China's Dedollarisation Strategy

1. Introduction

The US appears to be retreating from its role as a reliable provider of a safe and stable international currency (Pforr et al. 2025). As the dominant international currency, the dollar grants the United States a unique position in the global financial architecture. More recently, the current US administration has shown an increasing willingness to exploit their power as an instrument of geopolitical pressure. Accordingly, the costs of dollar dependence are going up and countries around the world are looking for alternatives.

In this context, the Chinese government’s efforts to internationalise the renminbi (RMB) are an instructive case of a major economy actively attempting to loosen its dependence on the dollar (Eichengreen et al. 2018; 2024; Freymann & Heng 2025). What can Europe learn from China’s success and failures?

This report analyses China to understand how international currencies create dependencies in the first place, and which levers a dollar-dependent economy can realistically pull. It also allows for an empirical assessment of how far such a strategy can go: what has China achieved, by what means, and where has it run into limits? These insights have important implications for Europe’s own exposure to the dollar and options for managing them.

China’s policy mix has been broad, but it has made most progress on settlement. Policy measures include the build-out of an offshore clearing-bank network, the development of CIPS as a cross-border payment system, the introduction of a central bank digital currency (the e-CNY) with an explicit international ambition, and the negotiation of a large network of central bank swap lines.

The empirical record shows partial success. The RMB is now used to settle more than half of China’s own trade, a substantial achievement compared to a decade ago.

Settlement, however, has been the easiest function to shift. Invoicing in RMB remains rare; the currency is not used as a vehicle currency between third countries; and on the asset side China continues to hold a large stock of dollar reserves. The clear pattern is that progress on payment infrastructure has outpaced progress on the deeper use-cases of an international currency – invoicing and investment. The limitations of China’s current strategy reflect network effects that bolster the dollar’s dominant status and the lack of deep and liquid RMB markets.

Europe differs from China, but there are lessons to be learned. The European economy is similarly, if not more exposed to the dollar. European banks intermediate large volumes of dollar funding, European firms also invoice extra-EU trade predominantly in dollars, and European investors hold substantial dollar assets. At the same time, Europe lacks China’s ability to direct firms by political fiat, and faces its own well-known constraints in the supply of euro-denominated safe assets.¹

The Chinese experience therefore offers both a roadmap and a warning. It shows which instruments can move the needle, particularly with respect to reducing reliance on USD settlement. But it also demonstrates that USD centrality in invoicing and investing can survive meaningful dedollarisation in settlement. A serious strategy to manage the risks of dollar dependency must therefore analyse all three use-cases of an international currency; decide what level of USD dependence is appropriate in each case; and what policy levers can and should be employed to reach those levels.

The report proceeds in three chapters. Chapter 2 develops the theoretical framework, distinguishing the three use cases of an international currency (settlement, invoicing, and investment) and the network effects that link them. Chapter 3 turns to China: it documents the dollar dependencies China has sought to reduce (3.1), reviews the policy measures employed (3.2), and assesses what has and has not been achieved (3.3). Chapter 4 draws out the implications for Europe, asking which elements of the Chinese strategy are transferable, which are not, and what a credible European response to its own dollar dependencies might look like.

Dedollarising settlement is the achievable near-term win and need not pressure the euro’s exchange rate, because it works through payment infrastructure rather than engineered currency demand. In addition, the more systemic change of boosting invoicing and investment is more within reach than it is for a capital-controlled China, even though it still hinges on the structural supply of euro safe assets.

2. International currency use-cases

International currency systems have strong lock-in dynamics, leading to convergence on a single dominant currency. Historically, there has been a sequence of persistent and dominant international currencies, such as the Dutch Guilder in the 18th century, the Pound Sterling in the 19th and early 20th century, and the USD since the second half of the 20th century (Chahrour & Valchev 2022). To understand how an international currency creates dependencies and how China is trying to internationalize the RMB to evade these dependencies, it is helpful to understand the theory of international currency systems.

Convergence on a dominant currency is explained by network effects. To structure the analysis, we differentiate three broad purposes for which international currencies are used: i) settlement and payment, ii) invoicing, and iii) investment. There are network effects both within and across these three broad use cases (Figure 1).

The settlement currency is the currency in which the payment is conducted. Its choice is largely driven by transaction costs. How expensive is it for importer A to transfer 500 USD to exporter B? The utility of any given means of payment depends on widespread acceptance, which bring down transaction costs. Widespread acceptance benefits both the currency and payment infrastructures within that currency (Rochet & Tirole 2003). Since the cost per transaction goes down with scale, international currencies become cheaper with more widespread use. Building an alternative is not easy. In today’s international payment systems, a network of clearing banks is needed to which both firms have a connection. The banks not only need channels to send the money, but also to communicate securely with each other. Breaches in the system could result in large amounts of wealth being displaced, resulting in significant costs creating and using the system. It is unclear to what extent technological innovation can simplify such transaction (Bindseil & Pantelopoulos 2022).

Invoicing currency determines the value to be transferred on the bill. The invoicing currency does not need to be identical to the settlement currency. For example: importer A buys a product from exporter B and agrees to pay the value of 500 USD (invoicing) in six months. Six months later, it may conduct the payment (settlement) in RMB at the USD-RMB exchange rate. In this case the foreign exchange (FX) risk lies in USD. If the USD appreciates during the six-month, importer A has to pay more in terms of their domestic currency. Invoicing is also determined by its own network effects (Gopinath 2015; Gopinath et al. 2020; Amiti et al. 2022). Firms use the dollar as an invoice currency because other firms do so. For one, firms want to keep their prices stable relative to those of competitors, who price in dollars. Second, they are locked into input-output linkages of global value chains: firms want to keep their output prices stable relative to the input prices, which are priced in dollar. These benefits are further increased by the low cost of FX-hedging in widely used currencies.

Investors, as well as debtors prefer safe and liquid markets. The choice of a currency to invest and issue in is not just determined by returns on investments, but also the size and liquidity of markets. This gives rise to what economists call a convenience yield on USD-denominated assets (Krishnamurthy & Vissing-Jorgensen 2012; Gorton 2017). The depth of USD financial markets creates safety and instant convertibility that investors are willing to pay a premium for. Because major central banks and asset managers hold dollar assets to manage daily liquidity, the market becomes the most liquid. Such self-enforcing dynamics create deep financial markets, high market liquidity, and the creation of cheap hedging tools.

As mentioned, the three use-cases are interlinked with each other. For obvious reasons, invoicing and settlement currency typically coincide (Friberg & Wilander 2008). If importer A hedged their 500 USD payment in six months, by holding a USD bond of the same value, RMB settlement would imply additional transaction costs. Similarly invoicing and investments are connected, as deep financial markets reduce FX-hedging costs. Investment and settlement are also connected, but looser. To conduct the settlement in a respective currency, liquidity in this currency needs to be accessible for the payer. For countries with free capital markets, this is hardly restricting, as FX-markets are typically very liquid, but for countries with capital restrictions this can become an important determining factor, as we will see in chapter 3.

Although the interdependencies are significant, policy measures can target individual aspects of international use. Focused policy measures can promote a targeted use-case, while network effects or regulation can keep other use-cases from following. As outlined in chapter 3, China has made some success in disentangling settlement and invoicing use of the RMB.

The outlined network effects and interlinkages create strong lock-in effects, explaining the prevalence of dominant currency systems.² A dominant currency’s benefits for individual users are so big that switching to an alternative becomes too hard or difficult to do, even if the alternative currency has other benefits (stable value or political alignment). A country that is locked out of the system then also faces serious economic consequences, creating strong dependencies on the hegemon. However, these dependencies can go in both directions. If the issuer of the dominant currency takes advantage of the dependencies, or even only threatens to do so, they are weakening the network effects and undermine their own dominance. In other words: If the lever is pushed too hard, it might break.

Because of the network effects international currency systems typically exhibit tipping-point behavior. Forces pushing against the dominance of a currency may show little effect for a long time, before a rapid switch can happen (Farhi 2019). As soon as a significant number of economic agents switches currency, the remaining agents experience weaker network effects, incentivizing them to also subsequently switch towards the potential new hegemon currency. Vice versa, the hegemon experiences increasing network effects further galvanizing the switch.

Therefore, the US have to be careful in using the currency power they hold over China (Sigl-Glöckner 2026). An overly aggressive use of their power would undermine their own currency and the advantages that come with it. For example, if the US would seek to lock China out of the USD financial system, other countries would likely seek alternatives means to pay their imports. Global dollar use would decline.

Similarly, China and other to the US geopolitically distant countries hold a significant amount of USD assets, granting US investments a convenience yield. If these assets were dumped on the market, this would significantly destabilize the US-economy, drive up borrowing costs for the US Government, US firms, and US households, and create lasting economic damage.

3. China’s dedollarisation strategy

The Chinese export model makes it highly dependent on the US dollar, but such dependencies need not lock it into dollar dependence per se. Although China has presented its currency strategy as one of internationalising RMB use, turning its currency into a rival to the dollar would require two crucial conditions to be met. The US would not only need to continue its currency erratic policies, thereby making the USD less attractive not just for China but also for other countries. China would also need to largely, although perhaps not entirely, liberalise its capital account (Eichengreen et al. 2018; 2024; Freymann & Heng 2025; Hofman & Petry 2025). We show that Chinese policies have nonetheless been quite effective in reducing its dependence on the US, thereby increasing its relative leverage.

3.1. USD-dependencies

The Chinese government seems increasingly concerned about its dependence on the US and the USD – and for good reasons. A newspaper-based index reports that global sanction risk is higher than ever (Iacoviello & Tong 2026; Figure 2). A bilateral index, based on the same dataset, indicates unprecedented geopolitical tensions from the US to China (Figure 3). This matches with increased usage of sanctions by the USA and the threat they impose on China (Bianchi & Sosa-Padilla 2025; Garofalo et al. 2025). This raises the question of how exposed China is to the US by using the USD and USD payment infrastructure.

China’s authorities are secretive about their USD-dependencies. For example, they do not publish official data on the currency composition of their invoicing, or their reserves. Nonetheless, China remains heavily reliant on the USD for both trade and investment purposes.

In contrast to invoicing, official settlement data on Chinese payments is available. Here China has made significant progress in reducing USD-dependencies over the past years (Figure 4). In 2010 over 80 percent of payments were settled in USD (SAFE 2026). In 2025 this declined to 46 percent of payment inflows, and 40 percent of payment outflows. An important factor in increasing RMB settlement was the introduction of the clearing infrastructure called Cross-Border Interbank Payment System (CIPS; Figure 5). This offers a structured and unified way of settling RMB payments, and covers the vast majority of global RMB payments. CIPS also features its own inter-banking messaging capabilities. Since such messages initiate payments, they are an integral part of the settlement infrastructure. Currently, CIPS messaging only covers within-system communication, while for most RMB transaction one bank of the transaction is often not part of CIPS. This explains while most CIPS payments still rely on SWIFT messaging, which provides the much greater network of clearing banks (Beschwitz 2024).

For invoicing, the USD-share probably remains significantly higher than the 40 to 46 percent of settlement still conducted in USD. A recent analysis by Boz et al. (2025) infers that a maximum of 10-15 percent of Chinese trade is invoiced in RMB. Most of the remainder is likely invoiced in USD. An earlier BIS analysis estimates the USD invoicing share at around 75-85 percent (Ito & McCauley 2019).

Similarly, China’s reserve portfolio remains heavily reliant on USD. The last currency composition release of the official reserves managed by China’s State Administration of Foreign Exchange (SAFE), was for 2019. According to the official statistics the USD-share decreased from 79 percent in 2005 to 55 percent in 2019. However, this decline was primarily driven by the extensive reserve expansion between 2005 and 2012 from USD 800 bn to over USD 3 tn, resulting in an expansion of China’s USD reserves in absolute terms (Setser 2026). Moreover, China appears to have increasingly shifted USD assets from the official reserves into other portfolios, hiding their USD dependency. For example, China’s state commercial banks have recently expanded their USD asset holdings to almost USD 1 tn, while decreasing their USD liabilities (Figure 6). In addition, up until recently, most loans of China’s Belt and Road (BRI) have been denominated in USD (Greene 2024).

The Chinese private sector also remains heavily exposed to USD asset markets. More than USD 1.1 tn of dollar-denominated debt was owed by Chinese borrowers in late 2023 (Greene 2024). In addition, the market capitalization of Chinese firms listed on US stock exchanges amounted to USD 848 bn in early 2024, although raising capital via US markets has recently become less popular for regulatory reasons.³

China has significant dependencies on the USD across all three use-cases: settlement, invoicing, and investment. But where does dependency translate into US leverage?

Concerning invoicing, direct US influence is limited. It is hard to prevent the USD from being used as an invoicing currency, even among sanctioned countries, as the currency that commercial parties use on their bills cannot be controlled by the US. Invoicing is, however, tightly linked to settlement and investment usage, and on those fronts, sanctions, taxes, conditionalities and other actual or threatened policy interventions are feasible.

Second, concerning settlement, almost half of Chinese trade continues to settle in USD. China could in principle absorb large volumes of RMB settlement, and in the event of sanctions, USD settlement could be redirected to RMB. But the switch would raise transaction costs notably for firms in countries with weaker connections to Chinese payment infrastructure. It would also strain the global provision of RMB liquidity. If, however, under sanction pressure, RMB settlement were further expanded, the move could end up significantly weakening USD dominance. Firms that settle with some trading partners in RMB might extend the practice to others, especially if the infrastructure works well – and they may not switch back once sanctions are lifted.

Third, concerning investment. Sanctions targeting Chinese USD investments could hit China hard. As outlined, USD assets and debt are integral to the Chinese economy. The public side relies on them to manage the RMB-USD exchange rate, the private sector to hedge FX exposure. But such sanctions would not only weaken the USD’s international role, they would also reduce investment flows into the US. Demand for US Treasuries could plummet, and the same is likely to hold for private-sector assets. The result could be severe damage to US capital markets and to the US economy itself.

The extent of this damage depends significantly on indirect- and second-mover-effects, and particularly on the actions of European investors. Chinese holdings of USD assets are heavily concentrated in US Treasuries. If sanctions or other policy measures against Chinese holdings disrupt the US Treasury market, the Federal Reserve is well placed to preserve market functioning, since it has stabilised this market successfully in the past, e.g. during COVID (Tooze 2021). Disruptions to US equity markets, however, which could follow from European investors’ reactions to US sanctions, would be harder to address: historically, these are not markets that the Federal Reserve has stabilised directly.

Exploiting China’s USD dependencies would therefore come at significant costs and risks to the US and should not be read as easy leverage. In the short run, however, the damage the US can inflict on China arguably exceeds the damage it would absorb itself, particularly if European investors remain calm. Under severely escalating geopolitical tensions, it is therefore not unthinkable that the US would exploit the dependencies, despite the long-run harm that this may entail.

3.2. Policy measures to promote the RMB

China’s strategy to internationalize the RMB is frequently viewed as a means to reduce the outlined US dependencies (Freymann & Heng 2025). To boost international RMB usage, China has implemented a broad set of policy measures that try to tackle the settlement, invoicing, and investment use of the currency. It has done most to increase the RMB’s settlement share, thus reducing USD settlement of Chinese trade over the past two decades. In contrast, policies to foster RMB investments and invoicing remain limited, due to China’s highly restrictive approach to capital flows.

3.2.1. RMB settlement

As outlined in chapter 2, the use of a currency for settlement purposes depends crucially on transaction costs. The invoicing currency can always be exchanged into the settlement currency right before the trade – at a cost. Restricted liquidity can typically still be obtained – at a cost. If a local commercial bank has no direct connection to a clearing bank in the target currency, an additional intermediary bank must be interposed. Boosting settlement in a currency therefore requires efficient payment infrastructure.

China has built out RMB payment infrastructure steadily over the past two decades. In 2010, it established an offshore RMB market (CNH) outside the capital-control regime that governs onshore RMB (CNY), accommodating unrestricted trade in the currency (Eichengreen & Kawai 2014). This provided the liquidity needed for international RMB settlement and laid the foundation for the payment infrastructure projects that followed.

A central pillar of this push has been the global network of RMB clearing banks. So far, 31 offshore clearing banks have been introduced in 27 countries (Beschwitz 2024). Most but not all are subsidiaries of Chinese banks. The two US RMB clearing banks, for example, are subsidiaries of Bank of China and JP Morgan respectively. Most clearing banks were established between 2013 and 2016, but the network is still expanding. In early 2026, the PBoC authorised branches of Bank of China in the UK and Sri Lanka. Perez-Saiz & Zhang (2023) estimate that the establishment of RMB clearing banks raised the RMB share of payments to China by up to 6–7 percentage points.

China introduced its own payment infrastructure, CIPS, in 2015. CIPS unified the clearing and settlement of RMB payments, mirroring the USD’s CHIPS infrastructure. Alongside clearing, it offers a SWIFT-like messaging service. For now, CIPS messaging only handles transactions between two CIPS-connected banks. One-sided CIPS connections still route through SWIFT (80% in 2022) (Yeung & Goh 2022). Even so, the in-house messaging capability can be seen as insurance against potential sanctions targeting SWIFT.

Central bank digital currencies (CBDCs) have so far made an visible, but small dent in international RMB settlement. China has launched its own CBDC, the e-CNY. With a global transaction value of roughly USD 1 tn in 2025, it is the world’s largest CBDC (Chhangani 2026). Most e-CNY payments are domestic, but China has taken the main authority over mBridge, a cross-currency clearing system for international CBDC payments. 95% of mBridge’s transaction volume is in e-CNY, amounting to roughly USD 52 bn per year (Jones 2026). However, this amounts to only around 0.2% of the yearly RMB transaction volume via CIPS (Baker 2025). CBDC transactions are nonetheless cheaper and faster than classical settlement, which creates significant growth potential. Particularly since the US and euro area do not yet offer an international CBDC alternative to the e-CNY.⁴

PBoC swap lines, by contrast, have already raised international RMB payments. Swap lines are bilateral central-bank agreements that provide short-term liquidity in a foreign currency in an emergency. The receiving central bank can then on-lend the funds to domestic financial institutions needing access to that currency. The PBoC now operates swap lines with 32 countries, covering roughly USD 625 bn in total (PBoC 2026). Bahaj and Reis (2026) show that the introduction of a swap line raises the value of RMB payments from the counterparty country by 400 percent. Unlike Fed swap lines, which mostly cover advanced economies, the PBoC also extends swap lines to emerging economies. Microevidence on Argentina shows swap lines being used to access liquidity for trade (Benguria & Novy 2025). Regulatory restrictions on USD liquidity led firms to invoice their imports from China in RMB, drawing on the PBoC swap line. This drove the RMB invoicing share to around 50% of Argentine imports from China.

3.2.2. RMB invoicing

The Argentine case shows that swap lines can raise not only settlement but also invoicing. The mechanism works through two channels. First, higher settlement directly drives higher invoicing. Firms already settling in RMB face lower transaction costs and risks if they invoice in it too. Second, swap lines mitigate FX tail risk (Bahaj & Reis 2026). As outlined in chapter 2, the invoicing decision is heavily shaped by FX exposure. Commercial banks engage in maturity transformation, rolling over short-term financing for long-term investments. In a global emergency, short-term funding markets can seize up, forcing banks to dump illiquid long-term assets at a loss and triggering severe FX fluctuations. Swap lines provide emergency liquidity in such situations, capping the tail risk that would otherwise deter invoicing in the currency.

China has also concluded bilateral trade partnerships with Russia, Iran, Brazil, Argentina, and Venezuela, each carrying pledges to promote RMB invoicing. These are framework commitments rather than binding obligations, and their direct effect on invoicing remained modest until recently. Since 2022, uptake has accelerated, presumably partly driven by greater pressure on Chinese firms to invoice in RMB. The expansion of sanctions on Russia and Iran has also pushed those economies further into RMB-denominated trade. The result is visible in the invoicing data (Figures 4 and 7).

Talks with Saudi Arabia to invoice some oil exports in RMB have so far been inconclusive. Even so, the bilateral relationship is deepening. In 2023 a swap line was introduced and a Bank of China branch opened in Riyadh. A successful agreement on RMB invoicing of oil exports could prove pivotal. Saudi Arabia supplies roughly 21% of China’s oil imports, and were RMB invoicing to take hold there, it would likely cascade across energy supply chains globally, creating structural demand for RMB income and hedging instruments. A similar agreement between Saudi Arabia and the US in 1974, to globally invoice oil exports in USD, is widely seen as a key driver of the “petrodollar’s” dominance.

Promoting RMB invoicing is difficult for China in part because of its capital controls. Financial FX-hedging is not freely available and markets are less liquid. Moreover, due to its relatively low risk-free rate, the US dollar trades at a forward discount against the RMB. This makes FX-hedging using derivatives relatively expensive for Chinese firms, thus incentivising dollar invoicing as a natural hedge. These hedging costs also inhibit foreign firms from taking on RMB FX risk. Receiving payment in RMB, finally, is not a particularly attractive prospect without RMB-assets to invest in.

3.2.3. RMB investment

Asset markets are key to internationalising the RMB, not only to provide a store of value, but also because of their multiple synergies with invoicing and settlement as described in chapter 2. Liquid and deep financial markets enable efficient FX-hedging and provide liquidity for settlement when needed.

However, promoting RMB investment is challenging for China because of its capital controls. Beijing has nonetheless sought to work around them by creating distinct international asset categories and by granting qualified foreign investors extended access to Chinese asset markets.

Hong Kong government bonds provide an international safe asset. Dim sum bonds offer offshore RMB issuance. Panda bonds let foreign issuers raise funds within China. Qualified foreign investor status can be obtained by eligible institutional investors after in-depth checks of financial solvency and jurisdictional eligibility. Such investors enjoy extended access to Chinese stock and bond markets. In 2023, access was widened to swap markets to allow the hedging of duration risk. Rules and restrictions for qualified investors have since been progressively simplified or lifted, making Chinese asset markets more efficient for foreigners and building credibility with foreign investors.

These policy measures initially produced a rapid rise in foreign investment (Figure 8). After 2021, however, foreign investment declined again, driven by economic weakness, notably the Chinese real estate crisis, the interest rate spread to major advanced economies, the depreciation of the RMB, and rising geopolitical tensions with the West (BOFIT 2025).

A similar pattern can be observed for central bank RMB reserves. Following the RMB’s inclusion in the IMF’s Special Drawing Rights basket in 2016, its share in global central bank reserves rose from 1 percent to almost 3 percent by 2021 (Figure 9). It has since declined to roughly 2 percent.

Recent reports point to renewed growth momentum in foreign RMB investment, but levels remain far below those of USD markets (Zeng 2025). Total foreign holdings of onshore RMB bonds and A-shares amount to about 3 percent of foreign holdings of US securities (BBVA 2024; CRS 2025). Foreign onshore bond holdings, which are largely government-backed, are around 7 percent of foreign US Treasury holdings. The RMB’s share of allocated global FX reserves, at about 2 percent, compares with roughly 58 percent for the USD.⁵

The gap between US- and Chinese capital markets is not just one of size, but also one of depth and liquidity. The US Treasury market provides the world’s deepest pool of safe assets: even very large positions can be bought and sold quickly without causing major price moves. Given the smaller pool of RMB safe assets, the same is not possible here. This makes holding large positions a riskier proposition, and RMB assets less attractive for liquidity management.

With capital controls in place, systemic inefficiencies and barriers remain, despite China’s manifold workarounds. The layered access schemes raise the operational cost of investing. Short-selling remains banned, and derivatives use is restricted to hedging, even for qualified foreign investors. Furthermore, even as entry to foreign investment is increasingly opened, exit remains constrained. Repatriating proceeds out of China requires SAFE registration, tax clearance, and custodian-bank intermediation. In addition, the PBoC can throttle the schemes that run through Hong Kong infrastructure at will. All of this restricts efficiency and raises the tail risk of RMB asset markets. Under these circumstances, a catch-up to USD financial markets is hardly imaginable.

3.3. Discussion

To date, China’s policy measures have concentrated on settlement. Invoicing is inherently difficult to improve through direct intervention. Forcing firms to bill in RMB impairs their competitiveness. Some foreign importers would abstain from trade altogether; others would demand discounts to compensate for the higher FX hedging costs in less developed RMB markets.

Deepening those markets requires lifting capital controls. So far, China has loosened them only marginally. Open capital markets would force the PBoC to choose between exchange-rate control and monetary policy sovereignty, putting pressure on their export markets. They would also expose Chinese financial markets to foreign influence, for example, by hedge funds, or even other states.

These constraints explain the pattern: settlement has advanced, invoicing and asset holdings remain low. Considering the linkages between the three use-cases (Figure 1), the invoicing currency decision is highly dependent on the ability to hedge FX-risk efficiently. Without lifting capital controls, however, this will remain limited. This explains the unusual gap between invoicing and settlement currency in Chinese trade (Figure 4). Foreign firms accept RMB settlement because the payment infrastructure and liquidity are there. They reject RMB invoicing because the FX exposure is too costly.

Invoicing growth has nonetheless accelerated since 2022. The driver is most likely geopolitics. Following the Ukraine war and the sanctions that came with it, countries geopolitically distant from the US have raised their RMB invoicing share; countries close to the US have lowered theirs (Figure 7). For sanctions-exposed countries, USD FX hedging becomes increasingly expensive (Bianchi & Sosa-Padilla 2025), which makes RMB hedging comparatively cheaper, incentivising them to switch invoicing currency.

Absent major shocks, the RMB will not take on a significant international role in the near future. The low-hanging fruits have largely been harvested. The countries least aligned with the US may turn to China, but their combined economic weight is, at this time, comparatively small. The payment infrastructure, swap-line network included, has been broadly in place for a decade. It produced a sharp rise in RMB settlement of Chinese trade, but little vehicle usage of the RMB between third countries. This is what would be needed to push global RMB settlement further and to challenge USD network effects. International RMB asset markets remain restricted as long as capital controls remain in place.

The limited success in internationalising the RMB does not mean China’s policy failed. Beijing’s authorities may have achieved what they set out to achieve. Without altering the structure of their economy or opening it to foreign influence, they have largely insulated themselves from US payment sanctions. Exclusion from the USD payment system would not isolate China the way it initially did Russia or Iran. Half of China’s trade already settles in RMB. It is most likely that the infrastructure could also cope with full RMB settlement. Payment sanctions would still cause damage, especially through price negotiations, due to increased transaction costs – but they would damage the USD too, and the leverage the US once had on this front has declined notably.

On the asset side, China remains exposed, but it is unclear whether that exposure can be levered against it. USD holdings provide a target for sanctions, and China has barely moved to dismantle that exposure. The cost-benefit trade-off has not yet tipped: it seems that the Chinese government thinks lifting capital controls would cost more than the asset-sanction risk justifies. That calculation also reflects the US side of the same trade-off. Large-scale asset sanctions on China would inflict immediate and severe damage on the US itself, particularly if European investors react, too. This damage is potentially greater and faster than the collateral damage caused by payment-side sanctions. US leverage on the asset side is therefore more limited.

Limited, however, is not small. Asset sanctions are not unthinkable if geopolitical tensions escalate. It appears, however, that Chinese policy makers perceive the the risk of such sanctions as less costly than the political costs of opening up Chinese capital markets.

4. Implications for Europe

What can Europe learn from China’s experience? Under the current administration, the dollar has become an instrument of pressure on allies and adversaries alike. Under US pressure, European banks have refused to open accounts for judges of the International Criminal Court in The Hague. Spillovers from US financial and monetary policy are already felt at regular intervals. Banking crises, exchange-rate swings, and Federal Reserve tightening transmit quickly to Europe, whether macroeconomically appropriate or not. In a future financial crisis, Washington could weaponise swap lines and the USD’s central role in the global financial architecture for short-term leverage. Thus, Europe, too, should evaluate and manage its USD-dependencies.

At first glance, the European economy is even more exposed to the dollar than China’s is. Private and public investors alike hold large stocks of US assets. European holdings of US Treasuries amount to roughly USD 3 tn, close to 10 percent of the total supply (Sigl-Glöckner 2026). Beyond that, European investors hold some USD 9 tn of long-term securities, equivalent to 45 percent of EU GDP (“U.S. Liabilities to Foreigners from Holdings of U.S. Securities” 2026). Around 34 percent of euro-area equity portfolios are allocated to US markets, and European investment funds place roughly half of their capital there.

However, as the China case illustrates, deep exposure does not automatically translate into strong US leverage. Europe may in fact be better insulated against US sanctions than China, precisely because sanctioning Europe would impose more substantial costs on the US itself. For example, weaponizing the Federal Reserve-ECB swap line could result in financial turmoil, forcing European banks and asset managers into fire sales of long-term US assets. Furthermore, the more diversified European exposure to US assets is more challenging to handle for the Fed. As argued above, a decline in demand for US Treasuries could, at least temporarily, be absorbed by the Fed to cushion the transition. A collapse in demand for private-sector assets, by contrast, is harder to offset. It would weigh not only on US capital investment but also on US pension funds, with direct and visible consequences for the welfare of US citizens. Deploying the full US sanctions arsenal against Europe appears unlikely.

This does not, however, rule out the threat of sanctions or other adverse policy measures as a bargaining chip. The Trump administration demonstrated the power of such coercion in July 2025, when the threat of a trade war secured it a favourable deal. The EU committed to USD 750 bn in US energy imports and USD 600 bn in additional EU investment in the US over three years, alongside a 15 percent baseline tariff on most EU exports to the US (Krahé et al. 2025). A comparable use of leverage is conceivable around the Fed’s USD swap lines. A credible threat to withhold them in a future liquidity emergency would threaten severe costs to Europe, even if at significant cost to the US itself. Europe may be unwilling to test whether the US would follow through on such a threat. As Washington shows growing willingness to apply economic pressure even to close allies, sustaining Europe’s current level of USD exposure could become both riskier and more costly.⁶

The China case shows that focusing on payment infrastructure alone is not enough for a comprehensive dedollarisation policy. A concerted push across all three use-cases is needed. Payment infrastructure is nonetheless an integral part of that effort, and here Europe can learn from China’s policy measures. The empirical record suggests that reducing transaction costs can do a lot to shift the choice of settlement currency in trade. With T2 the EU already has a wholesale clearing system in place, but it is focused on intra-European clearing, in contrast to the global clearing networks of the US (CHIPS) and China (CIPS) (ECB 2026). Transaction costs could be lowered by widening the reach of the euro clearing network or through technological innovation. The Eurosystem has already begun to move in this direction, interlinking its instant-payment system (TIPS) with those of partner countries to cut the cost of cross-border payments, and, through its Appia initiative, exploring a digital form of central bank money usable across borders. As yet, however, none of these efforts amounts to a central bank digital currency for settling international trade, of the kind China is piloting through mBridge.

Without China’s ability to exert political pressure on domestic firms, influencing the settlement currency choice of European firms may depend even more on economic factors, such as the invoicing currency. This raises the question of how to approach the invoicing and investment use-cases, where China’s own progress has been more limited.

Europe faces a similar problem to China, but addressing it may be much easier for Europe (van ’t Klooster & Murau 2025; Rey & Subran 2026; Vallée 2026). Both suffer from a shortage of safe assets and shallower, less liquid financial markets than the US. Without good assets to invest in, including both safe assets and more high yielding ones, receiving payments in that currency is less attractive. As we saw, this hinders RMB invoicing. Not only is financial FX-hedging not always freely available; low liquidity and yields on RMB assets also make FX-hedging more expensive for Chinese firms.

China’s financial and economic structure, however, relies heavily on capital controls, which Beijing has shown no intention of dismantling. Europe instead already has free and much larger capital markets, while other inhibiting factors can be tackled with policy. Approximately 20% of all global external assets and liabilities are already denominated in euros (roughly 50% is dollar-denominated) (EC 2026, p. 10).

A full emancipation from the dollar system would be a world-historical shift, whose dynamics and preconditions are difficult to understand ex ante. But China’s experience points the way: initial progress to limit dependence on US settlement infrastructure is possible with a clear strategic focus.

Building on this success requires a broader push. Here, Europe’s open capital markets may enable faster and further progress than what China has been able to achieve to date. In the context of an increasingly unpredictable US Government, the riskiest path for Europe may be to do nothing about its US dollar dependencies.

Acknowledgement

Authors: Nils Gerresheim (Oxford, Dezernat Zukunft), Dr. Max Krahé (Dezernat Zukunft, EMPN), Dr. Jens van ‘t Klooster (UvA, Dezernat Zukunft, EMPN)

This report is the EMPN edition of “China’s Dedollarisation Strategy: Asymmetric Success and What It Means for Europe”, first published by Dezernat Zukunft on 7 July 2026. The contents of this report may be freely used or reproduced without permission provided the original meaning and context are not altered in any way. Where third party copyright has been acknowledged, permission must be sought from the third party directly. For enquiries relating to this report, please email contact@empn.eu

Bibliography

• Abadi, J., Fernandez-Villaverde, J. and Sanches, D. (2025) ‘International Currency Dominance’, in Working paper (Federal Reserve Bank of Philadelphia), Federal Reserve Bank of Philadelphia, 25–20, Working paper (Federal Reserve Bank of Philadelphia), pp. 25–20.
• Amiti, M., Itskhoki, O. and Konings, J. (2022) ‘Dominant Currencies: How Firms Choose Currency Invoicing and Why it Matters*’, The Quarterly Journal of Economics, 137(3), pp. 1435–93.
• Bahaj, S. and Reis, R. (2026) ‘Jumpstarting an International Currency | The Review of Economic Studies | Oxford Academic’.
• Baker, J. (2025) ‘Is China’s cross-border payments network on the rise?’, FXC Intelligence.
• BBVA (2024) ‘Riding the waves-Stocktaking RMB Internationalization Development’.
• Benguria, F. and Novy, D. (2025) ‘How to Grow an Invoicing Currency: Micro Evidence from Argentina’.
• Beschwitz, B. von (2024) ‘Internationalization of the Chinese renminbi: progress and outlook’.
• Bianchi, J. and Sosa-Padilla, C. (2025) ‘International Sanctions and Dollar Dominance’, The Economic Journal, 135(672), pp. 2567–77.
• Bindseil, U. and Pantelopoulos, G. (2022) ‘Towards the Holy Grail of Cross-Border Payments’, European Central Bank, 2693, Working Paper Series.
• BOFIT (2025) ‘Despite economic headwinds, China’s bond market continues to grow’, BOFIT.
• Boz, M. E., Brüggen, A., Casas, C., Georgiadis, G., Gopinath, M. G. and Mehl, A. (2025) Patterns of Invoicing Currency in Global Trade in a Fragmenting World Economy. International Monetary Fund.
• Chahrour, R. and Valchev, R. (2022) ‘Trade Finance and the Durability of the Dollar’, The Review of Economic Studies, 89(4), pp. 1873–910.
• Chhangani, A. (2026) ‘What to watch as China prepares its digital yuan for prime time’, Atlantic Council.
• Coppola, A., Krishnamurthy, A. and Xu, C. (2023) ‘Liquidity, Debt Denomination, and Currency Dominance’.
• CRS (2025) ‘Foreign Holdings of Federal Debt’.
• EC (2026) ‘Strengthening the international role of the euro’, in Note to the Eurogroup, European Commission.
• ECB (2026) ‘The Eurosystem’s comprehensive payments strategy’.
• Eichengreen, B. and Kawai, M. (2014) ‘Issues for Renminbi Internationalization: An Overview’, SSRN Electronic Journal.
• Eichengreen, B., Macaire, C., Mehl, A., Monnet, E. and Naef, A. (2024) ‘Currency internationalization with Chinese characteristics: Is capital-account convertibility required for the renminbi to acquire reserve-currency status?’, International Finance, 27(2), pp. 102–28.
• Eichengreen, B., Mehl, A. and Chitu, L. (2018) How Global Currencies Work: Past, Present, and Future. Princeton: Princeton University Press.
• Farhi, E. (2019) ‘Toward a Multipolar System’.
• Freymann, E. and Heng, C. (2025) ‘The Logic of Partial RMB Internationalization: PRC Perspectives on “Financial War”’, The China Quarterly, 262, pp. 429–44.
• Friberg, R. and Wilander, F. (2008) ‘The currency denomination of exports — A questionnaire study’, Journal of International Economics, 75(1), pp. 54–69.
• Garofalo, M., Rosso, G. and Vicquéry, R. (2025) ‘Sanctions and currencies in global credit’.
• Gerresheim, N., Krahé, D. M. and Klooster, D. J. van’t (2025) ‘Report: Die Folgen einer Euro-Internationalisierung’.
• Gopinath, G., Boz, E., Casas, C., Díez, F. J., Gourinchas, P.-O. and Plagborg-Møller, M. (2020) ‘Dominant Currency Paradigm’, American Economic Review, 110(3), pp. 677–719.
• Gopinath, G. and Stein, J. (2021) ‘Banking, Trade, and the Making of a Dominant Currency’, The Quarterly Journal of Economics, 136(2), pp. 783–830.
• Gopinath, G. (2015) ‘The International Price System’.
• Gorton, G. (2017) ‘The History and Economics of Safe Assets’, Annual Review of Economics, 9, pp. 547–86.
• Greene, R. (2024) ‘China’s Dollar Dilemma’.
• Hofman, B. and Petry, J. (2025) ‘Internationalization of the RMB: Status, Options and Risks’, Clingendael, Clingendael CKN.
• Iacoviello, M. and Tong, J. (2026) ‘The AI-GPR Index: Measuring Geopolitical Risk using Artificial Intelligence’.
• Ito, H. and McCauley, R. N. (2019) ‘The currency composition of foreign exchange reserves’.
• Jones, M. (2026) ‘China-led cross-border digital currency platform sees surge’, Reuters.
• Klooster, D. J. van’t and Vallée, S. (2026) ‘Dedollarising the European Union: A strategy for monetary sovereignty in a multipolar currency system’.
• Klooster, J. van ’t and Murau, S. (2025) ‘Rethinking currency internationalisation: offshore money creation and the EU’s monetary governance’, Journal of European Public Policy, pp. 1–27.
• Krahé, M., Gerresheim, N. and Schulte, S. (2025) ‘Deal ja. Gewissheit? Nein.’, Dezernat Zukunft.
• Krishnamurthy, A. and Vissing-Jorgensen, A. (2012) ‘The Aggregate Demand for Treasury Debt’, Journal of Political Economy, 120(2), pp. 233–67.
• PBoC (2026) ‘RMB Internationalization Report 2025’.
• Perez-Saiz, H. and Zhang, M. L. (2023) Renminbi Usage in Cross-Border Payments: Regional Patterns and the Role of Swaps Lines and Offshore Clearing Banks. International Monetary Fund.
• Pforr, T., Pape, F. and Petry, J. (2025) ‘Dollar Diminished: The Unmaking of US Financial Hegemony Under Trump’, International Organization, 79(S1), pp. S117–33.
• Rey, H. and Subran, L. (2026) ‘Our euro, your solution: How to strengthen the international role of the euro’, French Council of Economic Analysis, (No. 89).
• Rochet, J.-C. and Tirole, J. (2003) ‘Platform Competition in Two-Sided Markets’, Journal of the European Economic Association, 1(4), pp. 990–1029.
• SAFE (2026) ‘SAFE Releases Data on China’s Foreign Exchange Reserves at the End of April 2026_Data Interpretation_State Administration of Foreign Exchange’.
• Setser (2026) ‘China Didn’t De-Dollarize. It Just Hid Its Dollars. | Council on Foreign Relations’.
• Setser, B. (2023) ‘China Isn’t Shifting Away From the Dollar or Dollar Bonds | Council on Foreign Relations’.
• Sigl-Glöckner, P. (2026) ‘Europe’s Trump Cards’, Dezernat Zukunft.
• Tooze, A. (2021) Shutdown: How Covid Shook the World’s Economy. Penguin UK.
• U.S. Liabilities to Foreigners from Holdings of U.S. Securities (2026) U.S. Department of the Treasury.
• Vallée, S. (2026) ‘A European strategy for a global euro’, in Public hearing, European Parliament (submitted to the record), European Parliament.
• Yeung, R. and Goh, K. (2022) ‘Petroyuan will not bring about a regime shift soon’, in China Insight, ANZ Research.
• Zeng, H. (2025) ‘New opportunities drive new RMB growth momentum’.