Tokenized Deposits and Transaction Banking: The Mandate War Has Already Started

Of the 290 banks that lead the world in transaction banking, meaning those with material FX, card-issuing, cash-management, and trade-finance businesses, just ten (3.4%) have a live tokenized deposit or stablecoin capability today. Around 21% are live or committed to launch by June 2027, according to our analysis. 

That gap, 3.4% live against 21% committed, may be the most important number in transaction banking. The live base is small and adoption is early, but it’s the trajectory between those figures, not today’s count, that turns a capability from experiment into competitive qualifier.

For any bank outside the 21%, the implication is uncomfortable. Near term, a transaction bank serving multinational treasurers can’t compete if it can’t move value across borders in real time. Longer term the stakes are higher: digital money is the only form built for AI-agentic commerce, and a bank without it excludes itself from the majority of future revenue streams.

What Tokenized Deposits Deliver For Transaction Banking

In this piece, I focus on tokenized deposits, which are bank liabilities on a programmable ledger, distinct from stablecoins and central bank digital currencies. I’ll cover what they change for transaction banking, where the infrastructure stands today, and why the window to remain competitive is open now.

A tokenized deposit is a claim on the issuing bank, represented as a token on a ledger. In legal terms it is almost indistinguishable from an ordinary deposit in many jurisdictions: it sits on the bank’s regulated balance sheet and benefits from the same deposit insurance and prudential treatment.

What changes when a deposit is tokenized is not its character but its behavior: it becomes programmable, atomic and composable.

• Programmability: conditional logic is embedded in the instrument; a payment releases when a defined trigger is met, without manual instruction.
• Atomicity: multi-leg transactions settle simultaneously, eliminating settlement risk between legs.
• Composability: the deposit can simultaneously function as a payment instrument, a collateral posting, and the settlement leg of a securities transaction.

That similarity is exactly why it is the right place to start for many banks on their digital asset buildout. Because the deposit already sits inside the bank’s KYC perimeter, it is the easiest way to build the baseline and prove value to clients quickly. And once it exists, the platform that supports it becomes the foundation for many products that can follow, including intraday lending, bank-issued stablecoin payments, and real-time investment into yield products.

Start with the treasurer’s actual problem

To understand what that foundation unlocks, let’s start with a clear problem that tokenizing deposits solves. 

In Deloitte’s 2024 Global Corporate Treasury Survey, visibility into global operations, cash and financial risk ranked as the single biggest challenge treasurers face, named by 58% of them, ten points clear of digital capabilities. Treasurers know what they cannot see. 

The Bank for International Settlements counts more than $27 trillion sitting in correspondent and nostro accounts worldwide. Not all of that is trapped, but the scale is well documented: PwC’s latest Working Capital Study puts excess working capital tied up across the world’s listed companies at $1.84T of that $27T. 

For a firm operating across twenty or more jurisdictions, a material share sits structurally idle, stranded in the wrong entities, and sometimes funded by intraday borrowing at 40 to 50 basis points. On a $10 billion treasury, even a conservative 10–15% is $1–1.5 billion working suboptimally for no one.

It is worth being precise about what tokenized deposits do and do not do, because the hype is somewhat deafening on occasion. Regarding FX, it will not compress your spread: the bid-offer your desk charges on a conversion is unchanged. What tokenized deposits strip out is the correspondent FX markup layered on top, and the timing risk of a rate struck at initiation but settled days later. Nor is the prize continuous, retail-style 24/7. Few treasurers need to move money at 3 a.m. on a Sunday. What they need is for a payment to stop stalling whenever it was released within the bounds of a working day, but lands in a weekend or holiday window.

What tokenized deposits do solve for is real, and treasurers have already named it: visibility, trapped liquidity, and the settlement delay buried in every correspondent chain.

Where tokenized deposits and stablecoins are already delivering

Four use cases carry the near-term commercial argument: 

• trapped liquidity and cash visibility for multinational and HNW clients; 
• card-scheme and merchant settlement; 
• asset settlement on a delivery-versus-payment basis; and 
• cross-border value transfer. 

The last three of these are already in production. A cross-border payment with one leg falling on a weekend or public holiday no longer means a two-day delay; value moves continuously. Cross-border intra-bank transfers, value moving between two clients of the same bank, now have a blueprint to follow. JPMorgan’s Kinexys settles around $5 billion a day on its own ledger. The global banks with proprietary platforms have largely solved this.

The harder problem is inter-bank: a client of one bank paying a client of another in real time, with no correspondent chain in between. No single rail has won here yet, but it would be wrong to say no one is building. The gap is being closed from three directions at once.

From the platform side, the rails themselves are being laid. Partior is live for USD, EUR and SGD; the Cari Network is piloting inter-bank transfers across five US regional banks; and in June 2026 the US incumbents moved collectively, with The Clearing House announcing a bank-owned network to clear and settle tokenized deposits around the clock, linked to RTP and CHIPS.

From the orchestration side, Swift is taking a different route: rather than build a competing rail, it is constructing a shared ledger that lets banks’ own tokenized deposits interoperate, with banks retaining control of their keys, assets and settlement. Its MVP, involving more than 40 institutions, is targeting live transactions before the end of 2026.

None of this closes the opening for regional and super-regional banks. If anything, it sharpens it. The rails will exist, but a rail supplies neither the client relationship, the local expertise, nor the operating model. The qualifier is shifting from “can you build a rail” to “are you connected to the one your correspondents and clients already expect.” The banks in the middle that wait will discover, within five to seven years, that mandates have become very hard to win.

And the scale of the prize is becoming clear: Citi Institute projects that tokenized bank deposits could support $100–140 trillion in annual flows by 2030, rivaling or even surpassing stablecoins.

Examples from the production landscape, as it stands, look like this:

The inter-bank opportunity is open now, especially for regional and super-regional banks

The incorporation of digital assets in core banking  is already an institutional consensus rather than a frontier bet. 

We surveyed more than 600 senior decision-makers at financial institutions and corporates worldwide for our Financial Grid study, and 76% named other banks as a strong or primary source of demand for tokenized deposits, stablecoins and other digital asset solutions. Your own correspondents, prime brokers and liquidity providers increasingly expect you to connect, and the inability to do so becomes a reason for clients and counterparties to take their business elsewhere.

And precisely because no one owns the inter-bank problem yet, it offers perhaps the clearest opening for regional and super-regional banks. 

A treasurer at a Southeast Asian conglomerate with subsidiaries in six countries does not need a global bank’s scale; they need real-time liquidity management within the region, with full visibility across entities. The rails solve the plumbing, not the permissions: a tokenized rupee or renminbi still moves only where capital controls allow, but within those limits the regional bank already has the relationship and the local expertise. What it lacks is the real-time book-transfer capability that now turns up as a qualifier in global cash-management RFPs. 

Why so few transaction banks are live with tokenized deposits

The constraint is rarely the technology, and our survey is blunt about where the difficulty actually sits. The money is committed: 88% of institutions have funded or will fund digital-asset infrastructure this year, and more than half are already spending over $1 million, yet only 16% have reached production. That gap between budget and go-live is the defining feature of this moment, and what sits inside it is internal. 

The foundational custody and wallet-governance decision is still unresolved at 85% of institutions; only 15% call their infrastructure production-ready. Skills are the single biggest internal obstacle, named by 42% of institutions overall and by 67% of global transaction banks: the very firms whose clients are asking for this first. 

And among European banks, 47% point to their own standard operating procedures—the operating model itself — as the main thing in the way. The honest first step is therefore internal: a bank serious about building this resolves custody, talent, and process first.

Build the infrastructure for tokenized deposits today

The case for moving now rests on more than what tokenized deposits do today. They are the baseline for what comes next: AI-agentic payments and micro-lending can only be layered onto infrastructure that already exists. 

The capability is becoming a qualifier, not a differentiator. The window to build it is open and, as the jump from 3.4% to 21% shows, it is already closing.

More than 100 banks are already running their tokenized deposit and digital asset operations on Fireblocks. We have secured over $16 trillion in transactions to date. Talk to us to get into production.