For years now, Experian has listed the following as one of the top three factors pulling my score down I currently have the following Credit Cards: * Target - 200 * JC Penney - 500 * Capital One - 500 * Barclays Visa - 800 * Credit One - 1000 * Tribute - 1000 * HSBC - 1300 * Direct Merchants - 1500 * Lowes - 1500 * Home Depot - 4000 * WaMu - 5000 What is a limit sufficient to get past this - and does anyone know how much my score might be suffering because of low limits? I really have more CCs than I need and I could start weeding out the small ones over time if they hurt that much. I know everyone says never close accounts, but if it drags down my score more to have a $200 account and they won't raise it up to $5000 or whatever is not a "low limit", then maybe it's best to close it. Anyone out there in the know?
The most recent real FICOs are EQ-682, EX-699, and TU-706. I have a veritable graveyard of bad credit - 2003 foreclosure, 2001 judgment, and a handful of collections fromt that 'era'. I'm significantly cleaner than I used to be, but getting better. These limits are the best I've been able to do with preapprovals and the benevolence of existing creditors to raise lines. I'd like to know what they consider a low amount so that I can weed out the stragglers as I build my credit. If having a $200 Target card costs me 10 points while it's age only helps me marginally, it's best to dump it - particularly if they're not likely to raise the CL to an acceptable limit.
Ok there are no major players in your stack. You need to determine your chances of getting a big player and doing it. Chase,Citi,Discover,or Amex. Closing accounts will hurt you if they have age to them. Another thing hurting you right now is that is a thin list of positives. So to start,continue attacking your negatives rigerously and come up with a plan to add a major account.I guess this site has a creditpull database somewhere here that would give you an idea of which report they will look at. One thing to remember is you can get better offers calling the creditor and speaking with a underwriter.
First of all, you cannot trust "robot advice". Whatever your score is, automatically generated advice will give you something that it will claim is negative. The reliability and quality of that assessment of your credit factors is questionable. Although you might have higher scores if you happened to have higher limits, it is not clear that if you have a lot of small limit accounts, and have already take any inquiry or age hit for opening them, closing any of them will provide any net benefit to your scores and leave you any better off in a year from where you would be in a year leaving them open, allowing them to age, and getting any limit increases you could. That is in fact the realistic choice you are talking about, not instantly sqeezing some miraculouos boost in scores out of some short term action. It is possible that some action in closing some accounts now might slightly boost your current scores, but it is also possible that your scores in a year might be higher if you didn't. Stick to the basics, built up increased credit limits over time, and increasingly older accounts over time.
An automatic requirement in generating a credit score is that they have to provide both the POSITIVE and NEGATIVE factors which influence the score. If you're even 1 point below a perfect score, there will be negative factors. Consider this one that would do more 'harm' by correcting (by closing the 'low' CLs), than good. Think of the serenity saying "Change the things that you can change, Accept the things that you can not change"; the key is to have the WISDOM to know the difference. Worry about the true negatives, and ignore the false negatives. There will always be negatives, some you can fiddle with, some you can't, this is one that except for obtaining more cards with higher limits, you can't.
My mortgage hasn't yet reported which is hindering me from getting offers from the majors. They should report this month. I don't have that much age on any of my accounts. Here's approximately when each was opened: HSBC - 4/03 - (CL-1300) Credit One - 7/04 - (CL-1000) Target - 4/05 - (CL-200) Capital One - 4/05 - (CL-500) Direct Merchants - 3/06 - (CL-1500) JC Penney - 9/06 - (CL-500) Tribute - 10/06 - (CL-1000) WaMu - 11/06 - (CL-5000) Barclays Visa - 12/06 - (CL-800) Home Depot - 3/07 - (CL-4000) Lowes - 5/07 - (CL-1500) Most of the things giving my profile age are derogatory items like a foreclosure on an account opened in 1993. That will continue on record only through 2010. I've got a judgment from 2001 that will fall off next year. I have 5 paid collection accounts from 2002-2003 that will be dropping off between this month and next year with one hanging in through 2010. I do have one satisfactory Amex card opened in 1994 and since closed that will continue on my report through 2012. I have 22 accounts on record including the judgment, foreclosure, 5 paid collections, 4 satisfactory closed accounts, and the 11 satisfactory CC accounts above. I have the following new accounts yet to report: Mortgage of $262,000 opened 12/06 Installment loan of $45,000 opened 4/07 HELOC of $67,000 opened 5/07 I'm just wondering if a $200 Target card or a $500 Cap1 card that will never raise my CL will hurt my score down the road. If not, then it's no problem, but if it might, I figure now is the time to address it. Once my old derogatories start falling off, my age is going to be dramatically affected anyway - moreso than closing an account that's only 2 years old. Also, what's the effect of sub-prime cards? When my mortgage begins reporting, I'll probably start getting prime solicitations. Where will I be down the road with them still in my hand? Will they hurt, or will having a couple of prime cards cancel that out? Does anyone know if any of the sub-prime companies have a prime side that I could eventually convert my cards to?
As far as I know, FICO does NOT discriminate against "sub-prime" cards, in favor of "prime" cards. Frankly, I don't know how they could build this into their model in a way that would not get them in trouble with FTC for unfair trade practices, or with individual lenders for hurting their business. We would have a major scandal, lawsuits, investigations, or at least some hint that lenders are jockeying to better position their cards, and we see nothing of that. All FICO can look at is what it sees in your credit file. It cannot do some massive evaluation based on who is reporting, by name, even if a human might be able to extract useful predictive information thru such a process. It can only use the coded information presented, and it can't even guess much about whether any particular revolving accounts are "prime" or "sub-prime", since interest rates are NOT reported. The CC companies probably do NOT want interest rates reported, just as some balk at reporting credit limits, as that would allow their competitors to cherry pick their most profitable customers. The observation that people with lower scores tend to have what a human recognizes as "sub-prime" accounts, does not mean that FICO scores those "sub-prime" accounts lower than it would a "prime" account reporting identical information.
However, it is one of the FICO "positive/negatives used in scoring, whether you have a "status" card. The model does give weight to premier cards, as an indicator that you meet their model of "creditworthiness". I'd have to look up on my reports the exact wording. This implies the model is "circular", that is it bases a score on previous set criteria regarding credit worthiness. Sort of a "bye" into a higher level. I think this is how they do not run into trouble with the FTC, or unfair trade practices. Credit limits also play into this perception of credit worthiness by the FICO model. It seems to give weight to higher credit limits, inferring that it does try to "read" the data. But the fact I find interesting is the discrepancies between positive/negative factors between the three CRAs. The overall "same" information does get construed differently, resulting in different scores and "factors". After seeingt he CDIA Manual, I think it has to do with these "codes" and entry. What may "read" the same(as written words) on a credit report, may not read the same on the "numeric code" review.
But if there is no "code" in the CDIA Manual to encode "prime" or "status" card vs. "sub-prime", then there is no way for FICO to discern this. It would have to work indirectly from size of credit limit and history. To some degree it can do this, as for example, in the manner in which consumer files containing mortgage information are evaluated and scored somewhat differently from files not showing mortgages, but this is consistent with the sub-population adjusted scoring models they claim to use. There are thousands of potential lenders, each with multiple card products. There is no way a scoring system could encompass this breadth of "information", without either the willing assistance of the data furnishers themselves (by "codes"), or by only classifying some subset of all data furnishers. In many cases, there might not even be a large enough sampling of information available to evaluate risk based on data furnisher. And if data furnisher was used in evaluating quality of reported data, why is it not used to invalidate collection data furnishers known to report erroneous, false, and re-aged data? Collections, revolving account, installment account, mortgage: these are all "coded".
I don't know from firsthand experience, but the most knowlegeable person I have every spoken to on the matter of credit scoring was a broker in NC. Inasmuch as anyone can understand anything about FICO, he seemed to understand it much more than anyone else I've talked to. He told me categorically that the FICO score considered prime cards vs. subprime cards when it came to carrying a balance. His explanation was that a person with a subprime card was statistically more likely (to some extent I can't recall) to have a 90 day late when carrying a balance on the card than someone who carried a balance on a prime card. He specifically mentioned Household, Orchard, Providian, and he said Cap1 had a subprime card as well as a prime card. He told me that carrying one of those cards was OK as long as I kept it paid off and didn't carry a balance. He contended that my score would suffer more when carrying a balance on a Household than it would carrying a balance on Discover, Chase, etc. Again, I have no firsthand knowledge, just what he told me and he 'seemed' to know his stuff.
He may still not have first hand inside knowledge about Fair Issacs algorithm. His knowledge is probably based on his experience as a broker, and observations over time, and both hypotheses (sub-prime accounts lower scores, vs. lower scores tend to have sub-prime accounts) are probably equally consistent with what he observed. Two conditions that correlate closely does not tell you which causes which. There is one aspect that FICO could use to weight sub-prime accounts more heavily, on average. If, for example, it separately weights each account's debt to available credit, in addition to a total debt to available credit, as has been suggested by some, then carrying any significant balance on small limit cards would count negatively more than their representation in the total available credit pool. Since people with many sub-prime accounts tend to have many small limit accounts, this would pull them down, even if other large limit accounts with little use might reduce the total debt to available credit ratio.
The unfair trade practice / anti-trust problem that could be created by labeling certain accounts as "prime" vs. "sub-prime" is that by so doing, a given lender would be broadcasting to other lenders that he is giving bad terms to a consumer that are NOT based on credit history, and that they could in turn take advantage of that knowledge to do the same, resulting in less competitive offers to that consumer, and higher profits to all his lenders, in effect by agreement. In effect, it is a mechanism to allow collusion in the marketplace, at a similar level to "red-lining" and other practices that have resulted in regulatory intervention. If state AGs, FTC, or class action attorneys got even e whiff of this, they would have a field day. The fines, attorney's fees and penalties would make them very happy.
This could be a valid explanation. I personally experienced a 20 point drop in my FAKO one month where I had utilized one card on a business trip and used about 90% of it. My overall utilization was less than 30%, but on that one card - 90%. It would be a grievous error on my part to assume that FICO works in any way like FAKOs, but... if it did, that experience would dovetail into your explanation.
You can only keep secrets when you don't tell anyone. Kickbacks, and side deals may be rife in the mortgage and title insurance business, but if similar collusion were occurring widely in revolving credit, I think some part of it would come out. Even Providian's posting delays and payment shredding came out.
Couldn't they operate within fair trade practices if it were based on a consumer's credit history? I got a Household card when I had a score that had to be in the low 500's. No other company would touch me back then. From what I've heard, that's their niche. If that's the case, wouldn't it be easy to label HSBC as sub-prime without breaching fair trade? Every lender has their criteria for lending and they're free to set it at whatever level they want as long as they apply the criteria across the board to every customer without preference or prejudice, right? If HSBC made it a practice to offer cards to people with scores between 550 and 650, and Chase only offers cards to people with 700+ couldn't FICO segment them based on the reporting agency's ID? And would FICO be violating fair trade to do so if credit history was at the base of the segmentation?
Both of them have a variety of products, and various levels of scores to qualify. The more fragmented the market with product variations, the harder it is to infer meaningful information in an ever changing market. By the time you could determine correlations allowing you to infer different lending products matched different customer credit risk, new products with little risk history would start making your inferences obsolete. It would be a never ending race, playing catch-up. Once the rate of change of the "system" you are trying to track starts exceeding the time it takes to extract useful information out of the noise for the sample sizes you have access to, you reach a limit in how much confidence you can ever have in your estimate of the system from your sample population. Same problem comes up in statistical process control, or adaptive control. You can only do so much, and if you place too much "trust" in what can only be a small sample, you end up doing a random walk forever chasing what you believe is an ever shifting state of the system. Alternatively, you overadapt your model to the dataset you used to construct and validate it, with the result that it acts as designed with files similar to what you validated, but may not work as reliably with dissimilar files. The net result is you can only be in the approximate region of optimal, and you simply cannot do better. The extra "information" you are trying to take advantage of to estimate better doesn't really buy you anything, so as you evaluate it's contribution to your best estimate of risk, you conclude you might as well ignore it. Adding some factors to your model may increase their accuracy, but adding other factors may actually decrease their accuracy. If it adds nothing, you have to weight it as such to produce your best estimate. Note that I am not a statistician. The above is what my gut instinct tells me, based on experience in other areas.
This is true, but historical data may be based upon the strata of "generalities:, such that there is a general concensus that an AmEx card (overall) is harder to get than a Household card, et al.....and I think if "default data" is provided by lenders/issuers, than there is a reasonable presumption that you would expect higher "defaults" (in general) on Household or Cap1 cards than AmEx or a "Visa Signature" card. FICO does use this default data at the core of its algorithm, and there is an empirical soundness to this scenario. Once again, a throwback to the "rich get richer" cliche. Cap1 would be a prime example of "noise" in this calculation, they have tried to cover the market by issuing cards from the most sub prime of secured cards with a few hundred dollar credit limits, up to offering the most premier cards with the lowest rates and highest credit limits. I assume there may be something in the coding of an account number to signify "qualitative differences, and perhaps the FICO model can read these. You are correct about the data, and correlations, and bluntly, many times the algorithm must "chase" actual results data. A clear example is FICO's new model coming out in September for people with little history. Based upon all we know, and infer from the FICO model, I am anxious to see if there are any "shifts" in overall scoring as the subprime mortgage market issues unfold. "Predicting" the model, there should be an aggregate drop on the "average" FICO score, and the "buckets or scorecards" should change.
CapOne was the type of lender I was thinking of, with a full range of products across the market. If FICO weighting were adjusted for recent loan default experience, I would expect recent mortgages to be weighted more negatively, but less difference on older mortgages, say after a year. Many sub-prime defaults have occurred in less than a year, sometimes involving cooked applications, etc., while even though there may have been a shift in recent years toward ARMs, the bulk of older mortgages out there are probably still conventional. The effect of defaulted property on market prices is probably primarily a risk for recent mortgages, since if you go back further, the purchase prices would have been lower, and therefore the loan to current value would also be lower. In short, the major risk to lenders from the sub-prime debacle is probably still in very recent sub-prime mortgages, probably within a year. That is not to say that borrowers with mortgages back 2 or more years might not suffer loss, only that they have some equity to cover it, avoiding it's transfer to the lender.
