Cents-Per-Mile Car Insurance for Texans

Links to the sections below

How Insurers Can Use Odometers.
Insurer's top three excuses for not using odometers.
Three real reasons why insurers are afraid to use odometers.
How insurers sabotage compulsory insurance by refusing to use odometers.

As of January 1, 2002, auto insurance companies for the first time can offer Texas car owners a choice between:

[1] Continuing to pay fixed installments at an annual rate.
or
[2] Buying miles of insurance protection in advance as needed—like buying gasoline—at a cents-per-mile rate.

It’s a fact that the miles put on anyone’s car may vary widely from year to year as circumstances change. That’s why a one-size-fits-all rate cannot possibly fit most of the thousands of cars a company assigns to a rate group. No matter how little (or how much) your own car is driven, you still pay for the cost of the average number of miles driven by all of the cars in your group. Cents-per-mile choice would still keep cars in the rate group but would individualize what each owner pays.

Offering a mile rate requires a company to go one step beyond setting its dollars-per-year rate for each group. In the additional step, the company converts each existing annual rate into the group’s alternative cents-per-mile rate. To do this, a company only needs to divide the group’s year rate by the average annual miles of the cars in the group.

It works this way. An insurance company assigns your car to one of its rate groups according to your zip code, car use and type, driver type, and other information about your household the company obtains. Your car might be put into a group paying $400 a year. If the company determined that the average for cars in your group was 10,000 miles a year, the alternative mile rate for your group would be 4.0 cents a mile.

If you chose the mile rate instead of the annual rate, you might initially buy 2,500 miles for $100 (= 4.0¢/mi. x 2,500 mi.) plus a nominal expense fee. These miles would be added to your car’s current odometer reading. Before these miles were all driven, you would have to buy more miles to stay legally insured.

You drive, you pay - you don’t drive, you don’t pay. And you buy as many miles of insurance protection as you need when you need them.

Compulsory insurance would finally work. With insurance on all miles traveled, the cost of the insurance system in Texas would be reduced for the benefit of everyone.

With one additional step, insurers can now sell miles of car insurance at cents-per-mile rates. But Texans must demand that the legislature make the insurance companies take that small step.

Don't expect the excuses to make sense. Auto insurers rely on good driver versus bad driver sales talk and black box “risk factors” to keep their simple product complicated and confusing. The following excuses about loss of efficiency, loss of risk factors, and theft of insurance are more of the same.

“It will be administratively difficult and expensive to keep track of miles.”
-- Jerry Johns, Southwestern Insurance Information Service, Austin KVUE-TV News, February 19, 2002

Not so. Cents-per-mile customers buy miles of insurance protection as needed, like buying gallons of gasoline. Insurance automatically ends when the odometer limit (recorded on the car’s insurance ID card) is reached unless more miles are bought. Customers keep track of miles on their own odometer to know when to buy more. The company does no after-the-fact billing of the customer, and the customer doesn't have to estimate a "future annual mileage" figure for the company.

Customers will quickly learn to watch their own miles. For example, someone who buys coverage from 60,000 miles to 62,000 miles on the odometer, but inadvertently drives to 62,500 miles before buying more miles (risking a ticket) would probably not tell the company about driving uninsured miles. When buying more miles, however, it would dawn on them that they are paying for 500 miles of protection not received.

The company only needs to keep track of the miles purchased and add them to the car's odometer record. The company is then able to verify that insurance is in force if a claim is submitted. As a condition of issuing a policy and renewing it each year, the company has each odometer read at a company service center, by a licensed inspection station or even by an agent. This allows the company to base next year's cents-per-mile rates on its total cost of claims for each category divided by the total miles of exposure driven during the year by all of the cars in each category.

In fact, insurance verification at traffic stops will be significantly more efficient for both law officers and insurers. Today's ID card shows the policy term but not whether insurance has been cancelled within the term through failure to pay an installment. Verification now requires a time-consuming check with the agent or company. Under the per-mile alternative, checking the mile limit on the ID card against the odometer reading will show immediately whether insurance is actually in force.

“The Insurance Council of Texas says companies aren’t offering the cents-per-mile option because it doesn’t account for risk factors like where you drive or what time of day.”
-- Dallas WFAA-TV News, Jan. 30, 2002

It is not possible to lose any real factor that is causally related to the amount of risk (like car type and use, car value, and amount of insurance) because the choice of a cents-per-mile rate is offered only after the company's risk factors have been used to categorize the car.

If a company currently claims to use job locations, work hours, or other profiles as indications of average "where" and "time of day" driving conditions for assigning cars to its annual rate categories, then those risk factors—to the extent they are not false ones like zip code and credit history—are also incorporated in the alternative per-mile rates offered for cars already assigned to the rate categories.

“People will roll their odometers back to make their cars worth more. Why would people not be inclined to do that in this instance as well?”
-- Doug Johnson, Insurance Council of Texas, Dallas WFAA-TV News, Jan. 30, 2002

There are powerful legal and practical deterrents to this way of stealing insurance protection. Odometers have always served as the measuring device for resale value, rental and leasing charges, warranty limits, mechanical breakdown insurance, and cents-per-mile tax deductions or reimbursements for business or government travel. Odometer tampering—detected during claim processing—voids the insurance and, under decades-old state and federal law, is punishable by heavy fines and jail.

As a practical matter, resetting odometers requires equipment plus expertise that makes stealing insurance risky and uneconomical. For example, in order to steal 20,000 miles of continuous protection while paying for only the 2,000 miles from 35,000 miles to 37,000 miles on the odometer, the resetting would have to be done at least nine times to keep the odometer reading within the narrow 2,000-mile covered range. (This repeated lawbreaking defrauds not just an insurance company, but also carmakers and other warranty providers, plus all subsequent car buyers.)

But sauce for the goose is sauce for the gander: If insurers claim to be concerned about having miles of protection stolen from them, where is their concern about collecting money for driving protection when the car is not being used at all? If an owner gets sick and the car just sits, it means that the company rakes in money when there is absolutely no risk of paying any out.

Car insurers may say they don't trust customers with using odometers, but the truth is that customers can't trust insurers without using odometers.

So much for the excuses. In fact, what auto insurance professionals really fear is the threat posed to their credibility and livelihood by the greater efficiency, rate transparency, and direct individual control of insurance expense that per-mile rates provide.

For example, selling additional miles of protection only requires a company to have computer access to the odometer record for each car that it insures and to print out an updated insurance ID card showing the new odometer limit. Therefore, competition on customer convenience eventually may lead to Automated Mile Machines at banks, supermarkets, and filling stations that, for a nominal service charge, sell miles of protection for policyholder cars.

Insurers argue that cumbersome systems using global positioning satellites are necessary to track cars and generate bills at the end of each month for the amount of travel recorded by the system. But that argument appears to be merely a new attempt to preserve the use of arbitrary black box “risk factors” and the scope they allow for rate manipulation to favor target markets.

The selling of auto insurance has always subordinated measurement of risk to all-out price competition in which truth and some customers are the real losers. Hundreds of insurance lobbyists and thousands of company functionaries are able to stay employed by promoting the false "risk factor" profiling that keeps the public confused about what they are paying for.

While some current profiles might possibly become meaningful given a valid statistical basis in mile rates, these will not include the traditional ones used in scapegoating various groups, such as driver sex, credit history, claim record, no-prior insurance, driver record, and residence zip code. These profiles are merely proxies for group differences in annual miles per insured car. (For example, households with credit problems often have to give up a car and put all of their driving on remaining cars, thus raising the per-car annual mileage.)

These differences in annual mileage per car are increased when consumers who need to economize on insurance put more miles on fewer cars in reaction to the fixed annual per-car rates. The differences in cost companies experience are entirely the artificial products of basing costs on insured years, not insured miles. Therefore, group differences in annual cost that are attributed to the worthiness or moral turpitude of the groups distinguished by these false risk factors simply disappear on a per mile basis.

What also disappears is the credibility of an entire industry that routinely pits customers against each other by using hidden information to profit at their expense. Agents and companies refer to the practice of overcharging for cars driven less than average for their annual rate category as "skimming the cream.”

To compete for sales to desirable customers who may buy other kinds of insurance, companies set rates by selecting categories created by not measuring miles of exposure. They also carefully ignore any cost (and the underlying annual mileage) differences that don't conform to the marketing stereotypes they create.

For example, companies have always charged the same annual rates for liability insurance for old and new cars. At one time companies had planned to charge higher liability rates to insure older cars based on negative stereotypes of owners of old cars, so the companies could lower rates for newer cars. But when the actuaries found that older cars on average generate fewer liability claims than newer cars because they are driven less, the industry leaders boasted that "common sense" required companies to ignore these statistics and to continue to charge the same annual rate for cars of all ages. This instance of systematic overcharging by insurance companies (skimming the cream) has now gone on for over forty years.

Using odometers and cents-per-mile rates would call into question the inefficiency and stereotype manipulation that characterize the way auto insurance is currently sold. It is for these self-protective reasons that the auto insurance companies refuse to allow customers the choice of using their odometers to measure and pay for the insurance they need and are compelled by law to buy.

But this “some customers be damned” attitude of insurance companies also extends to a “public policy be damned” attitude about the state's compulsory car insurance law.

The solid idea behind compulsory insurance is: every mile a car is driven creates a risk, and that risk has a statistical, but very real, cost. That cost would not exist if the mile had not been driven. Therefore, each mile driven in a car that is not insured imposes a cost (3¢ to 6¢ a mile) on the insurance system that is not paid by the driver of that car.

Following this logic, Texas requires all registered cars to be insured. When lawmakers adopted compulsory insurance over 20 years ago, they expected that insurance would get cheaper for everyone because more drivers would be sharing the insurance cost of accidents. Resulting lower rates would enable more drivers to buy insurance, making rates even more affordable so that eventually all cars would be insured. But this has not happened.

Despite increased enforcement, the proportion of miles driven by uninsured cars has stayed at about 20%, while the proportion of registered cars that are without insurance is much greater, an estimated 35%. (Most uninsured cars are driven less than average and so are under-represented in Department of Public Safety accident statistics.) The fault is not with the logic of the public policy. Rather the barrier to the success of compulsory insurance is that insurance companies insist on using fixed annual rates to collect the cost of cents-per-mile risk.

The hindrance to getting more cars insured is the immediate legal saving that per-car rates encourage drivers to get from insuring fewer cars and driving them more miles. For example, a family using two insured cars to drive 20,000 miles can reduce their insurance cost 50% by insuring only one car to drive these miles.

One catch is that two drivers may need the car at the same time. Hence the family may illegally keep the uninsured car registered to use for limited driving on an emergency basis.

A second catch is that, while the same per mile cost of accident risk (say 4¢) is imposed statistically on the insurance system whether one car or two is driven 20,000 miles, the insurance system collects only half (50%) as much per mile (2¢) to cover this imposed cost when only one car is used.

As a consequence, this way of saving backfires if too many drivers in a zip code make the same decision to use fewer cars and drive them more miles. In this case, the average annual miles per insured car rises and the amount the company collects falls below the cents-per-mile cost to the company. This leads companies to raise their annual rate for the zip code in order to increase the amount they collect to cover the cost of providing protection for more miles per car.

Thus, the sensible decision consumers make to economize by piling more miles on insured cars sets off a paradoxical spiral of rising cost and fewer cents per mile paid to insurance companies, leading to higher rates per car that result in still fewer cars getting insured.

Tightening enforcement against uninsured cars aggravates this spiraling situation. The greater the risk of being fined, or even of having a car confiscated, the greater the economic pressure becomes, despite the inconvenience of sharing cars, to pile more miles on the cars that are insured. Less-driven cars are forced out of the insurance pool, taking with them more premium than miles (more $ than VMT).