Looking back over the course of more than one hundred years of modern economic development, asset price bubbles have periodically emerged, triggering systemic financial risks, leading to financial crises and economic recession. The United States subprime mortgage crisis sweeping the globe in 2008 corroborates this fact once again as the global financial crisis triggered by the over-bubble in the price of financial derivatives. In recent years, China has suffered from excessive stock market catastrophe and excessive real estate prices. The asset price bubble has had a considerable impact on the development of China’s real economy and the stability of the financial system. Holding the bottom line where systemic financial risks do not occur has become an important macro-stability in our country. Preventing systemic financial risk around the asset price bubble has become a hot issue at present, and it is important to further clarify the formation mechanism of the asset price bubble and its possible economic effects. Based on this background, this paper focus on the definition of asset price bubble, the recognition and measure of bubble, the formation mechanism and possible economic effect of asset price bubble, aiming at revealing the internal formation mechanism of asset price bubble and its possible economic effect, providing theoretical support and practical guidance for the orderly market of asset prices and the inherent stability of the financial system.
First, after the bubble definition and the analysis of the characteristics of the bubbles, this paper identifies and measures the asset price bubble. Based on the explosive character of bubbles, we use the GSADF method to identify the occurrence and rupture of bubbles in the stock price and housing price in China. Based on the definition of bubble as the part of asset price deviating from its fundamental value, we first estimate the fundamental value of stock price and house price, then separate the stock price bubble and housing price bubble from their real price, respectively. For the stock price bubble, the appropriate macro-variables including industrial added value, consumer price index and inter-bank lending rate are selected to measure the fundamentals, and the price bubbles are extracted using the error correction model. Furthermore, the Markov switching model is used to divide the stock price bubbles. It is found that the two longest survival periods of the stock price bubble are October 2006-November 2007 and May 2014-June 2015 respectively. For the housing price bubble, variable selection and iterative estimation, which is use to eliminate endogenous variable, are applied to separate the fundamental value and bubbles in the housing price. It was found that the housing price bubble in China already exists, and has formed a certain scale. The housing price bubble increased significantly in July 2005 and early 2009.
Second, from the behavioral finance perspective, we analyze the mechanisms of investors’ expectations of future market returns, the investors’ confidence, emotion and behavior on the formation of asset price bubbles. (1) The noise trader bubble model under heterogeneous expectation is established, and the asset price bubble, which is caused by the irrational behavior of two types of noise traders: information digger and momentum trader, is deduced. By numerical simulation, it is found that information digger and momentum traders influence the bubble evolution through price expectation deviation and historical price changes respectively, and the direction and degree of influence are closely related to their emotions. (2) Based on the measurement of investor sentiment and herding behavior, we use TVP-SV-SVAR model to explore the time-varying effect between investor sentiment, herding behavior and stock price bubble. The results show that, the lagged effect of investor sentiment on herding behavior is more long in the period of stock price fluctuating; investor sentiment can stimulate the bubble inflated in the current time, and its current stimulus has been reduced over time; the impact of herding behavior on stock price bubble is uncertainty, the direction of the impact change from negative to negative in the initial stage of rising stock prices, and from negative to positive in the period of price peak or the stock price decline. (3) The positive effects of overconfidence and market liquidity on stock market speculative bubble are revealed theoretically. Using the time-varying transition probability Markov Regime Switching (MS-TVTP) model, we find that investors’ overconfidence increase the probability of stock market bubble transforming from dormant to explosive regime, and the negative market liquidity increase the probability of stock market bubble collapsing, both of which are consistent with the theoretic analysis.
Third, this paper analyzes the mechanism of the economic factors of the lever, economic growth and monetary policy on the formation of asset price bubble. (1) Based on the construction of asset price bubbles model in which financial leverage is considered, the theoretical mechanism and asymmetric effect of leverage on the asset price bubble are analyzed from by using the quantile method. The results show that the effect of leverage on the stock price bubble and the housing price bubble varies with the stage of economic development, the degree of bubble evolution and the level of leverage. (2) Considering the cyclicality of economic growth, frequency-domain causality test and wavelet decomposition methods are used to explore the effects of of economic growth on asset bubbles price in frequency-domain. The study finds that the period in which economic growth has a significant Granger effect on the stock price bubble is about six months to a year, while the economic growth has obvious Granger effect on housing bubble in the period of 3-9 months and more than two years. (3) Based on the economic theory of monetary policy and asset price bubble, this paper uses threshold structure VAR model to analyze the effect of monetary policy shock on asset price bubble under different monetary policy conditions. We show that the impacts of interest rate shocks and M2 growth rate shocks on asset price bubbles depend on whether the monetary policy is easy or tight. Furthermore, the effects of positive and negative shocks have obvious asymmetry.
Fourth, considering government tax and financial frictional, the endogenous growth models with and without asset price bubbles are established, and the impact of asset price bubble on the economic growth rate is analyzed. Our findings question the traditional view that asset price bubble detrimentally affects the economy, we find the asset price bubble may harm or promote economic growth. Based on the above results, we analyze the effect of stock price bubble and housing price bubble on economic growth using time-varying VAR. The result shows that the bubble has uncertain effect on the of economic growth. Furthermore, the influence of asset price bubble on the representative index of economic welfare are studied by quantile method. We found that, under extreme conditions, the bubble has good forecasting effect on economic welfare.
This study expands the theoretical modeling and empirical analysis of asset price bubbles and forms a more systematic and complete analysis framework of the formation mechanism of asset price bubbles and their economic effects. According to the conclusion of our study, this paper puts forward the policy of preventing and coping with the asset price bubble from many aspects, such as paying attention to investor’s psychology and behavior, focusing on liquidity warning, maintaining moderate leverage, understanding the bubble effect of monetary policy and considering the economic effect of asset price bubble.