Money is beginning to acquire the logic of software. The contest is not simply over which digital currency wins, but over who controls the rules, settlement assets and programmable infrastructure beneath the next generation of finance.
For most of modern banking, money has been digital without being programmable. A bank balance can move electronically, but the money itself does not normally carry logic about when a transaction should settle, which asset should move against it, whether compliance checks have passed or whether a payment should release only after an event occurs. Those instructions live in separate systems. The emerging architecture of tokenised finance tries to collapse that distance: money, assets and transaction logic can sit on interoperable or shared ledgers and execute in tighter sequence.
That is the real significance of the phrase “programmable money”. It is often used loosely, sometimes to describe digital currencies with restrictions on how they can be spent and sometimes to describe ordinary money moving through smart-contract infrastructure. The distinction matters. Central banks in Europe and the United Kingdom have explicitly rejected the idea that retail central bank money should be restricted to approved uses. At the same time, they are actively testing conditional payments and interoperability with programmable ledgers. The race, therefore, is less about creating money that tells citizens what they may buy and more about creating payment and settlement systems that can execute rules automatically.
The contest is already global. According to BIS Papers No. 159: Advancing in tandem — results of the 2024 BIS survey on central bank digital currencies and crypto, 91% of 93 central banks surveyed were exploring a retail CBDC, wholesale CBDC or both, and wholesale projects were generally more advanced. Meanwhile, stablecoins have moved from the crypto economy toward a regulated payments category, tokenised bank deposits are being tested in real-value transactions, and central banks are exploring whether reserves themselves should become available on programmable infrastructures. The result is not one race but several overlapping ones.
The race is over the monetary stack, not a single coin
The easiest mistake is to frame the future of money as a winner-takes-all contest between CBDCs, stablecoins and bank deposits. In practice, the emerging system is layered. There is a settlement asset at the base, an instrument used by households or institutions, a ledger or network that records ownership and movement, and a services layer that supplies wallets, compliance, foreign exchange, liquidity, identity and automation. Different institutions can dominate different layers.
The IMF Note: The Rise of Tokenization — Deciphering New Trends in Payments and Asset Tokenization describes tokenised finance in similar terms: an infrastructure layer provides the rails and rules, an asset layer contains tokenised money and financial instruments, and a services layer provides the applications through which users interact with them. This framing is more useful than asking which digital currency “wins”, because a stablecoin can run on one infrastructure, a tokenised deposit on another, and central bank money can be used as the final settlement asset behind both.
For banks, that distinction is strategic. If commercial bank money remains central but settlement migrates to programmable rails, banks may preserve their role in deposit creation while losing control of parts of the payment interface. If stablecoins become a major transaction medium, banks could lose some deposits but gain roles in reserve custody, distribution and compliance. If CBDCs become important, banks may still act as intermediaries but face a new public-money alternative in digital form. Each outcome redistributes economics across the value chain rather than simply replacing one form of money with another.
Central banks are trying to modernise the anchor of trust
The central-bank response is increasingly focused on preserving the monetary anchor while making it compatible with tokenised markets. The BIS Annual Economic Report 2026, Chapter III: Anchoring trust in money: innovation beyond stablecoins argues that the next-generation system should retain the core properties of money — singleness, elasticity and financial integrity — while bringing tokenisation into the existing two-tier structure of central bank money and private intermediation. Its preferred direction is not an unregulated universe of private tokens, but a system in which tokenised central bank reserves, tokenised commercial bank money and tokenised assets can interact safely.
This is more than a conceptual debate. Project Agorá: a shared programmable platform for wholesale cross-border payments reported in May 2026 that a prototype combining tokenised central bank reserves and tokenised commercial bank deposits could achieve atomic, multi-currency settlement across jurisdictions. The project involved eight central banks, including the Bank of Canada after its later participation, and more than 40 regulated financial institutions. The next stage is intended to include real-value transactions for selected currencies and participants.
Atomic settlement is one of programmability’s most important institutional use cases. In today’s cross-border chains, messages, funding, foreign exchange and final settlement can occur at different times and through different intermediaries. A programmable platform can make linked legs execute together or not at all. The potential benefit is not simply speed: it is reduced principal risk, fewer reconciliation breaks, less trapped liquidity and a smaller operational gap between an agreement to transact and final settlement.
Retail CBDCs are becoming deliberately less “programmable”
Retail CBDC projects reveal a paradox. Governments want digital public money to support innovation, but many are wary of making the money itself programmable in a way that restricts its use. The euro area is the clearest example. The ECB: FAQs on the digital euro states that a digital euro would never be programmable money in the sense of units restricted to a predefined purpose, merchant or duration. It could, however, facilitate conditional payments such as pay-on-delivery or milestone-based transfers.
That distinction is not semantic. Money that expires, is geographically restricted or can only buy approved goods is economically different from generally accepted money. It risks fragmenting the unit of account and creating political questions about who writes the rules. Conditional payments take a narrower approach: the user chooses a condition for executing an otherwise unrestricted payment. The money remains fungible; the payment instruction becomes smarter.
The euro project itself has moved beyond its preparation phase. The ECB: Digital euro says the ECB published draft rulebook version 0.91 in July 2026 and aims to be ready for a possible first issuance in 2029, assuming EU legislation is adopted during 2026. No decision to issue has yet been taken. That timetable matters because it shows how much of the digital-currency race is now about legal and operational readiness rather than proof-of-concept technology.
The United Kingdom is taking a similarly cautious approach. The Bank of England: Progress update — Digital Pound Design Phase says the design phase ends in 2026 and that the Bank of England and HM Treasury expect to publish their assessment and decision on next steps later in the year. Meanwhile, the Bank of England: Digital Pound Lab — Phase 2 update has tested features such as conditional business-to-business payments, programmable locks and interoperability between a digital-pound architecture and external smart-contract platforms. Again, the policy direction separates an unrestricted public currency from programmable services built around it.
Stablecoins have already put programmability into production
Stablecoins entered this race from the opposite direction. They did not begin as a public-policy project; they emerged on programmable blockchains where transfers, collateral movements and decentralised-finance transactions can already be executed through software. Their advantage is distribution and composability. Their weakness is that the institutional guarantees associated with bank money and central bank money are not automatic.
The policy environment has shifted materially. In the United States, the U.S. Public Law 119-27 — GENIUS Act became law on 18 July 2025, creating a federal statutory framework for payment stablecoins. In the United Kingdom, the Bank of England: Policy statement and draft rules on regulating systemic stablecoins published in June 2026 set out a regime for sterling-denominated stablecoins judged systemic. These measures do not remove every policy question, but they move major jurisdictions from debating whether stablecoins belong in regulated finance toward defining how they may operate within it.
The BIS remains sceptical of stablecoins as the foundation of the monetary system. Its 2026 report argues that current designs can suffer from fragmentation, run risk, financial-integrity weaknesses and imperfect singleness with sovereign money. It also warns that widespread foreign-currency stablecoin adoption could intensify currency substitution and capital-flow volatility in some economies. Yet the same report acknowledges that stablecoins have demonstrated the usefulness of tokenisation and faster, programmable transfers. That combination — technological proof with institutional shortcomings — is why regulated stablecoins are unlikely to disappear even if they do not become the monetary system’s core.
Tokenised deposits may be the banking sector’s most powerful answer
For commercial banks, tokenised deposits offer a more conservative route to programmability. Economically, the liability remains a bank deposit. Technologically, the claim is represented on a programmable ledger and can potentially interact directly with tokenised securities, collateral or other deposits. This matters because the bank does not have to surrender the deposit relationship in order to offer many of the functions users associate with blockchain-based money.
The strongest evidence is appearing in wholesale markets. Hong Kong’s HKMA briefing to the Legislative Council, 4 May 2026 says Project Ensemble moved from sandbox experimentation to the EnsembleTX pilot in November 2025, enabling real-value transactions in tokenised Hong Kong dollar deposits and digital assets in a controlled environment. The pilot is being upgraded toward settlement of tokenised deposits in tokenised central bank money on a 24/7 basis.
The inference for banks is significant but should be stated carefully. Tokenised deposits are not guaranteed to defeat stablecoins. Their distribution is typically permissioned, their cross-bank interoperability is still developing, and they inherit the regulatory and balance-sheet economics of banking. But they have a structural advantage in markets where institutions want programmability without abandoning deposit money, prudential supervision and central-bank settlement. Project Agorá is important precisely because it tests that model across currencies rather than within a single closed bank network.
Wholesale programmability is likely to arrive before retail transformation
The public conversation about digital money tends to focus on consumer wallets, but the near-term economic impact may be larger in wholesale finance. The reasons are practical. Securities settlement, repo, collateral management, trade finance and cross-border treasury already involve rule-heavy institutional workflows. They contain conditional steps, documentation, reconciliation and time-zone frictions that smart contracts can potentially compress.
The IMF’s 2026 work on tokenisation points toward an architecture in which ownership, compliance and settlement can increasingly be embedded in programmable workflows. The payoff is potentially largest where there are multiple intermediaries and assets that must move together. A retail card payment is already fast and convenient in many advanced economies; shaving seconds from it may not transform consumer behaviour. Automating the simultaneous exchange of tokenised cash and securities, or coordinating multi-currency settlement across jurisdictions, can alter capital usage and operating risk in a more fundamental way.
This is also why the meaning of “24/7 money” matters. Always-on settlement sounds obviously superior, but institutions need liquidity outside traditional market hours, resilient operational staffing, clear legal finality and central-bank arrangements capable of supporting the new timetable. Programmability can remove waiting from the technology stack while exposing the fact that liquidity, law and governance do not automatically become 24/7.
The geography of programmable money is diverging
There is no single global model. Europe is treating the digital euro as a strategic public-payment option while deliberately excluding restricted-use money. The United Kingdom is exploring a multi-money system in which commercial bank deposits, a possible digital pound and regulated stablecoins can coexist. Hong Kong is pushing real-value tokenised-deposit experimentation in wholesale markets. Japan continues its retail CBDC pilot and wholesale settlement research. The UAE has gone further in law by recognising digital central-bank currency as part of the national currency framework and has developed the Digital Dirham programme.
The Central Bank of the UAE: Digital Dirham — A Primer on the UAE’s Central Bank Digital Currency sets out the Digital Dirham as a central-bank liability designed to remain fully convertible with other forms of dirham money, while the UAE’s broader programme has explored domestic and cross-border use cases. The Bank of Japan: Central Bank Digital Currency shows that Japan’s pilot programme remained active in 2026. These examples illustrate a broader point: jurisdictions are converging on digital experimentation but diverging on the institutional form, speed of rollout and balance between public and private money.
That divergence creates an interoperability problem. A programmable domestic system is only globally useful if it can exchange value with other systems without recreating the correspondent-banking frictions it was supposed to solve. Common technical standards can help, but interoperability also requires compatible compliance rules, legal recognition of settlement finality, foreign-exchange arrangements and decisions about which institutions may access central-bank money.
For banks, the strategic question is where value migrates
Banks should not assume that programmable money is merely a payments-technology upgrade. It can change the economics of deposits, liquidity and fee income. If corporate clients can move tokenised cash and collateral automatically across platforms, treasury balances may become more mobile. If stablecoin issuers capture transaction balances, banks may face more competition for low-cost funding. If tokenised deposits become a standard settlement asset, banks may need to operate technology and compliance controls continuously rather than around conventional market windows.
At the same time, programmability can reinforce banks that adapt early. Banks already possess customer due diligence, balance sheets, access to payment systems and the ability to create deposits through lending. Those institutional advantages are difficult for technology firms to reproduce. The opportunity is to make those advantages available through programmable interfaces: tokenised deposits, intraday liquidity, smart-contract settlement, custody, identity services and APIs that allow corporate clients to embed banking directly into operational workflows.
The likely competition is therefore not bank versus blockchain. It is between financial institutions that can make regulated money behave like modern software and those that cannot. A bank that keeps deposits on legacy rails while clients move tokenised assets elsewhere may remain financially sound yet lose strategic relevance at the transaction layer.
Fintechs may win the orchestration layer even if banks keep the money
Fintechs and infrastructure providers have a different opportunity. They may not need to issue money at all. If the future is multi-money, value will accrue to platforms that can route transactions among deposits, stablecoins and central-bank settlement assets while abstracting complexity for users. That includes compliance orchestration, wallet infrastructure, tokenisation platforms, identity, foreign-exchange execution, treasury optimisation and cross-chain or cross-ledger connectivity.
This is the same pattern that shaped earlier phases of financial technology: the most valuable position is often not ownership of the underlying asset but control of the interface and workflow. Programmable finance intensifies that effect because software can choose not only where a payment is sent but which form of money, which liquidity source and which settlement path should be used under a given set of conditions.
Regulators face a design problem, not just a licensing problem
Regulation can authorise issuers and impose reserve, capital or governance rules, but programmable money creates deeper design questions. Who is responsible when a smart contract executes the wrong instruction? What happens when code is legally valid on one ledger but settlement is disputed in another jurisdiction? Can compliance checks be automated without creating opaque exclusions? How should emergency liquidity work when transactions settle around the clock? And how much programmability should be permitted in instruments that are supposed to function as money?
The answer is unlikely to be maximal automation. Financial systems need reversibility in some contexts, discretionary intervention in crises, operational fallbacks and clear accountability. Code can automate a rule but cannot resolve every ambiguity in the legal or economic reality behind that rule. The more financial value becomes executable by software, the more important governance around that software becomes.
The counterargument: existing payments are already good enough
There is a credible sceptical case. Many domestic instant-payment systems already provide near-real-time transfers without blockchain or tokenisation. Card networks offer global acceptance. Banks can automate treasury processes using APIs. Some of the language around programmable money risks attaching a new label to functions that conventional databases and payment systems can already perform.
That criticism is strongest in simple retail payments and weakest in complex multi-asset settlement. Tokenisation is not automatically cheaper, and shared ledgers can introduce new governance, cyber and interoperability risks. The burden of proof should therefore be economic: does a programmable architecture reduce reconciliation, liquidity usage, settlement risk or operational cost enough to justify migration?
The projects now moving from prototypes toward real-value testing are important because they begin to answer that question. If the benefits appear only in controlled demonstrations, conventional infrastructure will remain dominant. If real-value experiments show that atomic settlement and embedded workflow logic materially improve treasury and market operations, programmable money will stop being a technology narrative and become an infrastructure decision.
What investors should watch
For investors, the key signal is not the number of CBDC pilots or token announcements. It is whether programmable money starts changing balance sheets and market structure. Evidence would include meaningful migration of transaction deposits, sustained institutional use of tokenised cash for securities settlement, growth in stablecoin payment activity outside crypto trading, lower collateral or liquidity requirements from atomic settlement, and new infrastructure providers gaining recurring revenue from regulated financial institutions.
The winners may be less obvious than the headline currency issuers. Banks with strong transaction franchises, market-infrastructure operators that can bridge old and new rails, custody and compliance providers, and software companies that become the orchestration layer of tokenised finance could capture more durable value than any single token. Conversely, institutions that depend heavily on payment friction, idle balances or closed networks may face pressure if programmability makes money easier to move.
Conclusion: the winning currency may be the one that disappears into the infrastructure
The global race for programmable money is not heading toward a single universal digital currency. It is producing a more complicated monetary system in which central bank money, commercial bank deposits and regulated private tokens may coexist on increasingly programmable infrastructure. The decisive contest is over interoperability, trust, liquidity and control of the transaction layer.
The evidence today is strongest in wholesale markets: central banks and banks are demonstrating atomic settlement, tokenised deposits are moving into real-value pilots, and regulators are constructing frameworks around stablecoins. Retail CBDCs remain more politically and institutionally cautious, with leading projects explicitly rejecting restricted-use money even as they explore conditional payments.
The deeper implication is that programmability may matter most when users no longer notice it. The future of money will not be defined by how often people say “token” or “CBDC”. It will be defined by whether payments, assets, compliance and liquidity can coordinate automatically without weakening the trust that makes money valuable in the first place. The institutions that solve that problem will shape the next monetary architecture — even if the customer still sees nothing more than a balance and a pay button.
References
1. BIS Annual Economic Report 2026, Chapter III: Anchoring trust in money: innovation beyond stablecoins — 23 June 2026.
2. BIS Papers No. 159: Advancing in tandem — results of the 2024 BIS survey on central bank digital currencies and crypto — 22 August 2025.
3. Project Agorá: a shared programmable platform for wholesale cross-border payments — 27 May 2026.
4. ECB: Digital euro — Digital euro project status, accessed 11 August 2026.
5. ECB: FAQs on the digital euro — Updated 5 March 2026; accessed 11 August 2026.
6. Bank of England: Progress update — Digital Pound Design Phase — Published 4 March 2026.
7. Bank of England: Digital Pound Lab — Phase 2 update — Updated 25 June 2026.
8. Bank of England: Policy statement and draft rules on regulating systemic stablecoins — Published June 2026.
9. U.S. Public Law 119-27 — GENIUS Act — Public Law 119-27, signed 18 July 2025.
10. IMF Note: The Rise of Tokenization — Deciphering New Trends in Payments and Asset Tokenization — IMF Notes, 2 July 2026.
11. HKMA briefing to the Legislative Council, 4 May 2026 — Briefing dated 4 May 2026.
12. Central Bank of the UAE: Digital Dirham — A Primer on the UAE’s Central Bank Digital Currency — Policy Paper No. 01/2025, updated 31 July 2025.
13. Bank of Japan: Central Bank Digital Currency — CBDC project page, accessed 11 August 2026.