« Back to Results

Regression Discontinuity Designs

Paper Session

Sunday, Jan. 4, 2026 8:00 AM - 10:00 AM (EST)

Philadelphia Marriott Downtown, Room 305
Hosted By: Econometric Society
  • Chair: Kirill Borusyak, University of California-Berkeley

Inference in Regression Discontinuity Designs with Clustered Data

Claudia Noack
,
University of Bonn

Abstract

Clustered sampling is prevalent in empirical regression discontinuity (RD) designs, but it did not gain much attention in the theoretical literature. In this paper, we introduce a general model-based framework for such settings and derive high-level conditions under which the standard local linear RD estimator is asymptotically normal. We show that clustered standard errors that are currently used in practice can be either inconsistent or overly conservative in finite samples. To address these issues, we propose a new nearest-neighbor-type variance estimator. We verify that our high-level assumptions hold across a wide range of empirical designs, including settings of growing cluster sizes, and demonstrate the estimator's finite-sample performance via simulations and a diverse set of applications.

Partially Linear Regression Discontinuity Inference

Aditya Ghosh
,
Stanford University
Guido Imbens
,
Stanford University
Stefan Wager
,
Stanford University

Abstract

Regression discontinuity designs have become one of the most popular research designs in empirical economics. We argue, however, that widely used approaches to building confidence intervals in regression discontinuity designs exhibit suboptimal behavior in practice: In a simulation study calibrated to high-profile applications of regression discontinuity designs, existing methods either have systematic under-coverage or have wider-than-necessary intervals. We propose a new approach, partially linear regression discontinuity inference (PLRD), and find it to address shortcomings of existing methods: Throughout our experiments, confidence intervals built using PLRD are both valid and short. We also provide large-sample guarantees for PLRD under smoothness assumptions.

An Empirical and Methodological Examination of Reforms to Maternity Leave in the United Kingdom

Clara von Bismarck-Osten
,
University College London and Institute for Fiscal Studies
Andrei Zeleneev
,
University College London

Abstract

This paper warns of attenuation bias in Regression Discontinuity Designs (RDD) when the running variable is measured with error––a problem to which studies of family policies are prone when the eligibility status is imprecisely observed. Using three reforms to the UK maternity leave legislation, we illustrate the concern, argue that standard correction approaches might be inappropriate and propose a simple and transparent solution. We establish identification of the conditional average treatment effect on the treated (CATT), leveraging availability of validation data and a control cohort. Our approach allows us to nonparametrically identify the CATT at and beyond the cutoff point, and, importantly, it remains valid even when the measurement error is differential (i.e., the counterfactual outcomes are affected by the error). We provide easy-to-implement nonparametric and semiparametric estimators of the CATT, and illustrate the usefulness of our approach by detecting effects which naïve approaches fail to find.

Regression Discontinuity Aggregation, with an Application to the Union Effects on Inequality

Matan Kolerman-Shemer
,
TBD
Kirill Borusyak
,
University of California-Berkeley

Abstract

We extend the regression discontinuity (RD) design to settings where each unit’s treatment status is an average or aggregate across multiple discontinuity events. Such situations arise in many studies where the outcome is measured at a higher level of spatial or temporal aggregation (e.g., by state with district-level discontinuities) or when spillovers from discontinuity events are of interest. We propose two novel estimation procedures — one at the level at which the outcome is measured and the other in the sample of discontinuities — and show that both identify a local average causal effect under continuity assumptions similar to those of standard RD designs. We apply these ideas to study the effect of unionization on inequality in the United States. Using credible variation from close unionization elections at
the establishment level, we show that a higher rate of newly unionized workers in a state-by-industry cell reduces wage inequality within the cell.

Discussant(s)
Zhuan Pei
,
Cornell University
Michal Kolesar
,
Princeton University
Yingying Dong
,
University of California-Irvine
Soonwoo Kwon
,
Brown University
JEL Classifications
  • C21 - Cross-Sectional Models; Spatial Models; Treatment Effect Models; Quantile Regressions