— Job Market Paper

Financing Care or Financialization?: An Empirical Analysis of Tax-Exempt Bond Use In Nonprofit Hospitals

Municipal bonds are loans that nonprofit hospitals use to purchase medical equipment, build medical facilities, and invest in other major infrastructure projects that increase access to care. These bonds are tax-exempt, and associated with lower borrowing costs than most commercial debt. Municipal bonds serve as a primary source of financing for the nonprofit health care sector. Despite this heavy reliance, there is almost no evidence on which types of hospitals use these bonds, how these bonds are used, and whether this financing expands hospitals' capacity to provide care.

We used 2010–2023 IRS Form 990 data to identify nonprofit hospitals that used municipal bonds and describe their characteristics, including size, location, and wealth. We calculated the total amount of annual debt that nonprofit hospitals together take on in municipal bonds. To assess the impact of use of municipal bonds, we examined changes in land, buildings, and equipment after hospitals borrow using a staggered difference-in-differences event-study design. We also measured changes in hospital stock portfolios and expenses like top executive compensations to assess whether the bonds were used for alternative reasons.

— Job Market Paper II

Go to a Busy Hospital: Stroke Volume and 30-Day Survival in Medicare Patients

Stroke is a leading cause of death and disability in the United States, with around 795,000 people experiencing a new or recurrent stroke each year. The advent of endovascular thrombectomy (EVT), supported by multiple landmark RCTs published in 2015, transformed acute ischemic stroke care, and has since driven a rapid reorganization of the stroke care system: the share of AIS patients treated at EVT-capable hospitals has grown from under 40% to nearly 60% between 2014 and 2019, and EVT use has more than quadrupled over the same period. Ambulance routing protocols in many U.S. states now direct suspected stroke patients to EVT-capable hospitals on the assumption that access to thrombectomy improves outcomes. However, it is unclear whether access alone is sufficient. Not all EVT-capable hospitals perform the procedure at equal volume, and roughly 35% of EVTs are performed at hospitals doing fewer than 25 procedures per year.

Using 2016–2019 Medicare fee-for-service claims, we study two related questions among acute ischemic stroke patients: whether admission to an EVT-capable hospital improves 30-day survival compared with a non-EVT hospital, and whether higher annual EVT volume predicts further survival gains and at what threshold. We use propensity score overlap weighting and an instrumental variable approach based on differential travel distance to address selection into EVT-capable hospitals, and sweep annual EVT thresholds from 25 to 70 procedures per year to identify where volume benefits emerge. We find that access to EVT-capable hospitals reduces 30-day mortality, but that the survival benefit strengthens substantially with hospital volume, with a meaningful threshold emerging around 50 procedures per year — a level that a large share of certified stroke centers do not consistently reach.

— Work in Progress

The Effect of Investment Gain and Loss on Hospital Financial and Operational Activities, 2010–2023

— Work in Progress

The Impact of Revenue Cycle Outsourcing on Hospital Financial Performance and Medical Debt