The World Bank and Natural Catastrophe Funding

Transcription

1 The World Bank and Natural Catastrophe Funding The Changing Risk Landscape: Implications for Insurance Risk Management Rodney Lester

2 The Changing Risk Landscape: Implications for Insurance Risk Management Edited by Neil R. Britton. Proceedings of a Conference sponsored by on Group Australia Limited THE WORLD BANK AND NATURAL CATASTROPHE FUNDING Rodney Lester ABSTRACT The World Bank (the Bank) is one of the major financiers of post-disaster reconstruction with 101 loans having been approved in the last 20 years for a sum in excess of $US7.4 billion. The Bank is also a major lender for practices of risk mitigation. However, this is heavily concentrated in a few large developing countries. While natural disaster risk management is not high on the Bank s explicit agenda it is implicit in two of the Bank s main development themes; namely private sector development and poverty reduction. This loose coupling is primarily due to the decentralised operating structure of the Bank. Natural disasters are also not normally given any profile in country assistance and loan preparation guidelines, although this is changing. The key to putting this issue on the Bank s agenda is through raising awareness within client countries. The Bank s Disaster Management Facility together with its Financial Vice Presidency working with the private sector have the greatest chance of achieving this. The payoff for the Bank will be a greater capacity to assist with client countries risk management. For the private sector the payoff will be the opening of new viable markets and fee income from bank projects. Traditional insurance infrastructures are currently being built, however, in the interim hedge instruments may serve both their natural function (transferring undefined and systemic risk) and as a substitute for reinsurance, possibly in conjunction with government sponsored funds or pools. INTRODUCTION The World Bank (the Bank) is a United Nations agency established at the end of World War II to aid in reconstruction, and foster stability and economic growth. Its Articles of Agreement are remarkably short and concise. Article One describes the purposes of the Bank. The first two of these are most relevant in the context of disasters: I. To assist in the reconstruction and development of territories of members by facilitating the investment of capital for productive purposes, including the restoration of economies destroyed or disrupted by war, the reconversion of productive facilities to peacetime needs and the encouragement of the development of productive facilities and resources in less developed countries. II. To promote private foreign investment by means of guarantees or participation s in loans and other investments made by private investors; and when private capital is not available on reasonable terms, to 2

3 supplement private investment by providing, on suitable conditions, finance for productive purposes out of its own capital, funds raised by it and its other sources (The World Bank, 1995). The Bank s shareholders are most of the countries in the world and its clients (category two countries) are societies that are still developing or in transition to a modern mixed economy. In other words, those countries that have yet to graduate to category one status. Aside from the Bank s mission, perhaps the most important point to note is that the World Bank is a fully functioning bank. It generates profits out of the margins between its cost of funds and its lending rates, and service fees. Concessional long-term loans to very poor countries are handled through a separate entity called the International Development Association (IDA). However, most lending is through the International Bank for Reconstruction and Development (IBRD) at a small margin (typically 0.5%) over the London inter-bank offer rate (LIBOR) for the currency borrowed. As at 30 June 1998, the Bank had net assets of $US26.5 billion, loans outstanding of $US103.3 billion, and uncalled capital of $US175.1 billion. Net profit in totalled $US1.2 billion. It has an AAA credit rating because it is conservatively geared and has substantial uncalled capital; the Bank can call on the wealthier countries (category one countries), if necessary, to replenish its financial resources. The World Bank Annual Report shows that IBRD lending has averaged $US15 billion in the last decade (The World Bank, 1998). Over the last few years this has increased by approximately $US5 billion per annum in response to the Asian financial crisis. IDA adds at least another $US6 billion each year. Investment lending is geared through other donors and the private sector. Total World Bank mobilised project investment is in the order of $US50 billion per annum. Some Bank loans, known as structural adjustment loans (SALs), are purely for purposes of macro-economic adjustment. However, these normally carry heavy conditionalities and relate to economic management. The IBRD/IDA make loans to countries based on the following per capita income guidelines (Delphos, 1997): $US765 or less for granting civil works preference to eligible domestic contractors in evaluating civil work bids under international competitive bidding. $US1465 as the historical ceiling for IDA eligibility and $US925 as the operational cutoff for IDA eligibility. $US1465 or less for 20 year IBRD terms; $US for 17 year IBRD terms; and $US3035 or more for 15 year IBRD terms. $US5295 or more for initiating the IBRD graduation process. 3

4 As part of the lending process the Bank usually provides technical assistance to its clients prior to the approval of a loan and will continue to be heavily involved in a supervisory and advisory capacity until the project is complete. In addition, as a general rule the Bank will not make major financial commitments to a country unless it has satisfied macro-economic settings that are within International Monetary Fund (IMF) guidelines (the World Bank s sister organisation, also established at the end of World War II). If a country is an eligible borrower the Bank s main role is to help build infrastructure and institutions that will lead to economic and social development. Since the Bank s establishment, the product suite available has grown substantially and there is an ongoing program for developing new products in support of the Bank s core objectives. The Bank now owns an insurance company, the Multi-lateral Investment Guarantee Agency (MIGA), which insures political risk. It also provides various forms of partial guarantee to assist its client countries borrowing programs. A bouquet of small loan products has recently been created to assist countries in developing necessary capacities, sometimes over extended time frames. Another important arm of the Bank is the International Finance Corporation (IFC), which invests equity and debt in various businesses within client countries to support the development of private sector infrastructure. Currently, IFC has interests in a substantial number of life and general insurers in emerging countries and is growing its portfolio in this sector. One fundamental difference to date between IFC and IBRD/IDA has been that IFC requires full risk management of the enterprises in which it invests and also maintains a full insurance department. This is because, unlike IBRD/IDA, it does not have a sovereign guarantee from its client. 4

5 The Banks Role in Disaster Management To date, the Bank has dealt with non-financial catastrophe risk by largely ignoring it ex-ante, except for some mitigation activity. Despite this, the Bank is effectively one of the major catastrophe reinsurers and risk managers in the world. Over the last 20 years, it has lent developing countries at least $US14 billion in disaster-related funding, $US9 billion of which was approved in the last decade. The relative experience of developed, compared to developing, countries with regard to catastrophes makes it clear why this has been the case. Table 1 shows the comparative losses between developed and developing countries from Table 1. Comparative Losses Between Developing and Developed Countries Country Type Developing: Number of disasters Number of lives lost Insured loss Developed: Number of disasters Number of lives lost Insured loss 40 Worst Disasters in Terms of Lives Lost $US2 billion * $US2.6 billion 40 Worst Disasters in Terms of Insured Losses $US8 billion** * $US99.6 billion Note: * The Kobe earthquake cost 6000 lives. ** Includes some USA losses from Hurricane George. Source: Swiss Re, If, as seems possible, the world is emerging from a relatively stable geological and climatic phase into a more dangerous environment the consequence is likely to be an ongoing series of catastrophic disasters; such as the floods in China during 1996 that caused 2700 deaths, destroyed over five million buildings, and had economic costs of $US24 billion; or Hurricane Mitch (1998) that cost upwards of lives, left nearly three million people homeless, and destroyed much of the infrastructure in Nicaragua and Honduras. Regardless of intensity and frequency, man-made and natural disasters, when combined with a frighteningly rapid growth in both urban and rural exposures in vulnerable and emerging economies, are inevitably beginning to strain the capacity of governments, development banks, and aid organisations to finance post-disaster recovery. For example, the population of Istanbul has 5

6 doubled to approximately 14 million in the last decade, with most new housing built to standards well below those appropriate for one of the world s known major transverse fault lines. Estimates of a category 9 earthquake PML vary between $US25 billion and $US50 billion. Similar, but less extreme examples can be found in many client countries of the Bank. One long-standing area of extreme vulnerability is the Caribbean, where the economic impact of hurricanes has been frequent and considerable, as shown in Table 2. Table 2. Economic Impact of Severe Storms in the Caribbean. Country Hurricane Year GDP Growth Previous Year (%) GDP Growth Following Year (%) Jamaica Gilbert St. Lucia Debbie Dominica Luis/Maralyn Note: Whilst many factors affect GDP growth there appears to be a clear correlation with natural disasters in these cases. A more recent trend has been a significant drop in the provision of emergency relief funds (see Figure 1) from wealthier countries because of fiscal belt tightening, a strengthening USA dollar, and concentration on a few countries by the USA. The Disaster Management Facility Recognising these issues, the Bank recently established a special facility called the Disaster Management Facility (DMF), under the leadership of Dr. Alcira Kreimer, to develop methodologies to address what appears to be a looming crisis. Its small staff act as the centre of a much larger virtual organisation, including specialists throughout the Bank, the public and private sectors. Aon was the first private sector partner of the DMF; currently most major players in the catastrophe risk market are now working with the Bank, occasionally on specific projects. 6

7 Figure 1. Provision of Emergency/Distress Relief Funds $US m illion Years Source: Development Assistance Committee, The key areas in which the DMF aims to add value are as follows: Raising awareness within the Bank of potential economic and social losses arising from natural disasters: Eventually bringing about desirable changes in the Bank s operating policies, as has already happened on issues such as the environment, tobacco, and gender. The inclusion of a catastrophe risk assessment in country assistance strategies and loan appraisals. Fostering the development of social mitigation, which relates to modifying or preventing the creation of catastrophe exposures by people; and physical mitigation, relating to the physical engineering of risk reduction for existing exposures. Working with the private sector to develop funding and risk transfer methodologies, and vehicles for those developing countries that have a major social and economic vulnerability to natural disasters. 7

8 The Financial Sector, which the insurance industry forms part of, is more specifically involved in developing products to fill in where insurance and capital market failure has occurred. This paper is largely concerned with the risk transfer and funding aspects of these objectives. However, for a public/private partnership to be effective it is important the private sector first understands the Bank, its role to date, and possible future roles it may play. THE WORLD BANK S POST-DISASTER ROLE TO DATE The Bank s Operational Manual covers emergency recovery assistance under operational policy (OP) 8.5. This states that: A country may request assistance from the Bank when: It is struck by an emergency. This emergency seriously dislocates its economy. The emergency calls for a quick response from the government and the bank (The World Bank, 1995). An emergency is defined under the manual as an extraordinary event of limited duration, such as war, civil disturbance, or natural disaster (World Bank, 1995). The policy affirms that the Bank finances investment and productive assets rather than relief and consumption. Relief is defined in the manual to include search and rescue, evacuation, food and water distribution, temporary sanitation and health care, temporary shelter, and restoration of access to transport (The World Bank, 1995). Thus, the Bank is far more aligned with the insurance industry in its natural disaster-related activities than with non-government relief organisations (NGOs) or emergency management agencies. However, the policy does specify that: Collaboration with the United Nations Development Programme and other international agencies, local NGOs, and donors is often helpful in devising the recovery assistance strategy under an emergency recovery loan (ERL) and in designing specific prevention and mitigation programs (The World Bank, 1995). The insurance industry is not mentioned, largely because of its limited role, to date, in client countries of the Bank. The Bank s financial support can be in four different forms: Immediate support in assessing the emergency s impact and developing a recovery strategy. Restructuring of the Bank s existing portfolio for the country to support recovery activities. 8

9 Redesigning of projects not yet approved to include recovery activities. Provision of an emergency recovery loan. Emergency recovery loans (ERLs) are the Bank s product that comes closest to a substitute for insurance. The loans are designed to rebuild physical assets and restore economic and social activities after emergencies (The World Bank, 1995). ERLs are normally funded from within a country s general lending allocation and for some disaster-prone countries this can limit the scope for long-term development funding. The criteria used by the Bank to decide whether to provide an ERL includes: Impact on economic priorities and investment programs. Frequency of impacts: For recurring events, such as annual flooding, a regular investment loan is more appropriate. Urgency of impact consequences: For a slow onset disaster, such as drought, the more thorough preparation of a regular investment project may be preferable. Sustainable hazard management: Prospects for reducing hazards from similar natural disasters in the future. Economic resilience: Expected economic benefits. ERL projects are required to use disaster resilient reconstruction standards, including emergency preparedness and technical assistance on prevention and mitigation measures to strengthen the country s resilience to natural hazards or lessen their impact (The World Bank, 1995). The list of possible deliverables under a mitigation and prevention project does not include developing risk transfer mechanisms, although neither is precluded. One of the two largest ERLs (Turkey s Flood and Earthquake Emergency Recovery Project) currently has a study of risk transfer mechanisms underway as a conditionality of the loan (refer to Appendix for further details). Dr Kreimer (an architect) and a colleague, Dr Roy Gilbert (an economist), have recently researched the Bank s commitments with regard to natural disasters. Their work shows that since 1980 the Bank has made 101 identifiable loans for reconstruction, amounting to $US7.4 billion; of which $US5.1 billion was for ERLs, spread over 69 projects (Gilbert and Kreimer, 1999). In addition to the amounts shown above, disaster-related loans have been created by diverting funds approved for other purposes. It is not possible to quantify this figure, although anecdotal evidence suggests that at least 30% of infrastructure loans have been diverted in disaster-prone countries. A recent quantifiable example followed Hurricane Mitch in 1998, when the Bank announced that it had reassigned $US200 million to finance part of the reconstruction in Honduras, Nicaragua, Guatemala, and El Salvadore. The 9

10 regional distribution of post-natural disaster projects since 1980 is shown in Table 3. Table 3. Distribution of World Bank Post-Natural Disaster Projects Since Region Number of Projects Percentage of Projects Still Active Africa East Asia East Europe/ Central Asia 9 44 Latin America Middle East/ North Africa 9 22 South Asia Total Note: The comparatively large proportion of active projects in East Asia reflects the relatively recent involvement of the Bank in disaster-related lending in this region. Source: Gilbert and Kreimer, Although reconstruction (post-natural disaster) projects are widely spread there are a number of countries that have suffered recurrent disasters and have called on the Bank to provide financial support several times. Yemen, Bangladesh, Chile, and India fall into this category with each having had five loans approved in the last two decades. Countries with three projects approved include China, Brazil, and Argentina. In terms of the main hazards, two leading causes dominate the Bank s portfolio; flood accounts for 35% of loans and earthquake 27%. Extreme weather, both drought and severe storm, account for 36%. Not surprisingly, these are the same major hazards the insurance industry also finds itself responding to. The majority of disaster-related Bank lending, as might be expected, is largely orientated with rebuilding, and repair of buildings and infrastructure (see Table 4 for a complete list of projects). Apart from housing (23 projects since 1980), the Bank is often involved in the re-establishment of a working health sector (for example, hospitals), an important component of post-disaster relief. 10

11 School reconstruction is often dealt with through bilateral assistance and NGO support. Table 4. Distribution of World Bank Disaster-Related Projects. Nature of Project Percentage of Projects Reconstruction 53 Rehabilitation 16 Flood Protection 11 Economic Recovery 10 Institutional Development 3 Other 7 Note: Institutional development refers to the building up of a capacity to respond to future emergencies. Source: Gilbert and Kreimer, Despite this long established and wide-ranging involvement in post-disasterrelated projects, the Bank s activity has not kept pace with the frequency and severity of events. The number of projects approved in the two decades since 1980 are almost identical and there are indications the Bank s proportional involvement in post-disaster funding is scaling down. Given the pressures on finances of other development organisations this must be an ominous sign for highly vulnerable developing countries. One particular group of vulnerable countries has recognised this by prevailing upon the British Commonwealth Secretariat to sponsor (with the Bank) a small states initiative. Under this scheme, vulnerabilities to economic dislocation (this would include natural disasters) would be quantified and allowed for, in addition to GDP per capita, in determining whether a country is eligible to borrow from the Bank. THE WORLD BANK S PRE-DISASTER ROLE The Bank s involvement in mitigation is considerable, although it shows somewhat different characteristics to its reconstruction portfolio. Gilbert and Kreimer (1999) found 97 projects amounting to $US6.5 billion have been approved since 1980; compared with 101 reconstruction projects amounting 11

12 to $US7.4 billion. The two largest pre-disaster projects were for flood (China) and drought (India), and together account for approximately $US700 million. Activities carried out in the mitigation portfolio are concentrated in a few countries: Brazil, China, Bangladesh, and India account for 40% of them. The largest borrower for natural disaster mitigation is Brazil with 15 projects, followed by China with 10, and Bangladesh, eight projects. Although these countries loom large in the mitigation client list it is of concern that many other countries with post-disaster projects (reconstruction) are not found in the mitigation portfolio. On a positive note, however, there is a clear growth trend for projects with a mitigation component; 55 projects were approved in the current decade (20 of these since 1998 alone) versus 40 projects in the 1980s. In terms of hazard agents for mitigation projects, the break up reflects the geographical location of the countries concerned and political reality (that is, frequency and visibility of losses); with flood and forest fire each accounting for 32% of projects and drought 29%. Earthquake accounted for 3% despite its leading role in losses actually incurred (23% of post-disaster projects). This accords with the author s own experience of the political difficulties inherent in dealing with the property sector in earthquake-prone countries. The main components of the Bank s mitigation projects are outlined in Table 5. Despite these activities, natural disaster management is not yet a central theme within the Bank. In part, this is a result of the Bank s operating structure. 12

13 Table 5. Distribution of Mitigation Projects. Component Percentage of Projects Forest Fire Prevention 31 Flood Protection 28 Institutional Development 11 Water Resource Management 8 Adaptation of Production 7 Other 15 Note: Adaptation of production refers to encouraging farmers to plant less disaster-prone crops. Institutional development includes disaster awareness programs for highly vulnerable populaces. Source: Gilbert and Kreimer, THE RELATIONSHIP BETWEEN THE BANK S OPERATING STRUCTURE AND ITS DISASTER MANAGEMENT ROLE The Bank has been the subject of severe criticism over the last two decades. Hence, it has made strenuous efforts to make itself both more effective and more relevant in recent years. This is clearly demonstrated by a strategic compact entered into between the Bank and its country shareholders. This compact covers a range of matters (refer to for further details), however, one of the most far reaching has been a decentralisation of power and a responsibility shift to the country level. In this move, key officers have become the country director and the country economist and, working with their counterparts in government and civil society (non-government community organisations), they set a program for the Bank s activity in the country concerned. Through this change, the successful introduction of a natural disasters mitigation and funding agenda to the Bank will require raising awareness at a country level. Fortunately there is already a good understanding of these issues at senior levels in the Bank. One tangible indication of this support has been the provision of seed money (awarded through a competitive bidding process) by the Bank s president and certain sector boards for the development of literature, case studies, models, and private/public sector partnerships in this area. 13

14 The DMF and the Financial Sector Vice Presidency carry out the bulk of this development work. The major objectives are to generate a number of disaster-related projects in key countries and to do research that will, in turn, lead to financial products (loans, credits, guarantees) being sought by the countries concerned. Through this mechanism it is expected that the momentum to achieve the objectives (listed earlier) will require the Bank s operational manual to be amended to necessitate a review of natural disaster risk, in both the country assistance strategy and before a loan is approved. Through this scheme, projects are currently under way, or being prepared, in Mexico, Turkey, Central America, and the Caribbean. There have also been expressions of interest from Eastern Europe and North Africa. Another practical issue has been a shift in career progression criteria. Personal progress in the Bank has historically been associated with an ability to generate loans and effect rapid disbursement, thereby directly increasing the profitability and durability of the Bank. In recent years, however, more emphasis has been placed on the quality of the Bank s lending and its value to the people of the country concerned. Despite this, the pressure to continue generating loans remains and any new natural disaster-related products would eventually need to find legitimate roles in the Bank s financial (as opposed to technical advice) product suite. There are various possibilities under the existing regime. Included among these are hazard mitigation foci becoming a required component of structural adjustment loans for countries whose economies are highly vulnerable to natural catastrophes; and special purpose investment loans dedicated to major mitigation efforts, such as retrofitting. Smaller technically orientated loans are also possible. However, these are often seen as having a low upside potential by both the Bank s operations staff and senior country officials. New products currently being considered by the Bank include guarantees to enhance funding mechanisms, such as catastrophe bonds, reinsurance swaps, and rapid release liquidity facilities. Other areas the Bank is working on are the design of weather indexes, supported by robust measuring technology on the ground; and ways of delivering claims without increasing moral hazard and slippage at the payment point. Given this, the private sector has a major role to play, both in helping to raise awareness within the Bank and client countries, and in subsequently providing the core products required to deliver risk management capabilities to client countries. The first of these activities can be seen as an investment towards achieving the second. 14

15 Partnership and Raising Awareness Recently, the Bank has had a number of major policy agendas in pursuit of its development agenda. Miller-Adams (1999) has listed these as: The environment. Poverty reduction. Developing in the private sector. Participation by civil society in Bank programs. The role of women. Govenance (that is, property rights, the legal system, and confidence in the political system). Of these the most energetically and consistently pursued has most likely been the private sector agenda. In part, this could be because it accords with a perceived belief (the so-called Washington consensus), on the part of the Bank, that private sector solutions are to be preferred where possible. During recent years, the poverty reduction objective has also moved to centre stage under the Bank s current President; the DMF s objectives are also consistent with and supportive of both these policy priorities. Miller-Adams (1999) points out that the private sector agenda has three sub-agendas: Improving the business environment. Restructuring or privatising public enterprises. Developing the financial sector. The last of these, in particular, would appear to be consistent with the development of risk transfer capabilities in client countries. Unfortunately, the training and background of a number of country directors and country economists does not lead them to automatically consider the private sector as potentially important in the development of the Bank s risk management agenda. In fact, it is only during the last decade that the importance of the banking system itself has been fully acknowledged by the Bank, which has traditionally had a strong bias to the real economy and basic infrastructure. The key to gaining country directors and economists attention is through good data and a well thought through case; supported as is inevitable by an ongoing stream of natural disasters. 15

16 In this context, the key support that the private sector can provide to the DMF is information. Possible strategies to gain the attention of country directors include the preparation of a short natural disaster profile of their country and an assessment of social and economic vulnerability. A major private sector player has also suggested that a loss potential index could be developed for each country that could eventually become the basis for a global hedging market in catastrophe exposures. It is the author s opinion that this recommendation has merit, especially as it links easily with aspects the Bank already has under consideration, and that are discussed later in this paper. For these strategies to be effective some of the traditional intense rivalries that characterise the insurance, the reinsurance, and reinsurance broking sectors will need to be ameliorated with respect to the World Bank partnership. The Bank does, after all, provide a legal and competitive neutral ground where these groups can consider the possibility of a non-zero sum game. In this case, a partnership with the Bank would lead to an opening up of the world market to the insurance sector. The formal mechanism through which this partnership is to progress by is a consultative group coordinated by DMF (see Table 6). Table 6. Consultative Group This is a global, tri-sector partnership (government, private sector and civil society) aimed at reducing disaster risk in developing countries and making disaster prevention and mitigation an integral part of development efforts. To achieve its goals, the following activities are focused on identified areas of priority: Strengthen donor coordination and promote policy aimed at reducing disaster risk. Promote a culture of safety through education and training, and dissemination of good practices for reducing vulnerability to natural and technological disasters. Promote linkages, both between the public and private sectors and between the scientific community and policy makers for disaster risk reduction. Support pilot projects that may help to demonstrate risk reduction or risk transfer strategies. 16

17 This group will meet periodically to review progress and discuss possible initiatives. The first such meeting was recently held in Paris and represented a unique coming together of development organisations, NGO s, and the private sector. Under this consultative model the various partners each have important roles to play in the developing world. CATASTROPHE RISK MANAGEMENT FOR DEVELOPING COUNTRIES The overriding issue for the majority of the Bank s client countries is that the reinsurance sector has no foundation upon which to build. Typically, less than 5% of properties in client countries are insured. Even government-owned infrastructure is subjected at best to the most rudimentary risk management program. For example, a recent Bank study of catastrophe risk in Mexico found that only homes, or 1.8% of the market, were actually insured in 1998, according to industry sources. More than 8 million homes in Mexico are thus insurable but uninsured (Kreimer and others, 1999). These figures are well supported by the well-known Swiss Re penetration numbers, illustrated in Table 7. Table 7. Non-Life Insurance Penetration Region Premiums as GDP % Premiums Per Capita $US North America Latin America Western Europe Eastern Europe Japan South East Asia Middle East Africa Oceania Source: Swiss Re,

18 The sectors normally engaging in sophisticated risk avoidance, reduction, and transfer in developing countries are industrial and major commercial segments of the private sector. These are usually well served by the major international insurance intermediaries and reinsurers. Thus, only two groups usually exist to absorb most losses in developing countries. Those suffering the loss; and the tax payer through government natural disaster funding arrangements (this could include borrowings from the Bank). If the timing is right and the country meets the criteria for strategic importance it may also receive grant moneys from bilateral donors. However, this is a diminishing resource as the world, in general, becomes increasingly more disaster weary from donor fatigue and foreign aid becomes the first victim of surplus-based fiscal policies. As Priest (1994) states, this last point is an unfortunate development for both technical and bureaucratic reasons. At most, governments should only be involved in residual uninsurable risks since they are less efficient than the insurance sector at dealing with risk transfer and reduction. Recent work by Chichilnisky and Heal (1998), and Croson and Kunreuther (1999) has further demonstrated that even catastrophe risks which are not amenable to the insurance process (for example, because there is no stable risk function), can be efficiently handled through hedging arrangements in financial markets using derivative instruments such as catastrophe bonds. A number of insurers in developed countries have encouraged the development of such markets, presumably as possible hedging against rising reinsurance prices. The reasons given for the non-existence of an insurance base in most developing countries vary from lack of education and knowledge to religious conviction. Although the Bank is only beginning to research this topic, the author s own observations in a number of countries, combined with empirical data on consumption patterns by income group in developed countries, points to low income levels as the main determinant of insurance purchase. Moral hazard (that is, an assumed government bailout) is also an important factor. In this situation, and barring compulsory schemes, the traditional voluntary insurance mechanism is unlikely to be effective for the bulk of those exposed to catastrophic loss in the foreseeable future. The Bank regards itself as having a legitimate role in finding solutions to this situation because it sees this as being untenable in the long-run for highly vulnerable countries. Thus, it is developing a generic model that commences with compulsory mitigation as an initiating process for a wider country risk management framework. This framework is based on mitigation through insurance coverage and price signalling not being generally available. This is combined with a yet-to-be-developed national disaster and/or reinsurance fund that, in turn, lays off risk and arranges liquidity facilities in developed countries capital markets. At the same time, it will encourage processes to promote the development of direct domestic insurance penetration. Within this framework, government disaster fund mechanisms should ensure those who are capable of buying insurance do so. 18

19 In these circumstances, hedging instruments based on objective criteria about a catastrophic event (such as intensity levels of an earthquake for a given locality) may be useful even where the insurance process would otherwise be more efficient if available. The main disadvantages of hedging instruments are that payout functions may not be closely correlated with actual losses incurred. In addition, capital markets tend to be more expensive than insurance markets. Nevertheless, they may provide the best options available in an incomplete market. The role of the insurance sector working with the Bank is to help with the development of domestic property insurance and to provide alternative risk transfer (ART) mechanisms in the interim. This opportunity arises primarily as a result of the Bank s ability to raise awareness in client countries through technical support and other sector work being undertaken. The Bank also has the ability to reinforce positive developments through conditionalities attached to loans. Studies underlying this work often provide opportunities for the private sector to participate on a fee income basis. The fact that major financial institutions are expanding their own network, by having more offices in countries rather than centralising operations, will enable the fostering of local awareness toward insurance (Roberts, 1999). The Bank is not, however, in a position to endorse any one supplier. This is usually the responsibility of the client country working within the framework of the Bank s rigorous procurement guidelines. The main advantage the Bank s partners receive, in addition to any income obtained, is an understanding of the issues and opportunities. CONCLUSION While the World Bank is a major player in both pre-disaster mitigation and post-disaster funding, natural catastrophes are not yet a formal part of the risk assessment and management process for client countries. However, many client countries are extremely vulnerable to economic dislocation due to their current natural hazard vulnerability status. This situation is likely to get worse with increasing exposures, the increasing frequency and severity of events, and a limited pool of funds available to deal with both natural and man-made disasters. In this situation, given the probable technical superiority of insurance and hedging instruments to government funding, the private sector has a major role to play in the risk management process for developing countries. This opportunity is more likely to be realised if the World Bank and major private sector players in the catastrophe risk transfer market work in partnership. ACKNOWLEDGEMENT Dr. Alcira Kreimer who heads the Bank s Disaster Management Facility kindly provided most of the information relating to the Bank s past involvement in 19

20 catastrophe management. However, the opinions expressed in this paper are those of the author and do not necessarily represent the views of the World Bank or any of its associated entities. REFERENCES Chichilnisky, G. and Heal, G Managing Unknown Risks: The Future of Global Reinsurance. The Journal of Portfolio Management: Institutional Investment. Summer: ii. Croson, D. and Kunreuther, H Customising Reinsurance and Cat Bonds for Natural Hazard Risks. Conference on Global Change and Catastrophic Risk Management IIASA. Luxembourg, Austria. June 6-9: ii. Delphos, W.A Inside The World Bank Group: The Practical Guide for International Business Executives. Venture Publishing. Washington D.C: 65. Development Assistance Committee Gross Disbursements : Table One. Organisation for Economic Coordination and Development. Gilbert, R. and Kreimer, A Learning from the World Bank s Experience of Natural Disaster-Related Assistance. World Bank Research Paper: DMF. Washington D.C. Kreimer, A., Arnold, M., Barham, C., Freeman, P., Gilbert, R., Krimgold, F., Lester, R., Pollner, J., and Vogt, T Managing Disaster Risk in Mexico. World Bank Research Paper: DMF. Washington D.C. Miller-Adams, M The World Bank: New Agendas in a Changing World. Routledge. London: 1-5, 22. Priest, G The Government, the Market, and the Problem of Catastrophic Loss. Conference on Social Treatment of Catastrophic Risk. Stanford University. October. Roberts, S Spotlight Report: Aon Corp. Business Insurance. July 19: 23. Swiss Re World Insurance in 1996: Modest growth in the Insurance Industry. sigma. Swiss Re. Zurich. 4. Swiss Re Natural Catastrophes and Man-Made Disasters 1998: Storms, Hail, and Ice Cause Billion-Dollar Losses. sigma. Swiss Re. Zurich. 1. The World Bank Emergency Recovery Assistance. World Bank s Operating Manual. August. The World Bank Trends in IBRD and IDA Lending. World Bank Annual Report. Washington D.C: 8. 20

21 APPENDIX This paper was written before the recent Izmit earthquake in Turkey. Prior to the earthquake, the Terms of Reference (TOR) for a study of natural disaster funding mechanisms were due to be released by the Turkish Government to a short list of consulting firms. The following is a generic form of TOR headings that will provide some idea of the direction the Bank is likely to take in client countries. In practice a TOR is much more detailed and specific to the country concerned. The study involves: Estimating direct loss potential arising from natural disasters and developing catastrophe cumulative loss stochastic models. Determining the current allocation of catastrophe risk between households, private firms, insurers, and the government and modelling the resultant burden of catastrophic events on the central budget under various impact scenarios. Based on this analysis the fiscal benefits (if any) that would emerge if catastrophe risk could be transferred from current public expenditure are to be quantified. Examining the capacity of the domestic and international insurance and reinsurance industries to assume natural disasters risk and developing alternative insurance schemes including funded national disaster schemes. 21

22 Examining the feasibility of alternative risk transfer mechanisms to supplement insurance/reinsurance capacity, including capital market liquidity support and risk transfer instruments. Examining mitigation measures to be taken to reduce direct loss potential and to establish a viable catastrophe insurance scheme. Reviewing the existing legal framework for payment of disaster losses, identifying any undesirable economic incentives emerging from this framework and recommending and drafting desirable changes in relevant regulations. 22

23 The views expressed in this document are those of the author, and not necessarily of Aon Re Australia Limited. The papers are intended to provide general advice in summary form. The contents do not constitute advice and should not be relied upon as such. Formal advice should be sought in particular matters. Aon does not provide warranties, guarantees or undertakings in relation to the accuracy, completeness or currency of the information provided and does not invite reliance, or accept responsibility for the information being provided.