Research

Working papers

Causal Inference for Spatial Treatments (pdf) [Code]
Abstract

Many events and policies (treatments), such as opening of businesses, building of hospitals, and sources of pollution, occur at specific spatial locations, with researchers interested in their effects on nearby individuals or businesses (outcome units). However, the existing treatment effects literature primarily considers treatments that could experimentally be assigned directly at the level of the outcome units, potentially with spillover effects. I approach the spatial treatment setting from a similar experimental perspective: What ideal experiment would we design to estimate the causal effects of spatial treatments? This perspective motivates a comparison between individuals near realized treatment locations and individuals near counterfactual (unrealized) candidate locations, which is distinct from current empirical practice. I derive standard errors based on this design-based perspective that are straightforward to compute irrespective of spatial correlations in outcomes. Furthermore, I propose machine learning methods to find counterfactual candidate locations and show how to apply the proposed methods on observational data. I study the causal effects of grocery stores on foot traffic to nearby businesses during COVID-19 shelter-in-place policies. I find a substantial positive effect at a very short distance. Correctly accounting for possible effect "interference" between grocery stores located close to one another is of first order importance when calculating standard errors in this application.

Causal Inference from Hypothetical Evaluations (html) [R package],
with B. Douglas Bernheim, Daniel Björkegren, and Jeffrey Naecker
Abstract

This paper develops a method that learns the relationship between hypothetical responses and real choices in observational data, and then uses that estimated relationship to predict the effect of counterfactuals. After developing the econometric theory for the estimator, we demonstrate that it can be applied in settings where standard methods are not applicable. In both a lab and field setting we show it can recover accurate estimates of treatment effects that are close to ground truth experimental estimates.

Semiparametric Estimation of Treatment Effects in Randomized Experiments (html) [R package],
with Susan Athey, Peter J. Bickel, Aiyou Chen, and Guido W. Imbens, Revise & Resubmit, JRSSB
Abstract

We develop new semiparametric methods for estimating treatment effects. We focus on a setting where the outcome distributions may be thick-tailed, where treatment effects are small, where sample sizes are large and where assignment is completely random. We propose using parametric models for the treatment effects, as opposed to parametric models for the full outcome distributions. This leads to semiparametric models for the outcome distributions. We derive the semiparametric efficiency bound for this setting, and propose efficient estimators. In the case with a constant treatment effect one of the proposed estimators has an interesting interpretation as a weighted average of quantile treatment effects, with the weights proportional to the second derivative of the log of the density of the potential outcomes.

Published papers

Matthias Bäuml, Tilman Dette, and Michael Pollmann (2022) Price and income effects of hospital reimbursements. Journal of Health Economics 81: 102576, DOI:10.1016/j.jhealeco.2021.102576.
[final manuscript (pdf)] [estimated elasticities (csv)]
Abstract

Health insurance systems in many countries reimburse hospitals through fixed prices based on the diagnosis-related groups (DRGs) of patients. We quantify the effects of price and income changes for the full spectrum of hospital services as average and heterogeneous elasticities of quantities (number of admissions) and quality-related outcomes. For our empirical analysis, we use data on over 160 million hospital admissions, constituting the universe of hospital admissions in Germany between 2005 and 2016. Our identification strategy is based on instruments exploiting a two-year lag in regulatory price setting. The strategy lends itself to a placebo test demonstrating that our instruments do not have substantive anticipatory direct effects. We find that the compensated own-price elasticity of quantity is positive (0.2), while the income elasticity is negative (-0.15). On net, increasing all prices increases costs due to a behavioral response of larger quantities in addition to the mechanical increase.

Susan Athey, Guido W. Imbens, and Michael Pollmann (2019) Comment on: "The Blessings of Multiple Causes" by Yixin Wang and David M. Blei, Journal of the American Statistical Association, 114:528, 1602-1604, DOI:10.1080/01621459.2019.1691008 (invited comment).