Four Things From the Fed's Custodia Order
Crypto activities as principal, Glass-Steagall, risks of uninsured member banks, and monetary policy implications
Allowing Custodia Bank to become a member of the Federal Reserve System could “pose significant risk to its community.” That is one of several independent conclusions cited by the Federal Reserve Board as reasons why the Custodia’s membership application is inconsistent with approval, per a redacted version of the Board’s January order denying that application released yesterday.
I have my own previously acknowledged crypto-skeptical biases, so the bottom-line conclusion of the Board’s order — specifically, that considerations relating to the managerial, financial and corporate powers factors the Board is required to consider under the Federal Reserve Act each independently warrant denial of Custodia’s application — seems reasonable enough to me.
On the other hand, perhaps with the potential for litigation in mind, the Board’s order does at times read a bit like it is stretching to include every single even vaguely plausible justification for denial, sometimes without much supporting argument. This makes the order in places read like something of a laundry list.1
There are also portions of the order that seem difficult for the public to fully evaluate. For instance, the Board says that its pre-membership examination of Custodia found what it called “significant risk management gaps,” including in relation to BSA/AML and OFAC matters. These deficiencies led the Board to conclude that at the time of examination “Custodia had not demonstrated that it could satisfactorily manage the significant ML/TF risks presented by its proposed business plan.”
On the other hand the American Banker yesterday reported that, per a Custodia spokesperson, “the bank engaged with its primary regulator, the Wyoming Division of Banking, as well as an independent compliance consultant last fall, both of which gave its risk management practices and controls a clean bill of health.”
Where there are differences of opinion like these, the Federal Reserve Board usually wins. I am guessing that is where I would side too, but without being able to dive into the granular details we are sort of forced to take the Board’s word for it. (TBD, perhaps, on how receptive a court will be to this sort of argument.)
For all these reasons, today I am going to take the easy way out. Rather than focusing on whether the Board’s ultimate conclusions about the merits of Custodia’s business model and risk management practices are accurate, today’s post instead focuses on four other things in the order I thought were interesting, particularly to the extent they could have implications for banks other than Custodia.
Holding Crypto Assets as Principal
On the same day in January that the Board announced its denial of Custodia’s membership application it also released a statement interpreting Section 9(13) of the Federal Reserve Act. That statement observed that the Board had not identified any authority permitting national banks to hold most crypto-assets, including bitcoin and ether, “as principal in any amount.” Thus, because national banks are prohibited from holding crypto as principal, the Board would likewise presumptively prohibit state member banks from doing so.
As relevant to the membership application, one of Custodia's proposed activities was to hold a small amount of bitcoin and ether as principal so that Custodia could pay certain transaction fees on behalf of customers. Custodia believes it must pay such transaction fees using crypto held in its own account because the transfers would be done at Custodia’s direction, rather than at the direction of the customer. Wyoming law also may require this.2
The Board ultimately concluded that such principal activities would be impermissible for a national bank, and thus under its policy statement would be presumptively prohibited for a potential member bank such as Custodia.
The Board’s order gives a few indications as to Custodia’s counterargument. As summarized by the Board, Custodia apparently argued that a line of OCC interpretations stand for the proposition that a national bank may, as part of conducting a permitted activity, also conduct a small amount of impermissible activities if conducting such impermissible activities is necessary to ensure the bank’s conduct of the permissible activity is not made “infeasible, burdensome or less advantageous.”
Again according to the Board, Custodia cited to the following examples:3
OCC Conditional Approval No. 578: Holding certain fund interests for a limited period of time is permissible for a national bank where the holdings are “integral to facilitating a recognized bank-permissible activity” (investment management). See also OCC Interpretive Letter No. 940.
OCC Interpretive Letter No. 901: A national bank that purchased life insurance policies from an insurance underwriter doing business in mutual form is permitted to retain shares in the underwriter when the underwriter converts to stock form, as the bank’s receipt of the stock “is probably most properly characterized as a byproduct of the permissible activity of purchasing life insurance for the Bank’s needs” and “does not raise any safety and soundness concerns, according to examining personnel who have considered it.”
OCC Corporate Decision #2006-06: Sometimes when a national bank purchases or holds certain tax credits (a permissible activity), the national bank is also required to acquire an interest in the limited partnership that manages the project generating the tax credits. Under OCC precedents it is permissible for a bank to hold this “severely circumscribed” limited partnership interest because doing so is “needed to facilitate the Bank’s participation in the permissible financial intermediary activities and, therefore, is incidental to the business of banking.”
OCC Corporate Decision #98-17:4 A national bank would extend credit to an LLC that operates oil and gas leases (the Borrower). An operating subsidiary of the national bank would then purchase for $1 from the Borrower interests in certain natural gas leases. The transaction would be structured in this way so that the leases would qualify for tax credits. Previous OCC interpretations had held that "while it might appear that the bank was acquiring an interest in real estate in violation of 12 U.S.C. § 29 such a narrow view of the statute would elevate form over substance." The OCC reached a similar conclusion here, saying that "in substance the Bank is extending credit to the Borrower, receiving a security interest in the natural gas reserves, and receiving repayment of the funds advanced from sale of the natural gas reserves.” Thus, the transaction is best viewed as a permissible extension of credit by the bank.
The Board dismisses Custodia’s reliance on these letters in pretty summary fashion, saying that the OCC precedents cited by Custodia only “recognize limited authority to conduct carefully defined activities.” It is not clear how significant Custodia’s principal activities would be (see the redactions in footnote 47), but the Board’s implication here seems to be that Custodia’s activities as principal would not be similarly limited or carefully defined.
Also unconvincing to the Board were Custodia’s arguments about the activities that bank holding companies have sometimes been permitted to conduct as principal in connection with BHC-permissible activities.
Custodia cites instances when the Board found that certain activities are permissible in limited circumstances in connection with a permissible activity under the BHC Act. Id. In addition to not being legally relevant to bank-level permissibility, decisions made by the Board in such contexts do not implicate the competitive parity considerations applicable with respect to bank activities supervised by various regulators.
Section 21(a)(1) of the Glass-Steagall Act
The late 1990s repeal of certain sections of the law receive the most attention, but various provisions of the Banking Act of 1933, commonly known as the Glass-Steagall Act, remain in place. One still effective provision of Glass-Steagall, Section 21 (12 USC 378), has occasionally been cited as having implications for certain crypto activities.
The focus in this context has typically been on nonbank stablecoin issuers, given that Section 21(a)(2) of the Glass-Steagall Act prohibits anyone from engaging, “to any extent whatsoever” in the business of receiving deposits unless licensed and regulated as a bank or otherwise subject to certain federal or state oversight.
Custodia arguably doesn’t have to worry about Section 21(a)(2) as it is chartered as a bank in Wyoming and supervised by the Wyoming Division of Banking. The Board’s order suggests, however, that it might be Section 21(a)(1) that poses a problem for Custodia. Subject to certain exceptions, that section prohibits any company engaged in the business of taking deposits from also engaging in the business of issuing securities.
Thus, if Custodia’s issuance of dollar-denominated tokens, which it calls Avits, would amount to the issuance of securities, then Custodia could not both issue Avits and take deposits.
A footnote in the Board’s order says that Custodia “has declined to conduct an analysis” of whether Avits are securities under Reves, and that instead Custodia believes that Avits “are more appropriately evaluated exclusively under Marine Bank v. Weaver.” The Board says it will not opine on this argument, but then seems to suggest that in any case it might not matter. According to the Board, a 1971 Supreme Court case stands for the proposition that Section 21(a)(1)’s reference to “securities” should not be interpreted based on “a narrow reading” of that term.
Ultimately, the Board punts on this question as quoted below, but I thought this was an interesting marker for the Board to lay down, given that, as it says, it believed there were already a multitude of other grounds on which Custodia’s application could be denied.
The Board has not reached a conclusion on this issue, but does not believe that a finding regarding whether Avits would be securities under the Glass-Steagall Act is necessary in light of the multiple, independent bases for denial of Custodia’s application established herein.
Membership in the Federal Reserve System by Uninsured Deposit-Taking Banks
The Board’s order states that permitting an uninsured bank that accepts deposits other than trust funds to become a member of the Federal Reserve System “would be unprecedented in modern times.”
The Board explains that the Federal Reserve Act requires it to promote the effective operation of the economy, which includes promoting stability of the banking system. According to the Board’s order, deposit insurance is “central to achieving that stability” and admitting an uninsured deposit-taking bank as member of the Federal Reserve System could therefore “present risks to the stability of the Federal Reserve System by omitting a critical tool in preventing bank runs.”
The Board also believes that admission of an uninsured bank that takes deposits other than trust funds would require the Board to “create bespoke regulatory and supervisory frameworks,” including in light of questions over how such an uninsured bank could be resolved, given that the FDIC’s receivership regime would not apply.
Though nominally about Custodia, I wonder if this discussion in the order also has implications for other proposed de novo banking activity that was recently under consideration and that may be again in the future.
A few years ago Figure Technologies applied to charter a national bank that would take deposits, but would limit those deposits to only uninsured deposits above $250,000. Various groups disputed the OCC’s authority to charter such a bank.
Custodia is a state-chartered bank and Figure, which later modified its application and business model in a way that mooted some of the legal challenges, would have been a national bank. As a result, the applicable provisions of the statutes, and thus the underlying issues, are not 100% the same between the two applications.
Nonetheless, I think this suggests that even assuming the OCC does indeed have the authority to charter a national bank with the business model that Figure at one point had proposed, the Federal Reserve Board would still have serious concerns about the implications that admitting a national bank like Figure as a member bank could have for financial stability.
The Board closes this section of this order by making clear that it is not categorically ruling out the possibility that an uninsured bank may be able to overcome the issues it has identified:
The Board acknowledges that, notwithstanding the lack of precedent, Congress has designed a legislative framework that makes it possible for an uninsured depository institution to become a state member bank. Future applicants may present business models and control frameworks to overcome the concerns presented by uninsured deposit-taking banks.
At the same time, the Board says that Custodia has failed to “provided sufficient justification to break from longstanding precedent.” One could reasonably suspect that other banks would find it similarly difficult to produce what the Board views as sufficient justification.
Monetary Policy Concerns
Toward the end of the order, the Board expresses concerns that admitting Custodia to the Federal Reserve System could potentially undermine the Federal Reserve’s dual mandate to conduct monetary policy so as to promote maximum employment and price stability. The Board’s reasoning goes like this.
Custodia would plan to issue Avits, dollar-denominated tokens that Custodia describes as the digital equivalent of a cashier’s check, which it would then treat as deposits for purposes of federal banking law.
All Avits would be backed with funds held in a Custodia master account (if granted) at a Federal Reserve Bank.
These funds would be assets on Custodia’s balance sheet and liabilities on the Federal Reserve’s balance sheet.
Because Avits would be backed by deposits in a Federal Reserve master account, the Board believes the general public could view Avits as implicitly backed by the Federal Reserve.
This could lead to Avits, or other similar products issued by other banks, quickly growing at scale and becoming a tool for people around the world to instantly get access to U.S. dollars.
Such growth could lead to “new, meaningful, and volatile” sources of demand for Federal Reserve liabilities.
Because the Federal Reserve “generally accommodates changes in the level or volatility of demand for its liabilities through changes in the amount of assets (generally U.S. Treasury securities) that it holds,” such changes in demand for Federal Reserve liabilities, if large enough, “could complicate the Federal Reserve’s ability to manage the size of its balance sheet and overall conditions in the federal funds market.”
The Board has expressed similar concerns in other contexts,5 so this is not necessarily surprising, but it is noteworthy to see these views articulated in a formal order.
Thanks for reading! Thoughts, challenges, criticisms are always welcome at bankregblog@gmail.com
For example, there is a reference to the Volcker Rule on pages 59-60 of the order that I thought was sort of out-of-left-field. The context of this is that the Board believes Custodia’s fundamental focus is to support investment in and trading of crypto-assets by institutional and high-net-worth investors. The Board (reasonably!) views crypto-assets as highly speculative instruments, and asserts that lawmakers and policy-makers have frequently taken steps to separate banks from such speculative activities. There is then a string of references to certain provisions of the Glass-Steagall Act, to federal prohibitions on banks dealing in or participating in lotteries, and to the Volcker Rule, which the Board describes as seeking “to prohibit banks and bank affiliates from engaging in speculation … by limiting proprietary trading.”
The Board’s order takes no position on whether this is in fact the case, instead describing it as a Custodia assertion: “Custodia has also asserted that Wyoming law requires it to hold certain crypto-assets as principal in order to custody crypto-assets because cryptoasset custodians must have on hand sums required for the execution of transactions.”
According to the Board’s order Custodia also cited to a 1986 OCC interpretation regarding riskless principal transactions. I’ve not included this interpretation in the bullets above because it does not appear to be publicly available online, but the Board characterizes this interpretation, OCC Interpretive Letter No. 371, as standing from the proposition that in connection with riskless principal activities national banks may inadvertently become a principal because of the failure of a transaction.
The Board’s order cites to Corporate Decision #98-18, but I think that is a typo.
For example, in its ANPR on whether it should amend Regulation D to address certain policy concerns that could be raised by narrow banks, the Board observed that, among other potential ways that narrow banks could complicate monetary policy implementation:
… PTIEs could potentially attract a large quantity of deposits and maintain very large balances at Reserve Banks. The ability of PTIEs to attract a very large amount of deposits at a rate above other key overnight money market rates could affect the FOMC's plans to reduce its balance sheet to the smallest level consistent with efficient and effective implementation of monetary policy. Specifically, if deposits at PTIEs were to become an especially attractive asset for cash investors, the demand for reserve balances by PTIEs could become quite large. In order to maintain the desired stance of monetary policy, the Federal Reserve would likely need to accommodate this demand by expanding its balance sheet and the supply of reserves.
See also this speech from former Governor Brainard on digital currencies and stablecoins.
Moreover, widespread adoption of stablecoins could have implications for the role of central banks and monetary policy. Payments are the economy's circulatory system. Large-scale migration into a new stablecoin network for purposes of payments may prove to be the leading edge of a broader migration. If a large share of domestic households and businesses come to rely on a global stablecoin not only as a means of payment but also as a store of value, this could shrink demand for physical cash and affect the size of the central bank's balance sheet. The central bank's approach to implementing monetary policy may be complicated to the extent that banks' participation in short-term funding markets is affected.