通知公告
   首页» 通知公告
通知公告
6.5经济系Seminar:Estimating the Production Elasticity of Labor: A Partial Identification Approach
发布时间:2014-05-22       浏览量:

  题:Estimating the Production Elasticity of Labor: A Partial Identification Approach

  间:201465 14:30-16:00

  点:后主楼1610

主讲人:Tadashi SONODA日本名古屋大学,副教授

  

内容简介:This study seeks a better estimate of the production elasticity of labor for Chinese farm households and improves the imperfect instrumental variables (IIVs) method of Nevo and Rosen (NR). Compared with the method of instrumental variables (IVs) and that of Olley and Pakes (OP), IIVs method has two advantages. First, IIVs play a similar role to IVs or investment demand in OP method. However, they can have positive, negative, or no correlation with the error term unlike IVs and do not require a strict relation with the error term unlike OP method. Second, IIVs method does not require panel data. Instead, it allows estimation of only ranges of parameters. Therefore, we need to improve the method to obtain narrower ranges of parameters. We choose four IIVs (costs of producing crops and livestock products and the number of male and female farm workers). We then examine correlation among endogenous labor input, IIVs, and the error term in the production function and use it to compare candidates of the upper and lower bounds of production elasticity of labor. When we use two IIVs together to have the upper and lower bounds of this elasticity, direct application of NR’s method produces implausible lower bounds of this elasticity. Based on the correlation among labor input, IIVs, and the error term, we explain reasons for this result and propose a better procedure to identify and estimate the upper and lower bounds of the elasticity. Applying our procedure to cross sectional data on 4242 Chinese farm households, we estimate production elasticity of labor to be between 0.12 and 0.20, 0.11 and 0.17, and 0.06 and 0.10 for three different regions. These ranges do not include the corresponding OLS estimates, 0.43, 0.21, and 0.16, implying implausibility of IVs to justify the OLS estimates in Durbin-Wu-Hausman test..

 

主讲人简介:Tadashi SONODA2000年毕业于日本筑波大学,现为日本名古屋大学副教授。其论文发表在American Journal of Agricultural Economics Applied Financial EconomicsSingapore Economic Review等国际期刊上。