Hugh Macartney

Principal Economist at Amazon

Specializing in structural econometrics, causal inference, and machine learning to solve high-stakes workforce challenges at scale. Leading teams to deliver data-driven insights that shape strategic decisions for hundreds of thousands of employees.

5
Years at Amazon
15
Years Combined Experience
10+
Scientists Led
3
Production Models

Labor Investment Flywheel

Experience

Combining rigorous econometric methodology with practical business applications to drive measurable impact

Principal Economist

Amazon | PXT Central Science
2021 - Present
Arlington, VA

Lead scientists delivering structural econometric models and causal inference analyses that inform critical workforce decisions affecting hundreds of thousands of employees. Work cross-functionally with senior leadership, product managers, engineers, and data scientists to productionalize science at scale.

Structural Model of Labor Supply

Built a discrete choice econometric model estimating labor supply elasticities for employees across locations. The model combines internal and external data sources to quantify workforce response to wage and amenity changes. Used by senior leadership to inform compensation strategy, enabling counterfactual analysis of inflation scenarios, attrition patterns, and policy changes.

Candidate Screening Algorithm

Led development of a causal predictive model for candidate screening that is in production and used at scale. Continuously iterated on model accuracy, driving alignment with senior leadership. Generated economically impactful savings through improved hiring efficiency and reduced turnover.

Employee Experience Analytics

Developed causal predictive framework to identify key drivers of employee experience, securing strong executive buy-in. Model outputs directly inform operational decisions across the network, combining traditional econometric methods with modern ML techniques including GenAI-powered text embeddings for feature extraction.

Structural Econometrics Causal Inference Machine Learning Cross-functional Leadership Stakeholder Management Team Management GenAI Applications

Assistant Professor of Economics

Duke University
2011 - 2021
Durham, NC

Conducted rigorous research in labor economics, public economics, and economics of education. Published in top-tier journals including Journal of Labor Economics and Journal of Public Economics. Developed expertise in structural modeling, causal inference methods, and computational economics that now underpins my industry work.

Applied Econometrics Labor Economics Public Economics Peer-Reviewed Research Stata Matlab

Interactive Demonstrations

Explore key econometric concepts through interactive visualizations

Labor Market Equilibrium

Explore how labor supply and demand determine market-clearing wages. Hover over wage levels to see workforce supply, demand, and any excess supply or demand.

Causal Inference Simulator

Explore confounding and causal identification through an interactive DAG. Toggle variables on/off to see how causal paths change and whether treatment effects can be identified.

C Confounder T Treatment Y Outcome Z Instrument
Identification Status:
⚠️ Not Identified - Confounding bias present. The causal effect of T on Y cannot be estimated from observational data alone because C affects both treatment and outcome.

Economics Concepts Q&A

Select any topic below to learn about key econometric methods and concepts.

Selected Publications

Research foundation in applied econometrics and causal inference

with Eric Nielsen and Viviana Rodriguez
Labour Economics, 73: 102073, 2021
with Gregorio Caetano
Journal of Public Economics, 194: 104335, 2021
with John Singleton
Journal of Public Economics, 164: 165-182, 2018
Hugh Macartney
Journal of Labor Economics, 34(1): 1-28 (lead article), 2016

Get in Touch