« Back to Results

Competition and Productivity in Health Care Markets

Paper Session

Saturday, Jan. 7, 2023 10:15 AM - 12:15 PM (CST)

New Orleans Marriott, Mardi Gras Salon F, G & H
Hosted By: American Society of Health Economists
  • Chair: Kate Bundorf, Duke University

Adverse Selection, Price Competition, and Limited Entry in Health Insurance Markets

Mark Shepard
,
Harvard University
Edward Kong
,
Harvard University
Timothy Layton
,
Harvard University

Abstract

Adverse selection is a classic insurance market failure known to disadvantage (or "unravel") generous contracts relative to stingy ones. We show that when differentiated insurers compete on price, adverse selection can also unravel all competition, even among similar-quality firms -- leading to few surviving competitors and in the extreme, natural monopoly. Adverse selection generates strong incentives for insurers to undercut each other's prices to "cherry pick" price-sensitive low-cost consumers. As with fixed costs in standard natural monopoly models, competition makes it challenging for multiple firms to sustain equilibria with prices that cover average costs. We show the empirical relevance of strong undercutting incentives using subsidy-driven price variation and a structural model of competition in Massachusetts' health insurance exchange. Our analysis suggests a new rationale for policies mitigating adverse selection: Without them, the market devolves to natural monopoly; with them, the market can sustain robust insurer participation.

What Does a Public Option Do? Evidence from California

Evan Saltzman
,
Emory University
Katie Leinenbach
,
Emory University

Abstract

Stimulating competition in health insurance markets as a mechanism for expanding health care access has been a longstanding objective of U.S. health care policy. One of the leading proposals for enhancing competition is the inclusion of a ``public option'' that competes with private insurance firms.

In this paper, we study how a public option affects the market equilibrium. A thorough analysis of the public option is challenging because it requires understanding how (1) consumers value public plans relative to plans sold by private firms and (2) the public firm will price. Our study overcomes these challenges by exploiting consumer-level data from the California ACA exchange where some, but not all consumers have access to a public option sold by their county governments. Heterogeneity in exposure to the public option across county markets and time provides identifying variation.

We estimate consumers' valuation of the public option. We find consumers are willing to pay an average of $39 more in premiums (9.5% of the average premium) for a private plan relative to a public plan, controlling for observables such as premiums and plan generosity. This average masks substantial heterogeneity; older, higher-income, and non-Hispanic White consumers have a much higher willingness-to-pay for a private plan.

We evaluate how well alternative objective functions for the public firm reconcile with our data. We consider several public firm objectives, including maximization of weighted social welfare. We find the public firm places substantial weight on consumer welfare, particularly that of disadvantaged subpopulations.

Finally, we find the public option reduces premiums & enhances social welfare. Product differentiation allows the market to support both public and private firms. Our work therefore provides support for the public option as a means of stimulating competition in health insurance.

The Corporatization of Hospital Care

Atul Gupta
,
University of Pennsylvania
Elena Andreyeva
,
Texas A&M University
Catherine Ishitani
,
University of Pennsylvania
Ben Ukert
,
Texas A&M University
Gosia Sylwestzrak
,
Anthem Inc

Abstract

Between 2000 and 2020, the share of US hospital capacity at independent hospitals dropped by half – a rapid corporatization of hospital care. The welfare effects of corporate ownership are theoretically ambiguous, leading to an important policy debate. We study this question using novel, patient-level claims data from one of the largest commercial health insurance firms in the US and the universe of hospital discharges from New York. We study 117 hospital deals between 2012 and 2018 using staggered differences-in-differences, comparing trends for independent hospitals newly acquired by systems to hospitals that remained independent. System ownership increases the revenue of the acquired hospital through both higher prices and greater treatment intensity, while simultaneously reducing operating expenses and staffing levels. However, we also detect a robust increase in hospital-wide readmission rates, suggesting a potential trade-off between efficiency and quality. We estimate that the increase in operating efficiency outweighs the cost to society of the quality decline across a range of estimates.

The Productivity of Professions: Evidence from the Emergency Department

Yiqun Chen
,
University of Illinois-Chicago
David Chan
,
Stanford University

Abstract

Professions play a key role in determining the division of labor and the returns to work. This paper studies the productivity difference between two distinct professions performing the same job---physicians and nurse practitioners (NPs)---but with stark differences in training and pay. Using data from the Veterans Health Administration and quasi-random variation in patient probability of being treated by physicians versus NPs in the emergency department, we find that, compared to physicians, NPs significantly increase patient length of stay (by 11 percent) and medical costs (by 7 percent). Despite higher medical resource use, NPs achieve less favorable patient outcomes: they increase patient 30-day preventable hospitalization rate by 20 percent. In unpacking mechanisms, we find evidence suggesting channels related to lower human capital among NPs relative to physicians, including that: lower experience widens the NP performance gap relative to physicians, NPs exhibit prescription thresholds consistent with lower skill, and NPs are likelier to gather information from external sources than are physicians. Finally, our estimation suggests a net increase in medical costs with the use of NPs, even if accounting for NP salaries that are half as much as physician salaries.

Discussant(s)
Shooshan Danagoulian
,
Wayne State University
W. David Bradford
,
University of Georgia
Michael Chernew
,
Harvard University
Stuart Craig
,
Yale University
JEL Classifications
  • I1 - Health