Monetary policy reestimated: replicatication package.

We study the US monetary policy before and during the Great Moderation. We propose to use a few assumptions on an abstract dynamic stochastic general equilibrium model to derive restrictions on the associated structural vector autoregression model that identify the monetary policy rule and monetary policy shock. Using the estimated monetary policy rule, we argue that the Federal Reserve implemented the Friedman policy of a steady money growth before the Great Moderation. During the Great Moderation, the monetary policy followed the Taylor rule with generalized interest rate smoothing. The use of smoothing is consistent with the hypothesis that the Federal Reserve applied the optimal information filtering during the Great Moderation, but not before. This and other policy changes account for most of the reduction in macroeconomic volatility in the 1980s. The estimated impulse response functions to the monetary policy shock are large and significant, even on the Great Moderation data.本文针对大缓和（Great Moderation）时期前后的美国货币政策展开研究。我们基于抽象动态随机一般均衡（dynamic stochastic general equilibrium）模型提出若干假设，以此推导对应结构向量自回归（structural vector autoregression）模型的约束条件，从而实现货币政策规则与货币政策冲击的识别。借助估计得到的货币政策规则，我们认为在大缓和时期之前，美联储（Federal Reserve）推行了以稳定货币增长为核心的弗里德曼政策。在大缓和时期，货币政策遵循带有广义利率平滑特征的泰勒规则（Taylor rule）。这种利率平滑操作的使用，与“大缓和时期美联储采用了最优信息滤波，而此前并未采用”的假说相一致。这一政策变化与其他政策调整共同解释了20世纪80年代宏观经济波动性下降的绝大部分成因。即便基于大缓和时期的样本数据，针对货币政策冲击的估计脉冲响应函数（impulse response functions）依然幅度显著且统计意义显著。