Andreas Siebert

In a nutshell: Geoinformation in the insurance industry

Geocoded data have mostly been used up to now for the purpose of managing natural hazard risks in property insurance. But the fascinating satellite images to which Google Earth has given us access indicate that the insurance industry can profit much more from the benefits of geotechnologies.

Since 11 September 2001 and the devastating hurricane Munich Re has already been applying the principle of years of 2004 and 2005, the insurance industry has been spatial information analysis for many years as a means of seeking ways to analyse and control its risk exposures representing natural hazards and catastrophe events. more efficiently. However, the ability to manage catastrophe risks depends to a great extent on how well insurers know the risk situation, the risk concentration, and the affected lines of business in the insured area.

The questions insurers and reinsurers must first ask in this are: “Where are the risks. What are the indemnity application: telecommunications, energy supply, and limits?” The answers are an important step towards improving risk transparency – something that is also required in the context of Solvency II. The most promising key to all this is geographical underwriting, a procedure that produces the transparency required to analyse an insurer’s risks in spatial terms.

Help from above

The world of geodata was given an enormous boost when Google Earth was launched in mid-2005. This linking of exact satellite data with specialist data and its visualisation have made this technology accessible to user groups not previously classed as traditional specialist users.

The true strength of GIS is to be found in its capacity to draw together a wealth of – initially unconnected – pieces of information by way of their spatial or geographical relationship, thus facilitating a holistic view. In the insurance industry, all portfolio, rating, and loss characteristics can be linked with exposure data or comparative client and market data.

First the graft ...

The starting point for cutting-edge portfolio management and optimisation is to determine as accurately as possible the geographical location of the risks – i.e. their longitude and latitude – by means of geocoding. This is a practice the industry introduced thirty years ago in the form of the so-called CRESTA zones for property insurance. CRESTA zoning, which is for the most part based on postal and/or administrative areas, is used for the exchange of liability information between insurers and reinsurers – particularly in connection with the accumulation control of large portfolios.

But the dramatic floods of recent years, the growing significance of the terrorism risk, and the need for better risk models have increased the demands in terms of the precision and spatial resolution of the data. With only a few metres deciding whether a flood will cause losses or not, even more detailed information was required on the risk situation.

The solution is to be found in precise geocoding at street or address level with GPS accuracy in the metre range – similar to that provided by car navigation systems.

Geocoding tools are now so advanced that entire portfolios can be visualised at address level. Munich Re’s Geo Data Service geocodes and analyses portfolios throughout the world. The results are then transmitted to our clients, who can then analyse their risks separately by hazard class.

A welcome development is that more and more insurers have acquired the facility to extract the necessary portfolio information from their own complex portfolio management systems This involves a considerable amount of work but the subsequent possibilities for analysis make this initial input more than worthwhile.

... then the knowledge gain in risk management

Whether dealing with facultative industrial business or large property portfolios, the geographical approach is a great boost to risk assessment. Geocoded data support more or less all of the steps taken by an insurer along the value chain. This already happens in the processes of product development and rating, when spatial differentiation plays a major role – as in the case of residential buildings and contents covers and motor insurance.

Indisputably, geocoding achieves the greatest added value in the observation and analysis of accumulation situations. In this context, the procedures Munich Re has developed provide first-class analysis options – whether for individual clients, entire markets, or scenarios like Winter Storm Europe or Earthquake Japan.

Furthermore, all the solutions for modelling and calculating scenarios in respect of natural hazards and terrorist attacks are based on GIS with geocoded data. But GIS can also lead to a considerable improvement in loss management, providing facilities for loss prognosis and prevention, support in claims handling, and faster estimates following catastrophes and detailed analyses in the subsequent review phase.

And last, but not least, client segments set up on a spatial basis optimise sales management. They pave the way for selective geomarketing measures and the evaluation of clients for client-scoring purposes.

continue »