The exchange rate plays a vital role in the healthy development of a country's economy, and the performance of the foreign exchange market has been paid more and more attention by the monetary authorities of various countries, because the exchange rate is not only the transmission channel of a country's monetary policy, but also the fuse of a country's monetary crisis. At present, since July 2, 20051China's RMB exchange rate system reform, the RMB exchange rate has generally shown an upward trend. By July 22, 2009, the exchange rate of RMB against the US dollar had appreciated by 2 1% in the past four years, but the international community still called for RMB exchange rate appreciation. Improper handling may lead to internal and external imbalances in the economy and aggravate trade frictions.
Literature review
The research on RMB exchange rate mainly includes the following two aspects.
The first aspect focuses on the exchange rate level, that is, the RMB equilibrium exchange rate level, which is the argument basis of whether the RMB exchange rate level is overestimated or underestimated and how much it is underestimated, and also the argument basis of supporting and opposing the appreciation theory. Robert mundell (2005) thinks that no matter how fierce trade disputes China faces, China should keep the RMB exchange rate stable. He even thinks that the policy of RMB pegged to the US dollar should remain unchanged for the next 20 years. Japanese financial economist and prime minister's think tank Haruhiko Kuroda (2005) thinks that RMB should appreciate slowly. Domestic scholars Shi Jianhuai (2005), Fan Conglai (2004), Zhang Bin (2003), Lin (2002) and Zhang Xiaopu (1999) estimated and discussed the equilibrium level of RMB exchange rate through relevant econometric models. It is worth noting that Zhang Shuguang (2005) deeply and systematically studied the macro-economic background of RMB exchange rate and the cost-benefit problem of exchange rate appreciation, and pointed out that the appreciation pressure was caused by a series of reasons, such as internal and external economic imbalance, sustainable tradable goods sector with persistent double surplus and labor productivity. While analyzing the cost-benefit of appreciation, he also put forward some suggestions on the choice of appreciation methods and countermeasures.
On the other hand, it focuses on the formation mechanism of exchange rate, that is, under the condition of open economy, the influence of different exchange rate systems on the internal and external equilibrium of macro-economy and the safe operation of economy. Some foreign scholars have studied the international exchange rate after the financial crisis in Southeast Asia, especially the changes of exchange rate system in crisis countries. They believe that the exchange rate system of various countries has a trend of bipolar development, that is, either a fixed exchange rate system such as the currency board is implemented, or the peg is abandoned and the floating exchange rate system is changed. The main representatives who hold this view are eichengreen (1999), Fisher (200 1) and Edwards (2006). Domestic scholars are more or less influenced by the "polarization" of the exchange rate system when demonstrating that the RMB should abandon its rigid peg to the US dollar and implement a managed floating. For example, both Wang Xuewu (2000) and Ding Jianping (2002) believe that the reform of RMB exchange rate should consider the development trend of international exchange rate system.
However, the above research mainly focuses on the RMB exchange rate level and exchange rate formation mechanism, especially the exchange rate level. However, in the short term, few people have analyzed whether China's money supply and foreign exchange reserves have an impact on the exchange rate of RMB against the US dollar, which is exactly what this article is going to elaborate.
empirical analysis
This paper mainly analyzes the short-term relationship between the target money supply and foreign exchange reserves of monetary policy and the exchange rate of RMB against the US dollar. The data comes from the statistical database of China People's Bank. The data selection period is from 2000 10 to 65438+ in February 2009, and the sample is * * 120. Granger causality test is the main method used. Before Granger causality test, time series stationarity test and cointegration test are needed. All calculations in this section are completed by the econometric software Eviews5.0.
(A) Time series stationarity test
The purpose of stationarity test of time series is to eliminate the phenomenon of pseudo-regression between regression-related variables of time series. The ADF(Augmented Dickey-Fuller) test is actually the least square regression of the following equation under the null hypothesis H0: ρ = 0 (RI is a first-order integer sequence).
In the following, LnM2, LnM 1, LnM0, LnFER and LnER are used to indicate the monthly growth ranges of M2, M 1, M0, FER and ER, respectively. Δ lnM2, Δ lnm 1, Δ lnM0, LnFER and LnER are used to represent the first-ordER difFERences of m2, m1,m0, fer and er respectively, that is, the monthly increase and decrease values of these variables. (foreign exchange resERve: exchange rate), in which M2 is the broad monetary quantity, M 1 is the narrow monetary quantity, M0 is the cash in circulation, FER is the foreign exchange reserve, and er is the exchange rate of RMB against the US dollar. The ADF test results of Ln M2, Ln M 1, Ln M0, LnFER and LnER and their differences are shown in table 1.
According to the ADF test results of table 1, M2, M 1, M0, FER and ER are all unstable at the significant level of 5%. First-order difference or logarithmic first-order difference is performed on er to obtain δ△ lner, and second-order difference or logarithmic second-order difference is performed on the rest parts to obtain D (δ LNM2), D (δ LNM 1), D (δ LNM0) and D (δ LNFER). Then, ADF test statistics are all less than the significance level of 65438+. It shows that the original hypothesis can be rejected at least at 99% confidence level, and the difference series D (δ LNM2), D (δ LNM 1), D (δ LNM0), D (δ LNFER) and δδlnER are stationary time series. ThERefore, the five sequences M2, M 1, M0, FER and er have the same integer order and are all second-order integer I(2) processes.
(B) cointegration test
Since the variables D (Δ lnm2), D (Δ lnm1), D (Δ lnm0), D (Δ lnfer) and lnER all pass the unit root test, they are unary variables, so we can carry out co-integration test on these variables. The difference method used in the unit root test of variables Ln M2, Ln M 1, Ln M0, LnFER and LnER will make the information of long-term relationship between variables lose. The purpose of cointegration test is to determine whether there is a long-term economic relationship between two variables. If two variables pass the cointegration test, we say that there is a long-term economic relationship between them. There are many methods of univariate cointegration test, such as Zt statistics and Zρ statistics of Philips-Perron)PP method, T statistics of ADF test, Johansen test and so on. Johansen cointegration test is used in this paper, and the test results are shown in Table 2.
According to Johansen co-integration test results in Table 2, variables d(δLn M2), d(δLn m 1), d(δLn M0), d(δLnFER) and d(δlnER) have a long-term equilibrium relationship at a significant level of 5%, which means that there is a long-term interaction between variables.
granger causal relation test
Whether China's money supply and foreign exchange reserves will interact with the exchange rate of RMB against the US dollar in the short term. Granger causality test is used here to determine the interaction between variables. The test results are shown in Table 3.
Granger causality test results show that:
1. When the lag period is 2, the probability that the change of the natural logarithmic second difference of foreign exchange reserves will not cause the change of the natural logarithmic second difference of RMB against the US dollar is 0.39366. The lag period is 6 and 12, and the probability rises to 0.52702 and 0.67309 respectively. It can be seen that foreign exchange reserves have a certain influence on RMB against the US dollar in the short term, but in the long term, this influence is opposite. When the lag period is 2, the probability that the change of the natural logarithmic second difference between RMB and the US dollar will not cause the change of the natural logarithmic second difference of foreign exchange reserves is 0. 102. In the short run, the change of RMB against the US dollar has a significant impact on foreign exchange reserves, but in the long run, this impact is also weakening.
2. From the point of view of money supply, when the lag period is 6, the probability that the change of the natural logarithmic second-order difference of broad money supply will not cause the change of the natural logarithmic second-order difference of RMB against the US dollar is 0.34284, while the change of the natural logarithmic second-order difference of RMB against the US dollar is the Granger cause of the change of the natural logarithmic second-order difference of broad money supply; When the lag period is 4, the Granger causes of narrow money supply and RMB exchange rate against the US dollar are just the opposite of those of broad money supply and RMB exchange rate against the US dollar. When the lag period is 1, the exchange rate between cash in circulation and RMB against the US dollar is not Granger causality. These conclusions clearly show that in the short term, the narrow money supply causes the change of the exchange rate of RMB against the US dollar, and the exchange rate of RMB against the US dollar causes the change of the broad money supply, and the cash in circulation is not affected by the exchange rate.
3. Similarly, as can be seen from Table 3, there is no Granger causality between broad money supply and foreign exchange reserves, while broad money supply and narrow money supply are Granger causality.
Conclusion and countermeasure analysis
By collecting the relevant data of money supply, foreign exchange reserves and the exchange rate of RMB against the US dollar in a short period of time, this paper uses econometric methods to test the stationarity, cointegration and Granger causality among related variables. The second-order difference of natural logarithm of each variable is stable, which shows that there is an economic relationship between them in a certain period of time. Granger causality test is carried out on the second-order difference of natural logarithm of variables, and the following conclusions are drawn:
(1) conclusion
In the short run, the valuation of RMB against USD has a significant impact on foreign exchange reserves, but in the long run, the increase of foreign exchange reserves is not the reason for the appreciation of RMB against USD. China's huge foreign exchange reserves accumulated over the past 30 years of reform and opening up are more due to non-exchange rate factors.
In the short term, there is no Granger causality between cash in circulation and the exchange rate of RMB against the US dollar, but narrow money supply will cause changes in the exchange rate of RMB against the US dollar, and RMB against the US dollar will cause changes in the broad money supply. With the improvement of China's exchange rate mechanism, the foreign exchange market is effective as the transmission channel of China's monetary policy, but it must also be recognized that this efficiency is limited in the long run.
There is no Granger causality between money supply and foreign exchange reserves, and China's monetary policy will not have a direct impact on foreign exchange reserves, which makes China's monetary policy ineffective in a certain range of open market business, which also reasonably explains that the People's Bank of China issued a large number of base money to absorb foreign exchange, but the result was to inject too much liquidity into the real economy and virtual economy.
(2) Countermeasures
Based on the above analysis, this paper puts forward the following countermeasures:
In the short term, the monetary authorities or government agencies can influence the exchange rate by adjusting the money supply, and promote the internal and external balance of China's economy and the healthy development of the foreign exchange market; But in the long run, these policies will be ineffective in a certain range, and the result will not only raise the expected level of inflation, but also put macro-control into trouble. Therefore, we should moderately control the broad money supply to keep it at a stable growth level; The real exchange rate of RMB against the US dollar should ultimately be determined by the differences in price level, economic growth rate and interest rate between the two countries, so as to reduce human intervention.