What Now for Crypto Banking?
Crypto banking is in a mess. The collapse of the FTX exchange caused damaging runs on two U.S. regulated banks. One of them – Silvergate Capital Corp. – had to sell assets at a loss to repay depositors and lenders, and as a result has now warned that it may not be able to continue as a “going concern.” A third bank had to issue an emergency warning that those who had deposited fiat funds with its customer Voyager Digital did not have deposit insurance.
U.S. regulators are pressuring banks to withdraw banking services for crypto platforms and exchanges. And in a decision that sent shock waves across the crypto world, the U.S. Federal Reserve rejected a membership application from Custodia Bank, a full-reserve bank providing payment and custody services to crypto businesses. Where can crypto-related banking go from here?
Frances Coppola, a CoinDesk columnist, is a freelance writer and speaker on banking, finance and economics. Her book “The Case for People’s Quantitative Easing” explains how modern money creation and quantitative easing work, and advocates “helicopter money” to help economies out of recession.
The regulatory clampdown on crypto banking began in earnest on Jan 3, when the Federal Reserve, the Federal Deposit Insurance Corporation (FDIC) and the Office of the Controller of the Currency (OCC) issued a joint statement on crypto-asset risks to banking organizations. We might wonder why they have only now, 14 years after the creation of Bitcoin and nine years after Tether first appeared, decided to cut the ties that bind the crypto ecosystem to the U.S. dollar system. But the answer is in their statement: “The events of the past year have been marked by significant volatility and the exposure of vulnerabilities in the crypto-asset sector. These events highlight a number of key risks associated with crypto-assets and crypto-asset sector participants that banking organizations should be aware of.”
They go on to give a comprehensive list of the risks they see in the present crypto ecosystem. Frauds and scams; legal uncertainties; inaccurate or misleading marketing (including claiming to have FDIC insurance) and “other practices that may be unfair, deceptive, or abusive”; crypto market volatility; run risk on stablecoin reserves; contagion risk due to extreme interconnectedness; poor risk management and governance; hacks and cyberattacks; generally heightened risk associated with open, public and/or decentralized networks. And they warn:
“It is important that risks related to the crypto-asset sector that cannot be mitigated or controlled do not migrate to the banking system.”
The message is clear. U.S. regulators think crypto is a serious threat to the traditional financial system. Not because it is going to take it over, but because it could bring it down.
For all their rhetoric about “unbanking yourself,” crypto exchanges, lenders and stablecoin issuers need access to banks. All dollar-denominated transactions except those made in physical notes and coins go through the U.S. banking system and are ultimately settled across the books of the New York Federal Reserve. Fintech payment apps such as Venmo and Zelle create the illusion that dollar payments can be made without involving banks. But a look under the hood of these companies reveals that they depend on a network of banks – indeed, Zelle is owned by a consortium of banks. And the same is true for international dollar payments.
Contrary to popular belief, international payments aren’t sent by SWIFT: SWIFT is merely a messaging service. Rather, just like domestic payments, international dollar payments are sent and received by banks and settled across the books of the New YorkFed. So it’s not possible for any crypto company, whether or not it has a U.S. presence, to accept or pay out fiat dollars unless it has a direct or indirect relationship with a U.S. bank.
Crypto companies also use banks to store the cash reserves that “back” customer deposits. But this isn’t strictly necessary: They could use money market mutual funds, for example, or hold short-term U.S. Treasury bills. And it’s also distinctly odd. Crypto companies like to claim that “unlike banks” they hold full reserves against customer deposits. But if the reserves are held in fractionally reserved banks, this isn’t true. It’s hard not to conclude that the real reason why many crypto companies hold cash in banks is so they can tell depositors they have some access to FDIC insurance. The FDIC has issued cease-and-desist orders to numerous companies, including the exchanges FTX and CEX, for falsely saying or implying FDIC insurance applied to their customer deposits.
But although payment and custody services are important banking functions, they are peripheral to the business model of a traditional bank. The business model of a traditional bank is to borrow at a low interest rate, lend at a higher interest rate and pocket the difference. Traditional banks are important liquidity creators and distributors, not only in financial markets but in the wider economy. As the U.S. discovered after the failure of Lehman Brothers in 2008, when banks stop lending the economy grinds to a halt. Crypto originally eschewed lending, but quickly discovered that without it the ecosystem was about as liquid as the surface of Mars. When there is much HODLing and little lending, liquidity is gold.
U.S. dollars are illiquid in the crypto ecosystem, so crypto companies don’t need lending services from traditional banks. And because they might have to pay out those scarce fiat dollars at a moment’s notice, they want any funds they deposit in traditional banks to be held in custody, not used as liquidity for the bank’s other activities. So crypto companies want a type of bank we don’t currently have: a “full-reserve” bank. In 2019, Wyoming created a charter for full-reserve banks. Its “special purpose depository institution” can receive deposits and provide asset management, custody and related services, but is not allowed to lend (though it can purchase certain types of debt securities) and must maintain unencumbered liquid assets of at least 100% of its total deposits.
Custodia Bank is a Wyoming SPDI. It does not lend, but provides payment and custody services:
“Custodia has all the benefits of being a bank with expertise in digital assets – plus, as a depository institution, we’re eligible to connect directly with the Federal Reserve payment system, removing middlemen and layers of fees.” In particular:
U.S. banks with direct access to the Federal Reserve via a master account can directly clear payments for their customers at the Federal Reserve, reducing the cost, delays, reconciliation headaches and counterparty risk involved with traditional intermediaries
Banks are defined as “qualified custodians” under the Investment Advisers Act and the SEC’s Custody Rule
Banks are defined as a “good control location” under the SEC’s Customer Protection Rule.”
A full-reserve, nonlending bank earning its profits solely from fees on payment and custody services sounds like a sensible solution to crypto’s banking needs. Indeed, many people might think it has wider application, too – after all, it’s not just crypto companies that need safe places to put their savings. If it is 100% reserved there’s no risk from bank runs, and if it is not lending it can’t become insolvent due to bad debts. Even though deposits in Custodia don’t have FDIC insurance, they are completely safe. So why has the Federal Reserve refused to accept Custodia as a member and denied it direct access to dollar clearing?
The problem is not Custodia, it’s the customers. Custodia offers banking and custody services exclusively to crypto businesses. And the Fed’s view of these is more than slightly jaundiced. It regards the entire crypto industry as a hotbed of financial crime. It’s hardly going to give the green light to a bank whose principal business is, in its view, enabling enterprises that are highly risky and at worst actually criminal to store and move U.S. dollars.
There’s a second problem: Custodia plans to issue its own token. The token would be a liability of Custodia exchangeable at par for U.S. dollars – a “tokenized dollar.” It would be fully reserved, and Custodia has applied for FDIC insurance. A regulated bank issuing fully reserved, FDIC-insured tokenized dollars that could be used on multiple blockchains seems like a great idea. It would make U.S. dollars considerably more liquid in the crypto space and reduce reliance on the likes of Tether. The problem is not the token, it’s the network.
Custodia plans to issue the token on Blockstream’s Liquid network and possibly also on Ethereum. These are public decentralized networks. Custodia would have no control over the ownership and distribution of such tokens. It would be as if it had issued its own banknotes. The joint regulator statement cited above says that issuing tokens on such networks is “highly likely to be inconsistent with safe and sound banking practices.” While regulators perceive crypto as a vehicle for money laundering, terrorist financing and ransomware, and the headlines are dominated by crypto-related frauds, scams and rug pulls, regulated banks are not going to be allowed to issue stablecoins on public networks.
If crypto companies stopped fighting the regulators and cleaned up their act, the Fed might look more favorably on banks providing services to them. But I’m not convinced Custodia’s business model makes commercial sense anyway. The reason why we don’t have full-reserve banks is that they are intrinsically less profitable than their fractional-reserve competitors. Historically, full-reserve banking has never lasted for long: Banks either find ways of leveraging customer deposits or they are bought by a fractional reserve bank or they go out of business.
The short-lived nature of full-reserve banking is already evident in the crypto world: Crypto lenders, exchanges and stablecoin issuers that promised depositors they were “fully reserved” have turned out to be anything but. In a world dominated by fractional reserve banks, I doubt if specialist full-reserve banks like Custodia would long survive.