Insured Loss Inflation: How Natural Catastrophes Affect Reconstruction Costs

• Author: Martin Hibbeln,Marc Gürtler,David Döhrmann
• Date: 01 September 2017
• Published date: 01 September 2017
• DOI: http://doi.org/10.1111/jori.12134

© 2015 The Journal of Risk and Insurance. Vol. 84, No. 3, 851–879 (2017).

DOI: 10.1111/jori.12134

851

INSURED LOSS INFLATION:HOW NATURAL CATASTROPHES

AFFECT RECONSTRUCTION COSTS

David D€

ohrmann

Marc G€

urtler

Martin Hibbeln

ABSTRACT

In the aftermath of a natural catastrophe, there is increased demand for

skilled reconstruction labor, which leads to significant increases in

reconstruction labor wages and hence insured losses. Such inflation effects

are known as “Demand Surge” effects. It is important for insurance

companies to properly account for these effects when calculating insurance

premiums and determining economic capital. We propose an approach to

quantifying the Demand Surge effect and present an econometric model for

the effect that is based on 192 catastrophe events in the United States. Our

model explains more than 75 percent of the variance of the Demand Surge

effect and is thus able to identify the key drivers of the phenomenon.

INTRODUCTION

In recent decades, dramatic increases in the number and severity of catastrophes have

been observed (Kunreuther and Michel-Kerjan, 2009). These developments are

accompanied by a drastic increase in catastrophe-related economic losses, which is of

particular relevance because growth in catastrophe losses is expected to continue for

the foreseeable future, at least if effective disaster mitigation efforts are omitted

(Pielke, 2005; Pielke et al., 2008).

David D€

ohrmann is at Deutsche Bundesbank, Markets Department, Wilhelm-Epstein-Straße

14, 60431 Frankfurt am Main, Germany. Marc G€

urtler is at Braunschweig Institute of

Technology, Department of Finance, Abt-Jerusalem-Str. 7, 38106 Braunschweig, Germany.

Martin Hibbeln (contact author) is at University of Duisburg-Essen, Mercator School of

Management, Lotharstr. 65, 47057 Duisburg, Germany. Hibbeln can be contacted via e-mail:

martin.hibbeln@uni-due.de. We would like to thank an anonymous referee, Keith J. Crocker

(the editor), Randy Dumm, Lorilee Medders, and Richard Sylves, as well as the participants of

the 2013 Annual Meeting of the American Risk and Insurance Association, the 2013 Annual

Congress of the German Insurance Science Association, and the 2013 Annual Meeting of the

German Finance Association for helpful comments. The article represents the authors’ personal

opinions and do not necessarily reflect the views of the Deutsche Bundesbank or its staff. David

D€

ohrmann acknowledges the financial support from the German Research Foundation (DFG)

and the German Insurance Science Association (DVfVW).

852 THE JOURNAL OF RISK AND INSURANCE

The basis for economic losses is reconstruction costs, which must be raised after a

catastrophe to restore the original state of buildings and infrastructure. To estimate

future costs, however, it is not appropriate to apply the expected price level under

normal conditions. Rather, it must be considered that, in the case of a catastrophe,

there is increased demand for skilled reconstruction labor and building materials.

Because this increase in demand is confronted with a constant supply of relevant

goods and labor, significant price increases are expected, which in turn should be

taken into account in the forecast of catastrophe losses. Such price effects are referred

to as “Demand Surge” effects. According to the literature, “Demand Surge occurs

when the demand for products and services exceeds the regional capacity to

efficiently supply them. The additional costs for these products and services are

directly passed on to the consumer (and the insurer)” (EQECAT, 2005). Demand

Surge is especially relevant for insurance companies because this effect may lead to an

inflation of insured losses. For example, it is estimated that the average Demand

Surge effect for the affected construction lines due to Hurricane Katrina is in the range

of 30–40 percent (Munich Re, 2006).

Although Demand Surge is highly relevant for determining the economic damage of a

catastrophe, there are only few contributions in the literature that address this

phenomenon. This fact is even more surprising because it is a phenomenon that is

neither new nor limited to a particular region or a particular type of catastrophe

(Olsen and Porter, 2011a). Though the scientific literature considers Demand Surge

exclusively on a qualitative level or only for a specific catastrophe type or event,

universally valid quantitative models for Demand Surge have not been published. In

contrast, the three main catastrophe modeling companies, Applied Insurance

Research (AIR), EQECAT, and Risk Management Solutions (RMS), consider the

Demand Surge effect within the framework of modeling direct catastrophe losses.

Even if some background material is available to customers that provides some

intuition about the Demand Surge models of these companies, information about the

models is neither publicly available nor is it clear which concrete empirical analyses

or results underlie their models.

Against this background, the present article provides two main contributions. First,

we propose an approach to quantify the Demand Surge effect. Second, we introduce

the first econometric model for the effect. In this way, the article provides a basis for

the quantitative assessment of Demand Surge for future catastrophes, which, on one

hand, is important for (public and private) insurance companies when calculating

insurance premiums and determining economic capital. On the other hand, such

information is also relevant for investors of insurance stocks and issuers and investors

of catastrophe-linked securities (such as Cat Bonds), who have to consider Demand

Surge within the framework of security pricing.

Our empirical study is essentially based on data for natural catastrophes from the EM-

DAT database and pricing information for the construction sector from Xactware. The

data set of EM-DAT has comprised worldwide information on natural catastrophes

since 1900, and Xactware has been the leading provider of pricing information in the

construction sector for more than 460 economic areas in the United States and Canada

since 2002. Our proposed Demand Surge model is able to explain more than