IMF working paper suggests using browsing history in credit score calculations
extremetech.com
extremetech.com
> Even a well-paid expatriate moving to the United States can be caught in the conundrum of not getting a credit card for lack of credit record, and not having a credit record for lack of credit cards.
> Fintech resolves the dilemma by tapping various nonfinancial data: the type of browser and hardware used to access the internet, the history of online searches and purchases.
I enjoy the framing of this; give us the keys to the kingdom and we'll give you even more credit (to get yourself further into debt!)
Of course it won't stop there, better pay down that mortgage before you search or buy the "wrong" thing if you want to keep your house (they won't take it outright, just raise the interest rate on your loan because you now represent a higher risk based on "new information")
[1] https://blogs.imf.org/2020/12/17/what-is-really-new-in-finte...
At some point in the past or maybe in another universe, a well known accommodation booking site might or might not have been offering better rates to people using Linux than OSX. Not confirmed, obviously.
BTW, I also believe that flight prices magically increase on iOS...
[1]: https://www.wsj.com/articles/SB10001424052702304458604577488...
Add to that how likely someone with your personality-type is to just decide to take a few months/years off between jobs living off of savings, and you've got a heuristic nearly as predictive of loan-repayment as all the rest of the information lenders use today put together.
Here's some more detail on this: https://www.netinterest.co/p/financing-the-american-home
Strange, because I have a fixed term for the entire length of my mortgage and I'm not living in those countries. Admittedly it's for a length of time shorter than 30 years, but even at that timeframe it would be possible to get one.
There is however one additional modifier in the statement of the article: "fully prepayable", that may not exist anywhere outside of Denmark and the USA (but still, your source doesn't really explain where it got it's list of countries that have it, so I have no way to know it's correctness).
I can't see how any new risk is added in a fixed rate. In fact I can't see how a non fixed rate would really be good for anybody, and would keep the risk fluctuating for all parties involved.
If you have an adjustable-rate mortgage, you bear the interest rate risk.
This is incorrect. It is very common for investors (yes, including commercial banks) to frame risk in the way I described. This link to investopedia will expound upon my admittedly short description of interest rate risk [1]. Another obvious risk in this context is prepayment risk [2]. Note that both of these risk models are not about whether or not the bank is losing the money used to pay for the home, but are about factors that could affect the return on investment. You can find even more examples of risk models exactly like this on investopedia.
[1] https://www.investopedia.com/terms/i/interestraterisk.asp [2] https://www.investopedia.com/terms/p/prepaymentrisk.asp
After all, even though their calculations were counting on the profits of that money, having it back now means they can loan it out again, and possibly will receive a lager return if interest rates have risen.
For instance, my mortgage is at 2.49%. Right now interest rates are at 3.04%, so if I paid my mortgage off today the bank should be able to find a borrower who would pay over 20% more for the same money, right?
I get that this could go in the opposite direction, though, so I know it's not a perfect answer.
People with the means can easily change their computer - You go from a budget pc running an old version of windows to a brand new Mac and your credit score goes up.
On the flip side your isp may indicate you live in a “bad neighborhood” and get penalized for it.
We have all got to hope this never gets implemented.
(Because the actually crazy people have killed satire, let me be clear: this is satire.)
https://en.wikipedia.org/wiki/Library_Awareness_Program
I ran into this myself, in my High-school Library and I'm sure may other children may have as well.
Not per se a social credit system, but, how different is it? One org/database, interpreted by people of org and then a similar enough inquisition/questioning of "why."
Maybe these clueless old men should stick to what they know - mountains of physical paperwork and pointless policies and procedures, and stay away from anything that uses electricity or internet.
Bitcon/DeFi will render them obsolete in a few years' time anyway.
>Bitcon/DeFi will render them obsolete in a few years' time anyway.
Credit scores exist because banks want to know how likely they're going to get their money back. I'm not sure how that'd be replaced with "bitcoin/difi". Most of the "loans" I see are difi are either flash loans or overcolateralized loans, which have limited application to typical consumer loans (eg. car loan or mortgage).
Can an autonomous bot on some smart contract have a credit score? Why is credit a privilege only humans can enjoy?
We should be able to spool up a persistent pseudonym that can have a credit score based solely on its selectively disclosed transaction history since inception.
If we screw up the credit score of that pseudonym, we should be able to create another and another. The market decides what score to give based whatever factor it feels relavant, which should result in abuse prevention over the long term.
companies and even countries can have credit scores, so it's not something that "only humans can enjoy".
>We should be able to spool up a persistent pseudonym that can have a credit score based solely on its selectively disclosed transaction history since inception.
Part of what gives credit scores value is that you only have one identity, so you can't create burner accounts and default all the loans on them.
>The market decides what score to give based whatever factor it feels relavant, which should result in abuse prevention over the long term.
The market will react by treating pseudonyms the same way websites treat tor exit nodes: with extreme suspicion. Using a pseudonym with little history and/or "real" identity tied to it would essentially put you in the "default imminent" category of credit ratings.
If new pseudonyms are treated with extreme suspicion then so be it, but there should be a way for the pseudonym to progress from there solely through transactional behavior.
Even if the amount of credit the market becomes comfortable with through this scheme is $100, that could make a big difference in developing countries.
If someone is operating under a pseudonym and can walk away from it, you're playing an iterated prisoner's dilemma where one party knows on what turn they will defect and the other doesn't.
With real identities, the iteration only ends in the death of one party, the timing of which is seldom known to anyone (although will statistically be taken into account).
Credit scores might be a good idea in some respect but it shouldn't be possible for anyone but lenders and the individuals those scores represent, and they should only ever cover whether the person or organization is a good or bad risk for loaning them money or credit. Employers and landlords should never be able to know what your credit score is. At best they should know about your employment history, but if they want your credit score then they might as well ask what shows you watch on Netflix or if you look at porn, you know, for moral reasons.
Hopefully the system will be kept in check by decentralized tech, but who knows how long it will be before social credit will bar you from using the same internet as everyone else.
Therefore, regardless of how many people they kill, and, yes, they have killed people due to statistics and economics illiteracy combined with blind career ambition, they will continue to be around for a long time.
You're right about this. People's access to necessities like housing shouldn't depend on whether they are making money for the finance industry. And developing countries' access to the resources needed to improve the quality of life of their citizens shouldn't depend on making the development process profitable to US/European institutions.
> Bitcon/DeFi will render them obsolete in a few years' time anyway.
Unlikely. Bitcoin and other proof of work systems are likely to be banned in many places because of their environmental impact. And systems that survive that are likely to be banned as vehicles for financial fraud. The future of blockchain currencies is probably in making transactions more traceable to shut down tax avoidance schemes.
> You're right about this. People's access to necessities like housing shouldn't depend on whether they are making money for the finance industry. And developing countries' access to the resources needed to improve the quality of life of their citizens shouldn't depend on making the development process profitable to US/European institutions.
While there's definitely issues with how credit scores are implemented, it's absolutely essential for something like them to exist. People who habitually default on their debts shouldn't be able to keep borrowing money as easily as the rest of us.
(I am assuming "Bitcon" is a typo for "Bitcoin" rather than some new cryptocurrency or blockchain thing I have not heard of).
This reminds me of a question someone once asked Miss Manners:
> Dear Miss Manners:
> Don't you think that nowadays, in modern life, the old-fashioned custom of the condolence call is out of date?
> Gentle Reader:
> Why it that? Is it because people don't die anymore, or is it because the bereaved no longer need the comfort of their friends? Miss Manners is always interested in hearing about how life has been improved by modern thinking or technology.
People lending money via Bitcoin/DeFi or any other mechanism now or in the future are still going to want a way to estimate the risks that the borrower will be able to pay back the loan on schedule.
I'm not really sure if more than last one is needed, maybe some overall registry of how much money single person has lend and what part is unguaranteed of it...
Cold fusion is coming in the next 30 years starting from datetime.now()
In financial analysis of a company, the number one item you tend to look at is "free cash flow" - the "cash available for the company to repay creditors or pay dividends and interest to investors." [1]
Why then do we not take income into account in credit worthiness? Why do I not periodically upload my W2 or tax return to the credit bureaus in order to show that yes, I'm credit-worthy because my income makes it possible to service my debt?
The entire idea of a credit score based on having had credit and paying past loans off is asinine. To start throwing nonsensical parameters like browsing history into the equation is second derivative asinine.
Here, "large" runs at least up to $50k-$100k per lender without any meaningful income verification.
By reducing the information taken, do we add sufficient additional users to overcome the X% loss expected.
Income verification is what happens when you buy a mortgage. It has nothing in common with the credit card process.
It seems like we're heading for a society lying and fraud are engaged in and almost tolerated a lot more, and certainly where resources for enforcement are stripped.
https://krebsonsecurity.com/2017/11/how-to-opt-out-of-equifa...
Even though the article is old, I saw my last few years' pay history in the links provided there.
Opting out feels like a devil's bargain: credit will presumably cost you more or be harder to get as a result. But there's no fine-grained way to choose who gets the info.
And trivially exploitable...
> Credit scores are so fundamentally broken anyway.
The bigger problem is that the organizations designing credit scores (and alternative metrics) are incompetent.
If Equifax went out of business today, you'd have 5 new equally shitty credit companies trying to get their market share.
Only up to a point, though. Restaurants can be shut down when they have too many health violations. Private medical practices can be sued into the ground. There are lots of types of incompetence in finance that can get you shut down / sued out of existence.
> The real problem is they have buy in from other industries with major decision making abilities and the government has provided us no protections from them.
Which is why I'm being vocal, where possible, about how incompetent these organizations have become.
Is this supposed to be better? Right now I can apply and get credit cards without mailing my paystub. That actually seems better than companies having regular updates of my income, which they can use for evil marketing purposes.
Your pay history is probably sent out to all your creditors, unless you've opted out.
I’m not using an EBITDA/MV equation on my personal valuation, nor am I using FCF for any credit calculation.
Credit calculations are based on the most reliable data points that correlate with outcomes I care about.
Frankly, new tech is helping credit calculations - people are using phone call data to find what correlates with credit worthiness. I believe this is already in production around the world.
Whether the term “credit scores” is used or not, there will always be some mechanistic way to value risk. It will also use whatever data we have lying around to do so.
If we have no data, then it goes back to the good old days of the boys club, which is even more restricting Thant today.
Income is definitely considered in large loans. You’re not going to get a mortgage or a car loan on your credit score alone.
Credit firms also collect a lot of payroll data, but it’s kept separate from what you know as your credit score because they inherently can’t have payroll data for everyone that also has credit history data. The scores must be kept separate as a result.
Here’s just one example of credit worthiness companies expanding their payroll data collection: https://krebsonsecurity.com/2021/07/intuit-to-share-payroll-...
They are trivial to default on. They are very hard to discharge.
At the end of the day, a credit score indicates how profitable you will be as a customer for a lender. Showing profitability for a lender in a past indicates that you will likely be profitable in the future as well. Thus, in effect, you do have to buy your score, although the method of purchase may not always be straightforward.
And how is Bitcoin going to tell whether you default on loans or not?
We...do. Any significant credit application will look at your employment history, income, and credit score, not just the latter.
> Why do I not periodically upload my W2 or tax return to the credit bureaus in order to show that yes, I'm credit-worthy because my income makes it possible to service my debt?
Credit reporting companies do gather and provide pay data. As fir other data, they don't gather it directly from the people about whom they keep files.
> The entire idea of a credit score based on having had credit and paying past loans off is asinine
No, its not. While past performance isn't a guarantee of future results, it is useful as a guide. And it isn't the only factor considered in most loans.
Consider the fact that conflating credit score to credit worthiness (respectfully, as your reply does) as the biggest indicator of the brokenness.
Current income, W2, working or not, all of this is a second factor, separately determined by the lender.
Web browsing habits are odd in either category.
No, it doesn't. My reply was to a question that did that, and explicitly corrected it, identifying other factors considered in creditworthiness besides credit score.
Then why did it explicitly say that income was not considered in creditworthiness? The conflation of that with credit score was from this line in your comment: “Why then do we not take income into account in credit worthiness?” My response was directed at the error in the explicit premise of this question.
> Three credit bureaus purport to have a comprehensive opinion of someone's credit worthiness
No, they don't, which is why they also gather and provide, as separate products, payroll data and other data, and even credit decision platforms that incorporate income and other data. Credit score is a single product with a more limited function (both as advertised and as used by customers) than you are representing.
Sounds unethical.
Why on earth does the IMF, an organization staffed almost exclusively by academic economists and political appointees, think they have any expertise whatsoever in consumer credit decisions? These are people whose understanding of business stops at "getting funding," from presumed and paternalistic granting agencies. The entire premise of their organization was to use lending as a globalized policy leverage tool, to issue bad loans to weak governments for political reasons, and then demand policy concessions in exchange for renegotiation terms. These have always been really bad people.
It does suggest we need better p2p lending instruments, and really, anything that these organizations think they can do, we should probably be developing alternatives that limit their influence.
For those not familiar, here is the beginning of the relevant passage[1]:
> 'It is impossible to see reality except by looking through the eyes of the Party. That is the fact that you have got to relearn, Winston. It needs an act of self-destruction, an effort of the will. You must humble yourself before you can become sane.'
> He paused for a few moments, as though to allow what he had been saying to sink in.
> 'Do you remember,' he went on, 'writing in your diary, "Freedom is the freedom to say that two plus two make four"?'
> 'Yes,' said Winston.
> O'Brien held up his left hand, its back towards Winston, with the thumb hidden and the four fingers extended.
> 'How many fingers am I holding up, Winston?'
> 'Four.'
> 'And if the party says that it is not four but five—then how many?'
> ...
[1]: https://2x2.rocks/
For a couple hundred years, Americans insisted 2+2=5. It was even enshrined in law for the first hundred or so years. We eventually agreed on a compromise, 2+2=4.5, but some of our lives depend on the answer to this math problem, so we're not going to stop fighting for 2+2=4. Sure enough, if you work as an engineer, you're going to be fired if your answer isn't within tolerance.
No. Please do not try to label me.
https://news.ycombinator.com/item?id=28275671
Orwell indeed.
I have made no statements regarding vaccination.
However, I do consider delaying until after the approval announcement the release of the update to what is known about how many people who are fully vaccinated against Covid19 are dying from Covid19 or being hospitalized due to Covid19 to be suspicious.
That is not a conspiracy theory. It happened. Information that was due to be released on Friday was delayed and released after the FDA approval on Monday. The updated information showed that known rates of death from Covid19 among those fully vaccinated against Covid19 continued to increase.
Plus, the CDC does not put out a proper time series table[1]. Every week, they overwrite the information from the previous week so one is forced to rely on external snapshots[2] to preserve what should be public record. Plus, many states don't even release or track information about the vaccination status of people who are hospitalized or who die from Covid19. For months now, CDC have been saying:
> Ultimately, CDC will use the National Notifiable Diseases Surveillance System (NNDSS) to identify vaccine breakthrough cases.
What is the holdup?
I summarized why you want to care about that here: https://news.ycombinator.com/item?id=28287440
None of this is a conspiracy theory: https://news.ycombinator.com/item?id=28280087
[1]: https://www.cdc.gov/vaccines/covid-19/health-departments/bre...
[2]: https://archive.ph/https://www.cdc.gov/vaccines/covid-19/hea...
Wrong, that was Kafkaesque.
But it’s a glass half full for those aware and able to see it for what it is.
The end game is me sitting under a tree somewhere with a picnic in the middle of nowhere with someone I have a genuine connection with and no technology at hand.
At least I have a voice in government. I don't get a voice when Experian (or Google, or any private company) decides to blacklist me arbitrarily.
Sounds like it's working to me. Uncooperative credit slaves pay less money in interest.
They help nobody but the bankers.
Personally, I'd like to have much more control over who can view my credit data and for what purposes.
If you can afford those things more safely, you should be considered less desirable? Instead, it's the people who are likely to be in debt until death that are better?
But I guess, yeah, fuck society, get rich while you can.
If you can afford to leave, then yes, you are less desirable than someone who needs to stick around. These services are priced based on the assumption of long-term engagement. If you cut that assumption short, the expected revenue is also cut short.
Finding new customers is expensive. When you leave, the provider needs to find a new customer to replace you. They are much better off keeping existing customers. The credit score gives an indication of how likely one is to remain a customer. Those who are most likely to stick around are the best kind of customer.
When you paid off your loan, you also stopped paying the associated interest. In other words, you retracted your patronage and stopped being a customer. This is not desirable for a business. The credit score was reduced to reflect that. 60 points is very little movement. It is not like they're writing you off as a compete undesirable with that, just statistically slightly less desirable than someone who continued to borrow.
Remember who pays for credit score information: Lenders, not you. They are the customer. It is there to benefit them, not you.
Realistically, paying off a loan rarely hurts credit scores. And realistically, someone with years of paying off credit cards on time is going to have a high credit score (at least high enough for things you describe), outstanding debts or not.
This is part of a larger trend of non-state actors with a huge variety of sometimes very pro-social missions (IMF, anti-child exploitation non-profits) advancing their respective missions through leveraging or targeting other non-state actors' platform control (tech companies, adtech infra) of platforms we totally rely on for day-to-day life (browsers, smart phones, OSs).
The connective tissue is the selective analysis, or less generously: magical thinking, of tech's beneficial features for a specific mission while ignoring (or not understanding) what their views for tech-use cause in aggregate.
The only counteraction that seems capable of effective reaction to this trend is forcing that "aggregate view." Developers on GitHub re: showing hash collisions for Apple's CSAM tech have recently lead the way in this space. Tools which passively search the web to generate browser histories which have beneficial credit score impact are another option.
It's time to return to the mindset cypherpunks created, but expand it from their hyper-focus on privacy: write and socialize tools which politely but directly force tech-lite researchers and non-state actors with powerful missions and influence to acknowledge their blindspots (hopefully), or screw up their products (more likely).
a) how do they acquire users' web history? Do they purchase it from ISP?
b) how do they know what ISP to ask?
c) how does the ISP know my web history? Last time I checked, almost all websites I visit use https, so the only thing they can see is domain, not /pathname/
d) how does it feel to poke your nose where it doesn't belong?
"We may also share personal information that does not identify you with third parties for their own marketing and advertising purposes, which you can opt out of. This mainly occurs when you interact with our websites and mobile applications that contain third-party cookies or other advertising trackers. To learn more about this, please read our Cookies Notice."
Coupled with: https://www.vice.com/en/article/gygx7y/your-anonymous-browsi...
b) Afaik there's data brokers that know this and collect data from many different kinds of sources (isps being only one of them) those then sell to people like private investigators and private intelligence companies. Sadly I don't have a source ready, I learned this on a podcast and completely forgot which one. But the vice article above goes into that direction too.
c) That's correct, but that's also already valuable on it's own. I assume the /pathnames/ can come from third party tracking / ad companies, but dns data alone is also valuable. It's all in the patters of the (meta)data
d) Very likely exciting to many. Voyeurism and the "forbidden apple" aren't exactly new concepts
Go to HackerNews, post a dissenting topic, enjoy no travel or credit!
Work in a high compliance job where you have to submit financials yearly, no more!
We should be a bit vigilant about such concepts creeping into freer societies. They will usually do so either under guise of combating crime, or increasing efficiency. But there is no freedom without some slack.
It's a way more stable system. Allow a very narrow band of opinions in practice while claiming that everyone's free to think whatever they want in theory.
The last very public case I'm aware of was Damore at google, which they must've settled outside of court.
Yet we can't get a microphone, airtime or column inches for anyone in favor of the withdrawal.
I'm free to post pseudonymously though!
First Korea, now this...
From your link: "By 2020,[needs update] it was intended to standardize the assessment of citizens' and businesses' economic and social reputation, or 'Social Credit'."
It's 2021 now and the "Social" part of "Social Credit" is still science-fiction, let alone somebody's web history affecting their ability to obtain credit.
The part that is in place already and works somewhat is a national registry of unpaid debts, so you can find out e.g. that the state-owned investment fund you're dealing with owned a large stake of a mining company that went bankrupt while owing a large sum of money to a state-owned bank, leading to a court case in which the fund was ordered to service part of the debt. Stuff like that.
I don't know how it is implemented, but I would be surprised if it wasn't.
Presumably being on such a blacklist will actually influence individual credit scores once they get rolled out nationwide, but unless I somehow missed the announcement, that hasn't happened yet.
https://en.wikipedia.org/wiki/Xu_Xiaodong
So it's a system that's created by the government to make people aware of unpaid debts, but it's the same government that decides to impose debt on said people whenever it wants.
Coolbeans.
Also, I'm under the impression that people imagine something more direct when they talk about web history affecting credit scores, like visiting a site automatically deducting points from a score, or something.
The underlying issue here is that democracy is not enough. To state it differently, democracy is a means to an end, with the 'end' being self-representation. Democracy is necessary to achieve self-representation, but it's not sufficient.
An example: remember when The Don wanted to buy Greenland? Well, let's say the US bought Greenland. Their people, then, would be granted US Citizenship, including the right to vote. Throughout the transition, they would reside in democracies. One day they would stop voting in Danish elections and would instead vote in US elections. They would always be able to participate in the democratic process. They would also be extremely pissed off. Why, though? They had their say, didn't they? They voted in the elections, didn't they? But even though the process was democratic, the right of the people of Greenland to self-govern was not observed. They had democracy, but not self-representation.
Folks in the US feel the same way. In the Pacific Northwest, there is a broad consensus that the state should run healthcare, and yet they don't have state-run healthcare. Why? Because they need the consent of folks who live thousands of miles away in Oklahoma, Texas, and Florida! They have democracy, but they don't have self-representation.
When the tide turns and there is majority support, on the national level, for state-run healthcare, what happens in Oklahoma? We're going to force a national healthcare system on them when the vast majority of the citizenry abhor the idea of state healthcare. And why? Because, thousands of miles away in California, Oregon, and Washington, folks support the idea. That's democracy, sure, but it's not self-representation.
The rash of democracies that have taken hold in the world over the last few centuries has been wonderful, but we'd be foolish to assume that we've perfected human governance. There are better, more representative forms of government out there that we need to seek out and implement.
To make the next great leap in human governance, we need to acknowledge the limits of the systems that we currently have. We need to admit that over-large democracies are not conducive to actual self-representation. And we need to exorcise the ghosts of our monarchical forebears. Our governments don't own us, do they? Then why do we allow them to stifle and suppress democratic secession movements, like the recent one in Catalonia?
To ensure that we are truly able to self-govern, we need to pick our governments, not just our representatives.
> To ensure that we are truly able to self-govern, we need to pick our governments, not just our representatives.
i guess im having a hard time imagining what that would look like in concrete terms...
Consider the situation in Oregon, where five rural counties have voted to leave the state of Oregon and to join the state of Idaho.[0] According to WaPo:
> "Lawmakers in Oregon and Idaho would have to enact bills to redefine the states’ boundaries and redistrict their legislatures. They’d also have to muster the votes to override a potential veto from their respective governors. And then Congress would have to sign off on the move."
In other words, these counties can only change states if they're able to get 320 million people to sign off on the change. That's crazy.
Now, how to achieve this in practice? Ideally we would revert to the original Federal system (with a small, weak federal government and strong state governments) and allow for counties to move between states, but that seems unlikely.
Alternatively, we could allow for counties to move between states, and allow States to self-organize into new federal/confederal nations, or to move between nations via some established process (i.e. referenda). These nations should coordinate at a supra-national level (like the EU) to facilitate commerce and free movement, and to allow for the settlement if nation-level disputes.
Huh? They need:
(1) a simple majority of each legislative house in Idaho, which is 18+36=54 people
(2) a simple majority of each legislative house in Colorado, which is 18+33=51 people, and
(3) a simple majority of each House of Congress, which is 51+218=269 people
for a total of 269+54+51=374 people. Your 320 million estimate is off by about 6 orders of magnitude.
(Even if you make the rather unrrasonable assumption that they need to convince enough unique voters to guarantee the election of a sufficient majority in each involved legislature—i.e., you assume that every eligible voter treats every Idaho, Colorado, and federal legislative election as a single issue referendum on the issue—you’ll find that the number is a very small fraction of the ~240 million eligible voters in the US.)
wouldn't such flexibility/decentralization hurt national competitiveness vis-a-vis say, u.s vs china, or russia etc?
It would also allow the states to serve their original role as “the marketplace of ideas”. One could imagine Oregon implementing state-run healthcare or a UBI, or Kansas implementing the “Fair Tax”, whereas now the states look to the federal government to implement these sorts of large policy changes.
Fact of the matter is if people want access to credit the creditor has a legitimate interest in making an informed decision about whether the individual is reliable to set a correct rate. There is no way around this.
The two other alternatives are, we don't give someone access to credit, or we hand out credit without any sort of metrics. The first one doesn't help anyone, the second one doesn't make any financial sense.
Consumer data (the article also mentions online shopping history, which makes more sense than 'browser history'), in contrast to institutional banking data is ubiquitous even in poor countries because people now have wide access to phones and services. It would enable a lot of people to get access to credit.
Pretty much everyone here is familiar with Goodhart's Law; I'm surprised the IMF isn't. Use browsing history for credit scores, watch for the immediate proliferation of apps that generate/clean your browser history.
Soon, privacy is for nerds only anyway.
Anyways, Whatever the law says, us nerds should be in charge anyway, right?! Or at least we should have some inherent privilege. These guys at the IMF have great ideas.
The working paper doesn't really contain any original research (at least on this topic); they cite a single paper by Tobias Berg, Valentin Burg, Ana Gombović, Manju Puri [3]. The IMF working paper/blog post concludes, on the basis of this paper, that a system "powered by artificial intelligence and machine learning" (ugh) can improve credit assessments using features such as "the type of browser and hardware used to access the internet".
I feel kind of depressed that this needs to be said, but...
1. There is no causal link between user-agent and loan repayment.
2. Attaching high-value signals such as credit scores to such easily spoofed data as user-agent and browsing history has some amazingly obvious problems that are never mentioned.
The level of technological and statistical incompetence on display here is rather astounding, and it's difficult for me to take any of this seriously even before the ethical questions begin.
This is part of an ongoing trend I've noticed: the quality of human capital in financial services seems to be in rapid decline. The cream of the crop was scraped off by huge tech companies on the industry side. On the academic side, all the top students are choosing CS PhD programs instead of econ/finance and top-tier faculty are leaving core econ/business/finance/etc. departments for specialized institutes/departments. And even within finance, already degraded, the under-60 talent that does remain isn't doing stuff like loan origination or credit risk assessment for obvious reasons.
I think it's time to start strictly regulating things like credit scores. If not for reasons of privacy/equality, at least merely because the entire industry seems to be incompetent and getting worse.
--
[1] https://blogs.imf.org/2020/12/17/what-is-really-new-in-finte...
[2] https://www.imf.org/en/Publications/WP/Issues/2020/08/07/Fin...
Really? Assume people who still use IE6 on a late 1990s computer are too poor? My millionaire BIL would flunk that test.
>IMF recommends using ISP/OS/browsing history metadata for credit reports
>salary data is being sent to credit companies
I wish i could just only pick ISPs and jobs and houses and lenders and such which don't send information to tracking companies.
Sadly working for someone who pays you is illegal unless you report legal names, tax forms, and such to the government... and even making an online purchase without revealing your identity is difficult and questionably legal.
How viable it is to rent an apartment or buy a house, without letting others vacuum your personal information?
In this sense digital finance will be created by decree rather than as the unavoidable culmimation of some particular "tech innovation".
Underneath the huge risks there is always the potential that easier collection and dissemination of information will support a "better" financial system, somehow combining the best rather than the worst of "fin" and "tech".
https://www.usatoday.com/story/money/personalfinance/2021/07...
(I'm from Europe and I've never borrowed money. I even only use a prepaid credit card, always wondered what benefits a real credit card would even bring.)
On the bright side, if I do I'll probably not realize it ...
Of course the IMF has it's own goals which are fairly easy to guess. It's no surprise that a 'credit score' is more a way to separate friends from enemies rather than a measure of creditworthiness.
It's interesting to think of machine learning let loose on the problem of a persons likelihood to pay back their bills. The uproar about the 'unfairness' of the decisions would be tremendous. No doubt that's already been done to a large extent, but the politics of the era won't allow it so there's bound to be a bit of tension in the system.
They have already come for individual bank accounts, in the UK for example an activist named 'Tommy Robinson' [1] is not allowed to open a bank account due to his politics [*].
"What business of mine is it
So long they don’t take the yam
From my savouring mouth?"
Then they came for Trump, the former president of the United States [3]. Irrespective of your politics, do you believe that he should not be allowed to have money or do business?
"What business of mine is it
So long they don’t take the yam
From my savouring mouth?"
Now they come after sex workers on websites like OnlyFans [4], threatening the livelihoods of people doing perfectly legal business. Sure, OnlyFans did need to clean up their act - but we're talking here about the fringe, not the majority. I really also suspect PornHub were also targetted by payment processors not so long ago, but they kept more quiet about it.
"What business of mine is it
So long they don’t take the yam
From my savouring mouth?"
"And then one evening
As I sat down to eat my yam
A knock on the door froze my hungry hand.
The jeep was waiting on my bewildered lawn
Waiting, waiting in its usual silence."
- (Interlaced poem by Niyi Osundare [5])
[*] I would suggest to hear his speech at the Oxford Union address [2], rather than a heavily biased Wikipedia entry. The first two references in the Wikipedia article are simply links to news articles calling him far-right - not evidence in the slightest of him having far-right opinions. If you even have the slightest doubt in the news media, please watch the video before casting judgement.
[1] https://en.wikipedia.org/wiki/Tommy_Robinson_%28activist%29
[2] https://www.youtube.com/watch?v=_YQ94jFg_4A
[3] https://www.bloomberg.com/news/articles/2021-01-12/more-of-t...
[4] https://edition.cnn.com/2021/08/21/tech/onlyfans-creators-re...
Maybe it's all more just a matter of opportunity than it is belief system.
What should scare them, and everyone else, is the distinct possibility that things won't simply swing back into a previous state but rather some new state will arise and use this power against everyone else. When the woke crowd loses power, and they will, it might not be "the alt-right" that gets power but something along a completely different axis that no one is even imagining today. And they'll have all of these tools of control and oppression ready made for their abuse.
No, the IMF/NYT/News channel (or whoever) publishes opinion pieces all the time without agreeing with them.
What you might think is provocative, I see simply as topics that are designed to induce 'engagement'. While the two overlap, those are not the same thing.
In the end, editors have to decide whether it goes in. There needs to be a sufficient 'will' to make it happen. My point stands. There needs to be an internal support for any opinion pieces ( since that is the counter example you went with ).
Even the vice chair of Berkshire Hathaway has said he wanted a Chinese style system (among many others):
https://www.cnn.com/2021/06/30/investing/charlie-munger-chin...
It’s been real interesting to watch the conspiracy theories blend and become reality in some cases.
Personally, I think it’s more of a lose collection of similar minded people leading to certain conclusions and/or outcomes. But it does feel like an economic victory in civilization ...
. Do the ruling classes benefit?
. Is it practical/possible?
In any case, the easiest response is to remove yourself from the system as best as possible. Slowly getting grey'ed out by participating with the hivemind is no way to live.
Even if this were credible I'm not worried. The WHO has shown itself incapable of leading anything.