Hacker News Reader

Comment by harryh | Hacker News Reader

The math behind credit scores is quite good. They're very effective at predicting credit risk.

This statement, which is currently greyed out, is an objective statement of engineering fact about one of the most important technologies introduced in the last one hundred years.

I hope I have bought enough street cred to say the following by ghostwriting hundreds of letters to the credit reporting agencies to fix their problems, which are numerous and essentially inevitable given their model and present operations.

Credit scores are important because they allow banks to do standardized, automated underwriting for effectively free at 2 AM in the morning, which is what makes credit abundantly available in the United States and one of the reasons why it is so cheap to the middle class. (This routinely escapes the notice of people in the middle class, but the ordinary operation of banks is to advance well-organized people money for free and pay you to be chosen to do so, partly due to credit scores making this a derisked proposition and partly due to interchange revenue.)

Credit scores decimated the costs of unsecured consumer loans, as is readily observable by seeing what loan availability, pricing, and the "credit box" looks like when they're not available for underwriting. (This is a term of art in consumer credit that I provide for your future Googling pleasure, not having enough time to explain it at the moment.) Compare the cost on a cash advance on any card in your wallet to a payday lender. That delta is substantially (not solely) due to credit scores, both in improved understanding of default risk and in reduced operational cost (of underwriting and servicing, both enabled by the scoring infrastructure).

Credit scores are an important justice-enhancing technology because they make the inputs to an underwriting decision objectively observable and for the first time in history those inputs provably do not include race, religion, etc.

There is an argument which is not immediately dismissible that credit scores are based on borrower behavior which reflects the socioeconomic realities of the United States and therefore somehow effectively encode race, but it is an obvious improvement that banks now mechanically reach the same decision on similarly situated white and black borrowers. Regulation did not make that happen. FICO did.

I think well-off computer programmers should understand that there is a societal tradeoff which buys "middle class black Americans have the same access to credit you do" at the cost of Equifax knowing your credit limits; understand that if you are advocating for rolling back the second you should accept that ceteris paribus we will experience material disimprovement on the first.

(Obligatory disclaimer: personal opinion here.)

Sorry, I am not that confident about that. I live in Europe, and to my knowledge we do not have that kind of credit score companies (there are databases about people who have defaulted their debts, though), and to my knowledge it is banks themselves that do the credit evaluation.[1] And credit for middle class is to my standards pretty cheap enough here as well.

[1] More than happy to be corrected here with sources, I just have not ever heard about Equifax-like company in Europe

Of course they do (exist in Europe).

Most companies giving you a credit of some sort (banks, phone contracts, ...) are required to do vetting of you.

To help them do that, they use credit rating bureaus.

In Europe, you as a private person just don't know it and don't have access to your own rating. But you are likely to be rated anyways..

Here is one danish example (at least three exists)

https://translate.google.com/translate?sl=da&tl=en&u=https%3...

> In Europe, you as a private person just don't know it and don't have access to your own rating. But you are likely to be rated anyways..

In the E.U., there are no private organisations who can process your personal data without making it available to you on request.

In Ireland, there's a state-run Central Credit Register, to which lenders must submit any consumer credit agreement above €200, and which they must check by law before issuing credit. Anyone can apply online to get their credit history from the CCR, and make submissions to correct inaccuracies.

There's also the Irish Credit Bureau, a private organisation with a similar purpose. They are also required to reveal data held about you on request, and correct inaccurate data held.

In the US you can request your credit score once per year from each credit rating angecy at no cost.

The thing you can request once per year from annualcreditreport is your credit report, which doesn't include your score. Your score is accessible for free from websites like creditkarma or through most credit card online accounts, usually once a month although I have seen ones which update twice a month.

In Italy there is a Centrale Rischi, but is a government/public service (run by the Bank of Italy):

https://www.bancaditalia.it/statistiche/raccolta-dati/centra...

and I believe that in the end (now it is partial/experimental/in the works AFAICU) there will be an European Registry at the BCE (Anacredit):

https://www.ecb.europa.eu/stats/money_credit_banking/anacred...

Though there is no "public" access to the data in the Centrale Rischi, you can have it going in person to the Bank of Italy, the issue is that - if you find an error/mistake - it must be corrected by the bank/agency/whatever that made the erroneous entry, and this can sometimes be a nightmare, not so much with banks, but with smallish financial service agancies (that might have - in the meantime - changed property or going in default, etc.).

AFAIK the KreditRegisteret you linked to is required because Denmark, not part of Euro, does not have access to AnaCredit. So it's more of a tool for financial stability, and not private loan estimates. What is EU's AnaCredit?

> AnaCredit is a dataset with detailed information on individual bank loans in the euro area. The name stands for “analytical credit datasets”. The ECB launched the project in 2011 – together with the euro area and some non-euro area national central banks. It uses data and national credit registers to achieve a harmonised database that supports several central banking functions, such as decision-making in monetary policy and macroprudential supervision.

In Denmark I'm mostly aware of RKI, which seems to be owned by Experian now. If you fail to pay your loans, you end up on that bad payer list. Otherwise getting a loan is a private matter between you and your bank unless you want to use a third party lender

I'd like to get some further sources. Below is what Wikipedia has to say about Danish credit scoring[1] (emphasis mine) I know there are databases the store your information if you have defaulted your loans, but that is (to me) very different thing to actual credit scoring, which obviously any entity in credit business is and should be doing. I would be very keen to know (some of) the companies that sell actual credit scores of individual people in (mainland) Europe, and possibly shoot a GDPR request at them.

The Wikipedia snippet:

The credit scoring is widely used in Denmark by the banks and a number of private companies within telco and others. The credit scoring is split in two:

    Private: The probability of defaulting
    Businesses: The probability of bankruptcy

For privates, the credit scoring is always made by the creditor. For businesses it is either made by the creditor or by a third party.

There are a few companies who have specialized in developing credit scorecards in Denmark:

    Experian (generic rating for business)
    Bisnode (generic rating for business)

The credit scorecards in Denmark are mainly based on information provided by the applicant and publicly available data. It is very restricted by legislation compared to its neighbouring countries.

[1] https://en.wikipedia.org/wiki/Credit_score#Denmark

Banks in Norway use Experian to check your risk. Anytime you apply for a loan you'll get a letter in the mail from Experian a few days later letting you know who made the request and what their response was.

Does Experian response with an actual credit score as opposed to information whether you have defaulted some of your loans recently or not?

The UK is in Europe (for now) and has Equifax and the other credit reference agencies.

The UK imports all the shitty things from the US, instead of looking at what the european are doing: - No national ID, so identity theft is easier - For a long time, no pin code on credit card, so that theft is easier Basically they keep the stuff which trick the consumer, whereas in Europe they make law to protect consumer.

I really hope they leave the EU and stop bothering us with their shitty ideas, credit reference beeing one

>No national ID, so identity theft is easier

We don't want national identity cards; the Blair government tried to introduce them in 2006 and failed due to massive public opposition. We think it's a bit weird that other EU nationals are so tolerant of the government having a massive centralised database tied to a single token.

https://en.wikipedia.org/wiki/Identity_Cards_Act_2006

>For a long time, no pin code on credit card, so that theft is easier

Most of our debit card market is Visa/Mastercard, so we switched to Chip and PIN in 2006 - the same time as most international markets. Some European markets had their own local debit card schemes but these schemes were far from perfect.

>Basically they keep the stuff which trick the consumer, whereas in Europe they make law to protect consumer.

Last time I checked, we were still in the EU and still subject to EU law. We implemented the GDPR (and our regulator is one of the most proactive in actually enforcing it), we implemented the Consumer Rights Directive and we were well ahead of the curve in many areas, particularly distance selling and consumer finance.

You have drivers licenses and bank cards and passports already. Having a national ID just makes a bunch of stuff easier and identity theft harder.

A national ID card is a single point of failure. If I need to verify my ID for something important, I'll usually be asked for two forms of ID documents which may be subject to additional verification.

How do any of your examples 'trick' the consumer?

It’s effectively for free, except for the massive society wide risk of fraud that they’ve foisted upon all of us. The stuff looks cheap because, like any polluting industry, the price does not reflect the true cost.

Surely there’s some third alternative other than “incompetent credit bureaus enable fraud” and “minorities have bad access to credit because lenders are super racist”?

Also the existence of a centralized credit system increases the cost of not participating in the system. The lack of real competition (eg. for middle class credit) is what essentially leads to a polluted industry.

> It’s effectively for free, except for

Also except for the risk that financial problems will force a cardholder into accruing interest, and often that interest will get a lot worse if payments are missed a 5% card changes into a 22% card if more than some number of payments are late and some number may be just 1.

Why do you believe lenders are super racist?

Most of what you say is correct but it's wrong to say that regulation isn't responsible for race being absent from FICO calculation. It's fair to say that lenders would use factors like race in a heartbeat if it weren't illegal, but it is illegal to do that. If FICO were race-based, using FICO would be depending on race indirectly, but explicitly, to make loan decisions. Thus FICO can't be dependent on race.

I just wanted to say thank you, because this is exactly the kind of HN discussion that I find valuable (and increasingly rare in public discourse). It's so easy to get on the "credit bureaus are evil" bandwagon, especially in this story where Equifax's statements are honestly downright insulting, and your comment made me put away my pitchfork for a second and reflect on the benefits that credit scores give us. Yes, there are certainly arguments that one can have about the respective costs and benefits of credit scores, but it's good to know we can have informed debate about those pros and cons.

Very thoughtful and interesting. From the perspective of credit monitoring though, while they provide a social good with credit scores and risk assessment, they do so for a profit on the back of consumer information. For something like fraud monitoring, it seems almost an ethical imperative (in the social contract sense) to share information about potential fraud with the people whose information powers your business.

I’m sorry, but it’s pretty absurd to boldly claim that credit scores are provably objective with regard to race, etc. I’d recommend reading _The Big Test_ for another history of a supposedly objective measurement of merit (the SAT) which turned out to be highly discriminatory. Removing discrimination from a system is not as simple as removing such info from its inputs.

I had a recent 30 day delinquent payment reduce my credit score by 80 points. And this was because the letter in the mail was mid-delivered. Not that the credit card issuer, nor the credit agencies give two shits. This was after about 5 hours of calls ping-ponging between the two.

Credit scores change constantly. What your 80 point drop told creditors is "This person might not have enough money to pay their bills, or is careless or forgetful.". You are perceived as higher risk because, statistically, you are a higher risk. Those of us who maintain excellent credit do it in part by making sure our money gets where it's supposed to. I check online to make sure payments post, or hand in payments in person, or call in the rare case these days when there's no other way to check.

Fortunately, credit scores usually recover rather quickly from one-off delinquencies if you stay current afterward. And I bet you'll be paying more attention to your bills now, won't you?

The credit score is doing exactly what it's supposed to.

I was surprised how much they change constantly, for silly things like how much balance you have on your credit cards.

My wife & I both have excellent credit scores, and we pay all of our bills in full every month. Nevertheless, I've noticed a ~50 point swing in credit scores from month-to-month, all dependent on whether airline tickets, furniture, or charitable contributions happened to make it onto this month's bill. Our debt-to-liquid-assets ratio is something like 0.1%, so there's never any real risk of not having money to pay it off, but of course the credit bureaus don't have information about our assets, so they evaluate us against what other people our age have, which (being Millenials) is not very much.

Knowing how the system works, we can take steps to game it, like not putting any major purchases on credit card in the 3-6 months before getting a mortgage. But still, it's slightly ridiculous that something that's supposed to measure your creditworthiness can swing so much over short time periods.

It's not that weird. If your score is, say, 750, 50 points is about a 7 percent change. My finances very easily vary 7% from month to month!

50 points also isn't likely to make much difference, especially if your score is already good. Loans basically go off tables. Essentially if you're between A and B, you get this rate for this losan, C and D gets this rate, etc.

Once you're past 700, you're already getting good interest rates and banks will fall over themselves to loan you money if your income supports the loan size. When I was at 780, the loan officer couldn't give me a lower rate on a mortgage - I was already getting a fraction of a percent over prime.

You can eliminate credit score swings by never generating a credit card statement with a balance. You can do this by paying your card off prior to the statement date.

I spend between $1000-$2000 on credit cards a month and collectively I generate statements with less than $10 spread across 4 cards. My credit score is rock solid from month to month.

If you have six figures of liquid assets (what 0.1% debt vs liquid assets implies) this should be no problem at all.

We usually just set all our cards on autopay-in-full and forget about it. Is there a way to have them autopay before the statement date, or do you need to proactively check every month and make a payment right before the statement date?

Being proactive is the only way, unfortunately.

You also believe the stock market is nonsense, because valuations fluctuate?

It's worth remembering that an important aspect of your score is your value as a customer. Risk is one element. Carrying a balance gives you more value than if you never carry a balance.

It's the other way around - your credit score is higher if you have a lower balance on your cards.

That is not correct. A small balance raises your score more than a zero balance. But you want to keep it under a third.

This is an oft passed around urban legend, but it's not true. You can try for yourself by opening up a free CreditKarma account and seeing what happens if you zero out all your balances.

I'd take that with an entire shaker of salt; you are assuming CreditKarma knows the FICO formulas. Even the prediction engines used by the actual credit bureaus are notoriously unreliable at telling you what will happen given a particular change in inputs. Probably not accidental, they don't want to expose the algorithm.

> we can take steps to game it, like not putting any major purchases on credit card in the 3-6 months before getting a mortgage

I wonder how much your ability to game it indicates whether you are likely to be a good or bad debtor?

Foisting responsibility onto customers with more important things to worry about, rather than into organizations whose only purpose is to keep track of these things?

It is not my job to go out and manage my customer's finances and take their credit cards or write myself checks from their accounts, and I'd be quite rightly arrested for theft if I tried.

It is, or should be, your responsibility to notify them when payment it due.

Ah, but there is no indication they didn't. What happened is that the check was sent, but did not arrive. In other words, the customer and his agent (the postal service, probably) in the exchange did not fulfill their job of getting the money to the business. That is not something the business can control, and if you try to make it their responsibility, they will simply start refusing payment by mail, since it would be a constant and high level source of fraud.

But making it the customer’s responsibility is cool?

In any case, I read this as the notice that payment was required is what got lost, not the payment itself.

Ah, you may be right, but even that still makes the report valid.

Let me tell you what I hear as a business person if you tell me my letter to you was misdelivered: "My mail delivery is unreliable so it's your problem."

Well, no, you didn't tell me it was unreliable, so I couldn't adjust my risk perception or take special precautions. And if you ever want to do business with me again, you're going to come in and execute the transaction on site and not leave with the goods until the bank confirms it's cleared.

On top of which, most billing contracts specify that payment is due whether or not you were notified. You are supposed to keep track, too.

And yes, this is okay, because until we have a proper socialist system, it is, in fact, our responsibility to pay our debts. Don't like it? Don't incur debts.

As a business person, you’d put a customer through the wringer for one missed payment due to communication problems? I hope you have a monopoly in whatever market you serve, because otherwise that’s a great way to lose all of your customers.

Anyway, we’re back to making it the responsibility of the customer with a thousand more important things to think about, rather than of the business that’s dedicated to the stuff.

And if you think it should be that way, great. But it’s obvious to me that it is that way because of the power imbalance, not because it’s right.

If the customer is using it as an excuse, yes I would. They’re trying to weasel out of their responsibilities, responsibilities they almost certainly incurred voluntarily.

If you don’t make payments as agreed, your credit score goes down. If you do, it goes up, and will soon recover from the perfectly reasonable dip it took from your mistake. It’s pretty simple.

So if the customer's cheque doesn't get to you, that's the customer's fault, but if your letter doesn't get to the customer, that's also the customer's fault?

> The credit score is doing exactly what it's supposed to.

policing people and make the poor to pay more.

If you don't pay your bills, I think it's entirely reasonable for businesses to tell each other about that so you don't defraud them out of a lot of money.

>the entire banking system promotes systematic moral hazards and has lowered the purchasing power of stagnating wages for most people

"Why are people defaulting on their bills!?"

The whole notion of transfer of majority responsibility to an individual is one of the slimiest things in this society. We get monitored by an array of hidden surveillance measures and algorithimic judgement we have no way to properly counter or defend ourselves against. Meanwhile central credit gets to borrow money and get bailed out.

Give me a fudging break.

This isn't complicated. You incur a debt, you pay it. If you can't, people aren't going to want to do business with you, and will quite reasonably refuse.

You are trying to make everyone else responsible for my finances and I'd damn well thank you to stop, since they're mine.

I understand that clearly. But clearly there are classes of institutions and people who...

1. Make a mockery of prudent allocation of money and get bailed out when their games get messed up

2. Mess up the purchasing power of any money you possess far more than any individual defaults or even class of individual defaults

3. Impose grave and hidden responsibilities on individual borrowers and mass surveillance...

They are messing up your finances far more than the most reckless individual borrowers ever could. *

PS - How is finance going to deal with the multitrillion dollar pension bomb? With prudence or with a combination of a game of musical chairs and chickens until stuff gets serious? How is your financial discipline, as an individual, going to protect you from manmade tsnumais?

What are these "grave and hidden responsibilities" you're talking about? Paying your bills is hardly an obscure pastime.

As for your PS, I don't respond to conspiracy theories.

https://worldview.stratfor.com/life-expectancy-pension-retir...

Is the World Economic Forum a purveyor of conspiracy theories? Is our reduced purchasing power at the level of food and rent a conspiracy theory?

Edit: My apologies, here's the actual WEF press release https://www.weforum.org/press/2017/05/global-pension-timebom...

Pffft. The WEF is a celebrity getaway in the Swiss alps for the same people who caused the 2008 crisis and got away with it. If they’re making that prediction in good faith, it’d be a preemptive confession of guilt. If not, it’s just a distraction from their routine crimes.

> statistically, you

Is an oxymoron. Statistics exist when "you" is unknown.

"You" have statistics attached to you. Default rates, lifespan, cancer risk. Given people A and B in similar life and financial situations, A or B might defy the statistics that say they'll both pay or default, but they probably won't. The only way a bank or anyone can predict this is statistically.

I've had credit card companies call me and straight up say "If you pay this right now over the phone with me and pay the $15 telephone transaction fee, this won't hit your credit report"

Between the "Pay for Delete" scams and the gamifying of your credit score through services like CreditKarma I really questions how close these scores are hitting anymore in relation to relative credit risk.

> I've had credit card companies call me and straight up say "If you pay this right now over the phone with me and pay the $15 telephone transaction fee, this won't hit your credit report"

Does that not qualify as extortion?

The contract you agree to for credit cards or other forms of debt you choose to take on will specify things like late fees and that it's up to the company's discretion whether they report you a late payment to the agencies. If they didn't have this clause, you'd just be complaining that they report everyone indiscriminately rather than giving people a break when it seems appropriate.

> The contract you agree to for credit cards or other forms of debt you choose to take on will specify things like late fees and that it's up to the company's discretion whether they report you a late payment to the agencies. If they didn't have this clause, you'd just be complaining that they report everyone indiscriminately rather than giving people a break when it seems appropriate.

Thanks for making ridiculous assumptions and putting nonsense in my mouth. Nobody opined on anything in the first place. I was merely asking a legal question, because calling someone to demand an immediate payment to avoid direct harm to them sounds an awful lot like extortion, whether I like the approach or not.

If you want to know whether I'm impossible to satisfy or whether I think the company could in fact cut people slack while getting their $15 and avoiding potential extortion, you could just ask me. Yes, I think that's perfectly possible. Call the person up, tell them you'll waive the credit report if they make a payment by the end of the day, and optionally remind them that according to their contract, there is a $15 fee if they pay by phone. There. Now the person gets to spend more than 5 seconds thinking about it to make an informed decision. No need to put nonsense in my mouth.

But you're putting nonsense in the creditor's mouth. You don't know what the conversation was like, you only have the summary of one person who feels wronged.

All I did was I took the story at face value and asked a legal question without saying anything else about anyone. You felt I was somehow "putting nonsense in the creditor's mouth". OK, so why make a ridiculous response instead of just telling me I should wait to hear the other side's story? Or even better, maybe asking what I was thinking in the first place instead of putting your own nonsense in my mouth?

This doesn't solve the problem that already happened, but I would take this as a sign that it might not be a bad idea to just opt into e-statements and avoid snail mail. The world is electronic now. If nothing else, you'll at least save some carbon, paper, etc. from being consumed along the way.

I have autopay set up on my Google store Synchrony account, this last bill, for whatever reason, it did not attempt to withdrawal the money then sent me an email saying I was overdue and I now had to pay DOUBLE so I did manually, two days later I get an email saying my adjusted auto-pay had applied.

Guess what popped up on my report this week?

A late payment.

It does not let you pick the autopay date, and instead puts it on the date the bill is due, otherwise I'd have it auto pay at the first of the month weeks before it was due. However, it is not worth my time to call and probably have to deal with a call center for an hour before I get someone that will promise to remove it and then have to write three letters to send off to the CRBs to dispute it and wait another couple of months for it to possibly get removed.

Your anecdote is interesting but does not constitute data on the overall predictive powers of credit scores.

It does when millions of people have similar anecdotes, and when the system is literally set up for it.

Places can report whatever they want to. And you can dispute it, but the person who report it will just say "no, we're right" and it'll stay up.

This

I've been fighting with Equifax and TransUnion for 5 years to remove 30 year old reports on my credit history (which are clearly not me because im less than 25 years old)

This is amazing. How are you running into trouble with this for 5 years? What is their counterargument?

If you could build a better algorithm to do underwriting, you could make billions. Companies try all the time like zest finance.

Not being the worst is not the same as succeeding.

Following that logic the Burj Khalifa is a space elevator because no one has been able to build a better space elevator than that.

I have general objections to this sort of underwriting assessment being an ethical business for a private organization to begin with due to the immense amount of possibility for discrimination it opens up.

On the contrary, the modern credit scoring system has been a huge success when it comes to fighting discrimination. Previously banks left underwriting decisions to the whims of individual bankers. Unsurprisingly, many of these whims were based on things like the color of an applicants skin.

While there are certainly some arguments to be made about disparate impact under our current system basically everyone agrees that access to credit for minorities is much much better than it used to be.

Why? Why feed into it? This is the mentality that kills me.

Why double down and try to make something that is already being gamed? The very definition of insanity is doing The same thing over and over again and expecting a different result.

There is no right to operate a business that exposes everyone to risk they have no choice in whether to accept or not. Centralizing excessive amounts of data with obtuse and dubious control mechanisms/capacity for redress is a disaster waiting to happen.

Publicizing the risk, and privatizing the profits at it's finest.

It wouldn’t put anyone at risk if we held lenders to a legal standard of proof that YOU signed a contract asking for money. It might be inconvenient but not compared to being stuck with debts you never agreed to.

The problem with that, is the story doesn't stop at "You" when another person can get enough information to act as you in a legal sense.

To me, this entire industry reeks of people engaging in risky behavior, but trying to externalize the costs/risks of said risky behavior, and consequemces be damned. Furthermore, they want the agency that gets externalized to "make money", something which encourages the minimum amount of investment humanly possible in making sure they are actually solving the problem in a way that doesn't merely create new ones.

Furthermore, it seems to me that the financial sector is eating the bloody world; as the metrics they gather are being gleefully used as discriminators in far more than just loan granting.

The system is either so critical to the way the economy works, that we should be willing to "sink" money in the interests of making the system as effective as possible (I.e. no dark pattern B.S., easy to use controls, easy to manage all interactions with, and maintained to the highest degree of security), or it isn't, and a discussion needs to be had whether having such valuable pots of exploitable data is something we should even tolerate as an acceptable exercise.

To be quite honest, I've seen more harm than good come out of the system given the ubiquity of the Credit history "bootstrap" problem, and now the compromise of a huge portion of the American population's personal data.

Trying to couch this as merely a case of "oh, we just need more contracts" without addressing the central problem of your identity essentially being hijacked by a bunch of for profit involuntary surveillance companies operating under an incentive structure pushing minimum viable effort in protecting your data, ensuring it's correctness, and restricting access to only appropriate reasons.

Throw in the failure of the FTC to clearly levy a strong enough penalty leaves me feeling this industry is a social liability in it's current form.

Better for whom?

I’m sure the existing credit bureaus are great for the people who give them money. They sure suck for the rest of us, though. Fortunately for them, we get zero say.

Market is efficient. If underwriting wasn’t efficient, a company could utilize better algorithms and offer better rates and better determine risk to minimize default. This a multi billion dollar opportunity. People don’t care about who is underwriting their car or home loan, they care about rates.

Equifax and their brother businesses give us low rates in exchange for exposing us to fraud risks whose monetary value is probably literally incalculable.

The thing is, we don’t get to choose whether we’re exposed to those risks. It happens whether we like it or not.

So let’s say I start a company that offers better TCO: my rates are a bit higher but this is more than made up for by a much lower fraud risk. Do I win the market? No, because Equifax’s fraud risk hits my customers just as much, so I’m not competitive.

Credit bureaus give us low rates because they externalize the costs. Since the costs are externalities, competition can’t beat them. It’s the financial equivalent of making cheap electricity by poisoning the community with emissions.

There are underwriting systems that don't pull credit reports. The loan rates are around 20% and go down as you develop a history with them. People have choices.

They don’t really. Let’s say I choose to avoid the traditional credit system and go with one of these other things. Then some fraudster opens an account in my name with my info and suddenly I’m in Equifax anyway. Then Equifax gets breached and even more fraudsters use that info. I’m stuck trying to prove my innocence to creditors despite having supposedly chosen not to participate in that system.

"Your kneecaps are your collateral..."

Why do you say that? I’ve seen some of their criteria and while I can’t remember a lot of specifics, it was a bunch of ad-hoc rules, not updated since before computers were a thing, only stuff you can calculate with pen and paper and all of it totally arbitrary.

Your statement is interesting but does not constitute nor provide any substantive information that is backed by evidence. Backup your 'anecdotal' claims.

Banks and credit card issuers all pay hefty sums of money to the reporting agencies to use in underwriting their loans. They wouldn't do this if they weren't helpful in modeling risk.

I was speaking to a loan officer recently, and the purpose of the credit score seems only vaguely related to how well it actually predicts risk.

The gist was that if all their rates are based on a standard formula with no individual discretion, they are safe from regulation around (1) discrimination, (2) risky lending such as led to '08. That sounded more important to them than actually accurate risk modeling, especially since their competitors are all using the same formulas.

(edit: In particular, even if you offered them access to much more effective/predictive data points than the credit score, they would not use them because of (1) and (2) above.)

The majority of people on this board work in tech - just because money is exchanged for a service does not mean that the service purchased is worth that money... so can many attest about terrible highly paid consultants.

Large organizations are astoundingly good at finding inefficient ways to spend money.

Large financial organizations have teams of analysts whose entire job is to relate consumer information, such as credit scores, to statistical outcomes. If the credit scores weren't giving them useful information, that would show up rather quickly.

I’m suddenly reminded of the ratings used to fuel the subprime mortgage crisis.

That was something subtly different. The banks more or less knew they were taking poor risks, the problem is it wasn't their risk to take. They quietly packaged these bad mortgages along with others and resold the mess to less sophisticated investors (like pension plans). Then the bubble burst, the well dried up, and banks started collapsing. It wasn't a mismanagement of risk, it was criminal fraud writ large.

I would argue that credit bureaus leaking all your information to criminals and then charging you money to protect you from those criminals is also criminal fraud writ large.

That doesn't seem relevant to the question of whether credit scores are statistically useful.

We've established that sufficiently advanced fraud can overcome the ability of organizations to vet the info that’s given to them. If the credit bureaus are actually organizations dedicated to fraud, then the fact that many other organization fall for it doesn’t tell us too much about their accuracy.

If Osama bin Laden's ghost appears to you and says "There will not be a terrorist attack in New York tomorrow.", and then there is, and this repeats a few times, the fact that his statement is, facially, a lie, and the fact that he can't even exist doesn't matter. You have an input that correlates with a certain output. The intent of the person who provided you the input, or even how nonsensical you think the input is, doesn't change its statistical usefulness.

So how did all these sophisticated financial organizations get taken by subprime mortgage bundles in 2008?

They didn't. Pension plans and individual investors are not sophisticated financial organizations. They rely on those that are to do their job in a non-fraudulent manner. They didn't.

Bear Stearns? Lehman Brothers? AIG? The hundreds of banks that went under?

Committed or were complicit in the fraud and got in too deep to recover from the inevitable crash. This has all been widely and loudly reported on for a decade. There's a reason people are pissed that the executives who ran these institutions into the ground and fucked up the economy were barely prosecuted.

I'm sorry, but is the reason that Equifax isn't in the same class as these guys because Equifax was totally upstanding, never cut corners or fudged numbers and were absolutely upfront when this security breach happened? Because if that's so then I'm curious about the breach they detected three month prior that they tried to cover up.

Equifax is just as shady as those lenders - more so IMO because they have absolutely no obligation or business relationship directly with the individual's whose private data they compromised.

So in addition to the info being valid or the organization evaluating it being too stupid to evaluate it correctly, we also have the possibility that these organizations are involved or complicit in the fraud.

Why am I supposed to take it as a given that if these organizations use the info it must be useful, then?

If it's not, they're wasting money for no reason, and someone will come along and do the same thing without giving the credit bureaus any money. They're not actually legally obliged to use credit bureau data. They could just throw money at people in the street and hope they get confused and throw back more.

A bank isn't an individual who is lending his own money. It is an organization that lends other people's money and needs a veneer of justifications for the choices they make, and credit reporting agencies are a perfect choice. You mentioned earlier that banks have statistical teams making these analyses, but most simply don't because that provides counter-information that undermines the whole value of credit reporting agencies (which is that they outsource the politically fraught choices and makes it easy to say "well we just followed their ratings, just like everyone else").

Regarding the subprime crisis, the biggest victims were the largest banks -- the most sophisticated being put completely out of business -- so not sure what the bit about pension plans comes from (many of those made a lot of money on it).

"They wouldn't do this if they weren't helpful in modeling risk."

This isn't necessarily true.

A counterargument is that credit scores offer banks an easy way to outsource what often ends up being a very contentious, politically-fraught process. Yet many analyses have found credit scores barely better than random dice -- the guy with the perfect credit score has a perfect credit score, until he doesn't and there is a wake of delinquency in his wake.

It's also helpful to assess real world motivations. Extraordinarily few of the employees, including at the executive level, at a bank are legitimately concerned about the long term risk to the bank. Success is measured at the quarterly interval, and if you can justify your actions on a quantifiable measure -- even if it's a measure that has little predictive value -- then that's just perfect.

The biggest indicator that someone is a credit risk is that they are maintaining or growing higher interest borrowing products, such as carrying a balance on a credit card. This is an absolute flashing light indication that someone is over-extended, yet the credit monitoring agencies would not bite the hand that feeds them by making too big of a deal about this. Indeed, gross over-borrowing is barely a blip on a credit report, because the people who lend the money ensure that it isn't. The world is absolutely awash in cheap cash and banks are desperate to lend it.

In the wake of the subprime crisis everyone said "oh yes, of course there's a problem there it was the credit agency that was just marking these all wrong", but exactly the same thing is happening on personal credit reports. Of course it is, because the credit reporting agency is there to legitimize whatever the bank wants to do.

This is a fallacious argument. Just because banks use credit scores doesn't mean they are accurate (see argumentum ad populum).

What we'd need to demonstrate your claim is for example some data linking credit scores at time of loan to default rate, adjusted for income and loan terms.

Maybe just go upthread to patrick's comment on the topic. He's a much better writer than me.

https://news.ycombinator.com/item?id=20580478

Reply on news.ycombinator.com