题  目1：Misallocation or Mismeasurement? Evidence from Plant-Level Data

报告人：董白羽  长江商学院
 

题  目2：Understanding Mate Preferences from Two-Sided Matching Markets: Identi?cation, Estimation and Policy Analysis

报告人：Yu-Wei Hsieh     University of Southern California 
 

主持人：叶  兵

时  间：2012年11月23日   星期五   14:00-17:00

地  点：浙大玉泉校区外经贸大楼418室

 
                                  欢迎广大师生参加！

内容概要1：When capital is mismeasured, two problems arise. First, the extent of and the effects of capital misallocation are overestimated. Second, unless one controls for the errors in variable, the parameter estimates of production function are biased. To address these problems, I develop a new method to estimate plant-level CES production functions. My estimator eliminates measurement-error bias. Simulations show that as the variance of the errors increases, the bias of the parameter estimates and dispersion of capital-labor ratios, output-capital ratios and estimated TFPs remain roughly the same when using the new estimator. By contrast, such bias increases when using the traditional method, especially when the variance of errors exceeds the variance of capital. Using Korean manufacturing data, I construct two measures of capital stock by considering embodied technology change and utilization of capital. By applying my new estimator, I find substantial errors in these two capital measures. I also find that, on average, the measured loss of TFP from capital misallocation using Hsieh and Klenow (2009) is overestimated by 10% and the elasticity of substitution between capital and labor is overestimated by 15%. Between 1991 and 1999, average embodied technology change in Korea is around 9%, comparably to the U.S. And in industries where technological change is faster, equipment is less substitutable with labor.

 内容概要2：In this paper we estimate the utility functions over partner’s characteristics from the aggregate matching patterns, using a structural two-sided matching model without transfer. A distinct feature of our approach is the ability to separately identify utility functions of men and women, which is absent in the matching models with transfer; e.g., Choo and Siow (2006). We argue that the deferred-acceptance algorithm can be represented as a special demand/supply system. It thus leads to a fast algorithm to compute the joint distribution of characteristics of married couples implied by the model without simulation. Based on it, we present a set of new nonparametric identification results and propose consistent estimators, which are free from the curse of dimensionality induced by large number of players. Testable implications of Men (Women) optimal stable matching are also investigated. Furthermore, we consider the inference problem without imposing equilibrium selection. We show that a set of moment inequalities can be derived from the no-blocking-pair condition without solving the game. Besides marriage market, our estimators can also apply to other two-sided matching markets in labor and IO. We use the data of Current Population Survey to estimate the utility functions over spouse’s education level. We find that men care more about the similarity in spouse’s education than women do. Meanwhile, women’s education level is becoming more attractive to men compared to 20 years ago. It is well known that marriage market is one of the key considerations of education choice, but such endogeneity problem in empirical two-sided matching models is usually assumed away. Therefore, we develop a structural pre-marital educational choice model that accounts for marriage market prospect. We use the estimated preference to compute the “return to schooling within marriage”, and estimate the cost distributions of education via the structural model. By shifting the cost distribution of women via the lump-sum tuition subsidy, we find that the college attendance rate increases for both women and men. Moreover, the new policy creates more college-educated couples, while the correlation of the joint distribution of education of married couples remains unchanged.

                                                                  CRPE秘书处
                                                                   2012-11-19