# Samuel Moy

Data Scientist, Uber

Dissertation Title:  "Essays on Regulation in Health Care Markets"This dissertation consists of three chapters studying regulation of payer and provider health care markets.

The first chapter studies how the market definitions of the Affordable Care Act health insurance exchanges affect the function of these insurance markets. Using a county-level matching design among states that use the federally facilitated marketplace, I show that counties included in the same market as large metropolitan areas tend to offer lower premiums and deductibles; however, this comes at the cost of narrower provider networks. Given the nature of premium subsidies on the exchanges, the benefits of lower premiums and deductibles primarily accrue to the federal government. These effects are consistent with a form of market "unraveling": the presence of greater insurer competition in larger markets creates conditions under which no insurer can find it profitable to offer a plan that is generous on non-price dimensions.

In the second chapter, I develop a micro-founded structural model of the exchanges in the spirit of (Ho and Lee 2017). I calibrate the model using data from Commonwealth Care - the Massachusetts precursor to the ACA; I then use the model to simulate a counterfactual policy change to the geographic insurance market definitions. My simulations show that larger markets significantly attenuate hospitals' market power by altering insurers' threat points, leading to lower prices paid to hospitals. When the state of Massachusetts is divided into five separate rating areas, the average price paid for an inpatient admission is $15,333, whereas under a single state-wide rating area, the average price paid is$11,085. While large and striking, these price differences are actually less extreme than the observed differences between commercial and Medicare inpatient prices in Massachusetts. Taken literally, my results imply that aggregating rating areas may be an effective means to curb health care provider market power.

The third chapter studies the effects of clinical outsourcing on quality of care. I identify more than 250 distinct events for which a hospital's emergency department transitions from being independent to being managed by one of two larger, national staffing companies. Using an event study regression framework, I find that the "treated" hospitals improve with respect to measures of thoroughness and timeliness of care. I also find an increased volume of patients admitted to the hospital from the ED. However, I am unable to detect statistically significant changes in clinical outcomes, such as readmission rates and mortality, nor am I able to detect changes in patient satisfaction. These results suggest that outsourcing of clinical hospital departments may lead to improved efficiency of care delivery; however, these benefits may not offset their negative effects, such as higher prices and higher rates of out-of-network billing.