How Much Should We Trust Regression Discontinuity Design Estimates? Evidence from Experimental Benchmarks of the Incumbency Advantage

34 Pages Posted: 8 Oct 2020

See all articles by Leandro de Magalhaes

Leandro de Magalhaes

Bristol University

Dominik Hangartner

London School of Economics & Political Science (LSE); Stanford - Zurich Immigration Policy Lab; Public Policy Group

Salomo Hirvonen

University of Bristol

Jaakko Meriläinen

Instituto Tecnológico Autónomo de México (ITAM) - Centro de Investigacion Economica

Nelson A. Ruiz

University of Oxford

Janne Tukiainen

VATT Institute for Economic Research; University of Turku - Turku School of Business - Department of Economics

Date Written: July 22, 2020

Abstract

Regression discontinuity designs (RDD) are widely used in the social sciences to estimate causal effects from observational data. Scholars can choose from a range of methods that implement different RDD estimators, but there is a paucity of research on the performance of these different estimators in recovering experimental benchmarks. Leveraging exact ties in local elections in Colombia and Finland, which are resolved by random coin toss, we find that RDD estimation using bias-correction and robust inference (CCT) performs better in replicating experimental estimates of the individual incumbency advantage than local linear regression with conventional inference (LLR). We assess the generalizability of our results by estimating incumbency effects across different sub-samples, and in other countries. We find that CCT consistently comes closer to the experimental benchmark, produces smaller estimates than LLR, and that incumbency effects are highly heterogeneous, both in magnitude and sign, across countries with similar open-list PR systems.

Keywords: Close Elections, Personal Incumbency Advantage, Regression Discontinuity Design

Suggested Citation

de Magalhaes, Leandro and Hangartner, Dominik and Hirvonen, Salomo and Meriläinen, Jaakko and Ruiz, Nelson and Tukiainen, Janne, How Much Should We Trust Regression Discontinuity Design Estimates? Evidence from Experimental Benchmarks of the Incumbency Advantage (July 22, 2020). Available at SSRN: https://ssrn.com/abstract=3664069 or http://dx.doi.org/10.2139/ssrn.3664069

Leandro De Magalhaes

Bristol University ( email )

8 Woodland Road
Bristol BS8 ITN
United Kingdom

Dominik Hangartner

London School of Economics & Political Science (LSE) ( email )

Departments of Government and Methodology
Houghton Street
London, WC2A 2AE
United Kingdom

Stanford - Zurich Immigration Policy Lab

30 Alta Road
Stanford, CA 94305
United States

Public Policy Group ( email )

Rämistrasse 101
ZUE F7
Zürich, 8092
Switzerland

Salomo Hirvonen

University of Bristol ( email )

University of Bristol,
Senate House, Tyndall Avenue
Bristol, BS8 ITH
United Kingdom

Jaakko Meriläinen

Instituto Tecnológico Autónomo de México (ITAM) - Centro de Investigacion Economica ( email )

Av. Camino a Santa Teresa #930
Col. Heroes de Padierna
Mexico City, D.F. 10370
Mexico

HOME PAGE: http://sites.google.com/view/jaakkomerilainen/

Nelson Ruiz (Contact Author)

University of Oxford ( email )

Mansfield Road
Oxford, Oxfordshire OX1 4AU
United Kingdom

Janne Tukiainen

VATT Institute for Economic Research ( email )

Arkadiankatu 7
P.O Box 1279
Helsinki, FIN-00101
Finland

University of Turku - Turku School of Business - Department of Economics ( email )

Rehtorinpellonkatu 3
Turku, 20014
Finland

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
12
Abstract Views
114
PlumX Metrics