I am interested in topics related to econometrics and applied microeconomics. Currently, my work focuses on production economics under the context of spatial interactions.
Job Market Paper
"Growing from Spillovers: A Semiparametric Varying Coefficient Approach " (Under Review)
Abstract: It is conventional wisdom that China's rapid industrialization in recent years is marked by increased agglomeration. This paper examines whether there are productivity spillover effects among clustered firms, and if so, whether human capital affects firms' absorption of spillovers. I propose a semiparametric spatial autoregressive production function in which coefficients are smooth unknown functions of firms' human capital. The varying coefficients not only allow for flexible interactions between human capital and other inputs, but also permit heterogeneous spatial dependence and spillover effects across firms. While the commonly used spatial weighted matrix captures possible learning opportunities among firms, I hypothesize that a firm with more human capital is better at seizing these opportunities. Furthermore, I tackle the simultaneity issue generated by the endogenous inputs using the proxy variable method. I implement this model empirically to analyze firm productivity of China's computer and peripheral equipment industry from 1998 to 2007, and examine the effects of human capital on output elasticities as well as productivity spillovers.
Work in Progress:
“A Spatial Autoregressive Stochastic Frontier Model with Spatial Correlated Inefficiency and Its Application”
Preliminary Draft available upon request
Abstract: This paper measures firms’ production inefficiencies while considering the spatial dependence between firms. Best to my knowledge, the stochastic frontier analysis literature has not given enough attention to interactions between firms. However, the omission of the spatial effect may bias our estimates of the efficiency scores. Therefore, I introduce the spatial autoregressive production function with spatial correlated error component to the stochastic frontier analysis to overcome the limitations of traditional stochastic frontier models. To identify proposed production function, I implement a two-step procedure to handle the special structure in the estimated equation with a focus on the issues of spatial dependence and semiparametric smooth coefficients. A likelihood function for the error component of the specific structure is derived to facilitate our MLE estimation in the second step. Monte Carlo simulations indicate that my estimator has a good finite-sample performance.
Work in Progress:
“Public Capital and the Productivity Puzzle”