Health insurance, market power, and social welfare

• DOI: http://doi.org/10.1111/ijet.12137
• Author: Run Liang,Hao Wang
• Published date: 01 December 2017
• Date: 01 December 2017

doi: 10.1111/ijet.12137

Health insurance, market power, and social welfare

Run Liang∗and Hao Wang†

This paper considers the welfare effects of health insurance. With a rather general model, we

show that the presence of traditional health insurance makes consumers worse off when the

marginal cost of health care is low enough, while always making the health care provider better

off. The presence of traditional health insurance increases social welfare, but cannot achieve the

socially optimal outcome. Furthermore, we consider a marketst ructurecalled “integr atedhealth

insurance” and show that in a health care market where a firm provides not only health care but

also health insurance, the socially optimal outcome can be achieved.

Key wor ds health care, health insurance, market power,social welfare, vertical integration

JEL classification D6, I1, L1

Accepted 7 June2016

1 Introduction

Covered by health insurance, consumers may face usage rates that arelower than the marginal costs

of health care. Hence, insurance tends to induce excessive demand for health care (e.g. Crew 1969).

This phenomenon is often referred to in the health economics literature as “moral hazard.” Pauly

(1968) suggests that there is a “prisoner’s dilemma” in health care markets: individual may well

recognize that ‘excess’ use of medical care makes the premium he must pay rise. No individual

will be motivated to restrain his own use” (p. 534).1Given the market structure of the health care

industry, health insurance boosts consumer demand for health care and thus drives up the price of

services. Higher health care prices, in turn, increase the financial risk faced by patients, and hence

stimulate more insurance purchasing. Therefore, the equilibrium level of insurance coverage tends

to be excessive from the perspectiveof consumers or even society.

Empirically, using panel data for individual states of the United States for the years 1959–1965,

Feldstein (1973) finds that American families were generally over-insured against health expenses.

If insurance coverage were reduced, the utility loss from increased risk would be outweighed by

the gain due to lower health care prices and reduced purchases of excess care. Feldman and Dowd

(1991) present a new estimate of the welfare loss of excess health insurance, using the Rand Health

Insurance Experiment results regarding price elasticity and consumer risk-aversion. They find that,

∗Institute for Advanced Research,Shanghai University of Finance and Economics, Shanghai, China.

†National School of Development, PekingUniversity, Haidian District, Beijing, China. Email: hwang@nsd.pku.edu.cn

The two authors have contributed to the paper equally. The authors thank the Ministry of Education of China

(16JJD790002) for financial support. The authors thank Ake Blomqvist, Simon Chang, Yuk-fai Fong, Huanxing Yang,

Lixin Ye,Zhong Zhao, an anonymous referee, and others for helpful comments. They also thank Sandra Gain for language

editing. The authors are responsible for any errors.

1Pauly also comments that “the responseof seeking more medical care with insurance than in its absence is a result not of

moral perfidy, but of rational economic behavior” (p.535).

International Journal of Economic Theory 13 (2017) 427–442 © IAET 427

Health insurance and social welfare Run Liang and Hao Wang

with Feldstein’s (1973) “most likely” parameter values, raising the average co-insurance rate from

0.33 to 0.67 would produce a net gain of $27.8 billion in 1984 prices in the private hospital sector.

And with the Rand parameter values, the range of welfare gain is from $33.4 billion to $109.3 billion

in 1984 prices, which is more significant.

Theoretically, Chiu (1997) and Vaithianathan (2006) show that (under symmetric information)

competitivehealth insur ancereduces consumers’ welfare when the supply of health care is sufficiently

price-inelastic. Both papers assume that consumers are ex ante identical and have an equal probability

of getting sick. In Chiu (1997), health care providers act as perfect agents for consumers, while in

Vaithianathan (2006) the providers are profit-maximizing firms engaging in Cournot competition.

Both papers find overpriced health care under health insurance.

Wigger and Anlauf (2007) point out that these welfare analyses ignore the profits of the health

care industry. They reconsider the question of whether consumers purchase too much insurance

from the perspective of social welfare. In their model, a drug producer is monopolistic and health

insurers are competitive. A risk-averse individual can be either healthy or sick. A drug can be used

to reduce the individual’s loss from illness. Wigger and Anlauf conduct marginal analyses on the

equilibrium outcome, leading to two major propositions. The first is that when the marginal cost

of a drug is small enough, a marginal increase in the co-insurance rate starting from the market

equilibrium level raises consumer welfare. The second is that when the marginal cost of a drug is

positive, a marginal increase in the co-insurance rate starting from the market equilibrium level

lowers social welfare. These findings imply that although the consumer welfare criterion suggests

that consumers purchase too much insurance coverage, the social welfare criterion suggests that

consumers should purchase even more coverage in equilibrium.

The present paper revisits the welfare effects of health insurance. The contributions are threefold.

First, this paper presents a technical improvement over most insurance models in the literature. In

contrast to the two-state models in Chiu (1997), Vaithianathan (2006), Wigger and Anlauf (2007),

and many others, we offer an insurance model that allows a continuumof health states. In a two-state

model of health care, consumers have demand for health care when and only when they get sick.

Insurance is against the possible loss arising from a specific illness. In our model, a consumer has

a higher demand for health care when she finds herself in a worse health state. The design of an

insurance policy has to take into account all possible states.

Second, instead of conducting the marginal analyses of Wigger and Anlauf (2007), we conduct

global analyses on the welfare effects of health insurance. Specifically, Wigger and Anlauf (2007)

demonstrate how a marginal change in co-insurance rate from the market equilibrium affects con-

sumers or society, while we consider how the “presence” of health insurance affects consumers or

society.The timing s ofthe games are also different. Wigger and Anlauf (2007) assume that consumers

choose an insurance contract offered by insurers before the drug provider chooses a price, while we

assume that consumers make their decisions after the health care price and insurance contracts are

determined.

The third contribution may be the most innovative one. Existing studies analyze various health

care industries, where health insurance is usually provided by independent insurance firms. Wesug-

gest a new perspective by considering not only traditional insurance, but also “integrated insurance”

in which a firm can provide both health care and health insurance. We show that with traditional

insurance, the socially optimal outcome cannot be obtained even with perfectly competitiveinsurers.

But with integrated insurance, the socially optimal outcome can be achieved.

The rest of the paper proceeds as follows. Section 2 depicts a model of the traditional health care

market. Section 3 characterizes the equilibrium of the tradition health care market. It examines how

the presence of traditional insurance affects the welfare of consumers and a health care provider.

428 International Journal of Economic Theory 13 (2017) 427–442 © IAET