摘要
借鉴Diebold和Yilmaz(2012)广义预测误差方差分解思想,本文利用14个主要经济体2003年1月至2017年5月经济政策不确定性指数数据,测算了其溢出效应,并对形成机理进行了初步探讨。结果表明:发达国家的经济政策不确定性溢出效应整体水平要高于发展中国家,该溢出效应对突发大事较为敏感,而对于可预期世界大事则较为迟钝。在控制了个体和时间效应后,进行面板固定效应的FGLS估计,实证结果发现,经济政策不确定性及其波动对溢出效应的偏效应均为正,且波动值的估计系数远高于水平值,经济政策不确定性主要体现为波动溢出效应。由分位数回归估计结果可知,随着分位数的增加,对外贸易、经济政策不确定性水平值、波动值及交互项估计系数的绝对值逐渐变大,对经济政策不确定性溢出效应起主要作用的因素随之变化。为了更好抵抗来自其他国家经济政策不确定性的影响,各国应找到推动经济增长的内生动力,减少国际依赖。
In view of the generalized forecast error variance decomposition(Diebold & Yilmaz, 2012), this paper measures the spillover effect of economic policy uncertainty(EPU) and explores its formation mechanism based on EPU index over the period from January 2003 to May 2017. It is found that i) the spillover effect in the developed countries is stronger than that in the developing countries in general; ii) it is more sensitive to the emergencies than to the expected events. Applied with the individual and time effect controlled and the feasible generalized least squares(FGLS) estimation on panel fixed effect, the empirical results show that the effect of EPU and its partial effect of its volatility on the spillover is positive, and the estimated coefficient of the volatile value is higher than the mean value, so the EPU is mainly objectified as a volatile spillover effect. According to the quantile regression estimation, the mean and volatile values of foreign trade and economic policy uncertainty, as well as the absolute value of the cross item estimated coefficient turns large gradually, and the factors that play a major role in the spillover effect change. Therefore, each of the countries need to seek the endogenous driving force for its economic growth and to reduce its international dependence, in order to fend off the spillover of the economic policy uncertainty from other countries.
引文
[1]Baker S R, Bloom N, Davis S J. Measuring Economic Policy Uncertainty[J]. The Quarterly Journal of Economics, 2016, 131(4),1593-1636.
[2]Diebold F X, Yilmaz K. Better to Give Than to Receive: Predictive Directional Measurement of Volatility Spillovers[J]. International Journal of Forecasting, 2012, 28(1):57-66.
[3]Jurado K, Ludvigson S C, Ng S. Measuring Uncertainty[J]. The American Economic Review, 2015,105(3):1177-1216.
[4]Born B, Pfeifer J. Policy Risk and the Business Cycle[J]. Journal of Monetary Economics, 2014, 68(1):68-85.
[5]Johannsen B K. When are the Effects of Fiscal Policy Uncertainty Large?[J]. FEDS Working Paper,2014, No.2014-40.
[6]Creal D D, Wu J C. Monetary Policy Uncertainty and Economic Fluctuations[J]. International Economic Review, 2017, 58(4):1317-1354.
[7]Mumtaz H, Surico P. Policy Uncertainty and Aggregate Fluctuations[J]. Journal of Applied Econometrics, 2018, 33(4):319-331.
[8]Colombo V. Economic Policy Uncertainty in the US: Does it matter for the Euro area?[J]. Economics Letters, 2013, 121(1):39-42.
[9]Bhattarai S, Chatterjee A, Park W Y. Global Spillover Effects of US Uncertainty[J]. Social Science Electronic Publishing, 2017.
[10]Claeys P. Uncertainty Spillover and Policy Reactions[J]. Ensayos Sobre Política Económica, 2017, 35(82):64-77.
[11]Koop G, Pesaran H, Potter S. Impulse Response Analysis in Nonlinear Multivariate Models[J]. Journal of Econometrics, 1996, 74(1):119-147.
[12]Pesaran H, Shin Y. Generalized Impulse Response Analysis in Linear Multivariate Models[J]. Economics Letters, 1998, 58(1):17-29.
[13]Kl??ner S, Wagner S. Exploring All Var Ordering for Calculating Spillovers? Yes, We Can!-A note on Diebold and Yilmaz(2009)[J]. Journal of Applied Econometrics, 2014, 29(1):172-179.
[14]顾夏铭, 陈勇民, 潘士远. 经济政策不确定性与创新——基于我国上市公司的实证分析[J]. 经济研究, 2018(2):109-123.
[15]Mumtaz H, Zanetti F. The Impact of the Volatility of Monetary Policy Shocks[J]. Journal of Money Credit & Banking, 2013, 45(4):535-558.
[16]张喜艳, 陈乐一. 经济政策不确定性差异对经济周期协同的影响研究[J]. 中国软科学, 2017(8):46-54.
(1)数据来源于经济政策不确定性指数网站:http://www.policyuncertainty.com/index.html。
(2)G20集团成员有20个,包含19个国家和1个地区,由于欧盟包含了法国、意大利、德国等国家,故将欧盟从样本中剔除。
(3)参考Baker等(2016)[1]绘制而成。
(4)计算过程为γ4+γ6×c2+γ7×c3=-2.9060+0.1290×0.330+0.1140×25.145≈0.0031,c1、c2、c3分别为经济政策不确定性水平值(lnepu)、波动值(volepu)、对外贸易(lntrade)的样本均值。
(5)计算过程为γ5+γ6×c1+γ8×c3=-1.6580+0.1290×4.759+0.0573×25.145≈0.3967。
(6)计算过程为γ2+γ7×c1+γ8×c2=-0.6410+0.1140×4.759+0.0573×0.330≈-0.0796。
(7)限于数据的可获得性,暂时没有样本国之间的双边贸易月度数据。
(8)结果略,有需求可向作者索取。
(9)计算过程与前文类似。
(10)对外贸易、经济政策不确定性水平值、经济政策不确定性波动值当中的两个变量。
(11)结果略,如有需要可向作者索取。