The Cash LegSettlement Assets for Tokenized Markets

Most writing about tokenized markets concentrates on the asset. The bond, the fund interest, the credit position, the unit of property. That attention is understandable, because the asset is the thing being bought. It is also where the least difficult problems live.

Settlement is an exchange. One party delivers an asset and the other delivers payment. A market improves when both deliveries become faster, cheaper, and more certain at the same time. Tokenizing only the asset addresses one half of the exchange. The payment half, the cash leg, decides whether the improvement is real.

This report argues that the cash leg is the binding constraint on tokenized settlement, and that the choice of settlement asset deserves the same scrutiny an institution would apply to a counterparty. Three instruments now compete for the role: public stablecoins, tokenized deposits issued by banks, and central bank money made available for wholesale settlement. They are not interchangeable. They differ in who owes the holder, what stands behind the claim, when payment becomes final, and whether a regulator will accept the result. An institution that picks the wrong instrument, or treats the choice as an afterthought, inherits the credit and liquidity profile of that instrument whether or not it intended to.

The misunderstanding this report corrects is the belief that putting an asset on a ledger shortens settlement. It does not, by itself. What shortens settlement is the ability to move asset and cash together, with finality, on shared infrastructure. The asset is the easy leg. The cash leg is the work.

Consider a simple secondary sale of a tokenized instrument between two institutions. The seller holds the token. The buyer holds dollars at a bank. For the trade to settle, the token must move to the buyer and the dollars must move to the seller, and ideally neither should happen without the other.

If the asset moves on a ledger and the dollars move by wire, the two legs run on different clocks. The wire clears during banking hours and carries its own cutoffs. The token can move at any hour. Between the two events, one party is exposed to the other. The seller may deliver the token and wait for a wire that arrives the next morning, or the buyer may send funds against a token that has not yet appeared. This gap is principal risk, the risk that one side performs and the other does not, and it is precisely the risk that modern settlement systems were built to remove.

Conventional markets manage this with a settlement cycle and a central infrastructure that nets and guarantees the exchange. The cost is time. A trade agreed today settles a day or two later, and capital sits idle against the exposure in between. The promise of tokenized settlement is that the two legs can be joined into a single operation that either completes in full or not at all, collapsing the cycle and releasing the capital.

That promise depends entirely on the cash leg being present on the same infrastructure as the asset. A tokenized bond with no tokenized cash to settle against is a faster record sitting on top of an unchanged payment process. The asset leg moves in seconds while the cash leg waits on a wire, and the trade settles at the speed of the slower of the two. The cash leg is the one almost no one is working on.

Three settlement assets are viable for institutional use today. The grid below compares them on the dimensions that decide settlement quality. Read each row as a question an institution should be able to answer before it commits to a cash leg.

No instrument wins every row, and the tradeoffs between them are the whole decision. An institution settling around the clock across many counterparties weighs availability heavily and may accept issuer credit risk to get it. A bank settling large value between known parties weighs credit and finality and may accept narrower hours to remove them. Define how and where you intend to settle first, and the right instrument follows from that rather than the other way around.

A public stablecoin is a token that an issuer promises to redeem one for one against a reference currency, backed by a reserve the issuer holds and manages. Its strengths as a settlement asset are reach and time. It moves on public networks that operate continuously, it is available to any party the network can reach, and it settles in seconds at any hour. For a market that trades outside banking hours or across borders, no other instrument matches that availability today.

The cost is credit. A holder of a stablecoin holds a claim on a private issuer, not on a bank or a central bank. The quality of that claim depends on what the reserve contains, how it is held, who has examined it, and how quickly redemption works when many holders ask at once. These are ordinary questions of credit and liquidity, and they do not disappear because the instrument is a token. A stablecoin that holds its value in calm markets can still gap from its reference price in stressed ones, and a settlement asset that can gap is a settlement asset that can fail a delivery at the worst moment.

Dedicated regulatory regimes are changing this picture. Where a jurisdiction sets reserve, disclosure, and redemption standards for stablecoin issuers and supervises them, the credit question becomes answerable rather than speculative. The result is a widening split. A regulated, supervised stablecoin with transparent reserves is a serious institutional settlement asset. An unregulated one with opaque backing is a convenience that institutions should not settle large value against. Treating the category as uniform is the mistake. What an institution actually settles against is the issuer and the regime that supervises it, not the token format.

A tokenized deposit is an ordinary bank deposit represented as a transferable token on shared infrastructure. The holder owns the same thing a depositor has always owned, a claim on a regulated bank, with the same protections that claim carries in its jurisdiction. What changes is the form. The deposit can move on the same ledger as a tokenized asset and settle against it in a single operation.

For institutions that already keep their cash in banks, this is the least disruptive cash leg. It keeps the money inside the regulated perimeter, it keeps the credit relationship with a known and supervised counterparty, and it raises no new question about what stands behind the claim. The bank that issues the token is the bank that already holds the deposit.

The constraint is reach. A deposit token issued by one bank settles cleanly inside that bank and across whatever network the bank and its peers have agreed to join. It does not move freely to a counterparty banking somewhere else unless the two institutions share infrastructure or a bridge between them exists. Settlement among clients of a single large bank is straightforward. Settlement across the wider market requires either a shared network that many banks join or a mechanism that lets one bank's deposit token be exchanged for another's at par. Building that connective layer is the open problem for tokenized deposits, and it is a problem of banking cooperation and market structure rather than of technology. Funding in and out of these networks also tends to follow banking hours, which limits the around the clock settlement that a continuous market wants.

Central bank money is the most senior settlement asset in any system. A claim on the central bank carries no credit risk in the ordinary sense, and large value payment systems have settled in it for decades precisely because it removes the question of whether the payer's money is good. Making it available on shared infrastructure, so that a tokenized asset can settle against central bank money in one operation, is the cleanest possible cash leg.

Several central banks have run experiments toward this, either by issuing a wholesale settlement token for use among regulated institutions or by linking conventional central bank accounts to a tokenized asset platform so that the asset moves on the ledger while the cash settles in the central bank's own system. Both approaches aim at the same outcome, settlement in the safest money without asking institutions to take on a new credit exposure to get on chain speed.

The constraint is access and hours. Central bank money is available to the institutions that hold accounts at the central bank, which is a limited set, and the wholesale programs that exist operate within defined windows rather than continuously. For settlement among large regulated institutions in a single currency, this is often acceptable and even preferred. For a market that wants continuous settlement across many participants and currencies, central bank money is the destination rather than the instrument available at scale today. Its trajectory matters enormously, because it sets the benchmark the other two instruments are measured against.

Delivery versus payment (DvP) is the principle that an asset should change hands only if payment does, and the other way around. Atomic settlement is the strongest form of it, where both legs execute as a single indivisible operation that either completes in full or leaves both parties where they started. This is the property institutions most want from tokenized markets, because it removes principal risk outright rather than managing it.

Atomicity has a precondition that is easy to state and hard to meet. Both legs must live on the same ledger, under the same definition of finality, so that a single operation can move them together. When the asset and the cash share one network, a smart contract can swap them in one step and the exchange is genuinely atomic. When they do not, atomicity is lost and something weaker takes its place.

Three weaker patterns are common, and each carries residual risk that programs should name rather than assume away. The first is cross network settlement, where the asset is on one ledger and the cash is on another, joined by a bridge or a coordinating agent. The exchange now depends on that connector performing correctly, and a failure between the two legs reopens the principal risk atomicity was meant to close. The second is the on chain asset against off chain cash pattern, where the token moves on the ledger but payment settles by conventional transfer. Here the cash leg is simply the old process wearing new clothes, and the settlement cycle returns. The third is conditional or coordinated settlement, where each leg is real but a sequencing protocol stands between them, reducing the exposure window without eliminating it.

The practical lesson is direct. Atomic settlement is a property of a specific arrangement, not of tokenization in general. An institution should ask, for any tokenized trade, whether the asset and the cash actually share a ledger and a finality regime. If they do, atomicity is available. If they do not, the program is doing coordinated settlement with residual principal risk, and it should be measured and disclosed as such rather than described as atomic because the asset happens to be a token.

Choosing a settlement asset is the beginning. Operating it is where programs succeed or stall, and the operational questions are mundane in the way that genuinely important questions often are.

Multi currency coverage is the first. A settlement asset that exists only in dollars settles only dollar trades. Markets are not single currency, and a cash leg that cannot represent the currencies a program actually trades forces those trades back onto conventional rails. The breadth of currencies an instrument supports, and the quality of each, is a gating feature rather than a detail.

Funding outside banking hours is the second. A continuous market needs cash available continuously. If an institution can only move money into the settlement asset during banking hours, then the around the clock settlement the token enables is throttled by the bank window on either side of it. Some instruments solve this and some do not, and the difference decides whether weekend and overnight settlement is real.

Redemption under stress is the third and most important. A settlement asset is only as good as the ability to turn it back into ordinary money when it is needed, especially when many holders need it at once. This is a liquidity question, and it is the question that separates an instrument that works in calm conditions from one that works in all conditions. An institution should understand, before it settles large value against an instrument, what redemption looks like on the worst day rather than the average one.

None of these problems is new. They are multi currency funding, after hours liquidity, and redemption under stress, carried into a tokenized form of money. The programs that treat them as first order, and select a settlement asset that answers them, build markets that settle better. The programs that treat the cash leg as something to add after the asset is live tend to discover that the asset settles exactly as slowly as the cash behind it.

Cash leg

The payment side of a securities trade, as distinct from the asset side. The instrument used to pay for a tokenized asset at settlement.

Settlement asset

The specific form of money used to settle the cash leg, for example a public stablecoin, a tokenized bank deposit, or central bank money.

Delivery versus payment (DvP)

A settlement principle under which an asset transfers only if payment transfers, and payment transfers only if the asset does.

Atomic settlement

Settlement in which both legs of a trade execute as a single indivisible operation that either completes in full or leaves both parties unchanged.

Principal risk

The risk that one party to a trade delivers its side while the other fails to deliver, losing the full value of the delivery rather than a price difference.

Public stablecoin

A token redeemable one for one against a reference currency, issued by a private party and backed by a reserve that party manages.

Tokenized deposit

An ordinary bank deposit represented as a transferable token, keeping the holder's claim on the regulated bank that issued it.

Central bank money

A claim on the central bank, the most senior settlement asset in a monetary system, used in wholesale settlement to remove credit risk.

Settlement finality

The point at which a transfer of value becomes irreversible, defined both operationally on a ledger and legally in the relevant jurisdiction.

Wholesale settlement

Settlement of large value transactions among regulated financial institutions, as distinct from retail payments.

Citation slots below mark claims and context that require source verification before this document is treated as externally citable. They are placeholders by design. This library does not assert sourced facts without sources.

1. 01

   Official-sector analysis of settlement in central bank money and the role of a unified ledger in tokenized markets.

   Citation pending[Citation needed: Bank for International Settlements work on tokenization and wholesale settlement]

2. 02

   Supervisory frameworks for stablecoin issuers, including reserve composition, disclosure, and redemption requirements across major jurisdictions.

   Citation pending[Citation needed: applicable stablecoin regimes, for example EU MiCA and equivalent regimes in other jurisdictions]

3. 03

   Bank consortium and industry work on tokenized deposits and shared settlement networks.

   Citation pending[Citation needed: published tokenized deposit pilots and shared ledger initiatives]

4. 04

   Central bank experiments on wholesale settlement tokens and on linking conventional settlement systems to tokenized asset platforms.

   Citation pending[Citation needed: documented wholesale central bank settlement trials]

5. 05

   Market infrastructure analysis of delivery versus payment design and settlement finality in distributed settlement.

   Citation pending[Citation needed: market-infrastructure publications on DvP and finality]

For adjacent BlockHedge work, see Tokenization: The Institutional Primer for the layered model this report sits within, and The Liquidity Problem in Tokenized Markets for why a credible cash leg is necessary but not sufficient for a market to trade.