Public Comments

Status: Original and derived artifacts are available for this document.

Download: pdf

PUBLIC SUBMISSION

As of: 4/22/26 10:00 PM
Received: April 13, 2026
Status: Posted
Posted: April 14, 2026
Tracking No. mnx-qb9m-ak35
Comments Due: May 01, 2026
Submission Type: Web

Docket: OCC-2025-0372
Implementing the Guiding and Establishing National Innovation for U.S. Stablecoins Act for the Issuance of
Stablecoins by Entities Subject to the Jurisdiction of the Office of the Comptroller of the Currency
Comment On: OCC-2025-0372-0001
Guiding and Establishing National Innovation for U.S. Stablecoins Act for the Issuance of Stablecoins by
Entities Subject to the Jurisdiction of the Office of the Comptroller of the Currency
Document: OCC-2025-0372-0064
BitGo Holdings, Inc.

Submitter Information
Address: United States,
Email: [email protected]
Submitter's Representative: Michael A. Belshe and Charles D. Thompson
Organization: BitGo

General Comment
See attached file(s)

Attachments
BitGo Holdings, Inc.

file:///C/...25-0372%202026-04-22%2022-00-23/OCC-2025-0372%202026-04-22%2022-00-23%20-%20TEST/OCC-2025-0372-0064.html[4/26/2026 11:30:47 PM]

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

April 13, 2026
Chief Counsel’s Office
Attention: Comment Processing
Office of the Comptroller of the Currency
400 7th Street SW, Suite 1E–216
Washington, DC 20219
Re: Implementing the GENIUS Act for the Issuance of Stablecoins by Entities Subject to the
Jurisdiction of the Office of the Comptroller of the Currency (Docket ID OCC–2025–0372; RIN
1557–AF41)
To whom it may concern,

I. Interest of the Commenter
BitGo Holdings, Inc. (“BitGo”) submits this comment on the Office of the Comptroller of the
Currency’s (“OCC”) proposed rulemaking to implement the Guiding and Establishing National
Innovation for U.S. Stablecoins Act (“GENIUS Act”) through new 12 CFR Part 15 as one of the
few institutions globally that operates a digital asset infrastructure under federal banking
supervision, multiple state licenses, and international regulatory frameworks. BitGo (NYSE:
BTGO) is a publicly traded digital asset infrastructure company that provides custody, wallets,
staking, trading, financing, and settlement services to institutional and retail clients worldwide,
with over $100 billion in digital assets on its platform. BitGo operates BitGo Bank & Trust,
National Association (“BitGo BT”), a national trust bank chartered and supervised by the OCC,
and maintains regulated licenses across multiple U.S. states and foreign jurisdictions, including a
MiCA license in the European Union, a Major Payment Institution license in Singapore, and
virtual asset service provider approvals in Dubai. We offer the OCC a perspective informed by
daily operational experience at the intersection of traditional banking supervision and digital
asset technology. 1
As an institution actively preparing to seek approval as a permitted payment stablecoin issuer
under the GENIUS Act, BitGo has direct operational experience with the reserve management
infrastructure, custody architecture, and regulatory compliance framewo rks that the proposed
rule contemplates. We offer these comments from the perspective of a regulated national trust
bank that has built enterprise-grade systems for digital asset custody, real-time reserve
monitoring, and multi-jurisdictional regulatory reporting. Our goal is to assist the OCC in
adopting a final rule that achieves the GENIUS Act’s consumer protection and financial stability
objectives while preserving the ability of U.S.-domiciled, heavily regulated institutions to
compete effectively against offshore stablecoin products.
1

See 91 FR 10202 (March 2, 2026) (hereinafter, the“NPRM” or “proposed rule”).

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

II. Permitted Issuers Should Be Allowed to Issue Multiple Brands of Payment
Stablecoin Through a Single Legal Entity
The OCC has specifically requested comment on whether to restrict each permitted payment
stablecoin issuer to issuing only one brand of payment stablecoin. 2 BitGo respectfully urges the
OCC not to adopt a one-brand-per-issuer restriction and instead to permit issuers to issue
multiple brands through a single legal entity, subject to appropriate safeguards. The proposed
restriction would impose substantial costs on the stablecoin ecosystem without yielding
commensurate supervisory or consumer protection benefits.
A. Multi-Brand Issuance Is Essential to the Stablecoin-as-a-Service Model
Multi-brand issuance is central to the white-label stablecoin-as-a-service model that BitGo and
other infrastructure providers are developing for institutional partners. Under this model, a single
permitted issuer with robust compliance infrastructure, reserve management systems, and
regulatory reporting capabilities issues co-branded stablecoins on behalf of multiple distribution
partners. BitGo is actively developing stablecoin-as-a-service infrastructure that would enable
institutional partners—including fintech platforms, neobanks, cross-border payment providers,
and enterprise treasury operations—to offer co-branded stablecoins to their customers while
relying on BitGo’s compliance infrastructure, reserve management systems, and OCCsupervised regulatory relationships. Requiring each of these partners to establish a separate
OCC-supervised legal entity would make this service economically unviable for all but the
largest partners and would deprive the market of a model that efficiently concentrates
supervisory oversight in a single well-capitalized, well-supervised institution. This model
concentrates regulatory expertise, compliance technology, and supervisory relationsh ips in a
single well-supervised entity—precisely the outcome the GENIUS Act seeks to achieve.
The white-label model has deep precedent in the financial services industry. 3 Banks routinely
issue co-branded credit cards, private-label deposit products (such as brokerage sweep accounts
issued by Goldman Sachs on behalf of firms like Wealthfront, or certificates of deposit offered
through IntraFi’s network under individual bank branding), and white-label payment solutions
(such as Marqeta’s card-issuing platform powering branded cards for DoorDash and Square, or
Green Dot’s banking-as-a-service infrastructure underlying Walmart MoneyCard and similar
retail products) through a single regulated entity. The OCC has long recognized that this model
enhances consumer access while concentrating supervisory oversight in regulated institutions. A
one-brand-per-issuer restriction would represent a sharp departure from this regulatory tradition
without any clear basis in the GENIUS Act’s text, which neither mandates nor suggests such a
limitation.
2

NPRM at 10213 (discussing white -label arrangements and requesting comment on whether to restrict issuers to a
single brand).
3
See, e.g., JPMorgan Chase’s issuance of co-branded credit cards under the American Airlines, Amazon, and
Southwest Airlines brands through a single federally chartered bank; Goldman Sachs’s issuance of the Apple Card
under Apple’s brand; and Green Dot’s banking-as-a-service platform issuing the Walmart MoneyCard and similar
products.

2

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

B. A One-Brand Restriction Would Impose Costs Without Corresponding Benefits
Requiring a separate legal entity for each brand would compel the creation of multiple shell
subsidiaries, each requiring independent OCC approval, independent compliance infrastructure,
and independent capital—without any corresponding improvement in consumer protection or
financial stability. Each subsidiary would need its own BSA/AML program, its own board of
directors and governance structure, its own reserve management infrastructure, and its own
reporting systems. This would multiply the OCC’s supervisory workload without generating
additional supervisory insight, since each subsidiary would rely on the same underlying
operational platform.
The OCC itself acknowledges this concern by proposing to “streamline the process for approving
applications to become a permitted payment stablecoin issuer if an affiliate has already been
approved.”4 This acknowledgment suggests the OCC recognizes that duplicating legal entities
adds cost without proportionate supervisory benefit. The more efficient approach is to allow a
single issuer to operate multiple brands while imposing safeguards at the brand level, rather than
requiring structural proliferation.
C. Contagion and Run Risk Can Be Addressed Through Per-Brand Safeguards
The OCC has expressed concerns that multi-brand issuance “may also foster uncertainty about
reserve assets and encourage contagion and run risk among brands of payment stablecoins.”5
These concerns are fully addressable through targeted, brand-level safeguards rather than the
blunt instrument of structural separation.
Specifically, the final rule should require multi-brand issuers to maintain: (1) per-brand reserve
segregation, ensuring each brand’s reserves are identifiable and not commingled with reserves
backing other brands, consistent with proposed § 15.11(a)(1)(ii); (2) per-brand attestation
reporting, providing the OCC and the public with brand-level visibility into reserve adequacy;
(3) where appropriate, a legal opinion or third-party attestation confirming that each brand’s
reserve pool is held in a bankruptcy-remote structure—for example, through a segregated trust
arrangement or special purpose vehicle that ensures the reserve assets are not property of the
issuer’s general estate in the event of insolvency, a well-established technique in structured
finance and asset-backed securitization—for each brand’s reserve pool, ensuring that the
insolvency of one brand’s distribution partner does not impair the reserves backing another
brand; and (4) clear disclosure to stablecoin holders of the identity of the issuer and the specific
reserve pool backing their brand of stablecoin.
These safeguards would provide the OCC with full visibility into the reserve adequacy of each
brand while preserving the operational efficiency and regulatory consistency that come from
4

NPRM at 10213 (proposing to “streamline the process for approving applications” for affiliates of existing
issuers). See also 91 FR at 10213 (“[M]ultiple permitted payment stablecoin issuers could share certain services and
back-office functions with each other and might operate under a common risk management framework, but each
issuer would be legally separate.”).
5
See NPRM at 10213.

3

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

maintaining a single supervised issuer. To the extent the OCC is concerned that shared back office infrastructure could create operational contagion across brands, BitGo submits that this
concern is misplaced. Enterprise-grade digital asset platforms such as BitGo’s are architected
with logical segmentation by design: shared infrastructure components—including blockchain
nodes, hardware security modules, and treasury management systems—process brand-specific
operations through isolated execution paths, ensuring that a liquidity run or elevated redemption
activity on one white-label brand cannot operationally impair the minting, burning, or reserve
management of any other brand utilizing the same infrastructure. Indeed, per -brand reserve
segregation under a single issuer with logically segmented operations may provide greater
transparency and stronger operational resilience than the alternative of multiple affiliated entities
operating under a common risk management framework. The multi-entity alternative creates its
own risks: cross-entity exposures, shared services arrangements, and duplicated but
inconsistently maintained systems can produce opaque interdependencies and fragmented
security postures. These structures are often far more difficult for the OCC to supervise than a
single well-organized issuer with clear internal segmentation.
D. Insolvency Concerns Are Better Addressed by Existing Statutory Protections
The NPRM suggests that separate legal entities “may provide more certainty with respect to the
rights of payment stablecoin holders in the event that a permitted payment stablecoin issuer
becomes insolvent.” Stablecoin holders care about one thing: whether their reserves are protected
if something goes wrong. Per-brand reserve segregation with bankruptcy-remote structures
provides that protection regardless of how many brands share the same issuing entity. The
GENIUS Act already establishes comprehensive resolution procedures and claims priority for
stablecoin holders.6 The corporate form of the issuer is irrelevant to holder protection; what
matters is whether reserves are legally segregated and held for the exclusive benefit of holders.
Multiple affiliated shell entities with shared risk management frameworks do not pr ovide better
protection than a single well-supervised issuer with rigorously segregated per-brand reserves.

III. The Rebuttable Presumption for Interest and Yield Payments Requires
Explicit Safe Harbors and Procedural Protections (Response to Questions
Regarding § 15.10(c)(4))
BitGo supports the GENIUS Act’s prohibition on paying interest or yield to stablecoin holders
and agrees that the OCC should adopt regulations to prevent evasion. 7 However, the proposed
rebuttable presumption in § 15.10(c)(4)(i) would capture legitimate business arrangements that
do not constitute prohibited interest or yield payments. 8 For example, the OCC’s preamble
acknowledges that the prohibition is “not intended to prevent a merchant from independently

6

12 U.S.C. § 5911.

7

See 12 U.S.C. § 5903(a)(11).
NPRM at 10212 (discussing the proposed rebuttable presumption in § 15.10(c)(4)(i)).

8

4

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

offering a discount to a payment stablecoin holder,” but the regulatory text does not codify this
limitation. 9
A. Independently Funded Affiliate Programs Should Be Expressly Excluded
The proposed presumption would apply whenever an issuer has a contract with an affiliate or
related third party to pay interest or yield, and that party in turn pays interest or yield to holders
of the issuer’s stablecoin. This two-step test could inadvertently capture independently funded
affiliate loyalty programs, merchant discount arrangements, and promotional incentives that are
financed entirely from the affiliate’s or related third party’s own revenue—not from stablecoin
reserve income or issuer subsidies.
Such programs are commonplace in the payments industry and serve legitimate competitive and
consumer welfare purposes.10 Credit card rewards programs, merchant cash-back programs, and
fintech referral and sign-up incentives (such as PayPal’s and Venmo’s new-user cash bonuses, or
Robinhood’s stock reward for opening a funded account) all involve payments from affiliates or
commercial partners to customers holding or using a particular financial product. None of these
programs constitutes the payment of “interest or yield” on the underlying financial product, and
subjecting their stablecoin equivalents to a rebuttable presumption would place payment
stablecoins at a competitive disadvantage relative to traditional payment instruments.
In direct response to the OCC’s Question 36, which asks whether the OCC should “propose any
safe harbor for arrangements that the OCC believes do not violate the statutory prohibition, ”
BitGo respectfully submits that the final rule should codify explicit safe harbors for: (1) affiliate
loyalty or rewards programs that are independently funded from the affiliate ’s own revenue and
not subsidized by the issuer from reserve income; (2) merchant-funded discounts offered to
stablecoin holders as consideration for the use of the payment stablecoin in a commercial
transaction; (3) promotional incentives of limited duration that are funded by the issuer’s
operating budget (not reserves) and disclosed to the OCC in advance; (4) revenue-sharing
arrangements between an issuer and a white-label partner that do not result in direct payments to
stablecoin holders; and (5) de minimis arrangements—such as rewards, incentives, or
promotional payments that fall below a specified per-holder dollar threshold (e.g., $100 per
holder per year) or represent less than a specified percentage of the stablecoin ’s face value—that
do not have a meaningful economic impact on the holder’s return. De minimis arrangements of
this nature are ubiquitous in the payments industry and pose no credible risk of evasion of the
GENIUS Act’s interest prohibition; subjecting them to the rebuttable presumption and rebuttal
process would impose compliance costs wholly disproportionate to any regulatory benefit and
would render routine promotional activity commercially unviable for stablecoin is suers
competing against traditional payment instruments that face no comparable burden.

9

NPRM at 10212 (“The prohibition is not intended to prevent a merchant from independently offering a discount to
a payment stablecoin holder for using payment stablecoins.”).
10
See, e.g., American Express’s Membership Rewards program, Chase Ultimate Rewards, and similar
independently funded loyalty programs.

5

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

B. The Rebuttal Process Requires Defined Standards and Timelines
The proposed rule provides that an issuer may rebut the presumption by submitting written
materials that, “in the OCC’s judgment,” demonstrate that the arrangement is not prohibited. 11
This standard is insufficiently defined and creates unacceptable regulatory uncertainty. Without
clear parameters, issuers cannot reliably structure business arrangements or assess compliance
risk ex ante, which will chill legitimate commercial innovation in the stablecoin ecosystem.
The final rule should specify: (1) the evidentiary standard the issuer must meet to rebut the
presumption—for example, a preponderance-of-evidence standard demonstrating that the
affiliate’s payments are independently funded and not derived from reserve income; (2) a defined
timeline of no more than 30 days within which the OCC must act on a rebuttal submission,
consistent with the APA’s requirements for timely agency action—a 60-day or longer review
period would be commercially unreasonable, effectively paralyzing basic marketing campaigns,
partner onboarding, and ordinary-course business operations that require timely regulatory
certainty to launch; (3) a deemed-approved mechanism if the OCC does not act within the 30day period, which is imperative to provide issuers with the regulatory certainty necessary to
compete effectively and to incentivize timely OCC review; and (4) a right to appeal an adverse
determination through the OCC’s existing administrative process, including the opportunity to
present additional evidence. 12
C. The Presumption Should Not Extend to Arm’s-Length Commercial Relationships
The proposed definition of “related third party” is arguably broad enough to capture entities that
issue payment stablecoins under a white-label arrangement, even where those entities are
sophisticated commercial counterparties operating at arm ’s length. BitGo recommends that the
OCC clarify that the presumption does not apply where: (1) the arrangement is between
unaffiliated entities transacting on arm’s-length commercial terms; (2) the funds used to pay
rewards or incentives are derived entirely from the third party’s own business operations; and (3)
the issuer does not direct, control, or fund the third party’s payment of rewards or incentives to
stablecoin holders. This clarification would preserve the OCC’s ability to police genuinely
evasive arrangements while providing the regulatory clarity needed for legitimate commercial
relationships to flourish.
Under the proposed rule, a “related third party” is defined to include two categories of persons:
(1) any person paying interest or yield to payment stablecoin holders as a service on behalf of the
permitted payment stablecoin issuer, and (2) any person on whose behalf, or under whose
branding, the issuer issues payment stablecoins (i.e., white-label partners). The breadth of this
definition is concerning because it could sweep in arm’s-length commercial partners whose
independent business activities happen to benefit stablecoin holders. BitGo recommends that the
OCC narrow the definition to require a meaningful nexus between the issuer’s payments to the
11

NPRM at 10212 (“a permitted payment stablecoin issuer may rebut the presumption by submitting written
materials that, in the OCC ’s judgment, demonstrate that the contract, agreement, or other arrangement is not
prohibited”).
12
See generally Administrative Procedure Act, 5 U.S.C. § 555(e) (requiring prompt notice of denial of written
applications with a brief statement of grounds).

6

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

third party and the third party’s payments to stablecoin holders—for example, by requiring that
the issuer’s payments be the predominant funding source for the third party’s yield or interest
payments, or that the arrangement be specifically designed to channel reserve income to holders
through an intermediary.

IV. BitGo Supports Option A for Reserve Asset Diversification (Response to
Questions Regarding § 15.11(c))
The OCC has proposed two alternative approaches to reserve asset diversification: Option A, a
principles-based general requirement with an optional quantitative safe harbor; and Option B,
mandatory quantitative requirements for all issuers. 13 BitGo supports Option A as the approach
most consistent with the GENIUS Act’s directive that diversification standards be “tailored to
the business model and risk profile of permitted payment stablecoin issuers.”
Option A appropriately recognizes that issuers with different business models, risk profiles, and
reserve compositions may require different diversification strategies. A large issuer with
complex multi-chain operations may need to maintain diverse reserve pools across multiple
eligible financial institutions, while a smaller issuer with a simpler profile may reasonably
concentrate reserves to a greater degree. The safe harbor within Option A provides a useful
benchmark while preserving the flexibility the statute contemplates. However, BitGo strongly
urges the OCC to exempt Federal Reserve Banks and Global Systemically Important Banks (GSIBs) entirely from the 40% single-institution concentration limit within the safe harbor.
Applying the 40% cap to deposits held at Federal Reserve Banks or G-SIBs would force a largescale issuer—for example, one with $50 billion or more in outstanding issuance—to disperse
reserve assets across dozens of smaller regional banks and custodians, each of which presents
materially greater credit and operational risk than a Federal Reserve Bank or a G -SIB subject to
the most intensive prudential supervision in the world. Far from promoting safety and soundness,
this result would actively undermine it: the OCC would be compelling issuers to replace the
safest counterparties in the financial system with a fragmented web of less-capitalized, lesssupervised institutions, multiplying counterparty risk, operational complexity, and the very
contagion hazards the diversification requirement is intended to mitigate. There is no principled
basis for treating a deposit at a Federal Reserve Bank—which carries zero credit risk—
identically to a deposit at a community bank for concentration -limit purposes.
This analysis points to a second structural deficiency in percentage-based concentration limits:
by reducing the maximum loss from any single counterparty failure without ensuring that the
reduced loss falls below the issuer’s loss absorption capacity, the rule increases the mean time
between catastrophic events without preventing catastrophe when a failure does occur. A
diversification requirement calibrated to defer rather than eliminate the risk of par -breaking
reserve losses is not a meaningful safeguard from the stablecoin holder’s perspective. Holders
care whether they can redeem at par, not whether the issuer’s eventual insolvency was caused by
one large failure or several smaller ones.

13

NPRM at 10216–17 (discussing Option A and Option B for reserve asset diversification under § 15.11(c)).

7

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

The 40% concentration limit is also poorly calibrated from a loss-severity perspective. A
concentration cap reduces the frequency of catastrophic loss events but does not prevent
catastrophic outcomes when a counterparty failure does occur. Consider the widely reported
example of Circle Internet Financial, which held approximately $3.3 billion —roughly 8% of its
then-$40 billion reserves—at Silicon Valley Bank when SVB failed in March 2023. Even at this
relatively modest 8% concentration, the potential loss would have exceeded Circle’s
approximately $310 million in equity by a factor of ten. For Circle to have absorbed an SVB magnitude loss without insolvency, it would have needed to cap exposure to any single
institution at approximately $310 million—less than 1% of reserves—requiring diversification
across more than 80 counterparties.
This analysis reveals a fundamental flaw in percentage-based concentration limits: if the issuer’s
equity is small relative to total reserves, any fixed-percentage cap that is commercially
practicable (such as 40%) will still produce catastrophic, equity-exceeding losses when a
counterparty fails. The 40% limit merely increases the mean -time-between-catastrophic-failures
without preventing catastrophe when failure occurs. For an issuer with thin equity relative to
reserves, diversification across multiple non-G-SIB counterparties may perversely increase
aggregate counterparty risk by forcing reserves into institutions with weaker credit profiles—
while still leaving the issuer exposed to catastrophic loss from any single counterparty ’s failure.
BitGo therefore recommends that, in addition to exempting Federal Reserve Banks and G -SIBs
from the 40% cap, the OCC adopt a principles-based alternative under which issuers may
demonstrate that their concentration limits are calibrated to their balance -sheet capacity to absorb
counterparty losses without insolvency. An issuer that maintains equity or subordinated lossabsorption capacity sufficient to cover a counterparty failure at any given concentration level
should not be forced to disperse reserves across weaker institutions solely to satisfy an arbitrary
percentage threshold that does not account for the issuer’s actual loss-absorption capacity.
BitGo acknowledges the intuitive appeal of the counterargument that diversification across
multiple counterparties is still beneficial because losing 40% of reserves through a single
counterparty failure is preferable to losing 100%. This argument has surface validity but does not
withstand scrutiny in the context of the OCC’s stated regulatory objectives. First, the relevant
question is not whether a 40% loss is preferable to a 100% loss in the abstract, but whether
forcing issuers to disperse reserves across weaker counterparties to satisfy an arbitrary
concentration cap actually reduces aggregate risk. If the 40% cap forces an issuer to move
reserves from a Federal Reserve Bank or a G-SIB into regional banks with materially higher
failure probabilities, the expected loss from counterparty failure may increase even as the
maximum loss from any single failure decreases. The OCC’s framework should optimize for
minimizing expected loss, not for capping maximum loss at a level that remains catastrophic.
Second, a 40% reserve loss would in most cases still render the stablecoin unable to honor
redemptions at par, triggering the very confidence crisis and run dynamics that the diversification
requirement is intended to prevent. Stablecoin holders care about whether they can redeem at
par, not about whether the issuer’s insolvency was caused by a 40% loss or a 100% loss. A
concentration limit that still produces par-breaking losses is not a meaningful safeguard from the
holder’s perspective. Third, the GENIUS Act’s framework already provides robust protections
through the 1:1 reserve requirement, the restriction to high-quality liquid assets, and the
8

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

mandatory redemption obligations. Layering an additional concentration requirement that forces
reserves away from the safest counterparties in the financial system—Federal Reserve Banks and
G-SIBs—does not enhance these protections; it undermines them.
Option B’s one-size-fits-all mandatory requirements, by contrast, may impose unnecessary costs
on simpler issuers or prove insufficient for more complex ones. To illustrate: if Option B
mandated that every issuer diversify reserves across at least three eligible fi nancial institutions
with no more than 40% at any single institution, a small issuer with $500 million in reserves and
a straightforward single-chain operation would be forced to establish and maintain multiple
custodial relationships—each with its own due diligence, custody agreements, and ongoing
monitoring—that add operational complexity and cost without meaningfully improving reserve
safety for a simple portfolio of Treasury bills. At the same time, a large issuer with $50 billion in
reserves, complex multi-chain operations, and concentrated counterparty exposures could satisfy
the same percentage thresholds while still maintaining risk concentrations that the fixed
thresholds fail to address. Flat percentage caps also disproportionately harm large, well capitalized issuers in a manner that perversely increases systemic risk. As an issuer scales, a rigid
40% cap forces it to move tens of billions of dollars in reserve assets away from the safest
counterparties in the financial system—Federal Reserve Banks and G-SIBs subject to the most
intensive prudential supervision in the world—and deposit those assets into progressively
smaller, less-capitalized regional banks and custodians simply to satisfy a mathematical fraction.
The marginal counterparty added at the 11th or 15th institution is, by definition, weaker than the
Federal Reserve Bank or G-SIB the issuer was compelled to reduce exposure to—meaning the
diversification mandate, at scale, systematically replaces safer counterparties with riskier ones.
This is the opposite of sound prudential regulation. The OCC’s diversification framework must
therefore be tiered or calibrated by issuer size and actual loss-absorption capacity, rather than
applied as a blunt, one-size-fits-all percentage that treats a $500 million issuer and a $50 billion
issuer identically. A principles-based approach under Option A avoids both problems by
allowing the OCC to tailor diversification expectations to each issuer ’s actual risk profile.
BitGo also supports the proposed 20-day weighted average maturity limit within the safe harbor
as a reasonable constraint on interest rate risk. This limit is appropriately calibrated to ensure that
reserve assets remain highly liquid while permitting issuers to earn a modest return that supports
operational sustainability. BitGo recommends that the OCC confirm that the weighted average
maturity calculation be performed on a portfolio basis, allowing issuers to hold a mix of
maturities so long as the overall weighted average does not exceed the 20-day limit.

V. The Redemption Framework Requires Calibration to Avoid Unnecessary
Market Disruption (Response to Questions Regarding § 15.12)
The proposed § 15.12 requirement that permitted payment stablecoin issuers redeem stablecoins
at par value within two business days of a valid request reflects sound regulatory judgment. This
standard achieves the GENIUS Act’s objective of ensuring that stablecoins function as reliable
payment instruments. However, the proposed stress-extension mechanism and the associated
notification requirement require refinement to function effectively in practice without
introducing unnecessary operational risk or market disruption.
9

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

A. The Automatic 10% Stress-Extension Threshold Is Insufficiently Calibrated
The proposed rule automatically extends the redemption window from two business days to
seven calendar days whenever redemption requests exceed 10% of an issuer’s outstanding
issuance value in any rolling 24-hour period. As proposed, this extension is non-discretionary:
once the threshold is crossed, all pending and subsequent redemption requests are automatically
subject to the seven-day window, without any assessment of whether the issuer’s reserves are
sufficient to honor requests within the standard two-business-day timeline.
BitGo respectfully submits that a flat 10% threshold, applied uniformly regardless of the issuer ’s
size, reserve composition, or liquidity position, is not well-calibrated to the actual risk that the
stress-extension mechanism is designed to address. For a large issuer with $10 billion in
outstanding issuance, a $1 billion redemption request in a single 24-hour period would trigger
the automatic extension—even if the issuer holds $2 billion in immediately available cash across
multiple custodians and can demonstrably honor the full request within the standard two business-day window. Triggering a seven-day extension under these circumstances would harm
stablecoin holders who submitted valid redemption requests, undermine confidence in the
stablecoin’s reliability as a payment instrument, and potentially exacerbate rather than contain
market stress. The automatic, non-discretionary character of the extension is particularly
problematic: it removes the issuer’s ability to demonstrate that its specific reserve composition
and liquidity position are adequate to honor requests on the standard timeline.
As a threshold matter, the OCC should explicitly exempt from the automatic seven -day extension
any issuer that maintains 100% of its reserves in overnight or immediate demand liquidity —such
as balances held at a Federal Reserve Bank, demand deposits at eligible financial institutions, or
uninvested cash. An issuer in this position can, by definition, honor any volume of redemption
requests on the standard two-business-day timeline regardless of the percentage of outstanding
issuance those requests represent. Forcing such an issuer into a mandatory seven-day freeze
would not protect stablecoin holders—it would actively harm them by delaying redemptions the
issuer is fully capable of honoring, signaling to the market that the issuer is under stress when it
is not, and manufacturing precisely the confidence crisis and run dynamics the stress-extension
mechanism is designed to prevent. A mandatory freeze on a fully liquid issuer destroys the
stablecoin’s utility as a reliable payment instrument and punishes the most conservatively
managed issuers for the sins the OCC fears from less prudent ones. For issuers that do not
qualify for the full exemption, BitGo recommends that the final rule modify the stres s-extension
mechanism to incorporate a reserve-adequacy assessment: an issuer that has crossed the 10%
threshold should be permitted to notify the OCC within 24 hours that it can demonstrate, through
a reserve composition report and liquidity attestation, that it is capable of honoring all pending
requests within the two-business-day standard. The OCC should then have a defined period—for
example, 12 hours from receipt of the notification—to affirmatively require the seven-day
extension or to confirm that the issuer may proceed on the standard timeline. In the absence of an
OCC instruction, the issuer should be permitted to proceed under the standard timeline while
maintaining enhanced monitoring and reporting. This approach preserves the OCC’s ability to
impose the seven-day extension when warranted while avoiding unnecessary disruption to wellcapitalized issuers during periods of elevated (but manageable) redemption activity.

10

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

B. The Threshold Calculation Must Be Clarified for Multi-Brand Issuers
As BitGo has urged in Section II of this letter, the final rule should permit multi-brand issuance
under a single legal entity with per-brand reserve segregation. If the OCC adopts this approach,
as BitGo recommends, the rule must clarify how the 10% stress-extension threshold is calculated
in a multi-brand context. BitGo submits that where per-brand reserves are fully segregated
pursuant to the safeguards recommended in Section II, the 10% threshold must be calculated on a
per-brand basis with reference solely to that brand’s dedicated reserve pool and outstanding
issuance. Imposing a cross-brand freeze under these circumstances would be operationally
unjustified and harmful to innocent stablecoin holders. A redemption surge in “Brand A” should
not trigger a mandatory seven-day freeze for “Brand B” whose reserves are independently
segregated, fully funded, and wholly unaffected by the stress event. Such a cross-brand freeze
would punish innocent stablecoin holders, breach the issuer’s contractual redemption obligations
to Brand B holders, and perversely manufacture the very contagion the stress-extension
mechanism is designed to prevent—by signaling to the market that all brands issued by a single
entity are operationally interdependent when, by regulatory design, they are not. The final rule
must address this question expressly and confirm per-brand threshold calculation to avoid
examination uncertainty and to preserve the integrity of the multi -brand issuance framework.

VI. Weekly Confidential Reporting Requirements Are Technically
Unworkable (Response to § 15.14(h))
The OCC’s proposed requirement under § 15.14(h) that permitted payment stablecoin issuers
report weekly on the “100 persons that hold or trade the stablecoin the most” is fundamentally
incompatible with the architecture of permissionless blockchain networks and must be struck
from the final rule or, at a minimum, strictly limited in scope. Payment stablecoins issued on
permissionless blockchains—which is the predominant and commercially essential deployment
model—are freely transferable on secondary markets without the issuer’s knowledge,
involvement, or consent. The issuer has no contractual relationship with, no KYC data for, and
no technological means of identifying the beneficial owners behind the vast majority of on -chain
wallet addresses that hold or transact in its stablecoin. This requirement is unworkable for three
independent reasons.
First, identification of beneficial owners behind on-chain addresses is technologically impossible
under current permissionless blockchain architecture. A wallet address on a permissionless
blockchain such as Ethereum is a pseudonymous string of 40-plus hexadecimal characters—e.g.,
0x4E83362442B8d1bEc281594CEA3050c8EB01311C—that reveals nothing about the identity
of the person or entity that controls it. The issuer has no technological means of linking that
address to a beneficial owner unless the owner has established a direct customer relationship
with the issuer through KYC onboarding. No mechanism exists on any major permissionless
network to query or compel the disclosure of the identity behind a wallet address, and the issuer
cannot manufacture information that the blockchain’s design does not produce.
Second, the only alternative—third-party blockchain analytics—produces probabilistic
attributions that are neither reliable nor comprehensive enough to serve as a basis for supervisory
11

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

reporting. Such vendors can offer heuristic assessments (e.g., “this cluster of addresses may be
associated with Exchange X based on observed transaction patterns”), but no vendor claims these
attributions are definitive. Many of the largest holders employ multiple addresses, cold storage
wallets, smart contract wallets, or privacy-preserving techniques that defeat attribution entirely,
meaning even probabilistic analytics cannot produce a complete or accurate picture of beneficial
ownership.
Third, requiring issuers to certify weekly reports based on speculative, probabilistic analytics
would create an untenable compliance standard that exposes issuers to unwarranted enforcement
risk. Issuers would be compelled to submit reports they know to contain unverified and
potentially inaccurate data, exposing them to enforcement action for inevitable attribution errors
that are not the product of negligence but of the fundamental design of permissionless networks.
The resulting supervisory data would be affirmatively misleading—providing the OCC with a
false sense of visibility into holder concentrations while obscuring the actual limitations of the
underlying data—and would impose compliance costs wholly disproportionate to any
conceivable supervisory benefit.
BitGo urges the OCC to limit this reporting obligation exclusively to direct, known customers of
the issuer who have been officially onboarded through the issuer’s KYC/BSA-AML processes—
persons whose identity the issuer actually knows and can reliably report. Extending the
obligation to secondary market participants whose identities are unknown to the issuer is not
merely burdensome; it is a technologically unfeasible mandate that would result in speculative
and misleading supervisory data, expose issuers to enforcement risk for inevitable reporting
failures, and deter responsible institutions from issuing stablecoins under the federal framework.

VII. Transitioning State-Chartered Issuers Must Be Subject to Rigorous PreTransition Standards (Response to Questions Regarding § 15.15)
Under the GENIUS Act, State qualified payment stablecoin issuers with outstanding issuance
exceeding $10 billion must transition to the federal regulatory framework. 14 This transition
requirement is essential to ensuring consistent federal oversight of systemically significant
stablecoin issuers. However, the final rule must ensure that transitioning state-chartered issuers
do not enjoy a competitive advantage over institutions like BitGo that have operated under
rigorous OCC supervision from inception.15
A. The Risk of Regulatory Arbitrage During the Pre-Transition Period
The transition framework creates a potential window for regulatory arbitrage. A state -chartered
issuer can grow to $10 billion in outstanding issuance under a potentially less rigorous state
supervisory regime before triggering the federal transition requirement. During this growth
period, the issuer may have operated with less stringent capital requirements, less comprehensive
BSA/AML programs, and less robust operational resilience standards than its federally chartered
competitors. Granting such an issuer automatic equivalence upon transition would reward
14
15

See 12 U.S.C. § 5903(d)(1) and (2) (transition requirements for state-chartered issuers exceeding $10 billion).
See proposed § 15.15 (transitioning state-chartered issuers).

12

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

regulatory arbitrage and penalize institutions that chose the more demanding federal path from
the outset.
B. Comprehensive Pre-Transition Examinations Are Essential
BitGo urges the OCC to require that transitioning state-chartered issuers undergo comprehensive
pre-transition examinations covering capital adequacy, liquidity management, operational
resilience, and BSA/AML compliance before they are permitted to operate under the federal
framework. A transitioning issuer that has operated under less rigorous state oversight should not
be granted automatic equivalence with institutions that have invested substantial resources in
meeting OCC standards from day one.
The OCC should also require that transitioning issuers meet the same capital, liquidity, and
operational requirements applicable to new federal applicants, with a defined remediation period
(e.g., 180 days) if shortfalls are identified during the pre-transition examination. This
remediation period should include enhanced supervisory monitoring and restrictions on growth
until the issuer demonstrates full compliance with OCC standards. The transition framework
should not create a “fast track” to federal status that bypasses the rigorous application and
examination process that new federal applicants must complete.
C. The Proposed Six-Month Examination Grace Period Must Be Eliminated.
The OCC’s proposed framework contemplates that the initial examination of a transitioning
issuer would occur no later than six months after the transition notification—meaning a statechartered issuer could operate under federal status, fully unexamined by the OCC, for up to six
months. This is unacceptable. The grace period must be collapsed to zero: the OCC ’s
comprehensive examination must conclude before or concurrently with the transition date, not
months afterward. Permitting state-chartered competitors to hold themselves out as federally
supervised institutions—enjoying the full reputational and market benefits of OCC oversight—
while remaining entirely unexamined by the OCC for half a year fundamentally undermines
competitive equity in the federal stablecoin framework. It creates a massive competitive
disadvantage for institutions like BitGo that have voluntarily subjected themselves to rigorous
OCC supervision from inception, invested heavily in building federal-grade compliance
infrastructure from day one, and undergone thorough examination as a precondition to
commencing operations. Competitive equity demands that transitioning issuers demonstrate to
the OCC that they satisfy the exact same standards that native federal institutions are required to
meet before they are permitted to operate—not six months after they have already begun
competing in the market under the federal imprimatur. The OCC should not permit the transition
framework to create a two-tier system in which state-chartered issuers enjoy the market
credibility of federal status without first earning it through the same rigorous examination
process that institutions like BitGo completed before commencing operations.
VIII. The OCC Should Subject Applications for Captive Self-Custody Charters to
Heightened Scrutiny (Response to Question 168 and Questions Regarding § 15.30)

13

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

The OCC should carefully scrutinize, and in appropriate cases decline, applications for national
trust bank charters where the applicant’s sole or predominant purpose is to custody stablecoin
reserves for an affiliated issuer rather than to provide custody services to unaffiliated clients in
the broader market.
A. Captive Charters Represent Regulatory Arbitrage Without Corresponding Public Benefit
A national trust bank charter is a grant of federal supervisory standing that carries significant
reputational and operational advantages: access to the federal payments system, recognition as a
regulated entity under applicable law, and the imprimatur of OCC supervision. These benefits
exist to facilitate the provision of financial services to the public —not to serve as a costreduction mechanism for affiliated stablecoin issuers seeking to eliminate third -party custody
expenses.
An applicant whose charter application contemplates serving only a single affiliated issuer ’s
reserve custody needs has not entered the financial services market in any meaningful sense. It
has not subjected itself to competitive pricing discipline, arm’s-length counterparty due
diligence, or the market accountability that comes from serving clients whose interests are
independent of the institution’s own. Such an entity imposes supervisory costs on the OCC—
regular examinations, ongoing reporting obligations, BSA/AML compliance oversight—without
generating the public benefits that justify the allocation of those resources.
The OCC’s supervisory capacity is finite. Every examination cycle devoted to a captive single purpose vehicle custodying reserves exclusively for its own affiliate is an examination cycle that
cannot be allocated to institutions providing custody, settlement, and financial services to the
broader public. The OCC should be explicit in the final rule and accompanying guidance that
captive self-custody structures are not the intended beneficiaries of the national trust bank charter
framework and will receive heightened scrutiny accordingly.
B. Self-Custody Does Not Achieve Meaningful Separation of Issuance and Custody Functions
One of the core structural objectives of a robust stablecoin regulatory framework is the
meaningful separation of the issuance function from the custody function. When reserves are
held at an independent, regulated custodian with its own fiduciary obligati ons, its own
supervisory relationship with the OCC, and its own financial interest in maintaining the integrity
of the custodial relationship, stablecoin holders benefit from a genuine check on the issuer ’s
behavior. The custodian’s institutional independence provides a layer of protection that is absent
when the issuer and its captive custody vehicle operate as a single integrated enterprise.
An issuer that establishes a captive trust bank for the sole purpose of holding its own reserves
has not achieved meaningful separation—it has recharacterized a single integrated operation as
two affiliated entities for regulatory purposes while retaining complete operational and economic
control over both. The custody function in such a structure is not independent; it is a wholly owned extension of the issuer’s own operations, dressed in the regulatory clothing of a national
bank charter.

14

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

C. Captive Charters Introduce Unacceptable Conflicts of Interest and Contagion Risk
Captive self-custody structures present a fundamental conflict of interest that is irreconcilable
with the safety and soundness objectives of the GENIUS Act. When an issuer establishes an
affiliated trust bank for the sole purpose of custodying its own reserves, the custodian’s fiduciary
obligations to stablecoin holders are structurally compromised by the custodian ’s economic
dependence on, and organizational subordination to, the issuer. An independent, third -party
custodian owes legally enforceable fiduciary duties to the beneficial owners of custodied
assets—duties of loyalty, care, and segregation that operate as a critical check against issuer
misconduct, commingling of reserves, or reckless reserve allocation. A captive custodian, by
contrast, has neither the institutional independence nor the economic incentive to challenge the
directives of its parent issuer, even when those directives conflict with the interests of stablecoin
holders. BitGo respectfully submits that this structural conflict of interest is incompatible with
the grant of a national bank charter.
The GENIUS Act’s framework contemplates a meaningful separation between the issuance
function and the reserve custody function—a structural safeguard designed to ensure that
stablecoin holders’ reserves are protected by an independent fiduciary whose interests are
aligned with holders, not with the issuer. A captive self-custody structure wholly defeats this
statutory intent. If the parent issuer faces financial distress, operational failure, or regulatory
action, the captive custodian—dependent on the same management, the same capital structure,
and the same operational infrastructure—is exposed to identical contagion risk. The custodian
cannot serve as an independent bulwark against the issuer’s distress when the custodian is itself a
subsidiary of the distressed entity. This contagion risk is not theoretical: in every significant
financial services failure involving affiliated custody arrangements, the custody function has
been compromised precisely because the custodian lacked the institutional independence to resist
the parent entity’s demands for access to, or recharacterization of, custodied assets. An
independent, regulated custodian with its own supervisory relationship, its own capital base, and
its own fiduciary obligations provides the structural separation that stablecoin holde rs require—
and that the GENIUS Act demands.
Multiple well-capitalized, OCC-supervised national trust banks already exist and are prepared to
provide stablecoin reserve custody services to issuers of all sizes. The market for compliant,
regulated, independent custody is competitive and accessible. Th ere is no regulatory gap that
captive self-custody structures are needed to fill. What these structures eliminate is not a market
failure but the independent fiduciary check that protects stablecoin holders —replacing it with an
organizational arrangement in which the entity responsible for safeguarding reserves is
economically and operationally controlled by the entity whose conduct the custodian is supposed
to monitor. The OCC should not grant national bank charters to facilitate this outcome.
D. Concrete Regulatory Recommendations
BitGo urges the OCC to address captive self-custody structures through the following measures
in the final rule and accompanying supervisory guidance:

15

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

First, the OCC should require any applicant for a national trust bank charter whose stated
business plan includes custody of stablecoin reserves for an affiliated issuer to affirmatively
demonstrate, as a condition of charter approval, a bona fide intentio n to provide custody services
to unaffiliated clients. A credible demonstration would include specific evidence of unaffiliated
client commitments, a business development plan identifying target unaffiliated clients, and
financial projections showing meaningful unaffiliated revenue within a defined period not to
exceed 24 months following charter approval.
Second, the OCC should impose enhanced minimum capital requirements on national trust banks
that, at any point following charter approval, derive more than 50% of their custody revenues
from a single affiliated issuer. These enhanced requirements should reflect the heightened
concentration risk, reduced market accountability, and potential for operational underinvestment
that characterize single-purpose captive structures.
Third, the OCC should impose examination fee surcharges on captive institutions that fail to
demonstrate meaningful service to unaffiliated clients within the timeframes specified at charter
approval, reflecting the supervisory burden of overseeing institutions that provide no
corresponding public benefit and have no external market accountability to supplement the
OCC’s supervisory oversight.
Fourth, the OCC should require captive institutions to disclose annually the percentage of their
custody revenues derived from affiliated versus unaffiliated clients, with this information made
available to the public as part of the OCC’s commitment to transparency in the national bank
charter framework.

IX. Custody Rules Must Accommodate Cryptographic Sub -Ledgering Within
Omnibus Accounts (Response to Question 158)
BitGo strongly supports the OCC’s proposal in § 15.22(b) to permit covered custodians to utilize
omnibus accounts for covered assets.16 We urge the OCC to go further in the final rule and
explicitly recognize that cryptographically segregated omnibus accounts satisfy the segregation
and separate accounting requirements of §§ 15.21 and 15.22. 17
A. The OCC Should Explicitly Recognize Cryptographic Sub-Ledgering Under § 15.22(b)
Modern digital asset custody relies on mathematically verifiable on -chain segregation through
omnibus wallets tied to strict internal ledger systems that track beneficial ownership with
cryptographic precision.18 In practical terms, this means that even though multiple customers ’
assets may be held in the same blockchain address (the “omnibus wallet”), the custodian
maintains real-time internal records—continuously reconciled against the blockchain itself—
showing exactly which customer owns which portion of the assets at every moment. This is not a
matter of accounting convention; it is mathematically verifiable by any party with access to the
16

See proposed § 15.22(b) (omnibus accounts for covered custodians).
See 12 U.S.C. § 5909(b), (c), and (d) (custody requirements under the GENIUS Act).
18
NPRM Question 158 (requesting comment on custody account structures).
17

16

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

distributed ledger. This architecture allows custodians to maintain verifiable segregation while
optimizing operational efficiency, security, and cost management. Without explicit regulatory
recognition that cryptographic sub-ledgering satisfies § 15.22(b), examination teams may
interpret the rule to require account structures that are operationally incompatible with digital
asset technology.
The distinction between traditional and digital asset custody is fundamental. In traditional
finance, segregation is achieved through separate account numbers in a centralized ledger
maintained by a single institution. In digital asset custody, segregation can be achieved through
cryptographic proofs that are mathematically verifiable by any party with access to the
distributed ledger—providing a higher level of auditability and transparency than traditional
methods. The OCC’s rule should recognize this technological advancement rather than requiring
custodians to replicate the limitations of legacy systems.
B. Fragmented 1:1 Wallet Requirements Would Impose Disproportionate Costs
If the custody framework were interpreted to require separate on-chain wallets for every
customer or reserve pool, it would impose significant operational burdens that far exceed any
corresponding consumer protection benefit. Maintaining thousands of separate on-chain wallets
creates three critical problems. First, network transaction costs become commercially
prohibitive—every deposit, withdrawal, or rebalancing operation requires separate on-chain
transactions, multiplying network fees substantially. On high-fee networks, these costs could
exceed the economic value of smaller custody relationships, effectively excluding smaller
institutions from the stablecoin ecosystem. Second, liquidity becomes fragmented across separate
wallet balances for each customer, creating artificial constraints that degrade service quality and
increase settlement times. Third, the security surface area expands significantly: instead of
securing a small number of omnibus wallets with enterprise-grade multi-signature controls and
hardware security modules, custodians would manage thousands of separate private keys,
increasing key management complexity and attack vectors.
C. Cryptographic Segregation Provides Strong Customer Protection
In BitGo’s omnibus custody architecture, customer assets are segregated through immutable
blockchain ledgers providing cryptographic proof of ownership and transfer history, real-time
internal sub-ledgers tracking beneficial ownership with precision and continuously reconciling
against on-chain positions, multi-signature wallet controls requiring cryptographic authorization
from multiple independent key holders, and automated compliance monitoring that flags any
deviation from expected customer balances. This architecture provides transparency, auditability,
and protection that meets or exceeds what legacy banking systems achieve through separate
paper accounts and manual reconciliation.
D. The OCC Should Adopt Principles-Based Segregation Standards
Rather than prescribing specific account structures, the OCC should establish principles -based
segregation requirements permitting custodians to demonstrate compliance through any method
achieving the regulatory objectives of customer protection and asset segregation. The OCC
17

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

should also confirm that covered custodians may hold covered assets (such as payment
stablecoins) and non-covered digital assets (such as other cryptocurrencies) within the same
omnibus wallet structures, provided that internal cryptographic sub-ledgering mathematically
segregates each asset class and each customer’s beneficial ownership at all times. This flexibility
is essential for institutional custodians like BitGo that provide integrated trading, settlement, and
custody services across multiple digital asset classes—requiring separate on-chain wallets for
covered and non-covered assets would fragment liquidity, increase transaction costs, and degrade
settlement efficiency without any corresponding improvement in customer protection, since the
same cryptographic sub-ledgering technology that segregates one customer’s assets from
another’s is equally capable of segregating covered from non-covered assets within a single
omnibus structure. The final rule should confirm that custodians satisfy segregation requirements
if they can demonstrate: (1) verifiable beneficial ownership tracking thro ugh auditable ledger
systems; (2) immediate identification and remediation of any commingling; (3) mathematical
proof that customer assets are segregated from custodian proprietary assets; and (4) real-time
reporting of customer balances and custodian positions. This technology-neutral approach would
allow custodians to utilize omnibus wallet structures that optimize security and efficiency while
achieving the OCC’s consumer protection objectives, and would ensure that the regulatory
framework remains adaptable as custody technology continues to evolve.
E. National Trust Banks Must Be Recognized as Eligible Financial Institutions Without
Qualification
BitGo has invested substantial resources in building enterprise-grade custody infrastructure
precisely because the national trust bank charter provides the regulatory framework appropriate
for fiduciary custody of digital assets. Permitting incumbent depository banks to exclude trust
bank competitors from the reserve custody market would undermine the competitive neutrality
that the GENIUS Act’s framework contemplates and would harm stablecoin issuers by reducing
their options for compliant, regulated custody.
Any effort to narrow the definition of eligible financial institution to exclude non -depository
national trust banks—whether through rulemaking, interpretive guidance, or supervisory
practice—would contravene the plain text of the statute, reduce competiti on in the reserve
custody market, and concentrate reserve custody in a small number of large depository
institutions that may lack the specialized digital asset custody infrastructure that national trust
banks like BitGo have developed. The OCC should affirmatively reject any such narrowing and
confirm that national trust banks are eligible to custody stablecoin reserves on the same terms as
depository institutions, subject to the same prudential requirements.
BitGo urges the OCC to confirm explicitly in the final rule that national trust banks chartered
under 12 U.S.C. § 27(a), including uninsured national trust banks that do not accept demand
deposits or extend commercial credit, qualify as “eligible financial institutions” for purposes of
holding stablecoin reserves under the GENIUS Act and Part 15. The GENIUS Act’s definition of
eligible financial institution encompasses federally supervised banking organizations, and
national trust banks are unambiguously within this category.
F. Reserve Custody Should Require Fiduciary Duties Regardless of Custodian Charter Type
18

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

The proposed rule creates a fundamental asymmetry in the legal treatment of stablecoin reserves
based on the type of custodian the issuer selects. When reserves are held at a national trust bank,
they are held in a fiduciary capacity, legally segregated fr om the custodian’s proprietary assets
and held for the exclusive benefit of stablecoin holders. When reserves are held as ordinary
deposits at a depository bank, by contrast, those funds become general liabilities of the bank —
assets of the bank’s estate available to satisfy all creditors in an insolvency, not a segregated pool
held in trust.
This distinction creates a competitive asymmetry that disadvantages trust banks operating under
heightened fiduciary standards. It also provides weaker protection for stablecoin holders, whose
interests are best served when reserves are held in a legally segregated, fiduciary capacity rather
than as general obligations of a credit-extending institution. The practical consequences of this
distinction are stark. When stablecoin reserves are held as general deposits at a depository bank,
those funds become assets of the bank’s estate in the event of the bank’s insolvency, subject to
the FDIC’s receivership and claims-priority framework. The stablecoin issuer (and through it,
stablecoin holders) would stand in line with other general creditors of the failed bank, potentially
recovering cents on the dollar over a multi-year receivership process. By contrast, when reserves
are held in a fiduciary capacity at a national trust bank, fiduciary assets are not property of the
trust bank’s estate in insolvency. Instead, they pass through to beneficial owners outside the
receivership process entirely, because they were never the trust bank’s assets to begin with. This
is not a marginal difference in legal treatment. It is the difference between full recovery and
potential total loss for stablecoin holders. The OCC should level the playing field by requiring
that any institution custodying stablecoin reserves—whether a trust bank or a depository bank—
do so in a fiduciary capacity, with reserves legally segregated and held for the exclusive benefit
of stablecoin holders. This requirement would ensure that stablecoin holders receive consistent
protection regardless of the issuer’s choice of custodian, while eliminating the artificial
advantage that depository banks currently enjoy by offering non-fiduciary deposit products that
provide weaker holder protections.
By contrast, when stablecoin reserves are held as ordinary deposits at an FDIC-insured
depository bank, the depositor (the stablecoin issuer) is a general creditor of the bank, not the
beneficiary of a fiduciary relationship. The deposit represents a liabi lity of the bank, not a
segregated asset held in trust. In the event of the depository bank’s failure, the depositor’s claim
is subject to the FDIC’s receivership and claims-priority framework, and any amount exceeding
FDIC insurance limits is at risk of loss.
As a national trust bank, BitGo BT holds client assets in a fiduciary capacity: reserves are legally
segregated from BitGo’s proprietary assets, BitGo owes enforceable fiduciary duties to the
beneficial owners of those reserves, and in the event of BitGo’s insolvency, fiduciary assets are
not available to satisfy BitGo’s general creditors. These protections arise automatically from
BitGo’s trust bank charter and the fiduciary relationship it establishes with its clients.
BitGo recommends that the OCC require all covered custodians holding stablecoin reserves —
regardless of whether the custodian is a national trust bank, a state-chartered trust company, or
an FDIC-insured depository institution—to hold such reserves in a fiduciary capacity, subject to
19

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

the duties of loyalty, care, and segregation that apply to fiduciary assets under applicable law.
G. Cryptographic Sub-Ledgering Satisfies Sub-Custodian Oversight Requirements
Proposed § 15.21(b) permits covered custodians to use sub-custodians, subject to the requirement
that the custodian maintain adequate safeguards and internal controls over sub -custodial
arrangements. BitGo strongly supports this flexibility and urges the OCC to confirm in the final
rule that cryptographic sub-ledgering (the mathematically verifiable tracking of beneficial
ownership through on-chain proofs and real-time internal ledger reconciliation) inherently
satisfies the adequate safeguards and internal controls standard for sub-custodian oversight under
§ 15.21(b). BitGo’s institutional custody architecture routinely involves multi-tier custodial
arrangements in which BitGo serves as the primary custodian for institutional clients while
utilizing sub-custodial relationships for specific operational functions such as cold storage
distribution, geographic redundancy, and blockchain-specific key management. In each of these
arrangements, BitGo’s cryptographic sub-ledgering system maintains continuous,
mathematically verifiable proof of the location, ownership, and segregation status of every
custodied asset across every sub-custodian in the chain—a level of transparency and auditability
that far exceeds what traditional manual compliance reviews and periodic reconciliation
processes can achieve. The OCC should not interpret § 15.21(b) to require duplicative manual
compliance checks—such as paper-based reconciliation reports, periodic manual audits, or
redundant human-review processes—where cryptographic sub-ledgering already provides
continuous, real-time, mathematically verifiable oversight that is superior to any manual
alternative. Imposing duplicative manual controls on top of cryptographic ve rification would add
cost without any corresponding improvement in asset safety or customer protection, and would
disadvantage digital-native custodians like BitGo whose technology inherently provides the
oversight that traditional custodians must achieve through labor-intensive manual processes.
BitGo urges the OCC to adopt maximum flexibility in its interpretation of sub-custodial
oversight requirements under § 15.21(b), and to confirm that covered custodians may satisfy the
adequate safeguards standard through any combination of cryptographic, technological, and
procedural controls that provides verifiable asset segregation, real-time beneficial ownership
tracking, and immediate identification of any deviation from expected custodial positions.
Overly prescriptive sub-custodial requirements risk inadvertently disrupting BitGo’s B2B
institutional custody operations, which serve financial institutions, asset managers, and enterprise
clients that depend on multi-tier custodial arrangements for operational efficiency and security,
without any corresponding benefit to the safety or soundness of the stablecoin ecosystem.

X. The Capital Election for Uninsured National Trust Banks Should Be
Adopted and Clarified (Response to Questions Regarding § 15.41)
The OCC’s proposal to permit uninsured national trust banks to elect to comply with the Part 15
minimum capital and backstop requirements in lieu of Part 3 is the right approach. 19 This
election is essential for enabling existing national trust banks to participate in the stablecoin
market on a level playing field with non-bank permitted payment stablecoin issuers that would
not be subject to Part 3.
19

See proposed § 15.41 (capital election for uninsured national trust banks).

20

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

A. Part 3 Capital Requirements Create an Unjustified Competitive Asymmetry
Under current Part 3 requirements, the capital a national trust bank must hold may be
significantly in excess of what would be required for other types of permitted payment stablecoin
issuers under the proposal. This asymmetry would disadvantage the most heavily supervised
institutions in the stablecoin market and create a presumably unintended incentive for institutions
to seek stablecoin charters outside the national bank framework—precisely the opposite of the
GENIUS Act’s objective of bringing stablecoin issuance under robust federal supervision.
The Part 3 framework was designed for institutions whose primary activity is extending credit
and accepting deposits—activities that generate the credit risk, interest rate risk, and leverage
risk that capital requirements are designed to mitigate. A payme nt stablecoin issuer’s core
activity is maintaining a 1:1 reserve of high-quality liquid assets against outstanding stablecoin
liabilities—a fundamentally different risk profile that does not generate the same categories of
risk. Applying bank-centric capital requirements to this activity would require stablecoin issuers
to hold capital against risks they do not assume, reducing the capital available for operational
resilience and technology investment.
B. The Bifurcation Alternative Is Operationally Impracticable
The OCC’s alternative approach of bifurcating stablecoin operations from other trust bank
operations for capital calculation purposes would be impractical to operationalize, as the OCC
itself acknowledges in the NPRM.20 BitGo’s stablecoin issuance activities share unified
blockchain node infrastructure, consolidated treasury operations, integrated hardware security
module (HSM) key management systems, and enterprise-wide cybersecurity controls with our
custody, trading, and settlement services. The same hardware security modules that would secure
private keys for stablecoin minting and burning also secure keys for custody client wallets,
staking operations, and trading settlement. These HSMs are distributed across multiple
geographically separated data centers operating under unified key-management protocols and
firmware. The same cybersecurity operations center monitors threats to all activities. The same
blockchain node infrastructure validates transactions across all service l ines. There is no
operationally meaningful way to segregate “stablecoin HSM capacity” from “custody HSM
capacity”—they are the same machines running the same software, and any attempt to artificially
divide them would require duplicating enterprise-grade security infrastructure at enormous
expense while actually fragmenting security protocols and weakening protection .
C. The Election Should Be Available Through a Streamlined Process
We request that the final rule: (a) confirm that the Part 15 capital election is available to all
uninsured national trust banks from the effective date of the rule, through a streamlined 30 -day
notice process with defined and limited “good cause” objection criteria; (b) decline to adopt the
bifurcation alternative; (c) specify that the OCC will provide written confirmation of the
election’s effectiveness within the 30-day period, or the election will be deemed effective absent
20

NPRM at 10244 (acknowledging the impracticality of bifurcation).

21

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

a written objection; (d) explicitly grandfather any existing regulatory capital instruments issued
by uninsured national trust banks that elect to comply with the Part 15 framework, so that
electing institutions are not forced to undertake costly and unnecessary restructurings of their
existing capital stacks solely as a consequence of the election; and (e) reduce the proposed 12 month operational backstop requirement to a maximum of six months of projected operating
expenses—a timeframe the OCC itself acknowledges falls within its standard range—and
explicitly permit the backstop to be held in yield-bearing U.S. Treasury bills rather than
uninvested cash, recognizing that short-dated Treasury bills are among the most liquid
instruments in the world and that requiring issuers to trap substantial working capital in nonyielding accounts imposes a material and unjustified drag on operational sustainability. This
approach provides regulatory certainty while preserving the OCC’s ability to object in cases
where legitimate safety and soundness concerns exist.
D. FDIC Insurance Provides No Meaningful Protection for Stablecoin Reserves
As detailed in Section IX.F, FDIC deposit insurance provides no meaningful protection for
stablecoin reserves at any commercially significant scale—coverage of $250,000 per depositor
represents a fraction of a percent of reserves for any issuer above $100 million. The critical
implication for the OCC’s capital framework is that the presence or absence of FDIC insurance
is not a meaningful differentiator of risk for stablecoin reserve custodians, and should not serve
as a basis for imposing compensating capital requirements or additional supervisory burdens on
uninsured national trust banks.
BitGo BT is ineligible for FDIC deposit insurance not because we declined coverage, but
because we do not engage in the credit-extending and deposit-taking activities that deposit
insurance is designed to protect against. National trust banks that custody assets in a fiduciary
capacity (as opposed to lending out depositor funds and relying on fractional reserves) present a
fundamentally different risk profile than commercial banks. We hold client assets in segregated,
fiduciary accounts; we do not lend them out; and we do not create the asset-liability mismatch
that makes deposit insurance necessary for credit-extending institutions. The absence of FDIC
insurance is a feature of BitGo’s safer fiduciary model, not a deficiency—and any capital
framework that treats it as a deficiency requiring compensating add-ons would perversely
penalize the safest custodial structure available for stablecoin reserves.
The OCC’s capital and supervisory framework for stablecoin issuers must reflect these realities.
An uninsured national trust bank that holds reserves in a fiduciary capacity, with full segregation
and no credit extension, provides stablecoin holders with protections that are superior to those
provided by an FDIC-insured depository bank holding the same reserves as general deposits
subject to the bank’s credit risk. Imposing additional capital requirements on national trust banks
solely because they lack FDIC insurance—insurance that, as demonstrated in Section IX.F,
provides negligible protection at scale—would create an unjustified competitive asymmetry that
rewards institutions whose business models generate the very asset-liability mismatch risks that
stablecoin holders should be protected from.
BitGo urges the OCC to confirm in the final rule that the absence of FDIC insurance for national
trust banks is not a basis for heightened capital requirements, enhanced supervisory burdens, or
22

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

any other compensating regulatory obligation. The OCC’s capital framework should evaluate
custodian safety based on actual risk profile—fiduciary segregation, absence of credit extension,
and lack of asset-liability mismatch—not on whether the institution participates in a deposit
insurance program designed for an entirely different business model.
E. The 12-Month Operational Backstop Is Capital Inefficient and Must Be Reduced to Six
Months
The OCC’s proposed requirement that permitted payment stablecoin issuers maintain a 12 -month
operational backstop—equivalent to a full year of projected operating expenses held in non yielding cash or cash equivalents—is without precedent in financial services regulation and
imposes a capital burden that is wholly disproportionate to the regulatory objective it is designed
to achieve. BitGo urges the OCC to reduce the backstop requirement to a maximum of six
months of projected operating expenses and to explicitly permit that backstop to be held in shortdated U.S. Treasury bills.
The stated purpose of the operational backstop is to ensure that a permitted payment stablecoin
issuer has sufficient resources to execute an orderly wind-down, restructuring, or operational
transfer in the event the issuer ceases operations. Six months is the established industry standard
for this purpose. Technology and digital asset infrastructure companies—including regulated
digital asset custodians, fintech payment platforms, and national trust banks —routinely plan
operational continuity and wind-down scenarios around a six-month runway. This standard
reflects the time actually required to execute an orderly transfer of customer accounts and digital
assets to a successor custodian, comply with applicable notification and regulatory approval
requirements, wind down or transfer contractual obligations to distribution partners, and
complete required regulatory reporting and final examinations. A 12-month backstop is not
calibrated to the time required for an orderly wind-down of a payment stablecoin issuer—it is
simply twice the operationally necessary amount, with no demonstrated regulatory basis for the
doubling. The OCC’s own NPRM acknowledges that a six-month backstop falls within the range
the OCC has considered, which confirms that six months is a reasonable and well-grounded
benchmark. The OCC offers no evidence that six months would be insufficient to protect
stablecoin holders in a wind-down scenario, nor does it articulate any wind-down activity that
would require the additional six months of capital that the 12-month requirement mandates.
The capital efficiency costs of the 12-month requirement are severe and fall disproportionately
on federal issuers—and are compounded by a fundamentally flawed assumption in the OCC’s
own Economic Analysis. The NPRM estimates the operational backstop at “0.40% of
outstanding stablecoins,” implicitly assuming that operational expenses scale linearly with
issuance volume. This assumption is demonstrably wrong. The core infrastructure required to
operate a payment stablecoin—enterprise-grade hardware security modules, blockchain node
clusters, cybersecurity operations centers, treasury management systems, compliance monitoring
platforms, and the specialized personnel who operate them—is substantially the same whether
the issuer has $5 billion or $50 billion in outstanding issuance. Safeguarding a $50 billion
stablecoin does not require ten times the HSMs, ten times the compliance officers, or ten times
the cybersecurity engineers of a $5 billion stablecoin; the marginal operational cost of each
additional billion in issuance is a fraction of the cost of the first billion. Tying a 12-month
backstop to this flawed linear assumption means that as a federal issuer like BitGo scales —
23

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

precisely the growth trajectory the GENIUS Act is designed to encourage—it is forced to trap
exponentially larger, wholly unjustified sums of working capital in idle, non-yielding reserves
that bear no relationship to actual operational wind-down costs. Capital that could otherwise be
deployed to fund technology investment, reserve management infrastructure, compliance
systems, and the operational resilience capabilities that stablecoin holders actually depend upon
is instead sterilized by a formula untethered from economic reality. State-chartered stablecoin
issuers below the $10 billion federal transition threshold face no comparable backstop
requirement under most state frameworks, and offshore stablecoin issuers face none at all. The
12-month requirement thus creates a direct and unjustified competitive disadvantage for the most
heavily supervised institutions in the market—precisely the institutions the GENIUS Act seeks to
encourage to issue stablecoins under federal supervision. If the OCC’s assessment fees and
operational capital requirements make federal stablecoin issuance materially more expensive
than state or offshore alternatives, the inevitable market response is regulatory arbitrage away
from the federal framework—the exact outcome the GENIUS Act was designed to prevent.
The OCC should also explicitly permit the six-month backstop to be held in short-dated U.S.
Treasury bills maturing within 90 days. The operational purpose of the backstop—ensuring the
issuer has liquidity to fund an orderly wind-down—is fully satisfied by short-dated Treasury
bills, which are the most liquid instruments in the global financial system. Treasury bills can be
converted to cash within one business day through any primary dealer or through the Federal
Reserve’s open market operations, and they carry zero credit risk. There is no plausible winddown scenario in which the difference between holding the backstop in cash versus 90-day
Treasury bills would impair the issuer’s ability to fund its operations. By contrast, requiring
issuers to hold the backstop in non-yielding cash imposes a material and ongoing drag on
operational sustainability: at current interest rates, trapping six months of operating expenses in
uninvested cash rather than Treasury bills represents a significant and entirely unnecessary
opportunity cost that compounds each year the backstop requirement remains in effect.
Permitting the backstop to be held in Treasury bills would eliminate this drag without any
corresponding reduction in the protection the backstop is intended to provide. The OCC has
implicitly recognized this logic in its treatment of stablecoin reserves themselves, which may be
held in short-dated Treasury securities under the GENIUS Act’s reserve asset framework. There
is no principled basis for applying a more restrictive standard to the operational backstop than to
the reserve assets themselves, when both serve liquidity-assurance functions and both are equally
well-served by Treasury bills.

XI. Assessment Fees Must Reflect the Passive Nature of Reserve Management
(Response to Questions 194 and 195)
The proposed 35% assessment fee discount for stablecoin reserve assets substantially understates
the actual supervisory burden differential between passive reserve portfolios and traditional bank
credit assets, and the assessment structure’s linear scaling above any issuance threshold creates
barriers to scale that directly undermine the GENIUS Act’s objective of attractive large-scale
stablecoin issuance into the federal supervisory framework. For the reasons set forth below, we

24

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

urge the OCC to adopt the maximum 55% discount contemplated in the NPRM and to impose a
hard cap on assessment fees above the $10 billion issuance threshold. 21
A. The Discount Should Reflect the Reduced Supervisory Burden of Reserve Assets
Reserve assets backing payment stablecoins are fundamentally different from traditional bank
assets in ways that substantially reduce supervisory complexity. 22 Under the GENIUS Act’s 1:1
reserve requirement, permitted issuers hold only cash, short-term U.S. Treasury securities, and
other high-quality liquid assets that pose minimal credit risk, require no loan-loss provisioning,
involve no underwriting decisions, and generate no complex valuation disputes. These are
passive, transparent holdings verifiable through automated systems and monthly attestation
reports.
In contrast, traditional bank supervision requires intensive examination of credit portfolios, loan
classification, borrower financial condition, collateral valuations, and complex risk management
frameworks. The marginal cost to the OCC of supervising an additional billion dollars in passive
Treasury holdings is substantially lower than for an equivalent amount of commercial loans or
other credit products.
The proposed 35% discount implies that supervising stablecoin reserves requires 65% of the
resources needed for traditional bank assets. 23 BitGo respectfully submits that this does not
adequately account for the reduced supervisory burden and urges the OCC to adopt the
maximum 55% discount contemplated in the NPRM.
B. Asset-Based Assessment Fees Should Be Subject to Caps
The OCC’s proposed structure of charging basis-point fees on total reserve assets creates a
scaling concern that the OCC itself identifies, acknowledging that fee caps may be appropriate
when assessment revenues become disproportionate to marginal supervisory costs.24 The OCC’s
assessment structure for traditional banks reflects the reality that larger loan portfolios require
proportionally more examiner hours to evaluate credit quality, collateral adequacy, and borrower
financial condition. That logic does not translate to stablecoin reserves. Unlike traditional bank
assets, where asset growth corresponds to increased lending complexity and supervisory burden,
stablecoin reserve assets scale without a corresponding increase in supervisory complexity. A
$50 billion portfolio of segregated Treasury bills is not fifty times more complex to examine than
a $1 billion portfolio—it is functionally the same examination, repeated against a larger balance.
Examining $10 billion in segregated Treasury securities held by a qualified custod ian requires
substantially the same OCC resources—the same attestation review protocols and reserve
verification procedures—as examining $100 billion in identical securities held under the same
21

See proposed § 15.11(c) (reserve asset diversification); NPRM Questions 194 and 195 (assessment fees).
See 12 U.S.C. § 5903(a)(1)(A) (setting forth the GENIUS Act’s 1:1 reserve requirement and permissible reserve
asset types).
23
NPRM at 10248 (discussing the proposed 35% discount and maximum 55% discount for stablecoin reserve
assets).
24
See NPRM Question 195 (“[F]ee caps may be appropriate when assessment revenues become‘disproportionate to
the marginal cost of supervising GENIUS Act-related activities.’”).
22

25

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

procedures. The reserve verification procedures are identical; the attestation review protocols are
identical; the only variable is the number of digits on the balance sheet. Linear fee scaling above
$10 billion would not constitute cost recovery for supervision—it would constitute a tax on scale
that punishes issuers for commercial success.
BitGo urges the OCC to establish a hard cap on assessment fees for reserve assets, such that fees
plateau once stablecoin issuance reaches $10 billion. The operational reality is straightforward:
reviewing a SOC 1 Type 2 attestation and monthly reserve composition reports for a $5 billion
portfolio of segregated U.S. Treasury securities held by a qualified custodian requires the exact
same number of OCC examiner hours as reviewing those same reports for a $50 billion portfolio
of identical securities held under the same custody procedures and attestation framework. The
attestation is the same length, the reserve verification protocols are the same, and the
examination procedures are the same—the only difference is the number of zeros on the balance.
There is no principled basis for scaling assessment fees linearly above $10 billion when marginal
supervisory costs are functionally flat. BitGo accordingly rejects any tiered scaling structure
above the $10 billion threshold as bearing zero relation to actual supervisory resource
requirements. Without a hard cap, assessment fees will inevitably become grossly
disproportionate to marginal supervisory costs at scale, which could make large-scale federal
stablecoin issuance commercially uncompetitive relative to stat e-chartered or offshore
alternatives—directly undermining the GENIUS Act’s objective of concentrating stablecoin
issuance within the federal supervisory framework.
C. Excessive Assessment Fees Risk Regulatory Arbitrage
Permitted payment stablecoin issuers will compete directly against state -chartered issuers (until
they reach the $10 billion federal transition threshold) and against offshore stablecoin products
that face no comparable assessment fees. If the OCC’s assessment fees make federal stablecoin
issuance significantly more expensive than state-chartered or offshore alternatives, the
predictable market response will be regulatory arbitrage away from the federal framework,
directly undermining the GENIUS Act’s objective of bringing stablecoin issuance under federal
supervision.
BitGo operates under multiple international regulatory regimes and observes that comparable
foreign digital asset regulatory regimes take a different approach to supervisory fees. For
example, under the EU’s Markets in Crypto-Assets Regulation (MiCA), the EBA determines its
aggregate supervisory fee pool for significant e-money token issuers using an activity-based cost
model, meaning the total fees collected reflect actual supervisory expenditure rather tha n scaling
mechanically with token issuance volume. Individual issuer fees are then allocated in proportion
to issuance size, but because the overall pool is cost-driven, the per-issuer burden does not
increase linearly with market growth. Similarly, the Monetary Authority of Singapore’s licence
fees for major payment institutions under the Payment Services Act are flat, fixed amounts
assessed per licensed payment service type, with no variable component tied to transaction
volume or the total value of assets under custody.

26

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

D. Specific Recommendations
BitGo respectfully urges the OCC to: (1) adopt the maximum 55% discount for stablecoin
reserve assets; (2) establish caps on assessment fees above specified issuance thresholds, such
that fees plateau rather than scale linearly with issuance volume; (3) co mmit to a transparent,
periodic review process under which the OCC publishes an annual report comparing assessment
revenues from stablecoin issuers to the actual marginal cost of supervision, with any fee
adjustments—upward or downward—tied to demonstrated changes in supervisory resource
requirements and subject to notice-and-comment rulemaking, thereby ensuring that the review
mechanism promotes accountability rather than serving as an unconstrained vehicle for fee
increases; and (4) publish a methodology for calculating the marginal cost of stablecoin
supervision, providing transparency and a basis for future fee adjustments.

XII. Summary of Specific Requests
In summary, BitGo respectfully requests that the OCC:
1. Decline to adopt a one-brand-per-issuer restriction and instead permit issuers to issue
multiple brands of payment stablecoin through a single legal entity, subject to per -brand
reserve segregation, per-brand attestation reporting, and, if appropriate, a legal opinion or
third-party attestation confirming bankruptcy-remote treatment.
2. Amend proposed § 15.10(c)(4)(i) to codify explicit safe harbors for independently funded
affiliate loyalty programs, merchant discounts, arm’s-length white-label arrangements,
and de minimis arrangements—such as rewards under a specified per-holder dollar
threshold—that do not have a meaningful economic impact and must not be subject to the
rebuttable presumption or rebuttal process, and specify the evidentiary standard, a 30-day
review timeline, deemed-approval mechanism, and appeal rights for rebuttal submissions.
3. Finalize the Part 15 capital election for uninsured national trust banks in proposed §
15.41, confirm that it is available through a streamlined 30-day notice process with
defined and limited “good cause” objection criteria, decline to adopt the bifurcation
alternative, explicitly grandfather any existing regulatory capital instruments issued by
uninsured national trust banks that elect the Part 15 framework, and reduce the proposed
12-month operational backstop to a maximum of 6 months while explicitly permitting
this backstop to be held in yield-bearing U.S. Treasury bills.
4. Require transitioning state-chartered issuers to undergo comprehensive pre-transition
capital, liquidity, and operational examinations that must conclude before or concurrently
with the transition date—not months afterward—to ensure they meet the same standards
applicable to institutions that have operated under OCC supervision from inception, with
a defined remediation period for shortfalls. The OCC’s proposed framework allowing up
to six months of unexamined operation under federal status must be rejected as creating
an unacceptable competitive disadvantage for native federal institutions like BitGo.

27

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

5. Explicitly confirm under Subpart C that covered custodians may utilize cryptographically
segregated omnibus account structures combined with verifiable internal sub -ledgering to
satisfy all segregation and separate accounting requirements, including by per mitting
covered assets and non-covered digital assets to be held within the same omnibus wallets
provided that cryptographic sub-ledgering mathematically segregates each asset class and
each customer’s beneficial ownership at all times, and adopt principles-based,
technology-neutral segregation standards.
6. Modify the proposed semiannual assessment fee structure to adopt the maximum 55%
discount for stablecoin reserve assets, establish a hard cap on assessment fees once
stablecoin issuance reaches $10 billion—reflecting the indisputable fact that marginal
supervisory costs are functionally flat above that threshold—commit to annual fee
reviews tied to actual supervisory costs, and publish a marginal-cost methodology for
stablecoin supervision.
7. Adopt Option A for reserve asset diversification under § 15.11(c), preserving the
principles-based general requirement with the optional quantitative safe harbor, exempt
Federal Reserve Banks and Global Systemically Important Banks (G-SIBs) entirely from
the 40% single-institution concentration limit within the safe harbor, and confirm that the
20-day weighted average maturity limit is calculated on a portfolio basis.
8. Modify the stress-extension mechanism in proposed § 15.12 to permit an issuer that has
crossed the 10% redemption threshold to demonstrate reserve adequacy and continue
honoring redemptions on the standard two-business-day timeline subject to OCC
concurrence; clarify that automated reserve reporting satisfies the 24-hour notification
requirement; and confirm that where per-brand reserves are fully segregated, the 10%
threshold must be calculated on a per-brand basis with reference solely to that brand’s
dedicated reserve pool and outstanding issuance, so that a redemption surge in one brand
does not trigger a mandatory extension for any other brand whose reserves are
independently segregated and unaffected.
9. Strike or strictly limit the proposed weekly reporting requirement under § 15.14(h) that
requires permitted payment stablecoin issuers to report the “100 persons that hold or
trade the stablecoin the most,” and limit any such reporting obligation exclusively to
direct, known customers who have been officially onboarded through the issuer’s
KYC/BSA-AML processes; the requirement must not extend to secondary market
participants whose identities are unknown to the issuer and cannot be reliably determined
from on-chain data, as any such extension would be technologically unworkable,
commercially ruinous, and productive of speculative and unreliable supervisory data
rather than meaningful insight.
10. Impose heightened scrutiny on applications for national trust bank charters where the
applicant’s sole or predominant purpose is to custody stablecoin reserves for an affiliated
issuer, including requirements to demonstrate a bona fide intention to serve unaffiliated
clients, enhanced capital or liquidity requirements for single-purpose self-custody

28

Docusign Envelope ID: EB8D87F1-B33D-4839-B9F0-EDCE62B2B43A

structures, and examination fee surcharges reflecting the supervisory burden of
overseeing captive institutions without external market accountability.
11. Confirm explicitly in the final rule that national trust banks chartered under 12 U.S.C. §
27(a)-including uninsured national trust banks that do not accept demand deposits or
extend commercial credit-qualify as “eligible financial institutions” for purposes of
holding stablecoin reserves under the GENIUS Act and Part 15, and affirmatively reject
any effort to narrow this definition to exclude non-depository trust banks from the reserve
custody market.
12. Require all covered custodians holding stablecoin reserves-regardless of whether the
custodian is a national trust bank, a state-chartered trust company, or an FDIC-insured
depository institution-to hold such reserves in a fiduciary capacity, with reserves legally
segregated and held for the exclusive benefit of stablecoin holders, ensuring consistent
protection regardless of the issuer’s choice of custodian and eliminating the competitive
asymmetry that currently advantages depository banks offering non-fiduciary deposit
products.
13. Recognize in the final rule and accompanying guidance that FDIC deposit insurance
provides no meaningful protection for stablecoin reserves at any commercially significant
scale-coverage of $250,000 per depositor represents a fraction of a percent of reserves for
any issuer above $100 million-and that the absence of FDIC insurance for national trust
banks reflects a business model that does not generate the credit-extension and assetliability mismatch risks that deposit insurance was designed to address, rather than a
supervisory deficiency requiring compensating capital or other burdens.
BitGo appreciates the OCC’s comprehensive and thoughtful approach to implementing the
GENIUS Act. As one of the few institutions globally operating regulated stablecoin
infrastructure under multiple federal and international frameworks, we stand ready to assist the
OCC in refining these rules to achieve their consumer protection and financial stability
objectives. We welcome continued dialogue on these implementation issues and would be
pleased to provide additional information or technical briefings on any of the matters discussed
in this comment letter.
Respectfully submitted,

Michael A. Belshe

Charles D. Thompson

Chief Executive Officer
BitGo Holdings, Inc.

Chief Legal Officer
BitGo Holdings, Inc.

29