Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/227241 
Year of Publication: 
2020
Series/Report no.: 
IZA Discussion Papers No. 13714
Publisher: 
Institute of Labor Economics (IZA), Bonn
Abstract: 
In this paper we highlight an important property of the translog production function for the identification of treatment effects in a model of latent skill formation. We show that when using a translog specification of the skill technology, properly anchored treatment effect estimates are invariant to any location and scale normalizations of the underlying measures. By contrast, when researchers assume a CES production function and impose standard location and scale normalizations, the resulting treatment effect estimates are biased. Interestingly, the CES technology with standard normalizations yields biased treatment effect estimates even when age-invariant measures of the skills are available. We theoretically prove the normalization invariance of the translog production function and then produce several simulations illustrating the effects of location and scale normalizations for different technologies and types of skills measures.
Subjects: 
children
human capital
dynamic factor analysis
measurement
policy
JEL: 
C13
C18
I38
J13
J24
Document Type: 
Working Paper

Files in This Item:
File
Size
363.19 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.