人民幣匯率動態與波動外溢效果

Renminbi Exchange Rate dynamics and the Volatility Spillover Effects

黃俊傑(Chun-Chieh Huang)

指導教授 ： 陳珮芬

國立暨南國際大學/管理學院/國際企業學系/博士(2015年)

摘要

本文主要是探討中國匯率採用浮動匯率以盯住一籃子的貨幣。同時，亦探討波動外溢效果。過去中國主要是以固定匯率為主，然而，隨著中國對外貿易順差加大，促使中國於2005年7月將原本的匯率政策改成浮動匯率，同時宣告盯住一籃子（參考）貨幣包含美元、歐元、日元及韓元，意謂不再以美元做為主要指標貨幣。因此，本文的研究重點首先針對人民幣匯率與其參考貨幣於匯率變動後之長、短期之影響。其次，進一步的以全球金融危機時間來分析人民幣與參考貨幣之間的跨市場之波動外溢效果。最後，根據中國人民銀行所採取廣義盯住11種貨幣的訊息來分析自人民幣改革後所所產生的波動外溢效果是否會直（間）接的影響到11種貨幣市場。
根據上述研究動機，本文提出三個相關的研究主題來驗證人民幣改革對於其參考貨幣之影響。第一篇、分成二部分來探討。首見分析人民幣改革與其參考貨幣彼此之間有無顯著長期共整合，其次，則檢驗人民幣與其參考貨幣有無短期調整至均衡。第二篇、驗證人民幣與其參考貨幣在三個不同的樣本期間彼此之間有無顯著波動外溢效果，特別是美元。第三篇、檢驗人民幣與其參考之11種貨幣之間有無顯著之波動外溢效果。
本論文採取用三種不同的方法來呈現。首先，藉由共整合方法來探討人民幣與其參考貨幣彼此之間的長期效果，同時使用特徵值及特徵向量來分析短期調整係數於人民幣與其參考貨幣的調整過程。其次，則是運用二元時間序列模型來分析人民幣與為其參考貨幣之間的波動外溢效果。最後，則是借由向量自我迴歸模型來探討人民幣與11種參考貨幣之間於改革後的外溢波動效果。三項主題分析頻率皆以日資料為主，期間以人民幣改革（2005/07/20）做為樣本起點含蓋全球金融危機及各種不同金融事件。
第一篇的研究結果顯示，人民幣與其參考貨幣在樣本期間存在著長期共整合關係。多種結構改變點檢測結果至少有三個轉變點。另外，使用滾動過程來估計誤差修正項，結果顯示短期人民幣均衡在全樣本及子樣子期間是細小且緩慢的調整的。這意味著短期人民幣匯率會做持續性的動態調整至均衡。
第二篇實證結果顯示人民幣與其參考貨幣在樣本期間存在著雙向關係。實證結果亦發現在危機前人民幣與美元存在著雙向的波動外溢效果。相反的，在危機後時期人民幣與美元則沒有任何的波動外溢效果。
第三篇實證結果發現人民幣與11種貨幣的報酬與波動外溢效果來自於人民幣有顯著的影響。美元相對應到其他貨幣亦存在著報酬與波動外溢效果，例如：人民幣和歐元。最後，在樣本期間亦發現各種不同的危機事件，例如：全球金融危機及歐債危機。
綜合上述實證結果，人民幣匯率改革在改革前雖美元為主要參考指標。改革後雖然人民幣宣告不再以美元為指標，然而，人民幣與美元關係並無明顯的下降。因此，本文建議本（外）國投資人對於人民幣匯率改變可以就政策宣告所參考各種不同貨幣做投資避險選擇，首選以美元為主，然而投資者亦可選擇其他貨幣例如歐元或日元。另外，提供政策制定者在做出匯率政策變革時更時思考其政策是否會產生利益缺口，避免投機者從中獲取不當利益。

關鍵字

人民幣 ； 共積檢定 ； 波動外溢 ； 多變量模型 ； 向量自我迴歸 ； 一籃子貨幣 ； 匯率政策

並列摘要

This dissertation mainly examines the foreign exchange policy of China under the target of a currency basket and the effects of the volatility spillover of Renminbi (RMB) toward the targeted currencies. Since 1993, the China government monetary policies employed the fixed exchange rate regime. However, under the pressure of the US government, the Chinese governments have announced a new exchange rate regime which refers to a basket of currencies concerning the importance of trade partners. The basket of currencies includes USD, European (EUR), Japanese Yen (JPY) and South Korea Won (KRW). The first issue of the dissertation focuses on the impacts of the long-run and short-run relationship between RMB and its reference currencies. The second issue extends the first issue and examines the cross -market spillover volatility effects between RMB and its reference currencies markets according to the global financial crisis. The third issue investigates the direct and indirect volatility spillover effects between RMB and a basket of 11 currencies.
I adopt a series of related methodologies to apply the empirical investigations. These methodologies include, rolling cointegration, VAR model and time-varying spillover with the concern of possible structure breaks.
Chapter 3 presents the first article. The rolling cointegration tests indicate the presence of long-run relationship between RMB and its reference currencies. Further, the multiple structural breaks are detected for at least three breakpoints. In addition, I estimate the error-correction model and the evidence shows the short-run adjustment for RMB toward it equilibrium is small and sluggish during full and sub-samples. This result suggests that the RMB dynamics is persistent toward the equilibrium.
Chapter 4 presents the second article indicates that the RMB and its reference currencies exists two-way relationship. Nevertheless, the evidence shows the bidirectional spillover volatility effects between RMB and USD during pre-crisis period. In contrary, there is no finding of the spillover volatility effects between RMB and USD during post-crisis period.
Chapter 5 presents the third article. The empirical investigation indicates that RMB has a strong spillover effect to the rest of currencies in both returns and volatilities. On the other hand, the returns and volatilities of the rest of currencies do not spillover toward RMB except the USD. Finally, the events of global financial markets are identified from the time-varying spillover index.
To summarized the findings of the dissertation, the RMB exchange rate dynamics are tied to the USD and other reference currencies. The spillover effects of exchange rate returns and volatilities are presented among RMB, the USD, and the currency basket. However, the relationship is subject to policies and financial events.

並列關鍵字

Renminbi (RMB) ； Cointegration test ； Volatility Spillover ； Multivariate Model ； Vector Autoregression ； a basket currency ； Foreign Exchange policy

參考文獻

Alaganar, V. T., Bhar, R. 2002. Information and volatility linkage under external shocks Evidence from dually listed Australian stocks. International Review of Financial Analysis. 11, 59-71

Antonakakis (2012) Exchange return co-moments and volatility spillovers before and after the introduction of euro. Journal of International Financial Markets, Institutions and Money. 22, 1091-1109.

Baak, s., 2008. The bilateral real exchange rates and trade between China and the U.S. China Economic Review. 19:2, 117–127.

Bai, S., Mollick, A. V., 2010.Currency crisis and the forward discountbias: Evidence from emerging economies under breaks. Journal of International Financial Markets, Institutions and Money. 20(5), 556-574.

Bai, J., Perron, P., 1998. Estimating and testing linear models with multiple structural changes. Econometrica. 66, 47-78.