Systemic Risk and The U.S. Insurance Sector

• Author: Mary A. Weiss,J. David Cummins
• Published date: 01 September 2014
• Date: 01 September 2014
• DOI: http://doi.org/10.1111/jori.12039

SYSTEMIC RISK AND THE U.S. INSURANCE SECTOR

J. David Cummins

Mary A. Weiss

ABSTRACT

This article examines the potential for the U.S. insurance industry to cause

systemic risk events that spill over to other segments of the economy. We

examine primary indicators of systemic risk as well as contributing factors that

exacerbate vulnerability to systemic events. Evaluation of systemic risk is

based on a detailed financial analysis of the insurance industry, its role in the

economy, and the interconnectedness of insurers. The primary conclusion is

that the core activities of U.S. insurers do not pose systemic risk. However,

life insurers are vulnerable to intrasector crises, and both life and property–

casualty insurers are vulnerable to reinsurance crises. Noncore activities

such as financial guarantees and derivatives trading may cause systemic

risk, and interconnectedness among financial institutions has grown

significantly in recent years. To reduce systemic risk from noncore activities,

regulators need to continue efforts to strengthen mechanisms for insurance

group supervision.

INTRODUCTION

The financial crisis of 2007–2010 is a classic episode of a systemic risk event, where

problems in one sector of the economy, in this case housing, spread to other sectors

and led to general declines in asset values and real economic activity. Because the

crisis began in the financial industry and one of the major firms that played a role in

the crisis was an insurer (American International Group [AIG]), questions have been

raised about whether the insurance industry is a major source of systemic risk.

Answering this question has important implications for policymakers, managers,

investors, and regulators. Indeed, the newly established Financial Stability Oversight

Council (FSOC) is charged with determining whether there are nonbank financial

J. David Cummins and Mary A. Weiss are at the Department of Risk, Insurance, and Healthcare

Management, Temple University, Alter Hall, 006-00, 1801 Liacouras Walk, Philadelphia, PA

19122. Cummins and Weiss can be contacted via e-mail: cummins@temple.edu and

mweiss@temple.edu, respectively. We thank Zhijian Feng and Yanqing Zhang for their

excellent and tireless assistance with the data analysis. Thanks are also due to Frank Nutter and

Scott Williamson of the Reinsurance Association of America for valuable input regarding

reinsurance counterparty relationships. Any mistakes are solely the responsibility of the

authors.

© The Journal of Risk and Insurance, 2014, Vol. 81, No. 3, 489–527

DOI: 10.1111/jori.12039

489

institutions that should be designated as systemically important financial institutions

(SIFIs).

The purpose of this article is to investigate whether the U.S. insurance sector poses a

significant systemic risk to the economy. Because systemic risk is discussed

throughout the article, we begin with our definition of systemic risk, which is

discussed in more detail later:

Systemic risk is the risk that an event will trigger a loss of economic value

or confidence in a substantial segment of the financial system that is

serious enough to have significant adverse effects on the real economy

with a high probability.

1

Importantly, an economic event is not considered systemic unless it affects a

substantial segment of the financial system and leads to a significant decline in real

activity. For example, an event such as the 1980s liability insurance crisis, which had a

significant effect on the property–casualty (P-C) insurance industry, would not be

considered systemic.

In our analysis of systemic risk, we identify primary indicators that can be used to

measure the degree of systemic risk posed by specific markets and institutions (i.e.,

size, interconnectedness, and lack of substitutability). We also identify contributing

factors that increase the vulnerability of markets and institutions to systemic shocks.

Next, data are presented on the macroeconomic role of insurers in the U.S. economy

and the recent financial history of the insurance industry in terms of leverage and

insolvency experience. Because interconnectedness is one of the primary factors

driving systemic risk, an important contribution of this study is to provide

information on a form of interconnectedness unique to the insurance industry—

reinsurance counterparty relationships. Finally, we draw conclusions regarding the

potential for systemic risk events originating in the insurance industry.

An important distinction in our analysis is between the core activities of insurers, such

as insurance underwriting, reserving, claims settlement, and reinsurance, and the

noncore or banking activities engaged in by some insurers (such as AIG). Noncore

activities include provision of financial guarantees, asset lending, issuing credit

default swaps (CDSs), investing in complex structured securities, and excessive

reliance on short-term sources of financing.

Our discussion of the core activities of insurers focuses on the U.S. life–health (life)

and P-C insurance industries.

2

However, in analyzing reinsurance counterparty

exposure, we consider interrelationships between U.S. licensed insurers and

reinsurers worldwide. Our discussion of noncore activities provides data on the

1

Our definition of systemic risk is analogous to the definition proposed in Group of Ten (2001,

p. 126). Similar definitions have been proposed by other organizations. See, for example,

Financial Stability Board (2009).

2

The article does not analyze the monolines, which are important but deserve to be analyzed

separately.

490 THE JOURNAL OF RISK AND INSURANCE

participation by insurers in the market for asset-backed (structured) securities and

discusses more generally CDSs, asset lending, and financial guarantees, where data

are not readily available. By way of preview, the analysis suggests that the core

activities of insurers are not a major source of systemic risk. However, to the extent

that insurers engage in noncore activities, they become more susceptible to and could

become a source of systemic risk.

The remainder of this paper is organized as follows. The second section summarizes

the literature on systemic risk in the insurance industry. The third section further

defines systemic risk by analyzing primary indicators and contributing factors. The

fourth section analyzes the macroeconomic role of insurers, insurer insolvency

experience, and interconnectedness through reinsurance. The fifth section analyzes

the noncore activities of insurers. The sixth section analyzes business segments and

derivatives activity for samples of systemic and nonsystemic insurers to search for

characteristics of systemic firms. The seventh section concludes and provides some

directions for future research.

LITERATURE REVIEW

There have been a few prior studies of systemic risk in the insurance industry. Swiss

Re (2003) and the Group of Thirty (2006) investigate whether reinsurers pose systemic

risk. Both studies generally conclude that reinsurance does not pose a systemic risk

because primary insurers spread their reinsurance cessions across several reinsurers

and because reinsurers are not sufficiently linked to the banking sector or capital

markets. Bell and Keller (2009) investigate the systemic risk of the insurance industry

and conclude that “classic insurers ...do not present a systemic risk.” However, they

argue that noncore activities can pose systemic risk.

Harrington (2009) conducts an extensive study of systemic risk in insurance, focusing

on the federal takeover of AIG. He concludes that “the AIG crisis was heavily

influenced by the CDS written by AIG Financial Products, not by insurance products

written by regulated insurance subsidiaries.” The Geneva Association (2010)

concludes that insurers did not play a major role in the financial crisis aside from

monolines and insurers’ noncore activities.

Based on statistical tests on insurer stock prices, Grace (2010) concludes that AIG was

systemically important but that insurers in general are not a significant source of

systemic risk. Baluch, Mutenga, and Parsons (2011) investigate the role of the

insurance industry in the financial crisis, emphasizing European markets. They

conclude that systemic risk is lower in insurance than in banking but has grown due

to increased noncore activities.

Although the prior literature raises few concerns regarding systemic risk originating

from insurance, there are several reasons to evaluate the issue in more detail. First,

recent “micro” analyses of stock market data suggest that the interconnectedness of

financial firms may be more significant than prior research seems to suggest (Acharya

et al., 2010; Billio et al., 2012). The implications of these two papers are considered in

more detail later in this article. Chen et al. (2014) provides econometric estimates of

systemic risk in the banking and insurance industries, using data on stock prices

SYSTEMIC RISK AND THE U.S. INSURANCE SECTOR 491