Planning and Financial Protectionto Survive Disasters

Kari KeipiJustin Tyson

Inter-American Development Bank

Washington, D. C.

Sustainable Development DepartmentTechnical Papers Series

Cataloging-in-Publication provided byInter-American Development BankFelipe Herrera Library

Keipi, Kari Juhani.

Planning and financial protection to survive disasters / Kari Keipi, Justin Tyson.

p.cm. (Sustainable Development Department Technical studies series ; ENV-139)

1. Risk management--Latin America. 2. Risk assessment--Latin America. 3. Natural disasters--Economic aspects. I. Tyson, Justin. II. Inter-American Development Bank. Sustainable Devel-opment Dept. Environment Division. III. Title. IV. Series.

658.155 K282—dc21

Kari Keipi, Senior Natural Resource Specialist of the IDB and Justin Tyson, Consultant wrote thereport. The idea for the paper originated from the preparation of two documents: (i) First Report ofthe IDB presided Working Group of Financing of the OAS coordinated Inter-American Committeeon Natural Disaster Reduction, May 2000 and (ii) presentation by Kari Keipi for the InternationalConference of Sustainable Development Financing in Tegucigalpa, Honduras, December 2000.

Many contacts with the academic, private commercial and public sectors, and the review of a hostof studies have contributed to the ideas included in the document. The different stages of the paperhave been peer reviewed by Gabriela Basurto, Diego Rodriguez and Kim Staking of the IDB andby Omar Dario Cardona of the Universidad de los Andes, Colombia, Ivar Petteresen, Center ofEconomic Analysis (ECON) of Norway and Thomas Schaef of GTZ of Germany. A workingdocument version of the paper was also made available for the review by more than 30 participantsin the second session of the Regional Disaster Dialogue organized by the Bank in Washington,D.C. May 2002.

The views and opinions expressed in this documentary those of the authors and do not necessarilyreflect the official opinion of the Inter-American Development Bank or the other organizationsmentioned above.

October 2002

This publication (Reference No. ENV-139) can be obtained through:

Publications, Environment DivisionInter-American Development Bank1300 New York Avenue, N.W.Washington, D.C. 20577

E-mail: mariak@iadb.orghttp://www.iadb.org/sds/env

FOREWORD

Natural disasters have always been a part of the physical and human landscape in LatinAmerica and the Caribbean and they will continue to have a significant impact on the de-velopment of the region. Experience has shown that the economies of developing countriessuch as those in Latin America and the Caribbean are more vulnerable to losses than thedeveloped economies. They also suffer more fatalities. There is an urgent need to managerisk in the region.

Risk management and emergency response need to be very clearly distinguished. Riskmanagement calls for ex-ante planning and investments to reduce vulnerability whileemergency response involves ex-post expenditures which may be greatly reduced throughex-ante planning and investments in prevention and mitigation.

While the occurrence of natural events can not be prevented, there is a possibility to reducethe vulnerability of populations through risk management. The IDB Action Plan for naturaldisaster risk reduction of 2000 included recommendations in two important areas: (i) plan-ning with the purpose of the identification and reduction of risk by integrating preventionand mitigation measures into development plans and programs and (ii) financial protection,provided by the transferring of risk to others or spreading it in time. This paper deals withboth of these aspects.

Without adequate risk management disasters may seriously threaten the development ob-jectives by the borrowing member countries and jeopardize successful implementation ofBank mission in the region. There is a special need to balance risk reduction with risktransfer.

An important incentive to the preparation of the present document was given by the re-quest to the Bank by the Hemispheric Summit in Quebec City of April 2001. It asked theIDB to analyze the applicability of various financial instrument for disaster risk reductionin Latin America and the Caribbean. This paper is the first in a series to address this issue.

The ongoing discussions of the Policy and Evaluation Committee of the Board of Directorsof the Bank on the theme of disaster risk management further inspired the writing of thedocument. It is one of the themes to be addressed in the carrying out of the Bank’s visionof environmentally sustainable economic growth in Latin America and the Caribbean inthe coming years.

Walter ArensbergChief, Environment Division

CONTENTS

Overview i

Disasters in Latin America and the Caribbean 1

Risk Management 4

Vulnerability Reduction 7

Financial Protection 10

Conclusions and Recommendations 21

Bibliography 25

Annex 28

i

OVERVIEW

The Impacts of Disasters

One of this century’s biggest challenges forsustainable development in Latin Americaand the Caribbean is how to reduce the risk ofnatural disasters. The fundamental problemsfacing development in the region contribute toits vulnerability in the face of natural disas-ters. It is estimated that damages from disas-ters in the region have reached some US$20billion annually over a ten-year period, leav-ing a toll of 45,000 deaths and 40 millionpeople affected (ECLAC, 1999). Major ca-tastrophes have impacts that go beyond thedirect damages (which are normally quanti-fied in loss of life, number of people injuredand economic damages to those who are di-rectly affected) given that they also havenegative consequences for GDP, the balanceof payments, the level of indebtedness, thefiscal balance and investment.

Prevention and Mitigation

The economic and social development of thecountries of the region will continue to beadversely affected if adequate disaster pre-vention and mitigation measures are notadopted. To this end, risk management is akey concept. In the case of disasters, risk isdefined as a function of two factors: hazardand vulnerability. The former is relativelyeasy to detect, but its impact is hard to reduceif no plans are instituted for reducing vulner-ability. Therefore, it is indispensable to adopta full-range approach to risk management, aprocess that consists in identifying, analyzingand quantifying the likelihood of loss and, onthat basis, undertaking preventive or correc-tive activities. These activities may includeboth structural investments and nonstructuralactions to reduce vulnerability; among the

latter are financial protection mechanismsagainst the potential losses resulting from acatastrophe.

Building Awareness for Making Decisions

During the nineties (which the United Nationsdesignated as the International Decade for theReduction of Natural Disasters), the regionmade big strides forward in terms of under-standing the concepts of vulnerability and riskand the need to plan prevention and mitiga-tion actions. Nevertheless, pertinent officialsappear to have little knowledge of the rangeof existing options for obtaining financialprotection against the risk of disasters.

In the scientific and academic sector, majorknowledge has been gained regarding phys i-cal and financial protection against risk but,by and large, this information has not beentransferred to governments or to the privatesector. This has led to a situation in whichdecisionmakers continue to be ill-equipped toreduce the risk of natural disasters or to adoptmeasures tending to establish financial pro-tection against risks that cannot be reduced.

Financial Protection

A cost-benefit analysis to evaluate their long-term profitability will determine the mitiga-tion and prevention measures that a countryought to establish in order to reduce the highcosts of responding to emergencies, as well ascosts associated with rehabilitation and recon-struction. However, financial protection sys-tems are required because preventive andmitigation measures cannot eliminate allrisks. As part of this process it will be neces-sary to assess the potential effects that thecatastrophes may have on the country’s econ-

ii

omy. This evaluation of the risk and of itspossible levels of transfer will affect the im-pact that a disaster may have on the govern-ment budget, since in Latin America and theCaribbean, governments frequently shoulderthe greatest part of the risk, acting as insurersof last resort. Clearly, the government’s fi-nancial burden can be lessened if other actorsassume part of the risk.

The Credit System

The development of an efficient savings andcredit system through the region’s banks aswell as informal and microcredit institutions,would contribute to the mobilization of theresources needed to finance investments inprevention and mitigation as well rehabilita-tion and reconstruction.

The contingent credit mechanism makes iteasier to obtain financing in the event of adisaster. Yet, this requires that the clientcommit to an annual expense, and although arelatively low-cost instrument will, neverthe-less, increases indebtedness. Thus, to avoidthis, clients often prefer to obtain loans after adisaster has occurred, especially if they cangain expeditious access to them.

However, because it is generally difficult toobtain new loans immediately following adisaster, clients seek to refinance existingloans. This option has several disadvantages.First, it distorts the goals of the original cred-its. Second, it reduces the efficacy of the exe-cution of the original projects by the clientswho now seek to make use of the resourcesfor emergency situations. Third, it contributesto bad management of the original credits bythe financial entity, if it expects that the fundsmay be used to finance emergencies ratherthan for the purpose for which it was origi-nally extended. Finally, it makes clients de-pendent on this kind of ex post financing,instead of encouraging them to carry out pre-

vention and mitigation investments to dimin-ish risk.

Risk Transfer

The countries of the region are interested increating risk transfer instruments based onmechanisms currently in use in developedcountries, especially insurance. Since insur-ance premiums are a function of preventionand mitigation measures taken by the insuredparty, the establishment of insurance mecha-nisms would increase awareness of the needto invest in such measures and to put theminto practice. Financial protection againstdisasters through insurance mechanisms isattractive since it offers the opportunity oftransferring part of the risk, while avoidingthe need to increase indebtedness as a resultof an emergency. Financing through ex antecredits would offer even more incentives tomitigate risk because risk transfer instrumentstend to offer opportunities to contain moralrisk or adverse selection problems. The prin-cipal benefit of the transfer mechanisms re-lates to the distribution of risk that results inthe reduction of risk management cost.

The development of insurance markets re-quires updating the pertinent legislation andnormative framework. Although most of theweaknesses that exist are on the demand side(such as the lack of enforcement of buildingcodes and difficulties in establishing assetvalues, as well as the generally low capacityof clients to pay premiums), supply-side ad-justments are also necessary. These includestrengthening independent supervision sys-tems to improve monitoring of the solvencyof insurance companies and eliminate cond i-tions that favor anticompetitive practices. Tomeet this need, the supervisory entity shouldenjoy adequate oversight powers as well asthe ability to exact appropriate penalties. In-surance companies should also be encouragedto adopt international standards in order to

iii

maintain solvency, increase efficiency, andpromote transparency through the timelypublication of detailed and precise financialstatements.

There is also a need to establish the necessaryconditions for the use of innovative capitalmarket mechanisms such as parametric catas-trophe bonds, and weather-related derivatives.These instruments, which hold the interest ofinternational financial entities, avoid the ma-jor obstacles related to asset valuation andloss settlement procedures, but have to im-plemented at pool or governmental levels.

Establishing Priorities in the Use ofAvailable Resources

In the event of a disaster, the resources ofexisting and lower-cost sources would typ i-cally be employed first. If a fund for calami-ties, with adequate resources, is already inoperation (as in the case of Mexico’s FON-DEN, see Box 4), or coverage is available bymeans of insurance or catastrophe bonds, theyshould be used first. However, since suchinstruments don’t yet play an important rolein most countries, the trend is to first transferfunds from the government budget and fromdevelopment funds (municipal, social, urban,rural, etc.) to meet critical needs. This is usu-ally followed by a search for internationaldonations. Another instrument that could behighly useful is contingent credit, although itentails an annual administrative payment andindebtedness if the emergency takes place.

It is noted that instruments such as insurancemechanisms, common in countries with de-veloped financial markets, are not yet beingemployed in Latin America and the Carib-

bean. Governments can adopt a range ofmeasures to promote their development, in-cluding opening up the countries to interna-tional competition. As a first step they canimprove existing practices by insuring theirown critical assets through international tend-ers. On the other hand, international financialagencies such as the IDB could adopt stricterpolicies and require mandatory catastropherisk coverage, especially for private sectorprojects.

Comprehensive Approach

Risk management implies the important taskof reducing vulnerability and, as a result, riskitself, an undertaking that has barely begun inthe region. Hitherto, the predominant para-digm in the region has been to respond toemergency situations when they occur. Thismeans that there is a response to the phe-nomenon’s effects but without eliminating ormitigating the causes of the vulnerability thatcauses these effects. Ideally, prevention andmitigation measures ought to be combinedwith financial protection instruments in a co-ordinated action by the government and pri-vate sector. In addition to having a nationalemergency plan or strategy for ex post ac-tions, each country ought to design a consis-tent national plan or strategy to manage dis-aster risk. Those plans should be developedjointly with input from finance and planningministries, sector ministries and local gov-ernments, the private sector and civil society.Providing effective mechanisms for ex antefinancial protection is essential because itfacilitates the availability of funds when theyare needed most, and can reduce the ex postfinancial burden of recovery and reconstruc-tion following a catastrophic event.

1

DISASTERS IN LATIN AMERICA AND THE CARIBBEAN

Frequency of Disasters

Given the upward trend in the frequency andseriousness of natural disasters in LatinAmerica and the Caribbean, one of the mostimportant challenges for sustainable deve l-opment in this century will be how to managethe risk that they pose.

The region is threatened by a wide range ofseismic and weather-related phenomena. Theincrease in the frequency of recorded disastersand in related damages in recent years couldbe a consequence of greater vulnerability.That increased vulnerability could, in turn,result from an growing concentration of peo-ple, structures and other assets in urban cen-ters, making them more susceptible to disas-ters. Likewise, the lack of general awarenessregarding the importance of preventing andmitigating catastrophes, as well as institu-tional weakness, the persistence of poverty,the degradation of the environment and, pos-sibly, climate change, are leading to an in-crease in the seriousness and frequency ofthese natural events.

The disaster recorded during the last ten yearsresulted in over 45,000 deaths and affected 40million people, causing more than US$20billion in damages in Latin America and theCaribbean (ECLAC-IDB, 2000). As over-whelming as these figures are, they probablyunderestimate the real cost of disasters in theregion, since they do not include the effects ofmany events of lower intensity that go unre-corded. The increase in the incidence of dis-asters is not limited to Latin America and theCaribbean. As Figure 1 shows, it is a world-wide phenomenon.

Direct and Indirect Damages

Direct damages to goods and assets includethe total or partial destruction of dwellingsand other buildings, infrastructure and facili-ties, machinery and equipment, means oftransport, warehouses, furniture, harvests,cultivated land, dams, irrigation systems, etc.These direct damages are a function of thetype, geographical range and magnitude of theparticular natural phenomenon, as well as of

I n c i d e n c e o f D i s a s t e r s i n L a t i n A m e r i c a a n d t h e W o r l d ( 1 9 0 0 - 9 9 )

0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

1 90 0

1 90 4

1 90 8

1 91 2

1 91 6

1 92 0

1 92 4

1 92 8

1 93 2

1 93 6

1 94 0

1 94 4

1 94 8

1 95 2

1 95 6

1 96 0

1 96 4

1 96 8

1 97 2

1 97 6

1 98 0

1 98 4

1 98 8

1 99 2

1 99 6

0

5 0

1 0 0

1 5 0

2 0 0

2 5 0

3 0 0

3 5 0

L A C W o r l d

Figure 1

Source: Charveriat (2000).

LA

CW

OR

LD

2

Box 1. The Impact of Disasters on GDP: Hurricane Mitch

It is calculated that the damages caused by Hurricane Mitch in Central America reached US$6 billion in 1998,equivalent to 16 percent of the region’s GDP that year, 66 percent of its exports, 96.5 percent of its gross fixedcapital formation and 37.2 percent of its foreign debt. The most affected sector was agriculture (49 percent), fo l-lowed by infrastructure (21 percent), the social sectors (13 percent) and industry (10 percent). In Honduras, thedamages reached almost US$4 billion or 81.6 percent of GDP, 174.3 percent of exports, 343.9 percent of grossfixed capital formation and 94.1 percent of the foreign debt.

GDP growth in Central America averaged 4.3 percent per year between 1992 and 1998. Before Hurricane Mitch,GDP growth for 1999-2003 was forecast at an average annual rate of 4.8 percent. This rate of growth meant thatby 2004 the region’s per capita GDP would have returned to the 1978 level (US$1,166). More current ECLACestimates place the regions average annual growth rate for that period at only 3.6 percent, 1.2 percentage pointsless than it might have been. Consequently, the region will take a further three years to recover the “lost decade”and to reach the level of per capita GDP it had in 1978.

Source: ECLAC (1999b).

the vulnerability of the affected area(ECLAC, 1999a).

The economic impact of a disaster goes be-yond direct damages and also encompassesindirect ones, which are a consequence of thedestruction of productive capacity and of thesocial and economic infrastructure (see Figure2). These damages, which are very often hardto quantify, are cumulative and persist untilthe reconstruction and restoration of produc-tive capacity are complete. The damage togoods and assets (buildings, machinery, live-

stock, cultivable land, etc.) may have a sig-nificant impact on poverty because it impairspeople’s ability to find employment and gen-erate income. It is precisely the lack of in-come that, in many cases, delays economicrecovery.

Some of the damage caused by natural catas-trophes are irreversible. This is the case of thedeaths as well as of the irreparable destructionof natural resources like forests and cultivableland. Consequently, they make poverty moreacute and may increase social tensions, with

4,927

3,596 12,651

5,170

9,455

9,546

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

20,000

1972-1980 1980-1990 1990-1999

Daños directos Daños indirectos

Figure 2. Losses Due to Natural Disasters in Latin America and the Caribbean

Source: ECLAC (1999a)

Direct Damages Indirect Damages

3

potentially negative repercussions on the sus-tainable development process. However, withappropriate planning and financial protection,even a country with a relatively high inc i-dence of natural hazards can safeguard itseconomic growth potential.

Macroeconomic Effects

Natural disasters of a sufficiently large mag-nitude may harm the economic developmentof an entire country, adversely affecting GDP,the balance of payments, the level of indebt-edness, the fiscal balance and investment (seeBox 1). ECLAC (1999a) and IIASA (2000)refer to these macroeconomic consequences

of disasters as “secondary effects.” Cata-strophic events that affect large sectors of theeconomy generally reduce the pace of growth.For example, following a disaster, govern-ment revenues tend to decline while expen-ditures increase, leading to an increase in thegovernment deficit. Additionally, the declinein productive capacity and large public andprivate investments for reconstruction arelikely to reduce exports and increase imports,leading to a trade and balance of paymentsdeficits (which may be partly offset by theinflow of public and private disaster relief).Table 1 presents examples of the magnitudeof losses resulting from past natural disastersin Latin America and Caribbean.

Table 1.

E x a m p l e s o f G D P R e d u c t i o n D u e t o N a t u r a l D i s a s t e r sC o u n t r y Year T y p e o f D i s a s t e r G D P l o s s e s

( % )Argent ina 1 9 8 5 F l o o d 1 . 4 8Barbados 1 9 8 7 Hurr icane 6 . 8 6Bolivia 1 9 8 2 F l o o d 1 9 . 8 0Chi le 1 9 8 5 Ear thquake 9 . 1 0Cos ta Rica 1 9 9 1 Ear thquake 8 . 8 7H o n d u r a s 1 9 9 3 Hur r i cane /F lood 3 . 3 9Jama ica 1 9 8 8 Hurr icane 2 8 . 2 1M e x i c o 1 9 8 5 Ear thquake 2 . 1 8Nica ragua 1 9 9 4 D r o u g h t 8 . 7 4P a r a g u a y 1 9 8 3 F l o o d 1 . 3 6Peru 1 9 8 3 F l o o d / D r o u g h t 5 . 9 6

Source: Charveriat, 2000

4

RISK MANAGEMENT

The Concept of Risk

The management of risk is essential to re-ducing the damages associated with disasters.Risk is defined as a function of two factors:the natural hazard and vulnerability (see Fig-ure 3). It is the probability and magnitude ofthe losses associated with a specific naturalevent at a particular geographical point andtime. The incidence of natural events thatcould cause disasters lies beyond human con-trol; however, vulnerability can be controlled.

The estimate of natural threats is character-ized by the analysis of the frequency, magni-tude and location of disasters. This exerciseshould include both historical data and con-tinual vigilance of current natural threats, andthe preparation of forecasts, making use of thetechnological progress that has taken place infields such as seismology, vulcanology andmeteorology to estimate future threats. InLatin America and the Caribbean, the mostcommon threats recorded in the twentiethcentury were (in descending order of fre-quency): floods, hurricanes, earthquakes,landslides, droughts, volcanic eruptions and

fires. Among these threats, the first three ac-count for 77 percent of all catastrophic events(OFDA-CRED, 1999).

Risk management is defined as the process ofidentifying, analyzing and quantifying theprobability of losses in order to undertakepreventive or corrective actions (Schaming,1998). This involves two types of activities:1) planning actions to reduce vulnerability inareas where risk can be controlled, and 2)establishing protection mechanisms againstthe potential economic losses resulting fromuncontrollable factorsin order to offset thefinancial threat. A comprehensive approach todisaster risk management should emphasizeboth ex ante measures (prior to a hazard) andex post activities (see Figure 4). The mainelements of the ex ante phases of risk man-agement are: 1) identification and analysis ofthe risk; 2) mitigation and prevention; 3) pro-tection via risk transfer; and 4) preparations(see the Annex). The subsequent phases ofrehabilitation and reconstruction are betterknown.

Natural Hazard (NH)

Natural phenomena potentiallycausing losses to human settle-ments and economic activities

Vulnerability (V)

Susceptibility to human, eco-nomic and financial losses result-ing from the risk of natural disas-

ters

UNCONTROLLABLE CONTROLLABLE

RISK F (NH, V)Probability and magnitude of the losses

Figure 3. Natural Hazard, Vulnerability and Risk

5

Lester (2000) argues that private companiestend to maximize the net current value of theircapital with horizontal planning variables.Politicians, on the contrary, work withchronological frameworks related to electioncycles. Both actors may prefer shorter periodsthan those that are required by disaster riskmanagement, since not only is it difficult topredict beforehand and with precision whenthey will strike, but their recurrence may takeplace at extended intervals. Typically, it isonly when a catastrophe with serious eco-nomic, social and, therefore, political conse-quences happens that governments are moti-vated to take precise and concrete actions torespond to the disasters and to prevent futureones. This means that risk management tendsto gain in importance only after the damagehas occurred.

Ex ante and Ex post Actions

Countries affected by recurring events oftenfind themselves in a vicious circle of disaster-reconstruction-disaster, wherein each newcatastrophic cycle can increase the vulner-ability to future threats, as well as increaseindebtedness owing to the necessary invest-ments in rehabilitation and reconstruction.

Emergency respon-

Reconstructionand rehabilitation

Identification ofrisks

Prevention andmitigation

Risk transfer

Preparations

DISASTER

Figure 4. Risk Management as a Virtuous Circle

Box 2Risk Management:

Implementing the IDB Action Plan

The mandate of the IDB Group (the Bank, the Inter-American Investment Corporation and the Multilat-eral Investment Fund) is to contribute to the long-term economic and social development in the region,which also includes sustainable growth. As a result,this implies helping member countries to create afavorable environment for efficient risk management.Over the last 40 years, the Bank has financed activi-ties for the construction and reconstruction of basicinfrastructure, urban and rural development, envi-ronmental protection and natural resources manage-ment, many of which have contributed to reducingvulnerability to disasters. The Bank has also helpedgovernments to establish sector policies and reformsand to reorganize the public institutions that mayplay a positive role in the prevention of catastrophicevents. Nevertheless, there have been relatively fewcases in which a reduction in vulnerability was amain objective of the investments backed by the IDBor by other financial entities. In order to have agreater impact, it is essential to go beyond the cus-tomary rehabilitation or reconstruction and to takeproactive measures to define, understand and reducerisk. Thus, it is necessary to carry out a wide range ofrisk identification, prevention, mitigation and transferactivities, as defined in the Bank’s Action Plan for2000.

Source: IDB, 2000

6

This evidently has negative repercussions foreconomic development over the medium andlong term. However, the economic lossesfaced by a country that is frequently affectedby disasters can be minimized through pre-vention, mitigation and risk transfer becausethe losses resulting from a disaster tend to begreater than the cost of preventive invest-ments and financial protection. For this rea-son, both prevention and mitigation actions toreduce the amount of capital exposed to natu-ral threats, and establishing financial systemsto transfer risk and increase available ex postfinancing, could be very profitable political,economic and social strategies. Countries mayhave different optimum combinations of miti-gation and risk transfer depending on theirsituation. Economically, risk transfer tends tobe favored, while socially, high levels ofmitigation may be more acceptable.

Although investments in prevention and miti-gation, and financial risk transfer arrange-ments are apparently highly profitabile, real-ity is sometimes more complex. For manygovernments in the region, ex ante risk man-agement measures may involve costly outlaysthat appear unjustifiable in terms of a normalfour-year political cycle. They also tend to beeconomically unattractive when internationaldonations are easily available for eventual

reconstruction. In a certain sense, currentpolicies for the allocation of donations bybilateral cooperation agencies and of low-interest loans by multilateral banks act againstinitiatives to boost comprehensive risk man-agement. Although the increase in the level ofindebtedness may have negative future reper-cussions, the availability of these emergencyresources may counteract any existing incen-tive to carry out preventive actions.

Nevertheless, ex post international assistancewill never be able to fully offset the negativeimpact of a disaster on GDP, which can bevery significant. In addition, the internationalresponse may be inadequate. International aidis frequently in-kind, which has two majordisadvantages. First, the goods received maynot necessarily be the ones that are most ur-gently needed and, second, dealing with in-kind aid uses scarce resources during theemergency (means of communications andhuman resources), which are then unavailableto deal with more highly desired assistance. Inother cases, the promised international assis-tance does not materialize in time. For exam-ple, the US$190 million committed by theEuropean Union for Central American recon-struction after Hurricane Mitch, had yet to bedisbursed 18 months after the disaster tookplace (BBC, 2000).

7

VULNERABILITY REDUCTION

Risk Identification and Analysis

To identify the risk of natural disasters at anindividual, local or national level, it is neces-sary to estimate the potential magnitude andprobability of natural hazards, as well as toevaluate the vulnerability to each of them.Insurance companies typically use simulationmodels in their efforts to try to quantify oflikelihood of occurrence of natural hazardsand the related vulnerabilities. However, themodels are useful only if the basic informa-tion on which they rely is adequate.

Vulnerability may be evaluated from variousstandpoints (physical, social, political, tech-nological, institutional, environmental, cul-tural and educational). Vulnerability to natu-ral disasters is the result of anthropic factors;that is, factors that result from the interactionbetween human beings and nature. Addition-ally, vulnerability is a consequence of theindividual and political decisions that a soci-ety makes before a hazard occurs, which areevident once the disaster takes place(ECLAC-IDB, 2000).

Freeman et al. (2001) analyze the componentsof different types of vulnerability and citestudies make an effort to measure the poten-tial physical, social and economic conse-quences of natural phenomena. Those whoconcentrate on physical vulnerability analyzethe impact on buildings, infrastructures andagriculture. For example, the Council on Ap-plied Technology publishes vulnerabilitystudies on the earthquake resistance of 50types of structures (ATC, 1985). Those whofocus on social vulnerability estimate the im-pacts on especially susceptible groups such asthe poor, single-parent families, pregnantwomen and infants, the handicapped, children

and youths. Those interested in economicvulnerability calculate the potential impactson economic processes and assets.

The results of the hazard analysis and of theevaluation of vulnerability are then combinedto yield an estimate of risk (defined as ex-pected loss per period). A full scope evalua-tion of risk encompasses the appraisal of po-tential losses generated by the disaster andidentification of those affected by the risk.The evaluation of the risk makes it possible todevelop risk management strategies with twobasic components: i) prevention and mitiga-tion actions to reduce potential human, socialor economic losses, and ii) measures to estab-lish financial protection against the risks thatcannot be reduced.

The availability of information is critical forany action aimed at reducing the impact ofdisasters. Projections of the likelihood of theiroccurrence and estimates of their impact al-low decisionmakers to evaluate the total riskto a country, a geographical area or a specificsector, as well as to establish concrete pre-vention and mitigation measures and invest-ments.

Currently, there is a scarcity of risk informa-tion in the region. Although shorter method-ologies exist that generate adequate results inthe field of project evaluation, formal riskanalyses tend to be lengthy and costly(Bender, 1991). In addition, before moving onto the design stage of a project, a cost-benefitanalysis can give a more detailed idea of theusefulness of investing in risk identification.

In Latin America and the Caribbean, severalattempts have been made at identifying andmeasuring disaster risks. Among them is the

8

Box 3Mitigation Works: The Case of the Sabaneta Dam in the Dominican Republic

The Sabaneta dam, built by the Dominican government in the 1980s to regulate the volume of the San Juanriver, presented a great danger because the unfinished spillway structures considerably increased the risk ofrupture in case of strong rainfall.

In 1993, at the request of the Dominican government, the IDB approved the financing of a US$48 millioninvestment project that encompassed preventive reconditioning work on the dike and other fundamentalstructures of the dam. The repair work, which cost US$10.7 million, was completed during the first half of1998, just before Hurricane Georges struck. The areas most affected by the hurricane’s rainfall included,precisely, the province of San Juan de la Maguana, where the Sabaneta dam is located. The San Juan riverbasin received an average 320mm of rain during the hurricane. Nevertheless, the dam, the spillways, thepower plant and the dampening basin operated normally and did not suffer major damage, leading a panel ofexperts to the conclusion that the work that had just been completed had been very successful.

joint undertaking of the World Bank and theOAS in Saint Lucia, Saint Kitts and Nevis,and Dominica (Vermeiren and Pollner, 1994).The World Bank study of Mexico (Kreimer etal., 1999) is also a good example of this typeof analysis.

Freeman et al. (2001) make reference to sev-eral additional sources in the literature on thesubject of vulnerability, among which areEnvironmental Hazards: Assessing Risk andReducing Hazards by K. Smith (1996) andthe brochures produced by Swiss Re and Mu-nich Re, available in their electronic pages(www.swissre.com and www.munichre.com).

Prevention and Mitigation

Prevention and mitigation actions require agood understanding of natural threats, vulner-ability and risk. The phenomena that haverecently affected Latin America and the Car-ibbean show that, on many occasions, invest-ments in prevention and mitigation in the af-fected countries were not adequate to with-stand the natural threats. However, positiveexamples also exist (see Box 3), and severalcountries in the region, in many cases with thesupport of the IDB, have already begun toadopt measures in this regard. Even so, theidea still persists that prevention and mitiga-tion represent a cost and not an investment.Among the most common mitigation meas-

ures are the construction of structural projectssuch as the reinforcement of bridges, hospitalsand other public buildings. They also includeurban improvement programs, mountainsidestabilization work and surface drainage toreduce the danger of mudslides and floods, aswell as investments to safeguard natural re-sources. The Flood Rehabilitation Project inArgentina provides an example of flood miti-gation efforts. The project invested US$153million in structural improvements that savedapproximately US$187 million (in 1993 dol-lars) in damages during the 1997 floods, gen-erating an internal rate of return on invest-ment of 35 percent (World Bank, 2000b: 161-176).

There are several nonstructural measures thatcan also be implemented, among them, adoptnational laws and regulations to better re-spond to risks, land use planning, establishingmunicipal zoning regulations and constructioncodes, as well as education, training and im-proving awareness of the general public ofrisk, prevention and mitigation. According toECLAC data (1999b), 75 percent of the dam-age caused by Hurricane Mitch was related tothe inadequate location of buildings and infra-structure, such as dwellings built in floodprone areas or roads and bridges built on un-stable soil. Adequate land use planning, zon-ing and construction codes require politicalbacking since they may imply restrictions on

9

the use, purchase and sale of property, andaffect its value.

To reduce risk it is necessary to develop miti-gation plans that determine who, what, whenand where regarding the implementation ofpreventive or corrective activities. Theseplans must include a cost-benefit analysis toevaluate the long-term profitability of theprevention and mitigation measures. Evenwhen the broadest prevention and mitigationplans exist, damages from some truly cata-strophic events may only be partly reduced. Insuch cases, at least part of the risk should betransferred by means of financial protectionagainst unavoidable damages.

Among useful documents regarding mitiga-tion are the work by K. Smith (1996) referredto by Freeman et al. (2001) and quoted aboveas well as FEMA’s electronic page,www.fema.gov.mit.indes.htm. The CaribbeanDisaster Mitigation Project has publishednumerous texts on practices in that region,which are available at www.oas.org/en/cdmp/publist.htm.

Disaster Mitigation in Asia and the Pacific,produced by the Disaster Preparedness Centerof the Asian Technology Institute identifiesexemplary mitigation measures in Asia (Davisand Gupta, 1990).

10

FINANCIAL PROTECTION

Multiple Sources of Financing

To correctly identify a country’s financialprotection options against disasters it is nec-essary to know its tolerance to risk. Without ameasure of tolerance, the cost of transferringthe risk may appear excessive in comparisonwith the alternative of absorbing cost in someother manner (for example through taxationor foreign aid). The governments of the re-gion will have to analyze the different optionson the basis of their level of risk aversion andof domestic and external constraints that af-fect their economic and financial behavior.The aim of such an analysis is to adopt themost appropriate financial protection, or totransfer or assume the risk and assure theavailability of financing after a disaster.

Traditionally it is assumed that, unlike theprivate sector, governments have a neutralattitude vis-à-vis risk, an assumption based ontheir capacity to impose taxes and transfer therisk to taxpayers. Government administrationsare less limited by solvency or liquidity con-siderations than major corporations. However,the government of Latin America and theCaribbean have not been able to absorb therisk of losses caused by disasters (maybe be-cause tax collection is inefficient or very lim-ited).

As a general principle, importance is given tomobilizing existing resources and seekingnew sources of funds. The latter includesidentifying new instruments to increase thesupply of financing. The countries of the re-gion have access to a variety of sources offinancing, among them government resourcesand national and international credits or do-nations, to finance both the ex ante and ex

post investments related to disasters. Never-theless, and as pointed out above, ex post in-ternational resources may have drawbacks(donations are often in-kind and the financingmay be slow for the handling of emergen-cies).

It is useful to strengthen national financing,especially through the banking system. Thedevelopment of an efficient savings and creditsystem, both by commercial banks andthrough microfinancing entities, will facilitateprivate financial protection against risk. It ispossible to turn to contingency credit foremergencies offered by international banks,or by governmental and private domesticones. However, this entails annual adminis-trative expenses during the period prior to thedisaster. It also increases the debt burden ofthe country.

Existing development and social funds mayserve as instruments to finance the preventionand mitigation of disasters and, to a lesserdegree, reconstruction. Specialized reservefunds for disasters have been created in sev-eral countries, normally for ex post financing.In the case of Colombia, these funds also pro-vide resources for ex ante investments.

In some countries of Latin America and theCaribbean, risk transfer contracts have beenestablished with insurance or reinsurancecompanies, basing their payouts either on reallosses or in some cases on parametric activa-tion triggering indicators (for example, pay-ment to the insured would be tied to the inten-sity of an earthquake, say, a ranking of 6.5 onthe Richter scale, independently of the reallevel of loss). Other instruments, such as ca-tastrophe bonds, also exist, although they are

11

still considered very new in the region. Theprivate sector also has the direct investmentoption.

The community-wide formal and informalfinancing instruments, perform a very impor-tant role at the local level by supplying re-sources, particularly in poorer areas. Regard-less of the source of financing, the imple-mentation of these mechanisms requires closecooperation between the public and privatesectors, especially in reference to the estab-lishment of the appropriate legal and regula-tory framework.

Table 2 summarizes the potential sources ofex ante and ex post disaster financing. The exante non-reimbursable and reimbursable fi-nancing mechanisms without risk transfer,include donations and credits. The corre-sponding risk transfer instruments encompassinsurance and catastrophe bonds, which cancover the damage based on real losses (in-demnification) or through the parametric acti-

vation of payments. Financing instrumentsestablished ex post include donations, taxesand emergency and reconstruction loans, andrefinancing of existing loans.

In the event of a disaster, immediately avail-able and lowest-cost financing options wouldtypically be used first. For example, financingthrough an existing calamity fund and/or in-surance, reinsurance or catastrophe bondswould have priority. Similarly, part of budg-eted resources from existing government pro-grams would be transferred to meet immedi-ate emergency needs. In some cases, deve l-opment funds (municipal, social, urban, rural)may also be used. At the same time, the gov-ernment would seek as much international aidand donations as possible and resort to con-tingency credits. If the government has accessto emergency credits such as the IDB’s Emer-gency Reconstruction Mechanism (IDB,1999), it would request them and would alsobegin negotiations to direct resources fromexisting loans to finance disaster recovery

Table 2. Provisional Classification of Disaster Financing Mechanisms

Ex ante Sourcesa) Ex post Sources

Inst

rum

ents

with

out r

isk

tran

sfer

Nonreimbursable re-sources? Calamity funds? Reserve funds or di-

version of nationalbudgetary resources

? Development and so-cial funds

Reimbursable resources? Contingent credits? Development and so-

cial funds

Inst

rum

ents

with

ris

k tr

ansf

er

? Insurance and reinsur-ance with damage cov-erage based on reallosses

? Insurance and reinsur-ance with parametricactivation of payments

? Catastrophe bonds withdamage coverage basedon real losses

? Catastrophe bonds withparametric activation ofpayments

Nonreimbursable resources? Emergency donations? Taxes

Reimbursable resources? Emergency credits (for

example the IDB’sEmergency Reconstruc-tion Mechanism)

? Reconstruction loans? Reformulation of existing

loans

Note: a To finance prevention and mitigation investments (not losses), special mechanisms may be used both asdonations or loans (through the IDB’s Sector Facility for Disaster Prevention, for example). In addition to thecontingent credit, private sector companies can use contingent equity as an instrument. The table excludes other,less common instruments such as weather derivatives, swaps and pools.

12

investments and request new reconstructioncredits. Special taxes can also be used to fi-nance reconstruction (this option was em-ployed successfully in Colombia followingthe 1999 earthquake in Armenia).

Mobilization of Government, Multilateraland Bilateral Resources

A posteriori financing is currently the mostwidely used instrument by the countries of theregion, mainly via the diversion of resourcesof the national budget toward rehabilitationand reconstruction activities. Even though theuse of these resources does not entail a finan-cial expense in the form of commission orinterest payments, its opportunity cost can beonerous for the national economy. Althoughnew taxes could be levied, as was the case ofColombia in 1999, this is not a viable alter-native for most Latin American and Carib-bean countries.

When large-scale disasters take place, gov-ernments assume de facto roles as financingsources of last resort. These financial com-mitments can be divided into three categories:i) funds allocated to cover the financial costof the damages to public sector infrastructure;ii) financing made available as a result of po-litical pressures to private businesses wholacked sufficient insurance coverage; and iii)funds to meet the government’s obligations tocare for the poor. The lack of comprehensiverisk management adequate insurance markets,together with the governmental’s inability toabsorb losses out of current revenues meansthat the public sector may take on new finan-cial commitments that increase significantlyits debt burden.

Bilateral and multilateral assistance may takethe form of reimbursable or nonreimbursablefinancing and, in some cases, the refinancingor forgiving of past debts. The availability ofex post financing via donations or through

low interest loans tends to create perverseincentives, insofar as governments may preferto depend on these soft foreign resourcesrather than adequately manage risk to forestallthe impact of disasters. As has been noted,foreign aid is not always immediately avail-able, nor does it always come in the form thatwould be required by the particular country ata time of crisis. One way to guarantee theavailability of a posteriori financing would bethrough a system of contingent credit linesthrough international banks. In the event thatspeedy access to financing has a major impacton the rate economic recovery (and hence onthe government’s budget), the call option maybe a valid alternative (Lester, 2000). Althoughinternational financial entities such as the IDBmay not formally offer a contingent creditinstrument, they may act as de facto sourcesof contingent lines of credit through the real-location of existing loans to rehabilitation andreconstruction. If these entities charged apremium for this service, it would help coun-tries to internalize the cost of this form offinancial protection.

Multilateral and bilateral international agen-cies (like the IDB) can offer major assistancein analyzing the feasibility of different fi-nancing mechanisms given a country’s eco-nomic circumstances and risk aversion. Thiswould enable the transfer and diversificationof risk. Additionally, basic studies can allowcountries to assess fundamental aspects ofrisk management, namely: i) risk identifica-tion, ii) prevention and mitigation to protectessential infrastructure and human lives, andiii) efficient disaster responses. The IDB canalso help countries to adopt institutional andpolitical mechanisms allowing the creation ofthe permanent technical and operational ca-pacity to guarantee a sustained investment inrisk reduction (see the Sectorial Facility forDisaster Prevention in Box 3). Lastly, andalthough it is not the preferred procedure, theBank and other agencies can help the coun-

13

tries of the region by providing immediatefinancial resources to meet the emergencyneeds and promote the speedy restoration ofbasic services (see the Emergency Recon-struction Mechanism in Box 4).

Financing Prevention and Mitigation

Prevention and mitigation funds finance thereduction of vulnerability. In order not to runthe risk of exhausting their resources in caseof an event that causes significant losses,these funds should not provide financial pro-tection to the public or private sector after adisaster. Consequently, their by-laws shouldclearly stipulate that their resources are notavailable to finance emergency, rehabilitationor reconstruction activities. The compositionof their governing boards must clearly reflect

this commitment (development agencies vs.civil protection agencies).

There is a wide array of development fundsthat can be used to finance investments inprevention and mitigation. In Latin Americaand the Caribbean there are municipal (urbanand rural) development and environmentalfunds that can allocate resources for the pre-vention and mitigation of catastrophic eventsin addition to their normal activities. Some ofthese funds operate with reimbursable re-sources and allow the financing of major in-vestments. They could be used for the recon-struction or rehabilitation of critical installa-tions like hospitals and schools, water andsanitation infrastructure, or major roads.Other funds operate with non-reimbursablefinancing and could be applied, for example,

Box 4The IDB’s Sector Facility for the Prevention of

Disasters and Emergency Reconstruction Mechanism

Sector Facility for the Prevention of Natural Disasters

In March 2001 the Bank approved a new financial mechanism to carry out and strengthen actions prevent and miti-gate disasters. The Sector Facility for the Prevention of Disasters will support the establishment of pilot programs inthe full-range management of risk reduction and disaster prevention. These programs will also help to reinforceinstitutional capacity before larger-scale programs are carried out.

The Facility will provide reimbursable resources of up to US$5 million per project that are aimed at strengtheningdisaster prevention and risk management systems. These investments cover many areas, including policy and insti-tutional development to prevent establishment and adaptation of innovative financial instruments (risk reductionfunds, contingent financing agreements and insurance systems, etc.); forecasting and monitoring of potential threats;early warning systems; and priority strategies and mitigation investments. Other areas include education and train-ing, development of risk reduction technologies and information systems for monitoring and evaluation. There areseveral projects in the Bank’s pipeline that request financing through this instrument.

Emergency Reconstruction Mechanism

The Bank created the Emergency Reconstruction Mechanism (ERM) in December 1988. The ERM allows the Bankto respond quickly to disasters with loans of up to US$20 million from ordinary capital or US$10 million in re-sources from the Special Operations Fund (concessionary resources). Its goal is the immediate availability of thenecessary resources to finance a pre-established menu of eligible activities, which includes assistance to re-establishbasic services, financing for immediate repairs and cleanup work in the period following a natural disaster. A coun-try’s request launches an accelerated process of approval of the financing which can take two to four weeks. Due tothe high demand for ERM resources, the original US$100 million endowment has already been exhausted. Theprogram is being reevaluated in 2002.

Sources: IDB (1999) and IDB (2001).

14

for studies to evaluate risk and the vulnerabil-ity of populations and assets, as well as forland use planning and zoning (see the Annex).In ex-post financing prevention and mitiga-tion investments are often included in an ef-fort to ensure that actions are taken to reducefuture vulnerability to disasters.

For each of these funds there is a wide arrayof basic considerations that require specialattention in the design stages. This refers totheir legal structure and sources of capitaliza-tion, operating policies and rulings for thefinancing of projects, etc. Administrative andfinancial independence from politicalauthorities (public or private administration)and the systems established to administer theresources are important additional planningfactors that affect the financial sustainabilityof these funds.

Insurance and Reinsurance

In developed countries, the traditional way oftransferring the financial cost of disasters overtime and among the various actors is throughinsurance offered by private companies. It is amechanism that allows the transfer of finan-cial risk from one entity or individual to a

collective. This group can be expanded evenfurther through international reinsurance. Inseveral countries of the region, the govern-ment also insures some of the public assets.Insurance coverage for the latter lessens thefinancial burden of reconstruction for theState after a disaster, thus allowing govern-ments to be less dependent on contingent in-ternational assistance and to avoid divertingalready allocated resources from the nationalbudget to emergency investments, rehabilita-tion and reconstruction.

In addition to guaranteeing quick access tofunds for reconstruction, insurance mecha-nisms have the advantage of promoting pre-vention and mitigation actions, for examplethrough discounts in premium levels and de-ductions offered to customers who have madeprogress in those fields. At the same time,insurance companies carry out inspections toreview the condition of the assets insured,both during and after a disaster, which pro-motes improved care of those assets. The useof insurance allows risk diversification amongthe parties and reduces variation among theinsured, in addition to facilitating the segre-gation of the risk (Freeman and Kunreuther,1997).

Box 5Mexico’s FONDEN: A Calamity Fund

Mexico’s Natural Disasters Fund (FONDEN) is a government effort to reduce the negative effects of naturalcatastrophes, strengthen disaster preparedness and response capabilities. The goal is to be able to deal with theeffect of unforeseeable natural disasters whose magnitude surpasses the response capability established by theauthorities. The aim is to create resources that make it possible to handle the damages without altering publicfinances for normal federal, state or municipal government programs.

FONDEN has been requested to undergo a decentralization process to reduce its bureaucracy, regulations anddiscretionality, and increase the agility of gaining access to its resources. Actions have already been under-taken to improve the transparency of the allocation of resources, and reduce the time needed to process, deliverand employ resources in the affected areas.

The Fund’s resources have averaged about US$100 million per year. Between 1996 and 1999, 70 percent ofthe payments to handle losses were for hurricanes and floods, 19 percent for droughts and frosts, 10 percent forearthquakes and 1 percent for fires.

Sources: EIRD (1999); Mexican Senate (2001); Freeman and Mechler (2001).

15

As is shown in Table 3, insurance coveragefor natural disaster losses varies significantlythroughout the world. It reached 3.9 percentin Latin America during 1985-1999, com-pared to 34.3 percent in the United States andCanada, 26.7 percent in Europe, and 4.3 per-cent in Asia. Penetration is even low insur-ance against natural disasters. For example,according to the World Bank, of the total 16million dwellings in Mexico only 150,000(less than 1 percent) had disaster coverage in1998 (World Bank, 1999). It is evident that inLatin America and the Caribbean there is onlyan incipient market for insurance in general.When measured in relation to the GDP, theproportion of premiums for all types of insur-ance in the two sub-regions reaches only be-tween 0.5 percent and 2.7 percent (in com-parison, in the U.S. it has reached 3.4 per-cent).

Various factors account for the scarce deve l-opment of the insurance market in the region,among them are: the lack of competitivenessand competition in the insurance and reinsur-ance market; the absence of control and su-pervision of the quality and stability of theinsurers; the exclusion of foreign investmentin the insurance sector; the lack of trainingand professionalism among insurance bro-kers; inefficiency in the timely payment of thecoverage; the financial weakness of the com-panies, and the fact that the legislation, regu-lation and supervision of insurance does not

adhere to international standards (IFC, 2000;Skipper, 1997).

To counteract these deficiencies and to de-velop the insurance market, actions of thefollowing kinds will be required: eliminatingnormative obstacles, including the access bar-riers that affect the entry of foreign companiesinto the market; establishing and enforcingconstruction codes; carrying out land-useplanning and zoning on the basis of potentialdamages; disseminating information regard-ing risk; and making governments and donorscommit themselves not to guarantee the fi-nancing and re-establishment of those lostassets that could be insured. These measuresto stimulate the markets would help to lowerinsurance premiums and make them moreaccessible to medium- and low-incomegroups.

The creation and strengthening of insurancemarkets is a long-term strategy for transfe r-ring and reducing risk. Such markets providesome payment transfer mechanisms and con-stitute an additional source of financial re-sources for the insured after a disaster. At thesame time, depending on the payment struc-ture of the premiums, these mechanisms canalso be used by the insurers to promote theadoption of preventive and mitigation meas-ures by their customers. Figure 5 describesthe process of defining insurance premiumsthat would constitute the basis for comparing

Table 3. Insured Natural Disaster Losses 1985-1999Insured losses(US$ millions)

Percent of total losses

North America 116,940 34.5Europe 29,990 26.7Australia and Pacific 4,330 25.7Africa 610 8.9Asia 17,640 4.3Latin America and Caribbean 420 3.9Source: Munich Re 2000.

16

that instrument with other available financingmechanisms.

Insurance markets can exist thanks to Ber-noulli law of large numbers, according towhich, for a series of independent variables,the variation of the average compensationdiminishes with an increase in the number ofclaims. However, the covariant nature of thistype of risk (i.e., the fact that many of theinsured will suffer the same calamity at thesame time and in the same place) may elimi-nate the benefits of the insurance if the com-panies do not diversify their portfolio to sev-eral geographical areas and over time. In ad-dition, it should be recalled that insurance

systems harbor four types of potential prob-lems: adverse selection, moral risk, base riskand credit risk (Andersen, 2002).

Adverse Selection. This is based on the as-sumption that the insured may know moreabout his own level of risk than the insurancecompany. Consequently, the demand for in-surance coverage usually comes from thosefacing higher risks, thus raising the probabil-ity of losses. The premiums may be increasedeven for those who face lower risks. Thisproblem can be mitigated through an im-proved gathering of information regarding theinsured (Rotschild and Stiglitz, 1976; Hillier,1997).

Vulnerabilitybecause of

location

Frequencyand size of

risk

Value of assetsat risk

Structural vulne-rability (infras-tructure, buil-

dings, etc.)

Probablelosses

Insurancepremium

Cost of otherfinancial mecha-

nisms

Comparison ofinsurance withother existing

financial mecha-nisms

Figure 5. Process for Comparing Risk Premiumsand the Cost of Other Financial Mechanisms

Prevention andmitigation mea-

sures

Source: Adapted from World Bank (2001).

17

Moral Hazard. This presents itself when theexistence of insurance coverage causes riskierbehavior by the insured, for example, bybuilding in an unstable location or with poorquality materials. Moral risk can be reducedby adjusting the premiums according to theconditions of the risk and the adoption ofmitigation actions (Grossman and Hart, 1983;Doherty, 2000).

Basis Risk. This emerges from discrepanciesbetween the risk indicators employed in thefinancial instruments and the positions thatthese instruments attempt to cover. For exam-ple, if the value of the index maintained bythe instrument to cover the risk differs signifi-cantly from the value of exposure to the riskthat it must cover, the instrument will face alarger base risk, since it is difficult to apply ageneral index to an individualized risk port-folio (Andersen, 2002).

Credit Risk. Concern may exist over the levelof credit risk of the solvency of the issues ofan insurance policy. For example, disasterrisk coverage provided by an insurance com-pany with a low credit reputation endangersfuture coverage, since a natural disaster en-tails additional pressures on the solvency ofthe weak companies in the sector. The use ofderivative financial instruments that aretraded on stock exchanges, or the issuing ofguarantees in capital markets, will reduce oreliminate the credit risk of the counterpart(Andersen, 2002).

The insurance industry has created mutualfunds that allow a group of insurance compa-nies (normally national ones) to transfer thehighest risks to a common fund; that is, tospread the risk among a larger number of ac-tors and a wider geographical area. This maybe voluntary, as in the case of the CaliforniaEarthquake Authority, or compulsory, as inthe Norwegian national system. For thesefunds to work a certain level of development

must exist within the insurance industry,which is normally more likely in big coun-tries, or requires a regional context to facili-tate certain levels of diversification.

The creation and strengthening of insurancemarkets is a long-term strategic goal for trans-ferring and reducing risk. Such markets pro-vide some payment transfer mechanisms andconstitute an additional source of financialresources for the insured after a disaster. Atthe same time, depending on the premiumpayment structure, such mechanisms can alsobe used by insurers to promote the adoptionof preventive and mitigation measures bytheir customers. Figure 5 shows how to es-tablish insurance premiums and comparethem to other available financing mecha-nisms.

Regional catastrophic events have surpassedthe capacity of national insurance and rein-surance industries to provide coverage at areasonable price. In industrialized countries,where coverage is high, only part of the riskof natural threats is transferred internationallyvia reinsurance. However, in Latin Americaand the Caribbean—where many local insur-ance companies are not sufficiently capital-ized to handle natural disasters—most premi-ums are turned over to international reinsur-ance companies. In the Caribbean, for exam-ple, the real estate insurance market turnsover approximately 80 percent of the premi-ums to reinsurers in Europe and the UnitedStates. For consumers, this tends to raise theprice of premiums, which instead of beingsubject to the risk conditions of the local user,are subjected to international appraisals (Poll-ner, 1999; see also Box 6). In some cases theinternational reinsurance companies acceptvirtually the entire risk at all levels coveredby insurance, while the local ones act more asbrokers, a practice that tends to increase theprice of the premium for the consumer.

18

A study by the International Financial Corpo-ration for Central America indicates that cov-erage for buildings losses caused by disastershas traditionally been combined with fire in-surance policies. In the case of Honduras,these policies amounted to 20.3 percent of thevalue of all the insurance premiums in thecountry in 1998. Payments for losses causedby Hurricane Mitch that same year reachedUS$160 million, while total losses wereUS$3.8 billion. As a result of this limitedcoverage, several of the affected companiessuffered economic losses for not having rein-surance (IFC, 2000).

As mentioned, in addition to indemnificationinsurance that covers real losses, there areparametric insurance instruments. Payment ofthe claim using parametric insurance is notcalculated on the basis of the real loss, butaccording to a triggering indicator; for exam-ple, an earthquake measuring 6.5 on theRichter scale would trigger a predeterminedpayment). Because the payment does not de-pend on an appraisal of real losses for eachinsured party, the high administrative costsassociated with damage inspection and ap-praisal are significantly reduced. This instru-

ment has the advantage that it does not de-mand title deeds to the insured assets. Para-metric instruments are employed in areas witha shortage of data because they rely on indi-cators that are easy to measure (such as windspeed in the case of hurricanes, rainfall in thecase of floods or droughts, or the intensity ofan earthquake) to determine the compensationowed to the insured. They are considered es-pecially important when coverage for thepoorer segments of society is involved.

Catastrophe Bonds

Recognizing the lack of financial capacity inthe face of natural disasters, the insuranceindustry has attempted to transfer part of theserisks through capital markets by floating ca-tastrophe bonds, options and similar instru-ments. Catastrophe bonds could also com-plement the use of the special funds (men-tioned earlier) to transfer the higher levels ofrisk faced by them (Pollner, 1999). This,likewise, involves derivatives indexed toweather conditions (although these often onlyprovide protection against events that are notvery extreme), as well as contingent financing(such as retroconvertible debt). Both allow therefinancing of the investments lost in a disas-ter, at low interest rates.

The most promising instruments in this sys-tem are variations of the catastrophe bonditself, the basic mechanism of which is verysimple. The party that wishes to transfer therisk would issue a special bond. In the eventof a disaster, interest payments by the issuermay be cancelled or s/he may receive a per-centage of the bond’s principal. The percent-age would depend on the magnitude of thecatastrophe (triggering indicator) and on theterms of the contract. In both cases, the fundsavailable to the issuer, be it for not paying theinterest or for receiving part of the principal,would act in practice as the payment of aninsurance claim. The investors (buyers of the

Box 6The Case of Insurance in Barbados

In 1990, the average premium for real estate insur-ance in Barbados was 0.4 percent of the estimatedvalue of the asset. This means, for example, that toinsure a property appraised at US$100,000 the ownerwould have to pay a premium of US$400. FollowingHurricane Andrew, premiums rose to 1.3 percent,despite the fact that Andrew did not hit Barbados.This 300 percent increase in the price of insurancewent into effect even though there were no changesin the underlying risk in the country. It is possiblethat the premium before the hurricane was too low;nevertheless, it is also likely that a full-range subre-gional management of the risk, as well as the avail-ability of better information on the particular situa-tion in Barbados, would improve the stability ofpremiums.

Source: World Bank 2000

19

bonds), for their part, run the risk of not re-ceiving the payment of interest or that theprincipal may have diminished when the bondis redeemed. Thus they would charge higherinterest rates for running these risks (see Box7).

Some of the new financial instruments mightbe applied in Latin America and the Carib-bean, as long as there exist legal and regula-tory frameworks, adequate institutionalmechanisms and the corresponding risk in-formation allowing the development of the

market. However, these highly specializedinstruments will not necessarily lead, on theirown, to a reduction in the losses caused bydisasters. For them to be successful, the pub-lic must become aware of the multiple effectsof catastrophic events and have incentives toadopt prevention and mitigation measures toreduce the costs of coverage.

Table 4 summarizes the advantages of rein-surance contracts and catastrophe bonds (bothparametric bonds and those based on reallosses) in the face of adverse selection, moralrisk and credit and basis risk. It also indicatesthe level of market acceptance of each in-strument. As shown, parametric instrumentshave lower adverse selection and moral riskthan those based on real losses. Credit risk isgreater for reinsurance than for catastrophebonds. At the same time, it is estimated thatcatastrophe bonds based on real losses mayhave lower market acceptance than reinsur-ance contracts or parametric bonds. Accord-ing to the author of the table, the base riskmay be relatively acceptable for all four op-tions analyzed.

Community Strategies

Community financing mechanisms channelresources to the populations that need themmost; they are particularly important for re-ducing the vulnerability of impoverishedgroups. Included in this category are: i) socialor municipal funds, which are frequently fi-nanced in part with multilateral or bilateralresources that are disbursed through munici-pal governments and other local entities; ii)community development projects financed bynational and international nongovernmentalorganizations ; iii) microenterprise credit pro-grams, and iv) informal financing mecha-nisms such as local informal credit markets.

Box 7Earthquake Bond: Tokyo Disneyland

In May 1999 the Oriental Land Company, the ope-rator of the Tokyo Disneyland, issued a bond tocover against the risk of financial losses in theevent of an earthquake. This bond is a type of ca-tastrophe bond. It does not offer financial protec-tion against direct damages like the destruction ofbuildings, but against cash flow losses that mightensue from a reduction in the number of visits. Thefunds obtained by floating the bond, equivalent toUS$200 million, were deposited in a special ac-count in the Cayman Islands. Oriental Land belie-ves that the Tokyo Disneyland infrastructure wouldnot suffer major damage even in the event of a veryintense earthquake. However, if transport meanswere destroyed or if consumer confidence werelacking, visits to the park could be significantlyreduced. The company calculates that it would take10 months to restore the public’s confidence andthat US$180 million would be needed to survivethat period.

If an earthquake takes place within a 75-kilometerradius of the location of Tokyo Disneyland, theterm s of the bond would allow Oriental Land toreceive a percentage of its principal, which woulddepend on the intensity of the earthquake. Inves-tors, for their part, run the risk that the principalmay have diminished when the bond falls due infive years.

Source: Nikkei (1991).

20

Decentralization policies in the region haveled to the delegation to local governments ofgreater responsibility for supplying publicservices. This opens the opportunity for chan-neling resources from multilateral or bilateralagencies directly to those local governments.(The limiting factor is often, however, theneed for sovereign guarantee of internationalfinancing through loans). This new role forlocal administrations also requires improve-ments in their fiscal management systems andthe diversification of their sources of income.Capable municipalities are better able to ad-minister resources aimed at mitigation andprevention activities and to enforce local rul-ings that reduce risk. It is the communitiesthemselves that, in many cases, can best iden-tify their vulnerability and thus direct the in-vestments toward those areas where they willhave the greatest impact.

Vatsa and Krimgold (1999) distinguish be-tween two types of risk management strate-gies at the individual or community level: i)

those designed to reduce risk and ii) thosedesigned to cope with risk. Risk reducingstrategies involve diversifying the family’ssources of incomeand include, for example,taking on a second (or third) job, avoidingdependence on one crop, etc. Families canengage in a variety of activities that generateadditional income, increase and diversify theirassets to protect against the seasonal eco-nomic shocks, and ensure an adequate flow ofincome after a catastrophic event. Strategiesdesigned to cope with risk seek to minimizelosses in well-being and prevent interruptionsin consumption These include, for example,having access to loans and savings and insur-ance.

The main disadvantage of community mecha-nisms is the difficulty in transferring riskgiven that all members of a community tendto suffer the same calamities. The geographiclimitation of a community would call for co-operation with other regions in the risk-taking.

Table 4. Risk Elements (+ = high risk level) and Expected Market Acceptance (?= acceptance) for Reinsurance Contracts and Catastrophe bonds

InstrumentAdverseselection

Moralrisk

Base risk Creditrisk

Marketreception

Reinsurance contract- Parametric - - - + ? - Real loss + + - + ?

Catastrophe bonds- Parametric - - - - ? - Real loss + + - - -

Source: Andersen (2002).

21

CONCLUSIONS AND RECOMMENDATIONS

Prevention and Mitigationversus Financial Protection

A cost-benefit analysis to evaluate the profit-ability of long-term mitigation and preventionmeasures will determine those activities thatwill help reduce the high costs of respondingto emergencies, as well as rehabilitation andreconstruction expenses. Since prevention andmitigation measures cannot eliminate all po-tential damages, risk cannot be totally sup-pressed, making financial protection systemsnecessary.

The participation of governments, the privatesector and civil society in prevention, mitiga-tion and financial protection schemes can bevery varied. In Latin America and the Carib-bean, the government frequently assume thegreater part of the risk, acting, to a certainextent, as insurers of last resort. Conse-quently, the evaluation of the risk and how itmight be transferred under different types ofdisaster scenarios will affect the potentialimpact of disasters on the government budget.Clearly, the government’s financial burdencan be reduced if the private sector assumespart of the risk

Government-backed prevention and mitiga-tion activities can be financed through thenational budget, with private sector resourcesthrough municipal, social, urban or rural de-velopment, or environmental funds. SomeLatin American and Caribbean countries (forexample, Colombia) have established specialprevention and mitigation funds.. Some mul-tilateral banks and bilateral internationalagencies have also established mechanismsthat can contribute to this type of financing.An example is the Inter-American Develop-

ment Bank’s Sector Facility for Disaster Pre-vention (IDB, 2001).

Specific country studies should be carried outto determine the risks faced and the likelyimpact of different disaster scenarios on thenational economy. This analysis should pro-vide answers to the following key questions:How would the rehabilitation and reconstruc-tion of public assets (like road infrastructure)be financed? How will the resulting increasein social spending be financed? Will the gov-ernment compensate the private sector andcommunities for damages to important assetssuch as dwellings? In order to reduce theirexposure to risk, national authorities mustdevelop a national risk management plan thatestablishes who would carry out preventionand mitigation activities, what activitieswould be involved, when and where theywould be carried out and what would be themost desirable financial protection strategy(Cardona, 2001).

Toward Development of a Risk MitigationProtocol by the Banking System

The mobilization of resources to finance pre-vention and mitigation investments, as well asrehabilitation and reconstruction could beenhanced by the development of an efficientsavings and loan system through the banks aswell as informal and microcredit entities.

Commercial banks are exploring measures toreduce the natural disaster risk in their portfo-lio. One of the mechanisms considered con-sists in establishing a self-imposed preventionand mitigation protocol on the financing ofinvestments in rehabilitation and reconstruc-tion. Such is the case of Brazil’s “Green Pro-tocol,” through which environmental risk

22

management measures were incorporated intofinancing programs through the banking sys-tem (Bayon et al., 2000). The goal of a disas-ter mitigation protocol is to prevent the repli-cation of vulnerability and promote sustain-able economic development; thus contributingto the financial health of the banking systemover the medium and long term. It wouldserve to break the vicious circle of providingfinancing for projects whose design does notincorporate appropriate measures to withstandthe impacts of a disaster, by making preven-tion and mitigation components a compulsorypart of loan projects. Disaster risk evaluationsshould be part of the feasibility studies ofdevelopment projects (EIRD, 2002).

The successful adoption of a protocol of thisnature would be based on existing legislationin each country. The protocols would not be alaw but a code of conduct and a set of bestpractices that would help the signatory insti-tutions to incorporate ecological concerns intotheir decision-making processes. Neverthe-less, some banks could be reluctant to adoptthis type of measure, perhaps because a pro-active approach on their part could have un-desired legal consequences. For example, inthe United States a bank that places specificconditions on risk management in the projects

it finances may also assume legal responsi-bilities if those conditions are fail to providethe protection envisioned. This is becausecourt decisions have equated the impositionof conditions with an implicit obligation tomake sure that those conditions are appropri-ately met.

Multilateral and bilateral banks can also helptheir public and private sector clients to iden-tify disaster-related vulnerabilities and risks inthe projects they finance. In the case of theIDB, the role of the Committee on the Envi-ronment and Social Impact (CESI) shouldinclude an assessment of measures to reducevulnerability and of the necessary financialprotection available to ensure the project’ssustainability. These evaluations will identifythe degree to which the investments could beaffected by natural disasters, as well as theappropriate preventive measures to reduce therisk to acceptable levels. Figure 6 shows howrisks analysis should be carried out in theBank’s project cycle.

Risk Transfer

The countries of the region are interested inestablishing risk transfer instruments, using asa model the mechanisms applied in developed

PROJECT CYCLE

Programming Identificationof Projects

Design ofProject

Execution of Projects Evaluations

Proposal for Analysis of Threats and Vulnerability

Evaluation ofThreats atCountryLevel

Identificationof SpecificRisks ofProjects

Evaluation ofRisks andMitigationOptions

Execution of RiskManagementMeasures

Evaluation ofRisk Deductionsand Monitoringof Performance

DuringDisasters

Figure 6. Proposed Disaster Risk Management in the IDB Projects Cycle

Source: IDB (2000).

23

countries, especially insurance. Financialprotection against disasters via the develop-ment of an efficient insurance market offersthe opportunity to transfer part of the risk, andavoids indebtedness as a result of an emer-gency. The development of this market wouldinvolve updating the legislation and the regu-latory framework, strengthen independentsupervision to monitor the solvency of insur-ance companies, and eliminate incentives foranticompetitive behavior. To this end, thesupervising entity should enjoy appropriatejurisdictional and punitive powers. The estab-lishment of insurance instruments also re-quires the existence of measures such as certi-fications of compliance with accepted designand structural guidelines for the project thatwill be insured.

Insurance companies would benefit from theadoption of international standards that pro-mote solvency, efficiency and transparency. Itwould result in the generation and timelypublication of detailed and precise financialstatements for public use. The creation of anappropriate technical infrastructure for theinsurance system to operate adequately con-stitutes an important part of the strategy topromote risk transfer. Such an infrastructureis needed, among other things, for carryingout appraisals and risk management. Amongthe staff needed are building inspectors, in-surance adjusters and risk rating agencies.Lack of information can be an importantlimitation for the development of the insur-ance market. Insurance and reinsurance com-panies can generate information. Internationaldevelopment agencies may help in the proc-ess. The IDB (2002), is an example as it willbe implementing the Information and Indica-tors Program for the Risk Management ofNatural Disasters for the benefit of some 10countries in Latin America and the Caribbean.

In addition to the above initiatives, it is im-portant to foster demand for transferring risk

by building public awareness of the benefitsof risk management and insurance. At thesame time, continued research should go intothe use of innovative capital market instru-ments such as catastrophe bonds.

Emergency Response:Available Instruments Should be UsedFirst

In the event of a disaster, existing and lowestcost resources should be used first. If a ca-lamity fund is already in operation (Mexico’sFONDEN, see Box 4), or coverage is avail-able by means of insurance or catastrophebonds, these sources would receive priority.However, since these instruments do not yetplay an important role in most countries, thetrend is to first rely on government resourcesby transferring funds from existing programsand development funds (municipal, social,urban, rural, etc.) to meet critical needs. Asearch for the greatest possible amount ofinternational donations usually follows. An-other instrument that could be highly useful iscontingent credit, although it entails an annualadministrative payment and indebtednessonce the emergency takes place.

Since prevention and mitigation actions can-not fully eliminate the risk of disasters, meas-ures for the financing of losses will be re-quired. Countries tend to give priority to theiruse according to vulnerability and to theavailability of the instruments. Although in-struments such as insurance have not yet be-come popular in the region, governments canadopt a range of measures to promote the de-velopment of insurance markets, includingopening them up to international competition.As a first step they could insure their owncritical assets through national or internationaltenders. Another instrument that could bevery useful? although it incorporates an an-nual administrative payment and indebtednessin case the emergency presents itself? is con-

24

tingent credit. This has the advantage ofavoiding the potential distortions that can becaused by the common practice of realloca-tion of resources from existing programs torehabilitation and reconstruction.

Comprehensive Approach

Until recently, the countries of Latin Americaand the Caribbean have tended to respond toemergencies rather than to prepare for themthrough the establishment of prevention andmitigation measures in order to reduce theirvulnerability to catastrophic events. It is herethat the comprehensive risk management thathas been described in the preceding pages

begins to play a fundamental role. Preventionand mitigation measures, combined with pub-lic and private financial protection, play asignificant role in the process of building sucha comprehensive approach. It is essential foreach country to develop a consistent strategyor plan to manage the risk of disasters, withthe participation of the finance and planningministries, sector ministries and local gov-ernments, the business sector and civil societyin general. It is also necessary to have astructured plan for meeting emergencies oncethey take place. The supply of effective exante financial protection mechanisms is es-sential because it makes funds available whenthey are needed most.

25

BIBLIOGRAPHY

Andersen, T. J. 2002. Economic Exposure to Natural Disasters: Opportunities and Instruments for Finan-cial Risk Management. Working document of the IDB Regional Policy Dialogue. Departament ofSustainable Development. http://www.iadb.org/int/drp/English/Red6/disastermain.htm

Applied Technology Council (ATC). 1985. Earthquake Damage Evaluation Data for California. ATC-13Report. Redwood City, CA and Washington, D.C.

Bayon, R., S. Lovink and W. Veening. 2000. Financing Biodiversity Conservation. Sustainable Devel-opment Department Technical Papers Series (No. ENV-134). Washington, D.C.: IDB.

BBC News Online. 2000. EU Overhauls Aid System. Global Policy Forum. May.http://www.globalpolicy.org/socecon/ffd/euaid2.htm

Bender, S. 1991. Primer on Natural Hazard Management in Integrated Regional Development Planning.Washington, D.C.: Department of Regional Development and Environment Executive Secretariat forEconomic and Social Affairs Organization of American States..

Cardona, O. 2001. Diseño del sistema nacional de prevención, mitigación y respuesta (gestión de riesgo).

Charvériat, C. 2000. Natural Disasters in Latin America and the Caribbean: an Overview of Risk. Inter-American Development Bank. Research Department Working Paper #434. Washington, D.C.

Consultancy report to the Technical Secretariat of the Presidency and to the IDB (1152/OC-DR loan).Santo Domingo, Dominican Republic.

Davis, I. and S. Gupta. 1990. Disaster Mitigation in Asia and the Pacific, prepared by the Asian DisasterPreparedness Center of the Asian Institute of Technology. Bangkok. Thailand.

Doherty, N.A. 2000. Integrated Risk Management: Techniques and Strategies for Managing CorporateRisk . New York, N.Y: McGraw-Hill.

ECLAC, 1999a. América Latina y el Caribe: el impacto de los desastres naturales en el desarrollo, 1972-1999. LC/MEX/L.402. Mexico.

ECLAC, 1999b. Centroamérica: Evaluación de los daños causados por el Huracán Mitch en 1998.LC/MEX/L.375. Mexico.

ECLAC-IDB. 2000. La reducción de la vulnerabilidad frente a los desastres. LC/MEX/L.428. Mexico.International Financial Corporation (IFC) 2000. The Central American Market for Insurance. Working

document. Washington, D.C.

Freeman, P. and R. Mechler. 2001. Public Sector Risk Management in Mexico for Natural DisasterLosses. Document prepared for the Wharton-World Bank Conference. Disaster Management Facility.World Bank. Washington, D.C.

26

Freeman, P., L. Martin, J. Linnerooth-Bayer, K. Warner, A. Lavell, O. Cardona and H. Kunreuther. 2001.National Systems and Institutional Mechanisms for Comprehensive Disaster Risk Management.Phase I. Background study for the Regional Dialog on Disasters. IDB. http://www.iadb.org/int/drp/Red6/Docs/Cook1.pdf

Freeman, P. and H. Kunreuther. 1997. Managing Environmental Risk through Insurance. Dordrecht, TheNetherlands: Kluwer Academic Publisher.

Grossmann, S. and O. Hart. 1983. An analysis of the principal-agent problem. Econométrica: 51 (1) pp.7-45.

Hillier, B. 1997. The Economics of Asymmetric Information. New York, N.Y.: St. Martin’s Press.

IDB. 2002. Information and Indicators Program for Disaster Risk Management. Plan of Operations.Document AT-1259. Sustainable Development Department. Washington, D.C.

IDB. 2001. Sectorial Facility for Disaster Prevention. Document GN-2085-5. Washington, D.C.

IDB. 2000. The Challenge of Natural Disasters in Latin America and the Caribbean: IDB Action Plan.Sustainable Development Department. Washington, D.C.

IDB. 2000. First Report of the Working Group on Financing for the Inter-American Committee on Natu-ral Disaster Reduction of the OAS. Washington, D.C.

IDB. 1999. Emergency Reconstruction Mechanism. Document GN-2038-1. Washington, D.C.

IIASA.2000. Integrating Natural Catastrophe Exposure into Development Planning: Case Study of Natu-ral Catastrophe Risk in Honduras. Vienna.

International Strategy for Disaster Risk Reduction of the U.N. (ISDR). 2002. Disaster Risk and Sustain-able Development: Understanding the Links between Development, Environment and Natural Haz-ards Leading to Disasters. Background document for the World Summit of Sustainable Development.April. Geneva.

ISDR. 1999. Boletín para América Latina y el Caribe No. 14. San José, Costa Rica.

Keipi, K. 2001. Financiamiento para la reducción de desastres naturales: Elementos de estrategia. Pro-ceeding of the International Conference on Financing Sustainable Development. CABEI-CONADES.Tegucigalpa, Honduras.

Kreimer, A., M. Arnold, C. Barham, P. Freeman, R. Gilbert, F. Krimgold, R. Lester, J.D. Pollner and T.Vogt, 1999. Managing Disaster Risk in Mexico: Market Incentives for Mitigation Investment. Wash-ington, D.C.: Disaster Management Facility, World Bank.

Lester, R. 2000. Policy Issues in the Choice of Funding Instruments for Natural Disasters. Washington,D.C.: Disaster Managent Facility, World Bank.

Munich RE. 2000. Insured Losses for Region 1985-1999. htpp://www.munichre.com

Nikkei 1999. Quake Bonds Issued for Disneyland. htpp://www.eqecat.com/nikkei.htm. June 2.

27

OFDA-CRED. 1999. EM-DAT: International data base. Université Catholique de Louvain, Brussels,Belgium. htpp://www.md.ucl.ac.be/cred.

Pollner, J. 1999. Proposed Strategy for a World Bank Role in a Small States Regional Catastrophe Insur-ance Program. Washington, D.C.: Disaster Management Facility, World Bank.

Rothschild, M. and J.E. Stiglitz. 1976. Equilibrium in competitive insurance markets: An essay on theeconomics of imperfect information. Quarterly Journal of Economics: 90 (4), pp. 629-650.

Skipper, Harold D. Jr. 1997. Foreign Insurers in Emerging Markets: Issues and Concerns. Washington,D.C.: International Insurance Foundation.

Smith, K. 1996. Environmental Hazards: Assessing Risk and Reducing Disaster. Rothledge Physical En-vironment Series. University of Cambridge. CITY.

Vatsa K. and F. Krimgold, 1999. Financing Disaster Mitigation for the Poor. Washington, D.C.:Documento of the Disaster Management Facility, World Bank.

Vermeiren, J. and J. Pollner. 1994. A Probable Maximum Loss Study of Critical Infrastructure in ThreeCaribbean Island States. World Bank-OAS Report.http://www.oas.org/en/cdmp/document/pml/pml.htm

World Bank. 2001. Honduras. Catastrophe Risk Exposure of Public Assets: An Analysis of FinancialInstruments for Smoothing Fiscal Volatility. Working document. Financial Sector Cluster LCSFP.Washington, D.C.

World Bank. 2000a. The Insurance Market and the Case of the Caribbean Region. Working document.Finance, Private Sector and Infrastructure Department. Caribbean Country Department. Latin Amer-ica and the Caribbean Region. Washington, D.C.

World Bank. 2000b. World Development Report 2000/2001. Washington, D.C.: Oxford University Pressand the World Bank.

World Bank, 1999. Managing Disaster Risk in Mexico: Market Incentives for Mitigation. WorkingDocument. Disaster Management Facility. Washington, D.C.

28

ANNEX

Key Risk Management Factors

Before the Emergency: Risk Prevention and Reduction After the Emergency: Recovery

Identification ofRisks

Mitigation andPrevention

Risk TransferEmergencyPreparedness

EmergencyResponse

Rehabilitationand Reconstruc-tion

Evaluation of natu-ral threats (fre-quency, magnitudeand location)

Physical andstructural miti-gation works

Insurance andreinsurance ofpublic andprivate assets

Early alert andcommunicationssystems

Humanitarianassistance

Rehabilitation andreconstruction ofdamaged criticalinfrastructure

Evaluation of vul-nerability (exposedpopulation andassets)

Land use, plan-ning and con-struction codes

Financial ma r-ket instruments(catastrophebonds, etc.)

Shelters andevacuation plans

Cleanup, tem-porary repairsand re-establishmentof services

Macroeconomicand budgetarymanagement(stabilization,protection ofsocial expendi-ture)

Evaluation of risks(hazardand vulner-ability)

Economic in-centives to pro-mote the adop-tion of mitigationmeasures

Development ofnew instru-ments: hedgefunds indexedaccording toweather condi-tions

Plans for unfore-seen events(public servicecompanies);network of in-stitutions thatrespond in emer-gency situations(local and na-tional)

Evaluation ofdamages

Revitalization ofthe affected sec-tors (exports,tourism, agricul-ture, etc.)

Monitoring ofnatural threats andpreparation offorecasts (SIG,cartography andformulation ofhypothetical situa-tions)

Education,training andawareness rais-ing on risks andprevention

Privatization ofpublic serviceswith regulationof safety issues(power, water,transport, etc.)

Calamity funds(national orlocal); contin-gent credit

Mobilization ofresources forrecovery (gov-ernmental,multilateral,insurance)

Incorporation ofdisaster mitigationcomponents inreconstructionactivities

Creation and strengthening of national systems for preventing and responding to disasters: These systems form acomprehensive and intersectorial network of institutions that broaches all the above phases of risk reduction and recov-ery after a disaster. Backing is required in the areas of regulation and planning, including the reform of legal and nor-mative frameworks, coordination mechanisms, strengthening of participating institutions, national action plans, pre-vention policies and institutional development.

Source: IDB 2000