Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
7 Insuring against sea level rise
Insurability
of losses
Oil
pollution
Nuclear
damage
Implications
The insurance scheme proposed by AOSIS
The
Climate Change Convention
Notes
and references
Appendix: Scheme proposed by AOSIS for inclusion
in the Climate Change Convention
Michael Wilford
The Reports of the Intergovernmental Panel on Climate Change (IPCC) recognize that the human settlements most vulnerable to climate change are those which are especially exposed to natural hazards, that is, to coastal or river flooding, severe drought, landslides, severe wind storms and tropical cyclones. The IPCC also recognized that the most vulnerable populations are in developing countries; and that in small island countries as well as in coastal lowlands, inundation due to sea level rise and storm surges are a particular hazard (see box on page 170).
The IPCC Working Group II (Impacts) used various scenarios based on a number of different scientific studies. The main scenario upon which Working Group II based its assessments was:
1 an effective doubling of CO2 in the atmosphere between now and 20252050;
2 a consequent increase of global mean temperature in the range of 1.5°C to 4-5°C;
3 a sea level rise of about 30-50 cm by 2050 and of about 1 metre by 2100, with a rise in the temperature of the ocean surface layer of between 0.2°C and 2.5°C.
This scenario pre-dated, but is in line with, the assessments of IPCC Working Group I.
In assessing the effects of climate change on the oceans and coastal zones the IPCC reports point out that global warming would not only accelerate sea level rise, but would also modify ocean circulation and change marine ecosystems, with considerable socioeconomic consequences. These effects will add to the present trends of rising sea level and damage to coastal resources from pollution and over-harvesting. A 30 cm sea level rise will threaten low islands and coastal zones. A one metre rise would render some island countries uninhabitable. The Coastal Zone Management Subgroup also emphasized that:
Vulnerability of the Maldive Islands to sea level rise
The current environmental problems of the Maldives are in large part the result of the high density population (650/km2) which is aggregated onto relatively few islands within each atoll. The problems of the capital island, Male, have reached a critical level in several areas increasing the island's susceptibility to episodic events such as storm generated high waves. In other areas and atolls the problems are also locally severe increasing the susceptibility of such areas to future climatic change and sea level rise. This increased susceptibility is due to
• coral mining for construction and road surfacing;
• land reclamation, particularly on the seaward edges of islands;
• construction of coastal infrastructure including sea walls, breakwaters, jetties, piers, groynes and harbours;
• aquifer depletion and saline intrusion.
All of the current environmental problems are exacerbated by:
• high population growth;
• a lack of mechanisms within government for taking environmental problems into consideration in the planning process;
• a lack of guidelines and procedures for the evaluation of environmental issues;
• a lack of an adequate in-country data base covering many physical and biological parameters;
• a shortage of trained manpower at all levels.
The current, environmentally unsound, practices will increase the susceptibility of the Maldives to changes predicted to occur as a consequence of global warming and the greenhouse effect. Assuming a sea level rise of 12-18 cm by the year 2030 one might anticipate profound effects on those islands of the Maldives which have been structurally modified, since the normal processes of sand genesis, deposition, removal and flux between sinks have been altered by changes to the micro-climate and current regimes.
The impact of 'high waves' will be greater with greater mean sea level and such increases must be taken into consideration in planning future coastal infrastructure. Changes to aquifer volumes may be expected under higher sea levels; however, such changes will be less important on islands where the aquifer is not currently over-exploited. Saline intrusion will be exacerbated in those aquifers which are heavily used for human consumption.
Increased temperatures (of perhaps 1.5°C by 2030) will affect the human environment, agricultural production, and marine ecosystems. Given the country's proximity to the equator, the Maldives can expect a lower than average temperature rise which may have little impact on the human environment but may be expected to result in some increased demand for air conditioning. Agricultural production and terrestrial ecosystems are likely to be less affected than marine organisms such as corals, many of which are currently growing at temperatures close to their upper thermal tolerance limits.
Perhaps the area of greatest current concern in the Maldives is the possibility of an increased frequency of storm-generated swells and high waves, particularly given the experiences of the country during 1987. Analysis of the meteorological patterns in the Indian Ocean is urgently required to predict the possibility of an increased frequency of such events.
Situations of high sea level at the coast of the atolls are caused by storm surges and waves setup. A degree of coastal flooding due to high tides has been experienced in the past at various places. Recent flooding has been made more noticeable by its impact on construction such as sea walls and houses near the shore and on low-lying reclaimed land. The July 1988 high water situation at Thulhadhoo (Malosmadulu Atoll) was caused by high southwesterly waves (5 m high, periods 12-15 seconds) in association with high spring tide and a southwesterly wind. The damage caused was enhanced by the absence of beaches and the presence of vertical low seawalls which magnified overtopping and flooding. These events are a reminder that occasional natural events of long distance swells and high water levels due to wave surge and/or high tides, would in themselves cause little damage to the Maldives atolls, were it not for the mismanagement that has taken place in recent years. This mismanagement includes stripping the islands of the natural defences afforded by the outer reefs, the reef flats and the beaches.
Social impacts arising from changes to island stability and/or habitability are likely to be extensive given the nature of Maldivian society which is characterized by generally low mobility and strong attachment to individual atolls and islands. Economic impacts will be most intensely felt if the tourist industry is adversely affected. The present structure of the tourist industry is based on 'resort islands' which are essentially self-contained and as a consequence pack considerable infrastructure on the land and coastline of very small islands. The present tourist industry is concentrated in the Central Maldives, hence increasing the risk to this sector of the economy.
Excerpt from J Pernetta and G Sestini, The Maldives and the Impact of Expected Climatic Changes, UNEP Regional Seas Report 104, UN Environment Programme, Nairobi, 1989, pp 27-8; 38.
Sea level rise could increase the severity of storm-related flooding. The higher base for storm surges would be an important additional threat in areas where hurricanes, tropical cyclones and typhoons are frequent, particularly for islands in the Caribbean Sea, the south eastern United States, the tropical Pacific and the Indian sub-continent ... Many small island States are ... particularly vulnerable. This is reflected in their very high ratios of coastline length to land area. The most seriously threatened island States would be those consisting solely, or mostly, of atolls with little or no land more than a few metres above sea level. Tropical storms further increase their vulnerability and, while less in magnitude than those experienced by some of the world's densely populated deltas, on a proportional basis such storms can have a much more devastating impact on island nations.
These predictions of the increasing impact of severe storms have not been lost upon the insurance industry.
In a review of the effect of climate change on insurance one of the world's largest reinsurers has recently written:
Detailed measurements in the Pacific show that the areas with water temperature at the surface above 27°C have expanded by about one-sixth in the last two decades. While substantial fluctuations from one year to the next and additional factors such as El Niño make it impossible so far to prove the effect of such higher temperatures on the frequency of tropical cyclones, super-hurricanes 'Gilbert' and 'Hugo' may certainly be regarded as a clear sign of an increase in hurricane intensity. According to estimates, hurricane activities in the Caribbean will increase considerably in the next two to three decades, the loss potential going up by more than 50%. Together with the rising level of the sea, this also means a much greater risk of storm surges in densely populated coastal regions in the tropics and sub-tropics.
These insurers predict that there will be a global increase in catastrophic losses from a current annual average of about US$20 billion to about US$100 billion per year in overall economic terms. And that, it may be assumed, does not take into account the potential longer term consequences of climate change.
In theory, any loss which is not inevitable is insurable under a commercial contract of insurance. A classic definition of the necessary legal elements of such a contract includes the requirement that the event insured against 'should be one which involves some amount of uncertainty'. The definition continues: 'There must be either uncertainty whether the event will ever happen or not, or if the event is one which must happen at some time there must be uncertainty as to the time at which it will happen. An element of uncertainty is therefore a necessary legal requirement for a loss to be insurable.
In commercial terms, more will be required if a risk is actually to be insurable on the world insurance markets. The requirements for the insurability of a flood or inundation risk commercially have been stated to be:
1 a sufficient demand for insurance of the risk and sufficient variation in the insurer's portfolio of such risks to ensure an adequate spread of risk;
2 the random occurrence of losses, but neither too often nor too seldom;
3 the ability to restrict accumulation of losses;
4 the traceability of the causes of a loss;
5 the availability of statistical records of losses from the risk, covering sufficiently long periods of time.
The randomness or uncertainty of the event to be insured against is here again an essential element. But for commercial insurers, it is also essential to be able to select the risks that they insure and to be able to calculate realistically what premium they should charge, in order to accumulate over the long term, sufficient funds to meet losses and to make a profit.
The IPCC Working Group I concluded that even if action is taken to limit emissions, there is enough momentum in the global climate system for a rate of accelerated sea-level rise to be inevitable. Gradual flooding and inundation which may be attributable to an inevitable rise in sea level will certainly not be capable of being insured on the world insurance markets.
Even in the sphere of catastrophe insurance (which in this context refers to insurance against damage caused by storm, windstorm, hurricane, tropical cyclone, typhoon, as well as flood and inundation) events that up to the present have been regarded as fortuitous, and therefore insurable, may cease to be so regarded in the future. One insurer, in considering the impact of global warming and climate change upon catastrophe insurance, has remarked: 'There should be further examination of what constitutes an "abnormal" loss, and we will have to alter our view of what is perceived as normal.
In considering the potential for heavy increases in catastrophe losses as a result of climate change, it is important to bear in mind that the current levels of catastrophe loss represent in turn a dramatic increase over the losses of two decades ago. Considering the major natural disasters over the last three decades and extrapolating figures to 1990 prices, it can be shown in the 1980s compared with the 1960s, that: overall economic loss from major natural disasters increased by a factor of 3.1; insured losses increased by a factor of 4.8; and major catastrophes increased by a factor of 5.0.
The world's insurance and reinsurance markets will find it difficult to cope with such rapid change in the incidence of natural hazards. For example, the reinsurers of first layer catastrophe insurance covers granted to United Kingdom companies for worldwide exposures paid out between 1979 and 1988 in claims resulting from 'abnormal' weather experienced in the UK alone, more than twice the total amount received by these reinsurers in premium.
In stable climatological conditions, an insurer or reinsurer can assess the probable 'return period' of a known catastrophe peril by reference to historical events and statistical records. The premium for the risk can then be calculated with some reasonable measure of accuracy. Rapid changes in climate make the calculation of premium for natural catastrophe risks a very much more difficult and uncertain exercise. In a 1963 study, for example, it was calculated on the basis of existing records that the return period for a flood which inundated over 800 square kilometres of the east coast of England in 1953 was I in 200 years. Yet between 1963 and 1989 (26 years), the 1 in 200 year storm surge level occurred eleven times.
Given the difficulties currently facing insurers and reinsurers in basing premium calculations on past climatological records and those posed by predictions of global warming and sea level rise, insurers and reinsurers will likely become more selective and restrictive in their offered cover.
Options for insurers
The options available to insurers and reinsurers in such circumstances include:
1 impose upon the insured, in the case of original insurances, substantial deductibles per policy or risk;
2 impose a limit figure on the amount of their liability;
3 charge sufficient premium to enable adequate reserves to be accumulated;
4 exclude particular risks (such as, for example, wind storm, flood, inundation) either absolutely or in particular localities;
5 avoid insuring any catastrophic risks in particular localities or limit overall catastrophe exposure in particular localities;
6 withdraw from the insurance or reinsurance of catastrophe risks altogether;
7 redefine categories of loss which in insurance terms are at present regarded as 'abnormal' or fortuitous, and therefore ordinarily insurable, and those which are regarded as 'normal' or non-fortuitous, and ordinarily not insurable.
Additionally, insurers and reinsurers can assist and encourage the adoption of loss-prevention measures either by advice or by the introduction of conditions into policies of insurance or reinsurance. But faced with the probability of heavy increases in catastrophic losses from so-called natural causes, insurers and reinsurers will be forced to take some, if not all, of the measures mentioned if they are to continue to provide an effective and responsible service to their insurers and if they are to remain in business as insurers.
This logic suggests strongly that insurers will not cover a substantial part possibly the most substantial part - of the potential loss and damage resulting from climate change and sea-level rise. In short, such risks will be uninsurable on the world's insurance and reinsurance markets. This outcome has indeed been recognized by the insurance industry itself. One company recently stated that:
On the level of direct insurance and reinsurance, the insurance industry fortunately has a wide range of 'tools' end instruments for supervising and controlling its exposure to catastrophe risks.
These underwriting facilities . . . must however be used in full and without delay if insurers are to bring the current adverse development to a halt. At the same time these activities must be accompanied by structural loss prevention measures and suitable precautions taken by the authorities.
Wherever the national insurance industry supported by worldwide reinsurance capacity (which, ultimately, is only available if premiums, terms and conditions are adequate) is not able to cover an extreme catastrophe risk, it may become necessary for the government to assume a share in catastrophe covers, at least for losses exceeding a certain level, or to guarantee indemnification in all events, as some governments already do in the case of the risk of flood and inundation.
It is evident that even in the case of fortuitous catastrophic losses, the world insurance industry will not be able to bear the burden of providing a global spread of compensation without state intervention. And in the case of those losses which the insurance industry will have to redefine as 'normal'- for example, flooding or inundation resulting from inevitable sea level rise State funding will be the only source of compensation of those who are affected by it. Since the least developed countries are those most vulnerable to future catastrophic loss and what may come to be regarded as the inevitable consequences of climate change, any state compensation regime will have to be internationally funded.
Two internationally funded compensation schemes operate at present. These deal with oil pollution and nuclear damage. Neither case, however, is analogous exactly with loss consequent upon climate change and sea level rise. On the contrary, the incidence of oil pollution or nuclear damage is more likely to be traceable to an identifiable source and referable to a discoverable cause.
Loss or damage of the nature associated with climate change, whether of a catastrophic or 'abnormal' nature or of a 'normal' or inevitable nature, is highly unlikely ever to be conclusively attributable to any legally definable source or cause. Thus, if a compensation scheme is to form any part of a Climate Change Convention, it could probably not be based upon ordinary legal criteria of liability or responsibility. Rather, it would have to be based on a criterion of responsibility in its broadest sense.
Compensation for loss or damage resulting from the escape or discharge of oil from ships is governed by two conventions. The first lays down a regime of civil liability for oil pollution. The second sets up an international compensation fund to cater for those instances where the civil liability regime does not afford compensation to the injured party.
The first Convention, the International Convention on Civil Liability for Oil Pollution Damage 1969, contains uniform international rules and procedures under which shipowners are made strictly liable for certain types of oil pollution damage. It imposes upon the bulk carriers of persistent oils the obligation to effect insurance against the liabilities imposed by the Convention. The Civil Liability Convention came into force in 1975 and there were amendments to its financial provisions by Protocol in 1976.
The 1971 International Convention on the Establishment of an International Fund for Compensation for Oil Pollution Damage ('the IOPC Fund') provides for the payment of compensation to any person suffering oil pollution damage, as defined in the Civil Liability Convention, if that person is unable to obtain full and adequate compensation for one of the following reasons:
1 that no liability for the damage arises under the Civil Liability Convention, because the shipowner can invoke one of the exemptions from liability set out in that Convention;
2 the shipowner is financially incapable of meeting his obligations under the Civil Liability Convention and his insurance is insufficient to satisfy the compensation claims;
3 the damage exceeds the shipowner's limit of liability under the Civil Liability Convention.
The IOPC Fund also provides an indemnity to shipowners against a proportion of their liabilities under the Civil Liability Convention. It is financed by levies on oil importers. By Article 10 of the Fund Convention, contributions to the fund are levied on any person who has received 'contributing oil' (crude oil and heavy fuel oil) in a quantity exceeding 150,000 tonnes in one calendar year in a Contracting State.
Under Article 11 initial contributions are payable by such importers when a State becomes a member of the IOPC Fund. Annual contributions to the fund are levied to meet the anticipated payments of compensation and indemnity by the IOPC Fund in the ensuing year and to meet also the administrative expenses (Article 12).
The functioning of the IOPC Fund is controlled by an Assembly, an Executive Committee and a secretariat headed by a Director. The Assembly comprises representatives of the governments of all Member States. The Executive Committee is elected by the Assembly and must approve settlements of claims against the Fund. The amount of annual contributions to the Fund is decided annually by the Assembly.
Contributions are levied directly from and paid by the individual importers and not by the Contracting States, unless a Contracting State declares that it assumes that obligation (Article 14). The Contracting States are required to communicate annually to the Director of the IOPC Fund the name and address of any person in that State who is liable to contribute to the Fund as well as details of the quantities of 'contributing oil' received.
Contributing oil is counted for contribution purposes each time it is received at ports or terminal installations in a Contracting State after carriage by sea. The place of loading is irrelevant: the oil may be imported from abroad, carried from another port in the same State or transported by ship from an off-shore production rig. Oil received for transhipment to another port or received for further transport by pipeline is also considered as 'received' for contribution purposes. Fears expressed at the time the Fund Convention was adopted that the Secretariat would have difficulties in collecting contributions have proved to be unfounded.
In 1984 two Protocols were adopted amending the Civil Liability Convention and the Fund Convention. They increased the limits of the shipowners' liability and also increased the limit of compensation payable by the IOPC Fund in respect of any one incident from an aggregate amount of 60 million SDRs (Special Drawing Rights) to 135 million SDRs. This latter figure will be increased automatically to 200,000,000 SDR when there are three Member States of the 1984 Fund whose combined quantity of contributing oil received during a given year in their respective territories exceeds 600 million tonnes.
The Protocols also extend the scope of the Conventions to oil spills from unladen as well as laden tankers and redefined the term 'pollution damage'. The Protocols introduce the concept of compensation for impairment to the environment, although they provide that such compensation shall be limited to loss of profit and to costs to reasonable measures of reinstatement actually undertaken or to be undertaken.
The 1984 Protocols are not yet in force. The 1984 Protocol to the Fund Convention enters into force when ratified by at least eight States and when the total quantity of contributing oil received during a given calendar year in all the ratifying States is at least 600 million tonnes. As of 1 May 1991, forty-five States had ratified the Fund Convention.
A different approach from that adopted by the Oil Pollution Fund Convention has been adopted in the case of nuclear damage. The 1960 Paris Convention on Third Party Liability in the field of Nuclear Energy, concluded under the auspices of the OEEC (now OECD) was the first international convention to regulate the liability for risks arising out of the peaceful use of nuclear energy.
The purpose of the Paris Convention is to harmonize national legislation with regard to third party liability and insurance against atomic risks and to establish a regime for liability and compensation in the event of a nuclear incident. The Convention generally applies only to nuclear incidents occurring, and damage suffered, in the territory of Contracting States. In 1963, a supplementary convention to the Paris Convention was adopted ('the Brussels Supplementary Convention'.)
Article 3(b) of the Brussels Supplementary Convention (as amended by the subsequent 1964 Protocol) provides that the contracting parties undertake that compensation in respect of damage caused by a nuclear incident (other than one occurring entirely in a territory of a non-contracting State) shall be provided up to an amount of 120 million units of account per incident as follows:
1 up to an amount of at least 5 million units of account, out of funds provided by insurance or other financial security, such amount to be established by the legislation of the contracting party in whose territory the nuclear installation of the operator liable is situated;
2 between this amount and 70 million units of account, out of public funds to be made available by the Contracting Party in whose territory the nuclear installation is situated;
3 between 70 million and 120 million units of account, out of public funds to be made available by the Contracting Parties according to the formula for contributions specified in Article 12 of the Brussels Supplementary Convention.
Article 12(a) of the Brussels Supplementary Convention provides that:
The formula for contributions according to which the contracting party shall make available the public funds referred to in Article 3(b)(iii) shall be determined as follows
(i) as to 50%, on the basis of the ratio between the gross national product at current prices of each contracting party and the total of the gross national product's current prices of all Contracting Parties as shown by the official statistics published by the Organisation for Economic Co-operation and Development for the year preceding the year in which the nuclear incident occurs;
(ii) as to 50%, on the basis of the ratio between the thermal power of the reactors situated in the territory of each Contracting Party and the total thermal power of the reactors situated in the territories of all the Contracting Parties . . .
The Brussels Supplementary Convention therefore creates an international pool out of which the highest layer of compensation is payable in the event of damage from a nuclear incident, and levies contributions toward that pool from Contracting States according to a formula based upon GNP and total capacity of their nuclear reactors.
Unlike the Paris Convention, the 1963 Vienna Convention on Civil Liability for Nuclear Damage is potentially of worldwide geographical application. The Vienna Convention also establishes a regime for liability and compensation for nuclear incidents, but it permits the Contracting State within whose territory the installation is situated to limit the liability of the operator.
By Article VII, the installation State undertakes to require the operator to maintain insurance or other financial security covering his liability for nuclear damage in such amount as the State shall specify. But the installation State is bound to meet compensation claims up to the operator's limit of liability, if the operator's insurance or other financial security is inadequate to meet such claims. There is, however, no provision similar to the Brussels Supplementary Convention either in the Vienna Convention or elsewhere for any international pooling arrangement to pay compensation above the operator's limit of liability.
I conclude that the preceding analysis poses at least seven implications for the ongoing negotiations under the rubric of the Climate Change Convention.
First, it is clear that the consequences of inevitable gradual sea level rise due to climate change induced by global warming will not be insurable on the world's insurance markets. Second, even in the case of the 'abnormal' catastrophic consequences of climate change and sea level rise - floods, inundation, storms, windstorms, hurricanes, tropical cyclones, typhoons insurers already acknowledge that there will be a substantial shortfall in their coverage of losses in the future. Insurers are already forced to decline to insure certain risks in particular areas and increasingly they will be forced to refuse to insure risks in the geographical regions most prone to catastrophic loss. As a result they will have to impose deductibles, limit the total amounts of their liability on individual risks and limit their liabilities in the aggregate.
If as predicted the incidence of catastrophes resulting from the combination of sea level rise, severe storms and storm surges increases substantially, a large proportion of catastrophic losses will be uninsured and indeed uninsurable on the world's insurance markets. The insurance industry recognises that even in the case of catastrophe insurance, government intervention will be necessary.
Third, losses of such magnitude cannot be carried by the governments of those countries most vulnerable to these hazards - the small island and lowlying coastal developing countries. There is a need for an internationally funded insurance pool.
Fourth, the oil pollution and nuclear damage Conventions referred to in this paper provide examples of international insurance pools that have been set up in other contexts. One is funded by the oil industry, the other by States on a basis which takes account of gross national product and nuclear energy capacity.
Fifth, the basis for contribution to an internationally funded insurance pool in the present context, and the criteria to be satisfied in claiming from such a fund, should form part of the negotiations leading to the Climate Change Convention.
Sixth, such an international insurance pool might be funded by the developed countries on a contribution basis related to gross national product and/or greenhouse gas emission levels and/or by the industries of such countries responsible for such emissions.
Finally, the Ministerial Declaration at the Second World Climate Conference recommended, inter alia, that stepped-up financial contributions be provided by developed countries to address the particular problems and needs, including funding, of low-lying coastal and small vulnerable island countries. Such countries must be the principal beneficiaries of an internationally funded insurance pool set up to cover losses resulting from sea level rise and related catastrophic events stemming from climate change.