The Cure Can Be Worse than the Disease: A Cautionary Tale Regarding Instrumental Variables

23 Pages Posted: 25 Aug 2000 Last revised: 26 Sep 2010

See all articles by John Bound

John Bound

University of Michigan; National Bureau of Economic Research (NBER)

David A. Jaeger

Ph.D. Program in Economics, City University of New York Graduate Center; National Bureau of Economic Research (NBER); University of Cologne - Department of Economics; IZA Institute of Labor Economics; CESifo (Center for Economic Studies and Ifo Institute); University College London - CReAM - Centre for Research and Analysis of Migration

Regina Baker

University of Michigan at Ann Arbor - Institute for Social Research (ISR)

Date Written: June 1993

Abstract

In this paper we draw attention to two problems associated with the use of instrumental variables (IV) whose importance for empirical work has not been fully appreciated. First, using potential instruments that explain little of the variation in the: endogenous explanatory variables can lead to large inconsistencies of the IV estimates even If only a weal< relationship exists between the Instruments and the error in the structural equation. Second. In finite samples. IV estimates are biased in the same direction as ordinary least squares (OLS) estimates. The magnitude of the bias of IV estimates approaches that of OLS estimates as the R[squared] between the instruments and the potentially endogenous explanatory variable approaches O. To illustrate these problems with IV estimation we reexamine the results of the recent provocative paper by Angrist and Krueger "Does Compulsory School Attendance Affect Schooling and Earnings?" and find evidence that their IV estimates of the effects of educational attainment on earnings are possibly both inconsistent and suffer from finite sample bias. To gauge the severity of both problems we suggest that both the partial R[squared] and the F statistic on the excluded instruments from the first stage estimation be reported when using IV as approximate guides to the quality of the IV estimates.

Suggested Citation

Bound, John and Jaeger, David A. and Baker, Regina, The Cure Can Be Worse than the Disease: A Cautionary Tale Regarding Instrumental Variables (June 1993). NBER Working Paper No. t0137. Available at SSRN: https://ssrn.com/abstract=240089

John Bound (Contact Author)

University of Michigan ( email )

611 Tappan Street
Ann Arbor, MI 48109-1220
United States
313-998-7149 (Phone)
313-998-7415 (Fax)

National Bureau of Economic Research (NBER)

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

David A. Jaeger

Ph.D. Program in Economics, City University of New York Graduate Center ( email )

NBER
5 Hanover Square, Suite 1602
New York, NY 10004
United States

HOME PAGE: http://www.djaeger.org

National Bureau of Economic Research (NBER) ( email )

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

University of Cologne - Department of Economics ( email )

Cologne, 50923
Germany

IZA Institute of Labor Economics

Schaumburg-Lippe 5/9
Bonn, 53113
Germany

HOME PAGE: http://www.iza.org

CESifo (Center for Economic Studies and Ifo Institute) ( email )

Poschinger Str. 5
Munich, DE-81679
Germany

University College London - CReAM - Centre for Research and Analysis of Migration ( email )

Drayton House
30 Gordon Street
London, WC1H 0AX
United Kingdom

Regina Baker

University of Michigan at Ann Arbor - Institute for Social Research (ISR) ( email )

426 Thompson Street
Ann Arbor, MI 48106-1248
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
52
Abstract Views
1,325
rank
376,768
PlumX Metrics