Most readers will be familiar with the system of insurance rate regulation that prevails in the U.S. Property/casualty insurers file rates with state insurance regulators that are based on projected loss costs and expenses, plus a reasonable provision for profit. Projected loss costs are based on specific risk factors whose utility in predicting the risk of loss must be verified to the satisfaction of the regulators. Charging different rates for the same coverage to individuals with similar risk profiles is considered "unfair discrimination" and is thus proscribed.

Judgment Involved in Ratemaking

This process, however, has never been mechanistic. Regulators typically understand that judgment is part of the ratemaking process, primarily due to the uncertainty surrounding the actuarially indicated, cost-based rates and factors. Thus, rate filings have often involved some deviations from actuarially indicated costbased rates. Regulators have frequently accepted such deviations with the implicit understanding that there is a "reasonable range" around the indicated rates and factors. Not surprisingly, companies report that deviations are far more likely to be approved when they fall below the indicated rate or fall between the current and indicated rates.  

Traditionally, the practice of subjectively adjusting or supplementing cost-based rates to align with customer conversion (that is, the likelihood that a prospective policyholder will accept an offer of coverage at a given premium), retention (that is, the likelihood that an existing policyholder will remain a customer if confronted with a premium increase), and similar market considerations has been referred to by terms such as "market-based pricing," "rate smoothing," and "rate tempering." 

One question, then, is whether price optimization truly "represents a departure from traditional cost-based rating," as the Ohio bulletin would have it. It's arguable that the perceived novelty of price optimization lies mainly in its use of formal modeling techniques, as opposed to managers' subjective judgment, to deviate from cost-based rates based on the market considerations identified previously—and that such deviations are not novel at all. 

A second question is whether an insurer's use of formal models, algorithms or other statistical techniques, including those that consider consumer price sensitivity, should cause regulators to alter past practices for reviewing rate filings. Regulators have historically required that each factor be correlated with future costs such as losses and expenses; that any deviation from the actuarially indicated cost-based rate must be actuarially justified as "reasonable"; and that all individuals within a given risk class must be charged the same rate.

For example, if an insurer were to use a customer pricing method that resulted in different rates or premiums for individual insureds within the same class, a regulator would rightly consider the result to be unfairly discriminatory. It should not matter whether the insurer's pricing method was called "price optimization," nor should it matter if the method involves formal modeling as opposed to subjective judgment. Likewise, the fact that an insurer may have taken customer price sensitivity into account when developing its rates should be considered immaterial to the question of whether the rate filing complies with existing insurance laws.

One could argue that explicit regulation regarding price optimization is unnecessary, given that most states have language in their insurance law prohibiting rates that are "inadequate, excessive, and unfairly discriminatory." However, if regulators feel compelled to explicitly regulate price optimization, it is imperative that they precisely define the meaning of the term. The states that have issued bulletins banning price optimization have tended to offer superficial definitions based largely on anecdotes about the presumed effect of price optimization on consumers and assumptions about insurer motives. 

If nothing else, the NAIC's white paper could help resolve the controversy over price optimization by reminding regulators that actuarial projections are a starting point and that market conditions, such as demand and competition, have always played a role in property/casualty insurance pricing.  

Robert Detlefsen, Ph.D., serves as the vice president, Public Policy at the National Association of Mutual Insurance Companies.