Facebook Will Make the Money Now - BLOOMBERG
Also cheating on tests and renting Buffett’s house.
By Matt Levine
Money is a technology. This is true, first of all, in a grand abstract sense: The human capacity to generate collective fictions is our most powerful and general technology, the thing that distinguishes us from other animals and enables long-term cooperation and complex societies, and money is one of the most important collective fictions. 1 Humans were collectively able to decide that discs of shiny metal or sheets of engraved paper could be used as substitutes for all other goods and services, not because they were equivalents of those goods and services but just because we wanted them to be, and this decision created vast new possibilities for storing wealth and planning for the future and motivating human behavior and all the rest.
But most of us don’t walk around with a lot of gold coins or even all that much paper money, and the main actual technology of modern money is a bit more prosaic. It is the list. For most people who have money, most of their money consists of an entry in a list at a bank. Each bank has a list of its customers, and next to each customer there’s a number, and that number is the number of dollars that that customer has. It doesn’t, like, correspond to the number of dollars that the customer has; it is not a handy reminder that the bank has a box somewhere with a bunch of gold coins that belong to that customer. There’s no box. The dollars consist solely of the ledger entries at the bank.
This, too, is immensely powerful, especially combined with modern digital technology that lets the bank (and the customer) access the list from anywhere. But there is something too powerful, too easy, about it. The advantage of using gold as money is that it is hard to get a lot of gold: Once everyone agrees that gold is valuable, everyone will want gold, but not everyone will have much of it, so it will retain its value. Once everyone agrees that having a number on a list is valuable, I mean, it’s pretty easy to write a number on a list; right here on my computer I have Microsoft Excel, which allows me to write down long lists of large numbers and put dollar signs in front of them.
So the principal form of modern monetary technology is really legal and regulatory technology. Sure the computer systems that keep the lists of numbers at banks are powerful and complicated, so they don’t mess up the lists, but the thing that keeps the system afloat is mostly a set of rules and norms about who gets to keep lists of numbers of dollars. If your bank says you have some dollars in your checking account, then you do. If I say you have some dollars in my Excel spreadsheet, then you don’t. There is a regulatory privileging of who gets to be a bank, of who gets to be involved in the creation of money. There are central banks that can create money from nothing, and there are regulated banks that can create money under the auspices (and regulation) of the central bank, and there are regulated entities that can transfer and store and otherwise minister to money. 2 Money is just a bunch of computer entries, but it has to be scarce to be valuable, and the computer entries have to reflect a social consensus of who has how much money. 3 And so the structuring and regulation of banking is about making sure that the money is scarce and is allocated in socially approved ways.
Cryptocurrency proposes a different technological solution to this problem. The traditional solution is: If money is just a bunch of computer entries, then it won’t be valuable unless there is careful government regulation of who can create those entries and how. Bitcoin and subsequent cryptocurrencies said: Actually computer technology itself can be harnessed to limit how many computer entries are created, and to allocate those entries based on a social consensus of who is supposed to have them, all in a totally decentralized way with no government or regulation.
That was a big interesting technological innovation. It had to be coupled with a big interesting social innovation, which is just that some critical mass of people needed to look at the computer entries created by the Bitcoin algorithm and say “okay sure these are money.” And that happened, though in an imperfect sort of way: Bitcoin’s value is very volatile, and you can’t really buy much with it, but people do ascribe many billions of dollars of value to Bitcoins.
And so now, at a very high level, the technology exists to create decentralized money, money that derives its status and value not from the banking-regulatory apparatus of national governments, but from a combination of (1) computer algorithms that limit and allocate the money and (2) reasonably widespread social acceptance that this sort of thing can be money.
The next thing that Bitcoin enthusiasts will usually say is that this is an essentially democratizing process. You can see why! Central banks are mysterious, and banks are widely mistrusted, and there is a popular and not unreasonable sense that the banking system gives special privileges to favored insiders in unfair and inegalitarian and politicized ways. Cryptocurrency disintermediates banks and governments and traditional favored incumbents, and rests the power to create money in a form of direct social consensus. It is, crypto advocates say, the most powerful democratizing force since the internet.
But, you know … look at the internet. In a real, true, important sense, the internet is decentralized and libertarian and empowering; the architecture of the internet is not controlled by any government or corporation, and everyone can access and build on it. But in another, also quite real, sense, the internet has been a powerful tool for the centralizing of power. There are a zillion websites but you find them using Google; anyone can publish on the internet but you get your news from Facebook; anyone can sell anything on the internet but mostly you buy stuff from Amazon; email is a free and open protocol but you use Gmail. The democratizing effect of the internet’s openness and decentralization is counteracted by its vast economies of scale and network effects, which tend to concentrate power in a few big winners.
Basically the point here is that if you replace the traditional social-regulatory technology of money creation with a new sort of computer technology of money creation, odds are that the power of money creation will end up not so much in the hands of free-spirited individual hackers around the world, but in the hands of some giant tech company.
In this case, Facebook Inc.:
Facebook Inc. unveiled plans for a new cryptocurrency that the social-media giant hopes will one day trade on a global scale much like the U.S. dollar.
Called Libra, the new currency will launch as soon as next year and be what's known as a stablecoin -- a digital currency that's supported by established government-backed currencies and securities. The goal is to avoid massive fluctuations in value so Libra can be used for everyday transactions in a way that more volatile crypotcurrencies, like Bitcoin, haven’t been.
Libra is the culmination of a year-long effort to devise an easy way for Facebook users to send and receive money through its messaging services.
Oh yeah. Here’s the white paper describing Libra, and here are some ancillary papers about its blockchain and its programming language and its cash reserve and its compliance policies. Everything is at a pretty high level of generality—you can’t go out and buy a Libra today—but I suppose we have to talk about some specifics anyway.
So, first, Libra will be a “stablecoin,” in the sense that its value will be pinned to conventional financial assets. Unlike most stablecoins, though, it will be pinned to a basket “of low-volatility assets, including bank deposits and government securities” in different currencies, so it won’t consistently be worth one dollar or one euro or anything else. “As the value of Libra will be effectively linked to a basket of fiat currencies, from the point of view of any specific currency, there will be fluctuations in the value of Libra.”
This strikes me as very annoying, but it has some obvious advantages for Facebook/Libra. For one thing, they’ll only need to manage one general Libra, rather than having different stablecoins corresponding to different national currencies. For another thing, if Libra gains widespread acceptance, its lack of one-to-one correspondence will give it a tendency to displace national currencies. If you mostly spend dollars, and Libra is always going up and down against the dollar, that will be annoying and you won’t want as many Libras. But if you mostly spend Libras—if Facebook is successful at making this the main currency of the internet—then that dynamic will reverse. If the dollar is always going up and down against the Libra, that will be annoying and you’ll want more Libras. The dollar will start to seem unstable and useless. If you buy most things online, and if everything online is priced in Libras, then you’ll end up living your life denominated in Libras, and only converting your Libras into dollars on your occasional touristic visits to the physical world. 4 The goal is for Libra to be more useful than any national currency, accepted in more places and with fewer complications; pegging it to a single national currency would only hold it back.
Second, Libra will be fully reserved: For every Libra that is issued, one Libra’s worth of dollar/euro/yuan/whatever assets will be in a bank account or government bond somewhere. “The reserve will be held by a geographically distributed network of custodians with investment-grade credit rating to limit counterparty risk.” The initial reserve will come from investments by Facebook and its partners in the Libra Association; later, when Libras are issued for cash, the cash will be added to the reserve. 5 This limits the amount of fun that Facebook and its partners can have: They are not creating money out of thin air based on their own money-issuing clout; they’re just transmuting regular old fiat money into shiny new Libra money. But, you know, the white paper is pretty general and it’s pretty early days for Libra; if it really works—if Libra becomes the dominant currency of the world—then you could imagine this aspect being less important. If you pay for your rent and coffee and internet shopping in Libras, who cares if Libra is backed by dollars? 6
In any case Facebook and friends do get to have some fun with money creation now:
Users of Libra do not receive a return from the reserve. The reserve will be invested in low-risk assets that will yield interest over time. The revenue from this interest will first go to support the operating expenses of the association — to fund investments in the growth and development of the ecosystem, grants to nonprofit and multilateral organizations, engineering research, etc. Once that is covered, part of the remaining returns will go to pay dividends to early investors in the Libra Investment Token for their initial contributions. Because the assets in the reserve are low risk and low yield, returns for early investors will only materialize if the network is successful and the reserve grows substantially in size.
Isn’t it a little bit amazing that the banking system both lets you have access to your money to buy stuff at any time and pays you interest on that money? (I mean, not very much now, but in general.) It is a weird nice fact that that norm evolved, but it’s not clear that it would evolve again independently. The Libra tradeoff is: You can use your money, but we’re keeping the interest on it.
Third, Libra won’t be controlled by Facebook, exactly; there is an ancillary paper about the Libra Association, a consortium of participants that will provide governance for Libra and also run the blockchain nodes that manage the currency. (Each member also “needs to make an investment of at least $10 million in the network through purchasing Libra Investment Tokens,” to fund the initial reserve so that, effectively, Libra can give away some coins to jumpstart usage.) Facebook is obviously the driving force, and “will spin off a unit, called Calibra, to manage its own digital wallet service, which will be integrated into its family of apps”; presumably the biggest early use case for Libra will be for making payments through Facebook. But there’s a whole list of founding members—payments companies like Mastercard and Visa and Stripe and Paypal, tech companies like Uber and Lyft and Spotify and eBay, plus blockchain companies, venture capitalists and nonprofits—and the possibility of more being added.
There is also a pretty vague plan for “Moving Toward Permissionless Consensus.” Bitcoin is a permissionless blockchain; transactions on its system are maintained by anyone who wants to maintain them, secured by cryptography and incentivized by “mining” rewards. Libra is a permissioned blockchain; transactions on its system are maintained by servers run by members of the Libra Association, i.e. big tech and payments companies. This makes the Libra people nervous:
Permissionless systems have low barriers to entry and innovation, are resistant to censorship attacks, and encourage healthy competition among infrastructure providers (e.g., who can participate in consensus) as well as the developers of applications on top of the network. Since nobody can exclude others from the market or censor their transactions, permissionless systems provide stronger guarantees to participants that no single party will be able to unilaterally change the rules of the network to their advantage at a future date. At their core, permissionless systems make irreversible commitments to operating as open networks where changes can only be implemented if they are democratically supported by a majority of constituents.
That’s Libra saying that its own architecture is the wrong architecture, and that its “ambition is for the Libra network to become permissionless.” But not yet! “One of the association's directives will be to work with the community to research and implement this transition, which will begin within five years of the public launch of the Libra Blockchain and ecosystem.” Libra knows that you don’t necessarily trust Facebook to run the currency. It’s just asking you to trust Facebook to turn the currency over to a truly decentralized system, sometime, somehow, when it figures out what that system might look like.
Fourth, there are probably some things to say about regulation? Is this thing a payments processor? A money-market fund? A bank? I dunno, maybe, a little bit. 7 I cannot get too worked up about the regulatory framework applicable to it. For now I would prefer to be swept up by Facebook’s grand vision for it. That vision seems to involve competing with or even displacing national currencies, and if you’re going that far why not displace national regulatory regimes too? If we meet back here in 20 years and Libra has become the currency of the internet, we’re not going to be talking about whether Libra complies with banking regulation, we’re going to be talking about how the Libra Association regulates and stress-tests the Libra banks that it licenses. Libra says now that it is “committed to working with authorities to shape a regulatory environment that encourages technological innovation while maintaining the highest standards of consumer protection.” As Izabella Kaminska points out, that doesn’t sound like Libra sees itself as a passive taker of regulation: It wants to figure out what the right regulation for it should be, and then work with authorities to shape that regulation.
We talked last year about some trouble that accounting firm KPMG LLC got into. Basically KPMG hired a former employee of the Public Company Accounting Oversight Board, which oversees auditing firms, and on his way out the door at PCAOB he took some secret oversight materials about PCAOB’s inspection plans for KPMG. KPMG used those materials to game the inspections, prettying up its work papers for the audits that PCAOB intended to inspect and making KPMG look better than it otherwise would. The obvious analogy is to cheating on a test by getting the questions in advance, and both I and the Securities and Exchange Commission made that analogy. The SEC, in its enforcement action against several KPMG employees, called it “shocking misconduct—literally stealing the exam,” etc.
Now the SEC has brought a separate enforcement action against KPMG, which KPMG settled by agreeing to pay $50 million. Most of this action is about the same stealing-the-PCAOB-exam stuff that we talked about last year. But some of it is for much more literal cheating on much more literal exams:
KPMG audit professionals – at all levels of seniority – engaged in misconduct in connection with examinations on internally-administered training courses that were intended to test whether they understood a variety of accounting principles and other topics of importance.
This misconduct took a variety of forms. KPMG audit professionals shared exam answers with one another. A number of audit partners gave exam answers to other partners, and a number also sent answers to and solicited answers from their subordinates. In addition, for a period of time up to November 2015, certain audit professionals made unauthorized changes to KPMG’s server instructions that allowed them to manually select the scores necessary to pass the tests, which they often lowered to the point of passing exams with less than 25 percent of the questions answered correctly. The exams related to a variety of subjects that were relevant to the test-takers’ audit practices, and included additional training required by a 2017 Commission Order after the Commission found that KPMG engaged in improper professional conduct and had caused a client’s reporting violations.
Oh man. These were internal exams created and administered by KPMG, but they had a regulatory power; the accountants used them to meet continuing professional education requirements to keep their licenses. Having satisfied a certain number of continuing professional education requirements over the course of my various professional careers, I cannot say I am completely surprised that not everyone took them completely seriously, but you do hate to see it in print in an enforcement action. And of course the willingness to cheat on not-especially-important internal exams looks particularly bad when it’s in the enforcement action about stealing secret regulatory materials to cheat on audit inspections. It makes it look like a cultural problem, you know?
Also this is amazing:
Prior to November 2015, KPMG hosted exams to training programs on an internal server with software provided by a third party. KPMG sent participants in training programs a hyperlink that directed them to the applicable exams. Embedded in the hyperlink was an instruction to the server that specified the score necessary to pass the exam. Thus, the characters “MasteryScore=70” meant participants were required to answer at least 70 percent of the answers accurately to pass the exam. By changing the number in the hyperlink, audit professionals could change the score required to pass.
Ahahahahahahaha what, what, what? Imagine if, like, the SAT worked like that, where they’d email you a link to the test with “?HarvardScore=1520” at the end, and you could just retype the link with “?HarvardScore=800” instead and the test would be scored differently. Really any test set up this way deserves to be cheated, and any accountant who didn’t notice this flaw shouldn’t be trusted to do an audit.
Here is a story about a woman who lives in Warren Buffett’s old house in Omaha, and who rented the house out to some people attending the Berkshire Hathaway Inc. shareholders’ meeting this year, and who charged them $3,000, and my immediate reaction was: You can get way, way, way more for Warren Buffett’s old house during the Berkshire annual meeting. People pay millions of dollars to have lunch with Buffett; there is a thriving business of monetizing anything Buffett-related. An then of course I saw that she knew all this and is in fact way ahead of me:
Monen reaped $3,000, and is considering hiking the fee next year to spread the wealth to a charity. She's thinking an auction format might work best.
"I'm crazy not to capitalize on that," Monen said. "Warren would want me to."
Yep yep yep this all checks out. Buffett visited the house this year and wrote “the birthplace of Buffett Associates May 1956” on the door to the sunroom, which probably raises the rent by at least $20,000 a night.
Deutsche Bank seeks breathing space by slashing trading arm. There Is No Easy Fix for Legg Mason and Other Money Managers. Saudi Crown Prince Says Aramco's Planned IPO Is On Track. Private equity groups prepare to raise mega funds. Blank-Check Boom Hits Snag as SEC Blocks NYSE Plan. The Once-Forbidden U.S. Dollar Is Suddenly Everywhere in Venezuela. “ Trees don’t move as fast as Pro Bowl linebackers.” Boaty McBoatface makes major climate change discovery.
Obviously there are historical examples of “private banking” in which banks were not all that regulated or state-chartered, and in which more or less anyone could set up as a bank and start issuing notes, but (1) that arguably didn’t work as well and (2) they were sort of norm-chartered; there was still social acceptance of who was and wasn’t allowed to keep the ledgers.
J.W. Mason describes the consensus-allocation algorithm of money like this: “In the classroom, one of the ways I suggest students think about money is as a kind of social scorecard. You did something good — made something somebody wanted, let somebody else use something you own, went to work and did everything the boss told you? Good for you, you get a cookie. Or more precisely, you get a credit, in both senses, in the personal record kept for you at a bank. Now you want something for yourself? OK, but that is going to be subtracted from the running total of how much you’ve done for the rest for us.”
I mean, if all goes well for Libra the physical world will mostly take Libra too; credit-card and payment-processor companies are Libra partners, and so presumably the vision involves taking Libra at stores that take credit cards.
There's a sort of exchange-traded-fund-like mechanismfor actually transacting with Libra, where you have to go through an “authorized reseller”: “Users will not directly interface with the reserve. Rather, to support higher efficiency, there will be authorized resellers who will be the only entities authorized by the association to transact large amounts of fiat and Libra in and out of the reserve. These authorized resellers will integrate with exchanges and other institutions that buy and sell cryptocurrencies to users, and will provide these entities with liquidity for users who wish to convert from cash to Libra and back again.”
The deeper point here is that a fully reserved currency prevents fractional-reserve banking, and fractional-reserve banking is a useful thing. Now, the Libra white paper doesn’t actually prevent fractional-reserve banking; it just does that banking in dollars (or whatever).You deposit some dollars (etc.) with Libra, you get Libras, Libra goes and deposits those dollars in a bank, and the bank then can lend them to others. Libra is just a sort of transitory interface for dollar (etc.)-based banking. But if Libra became the world currency, you’d still want some way to do *banking*, to lend out deposits rather than just keeping them in a virtual vault. It is not worth worrying too much about the mechanics here, since Libra is very very far away from being the world currency, but if I were at the Libra Association and making my plans for world domination, this is the sort of thing I’d be thinking about.
Is it a *security*? Surely, I think, not; you don’t buy a Libra in the expectation of profit, and Facebook et al. aren’t issuing it as a corporate financing tool. The initial $10 million investments in the Libra Association do look more like securities—they’re a fundraising, and they carry the right to the interest on the Libra Reserve balances—but those are separate Libra Investment Tokens, which are being issued in an exempt private placement to accredited investors. They are securities, but they comply with the securities law.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story: Matt Levine at firstname.lastname@example.org
To contact the editor responsible for this story: James Greiff at email@example.com