货币政策立场变化、消息冲击时变性与预期管理优化

Expectation management is a new direction of monetary policy regulation, but the specific ways for its improvement remain a mist. This project seeks to gain a better understanding of the above problems from an innovative perspective. This project tires to provide a rigorous definition of monetary policy expectation management, which reveals its intrinsic concept. Comprehensive analyses were then carried out from this new definition. Based on the needs of characterizing Chinese economy, this project derives from the perspective of monetary policy news shocks and develops a nonlinear stochastic volatility VAR model and a New Keynesian Markov Regime Switching DSGE framework to serve as the analytical foundations. This project aims to answer the following scientific questions: First, it identifies monetary policy news shocks and exams the dynamic relations between news shocks and uncertainty. By these empirical analyses, it investigates the time-varying characteristic of China's monetary policy news shocks and its causes. Second, to clarify the effectiveness of China's monetary policy expectation management, this project considers the time-varying characteristic of monetary policy news shocks and quantitively estimates their impacts on macroeconomic effects of expectation management. What’s more, it explores the effects of changing monetary policy stance on time-varying monetary policy news shocks. Finally, this project will conduct six counterfactual experiments based on optimal welfare criterion. Combining the empirical results from our experiments and the supportive tools from policy makers, this project proposes specific strategies for optimizing expectations management. This project’s contributions will be twofold. On one hand, it develops new concepts and innovative ideas, contributing to the literature on expectation management. On the other hand, it puts forward a series of suggestions on the strategies optimization of expectation management, which will be helpful for the practice of monetary policy operations in China.

预期管理是货币政策调控的新方向，然而如何对其进行具体优化仍是一个谜。本项目尝试创新研究思路，深入探讨预期管理的内涵，提出货币政策预期管理的规范定义。并基于刻画中国宏观经济的现实需要，构建包含货币政策消息冲击的非线性随机波动率VAR模型和新凯恩斯MS-DSGE模型作为分析基础。本项目旨在探讨以下科学问题：（1）对消息冲击进行识别，并检验消息冲击与不确定性的动态关系，考察我国货币政策消息冲击的时变性及其原因。（2）定量地估计时变性货币政策消息冲击对宏观经济波动的影响，并进一步探讨不同货币政策立场对消息冲击时变性的影响，明确我国货币政策预期管理的有效性。（3）基于福利优化标准进行六类反事实研究，由实证结论和决策层可以掌控的手段，引申出优化货币政策预期管理的系列具体建议。本项目的新概念和新思路，对于推进预期管理现有研究应有所贡献；系列预期管理优化策略，对于中国货币政策实践探索具有积极的借鉴意义。

预期管理优化是货币政策调控的新方向，在当前需求收缩、供给冲击、预期转弱的三重压力背景下具有重要理论价值与应用意义。本报告采用实证检验和理论分析相结合的方式，首先检验消息冲击与不确定性的动态关系，基于1992-2019年我国GDP增长率、CPI通胀率和M2增长率的季度数据对我国货币政策波动率进行估计，并运用马尔科夫链蒙特卡洛模拟方法对货币政策波动率进行测度。通过非线性随机波动率VAR模型识别货币政策不确定性与货币政策消息冲击，探讨货币政策消息冲击的时变性及其原因。然后，构建符合中国国情的动态随机一般均衡模型，将马尔可夫区制转移模型纳入到数量型货币政策规则中，并比较不同区制转移设定选择出拟合能力最优的模型。使用贝叶斯方法估计适用于我国的模型参数，利用脉冲响应函数和方差分解比较预期与未预期的货币政策冲击对我国宏观经济的作用，定量化地考察货币政策预期管理的有效性。最后基于福利优化标准进行六项反事实研究，包括改变消息期限结构、变化各种货币政策操作方式与政策目标，以及减少财政政策波动等，并根据模型的定量化输出、实证分析以及反事实模拟结果，引申出优化我国货币政策预期管理的系列具体建议。. 研究发现：（一）中国货币政策中确实存在噪声信息。如果将信号中的噪声的标准差降低到0，那么短期内消息冲击的作用效率会有明显的提高，且长期看来可以增大消息冲击对实体经济的累计影响。这说明货币政策中噪声信息的存在整体上削弱了货币政策的效果。（二）为提高货币政策有效性，应疏通货币政策利率传导渠道，继续重视并化解影子银行风险，并降低货币政策波动率，加强货币政策与宏观审慎政策“双支柱”调控。（三）当前中国公众消息预期行为的最优刻画为提前一个季度，并且含消息冲击的模型很好地拟合了现实经济波动。我国应增强货币政策的连贯性，避免在短期内货币政策立场出现明显的区制转换，进而避免由货币政策预期不一致而引发的宏观经济波动。

内容获取失败，请点击重试