This post is also a response to a comment that can be seen here. I am not haranguing replier. I sincerely appreciate the response because it brought up some salient points that I didn’t cover in my previous post on this subject.
Here’s an interesting piece of knowledge which I did not know:
Credit Karma gives you a Vantage score, Goldman was almost certainly giving you a FICO score, and vantage skews more towards people who are building/rebuilding credit.
When I read this, I am reminded of the fact that simply stating the way something works is not a way to justify it. Many people do not seem to realize this, and so the above may appear as though it is a rationalization for what I was irate over in my previous post, when in fact it is exactly the opposite.
If Credit Karma reports a Vantage score, then what is the purpose of the FICO score? Or alternatively, if the FICO score is the one that actually means something, why is there even such a thing as a Vantage score? It would seem to me that if the Vantage score is the one that nobody uses except for Credit Karma, then Credit Karma should put a massive disclaimer next to those numbers that says “also these are totes not the ones your banker is going to see, lol”.
If someone is checking their credit score, presumably they are doing it because they would like to know what their credit score is? Is that … am I being naive here?
If somebody tells you that your credit score is one number, but nobody who actually uses a credit score to make decisions gets the same number, then … you can see where I’m going with this. It’s misleading at best, blatant fibbery at worst, and no matter where it falls on that spectrum, it always has the quality of being worthless.
“Here are some numbers we made up. They’re not the real numbers, but we promise that they’re related to them in some way or another. If anyone else asks for your credit score, these probably aren’t the numbers they’ll get, but they’re the ones we’ll give to you.” Seems like a really bad pitch, but perhaps I’m just asking too much of the world.
Speaking of asking too much of the world, here’s the second part of the response to my original article, which is something that I did know, but that I probably didn’t go into enough detail in eviscerating:
As for the credit dip after you bought your phone, it’s possible you bought it the one day between billing cycles that your card reports to the credit agencies. Let’s say your billing cycle is the 11th through the 10th …
The user then goes on to do a terrific job explaining how this works. I recommend reading it (link at top of page) if you’re into billing cycles and whatnot.
Who cares though? As I said, stating the dumb reasons why something dumb happens doesn’t make it any less dumb. If someone chose to implement and operate an idiotic system, then explaining that idiotic system doesn’t justify its existence, it just exposes how idiotic it is in the first place.
The state of affairs would appear to be this:
- There is a “real” credit score out there that’s actually used for things, but that’s not the one that Credit Karma gives out.
- This “real” score is updated each month, but somehow the Credit Karma score is able to change every week, without actually reflecting what the “real” score is doing.
- All the major American financial institutions, who handle all the money, are able to tell me — the mere consumer — what my bank account balance and credit use are, within seconds of those things changing, yet they can’t figure out how to inform each other of this information.
- This would mean that, thanks to the wizardry of electronic stuff, I can wave my wristwatch vaguely in the direction of the self-checkout system at the grocery store, my watch will tell the machine how to get into my account and take the money it needs for my can of beans or whatever, and then about 1.2 seconds after the checkout is done I’ll get a notification on my phone that I’ve just spent x dollars on beans, but the companies that aggregate my credit score are only capable of updating their information once a fucking month.
Point three there is pretty hilarious to me though. I mean, it makes me angry, but it also makes me laugh.
Think about this! I can make a purchase with my credit or debit account at any time of day or night, online or in person, and that transaction can be approved or denied because the account balance or credit utilization is accurately known. This can be done in a fraction of a second.
I don’t mean “this can be done in a fraction of a second” as in, it’s technologically feasible and we have the ability to make it a reality. I mean that it is a reality already. We see it in the world around us every day. It’s one of the many things that computing and networking have made possible: the ability to reliably make transactions because information can be stored and retrieved at what is on the order of the speed of electrons in electrical transmission lines (fast), or even at the speed of photons in a fiber-optic cable (literally the fastest speed in the entire universe that information can possibly travel within that medium).
All of this is possible, but TransUnion can’t get a bead on what my credit card utilization is until a monthly statement is faxed to it? What are they doing? I said faxing as a joke, but are they literally printing out paper documents and faxing them over to each other? Are there really like six guys in white button-ups and black skinny-ties who read these faxes and quickly key the entry into an Excel sheet?
How can I sum all of this up? I suppose I’d start like this: the concept of a credit score is that there is a number which is based on a variety of quantitative facts about your financial history, and which is calculated using a formula the result of which determines a representative description of your ability to take on debt and meet obligations.
That’s the idea that’s floated to us, and it’s the concept upon which we, as consumers, have more or less begrudgingly accepted the idea that this system is fair. We are operating on the assumption that the data is accurate, timely, and consistent. If that is the case then sure, this is a system which provides some value, some insight, into the credit-worthiness of an individual.
In relation to the efficient-markets hypothesis, credit scores would be a good thing because they would be a form of reliable information. They could be used to make informed decisions about things with an increased amount of certainty.
Basically, all the financial institutions working together to provide a credit agency with all of their customers’ information would be a good thing for all of these players. The initial investment would be small, too, because as online banking proves, they’ve already got the ability to do it.
Yet, here we are, in a world where I’m supposed to simply accept that of course my credit score (the real one, not the one anyone tells me about, obviously) is only updated once per month, so clearly it’s not abominably moronic that your credit limit was reported on the one day of the calendar month in which your account balance had been corrected but not posted.
Going back to the efficient-markets hypothesis, the question becomes why would banks use this?
If I were in charge of a bank, and TransUnion came to me and said “Hi, we’re a startup and what we do is aggregate payment history and credit utilization for individuals. We use this information to assess how likely a person is to be able to meet debt obligations, that way you can make a more informed decision about whether or not you’d like to give a loan to this person, because now you’ll know if they default on other loans, or have a lot of debt stacked up already.” then I would be intrigued. This is valuable to me!
I’d ask them what the latency on this was. If they said “We get new information once a month.” I would be drastically less interested.
“But wait,” they’d say, “don’t give up just yet. See, the thing is, how often do you really need to know what a person’s credit looks like, right? It’s not like your customers are coming to you for a mortgage every other day, haha. Most people don’t acquire new debt more frequently than a month at a time, and those who did … well, they’d have a very low score on our books.”
This would seem to be a very good explanation at first, to me. Well, to the real me I’d see through the garbage before they finished talking, but we’re assuming that I’m someone who runs a bank, so clearly I’m not particularly intelligent.
It would appear that the financial institutions of the United States all had the same response, and it went something like this: “Oh wow, good point! This is the most astounding idea I have ever heard. We’ll take it. How soon can we start getting these credit reports? You’re the best. We’re going to begin using this immediately as more or less the sole piece of information that will inform our decisions on whether or not to throw money at people. I’m going to make sure we begin integrating this as the only metric in our automated systems as soon as possible. You’re going to save me, and our loan officers, so much time!”
Now, let me tell you how I would respond myself, as evidence of the sort of thinking that proves I am more qualified to be in charge of a bank than actual bankers appear to be: “Are you fucking kidding me? Get the fuck out of my office with this kakameme bullshit, you mamzer; you buffoon; you absolute kadokhes! You must think I’m fakakte. You think a number that updates once a month is helpful to us, making loans? How many credit cards could this person open up in that time? Their credit utilization could go from nothing to 100% in the course of ten minutes and you want I should derive information about a $400,000 floating-rate mortgage loan from information measured in calendar months? Wait, wait … hang on. Are these calendar months, or do you using a rolling month?”
“Um, I’d have to get back to you.” The terrified little pischer from the credit-rating agency whimpers.
“No, you’re right, that was an unfair question.” I say, more calmly, leaning back in my luxurious office chair, kicking my feet up on my fuck-off huge mahogany desk, lighting a stogy. “But the fact remains, this whole business of being monthly, it’s no good. I mean here, what if I were to have alright credit, and then I max out all my lines the day after the reporting happens? If I went into the bank, they’d think I had the same credit I did earlier that morning before I bought a new private jet and an island, wouldn’t they?”
Quaking, and now devoid of any confidence in the product he came here to hock, he tugs at his collar and squeaks, “Well gee mister, I guess when you put it like that, we really would be reporting useless information. But what are the odds of that happening?”
“Alright,” I take a puff and muse, “how about this then. What if I paid down all my debt overnight with the bequest of my grandfather (of blessed memory), and desperately wanted to get in on a new stock offering that I know from information that isn’t of the insider sort, because I am a banker and I don’t play that kind of funny-business, and I want my bank to triple the line on my personal loan so I can take a bigger long position — would my loan agent get a score based on the old data, or the new?”
“Well sir, um, technically they’d … it would … I mean it would depend whether or not the monthly data import had happened yet.”
“Bupkes.” I, say, in a cloud of expensive cigar smoke.
“What?” The shyster asks.
“Bupkes. It’s what your dipshit little scheme is worth. But hey, given that this has a very minor useful quality in assessing things, taken into consideration alongside actually helpful information we get from other sources, then you’re welcome to leave any brochures you might have about how to obtain your so-called ‘credit reports’, provided that they are free of charge. Otherwise, please, don’t let me detain you.” I gesture casually towards the door while jumping onto the company-wide Slack chat, knowing it makes me look busy, and making fun of this schmuck who just tried to pitch me information that’s a month out of date.
Equifax never gets off the ground, the founders instead fulfill a function in society as presumably people involved in multi-level-marketing schemes because that’s about the only other place they’d have valuable skills to contribute, and our bank continues to make wise lending decisions informed by relevant information instead of snake-oil. See? I really ought to be running a bank. If anybody’s looking for an apprentice and inevitable successor, I’m available.
Look, anyways, the point of that whole spiel was this: you can explain how credit scores are calculated, and you can explain that there’s a difference between a FICO and a Vantage score, and you can explain about how months work. These are aspects of how the credit-reporting agencies and methodologies function; they are not, however, acceptable rationales for why they ought to operate that way, or why their operating that way makes them useful or valuable.
All the evidence points to credit scores being without value. They are slow, they make it willfully difficult for a person to know what theirs is, and they can’t even bother to agree with each other.
Banks have the ability to tell me at any moment what my accounts look like. In fact, they both have this ability and they implement it. There is not a single reason why that can’t be replicated to the credit agencies, and in turn that means there is not a single reason why the credit agencies cannot replicate that information into a real-time, moment-to-moment representation of a person’s credit-worthiness.
A credit report which is based on real-time data and uses a specific formula would provide the financial markets with truly useful data. The value to the market would be spectacular, and it saves all the individual banks and credit-card issuers and loan providers and everyone else who has a need to assess an individual’s credit-worthiness with a centralized location for both retrieving that data, and reporting their own when they open an account. That last part is important, because otherwise you’d need all those places to interoperate, and even with modern technology that would be a technical and logistical nightmare.
There is tremendous value to a Platonic ideal version of a credit report. It coordinates information from disparate sources in order to provide querents with a real-time snapshot of a large amount of data about an individual that is directly relevant to making an assessment of that individual’s credit-worthiness.
Calling it a Platonic ideal is my own tongue-in-cheek joke though, because whereas a Platonic ideal is a thing which cannot exist perfectly outside of the realm of thought, the technology to do this is already humming along perfectly smoothly in the electronic home-banking systems that have been in use for at least a decade. Sure, it’s a Platonic ideal in as much as no information system is ever going to be 100% accurate, with 100% uptime, but as far as building a credit reporting system that actually provides real data in real time, there’s no reason that can’t be done. Hell, you could have it in beta within a week if you got a crew of 10 talented cloud services engineers together, and even if they were intentionally sloppy it would likely still be at least as careful with customers’ private information as the current credit-reporting agencies are, if not moreso.
Which is what makes it all the more frustrating to realize that the banks have decided instead to buy the shoddy, patchwork, entirely ineffectual snake-oil that Equifax, TransUnion, and whatever-the-other-one-is-called are schlocking.
I dunno, if you’re a big bank and you’re tired of basing your decisions on consumer credit on a score that’s neither consistent nor real-time, then give me a call. I’d love nothing more than to start a project to correct this problem. The cost would be minimal, shared by all the stakeholders, and it wouldn’t even require much since your systems are already doing all the important stuff already — you just need a place to aggregate it. The biggest cost would, and this is depressing to me to state, but the most expensive part would probably be the lawyers’ fees in making sure everything is compliant with relevant privacy laws.
As for the rest of us, I guess all we can do is be irritated that this idiotic system exists, because legislation sure isn’t going to fix it. Equifax leaked millions and millions of US citizens’ data to god-knows-who and they got off more or less just having to provide a shoddy credit-monitoring service to people for a year or something like that. Obviously nothing is going to force them to suddenly start reporting accurate, timely data, or to make the data they have on individuals available for free. Sure, they’ve got files on you, but it’s none of your business.
Then again, I guess if you want your Vantage score (which, remember, isn’t the one that anyone actually uses and so has no value whatsoever that I can discern) then you could sign up for Credit Karma. They’ve got numbers, and whether they’re relevant or not, they’ll give them to you.
You get what you pay for? Is that the takeaway here? If so, TransUnion and Equifax should be the ones paying financial institutions to use their trash.
n.b. — This article has a lot of rhetorical questions in it. If you read it and felt the need to say something like “Actually, it’s not just Credit Karma that use the Vantage score, places like x also … “ then don’t bother. I’m not actually interested in which primitive tribes use FICO and which are denizens of the Vantage region. I reiterate: if a credit score were a thing that had a meaning, then just like any other meaningful measurement (like temperature or mass or volume) it would always have the same meaning, even when translated from one system to another. I might say it’s 108 degrees fahrenheit outside, and my German friend would say that it’s about 42 celsius, but we’d both be getting the same result: it’s fucking hot out. This is precisely the sort of translational equivalence that FICO and Vantage do not have: a good Vantage is not necessarily a good FICO, so clearly either both are useless measurements or one of them needs to be scrapped.