Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/300984 
Year of Publication: 
2024
Series/Report no.: 
IZA Discussion Papers No. 17088
Publisher: 
Institute of Labor Economics (IZA), Bonn
Abstract: 
Instrumental variables estimators typically must satisfy monotonicity conditions to be interpretable as capturing local average treatment effects. Building on previous research that suggests monotonicity is unlikely to hold in the context of school entrance age effects, we develop an approach for identifying the magnitude of the resulting bias. We also assess the impact on monotonicity bias of bandwidth selection in regression discontinuity (RD) designs, finding that "full sample" instrumental variables estimators may outperform RD in many cases. We argue that our approaches are applicable more broadly to numerous settings in which monotonicity is likely to fail.
Subjects: 
monotonicity
selection
entrance age
regression discontinuity
instrumental variable
JEL: 
C21
C26
C1
I2
I28
Document Type: 
Working Paper

Files in This Item:
File
Size





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