The fintech infrastructure reshaping global monetary power

The Architecture of Central Bank Digital Currencies

When we talk about Central Bank Digital Currencies, or CBDCs, it is tempting to treat them as just another form of digital money. That framing misses something important. CBDCs are not simply a digitisation of cash; they represent a fundamental redesign of the infrastructure through which sovereign money is issued, distributed, and controlled. Understanding how they work from a technical standpoint is not a niche concern for engineers. It is the prerequisite for understanding the monetary and geopolitical stakes involved.

The starting point for any serious discussion of CBDC architecture is the distinction between the two main types: retail and wholesale. A retail CBDC, often abbreviated as rCBDC, is designed for the general public. Think of it as the digital equivalent of the banknotes in your wallet, accessible to individuals and businesses for everyday transactions. A wholesale CBDC, or wCBDC, is something entirely different. It operates exclusively between financial institutions, functioning as a digital form of central bank reserves used for interbank settlement and large-value transactions. As the Bank for International Settlements explained in its August 2023 executive summary on CBDCs, while retail CBDCs aim to provide a risk-free and digital means of payment for everyday transactions, wholesale CBDCs are designed for use among financial intermediaries and operate like central bank reserves but with added functionalities enabled by tokenisation. These two categories share the same name but serve fundamentally different purposes in the financial system, and their technical architectures reflect that difference.

Within the retail space, most central banks have converged on what is called a two-tier architecture. The logic here is practical: if a central bank were to directly manage the accounts of every citizen, handling onboarding, identity verification, customer service, and day-to-day transactions, it would need to become, in effect, a retail bank. This is widely considered undesirable, both because it would crowd out commercial banks and because it would require central banks to take on operational responsibilities far beyond their traditional mandate. The BIS, in its 2023 FSI executive summary, noted that a direct model may hinderprivate sector innovation and would force the central bank into an undesirable role as financial intermediary. The two-tier model solves this by splitting the work: the central bank issues the CBDC and maintains the wholesale ledger at its core, while commercial banks and licensed payment service providers handle consumer-facing services like onboarding, wallets, compliance checks, and transaction processing. The central bank sits at the top of the chain; end users interact with it only indirectly.

Once the distribution architecture is settled, the next critical design question concerns how ownership of the currency is recorded and transferred. This is where the distinction between token-based and account-based models becomes essential. In an account-based system, the CBDC is recorded as a balance in an account linked to the identity of its holder. When a payment is made, the system verifies the identity of the payer, debits their account, and credits the payee’s account. This is essentially how conventional bank transfers work today; the money does not move as a physical object, what changes is a database entry. In a token-based system, the logic is inverted. The CBDC exists as a digital token, similar in concept to a bearer instrument; whoever holds the token owns the value. Authentication focuses on the validity of the token itself rather than the identity of the holder. As AWS explained in its 2021 technical paper on CBDC design, tokens can operate as bearer instruments without requiring online validation of user identity or CBDC ownership, which is precisely what makes them capable of supporting offline payments.

The practical implications of this choice are significant. Account-based models offer better traceability and are easier to integrate with existing regulatory frameworks, since every transaction is tied to an identity. However, they require constant connectivity: you cannot make a payment if the system cannot verify your account in real time. Token-based models are more privacy-preserving and can function offline, but they introduce new risks, including the potential for double-spending and more complex regulatory oversight. The BIS in its 2024 retail CBDC architecture proposal suggested a hybrid approach, emphasising a modular, token-based model to enhance privacy while also accommodating account-based structures. The goal is to capture the privacy and offline benefits of tokens while retaining the regulatory clarity of account-based verification. A 2025 academic survey published on arXiv, which analysed 26 existing CBDC systems across 135 research papers from 2018 to 2025, found that the most common configuration among live or piloted systems is a two-tier architecture combined with distributed ledger technology and a token-based access model.

The third major architectural decision, and perhaps the most technically contested, is whether to build the underlying ledger on Distributed Ledger Technology, or DLT, versus a conventional centralised database. A centralised ledger is managed by a single authority, typically the central bank or a designated operator. It processes transactions through a traditional database, is generally faster and cheaper to run, and offers the central bank complete control over the system. A DLT-based system, by contrast, distributes record-keeping across multiple nodes, with transactions validated through a consensus mechanism before being recorded. In the CBDC context, the relevant implementations are almost always permissioned, meaning that only authorised participants can operate nodes, rather than the open, public blockchains associated with Bitcoin or Ethereum. According to the IMF’s 2024 survey note on CBDCs and digital payments, there is currently no clear preference for DLT over conventional centralised ledger technology among central banks; both approaches are in active use.

The choice is not purely technical. It reflects priorities. DLT offers immutability, meaning recorded transactions cannot be altered, which enhances auditability and settlement finality. It also enables interoperability between different institutions running nodes on the same network. The downside is that DLT systems are generally more complex and slower than centralised databases. For wholesale CBDCs, DLT is more clearly advantageous. The 2024 BIS survey found that 66% of DLT-based wholesale CBDC projects support multiple digital assets, facilitating tokenised settlements. In a wholesale context, where the participant set is small, trusted, and technically sophisticated, a permissioned DLT system can deliver near-instant atomic settlement, meaning the simultaneous and irrevocable exchange of assets. The Federal Reserve Bank of New York and the Monetary Authority of Singapore demonstrated this in May 2023, when they successfully tested the interoperability of two distinct CBDC networks, achieving end-to-end finality in under 30 seconds.

A useful concrete example of how these architectural choices play out in practice is China’s e-CNY, which remains the most advanced and extensively piloted retail CBDC in the world. As of November 2025, China had recorded 3.48 billion cumulative digital yuan transactions worth approximately 2.37 trillion US dollars. The e-CNY operates on a two-tier model: the People’s Bank of China issues the currency and maintains the core ledger, while commercial banks distribute wallets and manage customer relationships. Its access model has historically blended token and account features, with a tiered anonymity system. Starting January 1, 2026, China implemented a new framework under which the e-CNY transitions to an account-based deposit currency model, with wallets becoming liabilities of commercial banks protected by deposit insurance. This represents a significant architectural evolution and illustrates how even the most advanced CBDC systems are still being actively redesigned as real-world deployment reveals the trade-offs involved.

The broader picture is one of managed complexity. No single architecture has emerged as definitive because there is no universal answer to the trade-offs at stake. More decentralisation through DLT improves resilience and interoperability but introduces latency and governance challenges. Token-based models protect privacy and enable offline use but require more sophisticated mechanisms to prevent fraud. Two-tier distribution preserves the existing banking system but adds layers of intermediation that can reduce efficiency. Central banks are navigating these trade-offs in real time, with most still in research or pilot phases precisely because the decisions are consequential and not easily reversed once a system is live. What is worth appreciating is that these technical choices are not neutral. They encode policy decisions about surveillance, financial inclusion, monetary control, and the distribution of power within the banking system. That is why the infrastructure question sits at the very centre of a much larger debate about the future of monetary sovereignty.

China’s Digital Yuan and Cross-Border Ambitions

China’s e-CNY, or digital yuan, is the most advanced central bank digital currency programme among major economies. Launched as a public pilot in 2020 after six years of internal development, it operates on a two-tier model: the People’s Bank of China issues the currency, while commercial banks handle distribution to end users. This design deliberately preserves the existing banking structure rather than replacing it, addressing early concerns about financial disintermediation. Unlike Bitcoin or stablecoins, the e-CNY carries full legal tender status and is pegged one-to-one with the physical renminbi.

The scale of the pilot is notable. By mid-2024, total transaction volume had reached 7 trillion yuan, approximately 986 billion US dollars, across 17 provincial regions covering sectors from retail and healthcare to education and tourism. Over 225 million personal digital wallets had been opened as of 2025.

To drive adoption domestically, China integrated e-CNY with WeChat Pay and Alipay in 2022 and used high-profile events, including the 2022 Winter Olympics and the 2023 Asian Games, as showcases for the currency. Despite the scale, adoption has not been frictionless: user feedback has pointed to limited merchant acceptance and, until recently, the absence of interest on wallet balances as disincentives. Starting January 1, 2026, the People’s Bank of China addressed the latter by making the e-CNY interest-bearing, transitioning it from a cash-equivalent to a digital deposit model, a first for any CBDC globally.

The domestic rollout is only one dimension of China’s CBDC strategy. The more geopolitically significant development is Project mBridge, a multi-CBDC platform for cross-border payments built on distributed ledger technology. Originally initiated in 2021 through collaboration between the BIS Innovation Hub, the People’s Bank of China, the Hong Kong Monetary Authority, the Bank of Thailand, and the Central Bank of the UAE, mBridge was designed to address longstanding inefficiencies in correspondent banking, namely high costs, slow settlement times, and operational complexity. Saudi Arabia joined as a full participant in 2024, extending the platform’s reach into Gulf commodity markets.

The platform reached minimum viable product stage in mid-2024, at which point the BIS stepped back from the project, framing its exit as a graduation rather than a withdrawal. The BIS General Manager simultaneously distanced the institution from speculation that mBridge could function as a sanctions-evasion tool for certain nations. That concern has not disappeared, however: reports from the Wall Street Journal in November 2025 indicated that the platform had been used by firms to circumvent US sanctions. As of January 2026, mBridge has processed over 4,000 cross-border transactions worth a cumulative 55.5 billion US dollars, with China’s digital yuan accounting for approximately 95% of total settlement volume.

A concrete example of thee-CNY‘s trade application came in October 2023, when the China National Petroleum Corporation completed the settlement of one million barrels of crude oil in digital yuan through the Shanghai Petroleum and Natural Gas Exchange. This transaction, while small relative to total global oil trade, marked a deliberate policy step. CNOOC followed shortly after, settling a 65,000-ton LNG contract in yuan with a Singapore branch of French energy company ENGIE. These transactions fit into a broader shift in China’s trade settlement patterns. By 2024, nearly 30% of China’s foreign trade was settled in yuan, up from roughly 3 to 4% in the early 2010s. The yuan ranked fourth among global payment currencies in 2024 according to SWIFT data, surpassing the Japanese yen and the Canadian dollar.

In September 2025, the People’s Bank of China inaugurated an international operations centre for the e-CNY in Shanghai, describing the move as part of a historical inevitability in global payments. China’s CBDC programme has consistently been presented domestically as a financial infrastructure project, but its cross-border architecture reflects a longer-term strategic intent: to build credible alternatives to dollar-denominated payment rails, even if displacing the dollar outright remains a distant prospect.

The Western Response: Europe and the United States

The Western response to the global CBDC race has been anything but uniform. While China advances a fully operational digital currency and over 130 countries explore their own versions, the two largest Western economies have taken sharply divergent paths, shaped by different institutional mandates, political climates, and strategic priorities.

The European Central Bank has been the most committed Western actor in this space. Work on the digital euro began in 2020 with an initial scoping report, followed by a two-year investigation phase from 2021 to 2023, and a preparation phase that ran from November 2023 to October 2025. The ECB aims to be ready for a potential first issuance of the digital euro during 2029, based on the working assumption that European co-legislators will adopt the necessary regulation in the course of 2026. Should that timeline hold, a pilot exercise and initial transactions could take place as of mid-2027. The ECB’s October 2025 closing report on the preparation phase confirmed the completion of a draft scheme rulebook, the selection of infrastructure providers, and the results of an innovation platform through which around 70 market participants tested features such as conditional payments and offline functionality.

The strategic framing behind the project is explicit. International card schemes accounted for 69% of all card payments in the euro area in the second half of 2024, a dependence on non-European payment infrastructure that ECB officials have consistently cited as a motivation for a sovereign digital alternative. ECB President Christine Lagarde has described the digital euro as integral to promoting Europe’s strategic autonomy. This framing has gained political traction: European heads of state called for swift progress on the legislation at both their March and October 2025 meetings.

Resistance, however, remains. The project has faced pushback from parts of the banking sector over concerns about deposit outflows and implementation costs. The ECB’s own analysis estimates a potential deposit run of up to 699 billion euros under a 3,000 euro holding limit in a stress scenario. Legislative negotiations in the European Parliament’s ECON committee continued into early 2026, with the Parliament’s vote expected in June 2026. A group of over 70 economists, including Thomas Piketty and Paul de Grauwe, published an open letter arguing that a strong public digital euro is the only credible defence against the growing dominance of dollar-backed stablecoins in European payments.

The United States has moved in the opposite direction entirely. On 23 January 2025, President Trump signed an executive order prohibiting federal agencies from issuing, endorsing, or promoting central bank digital currencies, citing risks to financial stability, individual privacy, and US sovereignty. Shortly after, the House passed the Anti-CBDC Surveillance State Act, while Trump signed the GENIUS Act into law, establishing regulatory frameworks for private, dollar-denominated stablecoins. This represented a fundamental policy choice: to reject sovereign digital currency while embracing regulated private alternatives.

The US bet, in other words, is on private infrastructure. Dollar-backed stablecoins like USDT and USDC are positioned to extend dollar reach in digital payments without requiring the Federal Reserve to enter the retail money business. The divergence between Europe and the US creates a structural asymmetry in the global CBDC landscape. Europe is building public monetary infrastructure; the US is outsourcing that function to private companies. Whether stablecoins can substitute effectively for a sovereign digital currency, particularly in cross-border contexts where the absence of a US CBDC weakens interoperability standards, remains an open and consequential question.

Impact on Banks and Fintech Companies

The introduction of CBDCs is often framed as a modernisation of payments, but it is more accurately seen as a reallocation of monetary power between central banks, commercial banks, and a new generation of financial technology firms. For banks, CBDCs introduce a direct threat to the funding model. For fintechs, they create a publicly governed monetary infrastructure on which entirely new competitive models can be built from the ground up.

The central risk for banks is disintermediation. If households move savings into a digital wallet held directly on the central bank’s ledger, those funds effectively leave the commercial banking system. Under the current model, banks collect deposits and transform them into loans, financing households and firms across the economy. Deposits are not a passive liability; they are the core source of stable, low-cost funding that sustains credit creation.

This risk operates on two distinct timescales. Slow disintermediation is structural: households gradually migrate savings toward CBDC in normal conditions, compressing the stable funding base on which credit creation depends. Fast disintermediation is acute: in periods of financial stress, the instant convertibility of deposits into riskless central bank money can accelerate a bank run, transforming a confidence shock into a systemic liquidity crisis before supervisors can intervene. This dual vulnerability shapes the ECB’s design response. A holding limit of 3,000 euros per individual is currently under active legislative discussion; the ECB’s own analysis confirms that such a cap would reduce deposit outflows to less than 2% of retail sight deposits under normal conditions. Alongside the cap, the framework adopts an intermediated model in which individuals access the digital euro through commercial banks and licensed payment service providers, allowing banks to retain the customer relationship and remain responsible for identity verification and regulatory compliance.

Where banks see a structural threat, fintechs see infrastructure. The same intermediated architecture that constrains deposit migration simultaneously defines a new competitive frontier for technology-driven players. Today, payment markets are dominated by a small number of card networks, large banks, and technology platforms. Entry barriers are high and infrastructure remains proprietary. Programmable CBDC infrastructure disrupts this structure at the monetary layer. By embedding conditional logic into currency itself, it unlocks a product territory that legacy networks cannot support economically. Conditional micropayments, real-time supply chain settlement, and targeted government transfers become viable on public monetary rails for the first time. In May 2025, the ECB contracted UK-based firm Fluency to test offline and programmable payment functionalities for the digital euro, using its Aureum platform to enable atomic settlement and real-time interoperability across CBDCs, stablecoins, and tokenised assets.

Central banks are already demonstrating their dependence on specialised technology partners to build the functional layer of their digital currencies. With 66 countries currently in the pilot or active development stage, demand for wallet infrastructure, compliance tooling, and connectivity services is global and growing. These are precisely the services where agile fintechs hold a structural advantage. Unlike traditional institutions, they move faster and integrate into new infrastructure without the burden of legacy systems. Their competitive edge does not disappear with CBDCs; it shifts from payment processing to programmable infrastructure design. The most likely near-term outcome is a hybrid public-private ecosystem, where banks bring regulatory trust and established client relationships, and fintechs bring speed and cost efficiency. Those that establish distribution positions early will access a generation of users through regulated infrastructure that, until now, simply did not exist.

Cross-Border Payments, Dollar Dominance, and Geopolitical Fragmentation

Cross-border payments are financial transactions between parties located in different countries, requiring the transfer of funds across borders and often involving currency conversion. They are essential to the global economy, supporting international trade, investment, tourism, and remittances. 

However, these payments remain complex, as they rely on multiple financial institutions, diverse regulatory frameworks, and various instruments such as wire transfers or electronic payments. This complexity leads to higher costs, slower processing times, and limited transparency compared to domestic transactions. As global integration expands, their volume continues to grow, while inefficiencies persist as a major challenge.

These inefficiencies are closely linked to the dominance of the US dollar, which underpins most cross-border transactions and reinforces its central role in the global financial system. Despite increasing fragmentation in the global financial system, the United States dollar maintains a structurally dominant position. 

According to the International Institute for Strategic Studies, dollar-denominated liquidity remains unmatched: in October 2025, daily average turnover reached 3,811 billion dollars for swaps and 1,151 billion dollars for outright forward contracts. This dominance is supported by the depth of US financial markets and the dollar’s role as a primary safe-haven asset. At the same time, there are signs of gradual change, as the dollar’s share of global foreign exchange reserves has declined from about 70% in 2000 to roughly 58% today. 

This shift reflects a slow diversification toward alternative currencies and increased use of local currencies, such as the Indian rupee and Chinese yuan, in bilateral trade. Although these trends point toward a more multipolar system, current data confirms that the dollar’s central role and its structural privileges remain firmly intact.

This structural dominance is inextricably linked to the technical infrastructure of global finance, specifically the Society for Worldwide Interbank Financial Telecommunication, or SWIFT, which serves as the global backbone for secure payment instructions. Despite its central role, alternative systems are increasingly enabling institutions to bypass this traditional infrastructure. Blockchain-based networks like Ripple allow for transactions to settle in seconds rather than days, while China’s CIPS provides a framework for cross-border payments independent of SWIFT. As SWIFT itself acknowledges, its network must operate in strict compliance with international sanctions, a factor that has driven sanctioned states to actively explore alternative trade channels. These emerging technologies are expanding cross-border capabilities and fostering parallel systems, ultimately reducing global dependence on the established SWIFT messaging framework. The push for these alternatives is not merely an upgrade in efficiency; it is a strategic response to the fact that financial sanctions have become a key tool of geopolitical power.

A clear example is the 2022 exclusion of Russian banks from SWIFT, which significantly disrupted cross-border payments. This shows how payment systems act as chokepoints of geopolitical power. At the same time, sanctions weaken financial connectivity but also encourage the development of alternative infrastructures, potentially reducing their long-term effectiveness. Sanctions demonstrate that control over payment systems is both an economic and geopolitical instrument, while also contributing to global financial fragmentation.

This trend toward fragmentation is further reinforced by the development of multiple CBDCs, as the absence of common standards risks creating a divided global payment landscape. Multi-CBDC fragmentation refers to the risk that various central bank digital currencies, developed independently, will lack interoperability, creating isolated digital islands with incompatible technology, standards, and legal frameworks. According to SWIFT, the current CBDC ecosystem could become globally fragmented, with different central banks adopting different technologies, standards, and protocols. Similarly, the Bank for International Settlements highlights that there is no single model for cross-border CBDC integration, as countries are pursuing diverse designs and policy objectives, making coordination essential but difficult. Without harmonised design, this fragmentation increases cross-border payment costs, creates regulatory gaps, and hinders efficiency. Without effective interoperability mechanisms, the expansion of multi-CBDC systems may not unify global payments but instead deepen divisions, reinforcing the fragmentation already accelerated by sanctions and geopolitical tensions.

Privacy, Cybersecurity, and Financial Inclusion

The implementation of a CBDC creates a fundamental trade-off between state control and individual rights, particularly regarding financial privacy. In the current financial system, private commercial banks and payment applications serve as a buffer that prevents the government from having direct, real-time access to every transaction a citizen makes. A CBDC would eliminate this buffer by establishing a direct link between the state and the individual’s wallet, with all activity stored on a government-controlled central ledger. This shift could enable a system of near-omnipresent surveillance, where financial activity may be monitored or even frozen in real time. Legal precedents such as the third-party doctrine in the United States already allow the government to access bank records without a warrant, and a CBDC could further entrench this dynamic by making the government the primary holder of all financial data. As noted by the Bank for International Settlements, this technology could grant central banks absolute control over the rules governing how money is used.

This centralised architecture also introduces new systemic cybersecurity vulnerabilities that could jeopardise national financial stability. An August 2024 Fintech Note from the International Monetary Fund highlights that the CBDC ecosystem is a highly interconnected network of central banks, commercial banks, and technology providers, which amplifies existing risk exposures. While central banks may hope that CBDCs protect consumers from private-sector crypto risks, they remain vulnerable to cyberwarfare, that is, attacks carried out by organised groups or states aiming to disrupt critical infrastructure. Any major disruption or system-wide outage could trigger severe financial shocks, including liquidity shortages or stress on commercial banks. While CBDCs aim to modernise payment systems, they also create a high-value target that could be exploited to cause widespread economic disruption.

Moreover, financial inclusion is not automatically guaranteed by the introduction of digital currency; in some cases, it may even deepen existing social divides. Proponents argue that CBDCs will help the unbanked, but they still present the risk of technological exclusion. Citizens without access to modern digital devices or reliable internet connections may face exclusion from an increasingly digital economy. The IMF’s CBDC Virtual Handbook notes that, while CBDCs can reduce certain barriers, they are not a silver bullet, as they still encounter fundamental challenges such as digital illiteracy and infrastructure gaps in remote areas. Disparities in education and digital skills may advantage wealthier populations, leaving more vulnerable groups behind.

Finally, the complexity of these systems requires global regulation and international cooperation to prevent economic instability. The absence of shared international standards may create a legal vacuum, leading to jurisdictional conflicts between countries. While both retail and wholesale CBDCs are advancing, they raise unresolved questions about the legal nature of digital currency and regulatory authority over service providers. Cross-border spillovers and the lack of a global framework for anti-money laundering remain significant challenges. Without coordinated efforts to establish common standards and share knowledge, evolving national regulatory frameworks may create legal uncertainty, discouraging investment and slowing global economic growth.

What Comes Next

The CBDC landscape in 2026 remains in a period of active development. Decisions regarding ledger design, access models, and the appropriate balance between privacy and traceability are still being determined, and they extend beyond technical considerations. They require societies to establish what they expect from monetary systems: how much visibility the state should have into financial activity, and whether digital currency can expand financial access in practice or will largely reflect existing structural inequalities.

The divergence between China, the eurozone, and the United States demonstrates that there is no agreed answer to these questions. China has prioritised scale and international reach, accepting considerable trade-offs in terms of state oversight of transactions. The European Union is working to reconcile monetary sovereignty with banking system stability, though its implementation timeline depends on political consensus that has not yet been fully achieved. The United States has largely left the question to private sector development, a position whose broader consequences will become more apparent as cross-border CBDC frameworks continue to take shape.

The decisions currently being made across central banks, legislative bodies, and international standards organisations will have long-term effects on payment infrastructure and will be difficult to modify once embedded. CBDCs are not incremental adjustments to existing systems. They reflect specific assumptions about the relationship between individuals, financial institutions, and the state. For this reason, the current phase of design and negotiation carries significant implications for how the international financial system develops over the coming decades.

Written by:

• Domenico Agostino
• Keti Aznaurashvili
• Nicolò Ballabio
• Claudia Cristofolini
• Giorgio Signorile
• Matilde Sommese