So much for Wyoming blockchain regulatory capture — the Fed blocks Custodia Bank

By Amy Castor and David Gerard

• Writing this good stuff takes time, and time is money. Send Amy money here and David here. One weird trick that crypto doesn’t want you to know!

Caitlin Long had high hopes of bringing crypto into the mainstream financial system. Long’s Custodia Bank in Cheyenne, Wyoming, a state-charted special purpose depository institution, or SPDI, was spearheading those dreams for all of cryptocurrency.

Those dreams, and the clown car they rode in on, crashed into the concrete bollard of reality on January 27, 2023, when the Federal Reserve Board denied Custodia’s application to be a Federal Reserve member bank, and the Federal Reserve Bank of Kansas City denied its application for a master account. [Press release; American Banker, archive]

The Board’s full 86-page order, made public on March 25, details how Custodia failed to address “the heightened risks associated with crypto activities, including its ability to mitigate money laundering and terrorism financing risks.” (The Kansas City Fed’s denial letter was not made public.) [Order, PDF]

The Board’s order phrases it in bureaucratic euphemisms — but they understood extremely well that this was a gang of inept grifters. The fundamental issue is that Custodia have no idea what they’re doing. The order notes repeatedly that:

considerations relating to the managerial factor are so adverse as to present sufficient grounds on their own for warranting denial of the application.

 

 

The elusive Fed master account

If you want access to the Fed’s payment rails, including Fedwire, to settle transactions with other member banks in central bank money, you need your own account at one of the twelve Federal Reserve Banks.

A master account would be cheaper and much more convenient for Custodia, because they wouldn’t have to go through an intermediary bank.

Custodia applied for a master account with the Federal Reserve Bank of Kansas City, which covers Wyoming, on October 29, 2020. Separately, it applied to become a member of the Federal Reserve on August 5, 2021. State-chartered banks aren’t required to become Federal Reserve member banks — but if you want a master account, it really helps to become a member. [Custodia press release]

The local Federal Reserve banks (private banks, highly regulated) make decisions on granting master accounts — but it’s the Federal Reserve itself (part of the US government) that sets the rules.

The job of the Federal Reserve is to make sure that member banks don’t introduce risks to the US financial system or facilitate illicit activities.

You don’t get Fed membership just by saying your state calls you a “bank.” The Board evaluated Custodia’s application on four factors:

• Management: is the management competent?
• Financial: Does the bank’s funding model make sense?
• Corporate powers: Is the bank’s business consistent with the purposes of a Federal Reserve account?
• Convenience and needs: does the bank serve its community?

Traditional banks generally get approved for a master account within days. But Custodia was doing novel things. It wanted to specialize in crypto, an industry rife with fraud; it wanted to issue its own dollar substitute — a stablecoin called Avit; and it had no FDIC insurance. Custodia wanted to be a wildcat bank with central bank money.

As a result, it took more than two years for Custodia to get a response on its account application.

Prior to the Fed releasing its final guidelines, Custodia filed a suit in the Federal Court of Wyoming against the Federal Reserve Board and the Federal Reserve Bank of Kansas City in June 2022, alleging a “patently unlawful delay.” [complaint, PDF; case docket]

Custodia believes that the Fed is required to give a master account to any state bank that asks. Both the Kansas City Fed and the Board argue that the law really doesn’t say that, and the Fed has frequently exercised its discretion not to grant membership or accounts automatically. Custodia also believes there’s a conspiracy to exclude it, and crypto in general, from the financial system. The lawsuit is still ongoing.

The Fed had serious concerns about crypto getting mixed up with central bank money — and master accounts were becoming a contentious issue. The local Feds were increasingly getting applications for master accounts from fintechs and non-traditional banks.

So the Fed proposed a new master account guideline in May 2021 and released the final guideline on August 22, 2022 — to make it clear that “institutions that engage in novel activities and for which authorities are still developing appropriate supervisory and regulatory frameworks would undergo a more extensive review.”

The guidelines list three tiers. Tier 3, which Custodia fell into, would require the most scrutiny. [Press release; Guidelines, PDF]

On January 3, 2023, the Federal Reserve issued a joint statement with the OCC and the FDIC about crypto: “​​It is important that risks related to the crypto-asset sector that cannot be mitigated or controlled do not migrate to the banking system.” And on January 27, the same day that Custodia received its rejections, the White House released its “roadmap” to mitigating cryptocurrency risk, which directs regulatory agencies to “ramp up enforcement.” [FDIC; Whitehouse]

If you want to get your bank into crypto, you have to satisfy the Fed that you understand the risks. Custodia doesn’t seem to have understood the question.

The grand dream

Bitcoin was created by extreme libertarians who resented financial regulation as a moral offense against free individuals. Their economic ideas were based in conspiracy theories.

Even when the crypto world tries to work with the world of actual money, they still have this attitude — they fundamentally don’t understand how anything works, and resent the idea that any of the rules should apply to them.

But they still want the government to save them when it all goes south.

Long, a Wyoming native, came up with one weird trick to let crypto work around a whole swathe of regulations — the regulatory capture of the least populous state in the US, that already leaned libertarian. This would definitely work!

In November 2017, Long launched the Wyoming Blockchain Coalition. The 2017 bitcoin bubble was in full swing. The group’s stated goal was to promote crypto as a business opportunity for Wyoming. [CoinDesk, 2017]

The Coalition’s actual program was:

• Create a new Wyoming state bank charter for crypto;
• Launch a “bank” using this form of charter;
• Gain access to New York customers’ money without having to go through the rigmarole of applying for a New York BitLicense;
• Obtain a Federal Reserve master account, which would thus back deposits of crypto with US government dollars.

The Coalition promoted legislation to this effect through 2018. In February 2019, Wyoming created  SPDIs, a variant on trust companies, so that crypto companies could easily get a “bank.” These institutions would hold full reserves and not have FDIC deposit insurance. The structure is that of a state trust company. [CoinDesk, 2019; WyoFile, 2019; Wyoming Division of Banking]

SPDIs are only sort-of banks. They’re not allowed to lend dollars. Under the Bank Holding Company Act, to qualify as a bank, a firm has to make loans. [Federal Reserve History]

Custodia planned to pay zero interest on deposits — and charge users for transactions.

Lending money is the main way that banks make a profit. The lending that Custodia planned to do was in unregulated crypto assets — while exposing the central bank deposits to risk.

The crypto world hoped that a state charter for an SPDI would be a trump card that would let them work around all that tedious banking regulation, and all the problems that crypto companies had with getting banking at all. [CoinDesk, 2020]

Or as the American Banker’s Association put its concerns: [ABA, PDF]

The ultimate goal of the new charter is to gain direct access to the Federal Reserve and the US payments system without being subject to the same regulatory and supervisory framework with which banks must comply. Indeed the business model behind the charter is to intentionally sidestep this important framework.

At CoinDesk Invest: NYC in November 2019, the Wyoming crew — both the blockchain promoters and their pet regulators — talked up their goal to just waltz around New York’s “onerous” BitLicense. [CoinDesk, 2019]

The theory was that because New York exempts national banks from the requirement to obtain a BitLicense to operate, Wyoming’s SPDIs — which are state-chartered —  should also be exempted.

This is an untested idea. New York-based trust companies are automatically exempt from the BitLicense, but it’s not clear if out-of-state ones are.

The Kraken crypto exchange and Charles Hoskinson’s IOHK, the company behind the Cardano cryptocurrency, were also involved in putting together the legislation. [WyoFile, 2019]

Wyoming passed other laws to encourage the use of crypto — sports betting using crypto, a law to incorporate your DAO as an LLC, and Long even tried to sell an obsolete coal power plant to bitcoin miners. [WyoFile, 2019; CoinDesk, 2021; CoinDesk, 2021]

Wyoming locals weren’t happy with crypto industry people coming to town and writing laws to benefit only themselves. Wyoming entrepreneur David Dodson wrote: “By allowing the laws to be shaped and written by those who benefit from them, our legislative hard work accomplished everything the blockchain industry wanted but nothing the state needs. Namely, jobs and revenue.” [WyoFile, 2020]

Kraken Bank was the first to get an SPDI charter in September 2020. Custodia (which called itself Avanti until early 2022) followed in October 2020, then Wyoming Deposit & Transfer in June 2021, and Commercium Financial in August 2021. [CoinDesk, 2020; CoinDesk, 2021; CoinDesk, 2021]

Custodia promptly waved its shiny new SPDI charter at the Kansas City Fed and asked for a master account. This would allow a world of crypto innovation — backed by the Fed!

Custodia has so far received $45.5 million in venture funding to support its efforts. It has no other income as yet. [Crunchbase]

Custodia’s business plan

Custodia planned to offer services online in all 50 states, and later internationally:

• Core banking: Deposit accounts for crypto businesses and the wealthy — ACH, wire transfers, and so on. Cash deposits could only come from corporate customers in amounts of not less than $5,000. Custodia could not lend dollars.
• Custody services: Crypto would be held in a separate trust to ensure separation from customer deposits. Custodia calls this bank charter a “bridge” between crypto and dollar markets. Customers could use crypto to make direct payments, invest, and so on without having to convert their tokens into fiat.
• Prime services: Buy, sell, borrow, and lend crypto assets — crypto exchange services, with Custodia making its money on fees.
• Avit: A dollar stablecoin, running on Ethereum and Liquid.

Those first three add up to a crypto exchange whose US dollar banker is the Fed, with an attached US dollar bank for crypto businesses. The fourth proposes a stablecoin whose reserve is kept at the Fed.

The Fed said that it couldn’t let Custodia get into the last three businesses at all, and that Custodia lacked a sufficient business plan to do even just core banking sustainably — nor did it have the managerial competence to start an unusual new banking business safely.

A crypto exchange by any other name

Running a bank is a well-understood business, right? But first, you need competent bankers who understand the regulatory requirements and aren’t clowns. Per the Fed:

The findings of Federal Reserve staff’s pre-membership examination suggested significant deficiencies in Custodia’s ability to manage the risks of its day-one activities, which consist of limited basic banking services.

Specifically, the findings indicated Custodia’s risk management and controls for its core banking activities were insufficient, particularly with respect to overall risk management; compliance with the Bank Secrecy Act (“BSA”) and U.S. sanctions; information technology (“IT”); internal audit; financial projections; and liquidity risk management practices.

Custodia’s core banking business was projected to be a relatively small part of its overall business. The real business was the stablecoin and the crypto exchange — but core banking had no real income stream without those.

Custodia wanted to do novel — indeed, “unprecedented” — things. Allowing innovation to prosper requires understanding the risks. The Board saw no sign that Custodia understood the risks.

The lack of effective anti-money laundering seems to have been the main issue. AML is not a nice extra you can bolt on later. The Fed requires that you have your AML in place already.

Custodia planned to offer custody services for bitcoin and ether. The sticking point for the Board on this one was Custodia’s plan to keep crypto on its books “to pay customers’ transaction fees”:

The Board has not identified any authority to support the position that national banks are permitted to hold bitcoin, ether, or most other crypto-assets as principal in any amount or for any purpose … state banks have not been expressly permitted to do so by federal statute or part 362 of the FDIC’s regulations.

“Prime services” would allow customers to buy and sell cryptos via Custodia, who would get orders from other exchanges as the back end.

Custodia would also let customers borrow and lend cryptos that were held in trust accounts.

Avit — wildcat banking, but on the blockchain

Avit was the big one. Avit was Custodia’s plan to issue its own dollar stablecoin on Ethereum and Liquid — with the backing reserve stored at the Fed!

The Board compared Avit to Tether and USDC. Users who weren’t even customers of Custodia could hold, redeem or transfer Avits between themselves. Avit tokens would be wildcat banknotes — private dollar bills, flowing about freely outside any possible control by Custodia.

Custodia told the Fed that Avits would not be stablecoins, but “a new payment technology.” Nevertheless, the Board assessed Avit as a “stablecoin.” OCC Interpretive Letters 1174 and 1179 specifically permit a national bank to issue “stablecoins” for payments. [Letter 1174, PDF; Letter 1179, PDF]

The Board had “broader concerns” — a term that means “you fail”, much as “uncertain” means “you fail” — about a stablecoin that could be held by persons unknown to the issuer, or that ran on a decentralized network where the issuer had no control over the validators handling its token.

OCC Interpretive Letter 1174 specifically requires a stablecoin to be able to “obtain and verify the identity of all transacting parties, including for those using unhosted wallets.” That would include third-party users of your token on Ethereum, for example. Custodia could freeze or blacklist Avits after the fact, but the Board didn’t consider that sufficient.

As an Ethereum ERC-20 token, Avit would be tradeable outside the control of Custodia, and usable as a dollar in DeFi trading.

The Board was not happy that Avit would rely on unknown and unknowable validators, who might affect Avit holders in ways Custodia couldn’t easily remedy.

The Board’s biggest worry was the risk of bank runs on Avit — users rushing to cash out in bad times. All transactions would be visible on a blockchain — so anyone could see a bank run happening live.

Avit would be 100% backed, but that’s not enough — because bank runs are notoriously contagious. The Fed notes: “Runs on any bank or financial intermediary have been documented to lead to panic and contagion that spreads to other banks and financial intermediaries.” A bank run anywhere in the Federal Reserve system would spur wider panic.

Precisely such a series of contagious bank runs — Silvergate, then Silicon Valley Bank, then Signature — happened just weeks after Custodia’s applications were rejected.

Avit would be backed by dollars on the Federal Reserve’s balance sheet. That’s a feature for prospective Avit users — but not for the Fed. If Avit became popular, it “could generate particularly pronounced demand for Federal Reserve liabilities.”

The Avit would be sufficiently close to a Federal Reserve Bank dollar that it could “plausibly become a tool for persons around the world to access the stability of the U.S. dollar instantly and anonymously” — that is, to launder money at scale.

Custodia’s plan explicitly stated that there would be Avit users who were not known to them. Custodia thought that this unknowability would absolve them of BSA/OFAC responsibility for non-customer movements of Avit, and that they wouldn’t be required to file suspicious activity reports (SARs).

The Board was not buying this theory, and it was absolutely not going to enable a financial instrument that would set up a laundromat for US dollars.

Nor was the Board happy that crypto is a disaster area, exposing ordinary Avit holders to all its risks:

they could be traded on largely unregulated or noncompliant exchanges; lent on crypto-lending platforms; and invested in decentralized finance protocols. Each of these poses risks to Avit holders, as demonstrated by the bankruptcies of FTX, Voyager, Celsius, Blockfi, and the collapse of the Terra/Luna protocol.

These concerns strongly suggest that the Fed will never authorize a bank to issue a stablecoin on a public blockchain.

But what about the community?

Custodia doesn’t have a local community to serve. Instead, it defined its “community” as “the crypto-asset market.”

That’s fine. But Custodia’s business plan wasn’t sustainable, its deposits wouldn’t be backed, and its plan for resolution if it failed was to sell itself to other crypto companies — who would also be in trouble if crypto took a downturn. As we’ve seen happening through 2022, when would-be rescuers promptly fail in turn.

As such, Custodia would not be the savior of crypto — in fact, “the current record indicates Custodia could pose significant risk to its community.” Whoops.

Competence is required

The board wants to see bank proposals from companies whose management has actual banking experience. Custodia has none of that, and Custodia’s management team has very little understanding of how banks actually work:

The depth of banking experience and bank-specific risk management experience among the board of directors and management team is limited, and Custodia’s board, executives, and staff come from a variety of backgrounds that are largely outside of traditional commercial banking, which is the context in which the pre-membership examination was conducted based on Custodia’s proposed day-one activities.

… The number and degree of shortcomings identified in the pre-membership examination suggest that management’s experience is not commensurate with the firm’s intended risk profile.

The Board notes “significant turnover in management” at Custodia.

These experience-free aspiring bankers were going to specialize in one limited sector of the economy, which also happened to be a disaster area:

In general, the Board has heightened concerns about banks with business plans focused on a narrow sector of the economy. Those concerns are further elevated with respect to Custodia because it is an uninsured depository institution seeking to focus almost exclusively on offering products and services related to the crypto-asset sector, which presents heightened illicit finance and safety and soundness risks.

With the Fed/FDIC/OCC statement on banks and crypto, and the recent string of bank failures, there’s no way they’re going to let bozos like Custodia loose on the actual economy where people live. The US government will not become your bagholder of last resort.

What happens next?

The Fed has denied Custodia’s member application “without prejudice,” meaning they can apply again — but Custodia would be expected to show that they can address all of the Fed’s objections. The trouble is that Custodia doesn’t understand the questions.

The Wyoming blockchain plan was always delusional. They thought they could use one weird trick to get around regulation — any kind of regulation. This was never going to work even before crypto screwed up as hard as it did all through 2022.

Caitlin Long is still a crypto conspiracy theorist. As are the hosts on CoinDesk TV, because they know their viewers. That’s great if you need to market yourself to crypto cranks — but it doesn’t play so well in the world of real money. [YouTube]

Custodia won’t stop. They’ll see this through to the bitter end — because there’s nothing left to do except set the rest of their venture funders’ money on fire. There are executives and lawyers to pay, after all.

The Fed, the OCC, the FDIC, and the Treasury will be keeping an eye out in the future for attempts at regulatory capture of a small state by grifters trying to use it as a way into the real economy.