Charcterizing Nonlinearities Andpasesnusnessyce: Evidence From China

Business cycle is the law which macroeconomic operations are bound to obey and the phenomenonthat will happen inevitably in reality. Via jointly implementing fiscal and monetary policies inresponse to different stages of business cycles, policy makers realize their goals of curbing extremebusiness fluctuations and controlling the macro economy. Since China decided to adopt the Reformand Opening up policy 30 years ago, the average annual economic growth has been kept over 10percent for a certain period of time. In this process, China has gained enormous wealth andexperience, and has finally become the world's second-largest economy. The rapid economicgrowth along with sharp economic structural change imply that economic fluctuations may not inaccordance with classical business cycle theory, thereafter dating of China's economic cycle hasalways been under the hot debate in theoretical research. Thus, using modern economic businesscycle theory a priori combined with nonlinear econometrics techniques, we endeavored tocharacterize the nonlinearity in Chinese business cycle and measure the volatility patternsystematically.Before we formally address our empirical research, a brief review is given in Chapter 2 on thebusiness cycle theory from neoclassical macroeconomics and new Keynesian economicsperspective. Based on different empirical methodologies on dating the chronology of business cycle,Chapter 3 firstly uses Bayesian nonlinear modeling strategy, ad hoc, Bayesian thresholdautoregressive model to analyze the properties of economic cycle starting from 1992 with respectto China's real GDP growth. By employing the posterior distribution for threshold value, thecharacteristics of China's economic cycle were measured and analyzed, with specific attention onthe phases of contraction. In order to obtain a more accurate measure for China's business cycleand double check the results from TAR model, Chapter 4 uses the wavelet analysis to date China'seconomic cycle based on the level of real GDP, and then compare the result with previous chapter.It is found that the contraction phases of business cycle determined by time domain techniques vis-à-vis that detected by wavelet method are roughly the same. This shows that the division foreconomic recession in our work is relatively accurate and can undergo the examination of empiricalanalysis which shows that our work is empirically robust.From the empirical results, we believe that Chinese economic has experienced the followingthree contraction phases: the first period begins from the third quarter of 1997 to third quarter of2001, the second period starts from the second quarter of 2008 to the fourth quarter of 2009, andthe last begins after the third quarter of 2011 to present. This provides a necessary empiricalevidence for the nonlinear characteristics of China's Economic Fluctuations.However, the characteristics of China's business cycle cannot be confined to just study thechronology of the economic cycle, other issues like causes for the asymmetry cyclical fluctuationsstill require particular attention. As it is mentioned at the beginning, economic policies can haveimportant implications for economic fluctuations. How to make use of the macro-economic controlpolicies to stabilize the fluctuations of the economic cycle as well as to avoid the risksacerbated ?by economic contraction is critical to the stability of economic growth.Based on the above considerations, chapter 5 studies the relationships between the fluctuationsof business cycles and two aspects of fiscal policy and monetary policy, respectively. As for fiscalpolicy, the paper analyzes the effect of the nonlinear impact of fiscal policy on different phases ofthe business cycle. The results showed that the fiscal revenue and expenditure has significantasymmetric response to different levels of economic expansion and contraction. As for monetarypolicy, this paper focuses on the correlation between execution cycle of monetary policy andbusiness cycle. Our study finds that three recession phases of business cycle have occurred withinsix months after the end of"tightening-up"monetary policy. In the meantime, this paper adopts thediscriminant analysis and logistic regression and then shows M1 Y-o-Y growth and short-termloans between financial institutions, among the indicators of monetary policy implementation tools,are able to predict economic contraction after tightened monetary policy. Therefore, we need to payspecial attention to the economic situations when the short-term deposit is more than 164.824billion RMB or the M1 Y-o-Y growth is less than 15.16%.As it known to all, term spread is an important indicator for the liquidity in financial marketand several scholars have further pointed out that it can play an important role in predictingeconomic contractions. While the volatility pattern of business cycle changes, there are some potential structural breaks in the evolution path of economic growth. By taking the nonlinearity ofbusiness cycle in China into account, chapter 6 employs the nonlinear structural break frameworkand a real-time model choice strategy to study the nonlinear relationship between spreads andbusiness cycle fluctuations. For the sample from 1992 to 2011, we find that SBVAR model is theoptimal choice in the SBTVAR framework adopted in our paper, and it singles out a structuralchange point which occurred in the fourth quarter of 2009; from this moment, the economic growthrate decreased to some extent.. In the previous study, we have concluded that the contraction phaseof business cycle occurred during the fourth quarter of 2008 and the second quarter of 2009. Thisshows that under current economic situations, the model can detect the structure break points in theprocess of China's economic growth, and this result matches well with the fact that China's newround of economic restructuring begun in the fourth quarter of 2009. An indirect policyimplications implied from this result is that if proper macroeconomic regulations are executed,China's economy is expected to successfully achieve"soft landing"in 2012 again.In addition, this study found SBTVAR model family can better capture the dynamicrelationship between real GDP and spreads changes. A special issue regarding SBTVAR modelingneeds to be addressed here is that the choice of the optimal model from SBTVAR model familymay vary as new information is added into our information set. Thus, in the process of continuousevaluation for new information, we should also dynamically select a suitable model from thenonlinear model family so as to track business cycle fluctuations.Since China is still in the primary stage of market economy, not only the restructure of theeconomy but also the construction of market economies need continuous development and constantimprovement So the studies on the nonlinear characteristics of China's business cycle and businesscycle determination are bound to be a difficult and ongoing task. We will unremittingly endeavor todelve into the new problems arising, and thus provide valuable judgments and suggestions tosmooth and sustain economic growth of China.