Tokenization is taking hold. That much is settled. It is showing up in financial market clearing and settlement, payments, and in products that did not exist three years ago. What is not settled is what “tokenization” actually means.
My read: the majority of tokenization efforts are best categorized as updating existing flows between existing parties using new DLT technology. Improving flows is not the same thing as remaking a market, and the press releases rarely distinguish between the two. I recommend you parse the hype carefully as it is at or near its apex.
Regular readers will recognize the theme, and I laid out the taxonomy in 101 Update: CBDCs, Stablecoins and Tokenized Deposits, and also argued in Stablecoins: A New Model of Trust enabled by Technology that the defining innovation is the trust architecture rather than the technology, and applied it most recently to SWIFT’s blockchain announcement in Parsing the Hype.
Technology Alone Changes (Almost) Nothing
Historically speaking, new technologies typically gain traction first inside existing markets, as each competing entity seeks to build competitive advantage. EXISTING Industries do not change because a better database arrived. They change when: commercial agreements get rewritten, trust gets reallocated, or a new competitor operates on a substantially more efficient platform. Tech may ENABLE some of this change, but operationalizing is where value is created.
The IMF reached the same conclusion this April:
“This Note argues that tokenization constitutes a structural reallocation of trust within the financial system. In traditional architectures, trust is embedded in regulated intermediaries, layered institutional processes, and the sequencing of settlement over time. In tokenized systems, execution, settlement, and aspects of risk management migrate toward shared infrastructures and programmable logic.” — Tobias Adrian, IMF Notes No. 26/01, “Tokenized Finance,” April 2026
The same Note delivers the line that belongs on a wall in every innovation team: “Tokenization is a technological enabler, not a determinant of institutional design.” I’m glad Ajay’s team at the IMF is reading my stuff.. he is my old boss (LOL).
Where Trust Actually Moves: B2B Cross Border
The clearest case of genuine reallocation is B2B cross border payments, and it helps to be precise about what exists today. SWIFT is not a financial institution, and no money moves through it. It is a messaging network that banks use to transmit payment orders to one another. Where two banks do not hold accounts with each other, they route through correspondents, and each additional intermediary makes the transaction slower and more expensive. The chain reports on its own status poorly or not at all.
A stablecoin does something structurally different. It replaces trust in a chain of banks with trust in the issuer and the reserves backing the token. The counterparty question changes from “do I trust this correspondent network” to “do I trust these reserves and the regime governing them.” That is a real reallocation, which is why it qualifies as disruptive rather than merely efficient.
The stakes are large because banks own this flow. UBS, citing FXC Intelligence, puts 92% of all B2B cross border volume as bank driven, and roughly three quarters of cross border payments made by small and medium sized businesses. When trust shifts in a flow that concentrated, it matters.
Financial Market Tokenization Is a Different Animal
Financial market tokenization works differently. Canton Network manages a closed network, its rules, and its usage among existing participants conducting existing flows. JPMorgan’s Kinexys does something comparable from a bank member’s (or market maker) balance sheet outward. Both are genuinely well executed.
The numbers are not small. Kinexys has processed over $4 trillion since inception and now averages more than $7 billion in daily transaction volume (J.P. Morgan, June 2026). Canton has attracted the institutions that matter, including DTCC, which is tokenizing a subset of the Treasuries it custodies on it. In January, Digital Asset and Kinexys announced JPM Coin (JPMD) would be issued natively on Canton, phased through 2026.
These platforms deliver efficiency, transparency and speed to existing markets and existing trust relationships. I covered the architecture, including Canton’s need to know privacy model and Visa’s Super Validator role, in Distributed Ledger Governance, and the bank consortium dynamics in JPMorgan, Citi and TCH: Tokenized Deposits ON Chain.
But operational improvement inside an existing market is not the same as market change. For an industry to change beyond efficiency, you need new participants, new markets, and new commercial agreements. Not a new type of connection for managing existing flows.
What Investors Should Look For
Look for where trust is shifting, who is managing that trust, and what regulatory obligations that entity operates under. That last clause does most of the work, because an entity holding reallocated trust without a matching obligation is a risk, not an opportunity.
On that test, large banks hold the strongest position, particularly across multiple regulatory and risk regimes, and the enablers serving those banks are well placed too. The value, though, does not sit in the tokenization of the underlying asset. It sits in the supply chain or the market that exploits the improvement the tokenization created. Canton, Kinexys and Airwallex are all better understood this way.
There is a useful reality check here from UBS, which found that stablecoins have accelerated bank DLT adoption but that “their economics constrain widespread use (unattractive for ‘money at rest’ as they pay zero yield to users)” (Justin Forsythe et al., UBS, “The Institutionalization of Blockchain and Stablecoins,” November 2025). Adoption is real and it is bounded.
The Dollar, the GENIUS Act, and Europe
The US and the USD hold tremendous advantages in tokenization. Our markets and operators have far greater flexibility to act, and the GENIUS Act supplies the supporting regulation. The BIS data explains why Europe is right to be concerned:
“Stablecoins’ immediate impact appears likely to reinforce rather than challenge existing currency hierarchies, particularly dollar dominance.” — Iñaki Aldasoro, Jon Frost and Hiro Ito, BIS Papers No. 170, April 2026
The BIS puts dollar denominated stablecoins at roughly 98% of the stablecoin market by value, a share higher than the dollar’s dominance in traditional international finance. European concern is not protectionism, it is arithmetic.
The ECB’s answer is a digital euro, a CBDC that, in my view, neither banks, merchants nor consumers particularly want, but which gives the ECB the greatest control. UBS reached a similar conclusion from the commercial side:
“Merchants may have little incentive to encourage usage if their unit economics are only modestly different from existing methods and banks may be reluctant to support broad adoption if implementation, servicing, compliance, and dispute-management costs are not adequately compensated. Legal-tender-like status may guarantee availability, but availability alone does not create consumer preference.” — UBS, “Addressing investor questions around European payments efforts,” July 2026
China likes this model too. Enough said.
Wrap Up
Tokenization is real, it is working, and most of it is plumbing. Excellent plumbing in the case of Kinexys and Canton, and I would rather own the well run plumbing than the narrative. Ask of any tokenization announcement: who used to hold the trust, who holds it now, and what are they obligated to do with it? If the answer is that nobody moved, you are looking at an efficiency project. Those are valuable. They are not revolutions, and they should not be priced as such.