Marginal outcome tests compare the expected effects of a decision on individuals who are of different races but at the same indifference point of the decision-maker. I present a simple formalization of how such tests can detect racial bias, defined as a deviation from accurate statistical...

Economics tends to define and measure discrimination as disparities stemming from the direct (causal) effects of protected group membership. But work in other fields notes that such measures are incomplete, as they can miss important systemic (i.e. indirect) channels. For example, racial...

Many studies use shift-share (or "Bartik") instruments, which average a set of shocks with exposure share weights. We provide a new econometric framework for such designs in which identification follows from the quasi-random assignment of shocks, allowing exposure shares to be endogenous. This...

We develop new tools for causal inference in settings where exogenous shocks affect the treatment status of multiple observations jointly, to different extents. In these settings researchers may construct treatments or instruments that combine the shocks with predetermined measures of shock...

Competition in health insurance markets may fail to improve health outcomes if consumers are not willing to pay for high quality plans. We document large differences in the mortality rates of Medicare Advantage (MA) plans within local markets. We then show that when high (low) mortality plans...

We develop new quasi-experimental tools to measure racial discrimination, due to either racial bias or statistical discrimination, in the context of bail decisions. We show that the omitted variables bias in observational release rate comparisons can be purged by using the quasi-random...

There is growing concern that the rise of algorithmic decision-making can lead to discrimination against legally protected groups, but measuring such algorithmic discrimination is often hampered by a fundamental selection challenge. We develop new quasi-experimental tools to overcome this...

Many large urban school districts match students to schools using algorithms that incorporate an element of random assignment. We introduce two simple empirical strategies to harness this randomization for value-added models (VAMs) measuring the causal effects of individual schools. The first...

Conventional value-added models (VAMs) compare average test scores across schools after regression-adjusting for students' demographic characteristics and previous scores. This paper tests for VAM bias using a procedure that asks whether VAM estimates accurately predict the achievement...

A longstanding and influential view in U.S. correctional policy is that "nothing works" when it comes to rehabilitating incarcerated individuals. We revisit this hypothesis by studying an innovative law-enforcement-led program launched in the county jail of Flint, Michigan: Inmate Growth...