State-contingent insurance markets in everything

Fortunately, when you purchase a low-powered car from Young Marmalade, the free installation of a black box can cut your insurance premiums into half. By monitoring the driving behaviour such as acceleration, braking, what time of the day the car was driven and at what speed, Young Marmalade provides affordable telematic insurance premiums.

It’s simple. The black box data is used to calculate premiums, if the car was driven well, the lower the premiums will be and vice-versa.

According to MD Crispin Moger, “A young male driver who has Intelligent Marmalade in his car will pay on average £2,601/year for comprehensive insurance, while a young woman will pay £1,642/year. If the black box detects a bad driver, initial premiums will be subjected to a £250 and £500 increase or Young Marmalade will cancel the policy.”

Here is more, and for the pointer I thank KM.


Insurance premiums in UK seem crazy high; the article quotes figures of something like £8000/year for young male drivers. What's the median wage of a young male? How can anyone afford a car!

I hope the premiums come down a lot with age? I remember my premiums in the US were about $1500/year when I was in my-mid 20's, admittedly on a low-value used car.

Sounds about right. There is a reason car ownership is lower there.

I've been trying to find a cite for lower car ownrship/use in teen years means lower employment. If you can't drive to work...

Why are premiums so much higher? Is it replacement-cost liability? Doubt it. A baseline Prius (say) is selling at £21,000 (UK) versus $24,000 (US). That'd only explain a 1.5x premium rise.

American courts are notorious for high liability judgments so it is a bit hard to imagine UK-liability-risks are spiking costs.

What is it then? Higher theft rates? Risky roads? Or just the British industry making a killing on high margins?

I haven't been able to find American statistics for all accidents, but comparing British vs. US fatal accidents per 100 million miles, the UK does quite a bit better. See and Could be theft rates or that the markets are too regulated to enhance competition. Here's a shot of Aviva's financials (though they do a bunch besides insurance):

The fatality number is probably driven by the much lower average speed on UK roads. It's not clear that insurance costs are driven by fatalities - they might be driven by the much more numerous fender-benders.

I'm at a loss to explain the higher rates, presuming they are real and representative. It would be interesting if someone who knows could comment.

They are making a killing, just like they are the US.

You're supposed to lose money on the underwriting, and make money on the investment side.

Having just been backed into by a female, rear ended by a female, the victim of a red-light running by a female I can tell you statistics don't define individuals. I can't remember being hit by anyone but females. Same thing as healthcare, if you don't use the service then a rebate should be issued. It's not the best way, but it's not as discriminatory.

IIRC, females are in many more fender benders, but fewer very expensive accidents.

Makes sense. But if the insurance company is so sure it knows exactly which behaviors make drivers safer, then why the price disparity?

Assuming your story is representative, it means that, if you're a female, you're 3x as likely to have your car damaged in an accident involving a male than vice versa. Who's the real menace here.

As an aside, with free movement of EU citizens have car prices become essential uniform? Any differential should be arbitraged out by a quick drive or do the nations have a way of clamping down?

They may have higher minimum liability limits.

My understanding is that insurance premiums in the UK have been driven up recently in part by an increase in whiplash claims; see

for example. When I first drove (aged 18, 3-4 years ago), the extra insurance cost to cover me as a second driver was around £1,000/yr; to insure my brother now in a similar situation would cost over £2,000/yr.

Those prices are completely nuts. We currently pay about $600/yr (in Michigan) to insure the old car that our kids drive.

It's surprising to me that higher prices for males are legal in the UK at all - that's blatantly discriminatory. Atleast unless the same is true in reverse - do males get higher pensions due to their statistically shorter lifespans ? Do they pay less in health-insurance based on their statistically likely lower consumption of health-services ?

Things that are discriminatory against males are legal almost everywhere including the US. Males are not a legally protected class.

A good example: In most states, charging women more for health insurance is illegal. Charging men more for auto insurance is legal.

"do males get higher pensions due to their statistically shorter lifespans "
No, that would be illegal discrimination.

"Do they pay less in health-insurance based on their statistically likely lower consumption of health-services "

In most states/countries that would be illegal.

Well, technically they are a legally protected class- it's just discrimination on the basis of sex (without a good reason) that is illegal. That's less likely to be enforced for men, though. Actually, men certainly do pay less in the U.S. for health insurance due to their lower consumption.

Yes, in most places individual policies are cheaper for men than women, due to lower consumption (with the same overall health condition and history; men do still die earlier). You do occasionally see "investigations" and political campaigns against it.

The practice of charging men more for auto insurance or life insurance rarely comes up politically, though.

It depends on the state. In Texas the discriminatory pricing is legal. In some states its not.

I lived in CA and paid less in health insurance because I was (and still am!) a male. I daresay if it's legal in California it's legal neigh-on anywhere.

I'm not sure the current status of the law, but the EU is moving to make this sort of gender discrimination illegal - at least until recently, (young) men paid more for car insurance, and also got better annuity rates due to their statistically shorter lifespans.

>do males get higher pensions due to their statistically shorter lifespans?

That's not the way (corporate - as opposed to public) pensions work in the US. Basically, you're getting a single-life annuity, and the cost to the pension fund is lower for males than females. If you are allowed to get "survivor" benefits for your spouse (not all pensions do), then your pension payments go down a bunch. If your spouse is more than 10 years younger than you are, then your pension payments (for the dual-life, or "joint and survivor", annuity) go drastically down.

Half a century ago, 40% of the workforce in the US would be covered by a pension during their career. Today, less than 10% of the workforce has a pension, and most pensions are "frozen" (if you have X years of credit, you'll never get X+1 years credit). About the only folks with pensions these days are members of a union or work for the government.

>Do they pay less in health-insurance based on their statistically likely lower consumption of health-services?

Except for insurance covering pregnancy, no. And that's because most men don't go to a doctor until it is almost too late. This means that things that could have been easily treated if caught earlier, are now very expensive procedures (think cancer or chronic illnesses such as heart disease). Women seem to get trained to go to the doctor a lot more often, so things usually get caught at an earlier stage.

And the other half of your question is why insurance companies in the US are really unhappy about the "Affordable Care Act" (also known as "Obamacare") because it mandates that health insurance companies have to spend at least 85% of premiums on actual health care. When this kicks in (in 2014 if it isn't over turned beforehand), it will radically change the cost of health insurance policies, and folks not spending a lot on healthcare will be getting large rebates.

As someone who takes 'liberties' with the speed limits I deplore the use of these devices.

Unfortunately, given that we'll eventually be classed under a "no black box installed" group, we'll offset the insurance losses with or without one.

Fortunately, hacking software that will make the black box give exactly the answers you want it to will be a simple download away.

Oh no! A small dog just walked on the road. If I brake hard, I won't hit him. But then my insurance will go up. Poor dog.

His loss is the White House chef's gain.

So many people get into accidents because they do something stupid to try to avoid hitting an animal. In many cases hitting the dog is the smart choice.

No, the smart choice is to be aware enough to see the correct out. It's a very rare situation that doesn't have an out, unless the driver put themselves in the bad position.

Too bad the dog isn't on the roof of his owner's car

Progressive Insurance has been doing this in the US since 2008 or so.

According to Western zeitgeist:

Men subsidizing female birth control = a fundamental right of women

Men paying higher insurance premia = the price men pay for their reckless choices and dispositions

The comments above seem to say men actually pay *lower*.

Higher auto insurance, I was referring to.

So what?

It's "what's yours is mine, what's mine is mine" on a grand scale.

Except I'm a guy and I favor allowing auto liability insurance carriers to charge higher premiums based on gender, but don't favor alllowing health insurance carriers to charge higher premiums based on gender. Perhaps I suffer from false consciousness? Maybe I hate you and want to steal your money even if it means that I also suffer?

Alternatively, it's simply what many believe to be fair and reasonable, and because they believe that the two issues are not actually the same?

At least we know that our votes will cancel out.

It is precisely my point that some find it fair and reasonable to screw over a group of people for a more socially/politically favored one, with disregard to principles in favor of emotions.

So, according to you, people with whom you disagree are emotional and not principled, whereas you are principled and dispassionate. Well okay then.

I don't need to claim myself to be principled to believe that the ad hoc whims of political correctness shouldn't dictate whether insurance companies are allowed to discriminate in a given situation.

Rather, insurance companies should either be allowed or not to treat people differently based on actuarial data.

Nice strawman, though.

Yes, it was wrong of me to make a strawman based on reasonable inferences from the plain and unambiguous language of your previous post. I too hate it when my words are construed to mean exactly what I meant after I realize my words made me look like a sad little man raging impotently against the world.

I do like your belief, undoubtedly sincerely held, that it is "political correctness" - that great hobgoblin of conservative minds - that is the reason why we disagree. Rest assured, if women were more likely than men to cause high-value auto accidents, I would give not a single fuck about women paying higher auto insurance premiums. It's almost like paying higher premiums for causing harm with automobiles is sort of different than paying higher premiums for being a woman.

In fact, there are plently of rules based on public policy that restrict freedom of contract in the insurance setting in order to achieve goals society deems to be more valid than freedom of contract - does it make you mad that indemnification for wilfully causing harm is prohibited? (Okay, it probably does.)

You sound like you have an axe to grind, sputtering aspersions against an imaginary enemy.

I happen to agree men should pay higher premia. People should have to pay higher premia if they're statistically more likely to incur higher costs. So thus I believe birth-control consuming women should internalize the costs of their birth control, whether it be cash OTC or through commensurately higher insurance premia. Which seems to go against the mainstream, with lipstick feminist pundits clamoring for "free" birth control to little opposition. Insurance should be a transfer of risk from insuree to insurer, not a transfer of funding obligations from insuree to insuree.

So yes, your presumptions were unequivocally wrong, your attacks misplaced.

I assume you also think it is outrageous that men receive erectile dysfunction medication free of charge from many health insurance programs, without having to internalize the costs.
I assume you are also outraged that the validity of receiving ED medications is not even considered to be part of the debate, since the national political debate is controlled almost entirely by men.
I eagerly await your posts on this topic.

As you tacitly acknowledged, many do, but many don't cover ED. I would indeed think something's amiss if older men don't have commensurately higher premia as a result for policies that do cover it.

Men are responsible too for white-knight politics.

I eagerly await subsequent smug, sarcastic follow-ups from you on this topic.

You find those situations to be equivalent? Your framing is off by an absurd margin. Or does this misdirection profit you?

Ah, the ad hominem comes out.

With smartphones getting ubiquitous how long before a health insurer uses geolocation data to give premium rebates for walking enough or avoiding known bar coordinates or hitting a gym three times a week? Maybe even penalize late nights outs based on accelerometer data.

They'll plant a chip in your butt first to monitor your activities--eating, sleeping, excercize--first.

They will even monitor whether you attend restaurants with beautiful women as a proxy for the quality of the food.

Some health insurance plans will already pay for a portion of your gym membership - but only if you use it x times per month.

Appalling that safe drivers are free-riding on the higher premiums of the less safe. Appalling.

Is there anything barring insurers from offering package deals. For example, life insurance plus health insurance. That way, men and women pay the same prices and have the same costs?

All insurance faces a tension between risk-distribution and risk-identification. This is clear if one imagines the two limit cases.

1. Imagine an (unregulated) insurance market in which the seller has perfect information about the claims each buyer will ever make. Clearly, the cost of insurance will effectively be identical to the (time-and-interest-rate corrected) cost of the risk, meaning that no individual will benefit from buying insurance: they might as well self-insure, because their costs will be the same or a bit lower (no administration and profits to cover). So, insurance in a state of perfect information does no risk-pooling, and is therefore not very useful.

The closest real-world analogy to this is life insurance, since the insurers have perfect knowledge that everyone insured will in fact die, sooner or later. The variance in cost is all about investment growth and time uncertainty. Under these circumstances, there's a modest value to having life insurance in specific circumstances, but it's often not worth buying -- it will cost more than simply saving the same amount of money. Again, time shifting is the main value of carrying it: an earning parent may want to guarantee income until her children are 21, for example, and thus will pay enough to fund insurance for her (probably fairly low) risk of dying before that happens.

2. Now, imagine a situation in which risk distribution is completely unknowable (though aggregate risk can be calculated). Under these circumstances, insurance, if available, is likely to be at a flat cost. An example I recently dealt with was flight-cancellation insurance. Since the selling company asks no individualized questions about the flyer, they can offer the insurance at a flat cost based on the fare. They know how many people who buy this insurance cancel and make a claim, and they may even know roughly how many on any particular route do so, though that data is probably pretty low-N and noisy. So they can flat-price the insurance. (It's also quite expensive, meaning that only those with substantial risk of cancelling are likely to buy it -- a potential death-spiral, or at least as serious limiting factor).

When considering such mechanisms as the post described, one has to ask how much it moves the offer of insurance toward the perfect-information end of the scale, and thus makes the potential value of the insurance lower for everyone. (Note the cherry-picking built in, as well: drivers who appear risky in the black-box measurements will have their insurance cancelled, while demure cautious drivers are most likely to sign up). The demure drivers will benefit (but the insurance will serve more as a catastrophic-risk buffer then a risk-adjusted value proposition for them), while everyone else's insurance is likely to go up because the dropped drivers will drive up general rates, as will the presence of more likely-risky drivers in the non-black-box pool overall.

+1. Now if you could just explain healthcare and insurance reform to the class.

What about asymmetric? Could we foresee a situation where the insurer (via his black box and data archives) knows more about a persons driving risks than the person himself? The approach to perfect information can be one-sided.

Isn't the long-run equilibrium that the riskiest drivers simply stop driving entirely, thereby lowering the insurance costs for everyone? (Albeit at a serious inconvenience for the bad drivers)

We've been using this service from Progressive on our new car for about two months now. You can go to their site and see reports and graphs showing your speed at the time of day. It also identifies 'hard brakes' (decreasing 7mph in a second or less). That's it and we got a discount for agreeing to it and then after the first 30 days we got another discount (I think 12%) based on its record.

It has caused us to alter our behavior. We've used the new car less than we otherwise would have and try to be careful about hours we take it out (between 12 midnight and 4 AM is marked as the most dangerous time). We've also tried to avoid the hard brakes but always seem to get one or two. The device doesn't record your location (or if it does they don't share it with you).

I think it would be nice if they gave a bit more feedback. If they showed us how we compared to other users it would probably encourage more safe behavior. Also I'm not exactly clear these things actually track accident risk all that well.

I have the same system. During the calibration runs I've never run as many yellow lights -- to avoid the hard brake.

I do think the 12-4 thing makes a lot of sense.

My system didn't quite work -- for six months it only recorded 5 minutes of driving. However, progressive only offered a 27% discount. What do you have to do to earn the max 30% discount?

I really want to find out that these boxes are fake and don't measure anything. It would seem that the placebo / beneficial selection value of signing up for the box is 90% of the benefit to the insurer; the actual monitoring seems like more trouble than it's worth.

I wonder how high my premium conceivably could go if I hired a stunt driver to drive in my place while I had the black box in the trunk.

I wonder what the max is and what is the price, if there is one.

If this gains widespread use, I predict it will be sued on the basis of discrimination by race/gender.

It'd be even funnier if it'd be possible to win a judgement against the insurer for knowingly allowing a bad (drunk?) driver to continue driving and cause an accident. Could an ingenious lawyer wrangle this into some sort of vicarious or secondary liability statute?

If you have information on how a person drives, what is the purpose of charging men higher rates than women?

I thought the reason behind that was men are statistically more likely to be involved in accidents because they drive more aggressively. If you have information in the individuals driving habits then there reason for that assumption no longer exists.

Comments for this post are closed