Research

The Effects of Dialysis Firm Consolidation on Patient Health 

(with Adrian Rubli and Francisco Garrido)

Job Market Paper

Like much of healthcare, the U.S. dialysis industry has grown increasingly consolidated, yet evidence on the mechanisms and magnitude of consolidation's impact on patient health remains limited. Using 30 years of rich administrative data, we show that mergers trigger facility closures and short-term disruptions that harm patients. Patients receive fewer dialysis sessions and mortality rises by 700 deaths per 100,000. Over the longer term, patient health improves, with fewer hospitalizations, ICU stays, and blood transfusion events, even as effects on laboratory biomarkers remain mixed. Two concurrent Medicare policy changes affecting dialysis do not meaningfully alter these impacts. A cost-benefit analysis reveals that the short-term health costs of consolidation outweigh any long-term benefits.

Health Impacts of Federal Pandemic Aid to State and Local Governments 

(with Jeffrey Clemens)

Under Review

Media: VoxEu, NBER Digest

The COVID-19 pandemic led to unprecedented levels of federal transfers to state and local governments. Did this funding impact population health? To answer this question, we leverage the fact that U.S. states that enjoy excess representation in Congress received substantially more fiscal assistance than did relatively underrepresented states. We find that the aid driven by excess representation had substantial impacts on population health. For each $1,000 increase in federal fiscal aid per state resident, we estimate that states experienced 38 fewer deaths from all causes per 100,000 residents from 2020 through 2022, of which 2/3 came from reductions in COVID-19 mortality. Additional aid also reduced rates of COVID-19 related hospitalizations and emergency room visits, though not in the total number of positive cases detected. Plausible mechanisms for these improved outcomes include substantially higher rates of COVID-19 testing and moderately higher COVID-19 vaccination rates. Medicaid enrollments and hospital capacity do not appear to play substantial mediating roles. Our robustness analyses provide evidence that the effects we estimate cannot be explained by pre-existing mortality trends, by the pandemic's differential impacts on relatively dense vs. rural areas, or by the pandemic's differential impacts on populations with more elderly individuals or with higher prevalence of chronic conditions. The mortality impacts we estimate were substantially greater for non-Hispanic Black Americans than for non-Hispanic White Americans, such that federal funds are associated with a reduction in population-wide health disparities over the course of the pandemic.

Association Between the Growth of Accountable Care Organizations and Physician Work Hours and Self-Employment

(with Douglas Staiger, David Auerbach, Peter Buerhaus and Lucy Skinner)

JAMA Network Open, 2018

Media: Medscape, Healthcare Dive, Healthcare Economist Blog, AJMC, Haymarket 

Accountable care organizations (ACOs) were created as a provision of the Affordable Care Act in 2010 where a group of physicians, hospitals and other healthcare professionals collaborate to take responsibility for patient care and cost savings. The share of the population covered by accountable care organizations (ACOs) is growing, but the association between this increase and physician employment is unknown. In this paper we investigate the association between the growth of ACOs and changes in physician work hours, probability of being self-employed, and probability of working in a hospital. Using data from the ACS covering the years 2011 to 2015, we use  a cross-sectional fixed-effects design to compare changes in physician employment in hospital referral regions with high vs low ACO growth.  We find that a 10–percentage point increase in ACO enrollment in a hospital referral region is associated with a statistically significant reduction of 0.82 work hours in men and a decrease of 2% in the probability of all physicians being self-employed. The association with self-employment is strongest in physicians aged 50 to 69 years, who are also more likely to work in a hospital. These results suggest that ACOs may affect physician employment patterns.

Minimum Wages, Medicaid Expansion and the Mix of Benefits 

Under Review

Minimum wage increases are intended to raise wages and improve worker well-being. However, in a complex policy ecosystem where workers also receive government transfers and firm-provided fringe benefits, their impact on well-being likely hinges on interactions with other policies and may differ across states due to variation in broader policy contexts. I examine the empirical effects of state minimum wage increases between 2013 and 2016 on low-income workers’ participation in Medicaid. I explore the role of the Affordable Care Act Medicaid expansion in generating heterogeneity in these effects. I find that, contrary to mechanical predictions, minimum wage hikes increase Medicaid take-up in both expansion and non-expansion states. The most plausible mechanism behind increased participation during expansion is the crowd-out of employer-sponsored insurance (ESI), as workers substitute Medicaid for less generous ESI plans offered by firms in exchange for higher wages. In non-expansion states, gains in Medicaid participation may result from two primary mechanisms. First, workers lose benefits from other means-tested programs due to income ineligibility. Second, other members of their households experience job losses. Workers may exit uninsurance and enroll in Medicaid to offset reductions in total household resources, including their own government benefits and fringe benefits tied to others’ employment. Moreover, minimum wage increases do not appear to generate wage gains large enough to trigger Medicaid eligibility loss, and may induce strategic labor supply responses. Overall, the interaction between minimum wage increases and the Medicaid expansion affects the mix of benefits workers receive from both the government and firms.

Understanding Cross-State Variations in Medicaid

Enrollment During and After the COVID-19 Pandemic

(with Jeffrey Clemens and Helena Detering)

Due to the implementation and unwinding of a “continuous coverage requirement”, the COVID-19 pandemic gave rise to the most dramatic changes in Medicaid enrollments in the program’s history. Nationwide, enrollments rose by 23 million individuals from February 2020 through March 2023, then declined by roughly 15 million by late 2024. Notably, changes in per capita enrollments varied dramatically across the country, with several states experiencing net declines and several states seeing their enrollments rise, on net, by more than 5 percent of their populations. Through a mix of descriptive and causal analyses, we explore several hypotheses regarding the possible causes of these variations. We find that a wide range of provisions designed to ease the frictions of the continuous coverage provision’s winding down have surprisingly little predictive power.  Similarly, we find that variations in federal aid to state and local governments has no predictive power, suggesting that liquidity constraints had little influence on states’ management of Medicaid enrollments during this period. Variations in political preferences, as proxied by Trump’s 2016 vote share, have modest predictive power within the unwinding episode. Finally, states that enacted Medicaid expansions during the pandemic experienced relatively large net gains in enrollments. The baseline generosity of states’ eligibility thresholds also predicts relatively large run-ups and net increases in enrollments.