Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/228114
Authors: 
Elminejad, Ali
Havranek, Tomas
Horváth, Roman
Year of Publication: 
2020
Series/Report no.: 
IES Working Paper No. 32/2020
Abstract: 
A key parameter in structural models is the Frisch elasticity of labor supply at the extensive margin, but empirical estimates vary greatly. We provide a quantitative synthesis of the literature. To this end, we collect 723 estimates from 36 studies along with 22 explanatory variables reflecting studies' characteristics and address model uncertainty by Bayesian and frequentist model averaging. Using linear and non-linear techniques, we find that publication bias exaggerates the mean of reported elasticities in the literature from 0.25 to 0.49. Our findings also suggest that two principal characteristics affect the magnitude of estimated elasticities systematically. First, identification bias: studies that follow a quasi-experimental approach tend to report smaller estimates. Second, aggregation bias: studies using macro data tend to report larger estimates. Furthermore, estimates associated with prime age or male workers tend to be systematically smaller, while studies relying on specific-industry data, near retirement workers, and probit regression tend to be larger.
Subjects: 
Frisch elasticity
extensive margin
meta-analysis
publication bias
Bayesian model averaging
JEL: 
E24
J20
J21
C83
Document Type: 
Working Paper

Files in This Item:
File
Size





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