| Intertemporal substitution and risk aversion affect resource allocation decisions and determine asset prices through intertemporal asset allocation and risk premiums.If a consumer defers "today" consumption until "tomorrow",the losing utility of "today"requires certain compensation,and vice versa;risk aversion is reflected in the certain compensation when taking the risk of future uncertainty.These two aspects of compensation characterized by elasticity of intertemporal substitution and risk aversion coefficients respectively are important part of asset prices.By definition,elasticity of intertemporal substitution represents the consumption change caused by the change in marginal utility of intertemporal consumption;risk aversion coefficient represents the risk compensation by taking the uncertainty of hedging consumption.Intertemporal substitution and risk aversion jointly determine asset prices by guiding the savings and investment behavior through intertemporal asset allocation motivation and risk management motivation.Since elasticity of intertemporal substitution and risk aversion coefficients are so important in asset price decisions,accurately measuring and setting parameters should be a basic premise.However,accurately assigning these two parameters is a tricky problem.In the asset pricing model,the values of these two parameters are often adjusted to match the data.The values of these two parameters also have large differences.A common problem in the theoretical model is that if the definition of elasticity of intertemporal substitution and risk aversion coefficient is met,it is possible that the real data cannot be fitted,and the real asset price data is fitted,and unreasonable elasticity of intertemporal substitution and risk aversion coefficient are obtained.In particular,in order to match asset price historical data,these two parameters are often set to be more extreme in the asset pricing model.Although this approach allows the theoretical model to interpret the data,it also blurs the economic implications of the elasticity of intertemporal substitution and the risk aversion coefficient itself.In view of this,this article has carried out three aspects of work:Frist,this paper derive Euler equations from four different utility functions(CRRA utility,recursive preferences utility,habit formation utility,and labor indivisibility utility).The author also estimate the values of EIS by using data from China Family Panel Studies(CFPS)and China Household Financial Survey(CHFS),and study factors which changes EIS from three perspectives.Next,we consider the factors affects EIS in macroeconomics by using Chinese macroeconomic data and global cross country data.Second,this paper introduce monetary holdings into utility function to measure the risk aversion.Based on this model,the author define the absolute risk aversion,the relative risk aversion,and the consumption-labor relative risk aversion(sum of consumption and human capital as wealth),consumption-currency risk aversion(sum of consumption and holding currency as wealth),and consumption-labor-currency risk aversion(sum of consumption,human capital and holding currency are considered as wealth).This paper then calculates the values for risk aversion at both micro and macro level by using China Household Financial Survey data and data from 87 countries during 1950 to 2017.At last,this paper explores factors which cause changes in the risk aversion.Third,this paper is based on the Consumption-based Capital Asset Pricing Model(CCAPM)model,the Long Run Risk Model,and the Dynamic Stochastic General Equilibrium introducing long-term risk,testing the elasticity of intertemporal substitution and relative risk aversion coefficient measured by micro data.The author discusses the conditions under which the economic significance and fitting historical data of these two parameters can be taken into account.This paper finds that:First,the results show that the mean of EIS using CFPS data is between 0.0806-0.0879,compared to 0.1063-0.1270 using CHFS data.These estimates mean that one unit of future marginal utility is higher than the current marginal utility,families will prefer future consumption,and the total consumption growth rate(1+growth rate)will increase by an average of 0.08%and 0.11%.This paper also finds that wealth positive,age positive,and health negative impact the elasticity of intertemporal substitution.People richer,older,and in poorer physical condition are more likely choose "future consumption" due to lots of factors and the relativity of current marginal utility to future marginal utility.Second,the mode of relative risk aversion is 2,the average is close to 2.More than 70%of values during our study are between 2.0000-2.0001,which means that one unit of variance value is required for wealth growth as a risk premium or risk compensation when one's wealth changes with 1%standard deviation.These results are also consistent with our calculations using cross-countries macroeconomic data.The author also find that risk attitudes are affected by income and age,as the degree of risk aversion rises with increase in both income and age by using micro and macro data.Third,the author find that if the elasticity of intertemporal substitution and the relative risk aversion coefficient are calibrated to the values measured by China's micro-data,the risk premium calculated by the CCAPM model is smaller than the historical real risk premium.And the long-term risk model and the DSGE(introducing the long-term risk)can simulate stock returns and bond yield historical data,but this requires the theoretical model to set a higher risk size.The numerical simulation of the DSGE model introducing long-term risks also found that long-term technical shocks will have long-term trend effects on asset prices,while short-term shocks will only increase asset price volatility.The theoretical significance of this paper is the author provide reference for parameter setting of the elasticity of intertemporal substitution and risk aversion in theoretical model by following normal assumption and standard asset price model setting.Specifically:First,the elasticity of intertemporal substitution and risk aversion coefficient based on Chinese micro-subjects are measured,which can be used as a reference for the theoretical model parameter setting.In the theoretical literature,the parameters for elasticity of intertemporal substitution and risk aversion coefficient are mostly derived from empirical evidence from developed countries,while we measure these two parameters for Chinese microeconomic entities and provide empirical evidence for the value of Chinese theoretical model parameters.Second,we have taken into account both the economic meaning of the parameters and simulation of historical data.Based on microscopic data,this paper measures China's elasticity of intertemporal substitution and relative risk aversion coefficient,and discusses a trade-off of the theoretical model:either set a larger risk depending on the economic significance of the parameter,or ignore the economic significance of the parameter to match the history data.This paper attempts to discuss the asset price decision based on the full consideration of the economic significance of these two parameters.This is an attempt to balance the economic meaning of the parameters with the historical data of the asset pricing model.The practical significance of this paper is:First,intertemporal substitution and risk aversion vary from person to person.Under the general trend of China's population change,we can capture and predict future asset price trends from the perspective of intertemporal substitution and risk attitude.Second,intertemporal substitution and risk attitudes affect asset portfolio size and portfolio selection respectively.From the perspective of "people",it can help us understand China's macro capital market size and capital market structure. |
PDF Full Text Request