Help, not hope: managing catastrophe exposures has become more about preparing for the worst, and less about hoping for the best.

As 2005 showed us, the chances of experiencing multiple catastrophes in a given year are not as remote as we may have hoped. In recent years, managing multiple catastrophe exposures has become increasingly complex--necessitating an intimate knowledge of everything from changing modeling tools and climates to increasing regulatory oversight and varying legal developments. Insurers that can deftly navigate these exposures will have a better chance of making informed and, ultimately, profitable underwriting decisions.

Since 2001, insurers have become more aware of their property and casualty exposure aggregation and of developing or updating models to determine where these aggregations are concentrated geographically. Insurers not writing property insurance for a cluster of buildings in a major city still should be aware that they may have significant general liability and workers' compensation exposures in that same cluster. In the event of a terror attack, casualty risks may be as costly as property risks.

Evaluating aggregation isn't just important on the granular geographic level. Litigation and regulations also affect insurers' catastrophe management strategies. The passage of many new state laws--such as those that bar nonrenewals of policies after a catastrophe or freeze insurance rates--underscores the importance of considering not just property values but the potential negative effects of regulatory and legal trends. Determining what percentage of a book of business may be adversely affected by such trends is just as important as any other evaluations insurers conduct to predict loss.

Climatic changes represent another area of growing concern. Not only has there been a rise in the number of hurricanes and weather-related catastrophes, but droughts and wildfires have led to increased erosion exposures. Geographic locations that were considered safe are beginning to worry insurers, which should act now to incorporate those concerns into their models before a century-old fault line becomes active or a Category 5 hurricane slams into an unprepared locale.

Catastrophe models are critical to managing catastrophe exposure--but they are models, not oracles. For instance, models are typically based on decades of loss experience, but may not yet reflect recent significant loss potential due to the impact of changes such as climate shifts or state laws. Another limitation is that models may not take into account unlikely hut possible catastrophic loss scenarios, such as we saw with the models of New Orleans before Hurricane Katrina that did not adequately account for the levee breaks. The lesson is that insurers can get more out of their models if they use them as a starting point and then also draw on their own loss experience and knowledge of emerging trends affecting their exposures.

While hope is a good thing, it is neither a strategy nor a way of conducting business. In the end, managing catastrophes still hinges on three basic questions: What is being insured, where is it located and how much of it do we have? What's new is the need to evaluate these questions in significantly greater detail and complexity. For companies that don't, successful catastrophe management will remain an elusive goal.

Steven R. Pozzi a Best's Review columnist, is senior vice president, Chubb & Son, and chief underwriting officer for Chubb Commercial Insurance. He can be reached at [e-mail].