新兴经济体汇率制度选择:状态依存的视角

Exchange Rate Regime Choice in Emerging Economies:A State Dependence Perspective

本文在多元离散选择模型中引入真实状态依存和虚假状态依存,试图解释大部分新兴经济体长期采用爬行钉住汇率制度、波幅与管理浮动汇率制度以稳定名义汇率的事实。研究发现:(1)新兴经济体的汇率制度选择主要由状态依存决定,并且真实状态依存这一内在驱动因素是导致新兴经济体采用爬行钉住、波幅与管理浮动汇率制度的重要原因;(2)非状态依存所代表的外部决定因素对新兴经济体汇率制度选择的影响不显著也不稳健,这与经典汇率制度选择理论的预测不同;(3)具有稳定名义汇率功能的爬行钉住、波幅与管理浮动汇率制度并不必然具有危机倾向,相反在一定条件下具有较高可维持性和稳定经济的作用。中国2015年"8·11汇改"以来逐步增加人民币汇率稳定性的政策调整有内在合理性。最后,本文给出了有针对性的启示与政策含义。

The choice of exchange rate regime is a classic issue of open macroeconomics.Recent literatures still focus on the issue,but the main concern is the fixed and the floating exchange rate regime.Since the collapse of the Bretton Woods system,the number of emerging economies that choose the crawling peg and the band/management floating exchange rate regimes has continued to increase,and the above exchange rate regimes have shown strong sustainability.This paper,introducing true and spurious state dependence into multinomial discrete choice model,tries to provide new empirical explanations to the stylized facts of the long-term adoption of the crawling peg and the band/management floating exchange rate regimes in most emerging economies.This paper tries to identify the true and spurious state dependence by using dynamic multinomial discrete choice logit model with random effects.This paper brings in 14 explanatory variables and control variables such as capital account openness,foreign exchange reserve,and currency mismatch,and the dataset covered 25 emerging economies from 1980 to 2016.The data in this paper derives from Ilzetzki et al.(2019),Chinn&Ito(2006),Lane&Milesi-Ferretti(2018),the World Economic Outlook Reports of the International Monetary Fund(IMF),International Financial Statistics of the IMF,Direction of Trade Statistics of the IMF,World Development Indicators(WDI),and Center for Systemic Peace Database.The main findings are as follows.Firstly,the exchange rate regimes choice of emerging economies is mainly determined by the state dependence factors,and the true state dependence is the most important reason for the long-term and widespread adoption of the crawling peg and the band/management floating exchange rate regimes in emerging economies.Secondly,the impacts of non-state-dependence factors on exchange rate regimes choice are statistically neither significant nor robust,which is largely different from the prediction of the classical exchange rate regime choice theory.Thirdly,the existing literature may overestimate the true impact of the explanatory variables and control variables on the choice of exchange rate regimes because the state dependence is not identified by the multinomial discrete choice model.Fourthly,the crawling peg and the band/management floating exchange rate regimes do not necessarily show a crisis tendency and are sustainable and can stabilize the economy under certain conditions.China’s policy adjustments to gradually increase the stability of the renminbi exchange rate since the exchange rate reform on August 11,2015 are inherently reasonable.The main contributions of the research are as follows.Firstly,this paper advances the binary choice model to the multiple choices model.Specifically,this paper classifies the exchange rate regimes into fixed,crawling pegs,band/management floating,and floating exchange rate regimes.This is important for identifying and explaining the evolution of exchange rate regimes in emerging economies.Secondly,the research promotes the identification method of the spurious state dependence in the evolution of the exchange rate regime.Hagen&Zhou(2007)used random draws method to identify spurious state dependence.Berdiev et al.(2012)used the within-group mean differencing method to estimate fixed effect,thus identifying spurious state dependence.Different from Hagen&Zhou(2007)and Berdiev et al.(2012),this paper uses Halton draws method with better sample coverage and convergence properties to identify spurious state dependence.Thirdly,this paper advances the identification method of the true state dependence in the evolution of the exchange rate regime and considers the calculation of marginal effects in parameter estimation.Berdiev et al.(2012)did not include the true state dependence into the discrete choice model.Hagen&Zhou(2007)incorporated the true state dependence into the discrete choice model,but did not discuss its identification methods in depth.Different from Hagen&Zhou(2007)and Berdiev et al.(2012),this paper used the method in Heckman(1981)and Gong et al.(2004)to obtain the consistent estimator of the true state dependence,and found that the introduction of the true state dependence can change the estimation results in the multiple choices model greatly.Fourthly,this paper finds that the choice of exchange rate regime in emerging economies is mainly determined by state dependence.The true state dependence is the important reason for the long-term and widespread adoption of crawling peg and the band/management floating exchange rate regimes in emerging economies.The impacts of non-state-dependence factors on exchange rate regimes choice are statistically neither significant nor robust,which is largely different from the prediction of the classical exchange rate regime choice theory.