Natural hazards and insurance

As a reinsurer, Munich Re is particularly affected by the growing losses from natural catastrophes. On the one hand, as a reliable partner, it is intent upon meeting its indemnification obligations. On the other hand, and this is in the insureds' interest too, it has to make sure that claims payments do not get so out of hand that it cannot continue to operate effectively as a reinsurer. After all, the most expensive insurance is the one that does not pay up.

The following insurance techniques and mechanisms help to cope with the increasing loss burdens:

• —
  Risk-adequate premiums: Risk-adequate prices and conditions are a central feature of a reinsurer's quality in that they guarantee the durability of business relations. The basis for premium pricing is the average annual loss potential. MRHazard, the computer model developed by Geo Risks Research to support the assessment of natural hazard risks, is used to calculate PML rates and premiums from past loss experience (retrospective underwriting). Besides this, Munich Re advocates prospective underwriting, which takes future loss trends into account as well.
• —
  Loss potential: One of the important factors in this context is the probable maximum loss (PML). This is an estimate of the largest probable loss amount an insurer will have to reckon with from a certain event. It is calculated using the ratio of event magnitude to expected damage. Note that the PML is not a worst case scenario in the sense of an absolute maximum loss. It is a probable maximum loss, the loss which, in other words, could materialise given a still realistic occurrence probability (which is defined, incidentally, by each insurer in the light of its own willingness to carry risk and its own financial resources). The PML is an imperative parameter for a company wishing to protect itself against exceptional losses which would exceed its capital strength and could therefore result in its financial ruin. The expected average annual loss (AAL) corresponds to the size of the net premium and is developed in an integration process across all losses.
• —
  Accumulation control: The process of accumulation control entails compiling risks in terms of their geographic distribution and risk class (e.g. residential building or industrial facility). Geo Risks Research defines accumulation assessment zones as separate spatial units for the purposes of accumulation control. The compilation of liabilities provides the input data for the geo-risk model (MRHazard) and thus forms the basis for the calculation of PMLs and prices. The future of accumulation control points in the direction of increasingly precise recording of insured objects with their exact addresses. We call this process of spatial coding "geographical underwriting", for which special tools are being developed in Geo Risks Research.
• —
  Deductibles: Deductibles have two functions. On the one hand, they act as an incentive for policyholders to prevent losses. On the other hand, they reduce the number of small losses that have to be adjusted, which in the case of major catastrophes may be very large. In return, it allows the premiums to be reduced accordingly. The size of premium reductions and the alternatives for various deductibles can be determined by Geo Risks Research using MRHazard.
• —
  Limiting liabilities:
• —
  Reinsurance: This is based on the PML calculated using MRHazard.
• —
  Loss prevention and loss avoidance: Reducing the risks present at the insured location or of the insured object itself. This requires a high level of risk awareness on the part of the policyholder and/or the authorities. Geo Risks Research contributes to a heightening of risk awareness in the form of publications, seminars, presentations, and media appearances, with the aim of transmitting its specialist knowledge of the insurance sector to the public at large and the decision-makers on political committees.

next page »

Hazard maps

Vital assistance in coping with increasing risks is provided by hazard maps. Hazard maps range from world maps containing information on a wide variety of natural hazards to analyses of the concrete exposure of a specific location to, say, earthquake or flood. The represented exposure may relate to a single type of event (like the new earthquake hazard map for Germany, Austria, and Switzerland, which is used as a basis for construction standards) or may consider all the extreme natural hazard events that could occur. Hazard maps are produced by statistically processing instrumental recordings, observations, or historical accounts of events.

Hazard maps are the basis for allocating areas to tariff zones. Each individual zone has its own tariff which is applied throughout that zone. Germany, for example, has a zoning system for flood, backwater, and heavy rain, called ZÜRS.

GIS

GIS can be used not only to overlay geographical data with event data (superimposition) but also and, in particular, to visualise underwriting data like average building values or liability distributions. This makes it easier for the insurer to assess the danger of accumulation in its book of business and also for public authorities to gain an overview of the loss potentials, regardless of whether they are insured or not.

The planning of public preparedness or escape routes can thus be much more effective. Particularly when it comes to providing cover for flood damage, it is essential that the often small-scale but nevertheless major differences in the hazard involved are compiled and evaluated accurately so that appropriate steps can be taken in the design of the insurance cover.

Geo Risks Research also has a unique archive of natural catastrophes, and this provides an important starting-point for the evaluation of new catastrophe losses: the MRNatCatService database. It contains reports and analyses of losses connected with natural hazard events throughout the world since 1975. "Great" natural catastrophes that occurred before 1975 are also researched and documented. The more losses that are recorded from the past, the easier it is to assess how losses will develop in the future and to identify "trends" in natural hazard events.

Extracts from the MRNatCatService database may be found on the 2004 edition of our CD-ROM "World of Natural Hazards".

« previous page