S THE incidence and intensity of
natural disasters have increased,
the resulting economic losses
have soared. During the past
10 years for which comprehensive data are
available (1992–2001), losses stemming from
natural disasters have averaged about $65 bil-
lion a year—more than a sevenfold real
increase since the 1960s (see pages 40–41)—
and they are expected to increase another
fivefold over the next 50 years. A comprehen-
sive study by Munich Re, a reinsurance com-
pany that specializes in disaster business,
estimates that the global direct costs of nat-
ural disasters will top $300 billion annually
by 2050, about 750 percent, in real terms, of
current levels. Munich Re estimates that aver-
age losses will range from a few tenths of a
percent of GDP to a few percent of GDP and
that some countries, especially small island
states, could face losses exceeding 10 percent
of GDP.
Many scientists believe that global warm-
ing is responsible for the increasing fre-
quency and severity of extreme weather
events like floods, hurricanes, windstorms,
and droughts. While few places will be
spared, Asia and Latin America will probably
feel the effects of climate change the most.
The Intergovernmental Panel on Climate
Change (IPPC) forecasts a 90 to 99 percent
chance that, over the next 50 years, floods
and droughts will become more common in
Latin America, rising sea levels will threaten
the survival of some island states, and tropi-
cal cyclones will become more intense.
The growing urbanization of the world’s
population has compounded the problem:
even a minor event can cause significant
damage in a heavily populated area. The pro-
portion of people in developing countries
who live in cities has doubled since 1960.
More than 40 percent now live in urban
areas, and this figure is expected to surpass
55 percent by 2030. Nearly half of these cities
are subject to extreme weather events
because of the same features that made them
attractive to settlers—such as natural flood
plains, alluvial soil, and river or sea access.
Fourteen of the world’s 19 megacities—cities
with 10 million or more inhabitants—are in
coastal zones, and over 70 of the world’s 100
largest cities can expect a strong earthquake
at least once every 50 years (see table).
Ninety-four percent of the world’s major
disasters in 1990–98 were in developing
countries, according to the World Bank’s
World Development Report 2000/2001.
However, these countries have made fewer
efforts than developed countries to adapt
their physical environments to mitigate the
impact of natural disasters or to insure
themselves against disaster risk, partly
because of the disincentive known as the
“Samaritan’s dilemma.” The dilemma arises
whenever those at risk (whether private sec-
tor parties or governments of vulnerable
countries) expect to receive support if disas-
ter strikes (from their national government
or foreign donors) and therefore underinvest
in protective measures—physical and finan-
cial—to reduce the costs they will incur
when it does strike. And, given the humani-
tarian imperative, it is hard for those in a
Being Prepared
Natural disasters are becoming
more frequent, more destructive,
and deadlier, and poor countries
are being hit the hardest.
Paul K. Freeman, Michael Keen,
and Muthukumara Mani
Finance & Development September 2003
42
A
position to help to make a credible commitment to scale
back postdisaster assistance even if those suffering did not
take appropriate protective measures.
The vulnerability of the poor
Twenty-four of the 49 poorest countries face a high level of
disaster risk; at least 6 of them have been hit by 2–8 major
disasters in each of the past 15 years. The chart on page 44
shows that the frequency of disaster in the 77 poor countries
eligible for support under the IMF’s Poverty Reduction and
Growth Facility (PRGF) is both high (with an average of
nearly three disasters each in 2002) and apparently rising.
Small island states are also especially at risk because of their
small economies, dependence on agriculture and tourism,
and narrow resource base. Some may eventually disappear.
Although economic losses stemming from natural disas-
ters are smaller, in absolute terms, in developing than in
developed countries, because of lower levels of infrastructure
and capital stock, they are far higher relative to GDP.
Between 1985 and 1999, the world’s wealthiest countries sus-
tained 57.3 percent of the direct losses caused by disasters,
representing 2.5 percent of their combined GDP; the world’s
poorest countries accounted for only 24.4 percent of losses,
but their losses represented 13.4 percent of their combined
GDP. Loss of life, moreover, is far greater in developing coun-
tries. More than 97 percent of all deaths from natural disasters
in 1990–98 were in developing countries.
Within developing countries, the poor are more likely to
suffer than the rich. First, they often live in areas especially
vulnerable to destructive events such as floods, hurricanes,
and landslides (not least because high-risk housing is more
affordable). Second, disasters can severely depress the food
production of the rural poor. Third, even small reductions in
income can have a dramatic impact on the poor: their sav-
ings are unlikely to be adequate in the event of large-scale or
multiple catastrophes, and they may be forced to sell real
assets such as agricultural land and livestock. Fourth, damage
to water supply and transport infrastructures hurt the poor
more than they hurt the rich. Finally, the poor are less likely
to have access to risk-sharing mechanisms like insurance.
Disasters can substantially increase measured poverty. It
has been estimated that El Niño increased headcount poverty
in affected areas in Ecuador by more than 10 percentage
points, and about 50 percent of the increase in headcount
poverty in the Philippines during the 1998 crisis has been
attributed to El Niño.
Macroeconomic impact
The destruction of physical assets, including capital stock,
infrastructure, natural resources, and, not least, labor, has
both a short- and a long-term impact on macroeconomic
performance in some countries, while natural disasters have
caused only minor economic disruptions in others. In
25 country-specific studies, the United Nations Economic
Commission for Latin America and the Caribbean (ECLAC)
found that the worse socioeconomic conditions were at the
time a disaster struck, the stronger its impact.
In the wake of a natural disaster, a country’s tax base
shrinks while its spending needs rise. Because disasters are
small relative to their economies, developed countries have
the flexibility to meet the cost of disaster relief by some mix
of temporary tax increases—Germany temporarily raised
taxes, for instance, after the 2002 floods—and borrowing. It
is far more difficult for developing countries to raise taxes.
Unless they receive external grants, they must either increase
borrowing or resort to monetization.
Along with the deterioration of their fiscal position, affected
countries may suffer a weakening of their trade balance, as
declining production of export goods and postdisaster recon-
struction boost demand for imports and divert tradables to
the home market. These developments, combined with the
flight of panicky foreign investors, put downward pressure on
the exchange rate, resulting in inflationary pressures. Disasters
depress not only the immediate macroeconomic outlook but
also the balance sheet positions of key economic sectors.
Public sector debt ratios are likely to worsen and domestic sav-
ing to decline, forcing both the private and the public sectors
to increase their borrowing abroad.
Preparedness
To some extent, countries can prepare themselves for natural
disaster by adapting their physical environment and their
economies and by purchasing insurance. (Measures aimed at
reducing the risk of natural disasters, like lowering carbon
dioxide emissions, are not addressed here.)
Finance & Development September 2003
43
Megacities at risk
Cities with 10 million or more inhabitants, 2000 and 2015
2000 (millions of people)
2015 (millions of people)
Tokyo*
26.4
Tokyo*
26.4
Mexico City
18.1
Bombay*
26.1
Bombay*
18.1
Lagos*
23.2
São Paulo
17.8
Dhaka*
21.1
Shanghai
17.0
São Paulo
20.4
New York*
16.6
Karachi*
19.2
Lagos*
13.4
Mexico City
19.2
Los Angeles*
13.1
Shanghai*
19.1
Calcutta*
12.9
New York*
17.4
Buenos Aires*
12.6
Jakarta*
17.3
Dhaka*
12.3
Calcutta*
17.3
Karachi*
11.8
Delhi
16.8
Delhi
11.7
Metro Manila*
14.8
Jakarta*
11.0
Los Angeles*
14.1
Osaka*
11.0
Buenos Aires*
14.1
Metro Manila*
10.9
Cairo*
13.8
Beijing
10.8
Istanbul*
12.5
Rio de Janeiro*
10.6
Beijing
12.3
Cairo*
10.6
Rio de Janeiro*
11.9
Osaka*
11.0
Tianjin*
10.7
Hyderabad
10.5
Bangkok*
10.1
Source: UN Population Division, March 2000.
* Cities located in coastal areas.
Adaptation. Possible measures include land-use planning
to avoid construction on seismic fault lines, in vulnerable
coastal regions, and on river shorelines; adoption of stan-
dards aimed at ensuring that buildings are resistant to shocks
like earthquakes and hurricanes; mitigation of environmen-
tal degradation like soil erosion that can increase the impact
of disasters; and engineering interventions, such as the
creation of dams for flood control, dikes to reroute flood
waters, and seawalls to break storm surges. In its 2001 World
Disaster Report, the Red Cross argued that investments of
$40 billion in disaster preparedness, prevention, and mitiga-
tion would have reduced global economic losses in the 1990s
by $280 billion.
Governments can also promote farming practices that
enable farmers to weather climatic variations—using
drought-resistant crop varieties, for instance—and help
farmers increase their ability to adapt to long-term change.
To ensure adequate water supply, governments may need to
anticipate increased seasonal variation and greater frequency
of both storms and dry periods.
Developed countries have done much more than develop-
ing countries to protect themselves against natural disasters.
The United States, in particular, significantly increased
expenditures on preparedness, mitigation, and recovery dur-
ing the disaster-prone 1990s—$1.9 billion in fiscal 1999
alone, according to the National Emergency Management
Association. In contrast, many developing countries lack the
financial resources, technical knowledge, and political will
to mitigate physical vulnerability. Moreover, in many
countries—especially the small island countries—physical
adaptation is inordinately expensive. And mitigation mea-
sures can eliminate only some of the risk, much of which
stems from the movement of people to disaster-prone areas.
Insurance. Even the best prepared will not be able to avoid
all natural disaster–inflicted damage. While disaster insur-
ance is fairly extensive in the United States—more than
50 percent of direct losses from catastrophes is insured—it is
much less commonplace in other developed countries. In
Austria, the Czech Republic, and Germany, for instance, only
about 10–20 percent of the losses from the floods in 2002
were insured. And, in countries with per capita incomes
below $10,000, insurance coverage is less than 10 percent; in
those with per capita incomes under $760, it is about 1 per-
cent. Asia, which suffered half of all the damage caused by
natural catastrophes and two-thirds of all the casualties from
catastrophic events in 1997, accounted for only 8 percent of
global purchases of catastrophic insurance, whereas Japan,
the United Kingdom, and the United States—less than
2 percent of the total market—accounted for 55 percent of
the total.
The insurance sector is still rudimentary in many develop-
ing countries. Disaster insurance is largely confined to
wealthy individuals and large enterprises, such as utilities
and hotels, that are strongly affected by the weather. Such
insurance schemes as have existed—usually offered by the
public sector—have often failed because of high administra-
tive costs, inefficient loss calculation, and inadequate pre-
mium charges. In many countries, governments have stifled
the development of innovative insurance products by oper-
ating highly subsidized public insurance programs.
There are important failures in the market for disaster
insurance. Adverse selection—the purchaser of insurance
knowing more about the underlying risk than the seller—
may be less of a problem than in other insurance markets, in
that the likelihood of disaster is, in principle, knowable with
some accuracy, as is the value of the property at stake. In the
San Francisco area, for instance, insurers are able to differen-
tiate risk by zip code. Risk predictions are less accurate in
many developing countries; however, this is due not to inher-
ent technical constraints but to the thinness of existing mar-
kets. This is not to say that assessing risk is easy—indeed,
climate change is adding to the difficulty of assessing the
probability of extreme weather events, and ill-defined prop-
erty rights make assessment even more complicated in devel-
oping countries. Nevertheless, two other problems appear
more fundamental: the difficulty of risk spreading (given the
size of the loss relative to that of the affected economies) and
the Samaritan’s dilemma.
New financial instruments tailored to extreme natural
events have been developed but have had little impact as yet
(see box). And, although governments could ease market
failures in the provision of disaster insurance, they are gener-
ally not positioned to act as insurers of last resort. However,
by strictly enforcing building and zoning regulations, they
can monitor firms’ and households’ preparedness in a cost-
effective way. Perhaps more important, governments can
address their Samaritan’s dilemma by making the purchase
of insurance compulsory or providing a premium subsidy.
France provides for catastrophe insurance using the existing
fire insurance program backed by government guarantees.
Other developed countries have established public-private
collaborative schemes to insure catastrophic events through
risk pooling, coupled with group reinsurance arrangements
and last resort credit backup. Turkey’s recently established
Finance & Development September 2003
44
Source: Center for Research on the Epidemiology of Disasters database and
IMF staff calculations.
Hardest hit
The number of natural disasters in PRGF-eligible countries is
high and rising.
0
50
100
150
200
250
02
01
2000
99
98
97
96
1995
insurance program for home-
owners’ losses from earthquakes
is based on this model. While
such measures can be an
efficiency-enhancing response to
market failure, they have a seri-
ous drawback—they may perpet-
uate inadequate adaptation. It is
widely believed, for instance, that
government support for subsi-
dized insurance (and relief when
disaster strikes) has encouraged
inefficient migration in the
United States to disaster-prone
coastal areas on the eastern
seaboard.
Fiscal implications
Countries at risk must also pre-
pare themselves fiscally for nat-
ural disasters. The risk of natural
disaster creates a contingent lia-
bility of a particularly difficult
kind, given the implicit guaran-
tees given to—or at least per-
ceived by—the private sector.
Because of the lack of demand
for, or unavailability of, insurance
in many developing and emerg-
ing market countries, govern-
ments assume substantial risk for
reconstruction after a disaster.
(For example, the World Bank
estimates that approximately 50 percent of its postdisaster
funding goes to restore damaged housing.) More generally,
the protection of those affected by disaster is widely seen as a
basic duty of government.
Countries must identify and acknowledge such contingent
liabilities. This requires assessing the probabilities and costs
of the various natural disasters that might befall them.
Clearly, there is considerable uncertainty as to, for example,
the maximum loss that might be suffered. There is, however,
considerable historical information to build on. Combining
this with evidence on current trends, a rough estimate
should be feasible. A recent study by the World Bank, Swiss
Re, and the International Institute for Applied Systems
Analysis illustrates how natural catastrophe loss calculations
can be integrated into the World Bank’s macroeconomic
planning model for countries.
Some provisioning against the risk of disaster may also be
appropriate. While governments typically set up contingency
funds to deal with unanticipated spending needs, natural
disasters are not systematically factored into the calculations.
A number of countries have been exploring the use of
reserve funds for postdisaster financing. Mexico’s FONDEN,
for example, is an annual bud-
getary allocation for natural dis-
aster expenditures. While it is
evidently impossible to set aside
enough to meet the costs of all
conceivable disasters—and this
would, in any event, not be the
best use of scarce development
funds—the importance of cover-
ing the immediate costs makes it
prudent to adopt a fiscal stance
that provides some degree of
self-insurance.
Disaster risk makes it appro-
priate for a government to adopt
a tighter long-term fiscal stance
(with possible exceptions for
investment in measures of adap-
tation). In general, one would
expect the amount set aside to be
higher (relative to the expected
loss) the larger the potential loss
relative to national income, the
greater the likelihood of disaster,
the more expensive the insur-
ance, and the more risk-averse
the government. Governments,
especially in developing coun-
tries, can also offset failures in
local insurance markets—for
example, by issuing guarantees to
insurers and reinsurers (guaran-
tees that might then be hedged in
world reinsurance and capital markets). There may also be
scope for simply requiring certain parties to buy at least a
minimum amount of insurance coverage. Not least, govern-
ments might do more to insure their own property.
The international financial institutions can support these
efforts by tackling market failures and helping to finance
relief and adaptation efforts. Prompt foreign assistance can
reduce both the long-term macroeconomic damage caused
by natural disasters—although some physical losses may be
permanent (for example, irreversible soil erosion caused by
severe flooding)—and the costs of recovery. The quicker out-
put recovers, the less the public sector will need to borrow or
monetize, and the sooner clean water is restored, the lower
the morbidity impact.
Michael Keen is a Division Chief and Muthukumara Mani is
an Economist in the IMF’s Fiscal Affairs Department. Paul K.
Freeman is an independent consultant.
For a more detailed treatment of this issue, see the authors’ paper “Dealing
with Increased Risk of Natural Disasters: Challenges and Options” (forth-
coming; Washington: International Monetary Fund, 2003).
Finance & Development September 2003
45
Risk management
New financial instruments for hedging against
weather and natural disaster risks are available in
international capital markets.
• Catastrophe bonds are subject to default if a
defined catastrophe occurs during the life of the
bond but are attractive to investors because of
their correspondingly high yields.
• Contingent surplus notes are essentially
“put” rights that allow the notes’ owners to issue
debt to prespecified buyers in the event of a cata-
strophic event.
• Exchange-traded catastrophe options allow
their purchasers to demand payment under an
option contract if the index of property claims
service options traded on the Chicago Board of
Trade surpasses a prespecified level. Indexes cover
different areas of the United States and reflect
insurance industry aggregate reported claims.
• Catastrophe equity puts are a type of option
that permits the insurer to sell equity shares on
demand after a major disaster.
• Catastrophe swaps are derivatives that use
capital market players as counterparties. An
insurance portfolio with potential payment lia-
bility is swapped for a security and its associated
cash flow payment obligations.
• Weather derivatives are contracts that pro-
vide payouts in the event of a specified number
of days with temperatures or rainfall above or
below a specified trigger point.