RISK GOVERNANCE IN THE INSURANCE SECTOR—DETERMINANTS AND CONSEQUENCES IN AN INTERNATIONAL SAMPLE

• DOI: http://doi.org/10.1111/jori.12218
• Author: Cornelia Schilling,Elizabeth Sheedy,Shane Magee
• Published date: 01 June 2019
• Date: 01 June 2019

381

RISK GOVERNANCE IN THE INSURANCE

SECTOR—DETERMINANTS AND CONSEQUENCES

IN AN INTERNATIONAL SAMPLE

Shane Magee

Cornelia Schilling

Elizabeth Sheedy

ABSTRACT

We analyze the relation between risk governance, risk, and performance

measures for a global sample of 107 insurance companies from 2004 to 2012.

Our risk governance index (RGI) covers several Solvency II provisions and

includes the existence of chief risk officer on the executive committee, risk

committee characteristics, and board industry experience. We find that in the

crisis period 2008–2009, firms with a higher RGI generally have lower

expected default frequency. We conclude that during noncrisis years, risk

governance does not have a risk-reducing effect but is positively associated

with buy-and-hold returns, risk-adjusted performance measures, and

Tobin’s Q. Our findings therefore support the role of risk governance as a

business enabler rather than inhibitor. Insurance companies typically

upgrade their risk governance following a negative shock, especially in

countries that are well regulated and have weaker shareholder rights.

INTRODUCTION

Following the financial crisis, the issue of risk governance in the financial sector has

risen to prominence. In an OECD report, Kirkpatrick (2009) concludes that “... the

financial crisis can be to an important extent attributed to failures and weaknesses in

corporate governance arrangements which did not serve their purpose to safeguard

against excessive risk taking in a number of financial companies.” If the statement

from the OECD report holds, better risk governance should reduce downside risks

and produce better performance during the crisis through more effective risk

management.

Elizabeth Sheedy is the Program Director (Risk) for the M. Applied Finance at Macquarie

University in Sydney, Australia. Shane Magee, also from Macquarie, is the Program Director of

the Graduate Certificate of Finance and also teaches in the M. Applied Finance program. At the

time the research was done for this article, Cornelia Schilling was a graduate student from the

Chair of Mathematical Finance at the Technical University of Munich. Sheedy can be contacted

via e-mail: esheedy@mafc.mq.edu.au

© 2017 The Journal of Risk and Insurance. Vol. 9999, No. 9999, 1–33 (2017).

DOI: 10.1111/jori.12218

Vol. 86, No. 2, 381–413 (2019).

2THE JOURNAL OF RISK AND INSURANCE

382

The effect of risk governance on risk outcomes and performance in noncrisis periods

is also of interest to policymakers and corporate decision makers. Skeptics of risk

governance may question whether initiatives designed to strengthen risk oversight

and provide protection in periods of turbulence may unduly stifle business initiatives

and dampen performance in more benign periods. Some have argued that a focus on

risk management implies lower risk (e.g., Erkens, Hung, and Matos, 2012) but we

hypothesize that this does not necessarily follow, especially during periods of benign

business conditions. A high-risk strategy to achieve higher expected returns is a

rational and legitimate choice provided relevant stakeholders agree and adequate

mitigants are in place. In such a context, effective risk management becomes an

enabler to support the firm in achieving its objectives. A primary hypothesis of this

research is that risk governance increases risk-adjusted performance through more

effective risk management.

We recognize that in this study, as in numerous other empirical articles in corporate

governance, endogeneity may be an issue, particularly with regard to the direction of

causality. Better risk governance may lead to better outcomes, but risk and

performance might also affect the level of risk governance. In order to investigate

possible endogeneity, we also investigate the determinants of risk governance.

The insurance sector provides an ideal laboratory for testing hypotheses regarding risk

governance. The long-term profitability of insurance companies strongly depends on

their ability to assess, correctly price, and manage underwriting risk. In addition, insurers

are exposed to a wide range of business, conduct, reputational, and operational risks.

Hence, for insurance companies, risk management is a core competency and potentially

serves as a business enabler beyond its role as a defense in periods of turbulence.

Whereas the importance of the risk governance function and risk oversight of the

board of directors has been examined in the banking industry (see, e.g., Aebi, Sabato,

and Schmid, 2012; Ellul and Yerramilli, 2013), related studies in the insurance

industry have focused on the effect of traditional corporate governance mechanisms

(such as ownership, compensation, and board monitoring) on risk (see, e.g., Cheng,

Elyasiani, and Jia, 2011; Eling and Marek, 2013). Our article fills that gap by

investigating the effect of risk governance on risk and performance in the insurance

industry. The study is particularly relevant given the recent introduction of Solvency

II

1

with its focus on risk governance.

For a sample of 107 publicly traded insurance companies from around the world

during the period of 2004–2012, we analyze the determinants of risk governance and

the effect of risk governance on risk and performance. Consistent with Ellul and

Yerramilli (2013), we construct a risk governance index (RGI) to measure the

implementation of key risk governance mechanisms such as a chief risk officer (CRO)

and board risk committee. Principal components analysis is used to aggregate five

dimensions of risk governance into an RGI. We find that the RGI increases over the

sample period.

1

Solvency II is a directive that seeks to harmonize insurance regulation within the European

Union. It took effect January 1, 2016.

RISK GOVERNANCE IN THE INSURANCE SECTOR 3

383

To examine the relationship between RGI and risk, we use two shortfall risk measures

(tail risk and EDF). We find some evidence that firms with greater risk governance in

2007 had lower EDF during the financial crisis year of 2008 but the evidence is mixed

for tail risk.

When we examine the effect of risk governance throughout 2004–2012, we conclude

that risk governance has no risk-reducing effect but is positively associated with

performance, measured by buy-and-hold returns, two risk-adjusted performance

measures and Tobin’s Q. These findings are consistent with the argument that the

purpose of risk governance is not to reduce risk per se, but to promote business

opportunities consistent with the companies’ risk appetite (Stulz, 2015).

For the 18 insurance markets we cover in our study, the findings suggest that high tail

risk (i.e., large downward share price movements) in the immediate past predicts an

increase in risk governance. In addition, our multicountry research design highlights

the regulatory environment as an important explanatory variable for the adoption of

advanced groupwide risk governance structures. We highlight the risk-reducing

impact of risk governance in a crisis, but we demonstrate the benefits of risk

governance for performance in more prosperous times. In sum, we provide evidence

for the value-enhancing effect of risk governance for a sample of global insurance

firms, which may encourage implementation of such structures.

BACKGROUND,LITERATURE,AND HYPOTHESIS DEVELOPMENT

Our article relates to the broader literature on corporate governance in the insurance

sector reviewed by Boubakri (2011). In a recent study, Eling and Marek (2013) use a

structural equation model to explore the relationship between corporate governance

factors (compensation, monitoring, and ownership structure) and insurance-specific

risk measures for a sample of 307 publicly traded insurance companies from the

United Kingdom and Germany throughout 1997–2010. They find that all three

dimensions are negatively associated with risk taking. Fields, Gupta, and Prakash

(2012) study how investor protection, government quality, and contract enforcement

affect risk taking and performance of insurance companies from around the world.

They find evidence that the aforementioned environmental factors decrease risk

taking without impacting performance. With the exception of these two studies

(Fields, Gupta, and Prakash, 2012; Eling and Marek, 2013), recent research on

governance in the insurance industry has analyzed single-country samples (see, e.g.,

Cole, He, and McCullough, 2011; Ho, Lai, and Lee, 2013).

“Risk governance” emerged from the financial crisis of 2008–2009 and the observation

that traditional approaches to corporate governance had failed in financial

institutions (Beltratti and Stulz, 2012; Erkens, Hung, and Matos, 2012). Financial

institutions are unique corporations that arguably require different governance

mechanisms placing greater emphasis on risk management for the benefit of

customers and the wider community.

We define risk governance as the framework of rules, relationships, systems, and

processes within organizations with regard to the management and control of risk.

The Financial Stability Board (2013) provides a useful summary of the key principles