Structural estimation of firm productivity with endogenous determinants
指出代理变量方法中生产率冲击与过去信息独立的假设在实证中可能过于严格，并提出了五种结构性框架下的解决方案，通过蒙特卡洛模拟评估其有限样本表现，为实证研究者处理内生性问题提供实用指南。
Proxy-variable methods (also known as structural approaches) for estimating production functions and firm productivity have developed rapidly and become a dominant empirical tool for analyzing firms’ production technology and productivity dynamics. A central identification assumption in this literature is that in the productivity process the contemporaneous productivity shock is independent of past information, which ensures the econometric exogeneity of productivity shifters, provided they are lagged by at least one period. However, we argue that this assumption is ad hoc and potentially restrictive in empirical contexts such as those examining the productivity effects of foreign investment, technology adoption, and similar factors. To address potential endogeneity concerns, we propose five solutions within the structural framework, each based on different assumptions about the data-generating process. Using Monte Carlo simulations, we assess the finite-sample performance of these approaches. Our study aims to serve as a practical guide for empirical researchers applying proxy-variable methods to estimate productivity effects while mitigating endogeneity issues.