Testing For The Validity And The Asymmetry Of Monetary Policy In China

As relationship between monetary variables and the actual output variables are not only related to the classical dichotomy, but also come down to the effect mechanism between the nominal economic and the real economy, as well as the effectiveness of monetary policy and other important issues of economic theory, the research on the relationship between two variables has been the core issue of modern macroeconomic theory and empirical analysis (Walsh, 1999). If there doesn't exist the stable effect mechanism or transmission mechanism between actual output variables and monetary variables, the monetary variables have no significant impact on real output variables, that may appears of monetary variables neutral and monetary policy neutral, resulting in the classical dichotomy. The monetary variables are only the symbols in the economy operation, the impact of monetary variables will lead to other nominal variables (prices, the nominal wage and nominal interest rates, etc.) changes in proportion, the real economy operation will not be affected by monetary shocks. Under long-run monetary superneutrality (LRSN), a permanent increase in the growth rate of the money stock has no real effects—apart from real balances—in the long run. as in the classical Sidrauski (1967) models, a permanent increase in the growth rate of the money stock leads to an equal increase in the long-run inflation rate, but leaves the long-run real interest rate, capital stock, and real output level unchanged. While there is general agreement that permanent changes in inflation arise solely from equal permanent changes in money growth. If permanent changes in money growth and inflation can alter the steady state values of the real interest rate, capital stock, and real output level, money is not superneutral in the long run.If the monetary policy is effective, people thought that tight monetary policy has reduced economic growth, and easy monetary policy has increased economic growth. However, if use some sort of quantitative indicators to measure monetary policy, the same magnitude of the monetary contraction and monetary expansion has different effects on economic growth. It's generally believed that the effect on the economy of easy monetary policy is less than the tight monetary policy, that's the monetary policy is non-symmetry.In this paper, we use a variety of research methods, analyse on the effectiveness and the non-symmetry of monetary policy. And the conclusions are following:First, we use the modified ICSS algorithm, as well as structural VAR model to measure the inflation rate, nominal interest rates and the actual output level, to testing the existence of long-run monetary superneutrality or Mundell-Tobin effect.We first use modified ICSS algorithm of Inclan and Tiao (1994) the to search for the structural changes of time series that may exist, the results showed that the inflation rate sequence of China has significant change. By the test results, from 1983 to 1995, the dynamic path of inflation rate of China showed characteristic of severe ups and downs. Since the first quarter of 1996, the inflation rate tends to smooth trajectory significantly, is consistent with the successful implement "soft landing" of economic of China. We divided inflation rate, nominal interest rates and the actual output level into two sub-samples. In addition, we use QMLE estimation methods and the GARCH (1,1) model parameter estimation results of the two sub-samples, proved once again that volatility of inflation rate has showed significant differences.In the process of testing structure VAR model, we first test the unit root of the inflation rate, nominal interest rates and the actual GDP series, the test results show that these three variables were significant in the standard level. In the test of trends ingredients of inflation rate, nominal interest rates and the actual GDP, we found the structural changes occurred after the 2rd sub-sample data, the nominal interest rate and real GDP components include the trend ingredients. Finally, we estimate the long-term response of the structure of three-variable VAR model including the inflation rate, nominal interest rates and the actual output level, we found the adjustment of nominal interest rate corresponds to a lasting change of the rate of inflation is less than 1/2. The long-term response of actual output to the rate of inflation is estimated positive. These results indicate that the positive impact of inflation rate on long-term real interest rates can have a negative response, while the long-term actual output level has a positive effects. This shows that long-run monetary superneutrality in our country doesn't exist.Secondly, we use the rate of inflation, nominal interest rates and money supply growth rate as the explanatory variable, and the actual GDP growth as explained variables. By constructing bivariate structural changes model, as well as multi-variable model, we found that both the bivariate structural changes model or multi-variable model show that inflation rate, nominal interest rates and money supply growth rate have multiple structural changes, the total sample interval is divided into a number of sub-interval. By bivariate and multi-variable model of structural changes that we can clearly see the results: the rate of inflation, nominal interest rates and money supply growth rate have different impacts on actual output level in different sub-samples within the same interval, the direction and size have taken place in different degrees of change. This is a strong evidence to the non-symmetry effect of inflation rate, nominal interest rates and money supply growth rate and other monetary variables on the real output.The estimated results of bivariate structural change model show that the positive money supply growth have driven actual output, but the effect is unstable. If just use simple money supply growth to promote positive output growth, it is will give rise to a high degree of monetary expansion in turn can lead to severe inflation. Further, the inflation rate, nominal interest rates and money supply growth rate affect to the real output levels through the combined effect of the cumulative positive and negative. Therefore, the monetary policy in China should not only focus on the sustained and stable growth of output, but also consider the non-symmetric relations of monetary variables affect the output.At last, this paper uses Leamer's sensitivity test in a VAR framework and examines the robustness of the relationship between different monetary and output variables through the different. regression equation of monetary and output variables based on our real GDP, consumption, investment, M0, M1 and real interest rates from 1991Q4 to 2008Q2 of the quarterly data. We get the conclusion that consumption can be Granger caused by M0 robustly, and our real GDP, consumption, and investment are Granger caused by M0,M1 and real interest rates fragilely.