On the Corporate Demand for Insurance: Evidence From the Global Reinsurance Market

• Author: James Garven,Jannes Rauch,Muhammed Altuntas
• Date: 01 September 2018
• DOI: http://doi.org/10.1111/rmir.12107
• Published date: 01 September 2018

Risk Management and Insurance Review

C

Risk Management and Insurance Review, 2018, Vol.21, No. 2, 211-242

DOI: 10.1111/rmir.12107

FEATURE ARTICLE

ONTHECORPORATE DEMAND FOR INSURANCE:

EVIDENCE FROM THE GLOBAL REINSURANCE MARKET

Muhammed Altuntas

James Garven

Jannes Rauch

ABSTRACT

In this article, we view the demand for reinsurance as a “special case” of

the corporate demand for insurance. We analyze the extent to which reinsur-

ance purchases by the global property–liability insurance industry vary across

countries and assess the relative importance of country-level factors compared

with firm-level factors. Using a data set consisting of 21,814 firm-year observa-

tions from 33 (developed and developing) countries during the period 2000–

2012, we find that after controlling for firm-level factors, country-level factors

have economically as well as statistically significant effects on the demand for

reinsuran ce.

INTRODUCTION

Corporate risk management is an important issue for virtually all firms. Not surprisingly,

this topic has been explored in a broad range of studies. Much of the financial theory

and empirical research has focused attention on either corporate hedging (Smithson

and Simkins, 2005; Aretz et al., 2007; Smith, 2008) or corporate insurance purchases

(MacMinn and Garven, 2013).

Since disclosure requirements for corporate risk management decisions involving

hedging and insurance vary widely, a significant challenge for risk management

empirical research has been a paucity of cross-sectional and time series data. A popular

strategy for addressing this problem has been to study risk management decisions

Muhammed Altuntas is in the Department of Risk Management and Insurance at the University

of Cologne, Cologne, Germany; e-mail: Muhammed.Altuntas@uni-koeln.de. James R. Garven

is in the Department of Finance, Insurance and Real Estate at Baylor University, Waco, TX;

e-mail: James_Garven@baylor.edu. Jannes Rauch is in the Department of Risk Management and

Insurance, at the University of Cologne, Cologne, Germany; e-mail: jrauch@uni-koeln.de.

This article has benefitted from helpful comments offered by Thomas R. Berry-St¨

olzle, Kathleen

McCullough, Heinrich R. Schradin, Sharon Tennyson, by seminar participants at the American

Risk and Insurance Association and the World Risk and Insurance Economics Congress, and by

two anonymous referees.

We would also like to thank Allianz SE for their support in providing the AM Best’s Statement

File-Global database used in this study.Of course, the usual disclaimer applies.

211

212 RISK MANAGEMENT AND INSURANCE REVIEW

within the context of industries where data on corporate risk management decisions

are readily available.1In the case of the insurance industry, reinsurance purchases by

insurance companies are systematically reported in developing as well as in developed

countries, so data availability on insurance usage is less problematic for the insurance

industry compared with most other industries.2Furthermore, since a rich variation in

risk taking occurs within a single industry, the insurance industry provides a nearly

ideal laboratory setting for studying the corporate demand for insurance (Mayers and

Smith, 2013). Following Mayers and Smith (2013), we view the demand for reinsurance

as a “special case” of the corporate demand for insurance.

Previous empirical research on the demand for reinsurance typically relies on data from

a single country (usually either the United States or the United Kingdom; Mayers and

Smith, 1990; Shiu, 2011) and focuses attention on identifying firm-level determinants of

the demand for reinsurance. However, based on theory and evidence, we expect that

reinsurance demand is affected not only by firm-level factors but also by country-level

factors. Indeed, the transaction cost theory proposed by Coase (1937) and Williamson

(1985) implies this by highlighting how external institutional factors influence the design

of corporate contracts and organizations. Altuntas et al. (2015) apply this line of reasoning

in their empirical study of how capital structure policies of insurers are shaped by

country-level factors, which cause the costs and benefits of holding capital to vary across

countries. In the empirical risk management literature, Fields et al. (2012) document

the importance of country-level factors in their study of risk taking by publicly listed

insurers. Country-level factors have also been shown to affect capital structure policies

of nonfinancial corporations as well as insurers; for example, see Gungoraydinoglu and

¨

Oztekin’s (2011) study of capital structure policies for nonfinancial companies. Moreover,

Cole et al. (2007) and Cole et al. (2012) document that country-level as well as firm-level

factors affect the supply of reinsurance provided to global markets by U.S. reinsurers.

Using a data set consisting of 21,814 firm-year observations from 33 (developed and de-

veloping) countries during the period 2000–2012, we document how the global demand

for reinsurance by property–liability insurers is affected by country-level as well as firm-

level factors. Since most (single-country) empirical reinsurance studies use data from

1For example, see Tufano’s(1996) study of hedging in the North American gold mining industry,

Jin and Jorion’s (2006) study of hedging in the U.S. oil and gas industry,Lin et al.’s (2012) study

of corporate property insurance demand in China, and Mayers and Smith’s (1990) study of

reinsurance purchases by U.S. property–liability insurance companies.

2It is worth noting that in the United States and the United Kingdom, insurers are also requiredto

disclose details concerning derivative usage. This U.S./U.K. regulatory disclosure requirement

has spawned several papers (Cummins et al., 1997; Drechsler and Cummins, 2008; Hardwickand

Adams, 1999; Shiu, 2016), which document that reinsurance purchases are negatively correlated

with derivatives usage by insurers; that is, the more insurers rely on reinsurance, the less

they hedge using derivatives (and vice versa). This empirical result comports well with the

notion that insurers coordinate the management of underwriting and investment risks within

an integrated/enterprise risk management framework. However,since insurer derivative usage

disclosure requirements are less detailed for the majority of the firms from countries outside

the United States, it is not considered here. Indeed, the data source used for this study (A.M.

Best’s Statement File Global) does not provide this kind of information for any insurers that are

included in the database.

ONTHECORPORATE DEMAND FOR INSURANCE 213

the property–liability insurance industry,3this makes our results more directly com-

parable with the prior body of research. Furthermore, reinsurance plays a particularly

important risk management role for property–liability insurers due to the unpredictable

nature of exposure to low-frequency, high-severity catastrophe risks (Froot, 1999; Froot

and O’Connell, 2008). Although life insurers have traditionally used reinsurance to

manage longevity risk, Biffis and Blake (2009) argue that mortality-linked securities and

derivatives are better suited for that purpose.4

Our results indicate that after controllingfor firm-level f actors, country-level factors have

economically as well as statistically significant effects on the demand for reinsurance. To

the best of our knowledge, ours is the first article to analyze the effects of country-level

as well as firm-level factors on the demand for reinsurance. Thus, we extend previous

research, which only considers firm-level determinants of the demand for reinsurance

within a single country. Moreover, knowledge concerning the impact of country-level

factors on the demand for reinsurance is important for insurers, reinsurers, and poli-

cymakers alike. For insurers and reinsurers, our study provides evidence concerning

how institutional environments influence reinsurance demand and supply in the coun-

tries in which they operate. Policymakers are provided with evidence concerning how

country-level regulatory environments influence insurers’ risk management strategies.

Furthermore, since many large nonfinancial corporations own captive insurance sub-

sidiaries, our research also has potential implications for the risk management strategies

of corporate owners of captives (Adams et al., 2008).

The article proceeds as follows. In the next section, we discuss the role of reinsurance

in the insurance industry. In the section “Data and Methodology,” we describe the data

and methodology that we use for our empirical analysis, and present the firm- and

country-level factors that we include in our analysis. In the section “Empirical Results,”

we present the results. We provide concluding remarks in the “Conclusion” section of

the article.

THE ROLE OF REINSURANCE IN THE INSURANCE INDUSTRY

Reinsurance is insurance purchased by one insurer from another insurer. The company

that buys reinsurance is the ceding insurer or cedent, and the company that sells rein-

surance is the reinsurer. Thus, the cedent “cedes” reinsurance to the reinsurer.

Reinsurance plays a crucial role in the risk management strategies of insurers. Insurers

manage risks by relying on the law of large numbers in that they pool many risks,

3See Mayers and Smith (1990), Garven and Lamm-Tennant (2003), Cole and McCullough (2006),

Shiu (2011), Manka¨

ı and Belgacem (2016), and work by Adams (1996) and Adams et al. (2008)

on life reinsurance are notable exceptions.

4Holzheu and Lechner (2007) report that on average, life insurers reinsured, or ceded, only 1.9

percent of their business during 2003, compared with 13.1 percentfor property–liability insurers.

They also report that 83 percent of ceded global reinsurance premiums that same year were

paid by property–liability insurers, compared with 17 percent by life insurers. More recently,

in its 2015 annual report (http://reports.swissre.com/2015), Swiss Re put global premium

income during 2015 for property–liability reinsurance at 170 billion USD, compared with 2015

global premium income for life reinsurance at 65 billion USD. These statistics provide evidence

concerning the predominant role played by the property–liability insurance sector in the global

reinsurance market.