4 ms·

On the other hand you are just as likely to get a loan you otherwise wouldn't have because you haven't been seen with questionable folks. In any case that kind

On the other hand you are just as likely to get a loan you otherwise wouldn't have because you haven't been seen with questionable folks. In any case that kind of policy is already illegal. And it's not even technically feasible at the moment. Even if it was, it's very unlikely there is a strong correlation between being seen with some random person in a photo, and your likelihood of paying a loan. There is plenty of questionable shit happening right now in the NSA and stuff. I don't see why we need to resort to fearmongering about absurd hypotheticals.

terryf 11y ago
The banks are doing things, perhaps not exactly what was described, but your facebook history, linkedin, other social media accounts are taken into account when applying for a loan. Even when you browse the bank's web page, where you hover the mouse, what pages you look at and variour other metrics. One example of a company providing such service is http://www.bigdatascoring.com/ http://www.bigdatascoring.com/
- Houshalter 11y ago
  In my opinion that is a very good thing. If they can make fewer loans to people who are high risk and lower interest loans to people who are not, everyone wins. The source of the data doesn't matter.
tbrake 11y ago
> On the other hand you are just as likely to get a loan you otherwise wouldn't have because you haven't been seen with questionable folks. I would imagine the list of things that will count against you in those situations will always be much longer than the list of things that can count for you, i.e. just because A can count against you doesn't mean that ¬A will count for you; ¬A will just be 0 points and the assumed baseline.
- msellout 11y ago
  They're actually trying to give you the lowest rate they can while staying profitable, you know, to be competitive with other lenders. There's no baseline, all our characteristics are inputs to a function that estimates risk of default. Then the lender chooses what rate to offer based on their portfolio design. In one sense, it's a zero-sum game. If person A is a better debtor, A gets a better rate than B. However, a more sophisticated portfolio will try to find anti-correlated debtors. Rather than lending to 100 low-risk people all similar to each other, I'd rather diversify and include some very different people that may be higher risk on an individual basis; the aggregate risk can be much lower than individual. To that end, the more the lender can learn about me, the more chance I have to reveal how anti-correlated I am with others.

terryf 11y ago

The banks are doing things, perhaps not exactly what was described, but your facebook history, linkedin, other social media accounts are taken into account when applying for a loan. Even when you browse the bank's web page, where you hover the mouse, what pages you look at and variour other metrics. One example of a company providing such service is http://www.bigdatascoring.com/ http://www.bigdatascoring.com/

Houshalter 11y ago
In my opinion that is a very good thing. If they can make fewer loans to people who are high risk and lower interest loans to people who are not, everyone wins. The source of the data doesn't matter.

tbrake 11y ago

> On the other hand you are just as likely to get a loan you otherwise wouldn't have because you haven't been seen with questionable folks. I would imagine the list of things that will count against you in those situations will always be much longer than the list of things that can count for you, i.e. just because A can count against you doesn't mean that ¬A will count for you; ¬A will just be 0 points and the assumed baseline.

msellout 11y ago
They're actually trying to give you the lowest rate they can while staying profitable, you know, to be competitive with other lenders. There's no baseline, all our characteristics are inputs to a function that estimates risk of default. Then the lender chooses what rate to offer based on their portfolio design. In one sense, it's a zero-sum game. If person A is a better debtor, A gets a better rate than B. However, a more sophisticated portfolio will try to find anti-correlated debtors. Rather than lending to 100 low-risk people all similar to each other, I'd rather diversify and include some very different people that may be higher risk on an individual basis; the aggregate risk can be much lower than individual. To that end, the more the lender can learn about me, the more chance I have to reveal how anti-correlated I am with others.