評估量化寬鬆政策之衝擊:對貨幣政策變數之緩長記憶和結構性改變之檢測

An Empirical Evaluation of Quantitative Easing: Testing for Long Memory Effect and Structural Breaks

王前崑(Chien-Kun Wang)

指導教授 ： 陳若暉

中原大學/商學院/企業管理研究所/碩士(2012年)

https://doi.org/10.6840/cycu201200301

摘要

2008年由美國次級房貸市場所引發的全球金融風暴造成全球經濟環境的緊縮。為防止通貨緊縮和流動性危機的發生，許多國家的中央銀行紛紛祭出擴張性的貨幣政策來拯救經濟體系和舒緩流動性不足的危險，但傳統的貨幣政策不足以解決此問題。為此，聯準會、英格蘭銀行、日本銀行和歐洲中央銀行紛紛實施了量化寬鬆政策，量化寬鬆政策是透過資產購買計劃，向市場注入大量的資金以壓低長期利率，以此來舒緩市場對資金的需求以及避免經濟走向更緊縮的狀況。
本研究利用ARFIMA-FIGARCH模型對美國、英國、日本及歐洲的公債殖利率、通貨膨脹、貨幣供給及匯率作檢測，分析量化寬鬆政策是否對其時間序列變數造成緩長記憶的現象。本研究也利用ARFIMA-FIGARCH模型對全部的期間作檢測(1995.01-2011.07)，也利用ICSS的檢測方法檢測量化寬鬆政策是否對時間序列變數造成結構性的改變，亦或者是多重的結構性改變。最後利用GARCH模型加上結構轉換之時間虛擬變數來檢測量化寬鬆政策是否有減少其時間序列變數的波動性。
實證的結果顯示緩長記憶的效果只存在日本和歐洲的通貨膨脹。量化寬鬆政策並沒有對公債殖利率、貨幣供給和匯率產生緩長記憶的效果。以全部期間作檢測的結果則顯示長期記憶存在於日本的通貨膨脹、歐洲的貨幣供給和歐元。在結構性改變的檢測上，量化寬鬆政策後，貨幣供給、通貨膨脹和匯率產生結構性的改變。但量化寬鬆政策的其中一個目標，即公債殖利率(歐洲)並沒有產生結構性的改變。另外，聯準會和英格蘭銀行曾執行過兩次的量化寬鬆政策，但實證結果顯示並沒有因此而產生兩次的結構性改變，意謂連續執行兩次的量化寬鬆政策的效果並不一定會比執行一次來得好。最後，GARCH模型的結果顯示透過長期傳統貨幣政策的調節(1995.01-2011.07)，結果顯示美國、英國、日本和歐洲的通貨膨脹和歐洲的貨幣供給有變得更穩定。但當只有考慮在執行量化寬鬆政策的期間時，量化寬鬆政策在使時間序列穩定上，不如傳統貨幣政策長期的調節有效率。

關鍵字

結構性改變 ； 量化寬鬆政策 ； 緩長記憶

並列摘要

The global financial crisis resulted by the U.S. sub-prime mortgage caused global economic deflation n 2008. Thus, many central banks were forced to implement expansionary monetary policy to rescue the economic system and deal with the liquidity risk. But the effect of convention monetary policy is ineffective. For solving that, the Fed, Bank of England, Bank of Japan and European Central Bank implemented the unconventional monetary policy – the Quantitative Easing (QE). Through large-scale asset purchase programs to decrease the long-term interest rate, the implementation of the QE relieved the pressure of demand of capital market and avoided deflation.
This study testes bond yield, inflation, money supply and exchange rate including the U.S., U.K., Japan and Europe. Using ARFIMA –FIGARCH (Autoregressive Fractional Integrated Moving Average- Fractional Integrated Generalized Autoregressive Conditional Heteroskedasticity) models to examine whether the time series before and after the QE and the whole period (1995.01-2011.07) have long memory effect. Then, this study also applies the iterated cumulative sums squares test (ICSS) to examine whether the QE results in the time series structural break or even multiple structural breaks. Finally, through GARCH model with the time dummy variable for structural break, this paper checks whether the QE decreased the volatility.
The results showed that QE had resulted in long memory only existed in inflation of Japan and Europe. Other factors like bond yield, money supply and exchange rate didn’t show the importance of modeling long memory effect. In the whole period, the long memory existed in the inflation of Japan, money supply for Europe and exchange rate for Europe. In structural break test, the QE had let money supply, inflation and exchange rate produced structural breaks. However, one of major goal of the QE, reducing the long term bond yield didn’t produce structural break in Europe. And there wasn’t happened multiple structural breaks, meaning the effect of twice QEs is not better than single impact of the QE. Finally, compared the whole period and the QE-period, through a conventional monetary policy in long term, the inflation for the US, UK, Japan and Europe and the money supply for Europe had became more stable. But when only considering the QE-period, the effect of unconventional monetary policy of the QE implementation is less effective than usual.

並列關鍵字

Structural break ； Long memory ； Quantitative Easing

參考文獻

14. Hsieh Jiun-Kuei and Lin Chien-fu (2004), “Long memory and regime switch”, Taiwan Economic Review, Vol. 32, 193-232.

20. Lin Shy-Chern (2005), “The relationship between monetary policy asymmetric and output: An application of Markov Switching model”, Master Thesis of Department of Economic, National Dong Hwa University.

36. Xiao Renwei (2006), “Multiple structural breaks in the real interest rate and inflation: Evidence from Asia countries” Master Thesis of Department of Applied Economics, Nation Chung Hsing University.

2. Beine M., A.B. Quere and C. Lecourt（2002）, “Central bank intervention and foreign exchange rate: New evidence from FIGARCH estimations”, Journal of International Money and Finance, Vol. 21, 38-115.

3. Huang Bwo-Nung and Yang Chin-Wei (2001), “The impact of settlement time on the volatility of stock market revisited: An application of the iterated cumulative sums of squares detection method for changes of variance”, Applied Economics Letters, Vol. 8, 665-668.