On stacking, discrete outcomes, matched inference, and data-driven controls

New papers on compositional shifts, coarsened outcomes, pre-trend testing and event-study identification

What to consider when reality has stopped cooperating

A DiD horror story?

Attending college is a significant human capital investment, but only about 60 % of those who start college will have a completed degree six years lat…

Universal treatments without clean controls, continuous policies without untreated groups, reversible reforms over time and statistical inference with handful of cases

America's largest non-profit had a broken distribution system. University of Chicago economists fixed it.

Emma Harrington and Matthew Khan recently posted their paper "Has the Rise of Work-from-Home Reduced the Motherhood Penalty in the Labor Market?" as an NBER working paper. I first encountered this pap...

Founded in 1920, the NBER is a private, non-profit, non-partisan organization dedicated to conducting economic research and to disseminating research findings among academics, public policy makers, an...

The devil in the ~DiD details~

The devil in the ~DiD details~

In recent times my online sphere has had a fair few people make the claim that, in some sense, the culture has moved on from Woke. The d...

What to do when data can't be pooled, timing gets messy, and TWFE isn't enough

YouTube video by NBER

Choice architecture interventions (or ‘nudges’) aim to guide behaviour by changing the proximal physical, social or psychological environment. In this Review, Szaszi and colleagues show that the avera...

On uniform confidence bands for conditional effects, a simple approach for small-N inference, refining the

Mining is often seen as having opposing local economic and environmental effects. This study models these dynamics across Brazilian municipalities and shows that outcomes vary over time and by type of...

Some practical insights for a more precise and robust estimation.

(July 2025) - I examine teacher preferences using a discrete-choice experiment linked to data on teacher effectiveness. I estimate willingness to pay for a rich set of compensation elements and workin...

The power of disaggregated causal effects.

How does withdrawing from a free trade agreement affect maritime cargo volumes? Are there external effects on freight transport between third-party countries? Using Brexit as a case study, in a diffe...

Why the common criticisms are outdated, and how new methods are making causal inference stronger.

How new methods handle treatment intensity, small samples with staggered adoption, and selection bias in observational studies

New methods and perspectives in causal inference research

What to do when your treatment isn’t binary, your sample leaks, and your SEs collapse

ESCoE will hold its annual conference on 21 - 23 May 2025 at King’s Business School, King’s College London.

What to do when your treatment isn’t binary, your sample leaks, and your SEs collapse