A key group of regulators has decided that insurers should use traditional formulas as well as more complicated statistical models when establishing reserves for variable annuity guarantees.[@@]

The Financial Conditions “E” Committee, an arm of the National Association of Insurance Commissioners, Kansas City, Mo., voted here at the NAIC’s summer meeting to approve a proposal for VA risk-based capital requirements that includes use of both a “standard scenario” and “principle-based,” “stochastic” methods in efforts to determine the RBC levels that insurers need to back VA guarantees.

The NAIC’s executive committee and the plenary, or body that includes all voting NAIC members, still must approve the VA guarantee reserves proposal, also known as the C3-Phase II proposal, before the proposal officially can take effect. But many observers say they are sure the proposal will be adopted by the NAIC plenary, and states already are preparing to add the new rules to the instructions for 2005 life insurance company annual reports.

NAIC panels also are holding similar discussions about Actuarial Guideline 38, a model for setting reserves for universal life products with secondary guarantees.

The VA guarantee proposal that the “E” Committee has approved seeks to address the stock price risk, interest-rate risk and expense-recovery risk associated with variable annuities and group annuities that offer death benefits or living benefit guarantees.

Traditionally, regulators have used relatively simple formulas to set VA product reserve requirements.

Some individuals in the insurance community want to stick with a traditional “formulaic approach,” but many insurers and regulators want to shift to a reserve-setting system that would combine use of basic reserving principles, sound actuarial judgment and stochastic models.

“Stochastic” is a term used to describe equations that include randomly changing variables, such as variables that represent shifts in interest rates and stock prices. When actuaries use stochastic models, they can test how product guarantees will perform under hundreds or even thousands of scenarios.

Supporters of a principle-based approach include many insurance commissioners, many life insurers and the American Academy of Actuaries, Washington, which has helped to develop the principle-based approach proposal.

New York regulators have developed a compromise approach that would phase in use of stochastic models but create a minimum floor for risk-based capital levels for VA guarantees, by requiring insurers to use a “Standard Scenario” to compute minimum RBC levels.

Lou Felice, a New York regulator and chairman of the NAIC’s capital adequacy task force, said New York is committed to a principle-based approach but also thinks that a certain conservatism and a phasing in of a principle-based approach is warranted.

The “E” Committee approved a measure based on the New York proposal.

But many insurance executives and regulators who support a principle-based approach have argued that the C3 Phase II proposal should be adopted without the Standard Scenario proposal.

At one NAIC meeting session, the American Council of Life Insurers, Washington and the National Alliance of Life Companies, Rosemont, Ill., called for the Standard Scenario to be removed.

Here are some other views on the C3 Phase II issue presented at the NAIC meeting:

- Ann Henstrand, a representative for Jackson National Life Insurance Company, Lansing, Mich., said the Standard Scenario did not have support in the industry. She called for a hearing, similar to the hearing that is being organized to review UL guarantee RBC proposals.

- Steve Johnson, deputy commissioner with the Pennsylvania Insurance Department, said change should be implemented carefully. He said insolvencies may have been prevented because of the conservatism of using a formulaic approach.

- Mike Batte, a New Mexico regulator, argued that a lot of resources have been expended to develop the C3-Phase II RBC proposal and that it would be a “mistake” to create a minimum floor by making one scenario a standard scenario. Rather, it would be better to use the Standard Scenario as an informational tool instead of a part of the formula used to calculate minimum reserves, Batte said.

- Doug Barnert, a representative for the National Alliance of Life Companies, Rosemont, Ill., recommended putting the original actuarial academy report in place for 2005 without the Standard Scenario. To include the Standard Scenario would be to take “the wrong step at the right time,” Barnert said. Work on the C-3 Phase II portion of the project has been under way for 3 years, he noted.

- Mike Akers, a representative for American International Group Inc., New York, said 13 letters to regulators supporting the actuarial academy’s work “is a solid indication of what the industry thinks.”

- John Bruins, a representative for the actuarial academy, said one Standard Scenario might not be a good measure of risk. He said the Standard Scenario could be used for informational purposes.