EU reluctance on stablecoins may backfire as US-backed dollar tokens spread, pushing demand offshore and importing dollar risks into Europe.

Stablecoins – a form of tokenised money issued primarily by non-banks – promise efficiency gains in payments but also bring new risks to financial stability and monetary control. Because of this, European Union policymakers have so far discouraged the development of stablecoins, instead favouring tokenised deposits issued by banks.  

This approach might well make sense in a one-jurisdiction world, but risks being counterproductive in the context of dollar-denominated stablecoins encouraged by the United States authorities. EU policy could push European demand for stablecoins outside the EU’s regulatory perimeter, leaving EU holders of dollar stablecoins exposed to US currency risks and US fiscal risks. The EU would also risk adoption by the digital settlement architecture of European financial markets of dollar-based conventions, and settlement activity may migrate away from the Eurosystem’s payments infrastructure in tokenised segments.

A strategy to contain these risks without a regulatory race to the bottom should include:

• Acceleration of the European Central Bank’s Appia project to establish interoperability between digital ledger technology platforms and ECB payments infrastructure;
• Dispensing with the regulatory requirement to hold a large share of stablecoin reserves in the form of bank deposits; 
• Allowing EU stablecoin issuers to remunerate stablecoin holders directly, as long as the remuneration rate is below the rate on ECB reserves and standard deposit rates; 
• Giving EU-regulated stablecoin issuers access to the ECB’s balance sheet, including to lending-in-last-resort facilities.

This paper was written for the May 2026 informal meeting of European Union finance ministers and central bank governors in Nicosia. We thank Ulrich Bindseil, Hans Geeroms, Konstantinos Kaniouras, Peter Kerstens, Mahmood Pradhan and participants at the Bruegel research meeting for their comments and suggestions

1 Introduction

Tokenisation – the digital representation of financial claims on programmable infrastructures – is transforming financial markets. Although many assets will be affected, the impact of tokenisation on money is the most immediate concern to regulators and central banks. Depending on which path it takes, the tokenisation of money could radically alter the system in which money is created, distributed and regulated, and via which monetary policy is transmitted. This ushers in new risks, not least to the sovereignty of currencies. These risks apply particularly to stablecoins: tokenised money backed by liquid pools of reserves and redeemable at par that is typically issued by private entities outside the regulated banking system. 

Based partly on the 2023 Markets in Crypto Assets regulation (MiCA, Regulation (EU) 2023/1114), the European Union has established a clear policy direction that favours the development of private-sector digital money within the architecture of the existing monetary system (ie tokenised deposits and stablecoins issued by banks), while discouraging the development of stablecoins outside that architecture. The European Central Bank has embraced the technological possibilities offered by digital money, while also warning of the risks to financial stability and monetary policy transmission created specifically by stablecoins (Lagarde, 2026).

The United States has made a different choice. While it has much in common with MiCA, the 2025 US GENIUS Act¹ encourages the development of US dollar stablecoins. US Treasury Secretary Scott Bessent has made the strategic intent clear: “We are going to keep the US the dominant reserve currency in the world, and we will use stablecoins to do that”².  

In this context, this paper makes two main points.
First, although EU policymakers want to protect the architecture of the euro-area financial system and maintain the sovereignty of the euro, they may unintentionally achieve the opposite. By discouraging the development of stablecoins, the EU could push demand to the US. This could expose EU holders to dollar currency risk and the EU to the risk of ‘infrastructure dollarisation’ (section 3.2.2). Europe could be left without a competitive distributed ledger technology (DLT) payment infrastructure – the technological backbone of tokenised finance³ – anchored in EU public money, during the critical period when global norms are established for tokenised payments.
Second, this problem can be fixed by supporting the development of euro-dominated, EU-regulated stablecoins, while ensuring their safety and preserving the role of central-bank money in the euro area’s financial architecture. This requires:

• Ensuring public settlement for tokenised markets, including by accelerating the European Central Bank’s Appia project to establish interoperability between DLT platforms and ECB payments infrastructure⁴;
• Measures to increase the liquidity of euro stablecoins on secondary markets;
• Removing MiCA requirements to hold a large share of stablecoin reserves in the form of bank deposits;
• Allowing EU stablecoin issuers to remunerate stablecoin holders directly as long as the remuneration rate is below the rate on ECB reserves and standard deposit rates;
• Giving EU-regulated stablecoin issuers access to the ECB’s balance sheet, including to lending-in-last-resort facilities.

Section 2 describes recent developments in tokenised money and compares EU and US rules. Section 3 explains why regulation that works well in a single-jurisdiction world may backfire in a multi-jurisdiction setting, and describes the risks of infrastructure dollarisation. Section 4 sets out policy recommendations and addresses concerns that the growth of EU stablecoins would divert deposits from banks to stablecoins. Section 5 concludes.

2 Tokenised money and its regulation

The rapid emergence of tokenised finance alongside advances in DLT and programmable financial infrastructures have enabled new ways of representing money. In this section we describe the two main forms of private digital money: tokenised deposits and payment stablecoins.

2.1 Tokenised deposits

Tokenised deposits, or digital representations of commercial bank deposits issued on DLT, remain at an early stage of development, largely limited to pilot projects and controlled experiments. These initiatives demonstrate technical feasibility but remain confined to permissioned networks and selected institutional participants.

One prominent initiative is Project Agora, led by the BIS Innovation Hub in cooperation with central and commercial banks⁵. The project explores how a unified ledger could combine on a single platform tokenised commercial bank money for payments with tokenised central bank money for settlement. Prototype development is expected to be completed in the first half of 2026.

Within the European Economic Area, the European Banking Authority (EBA, 2024) identifies three tokenised deposit initiatives, of which only one has reached operational stage. Beyond the EEA, the Regulated Liability Network (RLN) initiative, led by UK Finance, is testing a shared ledger able to support regulated digital money including commercial bank deposits, central bank money, electronic money and crypto-assets, and has assessed alternative models for tokenising bank deposits⁶.

Mastercard’s Multi-Token Network (MTN) is working on a blockchain-based infrastructure that would support tokenised deposits, stablecoins and central bank digital currencies⁷. J.P. Morgan, through its Kinexys platform and JPM Coin, has used tokenised deposit-like instruments for wholesale payments and intragroup transfers, including cross-border transactions and tokenised collateral settlement⁸.

Despite growing experimentation, tokenised deposits face structural challenges.

First, traditional bank deposits are fungible within each bank’s balance sheet and, through the payment system, across banks; but tokenised deposits issued by different banks on separate or permissioned ledgers may not be automatically interchangeable, potentially fragmenting liquidity unless interoperability mechanisms or shared infrastructures are developed. This problem, which arises from the absence of a common settlement anchor across DLT-based systems, is common to all digital assets including stablecoins (Shin, 2026). 

Second, scaling tokenised deposits requires costly and operationally complex integration with current core banking systems, payment infrastructures and liquidity management frameworks.

Third, banks must weigh the benefits of programmability and so-called ‘atomic’ settlement (simultaneous and instantaneous delivery of an asset and payment for that asset) against regulatory constraints, capital treatment and potential balance-sheet implications. Unlike stablecoins, tokenised deposits are embedded in prudential regulation, which limits flexibility but provides greater systemic safeguards.

Tokenised deposits are thus an incremental technological evolution of commercial bank money rather than a disruptive alternative. Broader adoption will depend on whether interoperability, standardisation and liquidity coordination challenges can be resolved.

Stablecoins have existed since 2014 but took off starting in 2020 (Figure 1a)⁹. Over 2024 and 2025, the volume of the top two stablecoins – Tether’s USDT and Circle’s USDC – increased by 126 percent. USDT and USDC are mainly backed by short-term US Treasury securities, reverse repurchase agreements and cash or bank deposits (Figure 1b). Only USDC complies at time of writing with the US GENIUS Act and MiCA (see below). By contrast, euro-denominated stablecoins, such as EURC and STASIS’s EURS, are backed almost entirely by cash and bank deposits.

Figure 1: Top seven stablecoins

^{Source: Bruegel based on CoinGecko (left chart) and transparency reports of Tether and Circle (right chart). Note: Top seven stablecoins in terms of current volume outstanding: USDC (USD Coin), USDT (Tether), USDS (Stably USD), USDe (Ethena USDe), USD1 (First Digital USD), DAI (MakerDAO Dai) and PYUSD (PayPal USD). Asset reserves as of 31 December 2022 and 31 December 2025.}

Stablecoins are finding increasing use in more mainstream applications. Stablecoin transaction volumes adjusted to exclude bot-driven transactions have grown at a compound annual rate of 133 percent since 2023, reaching approximately $28 trillion in 2025¹⁰. Retail-scale transactions (below $250 and excluding bot activity and high-frequency trading) increased from around $0.5 billion in 2019 to $69.13 billion in 2025, indicating growing use in smaller-value, consumer and business payments (Figure 2). Identifiable stablecoin payment volumes including larger-value transactions reached approximately $390 billion in 2025, more than doubling from 2024 levels. Stablecoin-linked card spending is estimated to have grown to $4.5 billion in 2025, up 673 percent from 2024¹¹. Visa has said it will expand stablecoin settlement support¹² and reports that stablecoin spending on Visa-linked cards surpassed crypto purchases in the second half of 2025.

Figure 2: Total retail-scale stablecoin transactions ($ millions)

^{Source: Bruegel based on Visa Onchain Analytics. Note: the 2026 figure includes data up to 21 April 2026.}

Stablecoin issuance and transactions are overwhelmingly dollar based (Figure 3). The largest euro-denominated stablecoin – Circle’s EURC – ranks only twentieth, and euro-denominated stablecoins account for just 0.3 percent of total supply. Measured by transactions, however, Europe does much better. Europe-based stablecoin transactions made up 38 percent of global transactions in the final quarter of 2025 (Figure 3). Given that almost the entire stablecoin market is denominated in dollars, the growing use of stablecoins in Europe indicates that an increasing share of crypto-asset transactions associated with Europe is conducted in tokenised dollars, rather than euros.

Figure 3: Dominance of dollar-denominated stablecoins and transactions

^{Source: Bruegel based on Artemis. Note: Figure 3b is based on a time-zone analysis to determine the regional breakdown.}

2.3 Regulation: Europe versus the US

Although the EU and US regulatory frameworks for stablecoins have much in common, there are also major differences in their scope and content. 

2.3.1 Scope 

For the EU, MiCA provides a full-spectrum regulatory regime covering three categories of crypto-assets: e-money tokens (EMTs), which are pegged to a single official currency, backed by assets in that currency and redeemable at par; asset-referenced tokens (ARTs), which are backed by pools of assets and redeemable at market value; and other crypto-assets that are not pegged to or backed by any other asset or value. MiCA also regulates crypto-asset service providers (CASPs), such as exchanges, custodians and portfolio managers.

The GENIUS Act only regulates ‘payment stablecoins’ – the equivalent of MiCA’s EMTs, defined as digital assets issued for payment or settlement and redeemable at a predetermined fixed amount, such as $1. Stablecoins defined as securities or other financial products issued by registered investment companies remain outside its scope, as are CASPs. The latter will be governed mainly by the Digital Asset Market Clarity Act that is at time of writing making its way through Congress¹³, or rely on individual state laws and oversight from Federal regulations  such as the Securities and Exchange Commission and the Commodity Futures Trading Commission.

2.3.2 Stablecoins

Despite their different scopes, both frameworks converge on several core principles for payment stablecoins/EMTs, while diverging on others.

On both sides of the Atlantic, regulated stablecoin issuers must hold reserves one-for-one against all stablecoins in circulation, held in a bankruptcy-protected structure. Both GENIUS and MiCA give payment stablecoin holders the right of redemption at par. Both frameworks ban interest-bearing stablecoins. Both require compliance with anti-money laundering and countering the financing of terrorism rules. Both have a tiered supervision model, with smaller issuers supervised locally and larger ones graduating to federal/pan-European oversight.

In terms of differences (Table 1), MiCA’s EMT regulation is generally more prescriptive ex ante and is friendlier to banks (Bindseil, 2026): 

• Capital requirements are imposed ex ante under MiCA but left to the responsible authorities under GENIUS;
• While both GENIUS and MiCA seek to set high quality and liquidity standards for reserve composition, only GENIUS prohibits issuers from holding longer-maturity bonds, while MiCA requires issuers to hold a large share of their reserves – either 30 percent or (for ‘significant’ issuers) 60 percent – in the form of bank deposits.
• Only MiCA has transaction caps. MiCA allows a maximum of 1 million transactions or €200 million daily on non-euro denominated EMTs used as a means of payment within the EU. This is a deliberate mechanism to protect euro monetary sovereignty and prevent dollarisation (in section 3, we explain why we think that it is insufficient). GENIUS has no equivalent, probably because US regulators do not consider the use of non-dollar stablecoins to be a risk to the US.
• Redemption fees. MiCA prohibits redemption fees except as part of a recovery plan approved by a competent authority, whereas GENIUS allows fees subject to disclosure requirements.
• Remuneration. While both MiCA and GENIUS prohibit issuers from paying remuneration, only MiCA extends this prohibition to crypto exchanges, broker-dealers and other CASPs. This creates a potential loophole through which holders of GENIUS-compliant stablecoins might receive remuneration.

Table 1: Comparison of EMT/payment stablecoin rules, EU, US and UK

	EU regulation14	US GENIUS Act	UK proposal15
Capital requirements	Highest of €350,000, 2% of reserve assets or a quarter of the fixed overheads of the preceding year (3% for systemic stablecoins)	To be specified by the responsible authority	Highest of £350,000, a quarter of the fixed overheads of the preceding year, or 2% of the value of stablecoin in issuance. For systemic stablecoins, principles for financial market approach
Liquidity requirements	High quality liquid assets, 30% (60% for issuers of significant stablecoins) minimum deposit in credit institutions16, and concentration limits	Cash, demand deposits, money standing to the credit with a Federal Reserve Bank, short term Treasury bills, and Treasury reverse repos	‘Core backing assets’ which include short-term deposits and short-term governmentdebt17 with a minimum of 5% in on-demand bank deposits18. For systemic stablecoins: 40% minimum deposits at the Bank of England withup to 60% short-term sterling-denominated UK government debt securities
Allowing commercial banks to issue stablecoins	Yes	No, only via subsidiaries (no direct issuance)	Implicitly allowed under banking regulation
Require issuers to be domestic legal entities	Yes	No	No
Access to central bank lender of last resort	No	No	No
Access to central bank deposits	EU regulations do not exclude access (but ECB rejected it)	An option; remuneration not yet specified.	Preferred; no renumeration
Redemption obligations	At par – no fees; No explicit definition of ‘timely’	At par – fees possible but need to be justified and transparent; The Office of the Comptroller of the Currency to issue regulations clarifying elements of redemption policies	At par – redemption by the end of next business day and fees allowed but limited to cost recovery; For systemic stablecoins: redemption by the end of the day in real time where possible and fees proportionate to costs incurred
Limits on holdings	No	No	For systemic stablecoins: £20,000 for individual holdings, £10 million for businesses. But expected to loosen and ultimately be removed
Transaction cap	For non-euro denominated stablecoins: 1 million transactions or €200 million	No	No
Explicit higher requirements on issuers deemed systemically important	Yes	No – regulators impose prudential requirements on a proportionate basis.	Yes
Prohibiting direct renumeration	Yes, broad prohibition covering direct and indirect incentives	Yes, but limited to issuers; indirect remuneration via third parties is not clearly prohibited	Yes

^{Source: Bruegel based on Bindseil (2026), Adrian (2025) and legal texts.}

2.3.3 Tokenised deposits

EU and US rules differ most over tokenised deposits, the most consequential area for the future of DLT-based payments.

MiCA excludes tokenised deposits from its scope. MiCA does not apply to crypto-assets that qualify as deposits under EU banking law (MiCA Article 2.4.c). Tokenised deposits are bank liabilities – they remain deposits in their legal character regardless of the technology used to represent them – and therefore fall under EU banking regulation rather than MiCA. The EU is thus channelling DLT-based bank money innovation through current banking supervision, not through the new crypto framework. This is one reason why EU policymakers – particularly the ECB and the Eurosystem – are more comfortable with tokenised deposits than stablecoins (Lagarde, 2026). Unlike stablecoins, tokenised deposits sit within the familiar, bank-centric regulatory perimeter and do not threaten the two-tier monetary system.

The GENIUS Act takes a different approach. It permits banks to hold stablecoins and reserves, use blockchains and issue tokenised deposits. The Federal Deposit Insurance Corporation (FDIC) is consulting (as of April 2026) on proposed rules that would clarify the treatment of tokenised deposits under the Federal Deposit Insurance Act, recognising tokenised deposits as a distinct and regulated instrument rather than leaving them in a regulatory grey area. The US framework hence accommodates stablecoin-native and bank-native forms of DLT money simultaneously, without privileging one over the other.
 

3 The threat from tokenisation to the European financial system

Compared to conventional finance, tokenised finance could lead to large efficiency gains linked to the automation of execution and compliance, atomic settlement and continuous operation across time zones, which reduces the cost of cross-border transactions. Tokenised finance could also support capital market integration in Europe (Box 1).

There is also a long list of risks. For the purposes of this paper, it makes sense to classify these into two groups: (1) risks that would arise even in a one-jurisdiction world (that is, if the EU were a closed economy); (2) risks linked to the global nature of digital finance.

3.1 Risks of tokenised finance: the single-jurisdiction case

Discussion of tokenised finance has overwhelmingly focused on risks that would arise even in a world consisting of a single central bank and a single regulator. Because these risks have been covered extensively in the literature and are mostly well understood (Arner et al, 2020; Ho et al, 2022; FSB, 2024; Adrian et al, 2025; Aquilina et al, 2025; ESRB (2025); Li et al, 2026; Gross and Senner, 2026; Altavilla et al, 2026), we only summarise them briefly. They fall broadly into two classes:

First, risks related to digital infrastructure and blockchain technology. These include cybertheft (including by future hackers benefiting from advanced AI tools and/ or quantum computing), code risk and smart contract failures, governance failures and risks related to platform concentration. A related risk is liquidity fragmentation: participants may be forced to hold liquidity buffers on multiple platforms, increasing costs and potentially amplifying instability (Shin, 2026).

Box 1: Benefits of tokenised money* 

Efficiency gains associated with DLT

• Automation and less operational friction. Parts of execution and compliance can be automated via smart contracts, reducing reconciliation across multiple ledgers and intermediaries when conditions can be verified on-chain. In traditional finance, asset transfers often involve multiple custodians, clearing entities and settlement sys-tems, each requiring its own record-keeping and verification processes. Tokenised infrastructures can consolidate these functions, reducing administrative complexity and potentially improving transparency.
• Atomic settlement. DLT platforms can enable atomic delivery-versus-payment (DvP), integrating asset and cash legs in a single transaction. This can shorten settlement chains and reduce counterparty exposures, meaning less need for liquidity buffers and collateral.
• Continuous processing and cross-border coordination. Tokenised infrastructures can operate continuously and across time zones. Large gains are likely in settings where legacy cross-border settlement remains slow, fragmented or costly.

Efficiency gains specific to private digital money

The efficiency gains from tokenised settlement depend not only on the technology, but also on the type of digital money used for settlement:

• Tokenised deposits extend familiar deposit money into programmable and interoper-able settlement environments. Tokenised deposits allow banks and market infrastruc-tures to integrate tokenised securities settlement with cash management, liquidity provision, compliance processes and client relationships. Tokenised deposits can therefore modernise the payments ‘rail’ of European banking, while preserving conti-nuity with the current institutional structure.
• Stablecoins are designed to be platform-neutral and easily transferable across networks, often operating on public or widely interoperable DLT infrastructures. This makes them attractive as a common settlement asset in multi-platform tokenised markets, where participants require a single instrument that can move seamlessly across venues, wallets and applications. Stablecoins can function as a liquidity layer that is less tied to national banking structures and more compatible with integrated, continuously operating digital market infrastructures.

Fostering EU capital market integration – up to a point

Euro-area payment systems are already highly integrated but capital markets remain fragmented, with intermediation and collateral pools segmented across jurisdictions. Standardised digital representations of securities and programmable settlement infrastructure may reduce some of these frictions. Tokenised assets circulating across interoperable platforms with integrated settlement could make cross-border participation operationally simpler. Atomic DvP and interoperable collateral management could support more fluid cross-border repo markets and securities financing activity within a common euro-denominated framework.

As a result, both tokenised deposits and stablecoins could potentially contribute to euro-area capital market integration. However, their effectiveness depends on overcoming the inherent fragmentation across blockchains (Shin, 2026). Interoperability solutions, such as cross-blockchain bridges or shared settlement infrastructures, can mitigate fragmentation but introduce new layers of technical and financial risk. Consequently, integration in tokenised markets is ultimately dependent on the design of safe and scalable interoperability mechanisms.

^{* This box draws on BIS (2023, 2025), Adrian et al (2025), OeNB and BCG (2025) and Pinna and Ruttenberg (2016).}

Second, risks specific to tokenised money, particularly stablecoins (as tokenised deposits remain part of the regulated banking system). These include:

• Risks to the stablecoin holder and to financial stability: stablecoins may not turn out to be money after all, in the sense that they are not always redeemable at par and their value deviates from parity in the secondary market. This might be the case, for example, if there are significant differences in the quality and liquidity of the reserves across issuers. Even when stablecoins are fully backed by high-quality reserves, mass redemptions may lead to a fire sale and collapse in value of reserve assets, undermining the promise to redeem at par.
• Risks to monetary control: stablecoins are analogous to money market mutual funds, which take cash and deposits and invest them in very short-term instruments. If this allows the banking system to finance itself at lower interest rates than the rate at which reserves held at the central bank are remunerated, the central bank would lose its role as the main instrument of monetary policy.
• Risks to financial intermediation: these changes may lead to a shift of deposits from banks, which can create credit, to stablecoins, which cannot, potentially hurting the real economy. We return to this risk in section 4.2.

3.2 Risks of tokenised finance to the EU in a multiple-jurisdiction world

Regulations might effectively mitigate the risks in a single jurisdiction. But tokenised finance is not bound to any single jurisdiction. This leads to additional risks.

3.2.1 The multi-issuer problem

One example is linked to ‘multi-issuer stablecoins’: identical stablecoins issued by different entities, presumably subject to the same controlling interest, in several jurisdictions. This could lead to two problems (Portes, 2025; ESRB, 2025a, 2025b; Lagarde 2025). First, the combined reserve pool is divided into several local pools, each of which may be insufficient because redemption claims may exceed locally available reserves. Second, regulatory arbitrage may produce a run on one of the local pools. For example, the EU prohibits redemption fees while the US does not (Table 1). In a stress situation, this could trigger a run on the reserves of the EU issuer, which would be unable to honour the requirement to redeem the stablecoin at par.

The European Commission has pushed back against these arguments on the grounds that MiCA’s redemption requirement applies only to coins minted by EU issuers¹⁹. However, an EU issuer may not be able to recognise if redemption claims are inflated by coins issued elsewhere, and it is unclear whether a US parent company would come to the rescue of an EU issuer facing a flood of redemption requests.

Regardless of how one views the magnitude of the potential problem, it can be solved. A simple solution, proposed by the ESRB (2025a), would be to ban stablecoins issued in multiple jurisdictions, meaning EU issuers would be prohibited from issuing coins that are fungible with those issued in other jurisdictions. To the extent that this solution is too radical – perhaps because of a public interest in maintaining the efficiency benefits of fungible coins minted by EU issuers – the ESRB (2025a, 2025b) offers a battery of recommendations to reduce the risks associated with multi-issuer stablecoins.

3.2.2 The limits of single-jurisdiction regulation

However, these solutions do not address a second, potentially even more damaging problem that could arise in a multi-jurisdictional setting. Although MiCA has very broad scope, EU residents retain the option to hold coins issued in (and regulated by) other jurisdictions. If EU regulation does not allow EU demand for stablecoins to be satisfied by EU-regulated coins, it will push this demand offshore. EU-regulated coins may be safe, but they may also become irrelevant. The alternative – stablecoins held offshore by EU residents – may expose the EU to even greater risks.

It might be objected that MiCA’s broad jurisdictional scope should prevent this outcome. It does not. MiCA’s reach depends on the presence of an EU-licensed intermediary – a crypto-asset service provider (CASP). Where such an intermediary exists, MiCA’s controls are effective: any stablecoin offered to the public in the EU or admitted to trading on an EU platform must comply, regardless of where its issuer is located. This is why Tether, which has not sought MiCA authorisation, has been progressively delisted by European exchanges from late 2024²⁰, while Circle has been able to maintain access for USDC by obtaining an Electronic Money Institution licence in France.

But EU-licensed CASPs are not the only access route to stablecoins. A European user can access a non-EU exchange through a self-hosted wallet or execute a peer-to-peer transaction directly on-chain. Such transactions are technically straightforward and increasingly common.

A related constraint shapes MiCA’s much-discussed transaction caps on non-euro stablecoins (Art. 23: 1 million transactions or €200 million per day). These caps apply only to transactions within the euro area, and only to payment transactions, not to investment or trading flows. Crucially, they bind issuers only to the extent that transactions can be observed. CASP-intermediated flows are visible and countable; peer-to-peer and self-hosted-wallet flows are not. Issuers are required to estimate them only on a ‘best-efforts’ basis using blockchain analytics. The cap is therefore a meaningful constraint on the regulated payments economy, and a much weaker constraint on the decentralised one.

The implication is that MiCA’s scope, however broadly drawn, depends on a CASP chokepoint that becomes less binding the more stablecoins circulate outside it. EU residents who want dollar stablecoins, particularly for purposes beyond intra-EU retail payments, can obtain them. The growth of European stablecoin activity outside the EU regulatory perimeter (visible in Figure 3) confirms that this is already happening.

3.2.3 The threat of ‘infrastructure dollarisation’ in the EU

A well-known fact of monetary economics is that the usefulness of a means of exchange depends on how widely it is accepted (Menger, 1892). Liquidity therefore concentrates where liquidity already exists. Once a particular asset becomes widely used as the principal means of settlement, or the principal asset for collateral or margining, then platforms, market makers and risk-management systems align around it. Technical standards, interoperability protocols and governance practices tend to follow the dominant liquidity instrument.

If US dollar stablecoins become the default settlement asset for tokenised securities, cross-platform liquidity and decentralised finance applications connected to Europe, digital infrastructures may gradually align around dollar-based conventions. Even euro-denominated securities could rely operationally on dollar collateral and dollar settlement within tokenised environments.

This process would not threaten the euro as legal tender or unit of account. Nor would it expose EU stablecoin holders to immediate risks other than currency risk, insofar as dollar stablecoins are well regulated by GENIUS and the reserve assets pools prescribed by GENIUS remain high quality. However, it would shift the operational centre of gravity of liquidity coordination toward dollar-based instruments. Such a shift would be difficult to reverse once embedded and would carry much broader risks.

We call this dynamic ‘infrastructure dollarisation’: not visible currency substitution in household deposits, but the embedding of foreign currency private money in the settlement architecture of capital markets. If tokenised capital markets increasingly rely on dollar stablecoins for delivery-versus-payment, margining and collateral transfers, then settlement activity could migrate away from Eurosystem infrastructures in the segments of the market that become tokenised. The euro would remain the domestic monetary anchor, but its operational centrality in digital market infrastructures would lessen. Over time, the default programmable liquidity layer in tokenised finance might not be euro central bank money or euro bank deposits, but dollar-based private instruments.

If this happens, European wholesale and retail financial markets would become critically dependent on instruments regulated outside the EU. Disruption affecting major US stablecoin issuers could transmit to European markets not only through price channels, but through settlement and collateral channels. The source of disruption could include the quality of the collateral held in the reserve pools of US stablecoin issuers. This could, in turn, create an additional channel for transmission of US fiscal stress to Europe.

4 A hybrid architecture for safe tokenised money in Europe

4.1 Core design principles²¹

In defining their policy plan, European regulators should start by agreeing on the ideal architecture for tokenised finance in the long run. The goal should be to preserve the efficiency gains of programmable finance while preventing prudential instability and infrastructural dollarisation.

To achieve this, policymakers must ensure that private stablecoins do not become de-facto settlement anchors detached from the public monetary framework. But they should also avoid giving a particular means of payment a structural advantage. The goal should be a hybrid architecture in which both tokenised deposits and euro-denominated stablecoins can operate within a common, prudentially anchored framework. Competition should take place at the level of service provision and technological efficiency, not at the level of access to the monetary anchor, or through regulatory or supervisory discrimination.

Europe’s digital monetary architecture should therefore rest on four principles:

1. Public money remains the final settlement anchor. Central bank money must remain the ultimate means of settlement in wholesale markets, preserving monetary sovereignty, singleness of money and crisis elasticity.
2. Prudentially supervised institutions continue to provide credit intermediation. Digital innovation should not undermine the regulatory perimeter that governs maturity transformation and credit creation.
3. Interoperability across digital platforms is ensured. Fragmented digital silos would replicate existing segmentation problems and weaken liquidity integration.
4. Euro-denominated liquidity is available wherever digital markets coordinate. The euro must be present not only at the core of regulated settlement systems, but also in programmable environments where market liquidity forms.

This model would prevent operational centrality from migrating entirely to private infrastructures. By making central bank settlement interoperable and usable within tokenised markets, the model reduces the likelihood that private stablecoins become entrenched as default settlement assets.

It would also address the risk of infrastructural dollarisation. If euro public settlement mechanisms and euro-denominated liquidity instruments are present and interoperable across platforms, markets need not coordinate around dollar-based instruments for operational convenience.

4.2 Policy priorities

The policy task in implementing the model is to align infrastructure development, sequencing and regulatory calibration so that tokenised finance evolves within a euro-anchored hybrid architecture. If euro-denominated options are not available at scale early on, coordination could default to foreign-currency instruments. Timing is critical because, once a particular settlement instrument becomes embedded, network effects reinforce its dominance, making subsequent rebalancing difficult.

Policymakers should work on four priorities, detailed in sections 4.2.1 to 4.2.4:

1. Ensuring access to public settlement for tokenised markets, including through interoperable links between DLT platforms and central bank infrastructures;
2. Supporting the development of euro-denominated stablecoins, including through adequate liquidity conditions;
3. Enabling the scalability and interoperability of both tokenised deposits and stablecoins, so bank money can function effectively across platforms;
4. Providing structured access to central bank reserves and lender of last resort facilities for euro-denominated stablecoins, to ensure their safety and economic viability.

Policy should act simultaneously on public settlement, bank-based tokenised money and euro-denominated stablecoins. If interoperable central bank settlement mechanisms become widely usable, and if European banks develop scalable and interoperable tokenised deposit frameworks, stablecoins are likely to remain complementary. They would serve cross-border or platform-based niches without displacing settlement in central bank money.

4.2.1 Ensuring public settlement for tokenised markets

The most important structural priority is to preserve central bank money as the default settlement anchor in digital financial markets.

Tokenised securities platforms require both delivery to the buyer of an asset (the asset ‘leg’) and payment for the asset (payment leg) in a transaction to happen within a programmable environment to achieve atomic delivery-versus-payment (see section 2.1 and Box 1). If central bank money cannot interface effectively with distributed ledger platforms, market participants will rationally rely on private on-chain settlement assets that can.

The ECB Appia project (see section 1) to enable interoperability between DLT platforms and TARGET services – or financial market infrastructure services, including synchronisation models and potential tokenised reserve mechanisms – is therefore vital²². The objective is operational usability: market participants must be able to settle tokenised assets in euro central bank money without friction or artificial constraints. For this reason, this project should be prioritised, with a particular focus on the part of Appia involving issuance of tokenised reserves directly on DLT – conceptually close to a wholesale central bank digital currency (CBDC).

4.2.2 Supporting the development of euro-denominated stablecoins 

Making tokenised deposits fully interoperable and integrated with public settlement rails will reduce the need for stablecoins as wholesale settlement instruments. But this outcome cannot be assumed. Historically, EU banks have not been strong first movers in digital innovation, particularly when coordination across institutions is required and private incentives to standardise are weak. Even with interoperable public settlement and tokenised deposits, digital markets may coordinate around the most liquid programmable instrument available: dollar stablecoins. Preventing this does not require foreign instruments to be suppressed, but it does require euro-denominated liquidity to reach sufficient scale and usability.

Stablecoins have a different logic to tokenised deposits. They are not embedded in the existing monetary hierarchy and do not benefit from an intrinsic link to central bank money. Their usability instead relies on a ‘promise of stability’, supported by reserves, redemption rights and regulatory frameworks. This is indeed the objective of MiCA. At the same time, stablecoins are characterised by strong network effects. Their value increases with the scale of adoption, liquidity and integration across platforms. This is where euro-denominated stablecoins and existing regulation fall short.

Market participants use dollar stablecoins because others already do. Dollar-denominated instruments dominate because crypto derivatives markets – in particular perpetual futures, or leveraged contracts without expiration that require continuous collateral and funding – are margined and settled in the top two stablecoins, USDT or USDC (see section 2.2). Funding rates are calculated in dollars and options markets are structured around dollar liquidity pools. Network effects reinforce this equilibrium. Breaking out of it requires policy, infrastructure and incentives.

Policy could encompass two classes of measures.

The first would support the use-case for euro-denominated stablecoins by increasing their liquidity in secondary markets, and hence making them more useful as a means of payment – in effect, creating ‘transactions demand’ for e-money tokens (EMTs).

Regulating EMTs as money through redemption and prudential requirements is not enough to make euro-denominated stablecoins competitive in real-use cases, such as purchases and sales of assets – crypto or not – or cross-border transactions. For these, they must achieve sufficient network adoption and market liquidity. Measures include:

• Allowing MiCA-compliant euro stablecoins to be fully usable on EU-regulated trading and tokenised-asset platforms, without operational or technical disadvantages rela-tive to foreign-currency instruments;
• Encouraging major European financial institutions to act as market makers in euro stablecoins, thereby ensuring depth and liquidity;
• Ensuring that regulated trading venues have sufficient access to euro-denominated digital liquidity to process large transactions smoothly.

The second set of measures would make MiCA and its implementation more supportive of the growth of compliant euro stablecoins, without lowering its prudential standard. This could be achieved by:

• Removing the requirement that 30 percent to 60 percent of reserve holdings must be deposits with financial institutions. Unlike the requirement to hold reserves in highly liquid, low-risk assets, forcing stablecoin issuers to hold a high level of reserves in bank deposits cannot be justified on prudential grounds. Furthermore, both requirements together limit issuers’ ability to optimise returns and may contribute to funding instability of banks in crisis situations. This constraint becomes more binding as issuance scales (because of the 60 percent requirement for significant EMTs), compressing margins in a way that would not be the case in the US. It should therefore be abolished. Meanwhile, MiCA could adopt the stricter liquidity standards that are prescribed by GENIUS.
• Allowing direct remuneration of stablecoins. The prohibition on direct remuneration is not justified by financial stability considerations. On the contrary: it is a source of volatility and run risk, as sharp interest rate increases could trigger redemption requests in large volumes (Bindseil, 2026). Stablecoin issuers should be allowed to pass through to users a limited share of their remuneration. To preserve monetary control and avoid an excessive decline in bank deposits, the rate of remuneration should be linked to the policy rate, but remain below both the rate on central bank reserves and standard deposit rates. This would ensure that euro stablecoins are sufficiently attractive to support adoption, while not fully substituting for bank deposits.
• Calibrating the ‘significant issuer’ designation so that it rewards scale rather than penalising it. MiCA requires issuers to hold own funds equal to 2 percent of reserve assets, rising to 3 percent once an issuer is designated as significant. Capital requirements therefore rise precisely at the point where an issuer is becoming commercially viable at scale. The underlying objective – ensuring that systemic entities hold larger capital buffers, including to enable an orderly wind-down if needed – is legitimate. But a restrictive interpretation of ‘significance’ would impose its heaviest burden on the issuers most capable of building deep, internationally competitive euro-denominated stablecoin markets.

The main argument against abolishing MiCA’s requirement to deposit 30 percent to 60 percent of reserves with banks, and to allow some direct remuneration, is the fear that these measures would tilt the playing field against banks, leading to financial disintermediation (see section 3.1). Given Europe’s bank-based economy, this may be why MiCA is more bank-friendly and less stablecoin-friendly than GENIUS²³. But the fear is overblown for several reasons.

First, the excess liquidity of euro-area banks is still substantial, at around €2.3 trillion at the end of the first quarter of 2026. The possibility that banks will be constrained in creating credit because of tight liquidity remains highly unlikely.

Second, unless stablecoin issuers are able (and willing) to hold a portion of their assets in the form of reserves with the central bank, the stablecoin issuance will not change the total volume of deposits in the commercial banking system: either stablecoin issuers hold bank deposits or they buy securities (eg government bonds) from counterparties that do²⁴.

Third, if stablecoin issuers were allowed to hold some of their reserves with the central bank (a possibility we advocate; see section 4.2.4), bank deposits could indeed decline. But any impact of this on bank liquidity could be neutralised by the central bank, which has the power to recycle the additional reserves to credit-creating institutions. Hence, while deposits may be replaced by central bank funding, the capacity of the financial system to provide credit would not change.

Fourth, banks would remain much freer than stablecoin issuers in managing their assets and liabilities. The capital and liquidity requirements they face are far less stringent than those imposed by MiCA on stablecoin holders. Furthermore, the stablecoin remuneration cap we propose – below the central bank reserves rate – would continue to disadvantage stablecoins relative to bank deposits (whether tokenised or not).

4.2.3 Enabling scalable and interoperable tokenised deposits and stablecoins 

Addressing the problem of the inherent fragmentation of blockchain technology, which limits scalability of both stablecoins and tokenised deposits, requires coordination mechanisms that go beyond purely market-driven solutions. Several complementary approaches can be envisaged:

• Standard-setting and interoperability requirements: authorities could promote or impose common technical standards for messaging, token formats and settlement interfaces (analogous to ISO standards in traditional finance). This would lower switching costs and facilitate native interoperability across infrastructures without reliance on ad-hoc bridges.
• Regulatory incentives for consolidation or common rails: frameworks such as MiCA could be calibrated to favour issuance on interoperable or shared infrastructures. For example, preferential treatment (in capital, liquidity or operational requirements) could be granted to arrangements that ensure cross-platform compatibility or settlement finality on common layers.
• Public infrastructure as a coordination anchor: central bank initiatives, such as wholesale CBDC platforms or unified ledgers, could act as common settlement layers, reducing fragmentation by providing a shared, risk-free asset via which private tokens (stablecoins or tokenised deposits) can interoperate. This echoes proposals for integrated platforms that enable multiple forms of digital money to coexist on a single infrastructure.
• Oversight and standards for bridges and cross-chain protocols: given their systemic importance, regulators could establish minimum requirements for security, governance and transparency of bridging mechanisms, reducing risks while interoperability technologies mature.
• Designation and supervision of systemic interoperability providers: as certain cross-chain infrastructures become critical nodes, they could be subjected to enhanced oversight (similar to financial market infrastructures), ensuring resilience and consistent access across participants.

These measures point toward a model in which interoperability is not left solely to technological evolution, but is actively shaped by policy to preserve the singleness of money in a multi-chain environment. Without such coordination, fragmentation is likely to persist as a structural feature, constraining the scalability and effectiveness of stablecoins and tokenised deposits, and potentially undermining their roles in the broader financial system.

4.2.4 Granting euro-denominated EMTs structured access to central bank reserves

Reserve access would materially strengthen the credibility and safety of euro stablecoins. Anchoring reserves directly in central bank liabilities, rather than commercial bank deposits or short-term securities, would reduce credit and liquidity risk, reinforce redemption certainty and support the singleness of money. Euro stablecoins could then become fully-backed programmable claims on central bank money, rather than private liquidity instruments exposed to banking-sector stress.

Work by the Bank of England (2025) provides a useful reference point for alternative design choices (Table 1). In its 2023 framework for systemic stablecoins, the bank originally envisaged issuers holding backing assets fully in central bank reserves, effectively anchoring stablecoins directly to central bank money. Building on this approach, a first policy option would be to introduce a fully backed, central bank reserve model, under which EMT issuers hold all their reserves at the central bank and receive remuneration at, or close to, the policy rate. This would eliminate credit and liquidity risk on the asset side, simplify supervision and ensure economic viability by providing a stable revenue base. It would also align stablecoins more closely with tokenised central bank liabilities, while preserving roles for private intermediaries in distribution and innovation.

A second, more flexible option (Bank of England, 2025) is a fractional central bank reserve model. A defined share of reserves would be held as remunerated deposits at the central bank, with the remainder invested in high-quality liquid assets such as short-term sovereign bonds. The proposed structure is at least 40 percent held as unremunerated deposits at the Bank of England and up to about 60 percent held in short-term UK government debt (gilts). The 40 percent portion is pure central bank money, which ensures immediate liquidity and convertibility, while the 60 percent portion is safe but interest-bearing, giving issuers a viable business model. This hybrid solution would not turn stablecoins into full ‘synthetic CBDCs’, but also would not allow them to behave like shadow banks.

These leaves two questions: if reserves should be remunerated, and if stablecoins should benefit from lender-of-last-resort facilities. The first depends in part on which model of reserve backing is adopted, while the answer to the second is an unqualified yes.

1. If the stablecoin is fully backed by reserves (the first option, above), remuneration is necessary to make the business model viable. The question is whether reserves should also be remunerated in the hybrid model (the second option). Doing so would improve monetary policy transmission by making the policy rate directly accessible to all holders of liquid money, thereby creating a uniform floor under returns on money. It would of course also imply that banks would enter into direct competition for deposits with stablecoins.
2. Introducing lender-of-last-resort facilities for EU-regulated stablecoin issuers is justified by the need to preserve par convertibility under stress. Even when fully backed by safe assets, stablecoins remain exposed to liquidity shocks and fire-sale dynamics that private buffers cannot absorb in systemic events. By providing temporary liquidity against high-quality collateral, the central bank prevents destabilising runs, ensures continuous redemption at par and aligns the regulatory treatment of stablecoins with their role as systemically important payment instruments.

Would the remuneration of stablecoins reduce central bank income? We do not think so. Allowing stablecoin issuers to hold remunerated reserves would not increase aggregate reserves but only change their distribution across private intermediaries. Moreover, restricting reserve access does not eliminate private interest income; it shifts it toward sovereign securities or bank deposits, potentially increasing market stress in periods of large-scale redemptions.

In addition to benefiting financial stability and monetary policy transmission, granting structured access to remunerated central bank reserves could also offset the potential competitiveness disadvantage implied by MiCA for euro stablecoins relative to dollar stablecoins. If euro stablecoin issuers can hold meaningful shares of reserves at the ECB and earn remuneration linked to the prevailing policy rate (possibly subject to constraints, such as a tiered approach, or maintaining remuneration below that of bank reserves), redemption risk will be reduced, fire-sale pressures on sovereign bond markets will be minimised and convertibility into public money would become credible and operationally robust. Unlike an unremunerated full-reserve model, remunerated reserve access would simultaneously enhance stability and support a sustainable business model.

5 Conclusion

The emergence of tokenised finance, and with it the invention of stablecoins, challenges regulators. The type of challenge is not new: the emergence of a new instrument, issued by entities outside the perimeter of the traditional banking system, which may be commercially successful enough to eventually force the regulator to bring it inside the perimeter. But in this case, market evolution is happening more rapidly than ever, the risks are potentially greater and the inherently cross-border and decentralised nature implies that tight regulation within a single jurisdiction is less likely to succeed.

In seeking to regulate payment stablecoins, the EU is pursuing the right goal: to protect financial stability and ensure that private digital money remains anchored in public money under ECB control. But the means employed – stringently restricting the development of private digital money-like instruments outside the existing banking system – are wrong.

The risk is that the demand for stablecoins will be pushed offshore. If this happens, EU stablecoin holders might not be safe, even if EU-issued stablecoins are entirely safe. EU digital infrastructures may align around dollar-based conventions, and disruptions affecting major US stablecoin issuers could transmit into European tokenised markets not merely through price channels, but through settlement and collateral channels. Applying stricter requirements domestically would not eliminate the regulated activity, but merely imply that the relevant infrastructure, governance and value capture remain outside the EU regulatory perimeter.

To prevent this, the policy goal should shift from containment to embedding. The conditions should be created for regulated stablecoins and tokenised deposits to be economically viable, interoperable and anchored – directly or indirectly – to central bank money. Steps to do this include giving EU-regulated stablecoin issuers greater freedom in how to invest their reserves, as long as these remain safe and highly liquid; paying stablecoin holders remuneration, as long as this remains strictly below the rate at which the ECB remunerates its reserves; and allowing access to the ECB balance sheet.

A potential objection to this approach is that the fast growth of EU-regulated stablecoins will lead to financial disintermediation by reducing deposits in banks. This argument is not correct. Banks would remain far less constrained in how they can invest deposits and what remuneration they can pay. Banks can fight back, by offering tokenised deposits and integrated digital settlement services. Most importantly, the need for stablecoin issuers to purchase reserves would maintain the overall liquidity of the system and the central bank has the power to recycle this liquidity to credit-creating institutions. While some bank deposits may be replaced by central bank funding, the capacity of the financial system to provide credit would remain unchanged.

The steps proposed in this paper do not predetermine the final equilibrium between various forms of digital money: inside the banking system, outside the banking system and CBDCs. But they ensure that the architecture within which they develop preserves monetary sovereignty, prevents parallel settlement systems outside central bank anchoring and avoids structural dependence on dollar-based systems.

Source : Bruegel