LUMP SUM AUTOMOBILE INSURANCE

a. How Rates Are Calculated
Automobile insurance is currently sold in lump sum (or all you can drive) pricing.A consumer goes to his or her insurance agent and purchases a policy tailored to his orher needs. The price of this policy is determined by several factors such as theconsumer’s age, gender, location, driving history, the type of vehicle insured, theaverage distance driven, the type of coverage purchased.8 This information is used toassign the consumer to a risk class which companies use to determine insurancepremiums. For each risk class companies determine an average annual mileage foreach driver.9 Drivers can get a discount for driving a certain percentage below theaverage annual mileage depending on

their insurer.10 However, the actual mileage is generally self-reported and not verified by the insurer. Ironically enough, the mileagedriven is one of the best indicators of the driver’s risk of being in an accident.11For example, consider drivers A and B. Driver A is a twenty-three year-old malewho lives in Tallahassee, FL. He has a safe driving history and drives approximately tenthousand miles annually. Driver B is exactly the same age as Driver A and has anidentical driving history. Coincidentally both drivers also drive the same vehicle,reported the same annual mileage, and purchased identical policies. Based on thesecharacteristics Driver A and Driver B will likely fall into the same risk class. Forsimplicity, we will assume that the average annual mileage for their risk class is fifteenthousand miles. However, Driver B’s annual mileage is up this year and he drivestwenty thousand miles. Based on the difference in this year’s annual mileage it appears that Driver B’s ata greater risk for being in accident. This is because at any given time there is “x”possibility of being involved in an automobile accident for each mile you drive.12 Thus,Driver A’s risk level is 10,000x and Driver B’s risk level is 20,000x. This would indicatethat Driver B is at a significantly greater risk to be in an accident than Driver A.Ironically, under lump sum pricing both men pay the identical premiums for theirinsuranc

b. Informational Asymmetry Creates Problems for Everyone

In the above example there are two possibilities for what is happening. First,Driver A purchased too much insurance. Second, Driver B purchased too littleinsurance. In practice, both are true. Before either A or B purchased their policies, thecompany determined the premiums for their risk class by taking the total miles to bedriven by all members of that class and dividing them by all members of the class. Thepremium is then determined by multiplying the average mileage driven by a member oftheir risk class and adding the premiums from the additional coverage purchased, if any.Thus, the company designed the policies so that low-mileage13 drivers like A subsidizehigh-mileage drivers like B. In fact, most consumers drive less than “average” and pay ahigher price per-mile under lump sum pricing than high-mileage drivers.14The automotive insurance market represents a market for lemons problem.15 Inthis market the informational asymmetry is in favor of the consumer. It is the consumerwho best knows his or her driving habits and likelihood of accidents. This disparity ininformation causes insurance companies to make educated guesses, based on actuaryscience, about the risk posed by insuring a particular driver. When determining the risklevel of a particular applicant insurance companies typically look at the coverageselected (e.g. incidents covered by the policy), the profile of the driver (based on age,gender, driving history, location of residency) and the usage of the car.16The informational asymmetry is furthered by some state policies. Some stateshave adopted policies that allow drivers who received tickets for violating the driving13 Low-mileage is only used to show that A drives below the average mileage in the example.14 Edlin and Karaca-Mandic.15 Akerlof.16What Determines the Price of My Policy

laws to complete traffic safety classes.17 These classes result in adjudication beingwithheld, the drivers not being assessed points against their driving licenses, and theincident not being reported to their insurance companies.18 While it is arguable thatthese classes will encourage future safe-driving habits from the ticketed driver, theymask, from the insurance company, the true risk posed by the drivers. Thus, the currentstate of automotive insurance and traffic law enforcement furthers the informationaldisparity against the insurance companies. In order to make the system more effectivethe informational asymmetry between the parties must be eliminated or at leastreduced. While the average driver may at first consider the informational asymmetry to bein his favor he would be incorrect. Insurance companies are very astute at assessingrisk. Through their actuaries, insurance companies come up with driver profiles basedon the factors mentioned above. In doing so they spread the risk among all of thedrivers of a particular group. For example, in the hypothetical example above, Driver Aand Driver B would pay the same insurance rate. On the surface this appears to be aperfectly reasonable system. However, when you add additional factors such asindividual driving behavior, the system fails to reflect actual risk.19 In fact, the higher risk driver pays a below-actual-risk price for his insurance while the safer driver subsidizesthe other driver’s bad driving habits.20

c. Lump Sum Pricing Leads to Economically Inefficient Behavior

One of the problems with this system is that it leads to economically inefficientdriving habits. Because B’s policy is priced as a lump sum he is not likely to realize themarginal cost of each mile driven. This is primarily because the costs of his drivinghabits are externalized to A. Thus, B might drive more than he really values. Theeconomist William Vickrey noted this in his 1968 paper on automobile insurance.The basic difficulty is that the insurance premium appears to the individualautomobile owner almost entirely as part of the fixed cost of owning a car . . . Theresult is that with the possible exceptions of the decisions as to whether to driveto work or use public transportation, and of the decision as to whether youngermembers of the family are allowed to drive at all, the added exposure to riskinvolved in added usage is not brought to bear on the decision.21For example, assume that the insurance company that sold A and B their policiesused the following pricing scheme: the company calculated the insurance premiums forthe desired coverage and sold coverage at a per-mile at price. Under this situation, B’sdriving habits would cost him 20,000m which would be greater than his lump sumpremium which was based on fifteen thousand miles driven. This could affect B’s drivinghabits in a number of ways. He could drive more, drive less, or drive the same. Any ofthese responses would be more economically efficient than the results under the currentpricing scheme. Under the new scheme B’s usage would correspond to his actual value of each mile driven. Assuming that his original policy cost 15,000m, B would not driveany additional miles unless he valued each additional mile to be worth at least “m”.Thus, any behavior that he makes under the new system is more economically efficientbecause it based on the actual per-mile value of the behavior

d. Negative Secondary Effects of Lump Sum PricingThe

behavior of drivers who drive more than they really value has numerousnegative effects on society. First, it increases insurance premiums for a majority ofconsumers. Second, it increases Green House Gases (GHGs). Third, it increasesreliance on foreign oil, which weakens the American Dollar. B’s increase in miles driven creates an increase in the probability of a vehicularaccident, as noted by “x.”22 It would increase “x” because it would increase the volumeof traffic on the road while B was driving.23 Thus, it is likely to increase the insurancepremiums of everyone else and also negatively impacts society by increasing the risk ofan accident and the anxiety associated with that for every member of society. Thisincrease in premiums has an especially significant impact on the poor. On averageindividuals with lower incomes tend to drive less than wealthier individuals.24 However,the increase in insurance premiums is the same for individuals in both groups. Thus,wealthier individuals have their driving habits subsidized by the poor. Furthermore, thissubsidization has a disparate impact on the poor because the increase in insurancepremiums is a larger percentage of their capital.25.