Suchoptionen
Startseite Medien Wissenswertes Forschung und Publikationen Statistiken Geldpolitik Der Euro Zahlungsverkehr und Märkte Karriere
Vorschläge
Sortieren nach
Massimo Ferrari Minesso
Lead Economist · International & European Relations, International Policy Analysis
Laura Lebastard
Economist · Economics, Supply Side, Labour and Surveillance
Olga Triay Bagur
Nicht auf Deutsch verfügbar.

Unlocking trade potential: the benefits of improving cross-border payments

Prepared by Massimo Ferrari Minesso, Laura Lebastard and Olga Triay Bagur

Published as part of the ECB Economic Bulletin, Issue 2/2026.

International trade could not happen without cross-border payments. Payment systems are the backbone of the financial infrastructure – the critical “plumbing” that underlies the functioning of modern economies by enabling the clearing and settlement of international transactions. This box aims to evaluate the economic benefits of technological innovations in cross-border payments by considering the case of interlinking fast payment systems between countries.

Many existing cross-border payments remain slow and expensive. Most international payments rely on correspondent banks – a global network that processes cross-border transactions for local banks that lack foreign accounts.[1] Payments often go through several intermediaries, making them slow and costly, owing to fees, currency conversions and operational frictions across different countries. For example, for nearly one-third of cross-border payments, the costs exceed 3% of the transacted amount, and on average only 40% of international business-to-business transactions are settled within one working day (Chart A).[2] Moreover, there has been a 20% decline in the global provision of correspondent banking services compared with the mid-2000s, increasing the cost of sending money across borders and, in some cases, leading to the complete disintermediation of certain payment corridors (stable payment linkages between pairs of countries).[3]

Chart A

Cross-border transaction costs and speed, by region

(upper scale: percentages of transaction value; lower scale: percentages of transactions, weighted by value)

Sources: Financial Stability Board (2025) and ECB staff calculations.
Notes: Sending country to the world, business-to-business and business-to-person transactions, for an amount of USD 20,000. Europe and Central Asia includes EU Member States, Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Georgia, Iceland, Kazakhstan, Kosovo, the Kyrgyz Republic, Moldova, Montenegro, North Macedonia, Norway, Russia, Serbia, Switzerland, Tajikistan, Türkiye, Turkmenistan, Ukraine, the United Kingdom and Uzbekistan. The latest observations are for April 2025.

Interlinking domestic fast payment systems has the potential to improve cross-border payments. New technologies have led to the development of a new generation of “fast” payment systems, capable of settling retail transactions in real time and at minimal cost. More than 80 countries have deployed a domestic fast payment system – such as the Eurosystem’s TARGET Instant Payment Settlement (TIPS), the Federal Reserve System’s FedNow Service, Brazil’s Pix and India’s Unified Payments Interface (UPI) – while many more are in development.[4] Interlinking these systems could reduce costs, increase the speed of cross-border payments and foster transparency; and, indeed, this has been identified as a priority in the G20 Roadmap for enhancing cross-border payments.[5] If payment systems are connected, banks in two jurisdictions can exchange funds using their respective domestic systems.[6] This avoids having several layers of correspondent banking that replicate processes and multiply costs, and would be of particular benefit to regions that are underserved by or excluded from correspondent banking.

Globally, there are already around 500 connections between fast payment systems, and more are in development. While TIPS interlinks euro area countries, Denmark and Sweden, interlinked fast payment systems have also been developed across Africa, Asia and South America (Figure A).[7] These cross-border arrangements vary widely: some support only retail payments, while others include wholesale settlement. Multilateral regional platforms emerged first, while bilateral links developed later and remain limited.[8] Overall, however, payment systems are still fragmented, and major economies continue to rely on correspondent banking. In line with the G20 Roadmap, and to strengthen cross-border payments and reduce fragmentation risks, the Eurosystem is working on new interlinkages (with India’s UPI) and exploring the potential benefits of connecting with Switzerland’s domestic fast payment system and Nexus Global Payments (NGP).[9] These initiatives are aimed at improving cross-border payments globally and reducing the risk of market fragmentation.

Figure A

Cross-border connections between fast payment systems

Source: Ferrari Minesso et al. (2025).
Notes: The figure shows cross-border connections between fast payment systems in 2024. It shows bilateral connections (split between unidirectional and bidirectional, depending on the currencies used to originate a payment through the link) and multilateral connections (also represented as bilateral connections between country pairs but coloured by regional platform).

Econometric evidence suggests that interlinking fast payment systems increases trade by about 4%. Iceberg trade costs depend partly on the efficiency of financial transactions.[10] Interlinking should reduce such costs and, consequently, improve bilateral trade. Including interlinking in a gravity framework suggests that connecting fast payment systems has a positive and economically meaningful impact on bilateral trade, even after accounting for potential endogeneity.[11] The average estimated effect (about 4%) is roughly half the boost to trade that comes from a formal trade agreement and a quarter of the impact of forming a common currency area (Chart B). Importantly, this estimate does not rely on specific case studies but is averaged across all interlinking initiatives, including those promoted by countries already well connected by global banking (such as the euro area or South-East Asia). This highlights the specific benefits for trade of interlinking payment systems, even when countries already have access to global markets through correspondent banks.

Chart B

Determinants of bilateral exports

(percentages)

Sources: Ferrari Minesso et al. (2026) and ECB staff calculations.
Notes: The chart shows estimates from a gravity model in which potential endogeneity is controlled for using the method in Carlson and Joshi (2024). The regression is specified as lnExporti,j,t=αi,t+αj,t+αi,j+β1Payment system interlinked+β2Common currencyi,j,t+β3Trade agreementi,j,t+β4Geopolitical distancei,j,t+Γ'Xi,j,t+εi,j,t. Payment system interlinked is a dummy equal to one if the fast payment systems of countries i and j are connected at time t. Xi,j,t includes the inverse Mills ratio that measures the expected value of the error term conditional on selection. The interlinking dataset covers 84 countries and 531 payment links. The model is estimated on annual data from 2021 to 2024.

The benefits of interlinking fast payment systems are larger in regions with high cross-border payment costs and for systems that allow the settlement of wholesale transactions. Focusing on the heterogeneity of payment systems, aggregate results are driven by payment systems that allow both wholesale and retail payments (Chart C). The benefits tend to be lower for payment systems that link only retail customers. This stems from the fact that aggregate trade is primarily driven by large-scale transactions of major firms, resulting in payment transfers whose values exceed the limits of retail payment systems. Moreover, interlinking is more beneficial in regions where the costs of cross-border payments are higher, suggesting that gains stem primarily from a reduction in fees charged by banks. This result is consistent with the interlinking of fast payment systems acting as a complement or alternative to more expensive payment methods, thereby reducing overall trade costs.

Chart C

Effect on exports by type of payment system and cost reduction

(percentages)

Sources: Ferrari Minesso et al. (2026) and ECB staff calculations.
Notes: The chart reports estimates from an augmented version of the equation used in Chart B. In the two columns on the left, the dummy for payment systems interlinking is divided between those systems that allow only retail settlement and those that allow both retail and wholesale settlement. The three columns on the right report the interaction coefficient of the interlinking dummy and measures of business-to-business cross-border payment costs in the region of the country of origin.

Interlinking initiatives may provide benefits that go beyond market functioning. The policy implications of these results could be substantial. First, they support ongoing international efforts under the G20 Roadmap to interconnect national payment systems, confirming that such initiatives deliver tangible economic benefits beyond their financial inclusion objectives. These gains are likely to be largest for countries that are poorly served by existing global payment networks and are therefore more excluded from international trade. Furthermore, these findings highlight the need for multilateral coordination to ensure the interoperability of technical standards and to address the legal and regulatory barriers (e.g. settlement finality rules, protection of personal data and fraud management) that still impede seamless cross-border settlement, thereby facilitating interlinking.

References

ACI Worldwide (2024), “Prime Time for Real-Time”.

Carlson, A. and Joshi, R. (2024), “Sample selection in linear panel data models with heterogeneous coefficients”, Journal of Applied Econometrics, Vol. 39, No 2, March, pp. 237-255.

Ferrari Minesso, M., Lebastard, L. and Triay Bagur, O. (2026), “Interlinking payment systems and trade flows”, Working Paper Series, No 3202, ECB.

Ferrari Minesso, M., Mehl, A., Triay Bagur, O. and Vanteenkiste, I. (2025), “Geopolitics and Global Interlinking of Fast Payment Systems”, CEPR Discussion Paper, No 20105, Centre for Economic Policy Research, April.

Financial Stability Board (2025), “G20 Roadmap for Enhancing Cross-border Payments – Consolidated progress report for 2025”, October.

Rice, T., von Peter, G. and Boar, C. (2020), “On the global retreat of correspondent banks”, BIS Quarterly Review, Bank for International Settlements, March.

  1. See Rice et al. (2020).

  2. See Financial Stability Board (2025).

  3. The retrenchment of correspondent banking has been driven by a combination of factors, including higher compliance costs, geopolitical risks and rising operating costs. Rice et al. (2020) provide survey evidence on the importance of each factor.

  4. See ACI Worldwide (2024).

  5. See Financial Stability Board (2025).

  6. Interlinking can be achieved either by connecting two payment systems directly through a shared infrastructure or by creating a hub that enables multiple systems to connect. Decisions to establish interlinking may be influenced by factors such as expected gains for trade, use for remittances, governance, technological preferences, cost-recovery prospects and geopolitical considerations. For more detail, see Ferrari Minesso et al. 2025.

  7. TIPS currently serves 23 countries with three currencies (the euro, the Danish krone and the Swedish krona), and more are in the pipeline.

  8. See Ferrari Minesso et al. (2025).

  9. NGP links the fast payment systems of India, Malaysia, the Philippines, Singapore and Thailand.

  10. “Iceberg trade costs” refers to a modelling assumption where a fraction of a traded good is lost during transportation, representing the cost of shipping.

  11. To account for potential endogeneity of payment system connections, the bias correction method of Carlson and Joshi (2024) is applied and the payment messaging standards of the domestic payment systems are used as an instrument. This method formally controls for the potential endogeneity of interlinking by modelling its probability in a first-stage regression. Results also hold when using semi-parametric methods (see Ferrari Minesso et al., 2026).