Empirical Studies On Business Cycle And Output Fluctuation In China

During the past several decades since the founding of China, the economy in our country has experienced nine cyclical fluctuations or business cycles. The economic growth is characterized by "big up and big down". Higher volatility will bring more risk in economic growth, and makes it difficult to constitute the optimum economic policies. Along with the deepening of the reform and opening to the outside world, Chinese economy becomes mightiness. Macroeconomic control policies carried into execution from 1980s prevented the large fluctuation in economic development, and smoothed the output volatility. The success of economic "soft land"is middle of 1990s proved the validity of these policies. Since the 1990s, the long wave situation of the business cycle is gradually emerging out and the fluctuation amplitude become smaller. In this thesis, we take some empirical analysis on the real output fluctuation in business cycle in China using quantitative methods, and test the asymmetry in business cycle. In detail, we mostly research on the aggregate variable of GDP. 1 Measuring the Output Volatility in Business Cycle In the measuring the change of the real output volatility, We use a markov regime switching model to estimate the variances of the output growth. The result reveals that the standard deviation of output in low volatility regime is 0.017, and 0.068 in the high volatility regime. In the past decade, the real output volatility show an obvious decline trend. So we measure the output volatility in two methods. We calculate the rolling standard deviation of the growth rate of real output, GDP, which is plotted in figure 1. From the rolling standard deviation curve, we find the deviation ascend after 1994 and reach the culmination in 1996, after this, it descend rapidly until the end of 1999. This volatility path indicates two things. First, the economic growth underwent a high fluctuation period. Second, in recent ten years, the output volatility descended largely. Figure 1 The rolling standard deviation of GDP growth We measure the volatility of real output, GDP, using a markov variance switching model and rolling standard deviation. We find a change mode that there is a large decline in output volatility, which is consistent with many researchers. This fact testify that the foregone fluctuation mode of economic cycles character with "big up to big down"has replaced by the new mode of steady growth. 2 The Reasons for the Changes in Patterns of Output Volatility In order to find out the cause of the decline in the output volatility, we first tested the correlation between the output volatility and inflation and monetary growth. The results suggest that the change of the price level and monetary supply have influence on the output volatility. The decline in the inflation volatility and the steady monetary policy may be help to reduce the output volatility, but the reasons are not enough. Output is an aggregate variable. Take the standard decomposition of GDP by type of purchase: consumption, investment, government spending and net exports. For each component we consider two measures of volatility. The first is the same as for the real GDP former, namely, the rolling standard deviation of these components growth rate. The second measures the volatility of avariable commonly called the "growth contribution"of each component, which adjusts for the share of the component in GDP. In the components analyses of output volatility, we find the change mode of consumption is very different from the real GDP. Before 2002, the consumption volatility contributes to the stabilization of GDP. But, henceforth, the action of consumption is weak. However, the descent in volatility of the investment contributes during all over the sample period. As a conclusion, the investment motion takes an most important role in declining the volatility of GDP. Furthermore, the net export motion is a reason that can't be ignored. According to the measure of the GDP components'volatilities above, especially the disaggregated look on the investment, we find the cause of the decline in the output volatility. The descends in the volatilities of investment, government spending and net export make up of the main reasons for the change in economic fluctuation mode. 3 Asymmetry in Chinese Economy Non-linearity time series models have been developing rapidly in the past 20 years. In the literatures about business cycle and output fluctuation, non-linearity time series models play an important role in testing the asymmetry in business cycle. Hamilton(1989) captured asymmetry and non-linearity modes in the US real GNP growth by an markov regime switching model, which is a very typical non-linearity time series model. Hamilton's research indicated that there were difference in duration and volatility between contraction and expansion. In addition, concerning the asymmetry in business cycle, Friedman advanced a famous "pluck"theory. He suggested that amplitude of cyclical contractions tended to be strongly correlated with amplitude of subsequent contractions, thus producing an asymmetry between succeeding phases of the business cycle. There is an upper limit, or ceiling, to real output, and the "strings"(the ceiling) may be plucked downwards at irregular intervals to represent recessionary shocks. The size of these shocks will vary across recession, but output will always rebound to the ceiling level, so that recessions can only have temporary effects on real output and subsequent...