货币政策的宏观经济效应——基于时变参数动态因子模型的分析
|
摘要
通过构建中国宏观经济月度数据集,运用时变参数动态因子(TVP-FAVAR)模型,考察了产出和价格部门分量的波动特征及其在不同外生冲击下的脉冲响应,探讨了主要宏观经济变量的动态波动及货币政策冲击对变量脉冲响应的时变特征。实证结果表明:产出和价格月度部门分量的波动主要源自特质成分冲击,其可分别解释产出和价格分量总波动的85.4%和57.3%;新时期货币政策规则转变的效果显著,短期利率冲击对经济变量的影响效应明显增强。
Through the construction of the set of China's macroeconomic monthly data,we applied the time-varying parameter dynamic factor(TVP-FAVAR) model to document the fluctuation characteristics of disaggregated output and prices,as well as the implulse response with different exogenous shocks. Here we aimed to analyze the time-varying characteristics of the variable impulse response with the dynamic fluctuation of macroeconomic variables and the shocks of monetary policies. We find that,the shock of trait components has mainly made the fluctuation of the monthly disaggregated price and output,which could explain the total fluctuation of output and price by 85.4% and 57.3%. Moreover,the change of monetary policy in new era has worked well,and the shock of shortterm interest rate has affected economic variables more significantly.
Through the construction of the set of China's macroeconomic monthly data,we applied the time-varying parameter dynamic factor(TVP-FAVAR) model to document the fluctuation characteristics of disaggregated output and prices,as well as the implulse response with different exogenous shocks. Here we aimed to analyze the time-varying characteristics of the variable impulse response with the dynamic fluctuation of macroeconomic variables and the shocks of monetary policies. We find that,the shock of trait components has mainly made the fluctuation of the monthly disaggregated price and output,which could explain the total fluctuation of output and price by 85.4% and 57.3%. Moreover,the change of monetary policy in new era has worked well,and the shock of shortterm interest rate has affected economic variables more significantly.
引文
[1]Zhang Wenlang.China’s Monetary Policy:Quantity Versus Price Rules[J].Journal of Macroeconomics,2009,31:473-484.
[2]胡志鹏.中国货币政策的价格型调控条件是否成熟?——基于动态随机一般均衡模型的理论与实证分析[J].经济研究,2012(6):60-72.
[3]马鑫媛,赵天奕.非正规金融与正规金融双重结构下货币政策工具比较研究[J].金融研究,2016(2):137-144.
[4]陈彦斌,郭豫媚,陈伟泽.2008年金融危机后中国货币数量论失效研究[J].经济研究,2015(4):21-35.
[5]Fan L,Yu Y,Zhang C.An Empirical Evaluation of China's Monetary Policies[J].Journal of Macroeconomics,2011,33:358-371.
[6]Ma Y.Monetary Policy Based on Nonlinear Quantity Rule:Evidence from China[J].International Review of Economics and Finance,2014,34:89-104.
[7]李成,马文涛,王彬.通货膨胀预期、货币政策工具选择与宏观经济稳定[J].经济学(季刊),2010,10(1):51-82.
[8]林仁文,杨熠.中国市场化改革与货币政策有效性演变——基于DSGE的模型分析[J].管理世界,2014(6):39-52.
[9]伍戈,连飞.中国货币政策转型研究:基于数量与价格混合规则的探索[J].世界经济,2016(3):3-25.
[10]Sun L,Ford J L,Dickinson D G.Bank Loans and the Effects of Monetary Policy in China:VAR/VECM Approach[J].China Economic Review,2010,21(1):65-97.
[11]Goodfriend M,Prasad E.A Framework for Independent Monetary Policy in China[J].CESifo Economic Studies,2007,53(1):2-41.
[12]王君斌,郭新强,王宇.中国货币政策的工具选取、宏观效应与规则设计[J].金融研究,2013(8):1-15.
[13]陈浪南,田磊.基于政策工具视角的我国货币政策冲击效应研究[J].经济学(季刊),2014,14(1):285-304.
[14]Sims C.Interpreting the Macroeconomic Time Series Facts[J].European Economic Review,1992,36(5):975-1011.
[15]Bernanke Ben S,Boivin Jean,Eliasz Piotr.Measuring the Effects of Monetary Policy:A Factor-Augmented Vector Autoregressive(FAVAR)Approach[J].Quarterly Journal of Economics,2005,120(1):387-422.
[16]Primiceri G E.Time Varying Structural Vector Autoregressions and Monetary Policy[J].Review of Economic Studies,2005,72(3):821-852.
[17]Gali J,Gambetti L.On the Sources of the Great Moderation[J].American Economic Journal:Macroeconomics,2009,1(1):26-57.
[18]Bernanke B,Boivin J.Monetary Policy in a Data-Rich Environment[J].Journal of Monetary Economics,2003,50:525-546.
[19]He Q,Leung P H,Chong T L.Factor-Augmented VAR Analysis of the Monetary Policy in China[J].China Economic Review,2013,25:88-104.
[20]Fernald J G,Spiegel M M,Swanson E T.Monetary Policy Effectiveness in China:Evidence from a FAVAR Model[J].Journal of International Money and Finance,2014,49:83-103.
[21]欧阳志刚,薛龙.新常态下多种货币政策工具对特征企业的定向调节效应[J].管理世界,2017(2):53-65.
[22]王少平,朱满洲,胡朔商.中国CPI的宏观成分与宏观冲击[J].经济研究,2012(12):29-42.
[23]杜海韬,邓翔.部门价格动态、特质冲击与货币政策——基于结构动态因子方法[J].经济研究,2013(12):93-105.
[24]蔡晓陈,蒋涛.特有冲击、共同冲击与通货膨胀持续性[J].统计研究,2014,31(5):48-53.
[25]Korobilis D.Assessing the Transmission of Monetary Policy Using Time-Varying Parameter Dynamic Factor Models[J].Oxford Bulletin of Economics and Statistics,2013,75(2):157-179.
[26]张龙,金春雨.数量型和价格型货币政策工具的有效性对比研究[J].中国工业经济,2018(1):119-136.
[27]Doan T,Litterman R,Sims C A.Forecasting and Conditional Projection Using Realistic Priordistributions[J].Econometric Reviews,1986,3:1-100.
[28]Kim S,Shephard N,Chib S.Stochastic Volatility:Likelihood Inference and Comparison with ARCH Models[J].Review of Economic Studies,1998,65:361-393.
[29]Gerlach R,Carter C,Kohn R.Efficient Bayesian Inference in Dynamic Mixture Models[J].Journal of the American Statistical Association,2000,95:819-828.
[30]Boivin J,Giannoni Marc P,Mihov Ilian.Sticky Prices and Monetary Policy:Evidence from Disaggregated US Data[J].American Economic Review,2009,99(1):350-384.
[31]Bils Mark,Klenow Peter J.Some Evidence on the Importance of Sticky Prices[J].Journal of Political Economy,2004,112(5):947-985.
(1)限于篇幅,本文未给出新凯恩斯DSGE模型的具体推导过程,有需要者可向作者索取。
(2)产出缺口(yt=yt-yn t)被定义为实际产出(yt)与中央政府制定的当期GDP增长目标(yn t)之间的缺口。
(3)若允许扰动项之间存在弱的相关性,则模型的动态性会更强,但这会使估计参数急剧增加,可能导致模型估计有偏。
(4)明尼苏达规则为:当Vij=1/c2时,参数自身滞后;当Vij=0.001s2i/c2s2j时,其他变量参数滞后且i≠j,logc=1,…,p。其中,s2i是p阶滞后对因变量i回归方差的残差,i=1,…,M,j=1,…,MP。
(5)两步主成分分析的具体方法是,首先从Xt中提取出K×M个主成分并估计出共同成分Ct,得到F′t对应的由C′t生成的空间部分,然后将F′t代入式(3)的估计模型。此外,给模型施加Λf′Λf/N=I或F′F/T=I约束,可以得到相同的共同成分FΛf′和因子空间,我们设定F′=(TZ)1/2,其中,Z为XX′的K个最大特征值对应的特征向量。
(6)由于检验结果数据庞大,此处未列出详细结果,有需要者可向作者索取。
(7)全部CPI部门分量是指包括居民消费者价格指数、城市居民消费者价格指数、农村居民消费者价格指数和居民消费价格指数(36城)的全部分量价格序列。
(8)Primiceri(2005)[16]的预置样本先验是将总体样本的前若干期作为训练样本,利用普通最小二乘回归(OLS)得到先验均值和方差,并将其作为待估样本的均值和方差代入模型进行估计。在稳健性检验中,我们使用60期预置样本先验。
(9)限于篇幅,本文未给出具体的稳健性检验结果,有需要者可向作者索取。
[2]胡志鹏.中国货币政策的价格型调控条件是否成熟?——基于动态随机一般均衡模型的理论与实证分析[J].经济研究,2012(6):60-72.
[3]马鑫媛,赵天奕.非正规金融与正规金融双重结构下货币政策工具比较研究[J].金融研究,2016(2):137-144.
[4]陈彦斌,郭豫媚,陈伟泽.2008年金融危机后中国货币数量论失效研究[J].经济研究,2015(4):21-35.
[5]Fan L,Yu Y,Zhang C.An Empirical Evaluation of China's Monetary Policies[J].Journal of Macroeconomics,2011,33:358-371.
[6]Ma Y.Monetary Policy Based on Nonlinear Quantity Rule:Evidence from China[J].International Review of Economics and Finance,2014,34:89-104.
[7]李成,马文涛,王彬.通货膨胀预期、货币政策工具选择与宏观经济稳定[J].经济学(季刊),2010,10(1):51-82.
[8]林仁文,杨熠.中国市场化改革与货币政策有效性演变——基于DSGE的模型分析[J].管理世界,2014(6):39-52.
[9]伍戈,连飞.中国货币政策转型研究:基于数量与价格混合规则的探索[J].世界经济,2016(3):3-25.
[10]Sun L,Ford J L,Dickinson D G.Bank Loans and the Effects of Monetary Policy in China:VAR/VECM Approach[J].China Economic Review,2010,21(1):65-97.
[11]Goodfriend M,Prasad E.A Framework for Independent Monetary Policy in China[J].CESifo Economic Studies,2007,53(1):2-41.
[12]王君斌,郭新强,王宇.中国货币政策的工具选取、宏观效应与规则设计[J].金融研究,2013(8):1-15.
[13]陈浪南,田磊.基于政策工具视角的我国货币政策冲击效应研究[J].经济学(季刊),2014,14(1):285-304.
[14]Sims C.Interpreting the Macroeconomic Time Series Facts[J].European Economic Review,1992,36(5):975-1011.
[15]Bernanke Ben S,Boivin Jean,Eliasz Piotr.Measuring the Effects of Monetary Policy:A Factor-Augmented Vector Autoregressive(FAVAR)Approach[J].Quarterly Journal of Economics,2005,120(1):387-422.
[16]Primiceri G E.Time Varying Structural Vector Autoregressions and Monetary Policy[J].Review of Economic Studies,2005,72(3):821-852.
[17]Gali J,Gambetti L.On the Sources of the Great Moderation[J].American Economic Journal:Macroeconomics,2009,1(1):26-57.
[18]Bernanke B,Boivin J.Monetary Policy in a Data-Rich Environment[J].Journal of Monetary Economics,2003,50:525-546.
[19]He Q,Leung P H,Chong T L.Factor-Augmented VAR Analysis of the Monetary Policy in China[J].China Economic Review,2013,25:88-104.
[20]Fernald J G,Spiegel M M,Swanson E T.Monetary Policy Effectiveness in China:Evidence from a FAVAR Model[J].Journal of International Money and Finance,2014,49:83-103.
[21]欧阳志刚,薛龙.新常态下多种货币政策工具对特征企业的定向调节效应[J].管理世界,2017(2):53-65.
[22]王少平,朱满洲,胡朔商.中国CPI的宏观成分与宏观冲击[J].经济研究,2012(12):29-42.
[23]杜海韬,邓翔.部门价格动态、特质冲击与货币政策——基于结构动态因子方法[J].经济研究,2013(12):93-105.
[24]蔡晓陈,蒋涛.特有冲击、共同冲击与通货膨胀持续性[J].统计研究,2014,31(5):48-53.
[25]Korobilis D.Assessing the Transmission of Monetary Policy Using Time-Varying Parameter Dynamic Factor Models[J].Oxford Bulletin of Economics and Statistics,2013,75(2):157-179.
[26]张龙,金春雨.数量型和价格型货币政策工具的有效性对比研究[J].中国工业经济,2018(1):119-136.
[27]Doan T,Litterman R,Sims C A.Forecasting and Conditional Projection Using Realistic Priordistributions[J].Econometric Reviews,1986,3:1-100.
[28]Kim S,Shephard N,Chib S.Stochastic Volatility:Likelihood Inference and Comparison with ARCH Models[J].Review of Economic Studies,1998,65:361-393.
[29]Gerlach R,Carter C,Kohn R.Efficient Bayesian Inference in Dynamic Mixture Models[J].Journal of the American Statistical Association,2000,95:819-828.
[30]Boivin J,Giannoni Marc P,Mihov Ilian.Sticky Prices and Monetary Policy:Evidence from Disaggregated US Data[J].American Economic Review,2009,99(1):350-384.
[31]Bils Mark,Klenow Peter J.Some Evidence on the Importance of Sticky Prices[J].Journal of Political Economy,2004,112(5):947-985.
(1)限于篇幅,本文未给出新凯恩斯DSGE模型的具体推导过程,有需要者可向作者索取。
(2)产出缺口(yt=yt-yn t)被定义为实际产出(yt)与中央政府制定的当期GDP增长目标(yn t)之间的缺口。
(3)若允许扰动项之间存在弱的相关性,则模型的动态性会更强,但这会使估计参数急剧增加,可能导致模型估计有偏。
(4)明尼苏达规则为:当Vij=1/c2时,参数自身滞后;当Vij=0.001s2i/c2s2j时,其他变量参数滞后且i≠j,logc=1,…,p。其中,s2i是p阶滞后对因变量i回归方差的残差,i=1,…,M,j=1,…,MP。
(5)两步主成分分析的具体方法是,首先从Xt中提取出K×M个主成分并估计出共同成分Ct,得到F′t对应的由C′t生成的空间部分,然后将F′t代入式(3)的估计模型。此外,给模型施加Λf′Λf/N=I或F′F/T=I约束,可以得到相同的共同成分FΛf′和因子空间,我们设定F′=(TZ)1/2,其中,Z为XX′的K个最大特征值对应的特征向量。
(6)由于检验结果数据庞大,此处未列出详细结果,有需要者可向作者索取。
(7)全部CPI部门分量是指包括居民消费者价格指数、城市居民消费者价格指数、农村居民消费者价格指数和居民消费价格指数(36城)的全部分量价格序列。
(8)Primiceri(2005)[16]的预置样本先验是将总体样本的前若干期作为训练样本,利用普通最小二乘回归(OLS)得到先验均值和方差,并将其作为待估样本的均值和方差代入模型进行估计。在稳健性检验中,我们使用60期预置样本先验。
(9)限于篇幅,本文未给出具体的稳健性检验结果,有需要者可向作者索取。