Productivity Decomposition in Heterogeneous Industries*
提出一种新方法，直接基于行业数据测算总生产率变化，并将其分解为企业生产率变化、行业结构变化和异质性效应，避免传统加权平均法在异质性行业中的偏差。
We propose a novel methodology for computing variations of aggregate productivity and its decomposition into changes of individual firms’ productivities and of industry composition. Industry productivity should be measured in principle on industry total inputs and outputs but it is believed that this measure does not allow for such decomposition. Thus, weighted averages of individual productivities are used which introduce serious biases especially when industries are heterogeneous. We propose a geometric approach where aggregate productivity can be measured directly on industry data, but nevertheless its variations can be decomposed into between and within effects plus a heterogeneity effect.