Speaking of LexisNexis, does anybody know if this is simply a rebranded ChoicePoint score? Google seems to indicate yes -- I remember reading about A-PLUS and CLUE reports along with their associated scores years ago, but at the time they were called ChoicePoint.
Also if anyone cares, here's a laundry list of all the reason codes for LexisNexis: https://consumer-solutions.custhelp.com/app/answers/detail/a_id/2523
(mods I apologize in advance if this external link is not allowed)
Edit: Looked through some of the reason codes a bit deeper, and they reveal some interesting data:
Codes 3004 to 3011:
2. How does this affect my insurance risk score? Insurance industry research shows that consumers whose accounts have been established on average for more than 7 years 7 months have fewer insurance losses.
Codes 3012 to 3018:
2. How does this affect my insurance risk score? Insurance industry research shows that consumers whose accounts have been established on average for more than 16 years 2 months have fewer insurance losses.
Wonder if these say anything about the break points for AAoA under the LN model.
@Anonymous wrote:The three penalty codes in question are 3489-3491:
3489 Ratio of Oldest Bank Revolving Account to Oldest Account is 44.55% or Less
3490 Ratio of Oldest Bank Revolving Account to Oldest Account is 44.56% to 53.28%
3491 Ratio of Oldest Bank Revolving Account to Oldest Account is 53.29% to 99.51%
Interestingly, there are some reason codes earlier in the series that attempts to cover every bracket it would appear:
3024 Ratio of Oldest Bank Revolving Account to Oldest Account is 37.34% or Less3025 Ratio of Oldest Bank Revolving Account to Oldest Account is 37.35% to 99.77%
3026 Ratio of Oldest Bank Revolving Account to Oldest Account is 71.54% or Less3027 Ratio of Oldest Bank Revolving Account to Oldest Account is 71.55% to 87.50%
3028 Ratio of Oldest Bank Revolving Account to Oldest Account is 87.51% to 91.67%
3029 Ratio of Oldest Bank Revolving Account to Oldest Account is 91.68% to 99.33%
Seems like under the LN model, there's just no escaping that "oldest account =/= BRA" penalty.
Then if you look through codes 3030 through 3045, even your Average Credit Line on Bank Revolving Accounts gets considered, with code 3045 topping out at ACL = $7,500.
Well, all the more reason to request CLI and get those limits up as high as possible. 
Man when I have time/feel bored over the holidays, I'm gonna have to sit down and go over those LexisNexis reason codes with a fine-tooth comb to dissect their model to the best of my brain's capabilities.
@arkane wrote:
Then if you look through codes 3030 through 3045, even your Average Credit Line on Bank Revolving Accounts gets considered, with code 3045 topping out at ACL = $7,500.
Well, all the more reason to request CLI and get those limits up as high as possible.
Man when I have time/feel bored over the holidays, I'm gonna have to sit down and go over those LexisNexis reason codes with a fine-tooth comb to dissect their model to the best of my brain's capabilities.
Yep. ACL matters in both LN and I think in Vantage. FICO doesn't care about size of CL at all.
The problem with trying to develop a practical LN score strategy is twofold. First, we don't have any idea what the scoring impact of these things are individually. Yes ACL matters. And yes, so does whether your oldest account is a bankcard. But conceivably these might cost you very little. Or maybe either or both matter a lot.
This first point is really serious. If all we had was a list of FICO 8 reason codes, we'd know very little about the model. The reason we do know as much as we do is because there are thousands of people on this forum alone regularly pulling their FICO scores -- often multiple times a month -- and comparing the changes in the score with changes in the reports and with the required four reason codes for the four biggest problem with your score.
Second, the LN reason codes are in tension with one another. When you optimize your profile to protect against reason code X, you end up realizing that you have put yourself in the penalty box for reason code Y. Not all of the codes are in tension with another code -- for example you can improve your LN score by having no lates, no collections, no chargeoffs, etc. By eliminating all lates, you won't be penalizing yourself somewhere else. But many of the codes are in tension with some other reason code somewhere else.
Still, going through the LN code list is a lot of fun and I encourage you to do it. It would be great to have somebody who makes it his mission to become an expert on it. You'll find a lot of fascinating stuff. Here are just a few differences between LN and FICO:
LN dislikes seeing you have a store card FICO doesn't care.
LN dislikes seeing you have an auto account (like Pep Boys, AutoZone, etc.). FICO doesn't care.
LN dislikes it when your Age of Oldest Account is substantially higher than your AAoA. No evidence that FICO cares.
LN is especially worried about the Mark of The Beast (666). Thus it looks at the total dollar value of all your debt (excluding mortgages, but with credit cards and loans combined) and if your total debt is $666 or more, it imposes a penalty. The developers at FICO spent less time watching The Omen as kids and are less stressed out by 666.
And so on.
Have fun!
@Anonymous wrote:
Yep. ACL matters in both LN and I think in Vantage. FICO doesn't care about size of CL at all.
The problem with trying to develop a practical LN score strategy is twofold. First, we don't have any idea what the scoring impact of these things are individually. Yes ACL matters. And yes, so does whether your oldest account is a bankcard. But conceivably these might cost you very little. Or maybe either or both matter a lot.
This first point is really serious. If all we had was a list of FICO 8 reason codes, we'd know very little about the model. The reason we do know as much as we do is because there are thousands of people on this forum alone regularly pulling their FICO scores -- often multiple times a month -- and comparing the changes in the score with changes in the reports and with the required four reason codes for the four biggest problem with your score.
Second, the LN reason codes are in tension with one another. When you optimize your profile to protect against reason code X, you end up realizing that you have put yourself in the penalty box for reason code Y. Not all of the codes are in tension with another code -- for example you can improve your LN score by having no lates, no collections, no chargeoffs, etc. By eliminating all lates, you won't be penalizing yourself somewhere else. But many of the codes are in tension with some other reason code somewhere else.
Ha yeah they have a smatter of codes on "department store cards", and it seems in particular code 0165 is in conflict with pretty much everything else. This is an excerpt:
Code 0165: Percent of All Department Store Accounts Reported in the Last 24 Mos to Total # of Accts
Property - More than 56% is better <--- what???
2. How does this affect my insurance risk score? Insurance industry research shows that consumers who with recent experience paying their department store accounts on time have fewer insurance losses
This is in clear conflict with codes 0134 (# of dept store accounts) and arguably 0120 (length of time dept store account has been established).
Still, going through the LN code list is a lot of fun and I encourage you to do it. It would be great to have somebody who makes it his mission to become an expert on it. You'll find a lot of fascinating stuff. Here are just a few differences between LN and FICO:
LN dislikes seeing you have a store card FICO doesn't care.
LN dislikes seeing you have an auto account (like Pep Boys, AutoZone, etc.). FICO doesn't care.
LN dislikes it when your Age of Oldest Account is substantially higher than your AAoA. No evidence that FICO cares.
LN is especially worried about the Mark of The Beast (666). Thus it looks at the total dollar value of all your debt (excluding mortgages, but with credit cards and loans combined) and if your total debt is $666 or more, it imposes a penalty. The developers at FICO spent less time watching The Omen as kids and are less stressed out by 666.
And so on.
Have fun!
It would appear that if you must have store cards, you better make sure of the following:
- less than 2 store cards (code 0134)
- less than 38 months (3 yr 2 mo) of history on each of the 2 store cards (code 0120)
- less than $284 total credit limit on both store cards combined (code 0146)
- have at least seven other types of open accounts (if you have 2 store cards) or five other types of accounts (if you have 1 store card) so your open store cards are <24% of all open accounts (code 0174) -- this is obviously in direct contradiction to code 0165
But most importantly, there is a separate code 0181 for "number of open department store accounts". So I say with all the potentially negative reason codes towards department store accounts, LN just doesn't like seeing any store accounts period.