5 ms·

I think you are confusing two different issues. It is more expensive to lend money to black people because race is a predictor of poverty. Wealth can be measure

I think you are confusing two different issues. It is more expensive to lend money to black people because race is a predictor of poverty. Wealth can be measured directly, so discriminating based on race isn't really an effective strategy (ethics not withstanding). In the cost of fuel argument the root cause of the increase in cost is the weight of the passenger (it isn't just a predictor). The gender discrimination case is more complicated, but if companies want to modify wages based on expected medical leave required, I don't really see a problem with that.

yummyfajitas 10y ago
No, that's simply not true. See the comment I linked to. A black person with a 600 FICO has a 40% chance of default, compared to an Asian person with a 600 FICO who has only a 20% chance of default. With identical financial indicators the black person is less likely to repay. Race discrimination is directly profitable, given the available data [1]. Now it's true that in this example it's just a predictor. The very nature of lending is stochastic. Then again, airplane crashes as a function of weight distribution are also stochastic. [1] It's theoretically possible - likely even - that some hidden data may reduce the influence of race. For example, lenders are required (more or less) to ignore location data since that could be "redlining".
- gragas 10y ago
  >A black person with a 600 FICO has a 40% chance of default, compared to an Asian person with a 600 FICO who has only a 20% chance of default. There's your answer. You have it all backwards and think that creditors are oppressing certain races, when in reality the creditors are merely quantifying hard evidence about each race. To be absolutely clear, I'm not saying being black definitively means you will default. But it's absolutely ridiculous to ignore hard evidence. Blacks default 40% of the time in that scenario and Asians default 20% of the time. That's not racist. Blacks are not inferior. Yet, it is untenably true. It is fact. If that upsets you, so be it. It is not the duty of creditors to ignore facts and appropriate costs.
  - yummyfajitas 10y ago
    That's fine. You seem to have figured out your ethics. But for those folks here who don't view this situation similarly, I think the airline example is also a tricky case. I certainly have no good answer.
    - snovv_crash 10y ago
      It isn't ethics or your thinly veiled accusations of racism, it is cultural differences. Asian cultures typically have a strong heirarchy, with respect being given to those to whom a debt is owed. Black culture in the US, for various valid reasons involving armed struggle and persecution, has developed a strong "stick it to The Man" mentality. Given these differences, repayment rates being higher for those who are culturally asian makes perfect sense. Thinking that attitudes towards authority don't bleed through into behaviour for things like debt repayment is naïve.
      - yummyfajitas 10y ago
        I'm not accusing anyone of racism. I'm saying simply that a consistent, principled ethical system which allows height discrimination also seems to allow race or sex discrimination. So if you reject the latter, it seems like you should also reject the former (or perhaps provide some more detailed explanation of your ethical system). My only point here is "X seems to go with Y, but Y seems problematic so maybe X is too? Lets think carefully." The particular theory you bring up doesn't change this. According to you, "black" -> "stick it to the man mentality" -> "not pay back loan". That doesn't answer the ethical question of "is it acceptable to make fewer loans to blacks on the basis of black -> {whatever, doesn't really matter} -> default?" This is more or less analogous to "is it acceptable to charge more to tall people because tall -> {whatever} -> bad thing?" Maybe there is some clear and relevant distinction to be drawn. If so, what is it? I.e., what's the ethical principle, and how does it distinguish between these two cases?
        snovv_crash 10y ago
        That's what we have "protected classes" for. It makes it illegal to discriminate on things which might otherwise give an imagined, or real, business advantage.
- thinkloop 10y ago
  Simplified comment: Is weight the cause of increased costs and decreased safety? Yes. Is blackness the cause of non-payment? No. Original comment: I think there's an important difference between directly consuming a resource, and being broadly lumped into a group that has an elevated chance of increased consumption as a proxy for other underlying factors: primarily socio economics. Why stop at that level of granularity when you can easily keep slicing. What's the comparison between college educated members of each group? What about children of professionals? I think the brutish imprecision of stopping at race, especially that it's almost certainly not the cause, makes it a very different argument. As an aside, wtf is the point of FICO if not to be a measure of precisely these outcomes?
  - yummyfajitas 10y ago
    Suppose the airline had a scale which worked in the following way. It will measure their weight, but add randomly generated noise of a few pounds either way. Suppose this scale is perfectly adequate for the airline's purpose - i.e. it provides a sufficiently accurate measurement to provide safety - but it no longer provides a "true" measurement. Based on your reasoning, since this scale is no longer a perfect measurement of the true causal factor, it shouldn't be used. In reality, passenger weight can vary between weigh-in and getting on the plane. The passenger might eat food or use the bathroom before boarding. Weight on-plane, rather than at boarding time, is the true cause of decreased safety. Weight at boarding is merely a statistical approximation to this (and admittedly a very good one). Reality is uncertain, so it seems we are back to the same place. Tangentially: Why stop at that level of granularity when you can easily keep slicing...As an aside, wtf is the point of FICO if not to be a measure of precisely these outcomes? The reason for this is that using more factors is often legally problematic. Such factors are often correlated with race and regulators therefore treat using them as "redlining." For example, (income = low) AND (location == oakland) might also be predictive, but that's strongly correlated with race, and regulators will rape you if you try and use it for predictive purposes. Also, from what I've seen, your suggesting that slicing further will eliminate the predictiveness of race doesn't seem to be true. At the very least, we often need to slice much further than our available data sets allow. For example, FICO includes a lot of data but still doesn't make race non-predictive. Similarly, in education, race is still predictive after accounting for parental income - rich blacks do worse than poor Asians: https://en.wikipedia.org/wiki/Achievement_gap_in_the_United_States#Income_and_class https://en.wikipedia.org/wiki/Achievement_gap_in_the_United_... The idea that we can avoid these ethical questions by slicing data in the exact right way is probably not true. We should instead figure out our ethics so that we can do things right even if the data doesn't turn out the way we imagine.
- danieltillett 10y ago
  It is a very interesting question what is the true underlying cause here other than FICO not being able to find a non-racial factor. I would imagine finding such a factor (or factors) would be incredibly valuable. If I were given the job the first place I would look is to see if the effect holds with immigrant black people verses those born in the USA.
- dtech 10y ago
  The only evidence I see here is that FICO is apparently a poor/insufficient predictor of default rate, which causes additional information (race => socioeconomic background => default rate) to increase accuracy of the prediction.
- tptacek 10y ago
  Again, I'll point out that you've cherry picked an idiosyncratic point on the FICO scale (600, vs. the 620 cutoff used by industry for subprime loans), and cherry picked the cohorts you're comparing (Asians being least likely in general to default), both of which has the effect of maximizing the delta you're citing as evidence for your argument. It would be a nit, except we had a long thread about this previously wherein I perceived you to have conceded these points, so I'm surprised to see you repeating them here.
  - yummyfajitas 10y ago
    I'm confused what you are arguing. What specific claim did you dispute (as I recall none), and what do you believe I conceded? In that thread I concluded you had no substantive disagreement and were merely looking for excuses to call me racist. (Why you enjoy that so much I don't know, but I don't mind so enjoy.) As far as the argument I make here, what do you believe this refutes?
    - dang 10y ago
      > In that thread I concluded you had no substantive disagreement and were merely looking for excuses to call me racist. (Why you enjoy that so much I don't know, but I don't mind so enjoy.) I haven't read all of your and tptacek's recent exchanges, but in the ones I have, he has been meticulously respectful to a fault, so this seems like an unfortunate lapse. Since I know you care about the principle of charity, I'm sure you won't mind my asking you not to do this again.
    - solutionyogi 10y ago
      I would agree with Dang here. I am a big fan of you and Thomas. And I love reading your discussions on a topic and have read most of them. I tend to agree with your arguments more but I must say that Thomas is very respectful while disagreeing and I haven't seen him call you a racist. Feel free to link to his comment where he does and I will stand corrected.
      - yummyfajitas 10y ago

yummyfajitas 10y ago

No, that's simply not true. See the comment I linked to. A black person with a 600 FICO has a 40% chance of default, compared to an Asian person with a 600 FICO who has only a 20% chance of default. With identical financial indicators the black person is less likely to repay. Race discrimination is directly profitable, given the available data [1]. Now it's true that in this example it's just a predictor. The very nature of lending is stochastic. Then again, airplane crashes as a function of weight distribution are also stochastic. [1] It's theoretically possible - likely even - that some hidden data may reduce the influence of race. For example, lenders are required (more or less) to ignore location data since that could be "redlining".

gragas 10y ago
>A black person with a 600 FICO has a 40% chance of default, compared to an Asian person with a 600 FICO who has only a 20% chance of default. There's your answer. You have it all backwards and think that creditors are oppressing certain races, when in reality the creditors are merely quantifying hard evidence about each race. To be absolutely clear, I'm not saying being black definitively means you will default. But it's absolutely ridiculous to ignore hard evidence. Blacks default 40% of the time in that scenario and Asians default 20% of the time. That's not racist. Blacks are not inferior. Yet, it is untenably true. It is fact. If that upsets you, so be it. It is not the duty of creditors to ignore facts and appropriate costs.
- yummyfajitas 10y ago
  That's fine. You seem to have figured out your ethics. But for those folks here who don't view this situation similarly, I think the airline example is also a tricky case. I certainly have no good answer.
  - snovv_crash 10y ago
    It isn't ethics or your thinly veiled accusations of racism, it is cultural differences. Asian cultures typically have a strong heirarchy, with respect being given to those to whom a debt is owed. Black culture in the US, for various valid reasons involving armed struggle and persecution, has developed a strong "stick it to The Man" mentality. Given these differences, repayment rates being higher for those who are culturally asian makes perfect sense. Thinking that attitudes towards authority don't bleed through into behaviour for things like debt repayment is naïve.
    - yummyfajitas 10y ago
      I'm not accusing anyone of racism. I'm saying simply that a consistent, principled ethical system which allows height discrimination also seems to allow race or sex discrimination. So if you reject the latter, it seems like you should also reject the former (or perhaps provide some more detailed explanation of your ethical system). My only point here is "X seems to go with Y, but Y seems problematic so maybe X is too? Lets think carefully." The particular theory you bring up doesn't change this. According to you, "black" -> "stick it to the man mentality" -> "not pay back loan". That doesn't answer the ethical question of "is it acceptable to make fewer loans to blacks on the basis of black -> {whatever, doesn't really matter} -> default?" This is more or less analogous to "is it acceptable to charge more to tall people because tall -> {whatever} -> bad thing?" Maybe there is some clear and relevant distinction to be drawn. If so, what is it? I.e., what's the ethical principle, and how does it distinguish between these two cases?
      - snovv_crash 10y ago
        That's what we have "protected classes" for. It makes it illegal to discriminate on things which might otherwise give an imagined, or real, business advantage.
thinkloop 10y ago
Simplified comment: Is weight the cause of increased costs and decreased safety? Yes. Is blackness the cause of non-payment? No. Original comment: I think there's an important difference between directly consuming a resource, and being broadly lumped into a group that has an elevated chance of increased consumption as a proxy for other underlying factors: primarily socio economics. Why stop at that level of granularity when you can easily keep slicing. What's the comparison between college educated members of each group? What about children of professionals? I think the brutish imprecision of stopping at race, especially that it's almost certainly not the cause, makes it a very different argument. As an aside, wtf is the point of FICO if not to be a measure of precisely these outcomes?
- yummyfajitas 10y ago
  Suppose the airline had a scale which worked in the following way. It will measure their weight, but add randomly generated noise of a few pounds either way. Suppose this scale is perfectly adequate for the airline's purpose - i.e. it provides a sufficiently accurate measurement to provide safety - but it no longer provides a "true" measurement. Based on your reasoning, since this scale is no longer a perfect measurement of the true causal factor, it shouldn't be used. In reality, passenger weight can vary between weigh-in and getting on the plane. The passenger might eat food or use the bathroom before boarding. Weight on-plane, rather than at boarding time, is the true cause of decreased safety. Weight at boarding is merely a statistical approximation to this (and admittedly a very good one). Reality is uncertain, so it seems we are back to the same place. Tangentially: Why stop at that level of granularity when you can easily keep slicing...As an aside, wtf is the point of FICO if not to be a measure of precisely these outcomes? The reason for this is that using more factors is often legally problematic. Such factors are often correlated with race and regulators therefore treat using them as "redlining." For example, (income = low) AND (location == oakland) might also be predictive, but that's strongly correlated with race, and regulators will rape you if you try and use it for predictive purposes. Also, from what I've seen, your suggesting that slicing further will eliminate the predictiveness of race doesn't seem to be true. At the very least, we often need to slice much further than our available data sets allow. For example, FICO includes a lot of data but still doesn't make race non-predictive. Similarly, in education, race is still predictive after accounting for parental income - rich blacks do worse than poor Asians: https://en.wikipedia.org/wiki/Achievement_gap_in_the_United_States#Income_and_class https://en.wikipedia.org/wiki/Achievement_gap_in_the_United_... The idea that we can avoid these ethical questions by slicing data in the exact right way is probably not true. We should instead figure out our ethics so that we can do things right even if the data doesn't turn out the way we imagine.
danieltillett 10y ago
It is a very interesting question what is the true underlying cause here other than FICO not being able to find a non-racial factor. I would imagine finding such a factor (or factors) would be incredibly valuable. If I were given the job the first place I would look is to see if the effect holds with immigrant black people verses those born in the USA.
dtech 10y ago
The only evidence I see here is that FICO is apparently a poor/insufficient predictor of default rate, which causes additional information (race => socioeconomic background => default rate) to increase accuracy of the prediction.
tptacek 10y ago
Again, I'll point out that you've cherry picked an idiosyncratic point on the FICO scale (600, vs. the 620 cutoff used by industry for subprime loans), and cherry picked the cohorts you're comparing (Asians being least likely in general to default), both of which has the effect of maximizing the delta you're citing as evidence for your argument. It would be a nit, except we had a long thread about this previously wherein I perceived you to have conceded these points, so I'm surprised to see you repeating them here.
- yummyfajitas 10y ago
  I'm confused what you are arguing. What specific claim did you dispute (as I recall none), and what do you believe I conceded? In that thread I concluded you had no substantive disagreement and were merely looking for excuses to call me racist. (Why you enjoy that so much I don't know, but I don't mind so enjoy.) As far as the argument I make here, what do you believe this refutes?
  - dang 10y ago
    > In that thread I concluded you had no substantive disagreement and were merely looking for excuses to call me racist. (Why you enjoy that so much I don't know, but I don't mind so enjoy.) I haven't read all of your and tptacek's recent exchanges, but in the ones I have, he has been meticulously respectful to a fault, so this seems like an unfortunate lapse. Since I know you care about the principle of charity, I'm sure you won't mind my asking you not to do this again.
  - solutionyogi 10y ago
    I would agree with Dang here. I am a big fan of you and Thomas. And I love reading your discussions on a topic and have read most of them. I tend to agree with your arguments more but I must say that Thomas is very respectful while disagreeing and I haven't seen him call you a racist. Feel free to link to his comment where he does and I will stand corrected.
    - yummyfajitas 10y ago