CHAPTER

9

Pay-As-You-Drive for Car Insurance

by Winston Harrington and Ian Parry

We urge, Mr. President, that your administration implement measures, particularly federal tax credits, to accelerate experiments in automobile insurance reform. The experiment of most interest is pay-as-you-drive (PAYD) insurance, which varies much more strongly with mileage than most automobile insurance plans currently do. If this policy reform works as its advocates predict, it would reduce the number of highway deaths without increasing overall insurance costs for the average motorist. Environmentalists also embrace PAYD, because reduced driving would moderate other problems associated with motor vehicle use, including oil dependence and emissions of conventional pollutants and greenhouse gases.

Background

More than 40,000 people die in motor vehicle accidents each year, and another 3 million suffer injuries. The costs to society from these fatalities and injuries, and additional costs from property damage, travel holdups, emergency services, and so on, have been measured at several percent of gross domestic product.

Although driving is much safer than in the past, thanks to improvements in vehicle technology (airbags, shatterproof windshields, and so on), better roads, and reduced drunk driving following raises in the legal drinking age, the number of road deaths and injuries is still substantially higher than it need be. One reason is that people do not consider the full societal cost of accident risk when deciding how much and how often to drive. Although they may take into account the risks of injury to themselves and other family members when making these choices, they are unlikely to consider other costs, such as the risk of injury their driving poses for other drivers and pedestrians, the costs of vehicular damage that is covered through insurance claims, and the costs to other motorists held up in traffic congestion caused by accidents.

Automobile insurance is the principal mechanism for coping with these accident externalities. Accident insurance premiums are “experience rated,” individually tailored to risk factors known to be associated with causing accidents, such as one’s age, sex, residential location, and prior crash record. However, almost universally, premiums are levied on an annual lump-sum basis and therefore vary greatly with the number of vehicles owned by a household, but only vary moderately with how much those vehicles are driven. Consequently, the current insurance system does not induce motorists to fully consider accident risks when deciding how much and how often to drive.

Unfortunately, per-mile insurance has failed to get much attention in the past because of the difficulty of measuring mileage reliably. A decade ago, a related idea arose, motivated primarily by environmental considerations.

In 1994, the EPA convened an advisory committee nicknamed “Car Talk” to attempt to find a consensus on the means of reducing greenhouse gas emissions from motor vehicles. Among the ideas that generated much discussion was a “pay-at-the-pump” (PATP) insurance scheme, wherein a portion of auto insurance would be collected by means of a surcharge (of about $0.25 to $0.50 per gallon) on the price of gasoline; lump-sum annual premiums would be reduced accordingly. This proposal, which traded on the commonsense notion that accident risk is strongly related to miles driven, and therefore gasoline consumption, had two attractive properties: it created an incentive to drive less without raising the cost of owning and operating a vehicle for the average motorist, and it was easy to collect and difficult for motorists to avoid. The proposal would have required each state to impose surcharges or risk losing some federal transportation funding.

However, the proposal was strongly attacked by the insurance industry, on the grounds that the surcharge per gallon would be the same for all drivers, regardless of a driver’s age, sex, location, and previous crash record. In its attack, the industry was enunciating an important and sensible principle: if insurance reform is adopted for environmental reasons, it must first make sense from an insurance perspective. The PATP proposal fails this test in significant ways.

The Case for Pay-As-You-Drive Insurance

With PAYD insurance, existing rating factors would be used by insurance companies to determine a driver’s per-mile rate, and this rate would be multiplied by annual miles driven to calculate the annual insurance bill. Thus by allowing companies to charge higher per-mile rates to drivers at higher risk of causing accidents, PAYD avoids the major flaw of PATP. PAYD relies on the notion that once individual risk factors have been controlled for, mileage driven is an important risk factor. Although the public literature contains surprisingly little conclusive evidence on this point, what does exist suggests that it is reasonable to assume that total accident risk is proportional to total mileage.

An additional advantage of PAYD is that, unlike PATP, it can be voluntary and therefore introduced gradually over time as people gain familiarity with the idea of mileage-based insurance. Insurance companies can market PAYD insurance alongside their conventional insurance plans. Drivers with low annual mileage would have an incentive to opt for per-mile insurance, as it would enable them to reduce their annual premium payments. Under the current system, both low- and high-mileage drivers pay approximately the same annual premiums, even though the former are less likely to crash and make claims on insurance companies. As low-mileage drivers begin to opt out of lump-sum insurance, over time this would increase rates for the remaining drivers, in turn providing them with more incentive to switch to mileage-based insurance.

We believe that the universal adoption of PAYD insurance would have a striking effect on motor vehicle accidents, perhaps saving up to 5,000 lives per year. For example, a driver between ages 25 and 70 with no recent crash record typically pays about $800 per year in liability and collision insurance premiums for a family car or minivan up to several years old. If the premium was levied on a per-mile rather than lump-sum basis, the per-mile costs for the driver would increase by 6.6 cents for a vehicle driven 12,000 miles per year. This would have the same effect on raising the marginal costs of driving as would increasing the federal gasoline tax from $0.184 to $1.50 per gallon, for a vehicle with fuel efficiency of 20 miles per gallon. The rise in per-mile driving costs would be higher still for drivers in a higher risk class, such as youths with prior crash records.

The political opposition by motorists to a sevenfold increase in the federal gasoline tax would be overwhelming. A very attractive feature of insurance reform is that it can achieve much of the desired effect of higher gasoline taxes without a politically suicidal increase in the tax burden on motorists. This is because the increase in the per-mile cost of driving is offset by the corresponding reduction in annual premiums; the average motorist could still choose to drive the same amount as before the insurance reform and be no worse off. In fact, many motorists—at least those with lower-than-average mileage—would end up paying less in driving costs under PAYD than they currently do.

Why hasn’t the industry itself introduced PAYD? For one thing, it is no panacea. Only the mileage of the vehicle can be monitored, not who is driving the vehicle. Thus when a youth drives his or her parents’ car, the rate per mile does not change, even though the driver is a much bigger statistical risk than either parent. More seriously, until recently, no reliable and accurate way has existed of collecting mileage information from motorists. However, new technology, in the form of Global Positioning Systems (GPS) and nearly tamperproof odometers, has largely eliminated this problem, though in the latter case motorists still would be required to bring in their vehicles for periodic odometer inspections.

More seriously still, a PAYD system has external benefits that cannot be captured by the issuing insurance company. When an insurer charges by the mile, its costs are reduced to the extent that its own customers reduce their accident risk by driving less. However, the costs of other insurance companies also are lowered, by reducing the risk of involvement in a two-car accident for their own customers. These other cost savings cannot be captured by the company offering the mileage insurance. This is a significant market failure, and it potentially justifies policy intervention to jump-start the transition toward PAYD insurance.

Civil libertarians have raised some concerns about privacy issues and the possible misuse of information obtained through mileage monitoring about individuals’ driving habits. GPS information, in particular, would allow a motorist’s movements to be traced. However, safeguards can be considered if issues actually do emerge, and, as long as PAYD insurance is voluntary, drivers can weigh the potential loss of privacy against the benefits of the policy, similarly to the decision about whether to use cell phones, which also raise privacy concerns.

Recommendations

To provide a jump-start toward using PAYD, we recommend altering the taxation of companies providing auto insurance in favor of per-mile insurance. You could do this by granting the companies an annual federal tax credit of at least $50 for each customer that is charged on a per-mile basis. Revenue costs could be capped by limiting the tax credit to the first, say, 25 percent of policies switched to a per-mile basis, and by including sunset provisions. The policy might be funded by a tax penalty for all remaining customers charged on a lump-sum basis; this penalty would initially be very small, but it would grow over time as more customers switched to mileage insurance, providing an ongoing impetus for the transition to PAYD.

This federal tax initiative would reinforce trends that are just beginning to emerge at the state level. In Oregon, insurance companies have been offered a state tax credit of $100 per motorist for the first 10,000 motorists who sign up for PAYD insurance plans. In 1999 in Texas, the Progressive Insurance Company of Ohio was granted permission to conduct an experiment with PAYD insurance. The company was able to sign up 1,100 customers. The experiment was canceled in 2001, apparently because the rarity and expense of the GPS devices required by the plan kept demand to a minimum. Nonetheless, the Texas legislature subsequently passed new legislation authorizing auto insurance companies to offer per-mile insurance. State governments in Maryland and Connecticut also are considering measures to encourage PAYD.

We believe that tax incentives for mileage-based insurance offer both a highly effective and politically feasible opportunity to effectively reduce the numbers killed on U.S. highways. And as a bonus, it would provide substantial reductions in greenhouse gas emissions and other externalities associated with auto use.

W.H.

I.P.