An interview with Dr. Torsten Jeworrek
Opinion: We need to reassess the risk
In an interview with Topics, Dr. Torsten Jeworrek argues that the more frequent extreme weather events throughout the world need to be given greater consideration in risk models. He also believes that an important lesson to be learned from Hurricane Katrina is to consider the inconceivable in risk evaluations.
Tropical cyclones caused enormous damage in 2004 and 2005 — US$ 80bn worth in the USA alone. Were these exceptional events or portents of a new trend?
Dr. Torsten Jeworrek: On the one hand, the record losses of recent years can be attributed to the fact that insured losses have increased enormously. However, the hurricane losses also reflect the general trend that weather events have been increasing in frequency and intensity for several years now. Today we are seeing areas affected that we previously considered to be devoid of exposure. For example, in the past two years tropical cyclones have caused losses on the Canary Islands and in Brazil.
Is climate change responsible for these developments?
Jeworrek: We have to distinguish between two important factors. Firstly, weather extremes are increasing throughout the world, which as far as we are concerned has been clearly shown to be a consequence of climate change and partially man-made. The second aspect concerns the North Atlantic. Surface temperatures in the North Atlantic are subject to natural fluctuation over a period of several decades. As we have been in a warm phase of the Atlantic multidecadal oscillation since the mid-1990s, we have a coming together of two factors which contribute to hurricane activity.
What consequences do these weather extremes have for the insurance industry?
Jeworrek: They bring increased loss potentials and loss frequencies. This results in increased demand for reinsurance capacity and a need for higher prices in many regions. Risk management is especially important in this connection. The bar is being raised continuously — state-of-the-art is the top priority. Starting with excellent risk assessment in underwriting through to risk management at Group level, we have to be extremely disciplined in consistently implementing our scientific and technical expertise.
What specifically can our clients expect: Higher NatCat prices? Event limits?
Jeworrek: Regions and hazards that are expected to produce changes in the risk will have to undergo price adjustments — based on the resulting higher technical demands and costs of the risk capital required. However, this will not mean an across-the-board increase for global catastrophe business. We will modify our models to ensure they are appropriate to a region's and a client's prospective exposure. Event limits to restrict a reinsurer's liability in proportional business are an essential part of modern risk management.
Do you expect less pronounced fluctuations in future price cycles?
Jeworrek: Yes, we do. Firstly, there is now a much greater awareness of the role of loss estimates and exposure. Furthermore, rating agencies and regulators are calling for improvements in processes for modelling, capital requirements and risk management, both for insurers and reinsurers. This pressure is bound to produce more disciplined treatment of the risk and flatter price cycles.
|
We readjusted our loss models for the renewal at 1 January 2006.
|
Is a general reassessment of the risk necessary?
Jeworrek: There is no doubt that hurricane exposure has increased significantly. We must therefore continue the process of factoring this increased exposure into risk management and risk models. Many models are based on retrospective analyses. However, recent climate research indicates that exposure can no longer be presented in terms of a long-term average but must be calculated on a prospective basis. Even if the necessary improvements are made, modelling capabilities will still be subject to uncertainties, and companies will have to make allowance for these uncertainties by taking a conservative view in their risk management.
A question on risk management: Was Munich Re overly optimistic in its risk assessments in the past two years?
Jeworrek: Our internal risk evaluations, especially for windstorm covers, have always been conservative. Hurricanes hit the North Atlantic in 2004 and 2005 with a frequency and intensity that we had not fully anticipated in our risk-measurement calculations. We therefore readjusted our loss models for the renewal at 1 January 2006. Also, our 30 geoscientists and meteorologists continue to work together with the world of science to develop our models further.
Although simulations had clearly demonstrated the consequences of a hurricane in New Orleans, nobody expected a catastrophe on this scale. What are the lessons from Katrina?
Jeworrek: We underestimated the flood risk in our model. However, large commercial risks and industrial risks in the USA nearly all have flood cover. This is what makes the modelling of flood risks so important. Katrina, and the flooding of large parts of New Orleans that it produced, involved loss elements that we had not seen with the major hurricanes of recent years, not even in the case of Hurricane Andrew in 1992. We may have placed too much faith in the technological flood control in New Orleans, which proved to be over-optimistic on our part. An important lesson from Katrina therefore is to include the so-called inconceivable in risk evaluations.
next page »