“The affects of a model change vary on a case-by-case basis,” Mr. Mount explained. However, it is not A.M. Best’s intent to “automatically downgrade a company just because of a model change.”

Mr. Mount said the rating agency works with a company to asses how it will address the adverse effects of a model revision, such as that announced earlier this year by Risk Management Solutions (RMS). The company can purchase more reinsurance, or it can outline a plan to cut back on exposure to reduce risk.

A.M. Best does not require the use of one particular modeling company, nor does it require the use of multiple models, according to a briefing released by the company. Therefore, if significant revisions are announced for one model, the affect on the overall assessment of risk should not severely impact a company’s capital position—or at least the chances of this occurring are reduced. A.M. Best does expect that companies incorporate model updates “as soon as practical.”

“Where there is the potential for significant increases in model results due to a model update, and the company has not yet run the model, A.M. Best will apply a factor to the probable maximum loss to more accurately capture the current view of risk,” the Oldwick, N.J.-based rating agency said.

A.M. Best also specifically advises companies not to rely too heavily on catastrophe models.

“As was evident in the past, an overreliance on models can be problematic,” the briefing said. “While catastrophe models provide an estimate, they cannot be expected to provide an exact loss figure.”

Karen Clark, president and chief executive of Karen Clark & Co., a risk and risk management firm in Boston, said rating agencies "need to stop using a formula based on a point estimate determined by a model. This puts insurers in a straightjacket, forcing them to buy reinsurance to the level the rating agency requires."

A change in loss results due to a model revision is “not necessarily a strike against the company,” said Mr. Mount. A.M. Best looks at other factors, such as the company’s track record with other storms.

Companies can also attempt to convince A.M. Best that their in-house PMLs (probable maximum loss) are better than the standard model assessments—or that one model fits them better than the other—in which case a weighted average can be used. The basic approach is to use a straight average of multiple models, A.M. Best explained.

Ms. Clark said she is "not sure reinsurance companies will give as much credibility to the new model; a 100 to 200 percent change can stop you in your tracks."

Modeler RMS said changes in loss results in the market portfolios it analyzed typically range between increases of 20 percent and 100 percent due to its revisions. In some cases, loss results can be higher or lower than the average range, said RMS, which revised its model based on new data derived from the 2004 and 2005 hurricane seasons as well as Hurricane Ike in 2008.

Reinsurers have the ability to take a look at all three models available (from RMS, AIR Worldwide and EQECAT). Most primary insurers can’t afford all three. Some of the smaller companies don’t license any but get access to the model loss results through brokers, said Ms. Clark.

Frank Pierson, executive vice president and chief technical officer of independent reinsurance brokerage firm Holborn Corporation, said reinsurers may be able to absorb the model changes without huge increases because they already thought the models were coming in low.

“Now that the models incorporate the [2004 and 2005] losses, the impact on primary insurance could be moderated by the prices reinsurers were already using.”

Mr. Pierson still cautioned that what he termed the “I Don’t Believe Factor” among reinsurers may soften as their confidence in models grows, resulting in upward pressure on pricing.

Moody’s Investors Service has said revisions in catastrophe models alone most likely won’t offset the lack of demand in the reinsurance market. Moody’s said reinsurance demand remains stifled because primary insurers have tighter reinsurance budgets and lower volumes due to reduced economic activity. This is not expected to change due to the model revisions, “despite the expectation of increased modeled losses in certain peril zones,” Moody’s said.

However the model revisions, when combined with catastrophes during the first quarter, could mean a change in the market, according to RenaissanceRe CEO Neill Currie. Mr. Currie’s comments were made before the magnitude 9.0 earthquake and tsunami that recently struck northern Japan.