Impact of National Health Insurance on Entry and Competition in Remote Areas, Evidence from Taiwan's Dental Market

• Author: Jitian Sheu
• Published date: 01 October 2017
• DOI: http://doi.org/10.1111/1468-0106.12137
• Date: 01 October 2017

IMPACT OF NATIONAL HEALTH INSURANCE ON ENTRY

AND COMPETITION IN REMOTE AREAS, EVIDENCE

FROM TAIWAN’S DENTAL MARKET

JITIAN SHEU*Chang Gung University

Abstract. This paper assesses the change in the supply of dental clinics after the implementation of

Taiwan’s National Health Insurance (NHI). Even though the government has enforced lower prices

of treatment since the implementation of NHI, the increase in use could actually result in higher rev-

enue for dentists. The higher profit could prompt dentists to attempt to enter markets that were pre-

viously unprofitable. This paper tested if NHI increased the dental clinic supply in geographically

isolated areas by investigating whether the required minimum market size was reduced. We found

that following the implementation of NHI, the per-dental clinic minimum market size significantly

decreased. In addition, the decreased minimum market sizes were enabled by a corresponding in-

crease in variable profits which is equal to the difference between price and average variable costs.

This implies that the NHI-enlarged health care demand compensates for the possible losses of dental

clinics due to price regulation. Furthermore, the results also suggest that the post-NHI dental market

becomes more competitive when there is a second entrant in the market, and the level of competition

approaches perfect competition when three and more dental clinics serve one market.

1. INTRODUCTION

One major objective of implementing National Health Insurance (NHI) in

Taiwan is to provide equal access to adequate health care for all citizens.

1

According to Chiang (1997), Taiwan’s NHI incorporates three features for

achievement of this objective: (i) compulsory universal coverage; (ii) uniform

comprehensive benefit packages; and (iii) the financing of NHI through a pay-

roll tax with a heavy governmental subsidy. The first feature of NHI dramati-

cally increased the insured population from 57% in 1994 to 92% by the end

of the inaugural year, 1996 (Chiang, 1997).

2

The NHI benefit package includes

ambulatory care, inpatient care, emergency care, prescription drugs, X-ray and

laboratory tests, rehabilitation, mental illness treatment, dental care, specified

preventive services and home care. According to NHI law, insurance premium

rates for all benefits should be identical and set on an actuarial basis. In addi-

tion, most of the premium is contributed via government subsidies. These three

*Address for Correspondence: Chang Gung University, Taiwan. E-mail: jtsheu@mail.cgu.edu.tw.

The research support for this work provided by the Ministry of Science and Technology (NSC94-

24158-H-182-001 and MOST 104-2410-H-182-030) in Taiwan is gratefully acknowledged. I would

also like to thank Randall Ellis, Albert Ma, Thomas McGuire, Chih-Nan Chen, Hsien-Ming Lien

and Martino De Stefano. All provided excellent and helpful comments and suggestions.

1

In 1990, the Council for Economic Planning and Development completed the fundamental docu-

ment regarding the implementation of a national health system. As stated in the CEPD report

(CEPD, 1986), there are three major objectives of the NHI: (1) to provide equal access to adequate

health care for all citizens in order to improve the health of the people; (2) to control health care costs

at a reasonable (or socially affordable) level and (3) to promote efficient use of health care resources.

2

The implementation of NHI was on 1 March 1995.

Pacific Economic Review, 22: 4 (2017) pp. 483–509

doi: 10.1111/1468-0106.12137

© 2016 John Wiley & Sons Australia, Ltd

bs_bs_banner

features provide all insured people with an opportunity to access required

health care.

National Health Insurance is expected to reduce health disparities through

universal insurance coverage; however, there may not be a sufficient supply of

providers to successfully carry out this endeavour. Achieving this goal is even

more unlikely in some rural and remote areas, which were previously unprofit-

able for providers. Before the adoption of NHI, most of the uninsured resided

in small towns or remote areas.

3

The number of physicians in these areas is far

below that in urban areas. If the implementation of NHI does not induce enough

of a supply response, the policy goal of equal access cannot be achieved. Hence,

from a policy perspective, this issue is critical because NHI is expected to solve

health-care inequalities between urban and rural areas.

Several studies have assessed the geographic distribution of medical resources

following the implementation of NHI in Taiwan. Huang (2000) used logit re-

gression and ordinary least squares (OLS) and found that after the establishment

of NHI, the change in the number of dentists was related to a town’s population.

Huang found that towns with larger populations attracted more dentists after

the implementation of NHI than towns with smaller populations. Because most

remote areas are small-sized towns, Huang’s results implicitly showed that NHI

does not induce the entry of dentists into rural and/or remote areas. However,

Lin (2002), taking into account spatial autocorrelation across towns in Taiwan

and using spatial statistics analysis, found that primary care providers tend to

gradually spread out to remote areas, while specialists choose metropolitan

areas. Both studies used individual geographic towns as units. For the identifica-

tion of the NHI effect, both papers used reduced form regressions. However,

their methods may introduce estimation biases. Both studies treated two adja-

cent towns as two independent markets, which may underestimate the NHI ef-

fect. Furthermore, the number of firms is the consequence of market

competition, and any NHI effect would be overestimated if one does not con-

sider the competition. Yet, neither study accounted for this factor. Finally, a

new dentist needs 6 years of in-school training before he or she can enter the

market. Hence, the number of new dentists in any year was determined 6 years

in advance. This 6-year lag causes another estimation bias because a high supply

of dentists entering the market may have been predetermined before NHI, yet

revealed only after NHI was established. This might introduce an overestima-

tion of the NHI effect if one just compares the difference between these two time

periods.

Using a simple and general entry condition, this paper has estimated the min-

imum market size required to support a given number of health providers be-

fore and after NHI. In addition, in order to avoid estimation bias, instead of

using all official townships in Taiwan, this paper has selected the geographically

isolated towns delineated by Bresnahan and Reiss (1991). To use only

3

Before the implementation of NHI, there was no private health insurance in Taiwan. The three

major social insurance programs providing health care coverage were Government Employer Insur-

ance, Labor Insurance, and Farmer’s Insurance.

J. SHEU484

© 2016 John Wiley & Sons Australia, Ltd