Yes, for exactly the same argument. It seems like you understand the point?

(In practice, I like any car I ride to have seat belts. But I support the argument that since there's no externalities from other people not using seatbelts, we shouldn't force them to have some. And in any case, the orthodox way to deal with externalities is via a tax, not via a ban.)

> But I support the argument that since there's no externalities from other people not using seatbelts, we shouldn't force them to have some

That argument is wrong. There are absolutely externalities to allowing other people's car rides to be less safe for them, even if no one else is injured or killed because of it.

Every injury or death makes health care more expensive for everyone, even in a place like the US that don't have socialized health care. The costs of these emergencies are largely borne by health and car insurance companies, and those costs mean everyone's premiums are going to be just a little bit higher.

Also consider the effects on family (or friends) when someone dies (or is much more severely injured) because they weren't wearing seat belts. Maybe the breadwinning spouse dies or becomes disabled, and now the remaining family has to go on welfare. Maybe it was a single parent who died, and the kids end up in the foster system. In the disability case, insurance (which might be Medicare) companies will have to pay for rehabilitation and possibly care for the rest of the person's life. Hell, just the emotional anguish of a loved one dying in a situation where it was easily preventable is an externality worth trying to eliminate.

> Every injury or death makes health care more expensive for everyone, even in a place like the US that don't have socialized health care. The costs of these emergencies are largely borne by health and car insurance companies, and those costs mean everyone's premiums are going to be just a little bit higher.

Insurance is a bet between you and the insurance company. If you decide to engage in extra risky behaviour, then a competent insurance company will charge you extra. (Or rather, more PR friendly but equivalent: your insurance company will offer you a discount, if you can prove that you are exercising, don't smoke and always wear a seatbelt, etc.)

Realised average accident rates for certain parts of the population can help you fine tune your risk models, but they aren't mathematically required.

That's why insurance companies can profitably insure one-off events just fine. Eg have a look at https://en.wikipedia.org/wiki/Prize_indemnity_insurance for example where you can get bespoke insurance for your one-off novelty promotion. Golf course A making their golf holes wider won't affect the hole-in-one insurance premium that Golf course B has to pay.

Of course, that's all unless there's some regulation that forces insurance companies to set premiums a certain way. But then, blame that price control regulation.

An insurance company that prices risks accurately will outcompete an insurance company that solely relies on population averages. The latter will overcharge less risky people who drive with a seatbelt on (so they will move to the competition), and will undercharge risky people without seatbelts, and thus lose money on them.

> Also consider the effects on family (or friends) when someone dies (or is much more severely injured) because they weren't wearing seat belts. Maybe the breadwinning spouse dies or becomes disabled, and now the remaining family has to go on welfare. Maybe it was a single parent who died, and the kids end up in the foster system. In the disability case, insurance (which might be Medicare) companies will have to pay for rehabilitation and possibly care for the rest of the person's life. Hell, just the emotional anguish of a loved one dying in a situation where it was easily preventable is an externality worth trying to eliminate.

You are proving too much here. Yes, this argument could apply to driving without a seat belt. But it could apply just as much to any driving at all. Or to lazing on the couch instead of exercising, or to living in the New Mexico instead of Maine, or to drinking or smoking, or working as a lumberjack.

---

Just to be clear, on the object level I'm ok with smoking bans (instead of just high taxes) and laws requiring seatbelts. But that's just because they are convenient for me. They aren't properly justified.

> An insurance company that prices risks accurately will outcompete an insurance company that solely relies on population averages. The latter will overcharge less risky people who drive with a seatbelt on (so they will move to the competition), and will undercharge risky people without seatbelts, and thus lose money on them.

Only if the price of costing risk accurately is less than just using population averages. So they look at cheap ways to cost risk, like credit scores and driving records.

Just give people a check-box to indicate whether they use a seatbelt. (And use the rest of the contract to create legal consequences for lying.)

That's a risibly untenable idea for several reasons.

For one, the company presumably finds out you were lying when you got hurt or died. So you -- the person who made the contract -- are not the one who gets punished for lying.

Instead, it's your family who's punished, because they don't get the monetary support after you die or to help take care of you, and likewise society is, because if you're still alive you're likely bankrupt and therefore fall into social support systems.

Second, putting something in a contract doesn't make it magically happen. If the company wanted to recoup anything for you lying, they'd have to take you or your estate to court. That's both extremely expensive for everyone involved and also overtaxes the courts, an already overtaxed public resource.

The damage to individual freedom is negligible (the right to die by windshield strike is not well-recognized) and the damage to innocent parties and society is much higher.

I don't see how you can say it's not well supported with a straight face.

> For one, the company presumably finds out you were lying when you got hurt or died. So you -- the person who made the contract -- are not the one who gets punished for lying.

> Instead, it's your family who's punished, because they don't get the monetary support after you die or to help take care of you, and likewise society is, because if you're still alive you're likely bankrupt and therefore fall into social support systems.

Huh, that's exactly the same situation as lying to your life insurance, eg about prior conditions or whatnot. And they handle that just fine.

> Second, putting something in a contract doesn't make it magically happen. If the company wanted to recoup anything for you lying, they'd have to take you or your estate to court.

They'll just don't pay out. No need to recoup anything.

> The damage to individual freedom is negligible (the right to die by windshield strike is not well-recognized) and the damage to innocent parties and society is much higher.

There's no damage to third parties. The damage is approximately all to the guy who's dumb enough to not wear his seat belt.

Every person who puts themselves at risk in the military, fire department or other rescue type situation does so because life/death outcomes of other people are not considered neutral events.

All our lives are interconnected and interdependent. Most untimely deaths will leave a financial, practical and mental health crater in many lives around them. When children are involved the fallout can last generations. Not to mention the lost investment of parents, the education system, etc.

That would be a good argument for not allowing people to make any independent decisions at all.

Compare and contrast https://www.astralcodexten.com/p/bad-definitions-of-democrac...

> That's a good argument for not allowing people to make any independent decisions at all.

No it isn't. Very false dichotomy.

The benefits of doing something have to be balanced against its costs. People's longevity and freedom both matter.

And something impossible can't have valid arguments for it. No independent decisions is an unattainable scenario even for maximalist autocracies.

What, that I actually said, do you disagree with?

> No independent decisions is a completely unattainable scenario even for maximalist autocrats.

Sure, but you could still try to minimise independent decisions as much as possible.

Have a look at the article I linked for a more eloquent argument.

Except nobody is actually trying to do that, so the argument is irrelevant.

This is wrong. If you don’t wear a seatbelt and you get into an accident, your whole body becomes a projectile. One persons dumb choice to not wear a seatbelt can have potentially fatal consequences to others riding in the car who are not responsible for that decision.

The energy has to go somewhere. So if you are not wearing a seatbelt and the car flings you out like a projectile, the car itself will have slightly less kinetic energy, being slightly less dangerous.

In any case, even if you discount energy conservation, the extra danger to other people from you becoming a projectile is likely tiny. You can run some cost benefit analyses, and I'm pretty sure you'll come to the conclusion that a Pigouvian tax of something like a dollar a year is enough to offset this.

> One persons dumb choice to not wear a seatbelt can have potentially fatal consequences to others riding in the car who are not responsible for that decision.

Are you talking about people in the same car as the guy not wearing a seatbelt? Then you can exactly identify the other parties, so the transaction costs for Coasean bargaining are very low. See https://en.wikipedia.org/wiki/Coase_theorem Basically, if you think it's dangerous to ride along with people who don't wear a seatbelt, then don't ride along with people who don't wear a seatbelt.

As a pedestrian you can't opt out of what drivers are doing. That's a real exernality. But as a fellow passenger, you know exactly who else is in the car, and you can refuse to ride with them.

If I understand your suggestion right, it's like marrying someone with bad breath, and then asking the government to make a law to make your spouse brush their teeth?

Not everyone riding as a passenger in a car has choice in these situations, namely children.

People are already allowed to eg feed their children unhealthy food etc. That's usually not classified as an externality.

Putting yourself and a child in a flying mechanical deathtrap and you're not wearing is an unnecessary risk to their life in the same way that leaving guns loaded and unlocked in a house is. These are differences in the magnitude of risk and a large enough difference in magnitude becomes a difference in kind.

That's mostly an argument for the law requiring children to be buckled in at most, perhaps.

> the car itself will have slightly less kinetic energy, being slightly less dangerous.

Extremely slightly less. What, 170lbs compared to 4,000lbs+? Is 4% weight reduction after the collision starts, minus the energy imparted to the seats and other passengers and the windshield and what not, really going to make much impact to the overall collision calculus here?

Yes, this won't have much of an impact. But neither is there much of an _additional_ danger to other people from you flying around in an accident.

There is, because people can and do go through glass. Sure, most of the time this means that their brains are just spread on the highway like nutella. But they could also go through YOUR windshield.

I'm trying to read your comments charitably and made in good faith but I'll admit it is proving to be a challenge here.

In the UK they ran a campaign to raise awareness that during an accident a passenger riding in the back can injure the passenger in front if they don't wear a seatbelt. That seems like a serious externality to me. I don't know if you just hate seatbelts or love arguing but there are absolutely reasons that people should wear seatbelts and I'm glad most people have accepted that the inconvenience of wearing them is pretty negligible now.

https://m.youtube.com/watch?v=mKHY69AFstE&pp=ygUeVWsgc2VhdGJ...

https://www.reddit.com/r/YouShouldKnow/comments/359am5/ysk_i...

It's not an externality if it's in the same car.

If you are a driver, and you want people in the back to wear a seatbelt, just ask them to wear a seatbelt. Duh.

> I don't know if you just hate seatbelts [...]

Why would I hate seatbelts? I wear them all the time, whenever I have to take a car. I just don't buy the usual justification for forcing other people to wear seatbelts (or have AM radios..)

> [...] and I'm glad most people have accepted that the inconvenience of wearing them is pretty negligible now.

I'm all for people wearing seatbelts, too. Just like I'm in favour of people eating their vegetables and flossing their teeth. Voluntarily.

Huge externality for not using seat belts. When your dead body closes the highway for several hours while they scoop up your carcass and do the required investigations. Road fatalities always take longer for first responders than simply injuries.

You can cost it out, and charge people an extra 5 dollars (or whatever the figure ends up being) in annual taxes for that externality.