Cyclicity in the French Property–Liability Insurance Industry: New Findings Over the Recent Period

• Published date: 01 June 2015
• Author: Catherine Bruneau,Nadia Sghaier
• Date: 01 June 2015
• DOI: http://doi.org/10.1111/jori.12027

©2014 The Journal of Risk and Insurance. Vol.82, No. 2, 433–462 (2015).

DOI: 10.1111/j.1539-6975.2013.12027.x

Cyclicity in the French Property–Liability Insurance

Industry: New Findings Over the Recent Period

Catherine Bruneau

Nadia Sghaier

Abstract

This article reinvestigates the presence and the causes of the underwriting

cycle in the French property–liability insurance industry as displayed by the

combined ratio for the 1963–2008 period. The question is still a timely issue if

we refer to regulation issues and the recent proposals in the Solvency frame-

work to take into account the fluctuations of the profitability in specifying

the solvency capital requirement. In the literature, two approaches are tra-

ditionally adopted to investigate the underwriting cycle. The first one refers

to an endogenous characterization of the cyclical properties from an AR(2)

model. The second one claims that the cycle in the property–liability insur-

ance has exogenous sources related to the financial markets and the general

economy.In this article, we reconcile the two approaches by using a smooth

transition regression model. This model shows that the AR(2) model is rel-

evant in a first regime where the capacity constraint is binding. In contrast,

the fluctuations in the combined ratio are positively influenced by the lagged

stock market return in a second regime wherethe capacity is not constrained,

as for the most recent period. Moreover, we find that the currentcapacity is

related to the lagged inflation rate in the latter case. These results confirm the

idea that the European rules regarding the solvency capital requirement for

insurance companies should take into account the state of the economy and

the financial markets.

Introduction

It is generally admitted that the fluctuations in the premiums and the underwriting

profits in the property–liability insurance industry exhibit a cyclical behavior, with

hard and soft market phases.

These findings can be seen in many papers in the related literature. The question is

still a timely issue if we refer to the current debates about regulation in this activity

Catherine Bruneau is Professor at the Panth´

eon-Sorbonne University and CES, MSE, 106-

112 Avenue de l’Hˆ

opital, 75013 Paris, France. The author can be contacted via e-mail:

cbruneau475@gmail.com. Nadia Sghaier is Associate Professor at the IPAG Business School,

IPAGLAB, 184 Boulevard Saint-Germain, 75006 Paris, France. The author can be contacted via

e-mail: nadia.sghaier@ipag.fr. The authors thank the anonymous refereefor helpful comments

on earlier versions.

433

434 The Journal of Risk and Insurance

sector.For example, we can find the statement that an insurance company, which starts

business at equilibrium, should not be asked to take corrective action if poor stock

market returns or a depressed economic environmentinduce a fall below the solvency

capital requirement, because an improvement in the business cycle is expected to

follow and, consequently,an improvement in solvency. Such a proposal makes it clear

that identifying the cyclical behavior of the profitability and understanding which

factors can explain this behavior could be crucial to forecast hard and soft market

phases and adapt the solvency capital requirement to these phases.

In what follows, we will reexamine the cyclicality question and its relation with sol-

vency regulation issues. Accordingly, we will mainly focus on the dynamics of the

combined ratio, defined as the ratio of losses including loss adjustment and under-

writing expenses to premiums, which is usually retained as a profitability measure.In

addition, we will examine the dynamics of a capacity measure, the capacity ratio, de-

fined as the ratio of capital to premiums, which is a key variable in solvency questions

and may play a major role in explaining the cyclical fluctuations in the profitability

as claimed, for example, by the capacity constraint theory.

As regard the empirical investigation of cyclical properties of the insurance activity,

the traditional approach consists of testing the existence of the underwriting cycle

and calculating its length from a second-order autoregressive process (Venezian,

1985; Cummins and Outreville, 1987). An alternative approach is based on the

spectral analysis (Doherty and Kang, 1988; Grace and Hotchkiss, 1995; Meier, 2006;

Venezian and Leng, 2006).

Once cyclical features are recognized, the question generally addressed is to know

which factors are responsible for this behavior. Although there are numerous studies

on this topic, there is no consensus on the causes of the cyclical features. However,we

can classify the studies into two main strands. The first one claims that the cycles are

endogenously generated, resulting from irrational behavior of the insurers (Venezian,

1985), capacity constraint (Winter1988, 1994; Gron, 1994a, 1994b), and financial quality

(Cagle and Harrington, 1995; Cummins and Danzon, 1997). The second assumes that

the underwriting cycle is caused by the external factors related, for example, to institu-

tional regulation or accounting interventions (Cummins and Outreville, 1987; Lamm-

Tennant and Weiss, 1997; Chen, Wong, and Lee, 1999), interest rate (Doherty and

Kang, 1988; Fields and Venezian,1989; Smith, 1989; Haley,1993, 1995, 2007), stock mar-

ket return (Cummins and Nye, 1980; Cummins and Harrington, 1985, 1988; Jawadi,

Bruneau, and Sghaier, 2009; Zhang and Nielson, 20091), reinsurance market (Berger,

Cummins, and Tennyson, 1992; Cummins and Weiss, 2000; Meier and Outreville,

2006, 2010), inflation rate (Bruneau and Sghaier, 2009), and general economy (Grace

and Hotchkiss, 1995; Leng and Meier,2006; Meier, 2006; Guo, Fung, and Huang, 2009).

Although the cyclical patterns in the underwriting profits have been largely exam-

ined in United States, few studies have focused on the underwriting cycle in France.

1In Zhang and Nielson (2009), the authors examine both the impact of interest rate and stock

market index.