They also have emerged as a new battleground in the ratemaking process.
The battle is this: Insurers want to charge enough to cover potential losses and maybe even make a profit. Regulators, however, are under political pressure to keep rates low.
Regulators want to keep insurers honest and ensure that catastrophe models aren't skewed in favor of the industry by modelers who want to keep their customers happy. But there's more than one way to skew that cat.
Models that show an increased frequency or severity of hurricanes, for instance, are never going to be popular with regulators and policy makers--especially those in politically-charged jurisdictions. And that is one of the big concerns about the idea of the National Association of Insurance Commissioners developing its own national multi-peril model.
The NAIC recently approved a $200,000 study to evaluate the possibility of leveraging the Florida Office of Insurance Regulation's Public Hurricane Risk and Loss Model to build a national multi-peril model. The NAIC's Industry Liaison Committee was expected to have a proposal in place by the time of the association's national meeting in mid-March.
A regulator-sponsored catastrophe model may provide an alternative to the private sector models on the market. But would it be any more accurate or fair than the models that are already out there? In fact, it could be far worse given that such a model would be much more vulnerable to political manipulation.
Consider what happened in Florida a couple of years ago.
In 2007, the Florida Legislature commissioned a study using millions of dollars to fund a public risk model, claiming that private models might overestimate risk.
Florida International University developed a model, but after a delay, newspapers reported that FIU researchers had found their public model's estimates for catastrophic risk were higher than those of private firms.
Additional tinkering only raised concerns that the process was tainted and that the model had been altered for political gain.
At about the same time, Risk Management Solutions developed a model that took a medium-term view of hurricane activity, reflecting the increase in hurricane frequency and intensity in the Atlantic basin.
But that version had to be modified before it could be used for residential ratemaking purposes so that it estimated hurricane activity on a straight historical average based on the number of hurricanes recorded since 1900.
While the potential for political manipulation is one issue, there's also the question of the cost of developing and maintaining this model. The NAIC's Property & Casualty Committee, which first came up with the idea of a model, initially asked for "significant funding". It would almost certainly cost tens of millions of dollars to develop and then maintain the model.
The NAIC, an insurance industry entity, does not have to take the Florida approach to catastrophe risk management. The function of assessing catastrophe risk is too important to allow it to become politicized and it's highly likely that any public model would be politicized.
The intentions may be good. But the road to hell is paved with them.
PATRICIA VOWINKEL has worked for national media outlets for more than 20 years.
April 1, 2009
Copyright 2009© LRP Publications