With auto insurers struggling to achieve top-line growth as pricing remains flat, no factor is more critical for healthy underlying margins than keeping loss costs favorable. Indeed, divining such trends has become a favorite pastime for equity analysts seeking to separate the winners from the losers in the world of personal lines insurers.

The carriers themselves shy away from offering their own projections of such trends for fear of tipping off the competition on what their own data tells them. So comments from them at reporting time remain fairly Delphic in their brevity.

Last month, Progressive posted some impressive profitability figures despite less than impressive premium growth–which the company in its commentary attributed to below-average frequency and severity trends.

Bear Stearns analyst David Small seized on that comment as the one bright spot given declining premium-per-policy figures.

“These comments should quell investor fears that these factors were changing directions,” he wrote in a recent report. “We have not been believers that a reversal of the declining trend for frequency and severity was imminent, and continue to believe the industry will benefit from a secular decline in auto frequency.”

Factors behind the decline in frequency over the past two decades remain solid enough to confound those skeptics who believe in general that in these matters what goes down must come back up, he added.

To figure out the future with most matters–but particularly with auto premiums–analysts start by looking at the present and how that came about. The short version for auto insurance is that while frequency has shown a favorable decline, just the opposite is true with severity.

Diana Lee, vice president of the Property Casualty Insurers Association of America, sees certain symmetry in the trend. “Over the last four years, the average cost of insured losses nationwide rose almost 15 percent for the major [auto] coverages–liability, collision and comprehensive–while the reporting of claims went down by the same amount,” she noted.

Automobiles rising in value–along with the cost to repair them–combined with higher medical costs to treat those injured in auto accidents have contributed to severity rates trending unfavorably, analysts contend.

But the same cars–while more expensive–are also safer thanks to features such as anti-lock brakes and airbags.

“Programs such as the graduated driver's license have helped to ease beginning, young motorists into the driving environment by restricting their time of driving and having adult supervision while in the car, unless they have more experience on the roads,” Ms. Lee noted.

Earlier this month, the New York-based Insurance Information Institute predicted that auto premiums will remain virtually flat this year, expected to rise a half-point after last year's 2.5 percent increase.

One factor altering the frequency-severity equilibrium of the past few years is the record catastrophe losses of 2005, with predictions of more of the same for the foreseeable future.

While homeowner insurance losses and disputes over flood versus wind coverage received the bulk of media coverage following Hurricane Katrina, insurers received nearly 674,000 claims for vehicles that were damaged or destroyed by last year's storms. “Those claims occurred across a wide swath of Southern states and cost insurers some $3.2 billion,” said Robert Hartwig, the Insurance Information Institute's chief economist.

Looking into her crystal ball, Ms. Lee said the current balance could be tipped in favor of greater severity and higher premiums in future years, with more common and severe catastrophe losses a major factor.

But on the positive side, the current spike in gasoline prices could curtail driving and with it auto frequency, although it will be several months before any reliable data on that can be assembled, Ms. Lee said.

Mr. Small said that so far for all the relative data concerning gas prices and miles driven, correlations show there is generally no decline in the latter when the former rises. “What generally affects miles driven is the Gross Domestic Product,” he said. “Only in the case of a shortage will people drive less.”

The abundant supply is one reason why Kim Hazelbaker, senior vice president of the Highway Loss Data Institute, feels Congress won't mount its own version of “That 70s Show” with a return to the 55 mile-per-hour speed limit. Such a move would presumably impact frequency favorably, but would not seem likely from lawmakers facing election in a few months.

The favorable frequency trend in auto thefts will also impact comprehensive coverage costs, analysts note.

In 2004, the number of stolen cars decreased by 1.9 percent–the first such drop in five years. “The good news is that preliminary FBI data for the first half of 2005 indicates that the auto theft rate fell by 2.1 percent,” Mr. Hartwig said, with declines posted in every region except the West.

Analysts credit new security devices–such as electronic tracking systems that help police find stolen vehicles–with helping to keep these premiums down.

One puzzling trend Mr. Hazelbaker hopes will continue is the sharp decline in collision costs. “Basically, when you don't really know why something is happening in cases like this, you always worry that it might end for the same unknowable reason,” he said.

New safety features just now going into production have a great potential to keep the frequency trend favorable–but worsen severity as they will increase repair costs when a collision does occur, according to Mr. Hazelbaker.

Fraud expenses continue to contribute to premium costs, but news on that front is improving as well. “However, crackdowns by law enforcement agencies and insurers have put a definite dent into organized insurance fraud,” Mr. Hartwig said.

In November 2004, then New York Superintendent of Insurance Greg Serio announced a major jaw-boning effort to get insurers to lower rates after the loss ratio in the state's private passenger auto market declined to .61 from .86 two years previous.

Among other factors, Mr. Serio said back then it was time for insurers to return the anti-fraud “dividend” to state policyholders.

Earlier this year, New York's current superintendent, Howard Mills, said state anti-fraud efforts in 2005 have cut premium costs by between 3 percent and 10 percent.

Perhaps one of the most important trends impacting loss costs is their overall diminishment in importance, as other factors such as credit and education assume new importance in ratemaking and underwriting.

Mr. Hartwig said efforts in states to ban credit scoring will have the effect of raising premiums for good drivers while lowering them for drivers involved in most accidents.

In its update on p-c insurance premium growth forecasts, the Tillinghast business of Towers Perrin said the increasing use of tiered pricing in the auto world has resulted in a general price flattening over the years.

“We do not necessarily believe that companies see a revenue-neutral impact from tier pricing,” the report said. “The competitive environment within this line is such that higher-rated insureds are able to move their business to other companies and find more favorable prices.”

Morgan Stanley analyst William Wilt said p-c insurers–particularly in auto–are moving into the top-tier of television advertisers, and he sees this as one of the new components to be studied when separating the wheat from the chaff.

“From our perspective, analysis beyond monitoring loss-cost trends and premium rates will be needed to pick the likely winners and losers over the next two-to-three years,” he said.

For Table:

Special Reports /

With Pricing Soft, Auto Insurers Hope To Stay In Driver's Seat On Loss Costs

Related Stories

Recommended For You