匯價相關因素變動對台灣外銷訂單金額影響之研究

The Study of the Effect of Exchange Rate Related Factors on Export Amount in Taiwan

本文以台灣主要外銷國家：美國、中國、日本、歐洲為研究對象，研究期間為2009年至2014年。採用似無相關迴歸(seemingly unrelated regression)探討匯價相關因素變動，包含國內生產毛額、貨幣供給、消費者物價指數、匯率、及匯率波動等對台灣外銷訂單量的影響。實證結果顯示，自量化寬鬆後，美國的貨幣供給對外銷呈正相關；但中國與歐洲貨幣供給對外銷呈負相關。消費者物價指數四個國家分析結果與預期一致，對外銷訂單量呈負向影響。在國內生產毛額中，美國、中國及日本，與預期一樣呈正相關；然而歐洲為負相關。在2009年至2014年間，日本與歐洲匯率升值，外銷訂單量增加，意味著匯價對出口呈正相關；然而美國與中國則呈負相關。匯率波動率即匯價跌則與匯率呈負相關。模擬在貨幣供給及匯率分別每季增加3%後，得知各國貨幣供給的外銷訂單量效用只有日本較明顯，為貨幣供給增加的1.3倍；各國匯價效用皆不低，尤以中國的匯率變動效用最大，為匯價增加的3.5倍，歐洲的二者效用最小。此外發現在2014 Q4外銷訂單量有明顯增加的情況，因此考慮到季效果加入季虛擬變數，顯示季效果明顯且外銷訂單量預測結果變好。

In this paper, the main exports from Taiwan to United States, China, Japan, and Europe are studied during the period of 2009 to 2014. The seemingly unrelated regression (SUR) approach is used to explore the impact of exchange rate related variables on Taiwan's export order volume, including gross domestic product, money supply, exchange rates, consumer price index, and exchange rate volatility. The empirical results show that the money supply of the United States has a positive correlation to exports since the quantitative easing policy started while China and Europe exhibit negative correlation. As expected, consumer price index of all four countries shows negative correlation to exports. Also as expected, gross domestic product of the United States, China and Japan has been evidenced positive correlation except Europe. During 2009 to 2014, the exchange rates of Japan and Europe were appreciated and export orders increased, showing that the exchange rates of both countries have a positive correlation to rxport orders. However, the United States and China expose negative correlation. As to the effect of exchange rate volatility resulting in dropping exchange rate, it acts negatively to the exchange rate accordingly. Simulating 3% increase of the money supply and exchange rate respectively for each country, the scenario results show that only Japan in four countries has significant money supply effect on export orders with 1.3-fold larger than money supply increase. The exchange rate effect of each country is not low especially for China having 3.5-fold larger than exchange rate increase. Europe is lowest for both effects. In fact, 2014 Q4 sales surged apparently. Thus, considering quarterly seasonal effect and adding seasonal dummy variable to SUR, it shows the seasonal effect is significant and the prediction of rxport orders becomes improved.