Research

Current and published research in econometrics, causal inference, machine learning, and applied microeconomics.

Current projects are grouped by research program.

Current Research

AI, Machine Learning, and Economic Measurement

Representation Risk in Pretrained Image Encoders

Submitted

Studies representation risk: different plausible pretrained image encoders applied to the same images can produce materially different downstream conclusions. The paper compares modern and legacy encoders across economic, medical, biological, and demographic tasks and studies representation selection, combination, robustness, uncertainty, computational cost, and a reproducible software framework.

arXiv · Software · Watch video

Images as Covariates: When do they Improve Prediction from Structured Data?

Ardyn Nordstrom, Morgan Nordstrom, Vamuyan Sesay, and Matthew D. Webb

2026 working paper

Examines when adding images improves out-of-sample prediction beyond structured covariates. It distinguishes image-only prediction, overlap with structured data, and incremental predictive value, using Monte Carlo experiments and applications to Toronto house prices, Thoroughbred workout performance, and Ontario academic salaries.

Working paper (GitHub)

Econometric Methods and Inference

Which Policy Works, and Where? Estimation and Inference for State-Level Treatment Effects in Difference-in-Differences

Nichole Austin, Sunny R. Karim, Erin Strumpf, and Matthew D. Webb

Submitted

Distinguishes aggregate from state-level and other sub-aggregate treatment effects, and studies inference for policy comparisons across jurisdictions, cohorts, periods, and policy types.

arXiv · Watch video

Improved Inference for CSDID Using the Cluster Jackknife

Sunny R. Karim, Morten Ørregaard Nielsen, James G. MacKinnon, and Matthew D. Webb

Working paper

Develops cluster-jackknife inference for the Callaway-Sant’Anna difference-in-differences estimator. Companion software includes fastcsdid for stationary repeated cross sections in Stata, R, and Python.

arXiv · Software · fastcsdid (GitHub) · Watch video

Good Controls Gone Bad: Difference-in-Differences with Covariates

Sunny Karim and Matthew D. Webb

Working paper

Studies identification with time-varying covariates and introduces the Intersection Difference-in-Differences (DID-INT) estimator.

arXiv · DID-INT software · Watch video

Jackknife Inference with Two-Way Clustering

James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Revisions requested at Econometric Reviews

Develops jackknife variance estimators and inference methods for linear models with clustering in two dimensions.

arXiv · Software · Watch video

Difference-in-Differences with Unpoolable Data

Sunny Karim, Nichole Austin, Erin Strumpf, and Matthew D. Webb

Submitted

Develops UN-DID methods for estimating treatment effects when data held in separate silos cannot be pooled.

arXiv · UN-DID software

Applied Microeconomics

Fewer Homes, Similar Neighborhoods: Revisiting the Evidence for Racial Housing Discrimination in ‘Sorting or Steering’

Shi Chen, Areez Gangji, Sunny Karim, Anthony McCanny, and Matthew D. Webb

Resubmitted; second round of revisions requested at the Journal of Political Economy

Previously circulated as ‘The Many Misspellings of Albuquerque.’

Current paper (GitHub) · Earlier version (PDF)

Gun Licenses, Restricted Firearms, Deaths and Crime: Results from Administrative Data in Canada

Derek Mikola and Matthew D. Webb

Working paper

Uses Canadian administrative data to study legal firearm ownership, deaths, and crime.

arXiv · Watch video

Published Research

A Comment on ‘Delivering Remote Learning Using a Low-Tech Solution: Evidence from a Randomized Controlled Trial in Bangladesh’

Lenka Fiala, Jack Fitzgerald, Essi Kujansuu, Derek Mikola, David Valenta, Juan P. Aparicio, Michael Wiebe, Matthew D. Webb, and Abel Brodeur

Accepted at Journal of Political Economy Microeconomics (2026)

Choosing Portfolios of Rival Risky Options: Evidence That Most People Violate the No Safety Schools Theorem

David B. Johnson and Matthew D. Webb

Journal of the Economic Science Association 11(2), 173-197 (2025)

Published version · Working paper · Watch video

Cluster-Robust Jackknife and Bootstrap Inference for Binary Response Models

James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Forthcoming, Econometric Reviews (2025)

Published version · Preprint · Watch video

Leverage, Influence, and the Jackknife in Clustered Regression Models: Reliable Inference Using summclust

James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Stata Journal 23(4), 942-982 (2023)

Published version · Preprint · Watch video

Testing for the Appropriate Level of Clustering in Linear Regression Models

James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Journal of Econometrics 235, 2027-2056 (2023)

Published version · Watch video

Fast and Reliable Jackknife and Bootstrap Methods for Cluster-Robust Inference

James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Journal of Applied Econometrics 38(5), 671-694 (2023)

Published version · Working paper · Watch video

Finish It and It’s Free: An Evaluation of College Graduation Subsidies

Derek Mikola and Matthew D. Webb

Economics of Education Review 93, 102355 (2023)

Published version · Working paper · Watch video

Reworking Wild Bootstrap-Based Inference for Clustered Errors

Matthew D. Webb

Canadian Journal of Economics 56(3), 839-858 (2023)

Published version · Working paper · Watch video

Cluster-Robust Inference: A Guide to Empirical Practice

James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Journal of Econometrics 232, 272-299 (2023)

Published version · Watch video

Clustering Methods for Statistical Inference

James G. MacKinnon and Matthew D. Webb

Handbook of Labor, Human Resources and Population Economics (2023)

Published version · Working paper

Bootstrap and Asymptotic Inference with Multiway Clustering

James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Journal of Business & Economic Statistics 39(2), 505-519 (2021)

Published version · Working paper

Impact of a Peer Comparison Intervention on Seasonal Influenza Vaccine Uptake in Community Pharmacy: A National Cluster Randomized Study

Matthew M. Loiacono, Christopher B. Nelson, Paul Grootendorst, Matthew D. Webb, Laura Lee Hall, Jeffrey C. Kwong, Nicholas Mitsakakis, Stacy Zulueta, and Ayman Chit

Journal of the American Pharmacists Association 61(5), 539-546 (2021)

Published version · Watch video

Randomization Inference for Accounting Researchers

Roger White and Matthew D. Webb

Journal of Financial Reporting 6(2), 129-141 (2021)

Published version · Watch video

Randomization Inference for Difference-in-Differences with Few Treated Clusters

James G. MacKinnon and Matthew D. Webb

Journal of Econometrics 218(2), 435-450 (2020)

Published version · Working paper · Watch video

A Simple, Graphical Procedure for Comparing Multiple Treatment Effects

Brennan Thompson and Matthew D. Webb

The Econometrics Journal 22(2), 188-205 (2019)

Published version · Working paper · Watch video

Fast and Wild: Bootstrap Inference in Stata Using boottest

David Roodman, James G. MacKinnon, Morten Ørregaard Nielsen, and Matthew D. Webb

Stata Journal 19(1), 4-60 (2019)

Published version · Working paper · Watch video

Wild Bootstrap Randomization Inference for Few Treated Clusters

James G. MacKinnon and Matthew D. Webb

Advances in Econometrics 39, 61-85 (2019)

Published version · Working paper · Watch video

Immigrant Category of Admission of the Parents and Outcomes of the Children: How Far Does the Apple Fall?

Casey Warman, Matthew D. Webb, and Christopher Worswick

Journal of Population Economics 32(1), 53-112 (2019)

Published version · Working paper · Watch video

The Wild Bootstrap for Few (Treated) Clusters

James G. MacKinnon and Matthew D. Webb

The Econometrics Journal 21(2), 114-135 (2018)

Published version · Working paper · Watch video

Wild Bootstrap Inference for Wildly Different Cluster Sizes

James G. MacKinnon and Matthew D. Webb

Journal of Applied Econometrics 32(2), 233-254 (2017)

Published version · Working paper · Watch video

Pitfalls when Estimating Treatment Effects Using Clustered Data

James G. MacKinnon and Matthew D. Webb

The Political Methodologist 24(2), 20-31 (2017)

Published version · Working paper · Watch video

Targeting Tax Relief at Youth Employment

Arthur Sweetman, Casey Warman, and Matthew D. Webb

Canadian Public Policy 42(4), 415-430 (2016)

Published version · Working paper · Watch video

Meta Papers

AI-Assisted Teams Outperform AI-Led Teams but Not Human-Only Teams in Assessing Research Reproducibility in Quantitative Social Science

Abel Brodeur, David Valenta, Juan P. Aparicio, many others, Matthew D. Webb, and many others (200+ authors)

Forthcoming, Proceedings of the National Academy of Sciences (2026)

Reproducibility and Robustness of Economics and Political Science Research

Abel Brodeur, Derek Mikola, Nikolai Cook, many others, Matthew D. Webb, and many others (300+ authors)

Nature 652, 151-156 (2026)