@Slabenstein wrote:
@SouthJamaica wrote:Actually, the credit report does show if the person was actively managing the account that month. It has nothing to do with the balance on the reporting date.
And the FICO algorithm does take into account historical data. In fact it looks back many many years, at such items as oldest account, average age of accounts, dates of negative items, dates of inquiries, dates of newest accounts, etc etc.
Yeah, but afaik it doesn't seem take that stuff into account for how it scores util or %/# of account with balance metrics, except maybe for weighting. I understood the AMA comments to mean that the AZ penalty works similarly, whether or not any of it should work that way. But that was just my reading of their vagueries.
Precisely.
But the absence of utilization at a given moment has no bearing whatsoever on your proven ability to properly manage credit. It may, however, reflect a desire on your part to avoid owing money. Which may make you a less profitable type of bank customer.
As between two completely equal, 100% financially strong and totally creditworthy customers, the one who likes to borrow is more profitable than the one who prefers to avoid it.
Looking at it from the perspective of risk v. reward
-it has no meaning in terms of 'risk'
-it may indeed have plenty of meaning for 'reward'
Barclays once put out a blog post, for its Ring card customers, which demonstrated how CLI and CLD policies were based on a risk vs reward strategy. People with too little utilization were as likely to be denied CLI's as those with too much utilization. Barclays was looking for the sweet spot, where people were borrowing, but not too much.
@Anonymous wrote:Right and the use of TD will in theory "fix" the issue. But, without TD being used, I don't see why it's all that difficult to understand where Fico is coming from. A single moment in time snapshot whether you're a Transactor or not doesn't differentiate between between someone that's at AZ for the first time in years or someone that's been at AZ for years. If those at Fico have determined that individuals that are at AZ for extended periods of time (clearly not using credit) are statistically slightly more likely to default than someone that reports small balances there's no reason they shouldn't be scored as such with an AZ "penalty" as it is called around here. If a single moment in time snapshot of another individual that just happened to report AZ results in the same penalty, it is what it is considering the current algorithm.
If Fico 10 "fixes" this issue by using TD and this penalty goes away, what will the narrative change to then? That it used to be about banks making more money, but all of a sudden they don't and that motive is gone? That sounds a bit silly to me.
FICO algorithms use trended data all the time. Their "snapshot" knows how old every reported account is, what its month by month record is, how old the inquiries are, how old any negatives are, etc. And it even knows whether your account had activity this month, regardless of whether it has a balance on the reporting date.
Your assumptions as to how people use credit cards do not take into account the wide swath of behaviors in the world. Some people use credit cards but don't use them every month. Some people use Chase cards, which report a zero balance whenever it occurs. Some people pay the charges they've made before the statement cuts.
And folks who use a lot of credit cards every month -- and don't pay them off before the statement date -- even if they always pay in full, are losing points for the number of accounts with non-zero balances and for the percentage utilization.
You keep defending FICO's all-zero penalty and no open loan penalty, even though the representatives from FICO have basically admitted that there is no logic behind them other than data showing a low balance is "slightly" less predictive of risk than a zero balance.
Well, in case you haven't noticed, the typical 30 point penalty for having no open loans, and 20 point penalty for having all zero revolving accounts, total 50 points on average, which is not "slight" by anyone's definition.
@Anonymous wrote:I don't really buy it SJ. The amount/percentage of people that PIF monthly and offer no profit (from interest) to banks is not directly correlated to AZ penalties. Plenty of people that PIF, in fact most get their statement, pay it off and go on with life without thinking for a moment about their reported balance and/or impact of it on their Fico scores. These people often in just naturally using their cards and not micromanaging them like many on this forum do will report non-zero balances monthly. The AZ penalty therefore IMO wouldn't serve the purpose of what you suggested earlier. To me I'm perfectly fine in the absence of TD usage saying that the penalty is due to the algorithm seeing no revolving credit use at that moment in time. If with the introduction of TD in Fico 10 the AZ penalty goes away it just further proves this.
Of course the penalties would not really serve the purpose of insuring profitability, and are easily gamed. I didn't say they were smart. In fact I think they're stupid. The avid MyFICO'er who doesn't want any installment loans or credit card bills knows to take out a meaningless unneeded loan and then pay it down to 9% or less to avoid the no open loan penalty, and to let their Netflix bill report on one bank card each month to avoid the all zero penalty.
This is what I know and what I don't know.
I know
1. there is an all zero revolving balance penalty.
2. there is a no open loan penalty.
3. FICO's publicly stated rationale for these penalties is that low balance accounts are slightly less risky than zero balance accounts.
4. their publicly stated rationales do not stand up to scruting.
I do not know that there are ulterior motives for the penalties, and if there are ulterior motives I do not know what they are. On these I am only guessing.
@Anonymous wrote:
@SouthJamaica wrote:FICO algorithms use trended data all the time. Their "snapshot" knows how old every reported account is, what its month by month record is, how old the inquiries are, how old any negatives are, etc. And it even knows whether your account had activity this month, regardless of whether it has a balance on the reporting date.
Right, but it doesn't use TD with respect to the AZ penalty, which is what we're talking about. My question was that if Fico 10 "fixes" this what happens to the bank profit narrative? Those that believe that (or some spin on it) will then say that banks suddenly don't care about the extra profit?
As I told you, my theory on the ulterior motives is purely a guess.
Also let's clarify the "30 point penalty" for not having an open loan. The ~30 points is related to going from an almost-paid-off loan to no open loan. Aside from those gaming the system with the SSLT, you'd figure that with naturally paid down loans, around 90% of people would have their loan sitting above the sweet spot break point and 10% of people below... give or take. That being said, I think the "30 point penalty" is over stated ~90% of the time.
Not in my case. In my case it exceeded 30 points. 30 points was my attempt to provide an average.
And "no logic" for the reason penalities shouldn't be used in the same sentence with anything about risk as predictive risk is the logic and frankly what the entire scoring system is about anyway.
If you're satisfied with the explanations given in the AMA thread, fine. To my mind they make no sense whatsoever.
FICO 8 penalizes scores when all revolving accounts are zero. The points penalty is profile specific. It is just a fact.
Why ask why?
