Local instruments, global extrapolation: External validity of the labor supply–fertility local average treatment effect

James Bisbee, Rajeev Dehejia, Cristian Pop-Eleches, Cyrus Samii

Research output: Contribution to journalArticle

Abstract

We investigate the external validity of local average treatment effects (LATEs), specifically Angrist and Evans’s use of same sex of the two first children as an instrumental variable for the effect of fertility on labor supply. We estimate their specification in 139 country-year censuses using Integrated Public Use Microdata Sample–International data. We compare each country-year’s actual LATE to the extrapolated LATE from other country-years. We find that, with a sufficiently large reference sample, we extrapolate the treatment effect reasonably well, but the degree of accuracy depends on the extent of covariate similarity between the target and reference settings.

Original languageEnglish (US)
Pages (from-to)S99-S147
JournalJournal of Labor Economics
Volume35
Issue numberS1
DOIs
StatePublished - 2017

Fingerprint

External validity
Average treatment effect
Extrapolation
Labor
Covariates
Labor supply
Fertility
Census
Instrumental variables
Treatment effects
Micro data
Integrated

ASJC Scopus subject areas

  • Industrial relations
  • Economics and Econometrics

Cite this

Local instruments, global extrapolation : External validity of the labor supply–fertility local average treatment effect. / Bisbee, James; Dehejia, Rajeev; Pop-Eleches, Cristian; Samii, Cyrus.

In: Journal of Labor Economics, Vol. 35, No. S1, 2017, p. S99-S147.

Research output: Contribution to journalArticle

@article{bcfd84ff924942f1966f108acddd5eed,
title = "Local instruments, global extrapolation: External validity of the labor supply–fertility local average treatment effect",
abstract = "We investigate the external validity of local average treatment effects (LATEs), specifically Angrist and Evans’s use of same sex of the two first children as an instrumental variable for the effect of fertility on labor supply. We estimate their specification in 139 country-year censuses using Integrated Public Use Microdata Sample–International data. We compare each country-year’s actual LATE to the extrapolated LATE from other country-years. We find that, with a sufficiently large reference sample, we extrapolate the treatment effect reasonably well, but the degree of accuracy depends on the extent of covariate similarity between the target and reference settings.",
author = "James Bisbee and Rajeev Dehejia and Cristian Pop-Eleches and Cyrus Samii",
year = "2017",
doi = "10.1086/691280",
language = "English (US)",
volume = "35",
pages = "S99--S147",
journal = "Journal of Labor Economics",
issn = "0734-306X",
publisher = "University of Chicago",
number = "S1",

}

TY - JOUR

T1 - Local instruments, global extrapolation

T2 - External validity of the labor supply–fertility local average treatment effect

AU - Bisbee, James

AU - Dehejia, Rajeev

AU - Pop-Eleches, Cristian

AU - Samii, Cyrus

PY - 2017

Y1 - 2017

N2 - We investigate the external validity of local average treatment effects (LATEs), specifically Angrist and Evans’s use of same sex of the two first children as an instrumental variable for the effect of fertility on labor supply. We estimate their specification in 139 country-year censuses using Integrated Public Use Microdata Sample–International data. We compare each country-year’s actual LATE to the extrapolated LATE from other country-years. We find that, with a sufficiently large reference sample, we extrapolate the treatment effect reasonably well, but the degree of accuracy depends on the extent of covariate similarity between the target and reference settings.

AB - We investigate the external validity of local average treatment effects (LATEs), specifically Angrist and Evans’s use of same sex of the two first children as an instrumental variable for the effect of fertility on labor supply. We estimate their specification in 139 country-year censuses using Integrated Public Use Microdata Sample–International data. We compare each country-year’s actual LATE to the extrapolated LATE from other country-years. We find that, with a sufficiently large reference sample, we extrapolate the treatment effect reasonably well, but the degree of accuracy depends on the extent of covariate similarity between the target and reference settings.

UR - http://www.scopus.com/inward/record.url?scp=85021732724&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85021732724&partnerID=8YFLogxK

U2 - 10.1086/691280

DO - 10.1086/691280

M3 - Article

VL - 35

SP - S99-S147

JO - Journal of Labor Economics

JF - Journal of Labor Economics

SN - 0734-306X

IS - S1

ER -