Externalities From Driving Luxury Cars

• Author: Sangeun Han,Sojung Carol Park
• Published date: 01 December 2017
• DOI: http://doi.org/10.1111/rmir.12085
• Date: 01 December 2017

Risk Management and Insurance Review

C

Risk Management and Insurance Review, 2017, Vol.20, No. 3, 391-427

DOI: 10.1111/rmir.12085

PERSPECTIVE

EXTERNALITIES FROM DRIVING LUXURY CARS

Sojung Carol Park

Sangeun Han

ABSTRACT

Driving luxury cars creates negative externalities. Driving a luxury car increases

property damage liability insurance costs for all drivers due to the striking

differences in repair costs of luxury cars and nonluxury cars in Korea. In this

study, we estimate the externalities related to auto accidents involving luxury

cars by running a two-part model using unbalanced individual-level panel data

on insurance claims and characteristics of the insured party. We find evidence

of negative externalities in all of our results. To be specific, a 1 percentincrease

in luxury cars raises the property damage liability costs by 1.9–2.6 percent per

claim. The estimated nationwide increase in the cost of liability due to driving

of luxury cars in Korea is USD 139–196 million per year. This cost is shared by

all drivers nationwide.

INTRODUCTION

Picture this scenario: Two vehicles get into an accident, and under the tort system, the

at-fault driver must pay for the repair and injury costs of both cars. Where comparative

negligence is applicable, the degree of negligence contributed to the accident by both

drivers is determined and both pay for their portion of the losses. This comparative

negligence tort system is common in many countries and is considered to be quite

fair and well accepted by insurance adjusters of auto accident losses. However, it does

present some challenges.

If the repair costs of two cars involved in an accident are fairly similar, the compar-

ative negligence system works well. However, when the values of properties vary

significantly, this may not be the case. Consider an accident involving an expensive

car and an economy car. Suppose that the driver of the luxury car is 90 percent at fault

and the driver of the economy car is only 10 percent at fault for inadequate defensive

Sojung Carol Park is an Associate Professor of Finance, College of Business Administration,

Seoul National University,1 Gwanak-ro, Gwanak-gu, Seoul 151-916, South Korea, phone: 82-02-

880-8085; e-mail: sojungpark@snu.ac.kr. Sangeun Han is with Samsung Fire & Marine Insurance,

32nd Fl, 14 Seocho-daero, 74-gil, Seocho-gu, Seoul 06620 South Korea, phone: +82-02-758-7385;

e-mail: sangeun0311@gmail.com. The support from SNU Institute for Research in Finance and

Economics is gratefully acknowledged. The authors also thank audiences at the Insurance Risk

and Finance Research Centre (IRFRC) and the World Risk and Insurance Economics Congress

(WRIEC), the Korean Risk Management Society, the International Academy of Financial Con-

sumers, and the SNU Institute for Research in Finance and Economics seminar meetings. Finally,

the detailed and constructive comments of the anonymous referees were extremely valuable. All

errors are our own.

391

392 RISK MANAGEMENT AND INSURANCE REVIEW

driving. Both cars have minor scratches, but the repair costs are USD 5,000 and USD 100,

respectively. In this situation, the driver of the expensive car must pay USD 90 to his

counterparty, while the driver of the economy car is required to pay USD 500. If the re-

pair costs of the luxury car were equal to those of the economy car,the driver of the latter

would only need to pay USD 10. In another imaginary scenario, suppose that suddenly

many drivers in your vicinity decide to purchase luxury cars. The drivers of economy

cars have not changed their driving habits, but their liability costs have increased as

a result of the increased number of luxury cars on the road. However, exercising their

personal preference for driving luxury cars does not necessarily involve an increase in

liability costs for those who purchased them. This example illustrates the concept of

negative externality, which in this case results from driving luxury cars, a concept that

the tort system is not designed to address.

The scenario above is not purely imaginary. When Germany reunified, the East German

cars were then surrounded by more expensive cars, like Mercedes Benz; as a result,

their liability rates went up. On the other hand, owners of Mercedes Benz in the former

East Germany could enjoy relatively lower rates (Fl˚

am and Wolfstetter, 2015). A similar

phenomenon is occurring in Korea. The Korean government recently loweredthe import

tax on foreign cars and this has resulted in an annual increase of more than 20 percent in

the number of foreign cars on Korean roads since 2005. As the costs of repairing foreign

cars in Korea are more than triple those of domestic ones, an increasednumber of foreign

cars means increased heterogeneity in repair cost, which has in turn triggered debates

on the fairness of the auto liability insurance system.

This important externality has by and large been ignored in the literature, with the

exception of a recent theory paper by Fl˚

am and Wolfstetter (2015). In their model, they

link the liability rules to the choice of car and show that the tort system does not

incentivize drivers enough to lower the overall liability costs and make people choose

more expensive cars. That is, because the cost of repairing luxury cars is not fully

borne by the owner of the vehicle but is instead shared by all other drivers on the road

through liability costs in the tort system, the expense associated with repairing a luxury

vehicle may not deter the purchase of such cars enough. This creates controversy, as the

purchase decisions of affluent people may circuitously burden people with relatively

lower income in the tort system.

A no-fault system, on the other hand, may favor negligent drivers and increase acci-

dents. This moral hazard issue related to the no-fault system has been examined by

many empirical researchers (e.g., Cummins and Tennyson, 1996; Cummins and Weiss,

1999; Cole et al., 2012), but the issue of negative externality under the tort system has

not been examined yet. This article contributes to the literature by showing the existence

of negative externality and estimating the size of this externality, using Korean data.

Although the negative externality may exist anywhere as long as the repair cost hetero-

geneity is severe, the rapid change in the composition of car values in Korea provides a

natural experiment environment for this particular study.

Wouldthis negative externality of driving luxury cars really exist in practice? That is, do

liability costs really increase when some people purchase luxury cars? It is, in fact, not

that obvious. Due to their awareness of the high costs of repairing expensive cars, both

luxury car and economy car drivers may drive more defensively in an effort to reduce

losses. For example, drivers may keep their distance from luxury cars on the road. Such

EXTERNALITIES FROM DRIVING LUXURY CARS 393

behavior may reduce the frequency of accidents. This is possible in Korea because of its

peculiar auto bonus–malus system. In most countries, the bonus–malus coefficient only

depends on the frequency of accidents, whereas under the Korean system, the malus

coefficient also depends on the severity of the accident. That is, if a driver is involved

in an accident with an economy car, it is possible that the insurance premium next year

would not be affected, but if the driver is involved in an accident with an expensive

car, it is very likely that the premium next year will increase. Therefore, people may

drive extra carefully around expensive cars. In addition, the extra sturdiness and safety

features of luxury cars may reduce the bodily injury of luxury car drivers and passengers.

The bodily injury liability cost, on the other hand, may actually increase if the luxury

cars’ sturdiness or safety features causes more severe bodily injury of others on, as is

sometimes the case in accidents involving SUVs and pickup trucks (Meyer and Gomez-

Ibanez, 1981; White, 2004; Wenzel and Ross, 2005; Anderson, 2008; Karaca-Mandic and

Lee, 2014).

Our article is not the first to examine externalities related to driving and car accidents.

Vickrey (1968) also discussed accident externality resulting from driving. Focusing on

two sets of California highways, he found an association between higher traffic density

and substantially higher accident rates. Vickrey’s work was extended further by Edlin

and Karaca-Mandic (2006), who improved the estimates of aggregated accident exter-

nalities resulting from driving. Huang et al. (2014) used individual-level data in Taiwan

to prove the existence of negative externality of driving. However, all of these studies

focused on the quantity of driving, not cost heterogeneity. Therefore, the welfare loss

of low-income families by the choice of high-income families or inefficiency created by

liability law system was never discussed in empirical studies.

Despite the significance of this topic and the associated controversy,the magnitude and

significance of externalities resulting from driving luxury cars have yet to be examined.

The purpose of this empirical article is to establish the existence of negative externalities

associated with driving luxury cars and estimate their extent. Specifically, we examine

the relationship between the ratio of luxury cars on Korean roads and claim frequency

and severity using a two-part model.

We find that negative externalities do exist. As the ratio of luxury cars increases, the

severity of claims regarding property damage liability increases significantly, whereas

accident frequency remains unaffected. The relationship between bodily injury and

luxury cars is also examined but it is insignificant. We found that a 1 percent increase

in luxury car ratio corresponds to a 1.9–2.6 percent increase in liability cost per claim.

The total increase in costs nationwide could therefore be as much as USD 139—196

million. In Seoul, which has the highest ratio of luxury cars, this means a 9–11 percent

increase or an approximately 118,876–169,802 won (approx. USD 108–154) increase in

per claim severity. If this cost were entirely borne by luxury car owners, each owner

would pay KRW 343,475–485,333 (approx. USD 312–441) more on average. Instead, the

cost is currently shared by all drivers in Korea, where territorial rating is banned, so the

increased liability cost for each driver is estimated at KRW 7,900–11,163 (approx. USD

7.2–10.1). Our conclusion is robust to all specifications.

In the next section, we provide a description of the Korean automobile market and auto

insurance system. In the “A Simple Externality Model” section, we suggest a framework

for externalities related to luxury car accidents. The data and methodology are presented