Production Function Estimation With Resource Misallocation
研究发现代理变量方法在资源错配下无法一致估计生产函数系数，并提出一种不依赖传统假设的新识别策略，通过蒙特卡洛实验和中国制造业数据验证了其有效性。
ABSTRACT We show that the proxy variable method fails to yield consistent estimates of production function coefficients in the presence of resource misallocation. This failure arises because unobserved firm‐specific distortions violate the assumptions of scalar unobservability and strict monotonicity. We propose a novel identification strategy that does not rely on these assumptions. Our method is robust to various distortions and production function specifications, and can be extended to accommodate serial correlation in unexpected productivity shocks. Monte Carlo experiments confirm the efficacy of our approach in consistently estimating production function coefficients under resource misallocation. Using a large panel of Chinese manufacturing firms—and employing the share of state‐owned firms as a proxy for industry‐level resource misallocation—we find that estimates obtained via the proxy variable method more closely align with those from our proposed approach when resource misallocation is reduced.