The end of sexual discrimination in Europe!

[KrazyHorse]

That calculation makes no sense whatsoever.

[Kuciwalker]

It doesn't work now.

If you're going to stereotype demographically, be more precise.

Oh? Why does your suggestion produce better outcomes, given that insurance companies are already being as precise as is economical and legal?

[Kuciwalker]

Asher's notion of "fairness" seems to be pretty biased in his own self-interest...

[Dinner]

Come to Kansas, I pay $360 a year.

Damn, that is cheap.

[Asher]

Oh? Why does your suggestion produce better outcomes, given that insurance companies are already being as precise as is economical and legal?

Uh...yes, isn't the entire point of this thread the legality?

Let them either discriminate against any demographic with any statistic they so choose, or forbid it entirely. This cherry-picking of statistics to satisfy law/PCness is absurd.

Asher's notion of "fairness" seems to be pretty biased in his own self-interest...

Fair is fair, I just happen to benefit in this case. Absolute fairness can never be biased, and I'm proposing letting the insurance companies discriminate using any statistics on any demographics they so choose.

[Asher]

BTW, I'm fairly certain it is not illegal for insurance companies to provide benefits based on long-term domestic partnerships when they already grant the same to married couples. I don't know of a single insurance company who does so. As a result, someone who has been dating a girl for 3 months, then gets married gets a discount for being more "responsible" while someone in an 8-year relationship gets nothing.

And since when is it illegal to discriminate against occupation when giving insurance rates?

[Kuciwalker]

Uh...yes, isn't the entire point of this thread the legality?

I added that in the sense that, even if the law does not allow them to discriminate in certain ways (e.g. race) that should not make it "unfair" for them to discriminate in other ways. Moreover, the law should not decide to be "consistent" and force the insurance companies to avoid ALL discrimination just because the law incorrectly prevents SOME discrimination. (It SHOULD decide to let them discriminate on pretty much any basis they choose, but that is not a realistic outcome.)

Absolute fairness can never be biased, and I'm proposing letting the insurance companies discriminate using any statistics on any demographics they so choose.

As I read it, you were arguing (in #11 and #15) that the insurance companies shouldn't discriminate based on demographics at all, if they don't do so to the furthest theoretically possible extent. If you instead believe that they should discriminate as much as is economical and legal, then I don't disagree with you.

Let them either discriminate against any demographic with any statistic they so choose, or forbid it entirely. This cherry-picking of statistics to satisfy law/PCness is absurd.

By "legal" I am also including practices that are not technically illegal but would immediately be banned if the companies actually used them, e.g. race (in some jurisdictions). I don't see any point in being more "fair" by doing that; the end result is identical, except they have also raised the ire of huge swathes of the public. Do you really think that they are obliged to do something with no substantive positive consequences just to show that they tried?

BTW, I'm fairly certain it is not illegal for insurance companies to provide benefits based on long-term domestic partnerships when they already grant the same to married couples. I don't know of a single insurance company who does so. As a result, someone who has been dating a girl for 3 months, then gets married gets a discount for being more "responsible" while someone in an 8-year relationship gets nothing.

And since when is it illegal to discriminate against occupation when giving insurance rates?

"as precise as is economical"

Regarding relationships, I expect it would be far too expensive for the insurance companies to collect that data in a fraud-resistant way to justify the possible savings. Regarding occupation, possibly the same, or there just aren't that many savings to be had.

[Asher]

I added that in the sense that, even if the law does not allow them to discriminate in certain ways (e.g. race) that should not make it "unfair" for them to discriminate in other ways. Moreover, the law should not decide to be "consistent" and force the insurance companies to avoid ALL discrimination just because the law incorrectly prevents SOME discrimination. (It SHOULD decide to let them discriminate on pretty much any basis they choose, but that is not a realistic outcome.)

I don't see why saying women drive differently than men is different than saying blacks drive differently than whites or gays drive differently than straights.

They're all stereotypes, with I'm sure ample amounts of data could very easily be mined to determine this. It's just not being done.

I resent being punished for being male when I'm not "rewarded" for being things that would statistically improve my risk factor with the insurance companies, just because it'd be a PR no-no.

This is my problem.

Regarding relationships, I expect it would be far too expensive for the insurance companies to collect that data in a fraud-resistant way to justify the possible savings. Regarding occupation, possibly the same, or there just aren't that many savings to be had.

Savings for who? Fairly certain a software developer with a university degree would drive appreciably different than a high-school dropout who works at a car shop installing turbos...would it not be fair to ask that the rates reflect the different risk profiles?

[Kuciwalker]

I don't see why saying women drive differently than men is different than saying blacks drive differently than whites or gays drive differently than straights.

It isn't. That's why the law shouldn't forbid either. But given that there is no chance that the law will allow discrimination based on race, we don't become better off by also proscribing discrimination based on sex. It really is a perfect example of "cutting your nose off to spite your face".

I resent being punished for being male when I'm not "rewarded" for being things that would statistically improve my risk factor with the insurance companies, just because it'd be a PR no-no.

It would be a "PR no-no" in the sense that it would immediately be banned and they wouldn't be allowed to do it anymore. So what, again, is the point?

Savings for who?

Savings to the insurance companies, savings to their customers, and savings to society; driving is a negative externality, and so pricing that externality (through insurance premiums) produces gains to society in proportion to the degree that those prices are accurate.

When prices are made more accurate, the lower-risk consumers save because they then pay lower premiums; the insurance companies save because reduced adverse selection lowers their unit costs; society gains because higher-risk drivers are more strongly deterred from getting on the road.

Fairly certain a software developer with a university degree would drive appreciably different than a high-school dropout who works at a car shop installing turbos...would it not be fair to ask that the rates reflect the different risk profiles?

Maybe, but maybe they don't drive differently enough for it to be worth the expense of 1) studying the precise degree of difference and how it interacts with other factors and 2) collecting the consumer data in a way that is fraud-resistant.

[Asher]

I don't see why saying women drive differently than men is different than saying blacks drive differently than whites or gays drive differently than straights.

It isn't. That's why the law shouldn't forbid either. But given that there is no chance that the law will allow discrimination based on race, we don't become better off by also proscribing discrimination based on sex. It really is a perfect example of "cutting your nose off to spite your face".

It would be a "PR no-no" in the sense that it would immediately be banned and they wouldn't be allowed to do it anymore. So what, again, is the point?

I'm saying it should, not that it will.

Maybe, but maybe they don't drive differently enough for it to be worth the expense of 1) studying the precise degree of difference and how it interacts with other factors and 2) collecting the consumer data in a way that is fraud-resistant.

I'm sorry, why is this so expensive? Am I missing something here?

When you get the people to fill out the paperwork to get their insurance, you ask an additional question: "What is your occupation?"

After a number of years, you data mine. Correlate costs with occupations.

The banks certainly find it's important enough to know...

[Kuciwalker]

If people know that their premiums will be based on their occupations they have a large incentive to lie or (more likely) describe it inaccurately. Employers have no incentive not to aid them in this (I'm sure you've seen plenty of job title inflation at some of the places you've worked). And you may just be wrong that there are large systematic differences in risk between different occupations, especially after other demographic factors are accounted for.

And of course it's possible that the insurance companies are just lazy and haven't done it - anywhere in the world.

[Asher]

If people know that their premiums will be based on their occupations they have a large incentive to lie or (more likely) describe it inaccurately.

Employers have no incentive not to aid them in this (I'm sure you've seen plenty of job title inflation at some of the places you've worked). And you may just be wrong that there are large systematic differences in risk between different occupations, especially after other demographic factors are accounted for.

It's not the job title, it's the occupation itself. Titles are useless.

And of course there's a potential for fraud. Just like everything else with insurance. You say you live somewhere you don't, etc.

I also find that excuse incredibly hard to believe because they ask inane, non-fraud resistant questions like "How far do you drive to work every day" and "Does your vehicle have a car alarm"? It doesn't seem to be something that concerns them very much.

And of course it's possible that the insurance companies are just lazy and haven't done it - anywhere in the world.

Given my experience with insurance companies, I'm fairly certain this is the real answer. To them it's needless cost to simply more precisely allocate income. It doesn't matter to them if $10M a month comes half from software developers and half from mechanics, or if the split is otherwise different. Insurance companies aren't out to make YOUR rate fair or reasonable.

[Kuciwalker]

Given my experience with insurance companies, I'm fairly certain this is the real answer. To them it's needless cost to simply more precisely allocate income. It doesn't matter to them if $10M a month comes half from software developers and half from mechanics, or if the split is otherwise different. Insurance companies aren't out to make YOUR rate fair or reasonable.

It is in their strong interest to make your rate more accurate, especially if their competitors don't. Example:

Women on average cost the insurance companies $90/year, men $110/year. All of the insurance companies charge $105/year (with the extra $5 going to overhead, profit, etc.) because they are too lazy to ask someone's sex. One company decides to discriminate, charging $95 for women and $115 for men. Obviously no men are going to switch, but almost all of the women will. As a result, the non-lazy insurance company gets a HUGE increase in market share and all of the lazy insurance companies end up going bankrupt because they are charging $105 for people who cost $110.

To the degree that insurance companies aren't doing this sort of thing, they are just leaving money on the table. I don't think it's a good idea for the government to micromanage industries that it believes aren't behaving as profitably as they could.

[Asher]

It is in their strong interest to make your rate more accurate, especially if their competitors don't. Example:

Women on average cost the insurance companies $90/year, men $110/year. All of the insurance companies charge $105/year (with the extra $5 going to overhead, profit, etc.) because they are too lazy to ask someone's sex. One company decides to discriminate, charging $95 for women and $115 for men. Obviously no men are going to switch, but almost all of the women will. As a result, the non-lazy insurance company gets a HUGE increase in market share and all of the lazy insurance companies end up going bankrupt because they are charging $105 for people who cost $110.

To the degree that insurance companies aren't doing this sort of thing, they are just leaving money on the table. I don't think it's a good idea for the government to micromanage industries that it believes aren't behaving as profitably as they could.

I don't think this makes sense whatsoever.

If they cut someone else's rates, they jack someone else's. It's zero-sum, is it not? Why would you make the case that one would switch from one provider to another to save money while the other would not? If I knew there was a company out there that did NOT discriminate on sex, I'd jump on that. So would almost all the men. The savings are huge.

Or are you saying once all of the women leave the company to join the company which offers women discounted rates, then the cost for men goes up because the women are no longer subsidizing. Worst-case, then -- isn't their cost going to be the same at company A or B?

The reason no one company is doing this is because they don't want to rock the boat. They're making good profits now, why risk it by complicating the industry even more?

[notyoueither]

Why is it OK to discriminate based on sex, but not on race?