« Back to Results

Advances in Causal Inference

Paper Session

Friday, Jan. 3, 2025 10:15 AM - 12:15 PM (PST)

San Francisco Marriott Marquis, Foothill B
Hosted By: International Association of Applied Econometrics
  • Chair: Francis Vella, Georgetown University

Assessing Exogeneity in Instrumental Variable Models without First Stage Monotonicity Restrictions

Paul Diegert
,
Toulouse School Of Economics
Matthew A. Masten
,
Duke University
Alexandre Poirier
,
Georgetown University

Abstract

N/A

Selection and Parallel Trends

Dalia Ghanem
,
University of California-Davis
Pedro Sant'Anna
,
Emory University
Kaspar Wuthrich
,
University of Michigan

Abstract

We study the role of selection into treatment in difference-in-differences (DiD) designs.
We derive necessary and sufficient conditions for parallel trends assumptions under general classes of selection mechanisms. These conditions characterize the empirical content of parallel trends. We use the necessary conditions to provide a selection-aware decomposition of the bias of DiD and provide easy-to-implement strategies for benchmarking its components. We also provide templates for justifying DiD in applications with and without covariates. A reanalysis of the causal effect of NSW training programs demonstrates the usefulness of our selection-based approach to benchmarking the bias of DiD.

Overidentification in Shift-Share Designs

Jinyong Hahn
,
University of California-Los Angeles
Guido Kuersteiner
,
University of Maryland
Andres Santos
,
University of California-Los Angeles
Wavid Willigrod
,
University of California-Los Angeles

Abstract

This paper studies the testability of identifying restrictions commonly employed to assign a causal interpretation to two stage least squares (TSLS) estimators based on Bartik instruments. For homogeneous effects models applied to short panels, our analysis yields testable implications previously noted in the literature for the two major available identification strategies. We propose overidentification tests for these restrictions that remain valid in high dimensional regimes and are robust to heteroskedasticity and clustering. We further show that homogeneous effect models in short panels, and their corresponding overidentification tests, are of central importance by establishing that: (i) In heterogenous effects models, interpreting TSLS as a positively weighted average of treatment effects can impose implausible assumptions on the distribution of the data; and (ii) Alternative identifying strategies relying on long panels can prove uninformative in short panel applications. We highlight the empirical relevance of our results by examining the viability of Bartik instruments for identifying the effect of rising Chinese import competition on US local labor markets.

Synthetic Difference-in-Differences via Distribution Regression

Ivan Fernandez-Val
,
Boston University
Jonas Meier
,
Swiss National Bank
Aico Van Vuuren
,
University of Groningen
Francis Vella
,
Georgetown University

Abstract

N/A
JEL Classifications
  • C0 - General
  • C2 - Single Equation Models; Single Variables