，亦即股市原本即存在對總體經濟變數的領先關係性，且經建會於2007年7月重新篩選構成項目時，亦將同時指標代表性下降之票據交換金額剔除，與本研究所獲之實證結果相符。

Business Cycle refers to the economy-wide fluctuations in production or economic activity over a period of time. Since its trough or peak, expansion or contraction, deeply relates to enterprises’ behavior of production and even the performance of stock market index, which is composed of the combined performance of a certain aggregate of individual publicly-listed company stocks. However, considering that no frictionless world exists, there should be a non-linear relationship between stock market index and macro economic factors in every business cycle. As such, we simply use the value of Pearson Correlation Coefficient along with the testing results of the correlation coefficient’s significance to review if the stock market index is an effective leading indicator to macro factors, and if the leading relationship differs in different business cycle. In summary, the findings are described as follows. First, the TAIEX effectively leads GDP, employment, export and import values of foreign trades and manufacturing variables, but only coincides with the bill exchange amount in relatively early stages. Second, no significant relationship exists between TAIEX and industrial production index except in the 10th business cycle. Third, the historical development and the relationship between TAIEX and macro factors can be partly explained by the empirical results. Finally, given that TAIEX and some macro variables selected in this research are components of official leading index and coincident index respectively, the empirical results are generally in line with pragmatic ways taken by CEPD.