An unnamed national bank has asked the Office of the Comptroller of the Currency for permission to hold crypto on its own balance sheet to support blockchain-based services. On Nov. 18, the OCC finally answered.
In Interpretive Letter 1186, the agency confirmed that national banks may hold the native assets needed to pay blockchain “network fees,” clearing the way for regulated institutions to run on-chain operations without external workarounds.
The letter says a national bank may pay blockchain “network fees,” commonly known as gas, as an activity “incidental to the business of banking.” It may hold, as principal, the crypto assets needed to cover those fees where it has a “reasonably foreseeable” operational need.
The ruling sits between plumbing and precedent. Network fees on public chains are paid in the native asset of the chain. Hence, any bank that wants to custody tokens, move customer stablecoins, or run tokenized deposits on Ethereum or similar networks needs some amount of ETH or an equivalent in hand.
Until now, many banks either stayed away from on-chain activity entirely or leaned on third-party providers to front gas and wrap it into a fiat fee.
The OCC is now saying banks can hold those native tokens themselves as principal if they’re only needed to run the pipes.
For large custodians, tokenization desks, and stablecoin issuers operating under the GENIUS Act framework, that shift means they can finally be full-stack on specific networks without outsourcing the last missing piece.
The agency said those activities will require banks to pay network fees “as agent for the customer or as part of its custody operations.”
Letter 1186 zooms in on a specific operational snag inside that new framework: you can’t do on-chain custody or tokenized deposits if you’re not allowed to hold the gas token.
American Banker quotes the letter’s logic directly. If serving as a node is permissible, then “accepting the crypto asset network fee” and holding it for some period must also be acceptable.
Otherwise, a bank could be “practically barred” from a lawful activity. That reasoning gives large custodians a cleaner path to maintain a small gas balance in-house rather than farming that function to fintech intermediaries or staying off-chain altogether.
The same letter confirms that banks can also hold limited amounts of crypto as principal to test otherwise permissible crypto-asset platforms, whether built in-house or bought from a third party.
In other words, the OCC is blessing small, working inventories of native tokens so banks can actually move transactions on the rails they’re allowed to use, and safely test those rails before committing customer funds or balance-sheet capital to production deployments.
For payments and settlement, this is about plumbing, not proprietary trading. The change matters most for banks running stablecoin operations or tokenized deposit programs that settle on public chains.
Those institutions now have explicit authority to hold the gas needed to process customer transactions without structuring workarounds or relying on external liquidity providers.
Several summaries stress that holdings are limited to “operational needs,” including fee buffers for settlement and for testing custody platforms, not open-ended speculative positions.
That’s the distinction between fee custody and balance-sheet crypto exposure: banks can hold enough ETH to cover foreseeable transaction volumes and platform testing, but they can’t build a speculative book or treat native tokens as an investment asset.
The OCC’s framing makes clear this is operational inventory, not a new asset class for bank treasuries.
For custody desks, the ruling removes a layer of counterparty risk and operational complexity.
Banks that previously relied on third parties to provide gas now have the option to internalize that function, which shortens execution timelines and eliminates intermediaries that might themselves face liquidity constraints during network congestion or market volatility.
It also positions national banks to compete more directly with crypto-native custodians that have always held native tokens as part of their service stack.
The OCC stresses that all of this must be done in a “safe and sound” manner and in compliance with existing law.
The agency’s press release and commentary from the American Bankers Association highlight that banks must narrow the size of these holdings, tie them to specific permissible activities, and run the usual market, liquidity, operational, cyber, and BSA/AML risk frameworks around them.
The OCC only oversees national banks, while the Federal Reserve has, in a separate policy statement, continued to describe holding crypto as principal as “unsafe and unsound” for state member banks, creating cross-regulator friction even after the OCC loosened its stance earlier this year.
That divergence means OCC-chartered banks have the green light to use operational gas balances. However, the broader US bank universe still faces mixed signals, depending on charter type and primary regulator.
Banks will also need to navigate price volatility. Native tokens like ETH fluctuate, which means the dollar value of a bank’s gas inventory can swing day to day even if the token quantity stays fixed.
The OCC’s “reasonably foreseeable operational need” standard implies banks should size buffers conservatively and avoid holding excess tokens that would expose them to speculative risk.
That creates a balancing act: hold too little and banks risk running out of gas during high-congestion periods. On the other hand, holding too much implies carrying volatile assets on the balance sheet without a clear operational justification.
The broader question Letter 1186 answers is whether US banks can participate in on-chain finance without regulatory workarounds or structural disadvantages relative to crypto-native competitors.
For years, the implicit answer was no: banks could offer crypto services only by staying off-chain, partnering with third parties, or seeking case-by-case approval for activities that involved direct token handling.
If the stance holds, expect national banks with existing tokenization or stablecoin programs to bring gas management in-house over the next year.
That shift won’t change the fundamental economics of on-chain payments. Still, it will consolidate more of the service stack inside regulated institutions and reduce reliance on fintech intermediaries for basic settlement functions.
It also sets a precedent for how regulators might approach other operational necessities that require holding native tokens, from staking for proof-of-stake networks to liquidity provisioning for decentralized-finance protocols that banks might eventually touch.
The risk is that this remains an OCC-only position. If the Fed doesn’t follow suit with similar guidance for state member banks, the result is a two-tier system in which charter choice determines whether a bank can hold gas tokens at all.
That would push more institutions toward national charters for crypto-related businesses, concentrating activity under a single regulator and leaving state-chartered banks at a competitive disadvantage for on-chain services.
For now, Letter 1186 is permission, not policy convergence, and the distance between those two will define how far US banks can actually go.
