Download Characteristics of Catastrophe Risk

Characteristics of
Catastrophe Risk
• Low frequency but high severity events.
• High exposures and vulnerabilities.
• Mismanagement of catastrophe risk can have highly
adverse social, economic and political implications
for the affected countries.
• Can strain local governmental and insurance sector
financial resources and often requires offshore risk
transfer.
• Some risks can not be hedged.
The Insurance and Contractual Savings Team
sees FSE’s Catastrophe Role as Follows
• Vulnerability of the world’s poor to natural disasters
should underpin the World Bank’s work on risk transfer
and risk financing.
• By ensuring that sufficient liquidity exists after a disaster,
risk transfer/funding mechanisms can help to speed
economic recovery and reduce government fiscal exposure
to natural disasters.
• Catastrophe risk management can also assist countries in
the optimal allocation of risk in the economy, thus
contributing toward higher economic growth, better
mitigation and more effective poverty alleviation.
Development and Concentration are two key
loss determinants
Est. Losses as percent of GDP
Country Type
Concentrated
Developed
Hanshin Earthquake,
Japan, 1995: 2.5
Developing
Mamara Earthquake,
Turkey 1999: 5.0
Hurricane Mitch,
Honduras, 1998: 20.0
Source: IIASA, Options, Fall/ Winter, 1999
Diversified
Hurricane Andrew,
USA, 1992: 0.5
Floods,
China,1998: 0.7
Floods,
Poland, 1998: 3.0
Assessing the Real Cost
• Little work done - reinsurers, Litan,
IISA/WB
• Three part model:
. Direct property loss
. Indirect losses
. Secondary losses
Insured and Uninsured Losses from
Natural Disasters (in US Billions)
US$ 160 bn
80
70
60
50
40
30
20
10
0
1950
1955
1960
1965
1970
Economic losses (2000 values)
of which insured losses (2000 values)
Trend of economic losses
Trend of insured losses
1975
1980
1985
1990
1995
2000
The bulk of the gap is in developing
countries: 1970 – 2000 analysis
Country Type
40 worst disasters lives lost
40 worst disasters insured losses
Developing:
No. of disasters
1
4
No. of lives lost
1,296,200
21,528
Insured loss
US$4.6billion
US$6.9billion
Developed:
No. of disasters
2
36
No. of lives lost
14,525
9,460
Insured loss
US$3.7B
US$113.7B
Insurance Penetration tells half the story
Country
GDP/
Capita
USA
UK
Sweden
Spain
Mexico
Argentina
Philippines
Zambia
26,030
19,720
28,271
14,810
3,512
8,584
1,197
325
Non Life
Insurance
Premiums %
GDP
4.8**
3.4
2.3
2.9
0.8
1.2
0.8
0.6
Why Market Failure?
• Loss of credibility - motor insurance
syndrome
• Lack of history of insurance
• Agency costs
• Reinsurer/ broker rents
• Moral hazard - government subsidies
• Religion?
• Uninsurability at acceptable prices
The other half - clients with least insurance
capacity are often the most exposed and
vulnerable
Index of Vulnerability to Natural Disasters:
Vanuatu
727.17
Bangladesh
539.16
Trinidad & Tobago 523.13
India
510.67
The Bahamas
491.28
Mauritania
487.55
Antigua & Barbuda 430.77
Botswana
418.03
The World Bank has helped to fill the
gap: 1980-2001
more than $30 billion
19
13 7
156
($5.3 b)
Africa
60
28
Latin America and Caribbean
88 ($8.1 b)
10
44
54 ($5.3 b)
South Asia
43
12
East Asia and Pacific
55 ($5.5 b)
25
10
35 ($1.8 b)
Middle East and North Africa
20
13
33 ($3.0 b)
Europe and Central Asia
0
20
Urban&T ransport
40
60
80
All other sectors
100
120
140
160
180
But may have also added to the
problem
‘..the World Bank, must increasingly incorporate
natural disasters and natural hazards into the
projects and programs they fund. Some of their
projects are not only silent on the issues of disaster
vulnerability but may actually serve to increase
exposure and vulnerability.’
Source: Berke and Beatley, ‘After The Hurricane’, John
Hopkins, 1997
FSE has developed a rigorous country risk
management approach – a new product - 1
• Independent Estimates of Countries’ Economic
Exposures and Vulnerability to Natural Disasters;
• Quantification of Economic Benefits from Different
Risk Transfer/Risk Hedging Arrangements;
• Selection of Best Risk Transfer and Financing
Programs
• Review of premium rates and assistance in the design
of risk transfer instruments
National Catastrophe Risk Management
Country Assets
(people, housing, factories, schools…)
Flood, Earthquake, Wind….
Risk Analysis
Expected Annual Loss
Loss Exceedance (PML’s)
Risk Transfer Cost/Benefit
Revise Strategy
Reinsurance/Alternative
Risk Financing Strategies
No
(Risk Transfer/Financing)
Achieve
Risk Management
Objectives?
Yes
Manage Position
Source: EQE
Lower Risk
Mitigation, Land use
planning
No
(Risk Reduction)
This involves a lot of technology
• Risk Identification and Measurement
– Extensive use of stochastic catastrophe risk models employing the
latest scientific research on natural hazards and utilizing stock
inventory and vulnerability data (EQECAT, RMS, AIR)
• Loss control programs
– Loss prevention programs/national mitigation efforts/enforcement
of building codes, construction supervision.
• Risk transfer/risk financing
– Reinsurance
– Government
– Insurance Industry
FSE has developed a rigorous country risk
management approach – a new product - 2
• Determination of expected survivability of
insurance/reinsurance pools for given levels of
exposure and capitalization
• Provision of risk funding facilities
• Design of Legal and Institutional Frameworks for
Risk Management
And we are developing a generic financing model
Capital markets
International R/I
WB
Budget
Country or Regional R/I Cat. Pool
Private Market
Industry and
the wealthy
Proxy Market
– pure cat.
Property owners
and SMEs, Cash
Farmers
Gov’t
Captives
Infrastructure
Welfare
transfers
The very
poor
When do the financial products
work?
• Relatively frequent, but not too frequent
(Boston EQ - Tunisian drought Bangladesh Flood) - cognitive effects
• The population has some experience of
insurance – otherwise tax perception
• The funding process will support mitigation
efforts - political cycle
• Reasonable data is available
Even when the basics are in place there are
challenges in building risk transfer systems
• Lack of risk awareness at the government level and
among population;
• Undeveloped insurance sector;
• Excessive reliance on the government as the reinsurer
of last resort – moral hazard;
• Low country incomes;
• High degree of uncertainty with regard to expected
economic losses.
• Distribution costs.
• Lack of public/ private trust.
The catastrophe work program to date includes the
Bank’s involvement in the design of the Turkish
Catastrophe Insurance Pool, preparation of the
Regional R/I Pool in the Caribbean and risk
management studies of natural hazards in Honduras,
India, Bangladesh, Pakistan and Sri Lanka. Feasibility
studies on parametric weather insurance have been
carried out in Nicaragua, Morocco, Mexico,Mongolia
and Turkey and tangible results already achieved in
Morocco and Mexico. Pre assessment work is
underway in Romania, Bulgaria and Iran.