Empirical Study On The Monetarism Model Of RMB Exchange Rate Determination

This paper investigates RMB exchange rate of China with methods of combining theoretics with practice, empirical analysis and canonical analysis, quantitative analysis and qualitative analysis.In the first section, we summarize recent development of the homogeneous literatures, and make comparison simply.In section 2nd, we briefly introduce the monetarism model of RMB exchange rate determination, including the flexible price monetary model and sticky price monetary model on the basis of the theory of purchasing power parity (PPP). Also some deduction and comparison are present.Econometric methods using in this paper are shown in 3rd section. In this section, we introduce two unit root tests of ADF and PP, and introduce two methods of cointegration in detail, involving Engle-Granger's (1987) test and Johansen's (1991) for cointegration. In the method of Johansen's test, a method of vector error correction model is present.Section 4th is the empirical part in this paper. Using the monthly data of China-America from January, 1990 to December, 2005, we study on the hypothesis of PPP and the monetarism model exchange rate determination in detail. Through this empirical analysis, some evidence can be drawn:1. The hypothesis of PPP for RMB holds in the long run. With an Engle-Granger cointegration test for residuals of PPP equation, we find that the null is significant rejected by unit-root test, so this suggests that there is a mean reversion process in residuals and more shows that long-run PPP holds.2. The flexible price monetary model is appropriate for RMB exchange rate determination. By analysis of the relationship among China-America relative money supply, relative real income (or output), relative interest rate level and nominal exchange rate, and with Engle-Granger cointegration test and Johansen cointegration test, the results show that the flexible price monetary model is appropriate, while the sticky price monetary model not.3. There are different postures of adjustment between the exchange rate and several fundamentals. We find that the (i) change rate of exchange rate, the growth rate of relative money supply, the growth rate of relative output and the change of relative interest rate appears to negative, positive, negative and negative adjustment, respectively.4. There is bidirectional relationship between the change rate of RMB exchange rate and the growth rate of relative money supply, and the growth rate of relative money supply is affected by itself in lags. Using the Granger causality test and impulse response function, we find that the change rate of RMB exchange rate responds positively to the growth rate of money supply, while the former affects positively weak to the latter; the growth rate of relative money supply is affected positively by itself in lag, while the response of the growth rate of relative output to this is obvious in short-term but not in long run; the change of relative interest rate doesn't affect the change of others all along in all dynamics of causalities.5. RMB exchange rate has undergone two periods of serious undervaluation (1992 to 1994 and 2001 to 2003) and one period of serious overvaluation (1994 to 1999). During the period from 1994 to 2003, RMB exchange rate was always lying in appreciation.