An Examination of the Geographic Aggregation of Catastrophic Risk

The debate in the United States about establishing a mechanism for insuring catastrophic wind risk at the national level pre-dates the intense 2004–2005 hurricane seasons. The prevailing argument against establishing any larger risk pool is that it would create a subsidy for the higher risk exposure...

Full description

Saved in:
Bibliographic Details
Published in:Geneva papers on risk and insurance. Issues and practice 2015-01, Vol.40 (1), p.159-177
Main Authors: Dumm, Randy E., Johnson, Mark E., Watson, Charles C.
Format: Article
Language:English
Subjects:
Capital markets
Catastrophes
Cyclones
Disaster insurance
Diversification
Economics and Finance
Finance
Hurricanes
Insurance
Insurance industry
Market penetration
Original Article
Property & casualty insurance
Public policy
Reinsurance
Risk exposure
Risk Management
Shin Research Excellence Award-Winning Papers
Studies
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The debate in the United States about establishing a mechanism for insuring catastrophic wind risk at the national level pre-dates the intense 2004–2005 hurricane seasons. The prevailing argument against establishing any larger risk pool is that it would create a subsidy for the higher risk exposures. To determine whether benefits do accrue by aggregating catastrophic risk across increasingly wide geographic areas, the paper uses catastrophe models to evaluate the behavior of residential property portfolios within the state of Florida and for a larger risk pool that includes multiple combinations of coastal states in the southeastern United States. We find that geographic aggregation does not inherently subsidise high-risk exposures, reduces uncertainty, and reduces required reserves relative to total exposure for the least frequent and more severe events. This finding holds true for all state combinations evaluated in this study.
ISSN:1018-5895
1468-0440
DOI:10.1057/gpp.2014.20