Adding an installment loan -- the Share Secure technique

[Whitebird]

Hi CGID:

To answer your question, no I had no installment loans on my reports, open or closed. Frankly, if I had I would not have even considered a SSL It was more about proving myself right than gaining in FICO scores--although I'll gladly accept the latter.

I'm referring to the fact that quite a long time ago I was told "in another place" that even though I had no installment loans--and that FICO was yelling at me for this "sin"--adding an installment loan would have little to no effect on my FICOs since "everyone" knows that revolving accounts are the important ones! I cannot tell you how much I believed I was correct and "they" were wrong. (They were not taking into consideration the effect of different report content, FICO scores and scorecards.) I was absolute perfection in every FICO category except mix of credit. Since I was not about to take out an installment loan--what? a car, a house--that I didn't want, it had to wait until the SSL technique was explained. Even then, I had to think about it for a couple months before I pulled the trigger. I "sacrificed for science" LOL!

I'm beginning to think we each have a slightly different idea of how Vantage approaches installment loans, but that in the end the outcome is the same! It's mostly all theoretical. You think they identify SSL specifically and then disregard it for scoring purposes. I'm more likely to think it's a bit of benign neglect, in that Vantage identifies car, house and student loans specifically, and that whatever else is on the report (other than credit cards) are shoved into the "other category" whatever they are, and don't need to be further identified for purposes of classification.  The two accounts I have in "other" have nothing in common. I'm having a hard time reconciling that fact. The bank LOC is obviously not a car, house or student loan. OTOH, it's not a credit card either. Voilà! It goes in the "other" pile.

Oh, for certain Vantage scores contribute, likely quite a bit, for those who have open car, mortgage and/or student loans. I would never be convinced otherwise. I could not venture a guess in terms of the impact those same loans reported as "closed" have on those same accounts. Frankly, I don't think we have an exact understanding of how even FICO handles closed installment loans, do we? I could certainly have missed that, however.

What I'd like to see: How Vantage treats and classifies unsecured installment loans that are NOT mortgage, auto or student loans. I'm thinking specifically of those furniture stores that offer installment credit on purchases. Idon't know much about them, but the payback period seems to last years. That would separate the "secured" tag from the "installment" tag and might lead to more understanding of Vantage. I'd speculate that they too would go into the "other" category. If so, the "secured" tag by itself would mean little. Yes, I'm getting ahead of myself here.

[Anonymous]

Great!  Very helpful.  So we can rule out the hypthesis that you had a closed installment loan on your report and that this was the reason that using the SSL technique did not help your Vantage scores at all.

The only reason I can think of is that Vantage does not consider secured personal loans (or perhaps, as you point out, any personal loans, secured or otherwise) as something that helps their credit mix and installment utilization.

If they ignore secured personal loans, it wouldn't be that puzzling, even though FICO 8 obviously considers them.  What an algorithm should really be looking for is the answer to two questions:

    * Do a significant number of people experience 120-day-lates (or worse) on this kind of account -- in general?

    * How did THIS consumer handle this particular account?

If the answer to (a) is "almost nobody" then (b) is irrelevant to his credit reliability.  What the algorithm ideally is looking for is a situation where people sometimes do get into trouble (which happens with credit cards, home loans, car loans, etc.) and yet this person did not.  My guess is that, with SS loans and credit builder loans, the very method by which they are secured precludes almost anyone from getting into trouble.

Of course we can conclude anything definite yet, since we have very little information to go on.  But thanks for raising a really interesting question!

[Thomas_Thumb]

For what it is worth here are a few things to file away.

1) Fico 04 also does not give points for adding an open share secured loan and paying it down to 9% (even if you have no open loans in file).

2) Vantagescore 3.0, unlike the Credit Karma presentation, does factor in closed accounts in it AAoA calculation per VS web site. Credit Karma's calculation is based on only open accounts.

3) For me and per the MyFICO pop up text boxes open AU credit card accounts are ignored in # open accounts with balances and AU balances are not considered in Ag UT%. Vantagescore 3.0 does look at the AU accounts as do the earlier Fico 04 and Fico 98 models models.

A VS3 score at 830 and above is quite good and getting there without an installment loan on file (open or closed) suggests that element is not a critical factor for VS3. Highest VS3 score I have heard reported is 839. As mentioned before based on my file behavior, VS3 appears to require quite a few accounts (many more than Fico 08) for maximizing depth of credit. Pasted below is some VS3 info for those who have interest.

[Whitebird]

Hi - Thanks for all the info.

Wasn't aware that the model of Vantage on Credit Karma is not the same one as on Alliant. Alliant specifically names the free score they provide as Vantage. Until CGID asked about SSL effects on Vantage, I was barely aware of Vantage scores. My interest now is more theoretical--their relative approaches to score development, how it differs from FICO in that respect, etc.; certainly no personal interest at this point--perhaps if/when Vantage is actually used by the majors in credit decisioning.

Nothing to do with this thread whatsoever, but it just occurred to me:   In case anyone is interested in FICO information outside that found online, you may find it by reading FICO's quarterly earnings call transcripts. I get mine at seekingalpha.com. Sometimes they drop interesting nuggets about future plans, what they're thinking, etc.

[Whitebird]

Just one small point to add...

Yes, what you propose is testing for adverse selection, but I wonder if this specific testing has been done for this factor by either Vantage or FICO. Intuitively, I'd say no for Vantage, maybe for FICO. Amazing that they treat this kind of account in entirely different ways. Assuming it has not been done for whatever reasons (? not enough participants for a valid outcome--especially for Vantage) what is left unsaid is the possibility that those with a SSL (I'd include bank passbook loans, if they still issue these antiques!) could actually show better outcomes than those in the same scorecard without a SSL. I readily understand that this theoretical outcome is not exactly intuitive to everyone but that matters not at all. It's all data-driven.

[Whitebird]

Post 445 is supposed to be for Credit Guy in Dixie - somehow it got posted out of order. 

[Anonymous]

Thanks WB and TT both!

Hey WB... TT is not saying that the scoring model used on Karma is anything different from a true Vantage 3 score.  Karma's scores really are the true V3 McCoy. 

What TT is talking about is the "presentation" of the data around the report.  This is the software they use to "summarize" the data on your reports.  Almost all credit monitoring sites have software like this (myFICO has theirs, Credit Check Total has theirs,  Credit.com has theirs, Karma has theirs, etc.)  Two things that "summary" software will often tell you is the "age of your oldest account" and the "average age of accounts."  For some reason I do not understand, Karma's summary software ignores closed accounts when summarizing these factors, even though Vanatge Score 3 considers open and closed accounts equally in its model for account age.  Karma has been using Vanatge 3 for a long time now so I have no idea why it "presents" account age in a way that is different from how the score it is using calculates it, but that is a fact nonetheless.

TT can correct me if I am misinterpreting him, but I am pretty sure he's only taking about this "presentation" software -- not the actual score itself.

Another question for WB.... how many open accounts do you have?  How many closed accounts do you have?  (As of April 2016.)  Just curious -- no need to tell me if you'd rather not.  TT is absolutely right about the need to know a bit about the "thickness" of a person's profile before inferring too much from their scores.  Knowing the number of accounts you have would be a big help.

[Thomas_Thumb]

Thanks for clarifying CGID

Neither Credit Karma, Credit Sesame nor Credit.com calculate credit scores. They all show bona fide VantageScore 3.0 scores that come from thr VS3 model. In Credit.com's case the top score they list comes from the Experian National Equivalency model.

What the sites DO themselves is generate their own customized summary reports (or presentation reports as some call them). Of the three sites listed above onlyCredit Karma calculates average age of accounts based on open only. The other two include closed accounts in their respective calculations. The presentation information CK displays for average of accounts came along before they converted to VS3

A year ago I too thought VS3 used open accounts only but, IV and Revelate provided information direct fro VantageScore which resolved my misunderstanding.

[Revelate]

Alliant is also on the Vantage 3.0 algorithm, have been for around 2 years perhaps more.  They used to be on VS 2.0.

They never updated their pie chart as they used to overlay their older VS 2.0 pull and the newer 3.0 one for comparison purposes.  Ignore the fact there's a tickmark for 900, you won't get there on either site.

Regarding Vantage score:

I didn't get points for installment utilization tricks on VS; I did though get a large score increase when my mortgage reported... unlike FICO 8 where I got hammered (lost all the points I'd gained from my pretty installment utilization) when it reported.  It's a substantial departure from every model of the FICO algorithm I've been able to test in that regard, mortgage isn't in the same class apparently as either closed auto loans, or open/closed personal loans.  At least on my data they are equivalent as far as installment utilization goes.

AU's do count under FICO 8 for some people, not all.  Wish we'd stop promoting bad information, though apparently FICO is too and unfortunately the admins aren't reading the PM I sent asking for clarification /shrug.

[Thomas_Thumb]

Revelate, Nice to know that VS3 favors having a mortgage on file. Ballpark - what was your point gain.

Note: I believe you don't have a closed mortgage. Wonder if it matters if the mortgage is open or closed on VS3 to realize the benefit.