Date of Degree

9-2018

Document Type

Dissertation

Degree Name

Ph.D.

Program

Economics

Advisor

David Jaeger

Committee Members

Michael Grossman

Partha Deb

Subject Categories

Health Economics | Industrial Organization

Keywords

vertical integration, full-line forcing, bundling, bargaining, hospitals, health insurance

Abstract

This dissertation examines the welfare implications of vertical restraints in the U.S. health care market. I focus on two phenomena that influence vertical relationships between hospitals and insurers: vertical integration and vertical bundling. Both of these practices have potential efficiency-inducing and anti-competitive effects from a theoretical standpoint, making welfare implications ambiguous. I analyze their impact on welfare in an empirical setting. This task requires structurally modeling the market by using data on hospitals, insurers, and consumers, and simulating counterfactual worlds free of vertical restraints.

I construct my dataset by combining data from multiple sources. Hospital characteristics come from the American Hospital Association (AHA) Annual Survey of Hospitals. I create an insurer dataset using three sources: Atlantic Information Services (AIS) Directory of Health Plans, WEISS Ratings Guide to Health Insurers, and National Committee for Quality Assurance (NCQA) Report on Health Plan Rankings. Patient level data comes from State Inpatient Databases published by Agency for Healthcare Research and Quality (AHRQ), and Public Patient Discharge data published by California Office of Statewide Health Planning and Development (OSHPD). I observe the universe of discharges from all hospitals in eight states. Finally, I manually collect hospital networks offered by insurers from individual insurers' websites.

Using this rich dataset, I separately model every component of the market, and use my results to predict the changes in actions of all agents under counterfactual simulations. In particular, I estimate structural models of hospital and insurer demand along with a bargaining model, and use results in counterfactual policy experiments that prohibit vertical restraints. Following the simulation of the counterfactual world, I measure the change in welfare. I use discrete choice models of demand when analyzing hospital and insurer markets, and a Nash bargaining framework when modeling hospital-insurer negotiations.

In the first chapter, I investigate how vertical integration between hospitals and insurers affects market outcomes and welfare. Hospital-insurer consolidation is expected to benefit consumers through improved coordination and reduced costs, but can also lead to a decline in consumer welfare through anti-competitive outcomes such as market foreclosure and barriers to entry. Using regression analysis, I demonstrate vertical integration is associated with lower premiums and higher quality. However, vertically-integrated entities engage in anti-competitive activities by denying their competitors access to their hospitals (upstream foreclosure) and health plans (downstream foreclosure).

Since I find evidence that both effects exist, I structurally model the market and simulate a counterfactual scenario to assess the overall impact on welfare. In this policy experiment, I remove vertical integration from the market altogether and measure changes in consumer welfare and producer surplus. Overall, consumers benefit from vertical integration, although the majority of hospitals and insurers are better off in its absence. Banning exclusionary restrictions gets rid of market foreclosure and gives access to vertically-integrated entities. As a result, insurers offer wider hospital networks but increase their premiums, which harms consumers. Producers, on the other hand, are better off in the absence of vertical integration as many hospitals and insurers enjoy higher profits driven by higher market shares and increased premiums. I also find evidence that vertical integration acts as an entry barrier to the downstream market due to the cost advantages vertically-integrated entities achieve through upstream foreclosure.

The second chapter analyzes bundling practices exercised by hospital systems in the negotiation process with insurers. Hospital systems offer full-line forcing contracts to ensure insurers carry all hospital system members in their networks. Vertical bundling can benefit consumers if it reduces unit cost for insurers, but it can also have anti-competitive consequences if bundled hospitals are included in insurers' networks at the expense of their rivals. I estimate reduced-form regressions that demonstrate efficiency gains present themselves through their impact on market coverage, but not premiums. I also find hospital systems are unable to use these contracts to gain leverage over their rivals. To quantify the overall impact on welfare, I estimate a structural model and simulate a counterfactual world free of full-line forcing contracts. This policy experiment breaks vertical bundling and allows system hospitals to negotiate individually. Results suggest that removal of full-line forcing contracts from the market leads to a $16 billion decline in overall welfare. A majority of insurers increase their profits upon removal of full-line forcing contracts, however many hospitals suffer losses. System hospitals lose the most since they no longer can use bundling to maximize their profits. Individual hospitals also lose as a result of increased competition in the upstream market. Consumers are worse off because they face higher premiums in the absence of vertical bundling.

This dissertation provides empirical evidence on potential welfare-improving effects of vertical restraints in the health care market. Results presented here imply increasing vertical and horizontal consolidation in health care is not completely detrimental to consumers, and can benefit them through unforeseen channels.

This work is embargoed and will be available for download on Wednesday, September 30, 2020

Graduate Center users:
To read this work, log in to your GC ILL account and place a thesis request.

Non-GC Users:
See the GC’s lending policies to learn more.

Share

COinS