从工厂层面数据估计加成率

ESTIMATING MARK‐UPS FROM PLANT‐LEVEL DATA*

Journal of Industrial Economics · 2009
被引 8
人大 A-ABS 3

中文导读

针对工厂数据不报告实际数量的问题,提出一个能控制未观测价格异质性的计量模型来估计加成率,并证明忽略价格异质性会导致估计值趋近于1,从而错误地推断企业缺乏市场势力。

Abstract

Typical plant‐level data sets do not report quantities. This paper shows that estimating mark‐ups (price‐cost ratio) in product‐differentiated industries using deflated sales to proxy quantity is not appropriate due to unobserved price heterogeneity. This paper presents an econometric model for estimating mark‐ups that controls for unobserved prices. The model shows that ignoring price heterogeneity results in mark‐up estimates that converge to one, whatever the value of the true price‐cost ratio. Estimates obtained using real data are consistent with this result, as they reveal that ignoring price heterogeneity leads to spurious evidence of firms with little or no market power.

加成率估计企业层面数据价格异质性市场势力