Do CC UW criteria change in different demographic areas?

[FinStar]

---

@CreditCuriousity wrote:

---

@FinStar wrote:

Just keep a few items in mind.  Lenders have extensive information from collective data-mining sources to determine specific UW criteria such as tiered or risk-based pricing, default rates, etc.

You'd be suprised what most major lenders have at their fingertips and the amazing nano-second processing speed using such algorithms to make a lending decision.  That being said, consider the aggregate information they obtain from census tract data, credit bureaus, lexisnexis, public records, the work number, etc. to taylor something specific with such granularity that "zip code" is not really that relevant when it comes to UW criteria per se.  While lenders are regulated by various federal agencies they also have proprietary information (i.e. algorithms) which are never shared with the public (let alone regulators) as long as they abide by the CRA, FHA and ECOA standards, including the Card Act reform (for CCs), especially in underserved goegraphical areas.

---

So you are kind of agreeing what I am getting from your post that they might have different UW standards for different demographic areas?  Correct me if I am taking what you said out of context..  I work with data warehouses, analytics and dashboarding and reporting for a living and know how powerful of tools these are and how you can slice the data in such ways that it is amazing what it can show you, especially with trends..  I agree alot of this stuff they don't have to share with such agencies.

Still wonder if they loosen their UW criteria to get into new geographical areas that are known(just pulling a number of the air) to have say a 30 lower average Fico scores on average due to whatever reason.

---

Statistical and risk models are constantly adjusted based on economic and market trends.  That being said, some UW can vary, yes.  Ask yourself this question, what's more prevalent in an underserved or depressed economic area, fraud or revenue?  

[bigDEElight]

I'm almost 100% certain this happens. If it doesn't, then the banks are failing to take advantage of Big Data and data science in general. (Tries not to nerd out with my obvious love of data science)

[takeshi74]

---

@CreditCuriousity wrote:

I would like to hear other peoples thoughts on this subjects. 

---

See if the data in the Credit Pulls Database supports your theory.

[kdm31091]

---

@takeshi74 wrote:

---

@CreditCuriousity wrote:

I would like to hear other peoples thoughts on this subjects. 

---

See if the data in the Credit Pulls Database supports your theory.

---

Yeah, that's a good way to check.

Honestly, I don't know if I think this happens or not. They are pulling each individual's credit and although the computer only reviews it for a nanosecond it is still reviewing it. Lower income areas will tend to use debit more because they can't always get approved for credit -- not because of discrimination, but just low income or other factors.

If you're talking about manual review, of course natural human bias will come into play. If you are from a high income, ritzy area the underwriter may be more swayed to approve you. But they also may not care. It's subjective, just like any other factor with manual review/approval. It's all up to their decision. I do think at the end of the day, your profile and credit history is what matters above all, not where you live.

[jamesdwi]

I think it mainly has to do with Income vs Rent + Payments on other TL's. Sure if you have a long standing relationship with a lender they may have all kind of details availible to them. But if you visit chase.com and apply for a freedom card with them they are mostly pulling your report, and using your income provided, and doing some quick comparison of previous payments on credit cards vs your payments and then comparing it to the demagraphic for your income. 

Rememeber that risk equals profit for lenders, sure lenders may like 800 fico scores with a 40k CL, but they relize they may end up making far more money on 650 Fico score that they gave a $500 CL. The 800 fico may use the card once every 6 months to pay home owners insurance which he promptly pays off on the due date. Where the 650 fico score, may charge the card up to $450 and pay interest at 22% then have a late payment or two, and then get hit with 29.9% penalty rate and only make minimum payments except once a year when he gets his tax refund where he pays the card off, and 2 months later the card is maxed out again. Who knows after 2 years,  the card might even be a charge off after the person has paid more than the $450 borrowed in fees and interest, then the lender writes off $1000 in loss really the customer was a net profit. 

[Anonymous]

They probably do but it will never be admitted as they will be foolish to do so. Publicly admitting to it is begging to be sued.... 

[B335is]

I lived in Detroit for 6 years until about 7 months ago. In that time I significantly built up my credit portfolio from a couple $300 Cap1 cards to well over 4x my annual income. With the exception of geographical limitations, I never found a lender to give me a hard time for living in Detroit with the exception of LMCU (ironically), but that was over excessive new accounts.

[Anonymous]

---

@B335is wrote:
I lived in Detroit for 6 years until about 7 months ago. In that time I significantly built up my credit portfolio from a couple $300 Cap1 cards to well over 4x my annual income. With the exception of geographical limitations, I never found a lender to give me a hard time for living in Detroit with the exception of LMCU (ironically), but that was over excessive new accounts.

---

The OP asks an interesting question, and the first thing I thought of when I read it was, "Lawyers would salivate over this: a class-action lawsuit against a redlining bank." It's probably a given that big banks use data-mining; I do wonder why or how much a zip code would factor against payment history and other established credit factors when it comes to giving out credit cards. I can understand if a bank wanted to identify a certain zip code to establish a bank that offers financial services tailored to folks who fit into a certain economic profile. 

Speaking of The D: I moved there ten years ago for work. When I went to get car insurance, the agent tried his best to get me to move out of downtown (I was living on Jefferson, right downtown, which, I learned later, got me alot of street cred). He was incredulous that a working professional would want to live downtown. (In two years living in downtown Detroit, nothing ever happened to me; however, I got pulled over by Grosse Point cops when I drove there to buy groceries.) Anyway, my car premiums were astronomical due to, apparently, past claims based on zip code. That's why they can "get away" with charging alot based on location, just as they charge a premium based on age, type and even color of car. I think it would get more complicated with credit cards and bank loans (redlining.)

[Anonymous]

I would be interested to hear from UWs. I worked for a CCC but I wasn't in underwriting; however, we were trained that using geographic information was not allowed but how much control do they have over the subtle actions of UWs? 

I live in nice neighborhood in a college town; however, the majority of my zip code covers a nearby slum in neighboring city. We're seperated by a commercial corridor but if you just google my zip it'll bring up the crappy town instead of my neighborhood next door. 

[taxi818]

---

@Anonymous wrote:

I don't think they do this. They track many different things that change your internal scoring, but where you live is rarely something that proves you are a good or bad person to lend to. There are many areas in which there are both very rich and very poor people in very close proximity. Why wouldn't they just use income data about the person rather than make assumptions about the person just because they live somewhere? I bet the majority of difference in default rates between regions can be explained by income and other factors they already know about. I'd be interested to know whether stating you are a student could change their decisions. I don't think it would, but they love students and you can see how they would want to be locking someone in who will have a lot of expenses and likely has some support from their family, probably won't have a bankruptcy soon.

---

Well if you told someone in the hood this. They would disagree. With a 80 percent decline rate.  Amex was fined for this years back.