FCHAPTER6ComingGame-ChangersBurgeoningCities,Cli.docx

F
CHAPTER6
ComingGame-Changers?BurgeoningCities,ClimateChange,and
Climate-InducedCatastrophes
uturedisasterrisks(acombinationofhazard,exposure,andvulnerabil
ity)may
change as a result of two powerful trends: burgeoning cities and
a changing
climate. The latest United Nations (UN) estimates suggest that,
globally, the urban
population exceeded the rural for the first time in 2008 (UN
Populations Division
2008).Inlessdevelopedregions,thisthresholdisexpectedtobereache
dbyaround
2020. How will changing distributions of population and income
in the context of
growingcitieschangeourexposureandvulnerabilitytonaturalhazard
s?Howwillthe
incidenceofclimateandweatherextremesaffectfutureeconomiesan
dwell-being?
Forexample,widespreadmigrationtocoastalregionsmaygreatlyincr
easeriskeven
if the climate were to remain constant, while increasing
prosperity may work to
reducerisk,eveniftheclimatehazardsthemselvesareincreasingorint
ensifying.
Andwhataboutclimate-
inducedcatastrophes,definedheretomeandisastersthat

occuronaglobalscaleandarelikelytobeirreversibleoveranyrealistic
timeframe
fordecision-
making?Forexample,themeltingoftheicesheetonGreenland,asa
consequence of climate change, could raise sea levels by seven
meters, and the
melting of the West Antarctic ice sheet could raise them by five
meters, flooding
manymajorcoastalareas.
The chapter starts with a discussion on cities, whose growth,
especially in the
developingworld,substantiallychangesexposureandvulnerability.
It thenanalyzes
howclimatechangecouldaffecthazardssuchastropicalcyclones,wit
haglimpseof
thesciencebehindtheprojections.Notethatthefocusisontheaddition
alhazards
induced by climate change, distinguishing them from changes in
hazards without
futureclimatechange.Moreover,thefocusonhazardsmeansthatthea
nalysisdoes
notaddressalltheeffectsofclimatechange.1
The risks and costs of climate-induced catastrophes, whose
global scale and
persistencedifferentiatethemfromdisastersonamorelocalandregio
nalscale,are
examinedlast.
Institutionsarethecommonthreadlinkingthethreepossiblegame-
changers.They
need to adapt to all risks—not just those from urbanization,
climate change, and
catastrophe—and function municipally, nationally, and globally.

There are no ready
recipestocreatethem,butmuchcanbedonetofosterthem.
The, World Bank, Bank World, and Nations (UN) United.
Natural Hazards, UnNatural Disasters, edited by World Bank
The, et al., World Bank Publications, 2010. ProQuest Ebook
Central,
http://ebookcentral.proquest.com/lib/apus/detail.action?docID=6
06027.
Created from apus on 2017-04-10 14:40:31.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k

P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
Cities:Risingexposure
Citiesareeconomicpowerhouses:theyoccupyonly1.5percentofthe
world’sland
areabutproducehalftheworld’sGDP.Andprosperouscitiesareecono
micmagnets,

attractingpeopleandinvestments.Theirprosperityarisesfromthediv
isionof labor
that thedensityofpeopleandassetsallows,andfromthe
lowercostofacquiring
productivity-increasing information and technology (“know-
how”). There are now 26
megacities(withmorethan10millionpeople),upfromeightin1950.T
he2009World
DevelopmentReportexaminedthese issues
ineconomicgeographyandconcluded
thatgovernmentsshouldnottrytopreventordiverturbanizationbutsh
ouldinstead
bettersupportcitiesandprovideneededservicestobothurbanandrura
lareas(each
has a different set of challenges). Building on the 2009 WDR’s
framework, this
sectionbeginsbyoutlininghowandwhycitiesgrowandwhyexposuret
ohazards
mayrisebutvulnerabilitymayfallintheaggregateasdensitiesandinc
omesincrease.
Citiesgrowfasterthancountries
Historically, output has grown by about 1 to 2 percentage points
more than
population,sopercapitaincomehasrisenalmosteverywhere.Muchof
thegrowth
has been in cities, where per capita income is higher. Among
150 of the world’s
largestcities,percapitaoutputisabout1.8timestheaveragenationalo
utput.And
urbanpercapitaincomeisonaveragetwicetherural.2Thisisnotnew:c
itieshave
long brought prosperity. Cities’ population is also growing. The
UN estimates the
worldurbanpopulation’ssharewillriseto70percentby2050.3About

halfthisgrowth
is “natural” (owing to the fertility of urban dwellers)
(Montgomery 2009), and the
remainderisduetoexpansion(whenadjoiningvillagesgrowtomeet)a
ndmigration
(map6.1).
Many cities are outgrowing the capacity of roads, water supply,
and sewage
disposal systems to serve their inhabitants. Services have not
kept pace largely
becausecitieshavenotinvestedenoughininfrastructure—
eveninthevauntedhomes
of high-tech industries like Bangalore, India. The reasons differ,
but many can be
traced to institutions that do not allow city administrations to
respond to people’s
needs:forexample,thefinancingarrangementsofIndiancitiescouldb
efaulted(Bahl
and Martinez-Vazquez 2008). Congestion, pollution, and
frustration may eventually
chokethecontinuinggrowthofsuchcities,but
theirpeopleareexposedtonatural
hazardstoday.Thesearethecitieswheredangersmaybeunnecessaril
yhigh.
Map6.1Citiesprojectedtohavemorethan100,000peoplebytheyear2
050
Central,
06027.

C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u
b
lic
a
tio
n

s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
Source:Brechtandothers2010.
Citygrowthwillincreaseexposure
Cities are largely founded at transport intersections—such as
ports, or at the first
bridgeonariverupstream.Somenaturalharborsreflectactivetectoni
cs.Flat land
close to the water was at a premium, reflecting unconsolidated
recent sediments,
often reclaimed for its value. Such land is vulnerable to both
flooding and ground
motionamplification.Forexample,SanFranciscowasoriginallyacit
ybuiltofwooden
buildings,largelydestroyedinthefirethatfollowedthe1906earthqua

ke.Thedebris
fromthatearthquakewasthenpushedintotheseatocreatemorereclai
medlandon
whichtheMarinaDistrictwasconstructed,onlytosufferhighlevelsof
damageand
ground settlement in the 1989 earthquake. Such growth
increases exposure and
vulnerabilitytohazardsunlesspeopletakeconsciousmeasurestoprev
entthem.
City-specific population projections to 20504 for this report are
combined with
geographicpatternsofhazardeventsrepresentativeof the1975–
2007period.The
projectednumberofpeopleexposedto
tropicalcyclonesandearthquakes in large
citiesin2050morethandoubles,risingfrom310millionin2000to680
millionin2050
fortropicalcyclones,andfrom370millionto870millionforearthqua
kes(map6.2).
Map6.2Exposuretocyclonesandearthquakesinlargecitiesrisesfrom
680millionpeoplein2000to1.5
billionpeopleby2050
Central,
06027.
C
o
p

yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u
b
lic
a
tio
n
s.
A
ll

ri
g
h
ts
r
e
se
rv
e
d
.
Source:Brechtandothers2010.
The growing exposure continues to vary by region. By 2050,
there will be 246
millioncitydwellersincyclone-
proneareasinSouthAsia,but160millioneachinthe
OECDand inEastAsia.AlthoughEastAsiahasfewerexposedpeople,
theurban
populationexposedtocyclonesisexpectedtogrowat2.2percentayear
,similarto
SouthAsia’s.Sub-SaharanAfrica’sexposuregrowthat3.5percent
isevenhigher,
reaching21millionurbandwellersby2050.
Exposure to earthquakes will likely remain the bane of East
Asia: 267 million in
2050,upfrom83millionin2000.ItisalsohighinLatinAmericaandthe
Caribbean
(150 million in 2050) and OECD countries (129 million in

2050). But the fastest
exposuregrowthisinSouthAsia(3.5percent),followedbySub-
SaharanAfrica(2.7
percent).
Thedensityofpeopleandeconomicactivitynotonlychangestheriske
quation—it
can also change the economics of disaster risk reduction
strategies. And what
applies
topopulationappliesevenmoretoeconomicassetsandoutput.Citiesa
re
enginesofgrowth,andfirmsprefertolocateinurbancenterswithgood
accessto
labor.Eachunitofareathereforegeneratesfarmoreoutputandhostsal
argerstock
of economic assets. This reflects the concentration and greater
economic value of
productiveassets—
aswellaspublicinfrastructureandprivateassetssuchashomes
—
incities.Theexposureofeconomicassetstonaturalhazardsincitieswi
llthusbe
considerably higher than in rural areas. But greater exposure
need not increase
vulnerability:muchdependsonhowcitiesaremanaged.
Citymanagementwilldeterminevulnerability
Acoretaskforcitiesistoprovide,coordinate,anddisseminateinforma
tionsothat
land, housing, and insurance markets can operate efficiently.
Data on hazard
probabilitiesandthevulnerabilityofstructuresandpeople feed
intocomprehensive
riskassessments.Theseshouldbemadeaccessibletoall.Suchinforma

tionallows
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k

P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
residents to make informed location choices and markets to
price hazard risk
appropriately. It also provides the basis for the emergence of
private insurance
markets.Anditservesasasoundbasisfortransparentzoningdecisions
andother

land use restrictions. And while hazard mapping has been
performed for many
decades,newtechnologiesallowconstantupdatingofinformationata
fairlylowcost.
Makingthesetechnologiesaccessibletocities—
notonlythelargest,butalsosmaller
andmedium-sizedcitieswithlimitedlocalcapacity—
shouldbeapriority.
Forlarge-
scalecollectivehazardriskreductioninvestments,thecostsandbenef
its
dependinlargepartonthedynamicsoftheurbaneconomy,particularl
yonthevalue
of land. In dynamically growing cities, where land is scarce,
large investments to
make landhabitableor reducesignificant riskmaywellbe
justified.Anexample is
large-scale land reclamation in Hong Kong SAR, China, and
Singapore. Limited
expansion options in the vicinity of high economic density raise
the value of land
significantly.Thisshiftsthecost-
benefitratioinfavoroflargeprotectiveinvestments.
Astricttestiswhetheradeveloperwould,inprinciple,bewillingtopay
apriceforthe
reclaimedorprotectedlandthatreflectsthecostoftheintervention.
All cities are not equal, and the viability of large-scale disaster
reduction
infrastructure will be different in cities with stagnant economies
and little or no
population growth. Today, this is a phenomenon in mature
economies with
demographicdeclinesor incountrieswithstronggeographicshifts
ineconomicand

population centers (Pallagst 2008). Examples are the former
socialist countries in
EuropebutalsopartsofScandinaviaandtheMediterraneancountries,
aswellasthe
oldindustrialcoreoftheU.S.midwest.Overtime,givendemographict
rendsinmany
middle-income countries, “shrinking cities” may also emerge in
some of today’s
emergingeconomies,suchasthoseinEastAsia.
Public investments in the wake of Hurricane Katrina sparked
debate over large-
scaleprotective
investmentstoencouragetherebuildingofNewOrleanswithinthe
pre-
Katrinacitylimits.Morethan$200billionoffederalmoneywillbeuse
dtorebuild
the city. Some have argued for providing residents of areas
behind massive flood
control infrastructure with checks or vouchers, and letting them
make their own
decisions about how to spend that money—including the
decision about where to
locateorrelocate.Thechoiceisbetweenspending$200billiononinfra
structurefor
residentsorgivingeachresidentacheckformorethan$200,000—
inaplacewhere
annual per capita income is less than $20,000 and which
reached its peak of
economic importance in 1840.5 There are, of course, political,
cultural, and social
factors that have to be considered in the decision whether to
reconstruct, but this
examplenonethelessshowsthedifficulttradeoffsthatshrinkingcitie
sface.
Reducing urban hazard risk through large-scale infrastructure

must consider the
dynamicsofcitydemand.Insomedevelopingcountries,
infrastructureinvestment—
long-livedcapitalstock—
islikelytopeakinthecomingfewdecades.Thesetasksare
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B

a
n
k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
perhapsmoredauntingthaninthepast,givencities’emergenceincoun

trieswhere
power is increasingly federal. The challenge is at all levels of
government—from
federal to urban development ministries to small-town mayors.
But the payoffs in
savedlivesandavoideddamageswillbehigh.
Climatechange:Changinghazards,changingdamages
Climate-
relatedhazards(“extremeevents”)haveresultedinanaverageof$59b
illion
ayearinglobaldamages(EMDAT2009)from1990through2008,or0.
1percentof
world product in 2008. Tropical cyclones account for 44
percent, and floods 33
percent.
Even without climate change, economic development and
population growth are
expected to increase the baseline damages from extreme events
over the next
century(figure6.1).Ifthereisnoconsciouschangeinadaptationpolici
estoextreme
events, baseline damages without climate change are expected
to triple to $185
billion a year from economic and population growth alone.
Floods and tropical
cyclonesareexpectedtocontinuetobetheprominentsources.Butheat
wavesare
expectedtobecomemoreprominent.
Thereiswidespreadconcernthatclimatechangecouldincreasefuture
damages
fromextremeevents(IPCC2007a,IPCC2007b,WorldBank2009).Ea
rlierstudies
projected increased tropical cyclone activity alone might result

in additional annual
damagesintheUnitedStatesof$100to$800million6andglobalannua
ldamagesby
$630million(Pearceandothers1996).Morerecentstudiessuggesttha
tadoubling
of greenhouse gas concentrations could increase tropical
cyclone damage by 54
percent to 100 percent in the United States and double tropical
cyclone damage
globally.7Somestudiesofhistorictrendsofextremeeventinsurancec
laimsfindthat
extremeeventsarerisingatarapidandevenexponentialrate(SwissRe
2006;Stern
2007).However,thesetrendlineanalysesdonotseparatechangesinth
eexposed
populationandchangesintheextremeeventsthemselves(Pielkeand
Downton2000;
Pielkeandothers2008).
Analysis commissioned for this report uses an integrated
assessment model
combiningscienceandeconomicstoestimatetheadditionaldamagefr
omhazardsas
aresultofclimatechange.8Whiletheanalysisattemptedtoestimateth
eadditional
damagefromallhazards,theanalysisofpotentialchangesinthelocati
on,frequency,
andintensityof
futuretropicalcyclonesisthemostcomplete.Box6.1explains the
methodologyusedfortropicalcyclones.
Figure6.1Current(2008)andprojected(2100)damagesfromextreme
eventswithoutclimatechange

Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u
b
lic

a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
Note:Damageswithoutclimatechangeareprojectedtoincreasebecau
seofincomeandpopulationgrowth.
Source:MendelsohnandSaher2010.
Afewcaveats:
Aspectsofthescienceremainuncertain.Althoughallclimatemodelsa
gree
theplanetwillwarm,theydonotagreeonthemagnitudeofthechangesa
nd

howtheywillbedistributedacrosstheplanet:theresultsarequitediffe
rent
acrosstheclimatemodelstested(boxfigure6.1).
Theanalysisdoesnotmeasurealltheimpactsofclimatechange,justth
ose
ofclimate-relatedhazards.9
Thestudyreportsonlythedirectdamagesfromclimate-
relatedhazards.For
example, the impacts on ecosystems are not measured. There are
other
indirecteffectsofdisasters,whicharedifficult
tomeasure,asdiscussedin
chapter2.
Theanalysisdoesnotaddresspossibleinteractionswithothereffectsf
rom
climate change. For example, although the tropical cyclone
analysis does
takeintoaccountstormsurge,itdoesnotconsidertheinteractionbetw
een
stormsurgeandsealevelrise.Whethertheinteractionbetweenarisein
sea
levelandstormsurgeis“additive”or“superadditive”woulddependo
nthe
assumptions about adaptation to sea level (for example, building
sea walls
wherepermissibleor locatingpeopleoutofharm’sway).Such
interactions
areanimportantareaforfuturework.
Box6.1Estimatingadditionaldamagesfromclimatechange-
inducedtropicalcyclones
TheanalysisbeginswiththeA1Bemissionscenariothatassumesamod
eratemitigationprogramwillstabilize
concentrationsat720ppm.Fourclimatemodelsarethenusedtopredic

tchangesinclimateby2100.Because
highlydamagingtropicalcyclonesaresoinfrequent,itmighttakehun
dredsofyearsofactualdatatobeableto
detectrobustandstatisticallymeaningfulchangesinthedistributions
ofstormfrequencyandintensityfrom
climatechange.Soforeachclimatescenario,tropicalcyclonesarepre
dictedbasedonaspecializedtropical
cyclone model that simulates the creation, development,
movement, and termination of storms (Emanuel,
Sundararajan,andWilliams2008).Tensofthousandsofstormsaresi
mulatedsothatevensmallchangesin
thedamagedistributioncanbedetected.Mostofthesimulatedcyclone
“seedlings”(potentialstorms)never
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0

.
W
o
rl
d
B
a
n
k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv

e
d
.
becometropicalcyclones.Theremainingeventsconstitutethetropica
lcycloneclimatologyassociatedwiththe
projectionsofeachparticularglobalcirculationmodel.
Climate change is predicted to have very different impacts on
tropical cyclones across the globe. The
intensity,frequency,andtracksoftropicalcyclonesaresensitivetoan
umberofenvironmentalconditions,not
allofwhichchangeinthesamedirectionwhenclimatechanges.Forexa
mple,anincreaseintemperature
increasestropicalcycloneintensity,otherthingsbeingequal,butwin
dshearcaninhibitstormformationand
development.Intensitiesandfrequenciesthereforechangeacrossthe
differentclimatemodels.Boxfigure6.1
showsthepercentagechange ofcoastalpower
dissipation,ameasureof thepotentialdestructivenessof
tropicalcyclonesoverthefourmodelsandfiveoceanbasins.Formosto
ftheclimatemodels,thecyclone
simulationindicatesasmallincreaseintheintensityofstormsintheAtl
anticandNorthwestPacificOceans.
One climate model predicts an increase in intensity at landfall
in the North Indian Ocean and Southern
HemisphereOceanbutmostofthemodelspredictadecreaseinintensit
yintheseoceansornoeffectatall.
Notethatincreases(decreases)instormintensityimplyclimatechang
ecausesdamages(benefits).
Boxfigure6.1Intensityoftropicalcycloneswillvaryoverthefiveocea
nbasinsby2100

Note:CNRM,ECHAM,GFDL,andMIROCaretheclimatemodelsuse
dfortheprojections.
Source:WorldBankstaff,basedonEmanuel,Sundararajan,andWilli
ams2008.
Thedamagefunctionisestimatedusinganinternationaldatasetofglo
balhazarddamagesfrom1960to2008
(EMDAT2009).Damagespereventareregressedonincomepercapita
andpopulationdensitytodetermine
thesensitivityindifferentlocations.Thedamageresponsetotheinten
sityofatropicalcyclonewasestimated
using US data from the National Oceanic Atmospheric
Administration. Future damages (without climate
change) are projected using predictions of future income and
population. The estimate of climate change
damageisthedifferencebetweenthedamagecausedbyalltropicalstor
msinthefutureclimateminusthe
damagecausedbytropicalcyclonesinthecurrentclimate.Notethatth
efactthatfuturebaselinespredictmore
peopleandcapitalwillbeinharm’swayimpliesthatclimatechangewil
lhavelargereffects.Empiricalresults
describedbelowrevealthatcyclonedamagesareahighlynonlinearfu
nctionofstormintensity.A1.1percent
declineinminimumatmosphericpressureatsealeveldoublesthedam
agesfromtropicalcyclones.
Theanalysismakescertainassumptionsofwhattheworldwilllooklik
ein100
years.Economicandpopulationgrowthmaybequitedifferent.
Relevantpoliciesthatwouldaffectadaptationmayalsochange.Forex
ample,
policies that encourage (discourage) risky development in
hazardous areas
wouldincrease(decrease)overalldamages.
Internationalreportingofextremeeventsanddamagesremainsuneve
n.As

datasets improve, itwillbepossibleto improvepredictionsof
international
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k

P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
damages.
Withthesequalificationsinmind,thekeyfindingsareasfollows.
Damagesareexpectedtoincrease

Withoutclimatechange,expectedtropicalcyclonedamagesincrease
from$26billion
today to $55 billion by 2100 because of the growth in income
and population.10
Climatechangecouldaddabout$54billionworthoftropicalcycloned
amageseach
year, doubling future baseline damage. The estimated increase
in damages from
climatechangevariesacrossclimatemodelsbetween$28and$68billi
on(or51to
124 percent of the future baseline). These estimates are
sensitive to the elasticity
between damages and income. If the income elasticity of
damages were unitary
(instead of 0.41, as estimated), future baseline damages become
$195 billion and
climatechangeaddsabout$178billion––
almostdoublethebaselinedamages.
Averagesmaskextremes
Theestimatesoftheabovedamagesarein“expectedvalue”termspery
ear.Butthe
damages are not expected to come in a steady stream. Even with
the current
climate, 10 percent of tropical cyclones are responsible for 90
percent of the
expecteddamages.Evenifclimatedoesnotchange,damageswillvary
agreatdeal
fromyeartoyearanddecadetodecade.Climatechangeisexpectedtosk
ewthe
damagedistributionoftropicalcyclonesandislikelytocauserare—
butverypowerful
—
tropicalcyclonestobecomemorecommon.Withawarmedclimate,th

e10percent
oftropicalcyclonesthatcausethemostdamagewillberesponsiblefor
93percentof
theexpecteddamages.
Climatechange“fattensthetail”ofthetropicalcyclonedamagedistrib
ution.Forthe
UnitedStates,destructivestormsthatwouldcomeevery38to480year
sgiventhe
currentclimate,wouldcomeevery18to89yearswithfutureclimatech
ange.Figure
6.2 illustratesthis
foronespecificclimatemodel(MIROC).11Mostof thecyclones
withandwithoutclimatechangeinvolvedamagesinthetensofbillions
ofdollarsor
less.Thesestormsmaybecomeevenlessfrequentwithclimatechange
.But,very
rarely,averypowerfulstormwillstrikeaveryvulnerablelocationcaus
ingdamages
uptoatrilliondollars.Thisseeminglysmallshiftinthetailofthedistrib
utionisshown
as“returnyears,”whichshowhowmanyyearswouldelapse,onaverag
e,between
occurrencesofastormcausingaspecificlevelofdamage(figure6.2).E
venthough
veryrareanddamagingstormsarepartoftoday’sclimate,theywillbec
omemore
frequentinawarmerclimate.Forexample,usingthefuturebaseline,a
$100billion
stormisestimatedtohappenonceinahundredyearsintheUnitedStates
giventhe
currentclimate.Withafuturewarmedclimate,itisexpectedtohappen
onceinabout
56years.
Figure6.2Climatechangeshortensthereturnperiodoflargestorms

Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u

b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
Note:Thefigureshowsthereturnperiodfortropicalcyclonesofdiffer
entintensityintheUnitedStatesforonespecificclimatemodel
(MIROC).A$100billionstormisestimatedtohappenonceina100yea
rsintheUnitedStatesgiventhecurrentclimate.Withafuture
warmedclimate,itisexpectedtohappenonceinabout56years.
Source:Mendelsohn,Emanuel,andChonabayashi2010a.
Damageswillvaryacrosslocationsandwithin-

countryvariationindamagesislikely
Thebulkofthetropicalcyclonedamagesfromclimatechangefallson
NorthAmerica
($30 billion) and Asia ($21 billion). Three countries bear 90
percent of global
damages:theUnitedStates($30billion),Japan($10billion)andChin
a($9billion).
However,whendamagesarescaledbyGDP,theCaribbeanislandsare
amongthe
worsthit.
Theglobaltropicalcycloneanalysisisbasedonnationaldatasetssotha
titisnot
possible to show how effects vary within most countries.
However, for the United
States, detailed data at state and county levels are available
concerning tropical
cyclonedamages,intensities,andfrequencies,allowingspatiallydet
ailedanalysisto
beconducted.Box6.2describestheseresults.AtleastfortheUnitedSt
ates,thereis
awiderangeofeffectswithinthecountry.Itislikelythatforlargecount
riesatleast,
therewillbesubstantialintracountryvariation.
Box6.2Withincountryeffects:ThecaseoftheUnitedStates
TheclimatechangestudyoftropicalcyclonesintheUnitedStatesused
informationaboutthecountiesthat
eachtropicalcyclonestruck.Thespatialscaleoftheanalysiswasmuch
finerthanthecountryscaleforthe
globalanalysis,permittinglargeintracountryvariationsindamagest
obeseen(boxfigure6.2).Mostof the
damagesfromtropicalcyclonesintheUnitedStatesoccurintheGulfst
atesandFlorida(87percent).The
damagesfallquiterapidlyasonemovesnorthalongtheAtlanticseaboa

rd.Atleastinlargecountries,therewill
besignificantintracountryvariationinextremeeventdamages.Thees
timateddamagesalsovaryagreatdeal
acrossclimatemodels.TheGFDLandMIROCmodelspredictmuchlar
gerdamagesthandoestheCNRM
model.
Boxfigure6.2TropicalcyclonedamagesintheUnitedStatesareconce
ntratedintheGulfCoastand
Florida
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W

o
rl
d
B
a
n
k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.

Note:Damagesareinbillions$/yrfor2100.
Source:Mendelsohn,Emanuel,andChonabayashi2010b.
These results provide insight into adaptation to tropical
cyclones. The
preponderance of damages from tropical cyclones is caused by
rare and very
powerfulstorms.Toadapt,onemaybetemptedtobuildextensiveseaw
allsalong
the coast as one might against sea level rise. However, very
powerful storms are
likelytooverwhelmsuchmeasuresmakingthemineffective.Building
higherfortified
sea walls in selected places of high value and population
density (cities) may be
justified but the costs have to be weighed against the damages.
Further, in some
locationsitwillbedifficulttobuildsufficientlyhighseawalls.Inthisc
ase,retreatmay
be the only option. In places that cannot be defended, land use
rules could be
developedtoencouragerobustlanduses,suchasopenspaceandagricu
lture,which
can survive occasional storms. Further research into efficient
adaptation to such
eventsisahighpriority.
The analysis also attempted to estimate additional damages from
other (non-
tropicalcyclone)extremeevents(box6.3).Forreasonsexplainedbelo
w,estimating
suchdamagesisinherentlymorechallenginganduncertain.Conseque
ntly,thispartof

theanalysisprovidesapointofdepartureforfurtherworkonlearninga
boutthelikely
directionandextentofthedamagesfromtheseevents.
Box 6.3 Estimating additional damages from climate change-
induced extreme events (other than
tropicalcyclones)
The analysis for the non-tropical cyclone extreme events
(floods, droughts, heat waves, and cold events)
followsasimilarthoughnotidenticalapproach.Itismoreuncertainbe
causethelinkbetweenclimatechange
and these extreme events is more difficult to discern
(Mendelsohn and Saher 2010). For the SRES A1B
emission scenario (IPCC 2000), three climate models (CNRM,
ECHAM, and GFDL) are used to estimate
changesinfuturetemperatureandprecipitationmeansandvariances.
Thelinkbetweendamagesfromthese
eventsandtheseclimatemeasuresisthenmeasuredusinginternationa
ldatafrom1960through2008.The
climateforecastsarethencombinedwiththecoefficientsfromthedam
agefunctiontopredictfuturedamages
in2100.Damagesfromclimatechangewerecalculatedasthedifferenc
ebetweendamagesin2100inthe
warmedclimateanddamagesin2100withthecurrentclimate,whileco
ntrollingforincomeandpopulation
growth.Changesintornadoes,thunderstorms,andhailareassumedtoi
ncreaseinfrequencyinthesame
proportionasfoundinaclimatechangestudyofthunderstormsintheU
S(Trappandothers2007).
Central,

06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u
b
lic
a
tio

n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
Following this approach (and related assumptions detailed in
the background paper), baseline damages
(withoutclimatechange)fromsuchextremeeventsareexpectedtoinc
reasefromtheircurrentlevelof$28
billiontoabout$113billionayearin2100.Climatechangeisthenexpe
ctedtoaddbetween$11billionand$16
billion a year of damages by 2100. The estimates presented in
this analysis are inherently uncertain. The
scientificresultsforthunderstormsintheUnitedStatesmaynotholdin
otherlocationsnormaytheyapplyto
hailandtornadoes.Thedamagefunctionlinkingdamagestoclimateva
riablesmayunderestimatedamages
becausetheavailabledataareatanational levelwhereasmanyof

theseextremeeventsoccuratafiner
spatialscale.Forexample,flashfloodsdependonprecipitationinsom
etimessmallareas,whicharepoorly
measuredbyvarianceestimatesacrossmuchlargerareas.Floodsdepe
ndonlocalhydrologicalconditions
that are only crudely accounted for in a global analysis. These
uncertainties are over and above other
uncertaintiesinherentinanyclimateprojections.Muchmoreworkisn
eededtogenerateprecisedamagesfrom
suchevents,anditisimportantnottoplacemuchweightonthespecific
numbers.
Source:WorldBankstaff,basedonMendelsohnandSaher2010.
Estimating impacts of climate change–induced extreme events is
relatively new.
Continued research will improve our understanding and ability
to estimate the
impacts.Betterdatawillalsolikelyhelp.Disastersarepoorlymeasure
devenunder
currentclimaticconditions.Severalcountriesdonotevenreportdama
ges,andthe
globaldamagedatasetsdonotreporteventintensity.Eventheverylarg
estextreme
events, tropical cyclones, are poorly measured on a global scale.
Although the
number of storms has been well documented since the advent of
satellites, the
intensity of these storms is still not measured globally. More
accurate and global
measurementsofbothstormsanddamageswilllikelyleadtobetterund
erstandingof
howclimatechangeleadstodamagesfromhazards.Finally,thereisthe
questionof
scale. It is likely that sub-national analysis would provide even
more accurate

estimates.
Climate-relatedcatastrophes:Deep-
futuredisasterswithaglobalfootprint
Theusualfinaleventofatragedyisacatastrophe(fromtheGreekkatast
rephein,to
overturn).Wedefineacatastrophehereasaneventthatisfairlytoextre
melyrare,
thatseverelyaffectsbroadswathsoftheworld,andthatislikelytobeirr
eversible
overanyrealistictimeframe:examplesincludeavirulentpandemic,a
nuclearwar,or
an asteroid collision. Climate-related catastrophes differ in
three ways: they unfold
moreslowly,providingapotentialopportunity toprepare;
theyresult fromacause
thatthepublicmaynotsoreadilygrasp;andnumerousactorsarerespon
sible.The
occurrence of nuclear war, perhaps the greatest threat in the last
half of the 20th
century, rested on the decision of a few people. So this was
analyzed in a game
theoreticsettingwithdifferentdegreesofcooperation.Theresultwas
deterrence—
mutuallyassureddestruction,withtheappropriateacronymMAD.In
contrast,climate
catastrophesresultfromtheconsciousself-
interestedbehaviorofbillionsofpeoplein
several countries living in different circumstances, so effective
international
agreementsaremoredifficult.12
The scientific community has identified several catastrophes
that climate change

a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
mighttrigger. It
isalsopossiblethatcatastrophescouldbetriggeredwhenseveral
smallerormorelocalizedimpactscascade,thoughthisremainsonlyat
heory.
Considerfourtypesofcatastrophe:
Drasticsealevelrise.Satelliteandtidegaugemeasurementsshowthat
sea
level rise has accelerated at about 3.4 millimeters per year since
satellite

measurementsbeganin1993.The2007IPCCreportprojectsagradual
rise
of 0.2 to 0.6 meters over the 21st century from thermal
expansion of the
oceans.ButthedislodgingandmeltingoftheWestAntarcticorGreenl
andice
sheetseventuallycouldraisesealevelsby5–
7meterseach.Thespeedof
suchamassiveincreaseinsealevelisasubjectofcurrentresearch.Itmi
ght
takecenturiesforanimpactofthisscaletounfoldfully,thoughitislikel
ythat
ameterofsealevelrisecouldoccurinthiscentury,withaprobableuppe
r
limitofabouttwometers(Rahmstorf2007).Ineithercase,theemissio
nsto
trigger large-
scalesealevelrisescouldbegeneratedinthiscenturyalone.
Suchriseswouldfloodlargeinhabitedareasanddramaticallychangeh
uman
activity. For example, a five-meter rise would require mass
migrations of
coastalpopulationsandtotalevacuationoflow-
lyingislands.Althoughhuman
societycouldadapt,thischangewouldbeextremelydifficultandcostl
y.
Disruptionofoceancurrents.Large-
scalemeltingofpolaricesheetswould
increasefreshwaterinthecoldNorthAtlanticOcean,weakeningthefl
owof
warm currents from lower latitudes. This diminution of the
Atlantic
ThermohalineCirculation(THC)couldaffecttheclimateofmuchofn
orthern
Europe.13

Large-scale disruptions to the global ecosystem. The impacts of
even
gradual climate change could suddenly disrupt a variety of
ecosystem
services.Thesecouldincludereducedbiodiversity,reducedaccessto
water
inthecurrentlocationsofsignificantpopulations,acidificationofoce
ans,and
rapid changes in land cover on a large scale. The social,
economic, and
environmentalconsequencesoftheselosses,notknown,couldbevery
large.
Accelerated climate change from large releases of trapped
methane.
Warming beyond a certain point could release into the
atmosphere large
quantities of methane in oceans and permafrost. This possibility
is an
exampleofa“tippingpoint,”whenlargeandpossiblyirreversiblecha
ngesin
the climate might result from exceeding a poorly understood
threshold.
Because methane resides in the atmosphere for only a few
decades, the
direct effect would be a temporary if powerful acceleration of
temperature
increase.Butsucha largeandrapid increaseintemperaturecould in
itself
leadtosevereandirreversibleconsequences.RapidmeltingofArctics
eaice
isalreadyhappening,andlargeandrapidwarmingcouldsetinmotiono
ther

Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u
b
lic

a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
factors (such as accelerating melting of heat-reflecting snow
cover) that
causeafurtheraccelerationinclimatechange.
A second concern is that multiple smaller hardships or
disruptions from climate
changeoverashorterperiodcouldcombinetocreateacumulativeeffe
ctworsethan
the sum of the independent hazards. For example, a worsening
of droughts and
damagestoecosystemsinmanyareasoverashortperiodcouldleadtoe

conomic
andsocialdisruptionforlargenumbersofpeoplefromthedirecteffect
softhemore
localized impacts. But it could also lead to forced migration,
armed conflict, and
widespreadfailuresofinstitutions.
Gradualorcascading,muchisstillbeingdiscoveredanddebated.Seal
evelrise
estimatesarethemostconcreteindicationofthepotentialforcatastrop
hicimpacts
from climate change. But even sea level rise scenarios involve
uncertainties about
vulnerabilityandadaptation.14Thesizeoflosseswilldependonthesp
eedofchange
insealevelriseaswell,onthedegreeofexposurerelativetocurrentcon
ditions,and
onmeasuresthatcanbetakentoreducetheimpacts.Thepotentialmagn
itudesand
likelihoodsofotherworrisomecatastropherisks,suchasabruptchang
esinlandand
oceanecosystemsorthepotentialfor“runaway”accelerationofclima
techangefrom
methanereleases,aredifficulttogauge.
Adecisionframeworkforcatastrophes
Thetriggersorthresholdsthatcouldsetoffcatastrophesareuncertain,
asarethe
probabilities of occurrence and the consequences, though recent
scientific
assessments indicate that the risks of climate change generally
look worse today
thansomeyearsago(Smithandothers2009).Expertjudgmentsmustb
ebroughtto
bear in the absence of more concrete information. How then

should policymakers
weighthecostsandbenefitsofalternativepolicyresponses?
Standardcost-
benefitanalysescanbeextendedtoincorporateriskswithknownor
subjectively specified probabilities, but both probabilities and
types of potential
outcomes are unknown for climate catastrophes. The possibility
of catastrophic
climate change is characterized by deep structural uncertainties
in the science
coupledwithaneconomicinabilitytoevaluatemeaningfullythewelfa
re lossesfrom
high temperatures. (Analyzing the most recent available climate
models, Weitzman
(2009)concludesthatthefutureholdsabouta5percentchancethattem
peratures
will rise by about 10 degrees Celsius—a world difficult to
imagine.) The costs of
mitigation also are uncertain, as they depend on the pace of
future technological
changeandthewaypoliciesandregulationsoperateacrosscountries.
Nevertheless,
some weighing of options by balancing pros and cons is
desirable and uncertainty
doesnotjustifyinaction.Butarguingfortoorapidandaggressiveinter
ventionscould
leadtomeasuresthatareverycostlyrelativetothepotentialreductions
inrisk.
Whileuncertaintycannotjustifyinaction,ithasimplicationsforhowd
ecisionmaking
Central,

06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u
b
lic
a

tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
is undertaken. Posner(2004) suggests a tolerable-windows
approach: a range of
plausibleestimatesareestablishedtoascertainalevelofrisk-
reductioneffortwhere
thebenefitsclearlyexceedthecostsandalevelwherecostsclearlyexce
edbenefits.
Policiesthencanbeadoptedthatfallwithinthiswindow.
Whencostsareincurredwellbeforethebenefits,asintakingmeasurest
omitigate
the potential for climate change catastrophes, the selection of a
discount rate to

compareearliercostswith laterbenefits
isafocusofuncertaintyanddebate.The
2010 WDR notes there is no consensus on the “correct” discount
rate for climate
change evaluation (and may never be). But decisions about
responses to climate
changecatastropherisksinvolvethepresentgenerationmakingaltrui
sticchoiceson
behalfoffuturegenerations.Thechoiceofalowerdiscountrateforval
uingreduced
long-term climate change hazards involves current generations
reducing their well-
beingforthebenefitoffuturegenerations.Thisisalsotrueforotherinv
estmentsthat
improvetheprospectivewell-beingoffuturegenerations.
Aportfolioofresponses
Dealing with catastrophic threats hinges on policies for dealing
with “fat tail” risks.
Climate change is expected to worsen the distribution of
damages from tropical
cyclonesandthisshiftwill takeplace in theextremerighthandtailof
thedamage
probabilitydistributionfunction,fatteningthetail.Policiestoaddres
stailrisksdepend
inpartonsociety’swillingnesstodevoteresourcestoreducetheproba
bilityandlikely
impactof therisk, relative tobenefits fromotherusesof
thoseresources.Sucha
comparison isverydifficult
toquantify,especiallywhenconfrontedwithwell-known
behavioralbiasesforcatastrophiceventsandwhentherearecompetin
gcatastrophic
risks.Withoutsuchestimations,prudenceinrespondingtocatastroph
icthreatscalls

for a portfolio of measures that emphasizes learning and mid-
course corrections
(noting however the tremendous inertia that exists in the climate
system, the built
environment,aswellasininstitutionsandbehaviors,WDR2010).Abr
oaderportfolio
of measures is desirable because of the uncertainties
surrounding the costs and
potential effectiveness of individual measures. Thus,
incorporating several distinct
measuresmakestheresultingsetofpolicyoptionsmorerobust.Thepor
tfolioshould
include:
Rapidemissionsreductiontostabilizegreenhousegasconcentrations
inthe
atmosphereatsomelevellowenoughtoachieveadesiredreductionint
he
perceivedriskofcatastrophe.Differenttechnologicalpathscouldbef
ollowed
toaccomplishthis,anditisvirtuallycertainthatnosingleapproachwo
uldbe
successful.Rapidscalingupofrenewableenergycertainlywouldbep
artof
the response. But given continuing uncertainties about the
future cost and
physicalavailabilityofdifferenttypesofrenewablesandourabilityto
store
energytooffsettheinherentintermittencyofmostrenewables,thisres
ponse
Central,

n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
also would require addressing expanded nuclear power and
introducing
carboncaptureandgeologicalstorageonaverylargeorevenglobalsca
le.
Variouslargescaleadaptationmeasuresimplementedacrosstheworl
dover
the medium term, beyond efforts by individuals and single
governments, to
anticipate and significantly reduce the potential impacts of a
climate
catastrophe.Prioritymeasureswouldincludeextensivechangesinlan
duse
policies and practices to limit further increases in coastal area

vulnerability
andtoexpandandfortifyprotectedareastosafeguardcriticalecosyste
ms.
The adaptation measures could even include large-scale
anticipatory
relocationsofespeciallyvulnerablepopulations,suchasthosevulner
ableto
anticipated sea level rises and increases in storm surges. With
such
relocation would come the need to rebuild infrastructure and
other fixed
capital.
These two categories of actions may not be enough to
satisfactorily lower the
chanceofcatastrophes,particularlyiftheworldcannotcometoanagre
ementabout
sharing the burden of mitigation efforts. It is therefore also
necessary to consider
geoengineeringasanotherpotentialmeasuretoreducetheriskofcatas
trophe(box
6.4).
Dealingwith the threatofcatastrophicclimatechange isanexercise
inreducing
uncertainty with only a limited ability to assess the results.
Considerations in
developingaportfolioof responses include thecostsof
thevariousmeasures, the
lead times needed (particularly important when some
uncertainties may decline as
scienceandtechnologyimprovebutinertiaremainsverylarge),andth
einformation
abouttheirprospectiveeffectiveness.Theportfoliocanchangeoverti
measmoreis
learned about the nature of catastrophe risks and the costs and

effectiveness of
differentresponses.Sincenoclimatecatastrophehasbeenexperience
dinrecorded
memory, people may underestimate or overestimate this “virgin
risk” (Kousky and
Zeckhauser2010).
Box6.4Geoengineering’spotentialandpitfalls
SomeeffectsofadoublinginCO2concentrationcouldinprinciplebeo
ffsetbyblockingasmallpercentageof
sunlight reaching the Earth’s surface. The most commonly
discussed option for reducing absorbed solar
radiationinvolvesseedingtheupperatmospherewithparticulatematt
ertoreflectsunlight.Otherapproaches
include increasing the reflectivity of the earth (massive rooftop
retro-fits), changing cloud cover, and even
building mirrors in space as a planetary “sunshade.” Other types
of geoengineering include increased
absorptionofCO2byoceansorgiantmachinestocaptureCO2fromthe
atmosphere.Allthesemeasures
haveknownsideeffects,someofwhichwould
induceunknownbutpossibly largechanges in theclimate
system.
Geoengineeringcouldarrestorpotentiallyavertcatastrophesinduce
dbyclimatechange.Butadjustingthe
earth’stemperaturebyreflectingsunlightmayadverselyaffectotherc
limatevariables,suchasprecipitation.
Oneclearpitfallofgeoengineeringistechnological.Launchingreflec
tiveparticlesintotheupperatmosphereto
increasetheearth’sreflectivitywouldneedtobecarefullycontrolledf
ortworeasons.First,theparticlesremain

a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
in theatmosphereonlybriefly,soonce
initiatedthemethodwouldneedtobesustained indefinitely.And if
stopped, the effects of manmade global warming would be felt,
essentially, all at once. The effect of
experiencingaccumulatedimpactsallatonceisunknown.Evenmoref
undamentalisthecurrentuncertainty
aboutnegativesideeffects,includingpotentialalterationofthehydro
logiccycleandoceanacidification.Based
oncurrentknowledge,thereisnowaytoknowifgeoengineeringcould
becarefullycontrolledtotheextent

necessarytoprovidesomeprotectionfromfurtherwarmingwhileeffe
ctivelylimitinganysideeffects.
Asecondreasoncomesfromthestrongincentivestodeploysuchtechn
ologiesunilaterally.Theproblemof
internationalcooperationinmanaginggeoengineeringistheinverseo
fachievinginternationalcoordinationfor
drastic mitigation. With mitigation, the incentives for acting
unilaterally are extremely weak because of the
strongincentivetofree-
ride.Withgeoengineering,givenapotentialforlowdirectcostsandfai
rlyimmediate
directbenefitsofimplementation,incentivestoactunilaterallycould
beverystrong—especiallyinthefaceof
severethreatsfromclimatechange.16So,itmaybeimpossibleforcou
ntriestocrediblycommittoabstaining
from geoengineering. Also, how might potentially beneficial
uses of geoengineering be distinguished from
hostilemeasurestoinflictharmonothercountries?Moreover,howwo
uldpotentialconflictsamongcountries
overtheimplementationofgeoengineeringberesolved?Forexample,
supposethatcountryAseekstolocally
coolitsclimateandstimulateraininanefforttoprotectitsharvestandst
aveofffamine.ButwhatifcountryA’s
application of geoengineering had side effects that threatened
crops or water supplies in country B? This
questionisparticularlytroublingifcountryBabutsA,andisahistorica
lrivalorenemy.Forthesereasons,it
wouldbepreferabletoundertakeinternationallyfundedandcoordina
tedresearchongeoengineeringprecisely
sothatitspotentialapplicabilityandriskscanbewidelyunderstood.
Source:WorldBankstaff.
Examiningcurrentandpotentialcostsofalternativemeasuresandcon
sideringtheir

effectivenesscanhelpprotectagainstpossiblebiases.Thepotentialfo
rcatastrophe
certainly makes aggressive action more desirable, but how much
more remains
uncertain. Postponing sound measures to curtail the growth of
greenhouse gas
concentrations will reduce the effectiveness of “crash”
emissions abatement and
massivelyincreasecosts.15Similarly,postponingstrongerlanduse
measurestolimit
growth of coastal settlements will greatly increase the cost of
later adaptation
throughrelocation.
Anyportfolio
foraddressingcatastropheriskwillneedtobeadjustedover time.
Onerobustconclusionfromthecomparisonofresponseoptionsisthat
asignificant
investmentinreducingthecostofimplementationandincreasingthee
ffectivenessof
eachoptionshouldbeahighpriority.Effortstoimproveunderstandin
gofthepotential
ofgeoengineeringandtolowerthecostsandpotentialrisksofveryrapi
dmitigation
options are a high priority. Given the likely high costs of large-
scale anticipatory
adaptationmeasures,amorecautiousapproachwouldfocusfirstonin
creasingthe
prospects for the survival of critical ecosystems and placing
some limits on the
growthofsettlementsinmoreat-riskareas.
ConnectingthethreeCs:Cities,climate,catastrophes
Thefutureisalwaysuncertain,yetitseemsclearthatcitieswillgrowan
dthatclimate
willchange,althoughdisparately.Wellmanagedcitiescanreducethei

rvulnerability
even
inawarmerworldwithstrongerstorms.Catastrophesarepossible,but
their
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a

n
k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
likelihood can be reduced with appropriate actions now and
preparations for
contingency actions later. Climate change poses a troubling risk

of increased
conflicts: armed struggles have historically been associated with
droughts and
desertificationinAfrica,forexample.Butsquabblingoverresourcesl
eadstoconflict
whencompetingclaimscannotbepeacefullyresolvedandwheninstit
utionstoresolve
conflicting claims are inadequate. There is thus a large premium
on strengthening
institutionsforresolvingtomorrow’sresource-
relatedconflictsmorepeacefully.
Theseoutcomesrequiremuch.Urbanizationshiftsthebalanceofprev
entionfrom
individualmeasurestocollectiveaction.Althoughgovernmentswill
havealargerrole,
they must harness the market in better ways, with greater
sensitivity to when and
how prices get distorted. For collective prevention to be
effective, national
governmentsandcitiesmustdeliverbetterservices,includingpreven
tion.Theymust
design,build,andmaintaininfrastructureandbemoreawareof—
andresponsiveto—
whatindividualscanandcannotdo:providingdetailedseismicmapso
ffaultlines,for
example,butallowingdevelopersandpeoplewholiveinbuildingstod
ecidehowto
construct safe structures. Knowledge and know-how are needed
more than funds;
without them, the funds would be poorly allocated. Global
institutions could also
spreadwordofwhatcanbedoneandhelpgovernmentsintheirtasks.
Whiletherearegoodreasonsforhope,therearealsoinstancesofconcer
n.Take

Jakarta,whereindividualpreventionmeasuresdependonwhetherthe
government,in
part,providesadequatewateranddrainage.Ifclimatechangewillwor
senJakarta’s
inundations,shouldinfrastructureandcitymanagementbeimprovedt
oday?
GreaterJakarta
isacoastalurbanareawith24millionpeopleandacatchment
arearimmedbyvolcanoes.Some13riversflowintoJakartabay,andth
ecityisinthe
lowestpartofthebasin.About40percentofthecityisalreadybelowsea
level,and
floods follow intense rains between November and April
(annual rainfall is 15–25
metersayearandupto4metersupstream).Majorfloodshitin1996,200
2,andthen
inFebruary2007,theworstinitshistory,whenheavyrainscoincidedw
ithapeakin
theastronomictidalcyclethatrecursevery18.6years.Yet,tidalsurges
andrainfall
alonedonotexplainthefloods’severity.Arecentstudyfoundlittlediff
erenceintotal
precipitationacross fivemeteorologicalstationsalong
theCiliwungRiver (Jakarta’s
mainriver)in1996,2002,and2007(Texier2008).Howsusceptibleisa
nincreasingly
urbanizingJakartatorainsandasealevelrise?Seefigure6.3.
Asinmanymajorcities,publicserviceshavenotkeptpacewithpopulat
iongrowth.
GreaterJakarta’spopulation,doublingfrom11.9millionto23.6milli
onbetween1980
and2005,isprojectedtoexceed35millionby2020.Upstream,numero
ussecondary
residenceshavebeenbuiltoverthepast50years.Teaplantationsrepla

cedforests
on the volcanoes’ slopes, reducing the capacity to absorb and
store rainwater,
increasingpeakrunoff flowsandsedimentation
inrivers.Downstream,uncontrolled
residential and commercial developments in lakes and
reservoirs, which once
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d

B
a
n
k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.

absorbedthestormwaterflowsintothecity,increasedfloodlevelswhi
leexcessive
abstractionofgroundwaterduetothelimitedsupplyofpipedwatercau
sedrapidland
subsidence.Injust15years,awaterabsorbingareainKelapaGading,a
subdistrict
inNortheastJakarta,becameaboomingcommercialandresidentialar
eathatfloods
everyyear(figure6.4).
Climatechange is likely toraisesea levelsand
increaseboththefrequencyand
intensityofstormsthatwillfloodJakarta.Peoplemayeventuallyhave
tomovefrom
Jakarta;soshouldefforts to
improvethecityberedirected?Atoughquestion,but
framedincorrectly.
Movingshouldbeanindividualchoice,notanexcuseforcollectivecoe
rcion.People
now living in Jakarta should not be forced to move, whether by
compulsion or by
neglecting infrastructure and public services to residents.
Investments in Jakarta’s
infrastructureshouldcontinueiftheypassthecost-
benefithurdle,andlargebenefits
intheimmediatefuturewouldweighheavilyinthedecision.Butitiseq
uallyimportant
for investment in and management of other cities to also
improve because their
growthcouldaccommodatemorepeopleandcommerce.
Figure6.3GreaterJakartaareaorographicmapwithrainfallregime
Source:Gunawan2008.

Climate change will not adversely affect all cities, and while
ports may be
important,itisdifficulttopredictwhichwillthrive.Inthe18thcentury
,fewthoughtthat
New York, which then ranked well behind Boston and
Charleston, would become
America’s largest and richest city, especially since Baltimore
and Philadelphia had
better ports. Jakarta’s prosperity may well continue (it
contributes 25 percent of
nationalnon-
oilGDP).Andifitdoes,itwouldbeinthesamesituationasRotterdam
today:havingtoconsiderexpensivemeasurestoprotectitspeopleand
assetsfrom
floods and sea surges. But such choices would be less stark, and
ultimately less
wasteful, if other Indonesian cities in safer locations grew.
Jakarta is not unique;
MexicoCity,Mumbai,andmanyothersaresimilar.
Central,
06027.
C
o
p
yr
ig
h
t

©
2
0
1
0
.
W
o
rl
d
B
a
n
k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts

r
e
se
rv
e
d
.
Figure6.4FifteenyearsofurbanizationinJakarta,beforeandafter
Source:HahmandFisher2010.
Cities, climate, and pending catastrophes are altering the
disaster prevention
landscape.Whilehazardswillalwaysbewithus,disastersshowthatso
methinghas
failed.Butdeterminingwhathasfailedanddecidingonthecorrective
measuresare
notalwaysobvious.AnddebatingwhetherHurricaneKatrinaorCyclo
neNargiswasa
resultofclimatechangedetractsattentionfrompoliciesthatcontinuet
omispricerisk,
subsidizeexposure,reduce individuals’
incentivestoreducerisk,andpromoterisky
behaviorinthelongrun.
Peopleriseoutofpovertythroughbettertechnology,greatermarketac
cess,and
more investment inactivities thatspillbenefits
fromonesetofeconomicactors to
othersthroughgreaterinterdependence,higherproductivity,andstro
ngerinstitutions.

Livingincitiesfacingseriousrisksofinundationisundesirable,butaf
ailuretoreduce
povertywouldbeevenmoreundesirable.Fortunately,neitherisinher
entlynecessary.
People acting individually and through responsive governments
can prosper and
survive.That,afterall,isthebasisofsustainabledevelopment.
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d

W
MEMORANDUMTOA
CONCERNEDCITIZEN
Subject:NaturalHazards,UnNaturalDisasters:TheEconomicsofEff
ectivePrevention
ehopeyoufoundthereporthelpful.Wecannotpresumetotellyouanyth
ing
specific:youknowwhatisbestforyouandyourfamily.Wesoughttobri
ngto
your attention the experiences of others that are embedded in
the statistics and
studies.
You may also be interested in hearing directly from two of our
colleagues who
recentlysurvivedharrowingdisasters.Thesearedeeplypersonalnarr
atives,difficult
for themtowrite.Wekept themin the firstperson.Ashortsection
following their
experiencepullstogethersomecommonthreads.
ReflectionsasaGujaratearthquakesurvivorandhumanitarianworke
r
January26,2001.Iwasauniversitystudentandapart-
timeemployeewiththestate
government’s remote sensing and communication center in
Ahmedabad (Gujarat,
India)whentheearthquakestruckin2001.Itwasalittlepast8:30inthe
morning,
and Iwasstillasleep in the flat that Isharedwith three friends. I
remember their

shoutingthroughthedoortowakemeupandtheswayingaswetookthes
tairsdown
four floors. Some things from that day are hazy, but other
details are etched
permanentlyinmymemory.
ItwastheRepublicDayholiday.Thegroundwasstillshakingwhenwe
emerged
fromtheflatandIcouldseethetalltelecomtowerswaying.Itriedtorem
emberwhat
Iknewaboutearthquakes,anditwaspreciouslittle.Immediatelyafter
realizingthatI
had survived a deadly disaster, and assuming that the
earthquake had struck just
Ahmedabad, I realized that my parents, who lived in the city of
Bhuj some 400
kilometersaway,wouldbeworriedwhentheygotnewsoftheearthqua
ke,especially
becauseIwasnotreachablethroughthemostlydisabledphonesystem.
Meanwhile,thedamageinAhmedabadwasbecomingapparent:peopl
ewentabout
onscootersandmotorcyclestomakesurethattheirrelativesandfriend
sweresafe.
Aswewereunsureofthesafetyofourmultistorybuilding,aclosefrien
dvisitedto
checkuponme.Hetookmeandmyroommatestohisuncle’sgroundflo
orhome,
whichwasrapidlybecominganemergencyshelterforothersintheneig
hboringhigh-
risebuildings.
News of the size and scope of the earthquake slowly began
trickling in through
transistor radios. I learned later in the evening that Delhi and
Mumbai had felt the

b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
quaketoo,andonlythendidIrealizethatits“epicenter”(wewerealllea
rningthese
terms)wasinBhuj,wheremyparentswere.Thetremorwehadexperien
cedwasof
muchlowerintensity.Myconcernimmediatelyswitchedfromtelling
myparentsthatI
wassafetoworryingiftheyhadsurvived,soIdecidedtoleaveforBhuj.
Day2.IhoppedonaprivatebusforBhujearlythenextday.Driversslow

edtoask
others(someincars,othersonfootandcarts)abouttheconditionofroa
dsandthe
towns they came from, and the news was disturbing. People
spoke of “all being
destroyed,”andthiswashownewsspread.Iwitnessedthedestructionf
irsthand,
andIhadastrangefeelingwhenIsawthemanygovernmentbuildingsd
estroyed,
includingpolicequarters.ThiswasstrangebecauseIalwaysthoughto
f thesarkar
(government) as invincible, so it was unnerving to see it as
equally vulnerable and
incapacitated as the rest of us. In contrast, I was overwhelmed
by trucks that
stopped to give people bottled water and packets of food, and
realized that
volunteerswerealreadyorganizingrelief.Someonethrewmeabottle
ofwater,andI
wasabouttohanditbackwhenIrealizedthatIshouldjustacceptit.
Thenormal6to8hourjourneytook12hours.EventhoughIwasusedtos
eeing
Bhujwithoutelectricity,itwaspitchdarklikeneverbefore.WhenIrea
chedmyhome,
to my utmost relief, I found my parents and neighbors
assembling a makeshift
tarpaulinshelteronthestreet.Theytoldmewhathappened:myfatherw
aspraying
indoorsandmymotherwas in thekitchenwhentheyfelt theshaking.
Instinctively,
theyranoutdoorsfromthebackdoortothegardenwheretheyheldonto
apapaya
treeuntiltheviolentshakingsubsided.
MyfatherhadthehousebuiltwellunderhissupervisionjustbeforeIwa

sborn,and
itwithstoodthequakebutacantileveredoverhangcamecrashingdow
n.Hadthey
escapedfromthefrontdoor,theoverhangwouldhavehurtthem.Thequ
akecracked
thewallsandhaddestroyedallelectronicsandcrockery.Ialsofoundou
tthenthat
becausepowerandphonelinesweredown,myfatherhadthepresenceo
fmindto
go to the intercity bus terminal and give slips of paper with my
name and phone
numberandthatofotherrelativestostrangersfleeingthecity,askingth
emtocall
andconveyhiswell-
being.Severalofthesemessagesgotthroughincomingdaysto
ourrelativesinvariouscities.
Day 3. Exhausted physically and emotionally, about 30 of us
slept outside that
night.Despitethechillywinternight,nobodywaswillingtoentertheir
homes.Early
next morning, we were awakened by an aftershock, only to see
an electric pole
hangingbythewires justaboveour tarpaulin.
Isupposeweweredoubly lucky: to
havesurvivedthe initialquakeandthentohavesurvivedpossible
injuries fromthis
aftershock.Werealizedthatsuchaftershockswouldcontinueforseve
raldays,sowe
decidedtolockupourBhujhomeandmovetoourancestralhomeinRaj
kot(some
240 kilometers away). It seemed unaffected by the quake, and
we monitored the
mediaandgotnewsfromfriendsaboutthescaleofthedevastation.

b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
Twoweekslater.Grateful
forhavingsurvived,wewereevermoreeagertohelp
thoseaffected.WeknewthatthecityofAnjar,around50kilometersfro
mBhuj,was
one of the worst affected areas, and our family friend and
former neighbor Mr.
Kathiwalahadrelocatedthereafewmonthsbeforetohelphissonsetup
abusiness.

Afterseveralinquiries,wefoundhim,fullybandagedfromthewaistdo
wnataprivate
hospitalinRajkot.Hiswifeandsonsurvivedwithminorinjuriesfromt
heircollapsed
home,buthis14-year-
olddaughternevermadeitoutofthebedroom.Mr.Kathiwala
wasburiedundertheoverheadtankforhoursbeforeneighborsrescued
him.
Even in the midst of such misfortune and the risk of losing a
leg, Mr. Kathiwala
recountedhowgratefulhewastotheDaudiVohracommunity—
acloselyknitgroupof
prosperoustradersthathebelongedto.
WhenDaudiVohramembersinothertownsheardofthediresituationi
nAnjar,they
hiredtruckstobringfirstaid.Theytransportedtheinjuredtohospitals
andthemore
severely hurt to larger cities with better care. In addition to
providing for medical
treatment,accommodation,meals,andbasichouseholdnecessities,t
heDaudiVohra
raisedfundstopayforthebestavailablemedicalcare.Theyalsogavefa
miliesRs.
5,000 in cash for incidental expenses. This humane support
greatly alleviated the
traumaoftheearthquake.
Threeweekslater.Lifehadtogoon,andIeventuallyreturnedtoAhmed
abadfor
thescheduledexaminations—
onlytofindanoticethattheywerepostponedbythree
monthsbecauseofthedamageduniversitybuilding.Istayedwithafrie
ndbecause
ourfourthfloorflatwasnotconsideredsafe.Icametoknowonedaythat

theUN
disastermanagementteamwasseekingvolunteerstoworkinareasrav
agedbythe
disaster,andIjoinedthemexactlythreeweeksaftertheearthquake.
Helping with post-disaster assistance enabled me to see things
from a different
perspective. Statistics cannot fully capture what happened. The
poorest suffered
mostandtooklongesttorecover.Inmanytownstherewaslittledamage
inwealthy
areaswithwell-builtbungalow-stylehouses,but
thepoorlybuiltstructuresof those
lesswelloffmostlycollapsed.Itwasamazinghowquicklythegovern
mentrestored
life-
lineservicesintheworstaffecteddistricts.Itembarkedonahugerecon
struction
program, and an efficient model for community-based recovery
and reconstruction
evolved.
Not all that I saw and heard was this inspiring. The old walled
city of Bhuj was
largelydestroyed,andIheardtalesthatinsonibazaar,wheregoldsmit
hsplytheir
trade, the survived shop owners offered passers-by money to
recover the gold
ornamentsandstrongboxesfrominsidetheteeteringbuildings.Ialsoh
eardthatthe
foreignfoodpacketswerenotservingtheirpurposebecausethelargel
yvegetarian
populationwasaversetoconsumingfoodinwrapperswithanundeciph
erableforeign
language.Whilemanyvolunteersweretirelesslyhelpingwithreliefa
ndrecovery,a

fewseemedinterestedonlyintakingphotographsandbeing“disastert
ourists.”
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k

P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
Eight years later. Now, after eight years of working in post-
disaster and post-
conflictsituations,notjustinIndiabutlaterinAfghanistanandSudan,
Iamstruckby
severalthings.Communitiesrespondfirstinthemidstofchaosbecaus

epeoplecare
forothers:butwithlimitedresourcesattheirhand,everyonehelpsthei
rcommunities
andfriendsfirst—
andonlythenanyothers.TheGujaratearthquakewaspivotalinthe
paradigmshiftfromemergencyresponsetoriskreductionandprepare
dness.Many
who believed that natural disasters like earthquakes cannot be
prevented are now
activelyhelpingreducingthedisasterrisksintheirownlivesandaroun
dthem.
Thelongest45minutesinAceh
December26,2004.ItwasSundaymorningaround8a.m.Myparentsw
ereaboutto
leave for the Hajj that week, expecting friends and relatives to
say goodbye. My
fatherwas in theshower,andmysisterandacousinwere in
thekitchenwashing
dishes.Then,theearthquakestruck—
bigandlong.MaybeitwasoneofGod’sways
toremindusofourinsignificanceinthelargerschemeofthings.Weran
outside.
Outside.Theshakingfinallystopped.Buthavingexperiencedearthqu
akesbefore,
weknewtherewouldbeaftershocksandwaitedoutside.Then,fivemin
uteslater,as
expected, another earthquake, this time smaller but longer.
More crying. I silently
recitedmyprayers,trustedthatGodwouldtakecareofthis,tawakkal
—that’swhat
wesayinIslam.Itrelaxesusslightly.Withthethirdquake,peoplestart
edtocryand
screamevenmore.

Thenallofasudden,wesawourneighborsrunningtowardus,screamin
g“RUN…
RUN…TOTHEMOSQUE.”Withoutknowingwhy,weallstartedtoru
n.Somepeople
tried to lock their houses before running. None knew what was
in store. We then
heardahorrible,helicopter-
likesound,butmuchlouder.WhilerunningIlookedbehind
andthere itwas.Darkbrown,high,amonsterwave3–
4metershigh!And itwas
approachingfast.
Wemadeit to themosque,whichwasnot far
fromourhouse.Themenquickly
askedall
thewomenandchildrentogoupstairs(themosquehadtwofloors).The
mosquewasbigandhadmanypillarswithnowallssothewatercouldjus
tflowin
easily. My dad insisted that he would stay downstairs, and the
rest of the family
insistedequally thatwewouldnotgoupstairs.
Itwasaverydifficultmoment.The
water,therealready,hadrisentomywaist.
We had to make a quick decision. Then, we compromised. Since
she was
physicallystrongerthanbothmeandmymom,mysisterstayeddownst
airswithmy
dadwhilemymomandIwentup.Wehuggedandkissedandcried.Thew
aterwas
nowuptomychestandtheearthwasstillshaking.Themosquecouldha
vecollapsed
butwereallyhadnooption.
Upstairs.Isawmanyofmyneighbors,cryingandpraying.Thoughmyh
eartwas

fullofpain,Ididnotcryatall.Onetearfulneighbortoldmeshedidn’tkn
owwhereher
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n
k

P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
sonandhusbandwere.Theyhadleftearlyinthemorningtogotothebea
ch:itispart
of Acehnese tradition to go to the beach on Sunday morning. I
felt relieved in my
heart,andthankedGodquicklybecausemyfamilywassupposedtobeo

nthebeach
aswell.Butmylittlebrother,whowasstudyinginJakartaandsupposed
toflyback
toAcehthatweekend,canceledhisplans.So,wedecidedtocancelourb
eachpicnic
thatSunday.
Waiting.Idesperatelywantedtogodownstairstoseehowmyfatheran
dsister
were.Butmymomstoppedme.Allwecoulddowastowait.Thensudden
ly,afew
menappearedupstairscarryingbodiesintheirhands—
myneighbors.Moreandmore
bodies were brought up. The upstairs was full of dead bodies. I
could not stop
thinkingthatthenextonemightbemydadormysister.Ijusthuggedmy
momtight.
Shekepthercomposure,comfortedus,andremindedustoreciteGod’s
name.
Downstairs. Someone finally yelled that the water had subsided.
We slowly
steppeddownthestairs.Thescenewasunimaginable.Watereverywhe
re.Bodies
coveredwithmud.Iwasexpectingtheworst.ThenIsawmydadandsist
er,clinging
tooneofthemosque’spillars,alive.
Finally the tears came. Never before had I cried so much. But
the men in my
neighborhoodwereamazing.Theyworkedhandinhandrightawaytoe
vacuateall
thebodies.Inlessthananhourthemosqueupanddownwascoveredwit
hthedead
bodies.
Icameacrossaneighbor,a17-year-oldgirl,

Iknewwell.Wefoundherwithno
clothes,entangledinthemudandelectricitywiresfromoutsidethemo
sque.Shehad
swalloweddirtwaterandcouldnotbreatheproperly.Bothherlegswer
ebroken.Her
headwasonmylap,andshekeptaskingwhetherwehadseenherfamily.
Sadly,her
entirefamilyhadperished.Butweliedtohertomotivatehertokeeponb
reathing
anditworked.Ourplanwastotaketohertoanearbyhospital.Somemen
founda
volunteerwithacarwhohadcometohelp. I left for
thehospitalwithouthavinga
chancetoinformmyparents.
Outside.Nearbyhospitals—fullofmudandwater—
werenotfunctioning.Wefinally
foundasmallcommunityhealthclinicbuttherewasnodoctor,onlyone
nursewithno
medicalsuppliesleft.Itwasfrustratingtothinkwehadcomealltheway
fornothing.
Wegavemyneighborsomewaterandcookies,whileafriendlefttofind
someother
help.KnowingthiswasthebestIcoulddoforher,Iwantedtogobacktot
hemosque
to inform my parents that I was safe. It was 4 p.m. already. But
there was no
transportationsoIdecidedtowalk.Itmusthavebeen100degreesthatd
ay,andI
hadnoslipperson.ThankGodIfoundaguypassingbyonamotorcycle.
Hedropped
me off at a family friend’s place. They got me a pair of slippers,
and I resumed
walking.
Someotherneighborspassingbyinatruckpickedmeupbuttoldmemyp

arents
werenolongerinthemosque.Theyhadsearchedformeandeventually
wenttoa
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n

k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
relative’splace.IsomehowreachedtherearoundMaghrib(sunset)tim
e.Myparents
wereupsetbutrelieved.Itoldthemthestoryandtheyunderstood.
The night. No electricity that night. None of us could sleep;

with at least 100
quakes.Wekeptrunningoutsidealmostevery5minutes—
sodepressing.Ikepton
hearingsounds.Helicoptersorwater?Notclear. I feltdeeplyguilty
for leavingmy
friendintheclinicandprayedthatshesurvived.LaterthatweekIfound
outthatshe
didn’tmakeit.Better,perhaps,sinceeverybodyinherfamilyhadalsod
ied.
Theassistance.Wehadtorationourfoodsupplies.Fuelwasscarce.My
mom—
so strong throughout—broke down when she found out that her
only sister had
passedaway.Shejustsatinthecorner,prayingeveryday.Shehadonlyo
nedress
towear—
theonethatshehadrunin.MysisterandIcouldatleastborrowsomeof
thegirls’clothes.Underwearwasabigissueforallofus.Idon’tneedtoe
xplainit
further.
We heard rumors that assistance had arrived but was piled up in
the airports.
Roadsremainedblocked,soonlyhelicopterscouldget
topeople.Allwecoulddo
wastobepatientandtightenourbelts.
Afewdayslater,mybrotherandunclecamewithacarfulloffood.They
hadflown
toMedan—theclosestcitytoBandaAceh—
anddrivenhome.Ittookthem14hours.
Theyalsobroughtsomeclothes,cleanunderwear,andcash.
Later, we received more cash and other types of humanitarian
assistance from

many friends from foreign countries. Each day, random people
came to the house
andbroughtusassistance.Wewillneverforgetthat.Indonesianvolunt
eers,national
andforeignsoldiers,localandinternationalNGOs,religiousgroups,n
ameit.Iwould
say the Red Cross, volunteers, and soldiers were crucial in
removing debris to
restoreroadlinks.
Things were a lot better after the second week. Among the
assistance we
received,theonlythingsIdislikedwerethefortifiedbiscuitsfromWF
P.Westayedin
thehouseforaboutamonth.Ithadtwosmallbedroomsbutsomehowwe
managed,
along with many others who came as well for shelter. We
wanted to rent another
place to lessen the burden but couldn’t find anything affordable.
It’s amazing how
rentalpriceshadsoaredsohigh.Peoplewouldrenttheirhomesonlyto
UNandNGO
offices.Amedium-sizehousewasaround100dollarsaday.
Home?Backhometocheckthedamage,wefoundoutwehadlostsevera
lwalls.
Twodeadbodieswerefloatinginthekitchen—oneofa5-year-
oldgirlandtheother
of a man. The house looked scary and dark—full of trees,
garbage, and water. I
lookedatmydadwithallhisgrayhairwithwateruphiswaisttryingtosal
vageour
belongings.Myfatherisacivilservantabouttoretireintwoyears,and
mymomisa
teacher.Wewerenotpoor,butwewerenotricheither.Thatwasouronly

home,and
myparentshadputtheirlifesavingsintoit.Everythingtheyhadworke
dforseemed
Central,
06027.
C
o
p
yr
ig
h
t
©
2
0
1
0
.
W
o
rl
d
B
a
n

k
P
u
b
lic
a
tio
n
s.
A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
gonein45minutes.Itwashardformetoseethefuturethatday.Therewa
snoway
forhimtocollectenoughmoneytorebuild.Buttheseareonlymaterialt
hings,hetold

me.Iwaswrong,hewasright.
Somereflections.Iwasrebornagain,eventhoughIdonotbelieveinrei
ncarnation.
Iseetheworlddifferentlynow.Lifeisshortandunpredictable.Mydad
says:“You
prayhard,youworkhard,youresthard,andyousocializehard—
otherwiseyouwill
neverbehappy.”Itrusthim!OnecouldnevertellexactlywhenGodwan
tstotakeus.
Insomeways,Iconsidermyselfveryluckytohavegonethroughthis.
Iwasdeeplytouchedbyallthecarethatcamefromallovertheworld.Iju
stknew
that everybody from Banda Aceh or outside was trying their
best. I am forever
gratefulforthat,eventhoughIhaveadifferentopinionabouttherecons
tructionphase
inBandaAceh.
Indonesia, including Aceh, is highly prone to almost all types of
natural hazards:
tsunamis,earthquakes,floods,droughts,volcaniceruptions,younam
eit.Thetsunami
shouldbeawake-
upcallforauthoritiesandcommunitiestoreducetheseveryreal
risks.IwishIhadknownmoreabouttsunamis.Perhapsmyauntwouldb
ealivenow
ifwehadanearlywarning.Theimportanceofbuildingdisaster-
resilientinfrastructure
shouldalsobeconveyedtocontractorsandconstructionworkers.Afte
rall,theyare
theoneswhoimplementpolicy.Sometimes,
theproblemisnotalwaysthebuilding
codesortheinstitutionalframework,buttheignoranceofworkerswho
feelthatitis

acceptable to reduce the amount of cement or concrete or steel
to cut down the
price.Weneedtorememberthatlocalengagementstendtoworkbettert
hanpaper
regulations.Weneedtoensurethatpolicies,regulations,andknowled
gearrivewhere
peoplelive.
Commonthreads
Notwodisastersunfoldthesameway,andnotwopeopleareexactlyali
ke.Butthe
twonarrativesrevealcommonthreads.Family,friends,andneighbors
arethefirstto
help.Aid,thoughuseful,comesmuchlater.Knowingthehazardsandb
eingprepared
(knowingwhattoexpectanddo)arereallyuptoyou.
Youcanalsoaskmoreofyourgovernment:notmorespending,butmor
eeffective
preventionmeasuresandmoreinformationabouthazards,suchasmap
soffaultlines
andfloodplains.Makingitreadilyaccessiblewouldhelp.Andwhendi
sastersexpose
weaknesses,makesureyourrepresentativeslookintotheunderlyingc
ausesandtell
youwhatisbeingdonetopreventitfromhappeningagain.
Central,
06027.

A
ll
ri
g
h
ts
r
e
se
rv
e
d
.
T
CHAPTER5
InsuranceandCoping
heearlierchaptersshowedthatindividualsgenerallytakeprevention
measures
withintheirchoiceset.Butfullpreventionisneitherattainablenordesi
rable,and
residualdisasterriskswillremain.Insuranceandcomplementarymea
surestohave
fundswhenneeded(suchasborrowingorsettingasidereservefunds),r
emittances,

andrelief“softentheblow,”andthischapterexaminestheirrolesintur
n.
Thechapterbeginswiththebasicsoftheinsurancebusiness:theadvant
agesof
poolingandtransferringrisktothosewillingtobearitandhowinsurers
dealwiththe
many complications that arise from adverse selection and moral
hazard. Insurance
clearly increases a person’s choice and thus well-being: the
contract specifies the
resources transferred from one person to another when the event
(such as a
disaster)occurs.Indoingso,
itshiftstheriskfromtheindividualtothepoolofthe
insured.Butsofteningadisaster’sblowconcomitantlydilutestheince
ntivetoprevent
—unless the premium reflects the risk and the prevention
measures a person
undertakes.
Commercial
insurancecompaniescalculatethepremiausingdetaileddataonthe
frequencies and intensities of hazards and how they affect
exposed assets. The
premia must also cover the considerable costs of administration,
marketing, and
monitoring.Manypeoplemayforgoinsuranceifthepremiaaretoohig
h.Andwhile
parametric insurance—a type of insurance that specifies the
payout based on a
parameterrelatedtothehazardbutunrelatedtoactualdamagesincurre
d—reduces
someofthemonitoringcosts,suchschemeshavelowpenetrationrates
indeveloping
countrieswheretheyhavebeenintroduced.

Whenaninsuranceindustrydoesdevelop,itinvariablydrawsthegove
rnmentinas
regulator, as provider (in many countries), or as reinsurer.
Governments inevitably
addapoliticaldimension,andpressurestosubsidizethepremiamayin
crease.The
U.S.experiencewithfloodinsuranceshowsthatthisisnotjustanissuei
ndeveloping
countries. Too low a premium encourages construction in
hazard-prone areas,
therebyincreasingexposureandvulnerability.
Thechapternextturnstowhethergovernmentsshouldbuyinsurancet
ohavefunds
to spend after a disaster, simply borrow, or set aside funds in
reserve. Many are
alreadyindebtedandeventhosewithlowdebtsmayfinditdifficulttob
orrowwhen
theymostneedto.Politicianswhowanttospendonworthyprogramste
ndtodeplete
fundssetasideinareservefund.Toavoidthis“honeypotsyndrome,”g
overnments
Central,
06027.
C
o
p
yr

ri
g
h
ts
r
e
se
rv
e
d
.
may purchase insurance. The World Bank’s Catastrophe Risk
Deferred Drawdown
Optionandothersuchfacilitiescanhelpcountries.
While individualsarerisk-averse, therearegoodreasons
forsomegovernments
actingontheirbehalftoberisk-neutral.Arisk-
neutralentitywouldbuyinsuranceonly
ifthepremiumwerelowerthantheprobabilitytimestheexpectedloss(
whichleaves
nothing to cover the insurer’s costs). But the likelihood of a
disaster that is large
relativetoaneconomy’ssize(asintheCaribbean,wherethemainunkn
owniswhich
island will be hit) may make some governments risk-averse,
especially when rapid
accessto
fundsafterdisasterscouldbedifficultorcostly.Suchgovernments,an
d
those seeking to avoid the “honey pot syndrome,” would benefit

from buying
insurance. The Caribbean Catastrophe Risk Insurance Facility
pools disaster risks
regionally, helping countries purchase insurance less
expensively than otherwise.
Comparing prices offered by insurance firms against those in
capital markets, as
Mexicodidwhenissuingcatastrophebonds,isalsoadvantageous.
The chapter then examines remittances sent by private
individuals and groups
abroadtohelppeoplecopewithadisaster.Remittancesaredirectedtov
ictimsand
theirsurvivors,evenwhenthedisasterdoesnotattractanymediapubli
city.Thefunds
arrive quickly without the involvement of governments or other
organizations. But
sometimes unnecessary government policies (controls on capital
flows, dual
exchangerates)impedearrivalofthefunds.Remittancesthatarrivebe
foreadisaster
also help with prevention. Although remittances augment
consumption, particularly
consumerdurables, theyarealsousedto
improvethequalityofhousing.Mudand
strawhutsgivewaytohousesbuiltofbrickandcement.Privateremitta
ncesalsohelp
develop banking and money transfer facilities, which in turn
strengthen the area’s
commercialtieswithotherpartsofthecountryandtheworld.
Last,thechapterexaminestheroleofaidinprevention.Post-
disasteraidcanalso
bedouble-
edged:whilesomeaidiswarranted,itcanalsogiverisetotheSamaritan
’s

dilemma—theinability tocrediblydenyhelpfollowingadisaster
tothosewhohave
not taken sufficient prevention measures. Some new but not
very strong evidence
shows that post-disaster aid could reduce prevention. Donors
should therefore be
awareof thedisincentives theymaycreate,andconcern for
thevictimsshouldbe
moderatedbytheeffectonincentives.
Insurance:Usefulifthepremiumispricedright
Aswithanyvoluntarytransaction,
insurancebenefitsallpartiestothecontract: the
insurerbenefitsfromthebusinessandtheinsuredreducetheadversity
oftheworst
statesofnaturebygivingupsomeofthebenefitsinthegoodstates.
Thebasicsofthebusiness
Insurerstakeprideincoveringuniqueriskslikeanoperasingercatchin
gacold,ora
Central,
06027.
C
o
p
yr
ig
h

ts
r
e
se
rv
e
d
.
racehorsebreakinga leg.Butsuch insurance
isasideshow,andthebulkof their
businesscoversmoremundane,predictable,anddiversifiablerisks(s
uchaslifeand
propertyinsurance).Considerinsuringhousesagainstfire:onecanno
ttellwhenandif
aparticularhousewillcatchfire,butdataonpastfiresallowthenumber
ofhouse
fires in an area to be reliably predicted with probabilities
attached. Making the
average lossmorepredictableallowsfirmsto insure
individualhousesagainst fire:
theinsurercollectsanannualpayment(premium)fromrisk-
aversehomeowners(the
insured) and promises to pay (the insured amount could be
actual damages or a
specifiedsum)iftheirhouseburnsdown(thetrigger).Theaggregatean
nualpremia
collected from the insured must cover the insurer’s operating
costs and the likely
payouts. And if unexpectedly fewer houses burn down, the
insurer has a surplus
beyonditsnormalprofits.

Thereisalwaysachancethatanunexpectedlylargenumberofhousesw
illcatch
fire(asinaparticularlydryyear),sotheinsurerhasabuffer,whichisthe
owners’
capitalandsurplusesaccumulatedfromearlieryears.Thisbufferisinv
estedandthe
ensuingearnings(dividendsorinterest)augmentthepremiathefirmc
ollects.Ifthe
surplusesaccumulateovertime,competitivepressureswouldprompt
theinsurerto
lowerthepremia;iftheyaredepleted,thepremiawouldberaised.Thisi
sthebasic
principleofinsurance;butcomplexitiesquicklymultiply.
Adverse selection arises when a person buys insurance knowing
that his risk
exceedsthatofthelargerpoolthatisthebasisfordeterminingthepremi
um.Ifonly
thoseknowingtheirriskisgreaterbuysuchinsurance,theinsurer’ssur
pluswillfallas
theriskofthepoolrises.Moralhazardariseswhentheinsuredtakeaddi
tionalrisks
becausetheyareinsured(notrepairingthebuildingsprinklersthatexti
nguishfiresif
the building is insured against fire). Co-payments (where the
insured bears a
specifiedfractionoftheloss)anddeductibles(wheretheinsuredbearl
ossesuptoa
specified amount) reduce but do not eliminate these difficulties.
Contracts become
complex and the costs of administering claims, resolving
disputes, and increasing
monitoring mount. Insurers continually seek observable proxies
of the risks they
insure, link their premia to these risks, and continually test the

insured’s price
sensitivitytopremiathatmustcoverthesecosts.
Thecostsresultinapremiumthatgreatlyexceedsexpectedlosses,buts
ufficiently
risk-averse people buy insurance all the same because it
protects them from the
devastating financial implications of a disaster. Insurance does
not “shift the loss”
collectively:theinsuredpayforthelossesthroughtheirpremia,andth
osewhodonot
makeaclaim,essentiallypayforotherswhodo.
Insurance can be a competitive industry, but some economists
find that there is
littlecompetition(insurersareexemptfromantitrustlawsintheUnite
dStates)orthat
costsarenotdrivendown.Administrativeandmarketingcostsareabo
ut35percent
of aggregate premia in the United Kingdom (the insured get
roughly 50 percent in
Central,
06027.
C
o
p
yr
ig

h
ts
r
e
se
rv
e
d
.
payouts,afigurethatisbroadlysimilarintheUnitedStates)wherepriv
ateinsurance
firms compete, in contrast to 10 percent in Spain where a state-
owned monopoly
provides coverage (Von Ungern-Sternberg 2004). Germany
privatized its provincial
monopolyprovidersunderaEuropeanUniondirectiveonlytofindthat
operatingand
administrative costs rose as a consequence and insurance premia
were raised
between35and75percentinfiveyears.
Regardless of who owns insurers, governments invariably get
involved as
regulatorsifnotasprovidersbecausebuyers“gettheproduct”(thepro
misedpayout)
only after a disaster; and the insurer may find some reason to
refuse payment,
reducecoverage,orgooutofbusiness.
Thegovernment’sinevitableinvolvement

Insuranceislimitedinthedevelopingworld,butalargeindustryinman
ydeveloped
countries.1 These countries’ governments are involved in each
of them, though in
differentways.Courts,notgovernments,enforcecontracts,but
thepayouts to the
insuredmaybedelayedordenied.Insurersinvariablywriteandinterpr
ettheclauses
to theiradvantage(“the fineprint”),andseemingly
fairclausesarenotalwaysso.
British insurers and the insured could cancel a policy with
seven days’ notice, and
insurers cancelled coverage in 1997 when it became apparent
that the erupting
volcanoonMontserratwoulddestroyeverybuildingontheisland,aki
ntocancelling
coverageafterafirehasstarted(VonUngern-Sternberg2004).
Inthe1800s,insurancewasfornamedperils,withcoveredlossesstem
mingonly
from the specified risk. In the 1930s, all peril property
insurance became more
common.Whateverthecoverage,governmentstrytoensurethatinsur
ershonortheir
contracts(consumerprotection),andwhenthisrequires insurers
tohaveadequate
funds(solvency),regulationsoftenextendtoapprovingthepremia.T
hepremiaare
sometimeshigh,andtoensurecoverage,propertyinsuranceissometi
mesmandatory
(asinGermanyandmanySwisscantons);butsometimespopulistpress
urescause
premia to be too low (as in the United States with flood
insurance), needlessly
increasingexposureinhazardousareas.
Thefertile lands in the floodplainsattract

farmers,andmanysettlements in the
United States are periodically inundated. Sympathy for the
victims would prompt
publicassistance,andsettlerswouldrebuild in
thesamearea.Afterseveralmajor
floodsinthe1950sand1960s,privateinsurerswerenolongerwillingt
ocoverfloods
(which became an “uninsurable” risk), and the U.S. government,
recognizing that it
wasunableorunwillingtodenyassistancetothoseaffected,establishe
dtheNational
FloodInsuranceProgram(NFIP)in1968.
Thepremiawereset lowto inducehomeowners tobuy theNFIP
insurance,but
veryfewpeoplevoluntarilypurchasedcoverage(KunreutherandMic
hel-Kerjan2009).
The federal government then required this coverage as a
condition for federally
Central,
06027.
C
o
p
yr
ig
h

ts
r
e
se
rv
e
d
.
insuredmortgages,butthemandatewaspoorlyenforcedandmanypeo
plecanceled
theirpolicies,especiallyiftherewasnofloodforseveralyears,andoth
erspurchased
insurance just after a flood (Michel-Kerjan and Kousky 2010).
They examine more
thanfivemillioninsurancepolicies,thelargestfloodinsurancesampl
eeverstudied,
and find thatof theonemillionresidentialNFIPflood
insurancepolicies inplace in
Floridain2000,athirdwerecancelledby2002andabouttwo-
thirdswerecancelled
by2005.Therewasnoeffectivemechanismtopreventordiscouragem
orepeople
from settling in the areas known to be hazardous: the NFIP is a
federal program,
whilezoningandinsuranceregulationarestateissues,andlocalpoliti
ciansreflected
the settlers’ desires. The number of policies nationwide
managed by the NFIP
increasedfrom2.5million in1992to5.6million in2007and,
innominal terms,the
property value covered rose from $237 billion to $1,100 billion

during the same
period.
TheNFIP’sothershortcomingswereexposedafterHurricaneKatrina
floodedmuch
ofNewOrleansin2005.TheNFIPcoversfloods,butprivateinsurance
coverswind
damage.Manydisputesaroseoverwhoshouldpaywhendamagefrom
windcould
not be easily separated from that by floods (Kunreuther and
Michel-Kerjan 2009).
Victimsweregiventherunaroundandpayoutsweredelayed.
In a background paper for this report, Kunreuther and Michel-
Kerjan note how
multihazard insurancecanaddress insurer-
insureddisputesbyhavinghomeowners’
coveragemovefromthetraditionalone-
yearinsurancecontracttomultiyearcontracts
(say10or15years)tiedtotheproperty(nottheownerasisthecasetoday
).The
premiawouldreflect insurers’bestestimateof theriskover
thatperiodandwould
assure policy holders of coverage. The possible denial of
coverage was a major
concerninhazard-
proneareasbecauseinsurerscanceledpoliciesfollowingthe1992
and2005hurricaneseasons.FollowingHurricaneAndrew,Floridapa
sseda law in
1992limitingthecancellationofpoliciesbyinsurersto5percentayear
atthestate
levelandto10percentatthecountylevel(JamettiandvonUngern-
Sternberg2009).
Bothinsurersandhomeownerscancelpoliciesfordifferentreasons,a
ndthepremia
aresubjecttopoliticalpressures.Thesemajorchangesingovernment

policyrequire
appropriateregulatoryauthorityanddecisions(KunreutherandMich
elKerjan2008).
Comprehensive,multihazard
insurancewillentailhigherpremia.Somepolicyholders
may think they are being charged for coverage they do not need
(a person in an
earthquakeareanotpronetohurricanesandfloodsmayonlyhavequak
einsurance),
buttheywouldnotbeoverchargedifpremiareflectriskaccurately.Wh
etherpremia
accuratelyreflectrisksbecomesallthemoreimportant.
Governmentinvolvementinevitablybringspoliticalpressures;andv
estedinterests
and populist pressures exist in all countries, though they
manifest themselves
differently.Insurancesubsidiesareusuallyregressive:thosewithass
etstoinsureare
generallybetteroffthanthepoorersegmentsthatoftenpayindirecttax
esthatpay
Central,
06027.
C
o
p
yr
ig

h
ts
r
e
se
rv
e
d
.
forthesubsidies.Butunderpricedinsuranceisnotalwaystheresultofg
overnment
pressures. Insurerssometimesmakemistakesormay
takeunwarrantedrisksand
then discover that the risks were greater than they assumed. To
compensate for
thesemistakes,insurersoftenfindreasonstodenypayments,redefine
therisksthat
are covered (terrorism was made a separate risk that got
excluded), and raise
deductiblesandpremia.2
Pricingthepremium
Thepremiumisanimportantprice:
toolow,andexcessiveconstructioninexposed
areasandinsufficientpreventionresult;toohigh,andfewbuyinsuran
ce.Calculating
theappropriatepremiumisnottrivial:probabilitydistributionsandlo
ssfunctionsmust
beestimated,andtherelevantpoolandobservablecharacteristicsthat
correlatewell

withtheunderlyingrisk(anunobservable)mustbeidentified.Thesees
timatesarea
firm’s “proprietary information.” And while competition may
drive insurers to
continually
improvethesecorrelatesandhencetheircontracttermsandprices,the
European experience (showing the lower operating costs of
monopoly providers)
suggeststhatthismaynotalwaysfollow.
Some additional complexities arise with infrequent hazards:
diversification among
manypolicyholders(contemporaneous)maynotsuffice,anddiversif
icationovertime
(intertemporal) is more difficult (box 5.1). Examining data from
the largest U.S.
catastrophic risk reinsurer for 1970 to 1998, Froot (2001) finds
that catastrophe
insurancepremiaare farhigher thanexpected losses(up toseven
timesgreater).
Themostlikelyreasonsarereinsurancemarketimperfections(suchas
government
intervention in insurance markets) and the market power exerted
by traditional
reinsurers.
Asnotedinseveralpartsofthisreport,governmentscandomuchtoimp
rovedata
qualityandaccessibility.Hurricanesaremorefrequentthanearthqua
kes,butconsider
what it
takestosetthehurricaneinsurancepremium:severalsetsofdetailedda
ta
areneeded includingthefrequency, the
likelypathsandseverityofhurricanes, the
value and type of construction of all structures in their path (so

accurate property
recordsareessential),andhowmuchdamageeachstructurewouldlike
lysufferat
various wind speeds (so local universities and engineering
associations must know
andtestthestrengthofmaterialsanddesignsofexistingbuildings).Cli
matescience
modelsestimatetheforces(suchaswindspeedandairpressure)anden
gineering
determines how buildings withstand them; allowing estimates of
loss exceedance
curves (insurers use this combination of cumulative probability
distribution function
withvaluesatrisk).
Even with good data, it is far from clear whether the frequency
and severity of
hurricaneshaschanged(chapter6discusseshowfrequencyandseveri
tymightbe
affectedinthefuturebecauseofclimatechange).InsurersintheUnited
Stateshad
taken note of the dangers to property after 1992’s Hurricane
Andrew, but were
Central,
06027.
C
o
p
yr

ri
g
h
ts
r
e
se
rv
e
d
.
nevertheless caught unprepared for the string of storms and
major hurricanes
(Katrina, Rita, and Wilma) in 2004 and 2005. Insurers incurred
large payouts and
raisedthepremia;butonecannottellifthestringofhurricaneswerealo
wprobability
drawingfromanunchangeddistribution(makingthepremiaincrease
unjustified)—ora
shiftinthedistributionitself.
Box5.1Catastropheriskininsuranceandfinancialmarkets
Poolingrisksreducesaggregatevariance;solossesthatarelargeandu
npredictableforavictimbecomesmall
andpredictableintheaggregatepool.Riskpoolingcouldbecontempo
raneousorintertemporal,butthelatter
requirestheinsurertohaveenoughcapitaltomakethepayoutsandrepl
enishitovertimewithannualpremia.
Catastrophesareinfrequent,andinsuringagainstthemillustratesthei
ssueswithintertemporaldiversification.
Theriskthataneventcouldgeneratealargelossforacountry(1998Hur

FCHAPTER6ComingGame-ChangersBurgeoningCities,Cli.docx

FCHAPTER6ComingGame-ChangersBurgeoningCities,Cli.docx

Recommended

Recommended

More Related Content

Similar to FCHAPTER6ComingGame-ChangersBurgeoningCities,Cli.docx

Similar to FCHAPTER6ComingGame-ChangersBurgeoningCities,Cli.docx (16)

More from ssuser454af01

More from ssuser454af01 (20)

Recently uploaded

Recently uploaded (20)

FCHAPTER6ComingGame-ChangersBurgeoningCities,Cli.docx