We both know I am unlikely to convince you to want your car insurance rate to go up, but I will try. Even if you disagree with that conclusion you will hopefully agree that the regulatory status quo for car insurance favors driving over other modes of transportation. At the least I hope that after you finish reading this, you will always consider the omission of compensation from every news story about a car crash, as if suffering ends as soon as cars come to a stop.

First, a story of how the insurance industry works, emphasizing ways that car insurance specifically is inadequate for protecting crash victims. Then suggestions for how regulations could be improved, and how those improvements are consistent with safer driving and a better transportation system.

Insurance is good. It helps people hedge risks so as to avoid the worst outcomes. None of us knows what tomorrow may bring. Often it is better to pay a little today to get a lot of help if something unexpected happens tomorrow. In some loose sense that’s what much of society is, us as individuals contributing and coordinating to take care of whoever needs it most. Insurance is also, in an equally loose sense, as much a way to express what matters to us individually as anything else we buy.

Maybe the simplest positive example is life insurance. Any unexpected early death is tragic. However, some people have particular concerns about leaving behind dependents. The material well being of children usually depends on the income their parents earn through work in the marketplace, plus the unpaid work parents put in at home. Term life insurance for the parent of a newborn provides assurance that if the worst happens to the parent before the child reaches adulthood, at least there will be some financial cushion to assist the child. Given declining marginal utility of consumption, it’s beneficial for families in this situation to pay life insurance premiums and slightly reduce consumption in order to have protection from catastrophic financial harm if a parent dies.

In practice life insurance gets more complicated, but that’s a basic story. It is as helpful of a product as anything else around. Getting life insurance is inherently a personal decision reflecting expectations for future earnings, risk tolerance, who the potential alternative caregivers are, etc. The overhead is fairly minimal. Premiums come in, they get invested in a hopefully prudent manner subject to reasonable regulations, and then get paid out in the clearly documented and undisputed event of death.

Health insurance, at least in America, serves that risk allocation function poorly. Calling it insurance is at least partially a misnomer. “Healthcare payment processing” would be as accurate a term for that industry. Everyone who uses it, and everyone who needs it but can’t afford it, probably has a worse view of insurance overall as a result.

For reasons that I’m not going to get into it serves largely as an inefficient, frustrating, and harmful method of payment for risks which are fairly well known in advance. It is also inefficient, frustrating, and harmful for unexpected medical needs. Running these convoluted health insurance companies is a lot of work, and it is likewise a lot of work for medical providers to interact with insurance companies. All of this overhead effort to process payments though insurance companies raises total costs for healthcare. All of this money spent on overhead means less is spent on actual service, i.e., many people don’t get the healthcare they need that we as a society could otherwise pay for from existing budgets. It is deeply bad.

Car insurance is structurally weirder than the two stories above. You may think of your car insurance policy as one more escalating bill, akin to a utility bill. But the particular structure of the industry is important for understanding what opportunities America has to change it.

Nominally, a “full coverage” plan with all the options can cover almost all of the harms that might result from owning or driving a car. Here is an explainer of terms from Allstate if you have slightly more detailed questions about anything in this section. This post idiosyncratically organizes coverage into four buckets of risk involved with cars. Here are brief descriptions along with a key point for each that will be relevant later.

As a driver, the people in your car including yourself are likely your highest concern in the event of a crash. They are also probably not covered by your policy.

Given how convoluted and expensive American health insurance is (see above), many people will not buy more insurance to cover medical care through their car insurance. I don’t have any precise statistics. Medical payment coverage does make sense for some drivers, and it is available, but I am very confident only a minority of car insurance policies include it.

Except, if your car’s occupants are unrelated to you and you are responsible for a crash, then they do get covered for injuries you caused through the bodily injury coverage which normally applies to people outside of the car.

Key point: Medical care for the driver and their family members are usually not covered by the driver’s insurance.

Bodily injury liability insurance is mandatory. If a driver is responsible for a crash, then their bodily injury coverage pays their victims. This includes pedestrians, occupants of other vehicles, etc.

Mandatory minimum coverage varies by state but is currently somewhere in the five figures per person injured. If you know anything about American healthcare costs you know five figures can run out fast if there’s a serious injury. That’s just for healthcare. Victims should also be compensated for loss of income, and in the event of death their estate should be compensated. Five figures is not much. Drivers can choose to pay more for higher coverage for victims.

Of course we want to avoid injury and death in the first place, and money cannot truly compensate. However, after a crash occurs financial compensation to the victim is very important to alleviate suffering. Driving is a serious responsibility that should entail both minimizing the risk of a crash and also seriously planning for what happens afterwards. When a driver gets behind the wheel they are putting other people at risk. Anyone upset about putting a dollar value on compensation for victims, or who feels that paying the corresponding premium is an unfair burden on drivers, should become an activist against car dependency so nobody drives in the first place. Keeping everyone safe and avoiding troubling questions of liability and compensation would be ideal.

Key point: Drivers choose their own liability coverage limits, not victims. Mandatory minimum coverage for hurting other people is absurdly low. So the expectation in our system is that drivers do not provide full compensation to victims who they kill or severely injure.

Comprehensive and collision coverage, together, will cover any costs associated with damage to a car. They are otherwise optional but likely required for drivers with a car loan. Car loans are secured by the car itself, so that if a borrower defaults then the bank can reposes the car. If the car is severely damaged then the driver will likely stop making payments, but in that case repossessing the car would not benefit the bank. Banks require this coverage so that if the car is damaged then the loan can get paid back.

This bucket of risk is relatively straightforward. Banks require it for reasons summarized above when issuing a loan. A driver without a car loan may choose to buy it, depending on their risk tolerance and personal circumstances. Like life insurance, losing a car is a risk that a person can choose to hedge at their own expense if it makes sense for them.

Key point: This insurance to protect property, specifically the bank’s interest in property, operates fairly well.

Like bodily injury, there is mandatory minimum liability coverage for property damage that the driver causes. These limits also vary by state, are probably too low, and it’s bad if a victim of a crash loses property (e.g., a car) that they depend on and can’t afford to replace. I acknowledge that but don’t particularly care so I won’t explore further.

Key point: Drivers choose their coverage limits, not victims, and mandatory minimum coverage for damaging property is somewhat low. So drivers do not provide full compensation for some property damage that they cause.

Note that car insurance ends up structurally unlike life insurance. Generally the person paying car insurance premiums is not compensated for their harms by their own insurance company. Car insurance generally does not help if the driver injures himself or his family in the car. The mandatory liability insurance is explicitly for the benefit of other people. Coverage for the driver’s own car is nominally aligned with the driver’s interests, but in practice largely operates to protect the collateral of lenders.

None of this is necessarily bad but it does suggest we could make improvements, as seen below.

The clearest difference, at least for the purpose of this post, is how absurdly low America’s mandatory minimums are. The EU mandatory minimum is "€ 6 450 000 per accident, irrespective of the number of injured parties, or € 1 300 000 per injured party". I'm not sure American insurers even offer that for anybody who would pay the premium. No country is perfect. I don’t know that € 1,300,000, or roughly one and a half million US dollars, is sufficient to cover fatal injuries. But when we see such stark policy differences between wealthy countries, we should be suspicious that one of them is making a huge mistake.

https://www.europarl.europa.eu/news/en/press-room/20211014IPR14929/parliament-adopts-new-rules-to-improve-protection-of-road-accident-victims

One change that has nothing to do with insurance policies but would make insurance work better, is enforcing insurance requirements and preventing hit and run drivers from escaping liability. I’ve written before about how we should do better at tracking drivers who commit harm on the road. Using similar technology plus databases of who is actually paying for car insurance, we could also stop drivers without insurance from driving before they cause a crash. This post is long enough without digging into enforcement, but if DOTs aren’t competent enough to do this they should shut down all road traffic. All car insurance policy collapses if drivers ignore insurance requirements altogether, just like progressive tax policy collapses if wealthy people are allowed to commit fraud with no audits.

Ideally the following changes would be adopted across the country. But insurance is regulated state by state. Luxury vehicle owners already register cars in Montana primarily to avoid sales tax, which is bad enough, but these fraudulent registrations can already be used to avoid paying for insurance in a state with riskier roads. So partial adoption on a state by state basis, increasing premium differences between states, would make it even more critical to crack down on drivers who intentionally register in the wrong state.

The most important regulatory change for insurance policies would be to raise mandatory minimums, or potentially replace the whole concept of minimums with a standard flat liability coverage. A driver has the physical power to readily kill a few people and destroy at least one vehicle. Coverage does not have to be unlimited, but it should cover that. If a priceless work of art is anywhere near a public road the owner can buy their own coverage for the art. But if a driver kills or severely injures a victim, the driver’s insurance should cover a fair payment. The DOT’s valuation of a statistical life is 13.7 million. I don’t know if that’s ideal but that’s a starting point. Multiply that by some number of potential victims. Make sure every driver has that coverage, somewhere in the eight figures instead of the current five figure minimum.

A secondary but still important change would be to make medical coverage for car occupants mandatory, the way liability coverage is mandatory today. That way car insurance covers all injuries due to crashes, either through liability insurance from another driver or through the driver’s own medical coverage. As described above, the American health insurance industry is bad, let’s try to avoid it. But also car insurance is the natural place for covering the medical costs associated with car accidents. The choice to drive leads to the medical cost of crashes. This would make car insurance more expensive for drivers, and make health insurance cheaper for everyone. Health insurance already has a medical loss ratio that “requires insurance companies to spend at least 80% or 85% of premium dollars on medical care” and if medical care costs decline then “the issuer is required to provide a rebate to its customers”. Shifting the costs of medical care due to crashes from health insurance to car insurance would decrease the incentive to drive, and benefit non-drivers who tend to be less wealthy.

There are roughly parallel possibilities for shifting costs and coverage from life insurance to car insurance, but that change would be significantly more complicated. This post is long enough so I’ll save it for later, if there’s interest.

Car insurance companies are in the business of predicting risks, because that’s the primary challenge in selling a lot of policies with premiums that are higher on average than payouts. That’s not everything, but it’s really important.

Predicting risks is so important that the insurance companies collectively fund the Insurance Institute for Highway Safety (IIHS) to help generate data about all vehicles sold in America, that can be inputs for company specific actuarial calculations. The IIHS arguably does a better job of tracking safety than the federal government’s National Highway Traffic Safety Administration (NHTSA). The NHTSA is fundamentally more lax than European regulators about vehicle safety, see for example this NJB video. Insurance companies also try to collect data on drivers, either through 3rd party data brokers or by offering incentives to induce drivers to share information about their driving.

The proposed regulations would increase the total amount of payments from car insurance companies to victims, thereby increasing the incentive for the industry to improve their predictions. In particular, the coverage for fatalities would be much higher. So while there would not be much more incentive in understanding the risks of fender benders, there would be much higher incentive for predicting which drivers in which vehicles will cause deaths.

This would likely lead to more crash testing of different models, and in particular testing for risks to pedestrians and bicyclists. It would also increase the incentive for insurance companies to understand particular drivers, so there would be even larger discounts for drivers who could prove less dangerous driving habits.

Drivers regularly modify their behaviors based on costs. It’s a cliche that elections depend on gas prices, but gas prices also change how far people are willing to drive and which vehicles they want to buy. Job applicants considering different job opportunities think about fuel costs for long commutes. Etc. Increased car insurance coverage and premiums would encourage drivers to find ways to bring their premiums back down, by way of reducing their crash risk.

Fleet operators already consider insurance rates in their operations. Specifically, many ambulances and fire engines have speed governors because given their use case speeding causes an unacceptable increase in crash risk, even for trained and attentive professional drivers. I suspect the considerations are similar for some commercial fleets. Given higher premiums, drivers might demand the same sort of safety features to reduce insurance costs. In addition to speed limiters, it could also include driver monitoring to preclude driving by phone users or inebriated people.

Drivers with safer habits would be more likely to share data about driving habits with insurance companies, due to the greater savings. That might teach some drivers to drive more safely, which is great. Some drivers do express concern with current tracking that they feel is unfair because it is beyond their control, such as being dinged for rapid braking in hectic traffic conditions or driving at night when their work commute forces them to do those things. Acknowledging the risk of those commutes is no more a moral judgment than saying a work commute that’s twice as long costs twice as much in gas money. The employer is already forcing the employee to be exposed to those risks and quantifying them in insurance is better than ignoring them.

Car buyers would become more concerned about insurance when buying vehicles. The IIHS already collects and publishes a lot of good data by make and model. In a world where insurance premiums were much higher for vehicles most frequently used in fatal crashes, buyers would have a greater incentive to choose safer vehicles.

Like congestion pricing, increased premiums will encourage people to find ways to reduce their vehicle miles traveled (VMT). But even better than congestion pricing, increased premiums tied to safety risk will also encourage drivers to be safety conscious while shopping for a vehicle and while driving. It is a multi-pronged strategy for reducing traffic violence even faster than VMT.

Among new vehicles, the share of giant trucks and SUVs will decline. Manufacturers will no longer be able to earn fat profits on those categories while ignoring the risk that their vehicles impose on people, because shoppers will prefer cars that are cheaper to insure the same way homeowners are starting to consider insurance costs before buying flood or fire prone properties.

Institutions such as employers, retailers, etc., will have a greater incentive to support mode shifting. Too often small business owners speciously complain that if the government replaces on-street parking with any higher value use for that land, that they’ll lose business. Many of those owners probably drive everywhere and genuinely don’t understand that there are tradeoffs to their behavior. Raising insurance costs to a fairer level will help make drivers more aware of their choice. By making drivers more aware, it will encourage drivers who are also business owners or sit in other positions of power more eager to enable alternative modes of transportation

Fairer prices for driving will encourage mode shifts, e.g., increase ridership for transit and biking. Plus, with drivers taking their responsibility more seriously due to the threat of higher premiums, it’ll be safer for people to walk or ride bikes in public, which will encourage even more active transportation. That in turn will make it easier to justify investments in better infrastructure.

Some drivers cannot afford or at least choose not to pay current premiums, and more will not be able to afford the raised premiums. Hopefully they could use alternative modes of transportation, but some will be unable to drive and have no way to get where they need to go. Letting them drive uninsured is bad policy generally, and bad anti-poverty policy in particular given how drivers are disproportionately likely to injure poor people. This insurance reform proposal is a non-carceral intervention that will raise costs on drivers, who are disproportionately wealthy, in order to directly raise payments to victims and indirectly to reduce traffic violence against victims who are disproportionately poorer. So opposing it on grounds of concern for the poor really reflects concern for the poor only if they drive, i.e., concern only for drivers. But for some impacted folks, maybe there could be rideshare credits or something to help them transition away from car-ownership.

Elderly drivers are the second highest risk age group per mile. Higher premiums, plus resulting recognition of the risk, will encourage more people to responsibly plan for retirement somewhere that they won’t be car dependent. With all the other changes in society, that retirement planning will get easier.

The current drivers most likely to stop driving are children, because their rates would be even higher and it’s easier for a child to delay driving than for an adult to stop. This would be for the good because children are the most dangerous drivers per mile. The more kids graduate high school without a car, the more of them might choose a college where having a car is not necessary, and then put off car ownership further still. As college graduates, maybe they’ll still move somewhere car dependent and with greater maturity be able to drive relatively safety. But just maybe they’ll prioritize moving somewhere that’s not car dependent, and help society move away from cars one generation at a time.

The answer will depend on your circumstances. If you have health insurance, that premium will go down. Risk of death or injury from a car crash will go down. CO2 emissions, microplastic pollution from car tires, and cardiovascular and respiratory disease due to particulates and NOx emissions will go down. If you are injured in a crash, you’ll likely get much closer to fair compensation, unlike now. With safer streets, you, your friends, your kids will go on walks or bike rides together that are too dangerous to try now. Kids will be able to take trips alone. People with disabilities will be able to cross intersections without worrying about needing to move fast to protect themselves from drivers. With a decline in car dependency, parking lots in cities can be used for infill housing. More affordable housing can be built for people who don’t want to pay for a parking space to be bundled with their apartment. These benefits would be widespread and significant.

Even though these policies are positive sum, helping society overall and making our world a less polluted place to live, not everyone will get a personal concrete benefit. If a given driver plans to keep driving the same car, never take trips by any other means even as more alternative infrastructure is built, and they hypothetically would never be in a crash under the status quo nor after these changes, maybe the only change they’d notice is paying a higher car insurance premium. Yet even that driver should want it.

Our car culture means ignoring so much pain. In a car dependent society with weak safety regulations, most drivers have to block out what their responsibility really means because otherwise the overwhelming risks would be debilitating. Improved insurance would not eliminate this. But these changes would significantly reduce the moral injury that American car culture imposes on drivers, and that’s a bigger benefit to them than a car looking cute or tough or whatever they spend so much money on already. Knowing that victims will be financially compensated is not a panacea for traffic violence. It’s not a ban on cars. But knowing that crash victims will get help is a morally and psychologically important benefit, especially for drivers who’ll never give up their keys.