人民幣升值對中美貿易差額影響的實證研究

摘 要：近年來，人民幣升值影響了全球經濟的神經。實證研究人民幣是否被低估，并進行了詳細的分析是人民幣問題影響到美國經濟衰退。為了最大化研究人民幣升值的后果，我們更使用月度數據而不是年度數據。考慮到中國和美國的實際情況，我們還利用VAR模型檢驗了人民幣升值的跡象，發現在長期兩者不存在任何因果關系。

關鍵詞：中美貿易；人民幣升值；VAR

中圖分類號：F74 文獻標志碼：A 文章編號：1673-291X（2013）13-0233-03

一、Introduction

Two years ago，the US Senate passed “Currency Exchange Rate Oversight Reform Act of 2011”，the debate about exchange rate between China and US becomes hot again.In recent years，the United States frequently puts pressure on China by the exchange rate issue.Since a series of currencies measures were implemented from 2005，the US dollar against the RMB exchange rate decreased from 8.27 to 6.29，and pushed up the RMB still further.They argue that the US in Sino-US trade deficit may be reduced if the RMB appreciate，so they force the RMB to appreciate.This article would not only find the relationship between the Sino-US trade balance and the RMB exchange rate fluctuation，but also discuss how the exchange rates impact our country’s import and export trade.At last，some views about the dispute of Sino-US exchange rate are presented.

二、The literature review

In the 1930s，Abba Lerner states that if the current account is initially zero，a real currency depreciation causes a current account surplus if the sum of the relative price elasticities of export and import demand exceeds 1.If the current account is not zero initially，the condition becomes substantially more complex，then the external devaluation of the national currency will certainly be able to improve its trade balance of payments，which is the famous “Marshall-Lerner condition”.After the 1980s，economists increasingly note the problem of exchange rate pass-through，forming the theory of incomplete exchange rate pass-through.As well as the changes in a national income，currency devaluation can improve the trade balance conditions，and devaluation caused more than national income to increase exports，which is called Lawson - Metzler effect.If we do not take into account the time lag of monetary depreciation effect，due to exchange rate movements，the change in financial asset prices can be completed in an instant.But the exchange rate shocks may affect the economy with some lags.So when the exchange rate changes，the change in trade and national income is relatively slow，so the devaluation of the currency could lead to trade balance deteriorate first and gradually improve subsequently，which is known as the J-curve effect.

In the view of academic circle in the USA，with the expansion of China’s current account balance of payments surplus in 2005，Joseph Stieglitz，Robert Mundell，Stephen Roach and Ronald McKinnon firmly opposed to the appreciation of the RMB，but after2005，these scholars change their points.Although these scholars do not believe that a substantial appreciation of the RMB is conducive to the development of Sino-US economic and in varying degrees，they recognized the independent and progressive RMB appreciation.

From the point of Chinese scholars，Lin Yifu（2007），Li Yang （2005），Fan Gang，Yi Gang，Cheng Side and He Liping were tend to support the policy of RMB exchange rate is relatively stable，while Yu Yongding（2003），He Fan and Hu Zuliu argued that the yuan should be revalued.

三、Empirical research

This paper selects monthly data of Sino-US trade balance and the exchange rate of USD to RMB，from January 2007 to January 2013.Because of seasonal fluctuations in trade balance，firstly，carry out X-12 seasonal adjustment to the data，secondly，analyse data in the use of unit root test，cointegration analysis，and Granger test.The analysis find out the impact on Sino-US trade，exerted by the exchange rate.The results show causal relationship between the exchange rate and the trade balance is not clear，and trade balance have an affect on the exchange rate，but not obvious.The exchange rate exert a great influence on trade balance，however，it can not decide the size of trade balance between two countries totally，there is no necessary connection between them.

The data of the Sino-US trade balance comes from the official website of China Customs，and the data of the exchange rate is from the State Administration of Foreign Exchange，the two Chinese official data is strongly persuasive.For analysising and processing data，Y is the logarithm of the Sino-US trade balance，that is lnnx，X is the logarithm of the exchange rate of dollar against the RMB，that is lne.

1.ADF stationary sequence inspection

Classic regression model is the smooth variable based on the non-stationary variables，using the classical regression model may appear spurious regression and many other issues.Therefore，the economic variables stationary test is necessary.As we can see from the figure 1，The 1 is not significant at the 1%，5% and10% level，the value of t is less than the statistical values，passing the test.Although the original time sequence of the variables are non-stationary series，the corresponding first-order differential sequence pass the 5% significance level test，and individual variables are integrated of order one sequence.Because of the regular use of the 5% level，we also analyze at a 5% level in order to be calculated later consistent.

2.Least squares method

The least squares method to test whether there is a linear relationship between the Sino-US trade balance and the exchange rate.If there is a linear relationship，the relationship between the two is apparent.The actual inspection，we found that the R-value only 0.000656，even neglecting the possible linear correlation of D（x） and D（Y）.That is to say，a linear relationship does not exist between the Sino-US trade balance and the exchange rate.

3.Granger causality test

In the first test，the concomitant probability P = 0.9941，it cannot reject the 1 hypothesis of D（X） is not D（Y） Granger cause，and in the confidence level of 95%，it reject to made Type I error is 0.0059.So that D（X） is not D（Y） Granger causes，that is，the exchange rate is not the causes of the Sino-US trade balance.

In the second test，the concomitant probability P = 0.9812，D（Y） is not D（X）Granger cause that can not reject the 1 hypothesis.In the confidence level of 95%，it is not that D（Y）Granger causes D（X），namely Sino-US trade balance is not the exchange rate Granger causes.

四、The conclusion

Firstly，there is no stable equilibrium relationship between the Sino-US trade balance and the RMB exchange rate in the long term.

Secondly，there is no causal relationship between the RMB exchange rate and the Sino-US trade balance.So，it is impossible to change the Sino-US trade deficit，using the adjustment of RMB exchange rate.Because of the strong complementarity in Sino-US trade，exchange rate has limit affect on the balance of trade between the two countries.

Finally，the Sino-US trade balance did not show significant J-curve effect，which means that there isn’t time lag effect of the exchange rate on bilateral trade balance.The conclusion of the exchange rate on the trade balance do not play an important role has been confirmed once again.Because of the RMB exchange rate is not the bilateral trade balance of payments reasons，and thus through the adjustment of the RMB exchange rate can not solve the problems of the current balance of trade between China and USA.As trade between China and the United States，there was not only complementary，and there is a lot of competition，so we can not wish to improve Sino-US trade payment by forcing the appreciation of RMB exchange rate.

References：

[1] Chen，Jinzhao，2010.Crisis，Capital Controls and Covered Interest Parity： Evidence from China in Transformation.CESifo Venice SummerInstitute，19-24 July 2010.

[2] Buigut，S.， Rao，V，2011.International Financial Integration of the Indian Money Market.International Journal of Economics andFinance，3（4），170-180.

[3] Liu Ganzhou，2006.The analysis of mechanism for interaction between exchange rate fluctuations and fluctuations in the stock marketprice，Guangxi College for Financial.

[4] Paul R.Krugman，Maurice Obstfeld，2003.International Economics Theory and Policy.Pearson Education.

[5] Zhang Xiaotong，2007.A Guide To Using Eviews.China Machine Press.