@Anonymous wrote:
@SouthJamaica wrote:In all fairness to OP, there are several aspects of the FICO 8 algorithm which lend support to the view that FICO is intended to keep us in debt:
1. The penalty for having all credit cards paid down to zero.
2. The penalty for not having an open installment loan.
These cannot be justified as gauging credit mix and experience, because paid down credit cards and paid off installment loans demonstrate these as well as -- indeed even better than -- accounts with balances.
Hi SouthJ!
Hi @Anonymous I've interspersed your characteristically thoughtful post with a few comments.
Remember, however, that our OP was claiming that FICO measured how much interest you paid to creditors. Having a small $5 balance on one card does not entail any significant debt and it does not entail even one penny per year of interest. (You could in fact report thousands of dollars every month and still pay zero dollars in interest, simply by paying the statement balance in full, which many consumers do via autopay.)
Yes of course that FICO algorithm does not measure how much interest you pay, but there is certainly greater likelihood of paying interest if you owe money than if you don't.
And FICO does indeed reward paying down credit cards -- just not paying all of them to zero. (And even when that happens the small penalty gets removed instantly as soon as one card reports a small balance.) We should keep in mind that it typically takes a lot of work to constantly keep all cards reporting zero.
I don't agree with that at all. That was true when credit card statements were printed on paper, and we had to wait for the paper statement to know what our balance was. Nowadays with online access to our accounts, it is easy to know one's balance at any moment, and it is just as easy to pay it before the statement cuts as afterwards. For me personally it is easier to pay asap rather than wait for the due date or the statement.
Most consumers don't put in the extra work to do it and so the small All Zero penalty affects few people.
(a) For some people, like me, it's not extra work. (b) For some people, who are extremely credit worthy, there are zero balances because they didn't use the card that month.
The more reasonable explanation for the AZ penalty is that it is a vestigial organ (as biologists would say) from a time (e.g. FICO 8 and earlier) when this was the only reliable way to determine whether a person was using his cards or not.
(a) I truly doubt there was ever such a time. (b) I don't see how the fact that one is using one's cards makes a person more credit worthy.
And that fact is relevant in assessing risk. If someone regularly uses his cards but has never been late even once, that is a more powerful indicator of low risk than a person who's absence of lates is caused by having never used his cards.
If someone has regularly used his or her cards and has never been late even once, that is a very powerful indicator of low risk. But whether he or she is using them now or not is no indicator at all of higher credit worthiness.
It's a bit like someone who uses his bathroom every day and yet it is spotless, compared with a bathroom that is spotless because the owners has been on vacation for the last few years: if you are trying to assess the likelihood that when he comes back he'll keep the bathroom clean it is the former behavior that you are going to care about.
As far as the open loan that is mostly paid off, it's hard for me to judge. It certainly seems to me as though that wouldn't have to be a marker for low risk, but then again I am not a statistician and didn't have access to the huge datasets that FICO's statisticians did when they developed and tested the FICO 8 model. So the simpler explanation is that they were able to show some value in risk assessment there.
In both cases the typical man on the street doesn't even know about these obscure corner cases -- he's usually surprised when he hears about them -- so if FICO is secretly in league with the Fat Cats of the Big Banks, then it's doing a pretty bad job about using its models to push people into debt. When someone doesn't even know about your carrot and stick, it's hard for that to be a significant influence on their behavior.
@Anonymous wrote:
@SouthJamaica wrote:If someone has regularly used his or her cards and has never been late even once, that is a very powerful indicator of low risk. But whether he or she is using them now or not is no indicator at all of higher credit worthiness.
SJ, you and I have gone to the mat on this topic before with respect to installment loans, but the same argument I made regarding installment loans can also be made with revolvers. As we all know with nearly all FICO scoring categories, the more recent the data, the more it "counts" toward your score and the more meaningful it is with risk assessment. Your utilization today means a lot more than your utilization from a year ago at the same time. A 90 day late payment from 6 months ago matters more than a 90 day late payment from 6 years ago, etc. That being said, showing the ability to make monthly payments and manage your accounts (installment loans OR revolvers) today/recently means more than this same data a year ago, 3 years ago, 5 years ago and so on.
I do think it would make sense if the algorithm could take into consideration to some degree "recent" revolving credit / installment loan activity that isn't happening currently, but what is happening currently IMO matters most... and, that's what makes the most sense in terms of risk. As I've said in the past, a lot can change in a year or two. One can lose their job/income, go through health issues, lose loved ones and incur expenses as a result etc. Basically, life can happen which could make the ability to manage accounts more difficult today than was the case in the past, so considering activity today is indeed important.
It's like a MLB starting pitcher that in 2015, 2016 and 2017 pitched to a 2.5 ERA all 3 seasons, but was on the shelf for the 2018 season. Now in 2019, we can't just expect that he'll pitch to another 2.5 ERA. That time off in between adds in an element of risk. I see the same thing when it comes to FICO scoring if you're talking time off from paying [as agreed] an installment loan or revolving accounts.
Well @Anonymous you and @Anonymous are very persuasive and eloquent; we now have the baseball analogy and the clean lavatory anology.
But I do not find either analogy persuasive.
A person who has demonstrated experience managing credit and has come out the other end, owing nothing, gets extra points for that -- not demerits -- in my book.
@Anonymous wrote:
@SouthJamaica wrote:A person who has demonstrated experience managing credit and has come out the other end, owing nothing, gets extra points for that -- not demerits -- in my book.
"Experience" comes in the form of age of accounts factors.
I don't think that past credit-managing should be ignored and don't think anyone is suggesting that... but I find it hard to believe that you feel that old data (old past payment history on closed accounts) matters more than current payment history. If you legitimately do feel that way, I'd like to hear your reasoning behind it. And, at what point does it not become relevant, if ever? How about an installment loan paid off 5 years ago and no open loans since then? 10 years? 15+? How about someone that hasn't used revolvers in 5 years... 10 years... 15+? At what point (in your opinion) does the past become less relevant than the present?
I do not think that past history is more important than current history. Current credit management history is of vital importance and of even greater importance.
But owing money is not a proxy for activity. Borrowing money and paying it off is activity.
Paying balances down to zero is a good thing. And not owing money is a good thing. And the converses of both of those statements are also true.
@Anonymous wrote:
@SouthJamaica wrote:I do not think that past history is more important than current history. Current credit management history is of vital importance and of even greater importance.
Agreed. So, if current history is of greater importance, the presence of a current installment loan or current revolver history (with a reported balance) matters more than the same from the past. Since the FICO algorithm takes into consideration with greatest signal strength what's going on now with one's profile, it makes sense that this current information is of greater importance than older information.
Borrowing money and paying it off is activity.
Paying balances down to zero is a good thing.
Agreed again, but it's all about what matters most now [relative to the past]. Perhaps you can provide an analogy to help me better understand your perspective, as CGID and I have provided analogies that seemingly haven't resonated well.
Maybe I'm not as creative as you guys 
I'll try.
If I see 2 guys climbing a mountain, and they both appear to be great mountain climbers, I'm not going to say the guy who is still climbing after the other guy got to the top is better because he's still climbing while the other guy is at the top in a lounge chair drinking orange juice.
If I see 2 guys who are great baseball players, 1 of whom is running at blinding speed after hitting the ball into the right field and pulling out a triple, while the other is just jogging because he hit a home run, I'm not going to say the guy who hit the homer is less of a ball player because he's not really running.
If I see 2 guys who pay their bills promptly and who manage their credit deftly, and one has a little bit of his debt left to pay off while the other guy has paid it all off, I'm not going to say the 2nd guy is less creditworthy because he doesn't owe any money any more.
Paying your bills down to zero is what you're supposed to do. People who do that are the most creditworthy.
If you do that, and then have to take on some fake debt to keep your credit score up in order to maximize your FICO 8 score, your FICO 8 algorithm is defective.
Unless of course the algorithm is not what it pretends to be, as the OP suspects.
And as I, I must confess, suspect as well.