Two scams, both bad

Good piece in Sunday's paper on the PPR -- the poor people's rate, or the premium that poor and working classes are charged over and above what the rich charge one another for the same goods and services.

If that sounds to you like contentious rhetoric that overstates the case, then please read the article: "Lower income can mean higher rates."

Insurers often factor in a driver's occupation and/or education level when setting rates.

The practice is legal in Delaware and all but a handful of other states.

But critics charge there's no valid reason why a lawyer should pay less for insurance than a waitress with the same driving record, and that the practice results in minorities and low-income drivers paying more. And the difference in rates is more than spare change.

A 2007 report by the Florida Office of Insurance Regulation found the premiums one company charged Florida drivers varied by as much as 200 percent, depending on a driver's occupation and education level.

The department found that under some plans, drivers "with more professional occupations (doctors, lawyers, architects), and advanced college degrees" received preferred rates while blue-collar workers and those with a high school education paid more.

"It's not obvious to me why a clerk is a worse driver than a CPA," said Robert Hunter, director of insurance for the Consumer Federation of America.

Let me repeat that one line:

A 2007 report by the Florida Office of Insurance Regulation found the premiums one company charged Florida drivers varied by as much as 200 percent, depending on a driver's occupation and education level.

The PPR, for auto insurance, can be as high as 200 percent. That's hundreds of dollars. Every year. Hundreds of dollars charged to people who, by definition, do not have hundreds of dollars to spare and charged because they do not have hundreds of dollars to spare.

The game is rigged.

* * *

John McCain seems to have forgotten the First Rule of Holes when it comes to his goofy proposal for a "gas tax holiday."

The Arizona senator's Big Idea is to help Americans at the pump by diverting an additional 18 cents of the price of every gallon of gas to the profits of the oil companies. The effect on what we're all paying to fill our tanks would be negligible -- far less than 18 cents a gallon. But the side effect -- an $8-$10 billion windfall for the oil companies taken directly from funds needed for highways and bridges -- would be six different kinds of bad. That side effect is so vastly disproportionate to the purported aim of this policy that it's difficult not to suspect that this side effect is really the proposal's primary purpose.

You'd think at least one of McCain's advisers would point out that transferring billions of dollars from the public coffers directly into the pockets of ExxonMobil and Royal Dutch Shell seems like a dumb thing to do while running for office. Trying to spin such corporate welfare as help for working families just adds insult to injury.

But McCain is defiantly proud of his hole and he's determined to keep digging. Economists have uniformly panned McCain's proposal. In the words of MIT's Joseph Doyle, economists "are as close to unanimous as it gets in viewing the proposal as a horrible idea.” Undeterred, McCain struck back: "I'm not going to put my lot in with economists," McCain said.

What more evidence do you need that this man is simply a carbon copy of George W. Bush? That McCain would stoop this low, cloaking himself in populism while sneering behind his hand that the people are too stupid to realize his shell game won't help them, just his corporate masters, just goes to demonstrate that ...

Wait, what? It was who?

You're kidding.

Well, that's just ... it's ...

Is there a word that conveys both extreme disappointment and outrage? Because that's the word I need here.

Comments

[Rosina]

Do no claims bonuses exist in the US insurance industry? In time, a poor person with an excellent driving record will pay much less than the full mark-up - possibly even find a better insurer - while a poor risk, wealthy person will find their insurance costs rocketing, as individual factors take over from the statistically general. Otherwise, I agree. As a low risk (despite those three claims in one year) I would not approach an insurance company which offered cover to those drivers not otherwise able to get insurance. I would prefer to ally myself with other low-risk drivers, in the hope that that would keep the premiums low.

When this was raised in the UK, quite a long time ago, the occupations considered high-risk did not map directly to wealth or education - if I remember correctly, journalist was high risk, as was being a furrier.

[cosmicdancer]

including intelligence, analytical skills

You seriously just said that. An awful lot of poor people are not poor because they're stupid or lack "analytical skills". I'm a high school and college drop out on disability and I've had fewer car accidents than my sister or my parents. Who all have college degrees and are employed.

I-I have to go away now and rant about this in some other blog.

[Lee Ratner]

Why would being a furrier be considered a high-risk occupation? The only danger I can think of are assaults from passionate animal rights people.

[hapax]

Other people have already pointed out that Mr. Clark totally ignored the question of whether the factors insurance companies base their premiums on corresponds to the likelihood of accidents. I suspect that's because Mr. Clark knows that they do.

And I suspect tht he ignored it because he not only doesn't know, it is IMPOSSIBLE for him -- or anyone else -- to find out.

Insurance companies consider their statistical algorithms to be corporate secrets. They will not give anyone the data they use to compute their rates -- not even state regulators. When California threatened to force insurance companies to open their books, the insurance companies retaliated by threatening to pull out of the state -- and the state legislature folded.

The best that anyone can do is try and extrapolate what these companies are doing by comparing the rates offered to different people. When investigators notice a pattern -- for example, people of one ethnic group being consistently charged higher rates than others -- they can use this information to try and make a legal case of discrimination (if the factor is an illegal one).

Supply and demand does not work in the insurance business any more than it does in the gasoline business. People cannot simply refuse to purchase car, health, or house insurance if the price is too high.

But your touching faith that the insurance companies are simply fairly computing rates on the basis of purely disinterested risk calculation is on par with Clinton's that the oil companies will pass any tax savings along to their consumers.

[Rosina]

Even if a poor person were 3 times as likely to get into an accident as a rich person (which I doubt), shouldn't the fact that the poor person is likely only totaling a $5000 used car vs. the rich person totaling a $25,000+ new sedan count for something???

Even if the insurance isn't just for 3rd party insurance, the major potential cost of claims is not related to the cost of the vehicle. A $5,000 car can cause as much mayhem as a $50,000 one. And since poor maintenance will factor into a number of accidents, the used car, owned by 'a poor person' may develop a number of faults which make it an accident waiting to happen. And that accident - the tyre blowing (couldn't afford a new one) or the brakes failing (missed the last check-up), or the entire steering system locking up (that was checked by cousing Terry) - could lead the £5,000 car to crash into a lorry carrying goods worth amillion, or even cause a train to derail, with multiple death and enormous costs to be borne by the insurance company. Only when we move on to fre, theft and fully comp does the value of the car become relevant in setting the premium.

The thoughts about cheap or ignored safety checks and repairs might go someway to explaining any waiting on the poor.

Why would being a furrier be considered a high-risk occupation? The only danger I can think of are assaults from passionate animal rights people. The article didn't explain. It might have been attacked by activists, or a tendency to carry furs in the cars, or just one very bad furrier driver, who warped their statistics. It was just an "isn't it strange" article, with a touch of warning about insurance companies.

[Ryan]

£5,000 car to crash into a lorry

whoa, we moved to britain all of a sudden

[PurpleGirl]

Re: Co-Op or not-for-profit insurance companies.

There once were such entities. They were called mutual companies and the policy holders were the owners. Many of the current for-profit insurance companies began as mutual companies -- you can tell from the name or what the prior name was. MONY was Mutual of New York (I think it changed it's name completely), Axa was a mutual, Mutual of Omaha, might still be, Lutheran Brotherhood was/is a mutual. I may be off in naming examples, as I'm going on memory, but the point still stands. After the manangement was paid, the profit was returned to policy holders in lower rates and/or as dividends. But a number of years ago, they started to be changed over as management did buyouts, etc. to make the companies for profits. (The medical field did the same thing.)

It rots.

[Tonio]

Isn't the whole argument against the market a claim that the smart people in DC know better...

No. The argument isn't an either-or proposition. One can point out that placing profit and stock price above everything else can have destructive consequences, while at the same time pointing out that government intervention in markets can have equally destructive consequences. In some cases, people do best when corporate power and government power balance each other out, neither one having an advantage.

[PurpleGirl]

A little research finds:

MONY has been taken over by Axa-Equitable (as it's now known, having put back in the Equitable because no one knew what Axa was supposed to be).

Axa-Equitable's former name was Equitable Life Assurance Society of the U.S. Axa-Equitable is a for-profit; Equitable Life Assurance Society was a mutual.

Mutual of Omaha still seems to be a mutual.

Lutheran Brotherhood also seems to still a mutual benefit society but it's changed its name to Thrivent Benefit Society.

New York Life Insurance is still a mutual. On their "about us" page they have a good explanation of what a mutual company is.

And, no, I don't work in insurance, but some commenters were talking about non-profit or co-op insurance and I know a bit about it from policies I've had over time.

[SweetCraspy]

Thanks for the info PurpleGirl. It's good to know that the model is around and working in some form. I guess real life is intruding and keeping it from being ideal, though. And it's sad to see that some don't last. It just seems like it could be a little piece of socialism working really well if it were set up right.

[Rosina]

£5,000 car to crash into a lorry

Ryan: whoa, we moved to britain all of a sudden

Congratulations. I'm sure you'll enjoy it here, where tax cuts for the middle income group are funded by taking it from the poor, the old and the single men in barracks.

[randlebrot]

And I suspect tht he ignored it because he not only doesn't know, it is IMPOSSIBLE for him -- or anyone else -- to find out.

Bullshit - mathematically correlating one stat (credit rating) with another (accident rate) is well within the range of possibility.

Ryan: whoa, we moved to britain all of a sudden

OK, replace 'lorry' with "school bus full of children". The point is, an auto accident can damage more than the driver's car, so how expensive a car that driver can afford isn't the only issue. The insurance is for the damage you do to others as much as to yourself.

[randlebrot]

close italics

[Hawker Hurricane]

Rosina, Single men in Barracks aren't no plaster saints.

[hapax]

mathematically correlating one stat (credit rating) with another (accident rate) is well within the range of possibility.

Why yes, it is. And is probably done. That is not the point.

The point is that neither you, nor I, nor Fred, nor government regulators, nor ANYBODY outside the company is actually making and using that particular correlation. Why not? Because it is considered proprietary business information, and they won't tell anybody.

We can suspect that they are. We can look at the results of rates offered to different persons and extrapolate that it is statisticallly likely that they are. But we don't KNOW what they are doing.

So, once again, your assumption that the insurance companies are using a fair, relevant, and legal correlation is exactly that -- an assumption, that depends on your oft-demonstrated blind faith that Teh Market can do no wrong.

[inge]

Especially since no matter what your occupation and education level, you have exactly the same likelihood of getting into an accident!

We've been through this before: Ratings rate the company you keep. I share two characteristics with young drivers (1. I do not have a garage; 2. I drive an inexpensive car), and as young drivers cause a lot of accidents, I pay high base rates on insurance. The facts that I'm not young and have never caused an accident are fortunately figured into the equation somewhere, too.

Correlation is not casuation, but insurance mathematics couldn't care less. The results suit them well enough.

[David]

Are you saying that the not-for-profit could afford to offer the lower rate to the middle class person, but that it can't know enough without unethical data mining? What if the not-for-profit uses the same statistics (ethical or un, so long as they are legal) that other companies use. Not the ideal case, but wouldn't it still be lower cost for everyone?

Eh, not exactly. You're probably correct in saying that a co-op could use unethical data analysis in order to still do a better job of providing for pretty much everyone. The primary obstacle then would be that even if the data was unethical, you would still need to ethically pay out reasonable claims -- insurance companies save a lot of money by avoiding doing that, thereby lowering premiums, which would attract many people even if it wasn't in their long-term interest.

Neglecting that aspect of the competition, though, I'm suggesting that offering the lowest rate possible is itself unethical in this case.

Here's the basic scenario (my example may be a little off, I don't know the actual statistics, but the principle is the same). I start an insurance company, and do a lot of data analysis. I find that the absolute lowest-risk group of drivers I can find is upper-middle-class white single childless women with good credit and a steady job who work in the suburbs, have short commutes, and only drive brand new cars which they sell after two years. Feel free to add more qualifiers. I decide that my insurance company will cater exclusively to this segment: if you belong to this hyper-specialized group, then on average you are very unlikely to file any car insurance claims. Thus, I can offer you extremely low rates, and still make a profit. The only way this strategy works, however, is by raising the average rates for everyone who is not an upper-middle-class white single childless etc.

Now, my rates for all those other groups might be absurdly high, but because my rates for this particular group are so low, many people in this group will come get insurance from me. Maybe beforehand they used an insurance co-op -- but now my commercial insurance company is siphoning off the rich, low-risk drivers from that co-op, forcing the average rates for the co-op to rise.

I'm suggesting that it is unethical to provide absolutely minimized rates for a privileged portion of the population at the expense of the rest, when the distinction is in no way based on relevant (and by relevant I don't mean statistically correlated, I mean actually how you drive) personal behavior. But companies that use that strategy can always undercut ones that don't for the low-risk drivers, so even companies that try to be ethical will be forced to charge higher rates based on statistically correlated but individually irrelevant categories.

[David]

@ randlebrot:

It's immoral for a business to predict its costs? That's a new one.

Two responses:

1. You're misrepresenting what I said. I will give you the benefit of the doubt and rephrase: I never said it was immoral to predict costs. Insurance will always be about predicting costs, even if it's funded by income tax and run centrally by the federal government. That's just what insurance is, and has nothing to do with market pathologies.

I said it was immoral to use those predictions to raise rates for non-privileged groups based on criteria that are statistically correlated with, but not actually indicative of, bad driving. Non-whites may statistically make more car insurance claims. That doesn't mean it's okay to charge non-whites more for insurance. It is possible to be aware of the correlations (predict the costs) without reflecting that in unethical fee schedules.

2. Actually, in the sense I meant it, it's a very old one. You might want to look through the Old Testament business regulations sometime, to see how amazingly anti-capitalist they are. (Or maybe I should say anti-libertarian, since capitalism is a spectrum on which American culture prefers a particular extreme.)

Fred-style disclaimer: I'm not suggesting a priori that modern law should be based on Old Testament law. I would at least suggest, though, that in cases where Old Testament law is more merciful and humanitarian than our own system, we should take a hard look at why that is.

[inge]

Lee Rathner: Why would being a furrier be considered a high-risk occupation?

Maybe it's a profession with a lot of over-sixties people. Or it's a very small group with one member who drives really badly. Or they have to drive a lot in unwieldy cars. Or it's a profession with very few women.

The causation behind a correlation can be extremely strange.

[David]

One more thing:

If it's immoral to base other decisions on someone's wealth, is it equally immoral to base means testing of govt social programs on someone's wealth?

The social programs you're referring to are there to (try to) ensure that people who are in the most dire need can still obtain food. It is reasonable, in trying to help people, to in some way evaluate who most needs that help.

Insurance is exactly the inverse. The evaluation here is to ensure that... err... corporate executives maximize their profit. In the former case, the evaluation is to maximize benefit to the person being evaluated. In the latter case, it is to maximize benefit to the evaluator which, on average, has an inverse correlation with benefit to the person being evaluated (some people benefit -- lots of people don't).

Now, if you were in fact trying to draw a more direct metaphor, and suggest that it should be legal for insurance companies to evaluate financial need in order to lower rates for individuals who most need help.... or if, in the other direction, you were suggesting we should evaluate the means of insurance executives, and their salary should be based on their financial need... well, in either case, you may be on to something :)

[Rosina]

Non-whites may statistically make more car insurance claims. That doesn't mean it's okay to charge non-whites more for insurance. It is possible to be aware of the correlations (predict the costs) without reflecting that in unethical fee schedules.

I suspect it is illegal in the UK to ask about race in insurance applications, and I've never been asked about my education. Occupation, age, gender, where you live, where you park the car, how far you drive, and driving history (length of time driving, claims and offences) seems to be about it. The actual rate you are offered is based (I think) on age and driving history, with 60% or more off if you have no or very few claims. This no claims bonus is carried over when you change insurers, your old insurer being obliged to give you a formal record of your 'no claims' status, so any disadvantage from being a furrier could be wiped out by a careful driver.

Is there any reason to suppose that the insurance companies are using unethical sources for their weightings? Much of it could be done from the records of the company itself.

[Lila]

Re insurance co-ops, the National Fraternal Society of the Deaf was formed in 1901 because U.S. insurers wouldn't issue policies to deaf clients.

[Jeff]

whoa, we moved to britain all of a sudden

No, some of the Slacktivists have always been in Britain (more accurately, the UK, as was discussed recently). Rosina, Jesu and Praline, among others, live in the UK.

[balt]

As was pointed out immediately above that post, the entire point of insurance, from the customer's standpoint, is that high risk people are subsidized by low risk ones.

This is absolutely incorrect. The point of insurance it to manage the costs of a risk, not to subsidize the risks. The point of insurance is not to charge the same cost to person X who engages in risky behahior as person Y who doesn't.

The basic model for insurance is for a group of people with the same risks (such that they know that it is likely that one of them will suffer the harm, but not all) and have each of them pay into a common fund so that when the harm does occur to one of them, that one is not ruined.

Once we've established that basic model, the amount paid in can be varied by estimates of the variance in the risk posed. In the terms of how modern insurance originally arose, covering losses in shipping, a shipowner sending his ship to India would have greater risk (and a higher premium) than one send his ship to Antwerp. And one with a bigger ship has a greater risk than a smaller one. The owner sending his sloop to Antwerp shouldn't pay the same as the clipper to Bombay.

In a world where purchasing insurance in voluntary (and most of it is), a structure built on subsidizing higher risk persons would fail, because it's against the self interest of the low-risk putative insured to obtain insurance under that circumstance. If competition is allowed the insurer that does not discriminate will loose low risk customers to its competitors who do discriminate. If competition is not allowed, insurance will only be obtained by the low risk customers if they are required to by law.

There is an argument that we should use different tests for evaluating risk. But I think people would object more to subjecting oneself to an annual insurance checkup where they measure reaction time, intelligence, risk tolerance and attention span. Just a guess.

[David]

"Other people have already pointed out that Mr. Clark totally ignored the question of whether the factors insurance companies base their premiums on corresponds to the likelihood of accidents. I suspect that's because Mr. Clark knows that they do."

And I suspect tht he ignored it because he not only doesn't know, it is IMPOSSIBLE for him -- or anyone else -- to find out. Insurance companies consider their statistical algorithms to be corporate secrets...

Well, that's only a little bit true. I mean, certainly insurance companies jealously guard their databases and algorithms. But I suspect Fred does in fact know that insurance companies do exactly what was claimed here (and if he or anyone here doesn't know it, let me assure you it is the case): compute premiums so as to minimize their risk and maximize their profit. It's remarkably easy to do, the more so as the size of your database grows. It's a strategy used by insurance companies and hedge funds across this great nation (and others, I assume, but I've never looked at the job market outside the US): find some CS / math / statistics PhDs from top universities, offer them $500,000 a year as a starting salary, plunk them down in front of a database, and just plug in the numbers they give you. This will result in the kinds of situations we've been discussing, where premiums for some groups go up for completely unethical reasons, but you'll still be able to sleep at night on a bed made out of money.

Like I mentioned, I'm in CS, and the general feel at my schools has been that Google is a pretty lucrative place to work, with salaries reaching up to six figures. I remember the point at which I found a webpage of job listings from financial companies and hedge funds, hiring CS people with comparable experience, and actually offering salaries of $500,000 and even more. Those places terrify me, but they sure have a lot of money to throw around, and I'm quite certain they know exactly what their return is on that investment...

[Majromax]

Yes, but if you're pooling a higher level of risk than me, it's not a conspiracy against you if you are charged more. Other people have already pointed out that Mr. Clark totally ignored the question of whether the factors insurance companies base their premiums on corresponds to the likelihood of accidents. I suspect that's because Mr. Clark knows that they do.

Does it matter? Suppose that for unknown biological reason, there is a gene on the 8th chromosome who's presence highly correlates with traffic risk. Would it be okay for insurers to demand blood tests for all applicants (or conversely offer "discounts" to those who prove that they don't have the gene?)

A person's wealth is not easily a controllable factor. Indeed, it's easier to change religion (which is a protected class) than wealth, which is not.

It may be capitalist, but a system which rewards and punishes people for decisions they didn't make is unfair at best, undemocratic at worst.

[not someone else]

Aren't furriers the ones who go out and catch animals to be skinned, not just the ones who sew them into coats and attractive handbags?

I would suppose anything involving the wilderness and things with pointy teeth would be of some danger, if I'm not thinking of something else.

[SweetCraspy]

The primary obstacle then would be that even if the data was unethical, you would still need to ethically pay out reasonable claims -- insurance companies save a lot of money by avoiding doing that, thereby lowering premiums, which would attract many people even if it wasn't in their long-term interest.

Point well taken. Trying to be duly diligent in providing the service people are paying for is a disadvantage that I hadn't considered.

I decide that my insurance company will cater exclusively to this segment: if you belong to this hyper-specialized group, then on average you are very unlikely to file any car insurance claims. Thus, I can offer you extremely low rates, and still make a profit. The only way this strategy works, however, is by raising the average rates for everyone who is not an upper-middle-class white single childless etc.

I'm not sure why this is the case. It seems to me that in order for any customer to be profitable and worth keeping, that customer has to pay for their own risk (on average). Other wise it's better to drop the low-rates people and keep charging high rates for the high-rates people. The only reason that I can see for raising one group's rates in response to another group's rate cuts is because the former are funding the risk of the latter.

In my ideal co-op, everyone pays in as much as they claim out (again, all on average). No one funds anyone else's risk. High risk people pay high rates, even if they are poor. Low risk people pay low rates, even if they are rich. In this case, the purpose of the insurance is to give the customers a money pile to pay for their disasters. It's not even doing any thing the people couldn't do for themselves, i.e. stockpiling for a rainy day. It's just fulfilling the legal requirement of having insurance, so that the stockpile is guaranteed to be big enough to cover any eventuality.

I think I may be talking parallel to balt here, but I'm not sure. I got lost right when you got to the good part about stealing low risk people.

[David]

"The only way this strategy works, however, is by raising the average rates for everyone who is not an upper-middle-class white single childless etc."

I'm not sure why this is the case. It seems to me that in order for any customer to be profitable and worth keeping, that customer has to pay for their own risk (on average).

Well, the problem is the "on average." What groups do you take the average over? As pointed out, going to the extreme, each individual will claim, "on average," exactly the amount that they claim. That is, if your "average" is over each individual separately, and you have an effective way of determining that value, then insurance becomes worthless for everyone except the seller.

Here's how it works in my example. I'll use unrealistic numbers because I don't know the realistic ones.

Supose that, on average, individuals in the nation of Erewhon make $1000 worth of car insurance claims per year. (Of course, most people aren't making claims at all, but the average is brought up by the proportion of people who actually get in accidents, etc.) The insurance market is just getting started, so company A starts up, looks at that average, and decides to charge every individual $1200 a year for insurance. Insurance is mandatory, so everyone gets insurance from company A.

Now, company B starts up, observes that the upper-middle-class white female single etc. group makes, on average, only $50 of claims per year. They offer coverage to this group for only $100 per year. Everyone in this group immediately switches to company B.

Now, in Erewhon, that group makes up 1% of the population. This means that the average claim over everyone who is not white female blah blah is (1000 - 0.01 * 50) / 0.99 = $1009.60. Reacting to raising claims costs, company A raises their prices to $1210.

Now, so far that isn't a huge difference, but as company B slowly eats away more and more high-income low-risk groups, company A will be forced to raise its rates more and more, unless they too implement similar profiling techniques.

And the upshot is that it's not okay to charge someone for "their own risk" incurred by the fact that they are not wealthy, white, etc. Only for the risk they incur by a poor driving record. But unless regulations are in place to prevent it, that's exactly what will happen.

In my ideal co-op, everyone pays in as much as they claim out (again, all on average). No one funds anyone else's risk.

But the whole point of insurance is to fund other people's risk... you can say that you have a lot of money and therefore are less at risk of a car accident, and statistically you might be right, but insurance is collective pooling of resources to handle dangers that we have no (or very limited) individual control over. It is certainly advantageous (for corporations, and for lucky individuals) to exclude from the pool people who have uncontrollable risks to which you are not personally subject, but it's an uphill ethical argument to make since they are, after all, uncontrollable risks...

[hapax]

In a world where purchasing insurance in voluntary (and most of it is)

Huh? In what sense, "voluntary"?

If you I have auto insurance, I can't get a driver's license. If I don't have health insurance, I can't get non-emergency medical care. If I don't have house insurance, I can't get a mortgage.

I suppose I don't, strictly speaking, "need" any of those things, so in Libertopia they would be considered "voluntary." But, alas, I (and most people) live in the real world, not on Second Life, and the only kinds of "voluntary" insurance I can think of (life, pet, superhero damage) I don't own anyhow.

[hapax]

If you I have auto insurance

*sigh* That should be "If I don't have auto insurance" etc.

Never let Preview get in the way of a good rant.

[SweetCraspy]

But the whole point of insurance is to fund other people's risk... you can say that you have a lot of money and therefore are less at risk of a car accident, and statistically you might be right, but insurance is collective pooling of resources to handle dangers that we have no (or very limited) individual control over.

I think that those two things are different. There can be collective pooling of resources without sharing of risk (as long as the claims never bankrupt the resource pool). You said that that if everyone claims out what they pay in, then no one wins except the insurance company. The insurance company does win because they get paid whatever their markup is, but if that markup is kept as small as possible by a not-for-profit, then the customer isn't paying for nothing. They are paying into the pool so that they can have easy access to gobs of money when they have a disaster.

I think I'm looking at this as kind of an instant-loan safety net rather than insurance in the proper sense. You buy into the system so that you can access the big pile. However, if you make a claim, then you ultimately have to pay back the difference between what you paid so far and what you took out. Otherwise, the system will eventually bleed dry. I think it would work well for for one-shot disasters that you can recover from, but wouldn't offer as much benefit for things like chronic illness or truly gargantuan disasters that no one person could cover in one working life.

[David]

I think it would work well for for one-shot disasters that you can recover from, but wouldn't offer as much benefit for things like chronic illness or truly gargantuan disasters that no one person could cover in one working life.

I'd draw the line rather considerably short of "one working life" -- and I also think you might not realize what sort of events can qualify as a "gargantuan" disaster. Chronic illness is certainly something current insurance systems handle very badly, but even a single serious incident can quickly add up to medical expenses in excess of a year's salary (I had a minor cut on my hand last year, and if it weren't for insurance and my insurance company's negotiated fees, it would have cost two months pay to have them look at it and put on some disinfectant and bandages). Most people don't have that much money sitting around, nor can they raise it even over the course of several years. Fortunately, most people also won't incur those kinds of expenses very often, if ever, so by risk pooling, paying slightly more than the expected costs, it can be ensured that no one will incur unexpected costs so disproportionately high that it will ruin their finances and credit for the rest of their life (or even, one hopes, for several years).

So, I kind of see what you're getting at, but I'm not sure I believe in it as a model -- the "very unlikely but life-ruining disaster" is the real purpose of insurance for a lot of people, and if it was just a (rather high-interest) loan, a lot of people would be better off just getting a loan when those things happen -- zero cost unless something bad does happen, and hopefully decent interest rates if it does. Insurance, on the other hand, is about improving the worst-case-scenario for unlikely events that most individuals in the pool would be incapable of recovering from alone.

[balt]

Huh? In what sense, "voluntary"?

If you I have auto insurance, I can't get a driver's license. If I don't have health insurance, I can't get non-emergency medical care. If I don't have house insurance, I can't get a mortgage.

These are the types of insurance I have (roughly):

Auto liability insurance, $20,000 in coverage - mandatory to operate the vehicle I own (but not to have a license in the US)
Home owners insurance - mandatory per mortgage company
Health insurance - not mandatory in the literal sense, but practically
Professional liability insurance - mandatory

Auto liability insurance, excess coverage over $20k - not mandatory
Uninsured motorist coverage - not mandatory
Auto comprehensive coverage (covers costs to my car in an accident that is my fault, plus vandalism, etc.) - not mandatory
Home owners insurance covering more than just the house - not mandatory
Jewelry rider - not mandatory
Long and short term disability insurance - not mandatory
Life insurance - not mandatory

If I owned a boat or recreational vehicle, I could obtain insurance for that, but it would't be mandatory. I also do not have long term care insurance.

Most of those things aren't mandatory, and except for the minimums for the auto liability and the home coverage, even the amounts and types of coverage aren't mandatory. And I've totally excluded extended warranties.

[balt]

One other thought: if I didn't own a car and rented an apartment instead of (nominally) owning a house, I'd be down to health and professional liability insurance as mandatory. Neither car nor home ownership are required, even in America.

[SweetCraspy]

So, I kind of see what you're getting at, but I'm not sure I believe in it as a model -- the "very unlikely but life-ruining disaster" is the real purpose of insurance for a lot of people, and if it was just a (rather high-interest) loan, a lot of people would be better off just getting a loan when those things happen -- zero cost unless something bad does happen, and hopefully decent interest rates if it does.

Thanks for talking through it with me. It's an idea that I had in a, "Wait, why doesn't it work this way?" moment and it was very roughly and arbitrarly put together. I think I understand what both of us are talking about better from going through some of the details. Cheers for now.

[balt]

But I think we've missed the forest for the trees. My point was that people have choice in what they insure, and for how much, and when you have choice, there is differentiation in costs based on different priorities by both the insured and the insurer - unless there is a regulation prohibiting differentiation.

This isn't a value judgement, this is simply a statement of reality.

[Lee Ratner]

The people who catch and skin animals are called trappers, assuming that the animal is hunted rather than raised. Furriers were the ones who crafted the coasts and other articles from the dead animals.

[hapax]

My point was that people have choice in what they insure, and for how much

That may have been your point, but it was (as I stated above) flat-ass wrong. I have never purchased any of those optional insurance coverage you outline above. More to the point, they are not the ones people are talking about in the original post.

Yeah, I have a CHOICE as to whether I have a house or a driver's license or rely solely on emergency care. I don't *have* to live with my spouse, where we can both get jobs, where there are no rental properties to speak of or mass transit of any kind; heck, it was my *choice* to have chronic health conditions.

If I really *cared* about the insurance companies, I would just lay down in a gutter and die, and so increase their prophet margin. It's the capitalist way, innit?

[MercuryBlue]

No, the capitalist way is for you to stay alive and feed them money and not do anything or endure anything that would require them to give you money, then die in a manner that doesn't require your life insurance provider to pay your survivors.

Also, there's a really good joke somewhere in "prophet margin", but I'm damned if I know what it is...

[hapax]

It's the money they make by correctly predicting the future.

[inge]

The interesting thing about the insurance market is that it functions on lack of information. If everyone knew everything that was going to happen to them, there'd be no market for insurance, because it would not be bought by people who didn't need it, and not be sold to people who did.

With more information available by the year, this is exactly what happens. Insurers try to keep an edge by keeping their information private, but as the rates are public, information seeps out. Customers go for the best rates and make it impossible for our hypothetical mutual insurance to stay both beneficially ignorant and in business.

It becomes a lot more intuitive when you remember that the whole field of statistics (which insurance so heavily relies on) was created by and for gamblers.

[Rosina]

The complaint - and what makes it potentially a scam - is that the insurance companies have decided that the group (or one of the groups) that make a poor risk correlates with the same factors that lead to low pay. If the insurance companies were wrong about this, you would see other companies entering that niche market (possibly African American/Hispanic drivers who did not go to college, or who have not held a regular job for x years) and able to offer lower rates than their 'unethical' competitors. If there is no such niche marketing, then it seems probable that the first insurance companies were right - as a group, this part of the 'low paid' are a high risk.

Would those protesting against this be equally ready to protest if a group were identified as statistically of 'lower risk', equally on the basis of unchosen characteristics, and offered a lower rate? Would it be unethical, even if that group were on average paid less, and poorer, than those outside the group? Or should insurance companies be free to continue to offer lower rates to women? Or to the healthy but elderly?

[Hawker Hurricane]

Insurance IS gambling. The company acts as the house, and sets the odds in thier favor. The bet is this: the company is betting that you won't get into a car accident, you are betting that you will. The odds are stacked against the customer intentionally as far as they can be without driving the customers away. In Vegas, the best odds are found in the scummiest casinos...

[Tonio]

If there is no such niche marketing, then it seems probable that the first insurance companies were right - as a group, this part of the 'low paid' are a high risk.

It's also probable that the insurance companies merely presume that niche markets are not profitable.

An analogy from retailing is the problem faced by Prince George's County in Maryland. Three-quarters of the county's population is African-American, and some of its communities are among the wealthiest black-majority communities in the nation. But some other communities are plagued with drug and crime problems, even the middle-income ones. As a result, upscale retailers avoid Prince George's because they look only at countywide statistics.

[Tonio]

Would those protesting against this be equally ready to protest if a group were identified as statistically of 'lower risk', equally on the basis of unchosen characteristics, and offered a lower rate? Would it be unethical, even if that group were on average paid less, and poorer, than those outside the group?

Yes. The whole point is that individual behavior should count more.

[Realist]

It seems that many people want insurance to act not as a hedge against risk, as it was designed, but as a means to combat social or economic or biological inequality. To take Majromax's example, you don't want insurance companies to draw blood samples to test your genetic susceptibility to getting into accidents in order to come up with a rate to charge you, despite the fact that this would be consistent with its function as a risk-reducing agent. Rather, you want to use insurance as a means for the genetically good drivers to subsidize the risk of the genetically poor drivers, charging everyone the same rate and making society more equal.

The motives here are reasonable--I too want a more equal society. But why use private insurance as the means of accomplishing this? Insurance is not the cause of inequality; insurance companies don't charge the poor more because they enjoy screwing them more than rich people. In a society with no insurance, genetically poor drivers would still be worse off than genetically good drivers simply because they would get into more crashes, through no fault of their own! So why use private insurance to combat a problem insurance never caused?

If your real goal is to reduce inequality as well as pool risk, why not just dispense with the pretense of a capitalistic system, and insure everyone in a state plan? Everyone pays the same rate; if you cause an accident you get a fine and raised rates. No need for generalizing data that doesn't necessarily apply to specific cases. And you actively help people who happened to be born with statistically detrimental traits, thereby making society more equal.

[Froborr]

Realist: First, I'm not clear of what genetic traits could make someone a statistically worse driver without making them entirely unable to drive. Do you have examples?

Second, the system you describe is all well and good in fantasyland, but the U.S., at least, has a large population of assholes, and they love capitalism. So, some compromise with them is necessary.

[Jeff]

a large population of assholes, and they love capitalism

Are balt's ears burning?

(I hate libertarians, possibly more than Falwellian Xtians.)

[Realist]

Froborr,

Impulsiveness, aggression, poor motor control, would all make someone a worse driver and all are probably partially under genetic control. This is probably not testable now, but probably will be in the future.

As for your second point, the problem is that the compromise is often worse than either extreme. Superregulatory capitalism, the system which Democrats and Republicans seem to agree on, takes the inefficiency of socialism and combines it with the inequality of capitalism. If you forbid insurers from discriminating based on statistical risk, they will have to raise prices for everyone to cover the extra cost, and it will no longer be worthwhile for the population at lowest risk to enter the system, raising the prices even more. The high-risk now have their original high costs (probably even greater actually), and the low-risk have lost their risk-hedging mechanism. Everyone is worse off than either the capitalist or socialist system.

This seems to be a general feature of our chimeric society. Think the government is getting too big? Privatize parts of it to make the government seem smaller, then pay the new private owners even more to do the same job, probably worse. In real capitalism there would be competition, but in our hybrid system usually only one company benefits from the government contract. In real socialism the system would be run by bureaucrats with a functional task, in our hybrid system it's run by profit-seekers with incentive to cut function to raise profits.

I think we should just choose one and stick with it. You're right, this is hard to do when so many people think they like capitalism but don't like many of its effects.