> “It’s not cost efficient” isn’t a valid reason to discriminate against people, if you ask me.
Moreover, it's not, in general, a legal reason in the US to discriminate where a protected class is involved, including, in employment, age over 40 or gender.
Well, I mean, if you'd rather a political than a mechanics-of-law explanation as to why the law treats two situations differently mo, it's legal because heterosexual couples plus women outside of such couples combine to form a powerful voting bloc, and gender rating, compared to it's absence with no other change in practice, is a benefit to that group.
Specifically, compared to perfect allocation of costs, it's a small subsidy from people outside of opposite-sex couples to opposite-sex couples, whereas not having gender rating with otherwise status quo policies is a large subsidy from women not in opposite-sex couples to men not in opposite-sex couples.
Under status quo policies, opposite sex couples are in practice fairly free to gender-optimize insurance pricing by putting the less expensive to insure driver as primary on the more expensive to insure vehicle, regardless of actual primary driver.
I worked for a car insurance company for ten years and the management would answer, “Because loss history shows that we pay more in claims for insured men.”
So that’s ok but for some reason if it were a particular race they wouldn’t think to charge people
More by race so honestly I don’t know how to answer your question in a logically consistent manner.
But no one said it is was ethical in the case where it is legal, so if that was the intended question, not only is it phrased wrong but it starts from an unsupported premise, to wit, that people in the discussion believe that one is ethical and the other not.
But of course, there could be laws that prohibit linking vehicle insurance cost to gender.
My solution, when I've been married, has been having my wife get insurance, and add me as a driver. But that doesn't matter so much now that I'm older.
Edit: I do agree that discrimination by gender is fundamentally unethical.
Because there is no specific law that bans charging men and women different prices for car insurance. There are specific laws regarding discriminatory advertising with employment and housing.
What if the bias was learned by a neural network or other algorithm (eg historical insurance costs, page rank, ...)? Does that somehow make it legal if they make an inference about a protected class and stop advertising to them?
That kind of thing is why the doctrine of disparate impact was developed. There are a lot of messy edge cases (everything has some correlation with one or another protected class), but if I end up never advertising to men the fact that I'm using a neural network won't protect me.
> That kind of thing is why the doctrine of disparate impact was developed.
It's really not; disparate impact deals with it, but the doctrine was articulated in a case with ample evidence that the policy at issue, though facially neutral, was adopted as a deliberate replacement for explicit racial discrimination with an intent to maintain it's discriminatory effect and without other business purpose, and the doctrine was clearly intended principally to address such veiled-substitute policies.
The doctrine was first articulated in a case (Griggs v. Duke Power Co.) where the Supreme Court explicitly held the employer didn't intend to discriminate.
> the Supreme Court explicitly held the employer didn't intend to discriminate.
The Supreme Court did not hold that, it merely noted that the lower courts had held that, and noted that a lack of discriminatory intent was “suggested” by the Company's efforts to help undereducated employees by financing a major portion of remedial education (part, but not all, of the policy held to be disparate impact was adoption of high school diploma requirements for positions for which such requirements has no substantial relation to job performance.)
The Supreme Court explicitly did not consider whether there was intent to discriminate, because it was not necessary to do so to resolve the case.
So then are algorithmic approaches to advertising for employment off limits? Facebook and google then really shouldn’t go near this kind of work since their tech is legally inappropriate.
> So then are algorithmic approaches to advertising for employment off limits?
Not if what the algorithm optimizes is a measure provably closely linked to job performance.
In fact, as the array of protected classes and the statistical means available will probably over time allow proving prohibited disparate impact for any hiring method other than an algorithmic method optimizing a true job performance measure, algorithmic methods may soon be practically mandatory.
OTOH, algorithmic methods targeting any figure of merit that isn't provably tightly linked to job performance are going to be problematic, sure.
I don't know how it works with Google, but the article says Facebook is building a new ad console specifically for the kinds of advertising where discrimination is illegal. Presumably they'll make sure ads run through it aren't producing any disparate impact. (But note that this wasn't the problem with the cases in the article; in those cases, advertisers were explicitly pressing a "don't send this to women" button.)
Moreover, it's not, in general, a legal reason in the US to discriminate where a protected class is involved, including, in employment, age over 40 or gender.