Typhoons and Opportunistic Fraud: Claim Patterns of Automobile Theft Insurance in Taiwan

• Author: Kili C. Wang,Tsung‐I Pao,Larry Y. Tzeng
• Published date: 01 March 2014
• DOI: http://doi.org/10.1111/j.1539-6975.2012.01498.x
• Date: 01 March 2014

© The Journal of Risk and Insurance, 2014, Vol. 81, No. 1, 91–112

DOI: 10.1111/j.1539-6975.2012.01498.x

91

TYPHOONS AND OPPORTUNISTIC FRAUD:CLAIM

PATTERNS OF AUTOMOBILE THEFT INSURANCE IN

TAIWAN

Tsung-I Pao

Larry Y. Tzeng

Kili C. Wang

ABSTRACT

We present evidence to support the existence of opportunistic fraud in the

automobile theft insurance market in Taiwan.After encountering a typhoon

hit, the insuredwho purchase automobile theft insurance but do not purchase

typhoon/flood insurance tend to have a significantly higher probability of

filing a total theft claim than other insured. The above relationship exists

mainly in places affected by typhoons. Such evidence does not exist in partial

theft claim. These claim patterns of automobile theft insurance provide us

with strong evidence that supports the existence of opportunistic fraud in

the market.

INTRODUCTION

Fraud has been a serious threat to insurance companies throughout the world.1

Cummins and Tennyson(1992, 1996), Tennyson (1997), Carroll and Abrahamse (2001),

and Dionne and Gagn´

e (2001, 2002) all provide evidence regarding the existence of

fraud in the automobile insurance market. Meanwhile, Dionne (1984), Sparrow (1996),

Tsung-I Pao is a Master’s Degree Holder at the Graduate Institute of Finance and Insurance,

Tamkang University. Larry Y. Tzeng is a Professor in the Department of Finance, National

Taiwan University and a Research Fellow at the Risk and Insurance Research Center, College

of Commerce, National Chengchi University.Kili C. Wang is an Associate Professor in the De-

partment of Insurance, Tamkang University and a Research Fellow at the Risk and Insurance

Research Center,College of Commerce, National Chengchi University. Kili C. Wangcan be con-

tacted via e-mail: kili@mail.tku.edu.tw.The authors gratefully acknowledge helpful comments

received from the two anonymous referees and the participants at the 2011 annual meetings of

the American Risk and Insurance Associations.

1The Automobile Insurers Rating Bureau of Massachusetts (1990), Florida Insurance Research

Center (1991), Weisberg and Derrig (1991, 1992), and Derrig et al. (2006) find fraud to exist in

many U.S. states. Foppert (1994) points out that the annual cost of fraud in the United States

amounts to nearly $70 billion. Dionne and Belhadji (1996) and Caron and Dionne (1999) find

fraud in Canada. Medza (1998) points out that the Insurance Bureau of Canada has estimated

the total annual cost of fraud to be about $2 billion.

92 THE JOURNAL OF RISK AND INSURANCE

Hyman (2001, 2002), and Sulzle and Wambach(2005) provide evidence of fraud in the

health insurance market. On the other hand, Dionne and St-Michel (1991) and Butler

et al. (1996) provide such evidence on the basis of workers’ compensation insurance.

Following this line of research, this article intends to provide other empirical evidence

to support the existence of opportunistic fraud in the automobile theft insurance

market.

Our article is closely related to that of Dionne and Gagn´

e (2002) who identify op-

portunistic fraud by finding that the insurance contracts with replacement cost en-

dorsement had a higher probability of total theft claims and that those claims were

particularly concentrated near the end of the policy period. By following their idea

that opportunistic fraud may result in a particular time pattern insofar as claims are

concerned, this article intends to investigate a special pattern of claims in the auto-

mobile theft insurance market in Taiwan. Although Dionne and Gagn´

e explore the

time pattern of opportunistic fraud near the end of the policy period, we study the

particular claim pattern of opportunistic fraud when the insured has encountered a

specific event, namely, a huge typhoon hit. In their article, Dionne and Gagn´

e iden-

tify opportunistic fraud by comparing claim patterns between insurance contracts

both with and without replacement cost endorsements. In this article, we examine

the opportunistic fraud problem between the insured who purchase only automobile

theft insurance and the insured who purchase both automobile theft insurance and

typhoon/flood insurance.

In Taiwan,automobile theft claims have sometimes experienced a sharp increase right

after typhoons, which caused floods. Why are the occurrences of automobile theft and

typhoon incidents correlated? There are two competing hypotheses. First, many of

the vehicles that have theft insurance have been damaged by flooding. This may

have given some opportunistic vehicle owners with only theft insurance (without

typhoon/flood insurance) an incentive to lie to the insurance company by stating

that their vehicle had been stolen.2They thereby filed a total theft claim. Second,

the typhoon has resulted in many flood-damaged cars being in need of repair. The

demand for automobile parts on the black market thus dramatically increases at such

times. This may have increased the incentive for car thieves to work particularly hard

after a typhoon hit and resulted in a higher rate of automobile theft claims.

The way we disentangle opportunistic fraud from ordinary theft is through the type

of insurance coverage that the insured had when he or she has encountered a huge

typhoon hit. If the insured vehicles were damaged by a typhoon, those insured

with only theft insurance coverage but without typhoon/flood insurance coverage

could not make a claim against the losses incurred due to the typhoon. Hence, the

opportunistic fraud could have been perpetrated by those with only automobile theft

insurance who did not purchase typhoon/flood insurance, in contrast to those with

both types of coverage. On the other hand, the increase in theft caused by the increase

in demand for automobile parts on the black market provided no basis for pointing

at either one of these two groups of insured. This therefore provides us with a good

2In Taiwan,car owners can voluntarily choose to purchase automobile theft insurance as well as

typhoon/flood insurance. Accordingly, some of the insured with automobile theft insurance

also have coverage for typhoon/flood insurance simultaneously,while some of them do not.