By Niranjan Govindaram 

Domestic real-time payments are a reality in most of the world. Many countries have functional low- and high-value domestic real-time payment platforms. Global real-time cross-border payments (RTCBP), however, are still a work in progress. Platforms like the Global Payments Initiative (SWIFT GPI) have helped to drastically reduce transaction times. Some pilot initiatives like immediate cross-border payments (IXB) look promising but are still experimental.

Cross-border payments: the journey so far

The Society for Worldwide Interbank Financial Telecommunication (SWIFT) was created in 1973. SWIFT established a common payment messaging format and model for payments data across the globe, which has served as the default network for communication-related to cross-border transactions. Since the 1990s, the ISO 15022 format has been the primary payments messaging format used by banks across countries, even for domestic payments.

Cross-border payment transaction times have shortened gradually over the last few decades—from several days to mostly single-day and sometimes to a few hours or minutes. For cross-border payments, the ISO 15022 format was the only way to conduct payment transactions. Cross-border payments are more complex than domestic payments and one added intricacy is the involvement of at least two currencies in any such transaction.

Here’s an example of how cross-border payment transactions work. Bank X in one country wants to transfer money to Bank Y in another country; both Bank X and Y need to hold an account with their counterparts. In an international transaction, funds are not physically moved. Instead, bankers credit accounts in one jurisdiction and debit the corresponding amount in the other jurisdiction. There are problems when banks want to make international payments to all global banks. For this to succeed, they would either need to set up their own branches everywhere or have thousands of direct relationships and hundreds of accounts to manage, which is nearly impossible. Only a few of the world’s largest banks come close to achieving this, but even they constantly struggle with many local requirement challenges from a technology, infrastructure, and regulatory perspective.

Because of these issues, banks often need to transact with intermediaries, also known as correspondent banks, and more than one correspondent bank may be involved. As the number of correspondents in the chain grows, transaction time and costs increase. Over the last several decades, this combination of SWIFT standards and correspondent banking relationships was the standard method for moving money across borders resulting in cross-border payments revenue being captured by banks. This model is not economical, especially for low-value transactions. This enabled non-bank providers to make inroads into cross-border payments to address the growing small and medium enterprise (SME) and consumer markets.

Challenges with the current model

Several challenges exist with the current model, including:

1. Cost. Since multiple stakeholders, like financial institutions, payment processors, and market infrastructures, are involved in cross-border transactions, coupled with foreign exchange components, transaction costs remain elevated. At a minimum, there are four stakeholders—the originating, receiving, and correspondent banks, and SWIFT in a transaction.
2. Speed. Cross-border transactions used to take three to five days but now take less than a day. In some locales, it can still take two to three days to complete a transaction.
3. Opacity. In the end-to-end transaction chain, each entity functions differently because it belongs to different jurisdictions and is in different operating environments. This results in opacity; one cannot know the transaction’s status easily, especially in an “exception” scenario.
4. Interoperability. Because of the nature of the current model, cross-border payments only work between banks and SWIFT. In today’s diverse financial ecosystem involving fintechs, payment service providers (PSPs), and many other players, this model does not offer suitable entry points.
5. Geographical reach and accessibility. Even though SWIFT provides connectivity to multiple countries, each bank offering cross-border transactions is limited by the number of correspondent bank relationships it has across countries. This also results in limited accessibility, especially since many countries still don’t have easy access to cross-border payments. Regulatory requirements also play a role in the reach and accessibility of cross-border transactions.

Recent initiatives in RTCBP

As part of the Single Euro Payments Area (SEPA), the Trans-European Automated Real-time Gross Settlement Express Transfer System (TARGET) was introduced in 1999 as a Real-Time Gross Settlement system (RTGS) for high-value payments. TARGET had a decentralized structure and linked together 16 national RTGS systems and the ECB payment mechanism (EPM). A more advanced version, TARGET2, was introduced in 2007. It was developed as a single shared platform with a high degree of centralization and standardization as compared to TARGET. Even though they are not truly RTCBPs, this is considered cross-border as more than 16 countries use this platform to process EURO-based payment transactions.

Other initiatives include:

• TARGET Instant Payment Settlement (TIPS) was developed as an extension of TARGET2 for low-value payments for anyone to use at any time.
• SWIFT GPI was launched in 2017 to improve speed and transparency in cross-border payments. It was developed based on the traditional correspondent bank model and helped in reducing transaction time to a couple of hours.
• IXB is a pilot service that is intended as a true RTCBP. It will leverage existing Real-Time Payments (RTP) platform in the United States and RT1 platform in Europe with support from SWIFT. This is expected to be launched sometime in 2023.
• The Unified Payment Interface (UPI) is a real-time payment platform developed by the National Payments Corporation of India (NPCI). The UPI app allows people to send money through their cell phones to other bank accounts. It is immediate and users can send money at any time. UPI is available in more than 10 countries and India has signed agreements with 13 more countries that want to adopt UPI.
• IXB and UPI are the only true RTCBPs, but both these initiatives are driven by a narrow group of stakeholders that are not truly global. They are not international RTCBP platforms that would ideally cover major economies. Other initiatives mentioned are not truly RTCBP.

Global RTCBP system

The IXB platform is a bilateral cross-border payments model and UPI is an enhanced version of this model. TARGET is a hub-and-spoke platform model while TARGET2 evolved into a common platform where only one central payment platform was used to process payment transactions across the SEPA region. The experiences from the implementation of all three of these platforms (IXB, UPI, and TARGET) can be used to develop a true backend multilateral RTCBP.

A major requirement that these platforms lack is the need for multi-currency support, where transactions can be carried out among multiple currencies at any given point in time. Other than this key requirement, the features of these systems are excellent and are essential for building a robust platform.

A backend multilateral cross-border platform, driven by a multilateral forum and endorsed by global financial organizations and watchdogs, is needed for the future of cross-border payments. For a system to be sustainable, efficient, and fully utilized, one central payment system will have to work in coordination with multiple regional payment systems. This backend platform should also support multiple commonly used currencies across the world.

Some key features that should be part of any global RTCBP are:

• Foreign exchange (FX): Exchange rate management and currency conversion
• Payment messaging format to support initiation, submission, authentication, and processing: Develop a global standard message format.
• Clearing including netting: Multilateral clearing with support for netting
• Settlement: Real-time gross or net settlement with transaction finality and ensuring technical settlement
• Liquidity management: Intraday credit facilities
• Compliance and data management: Know your client (KYC), anti-money laundering (AML), fraud monitoring, and data privacy.
• Anytime accessibility: Should be able to work round the clock to manage time zone requirements of different countries.

One key design consideration is to choose between the hub-and-spoke model or the central common platform model. This aspect may impact the development of platform rules and procedures. In a hub-and-spoke model, the hub can operate with minimal uniform rules if key issues of spokes across jurisdictions are consistent. In a common platform model, it’s important that the platform’s rules and procedures are consistent with the pertinent regulations and laws of the jurisdiction in which it operates.

A hub-and-spoke platform may be able to leverage existing relationships between

spokes and their participating financial institutions or payment service providers for domestic payments, thus allowing stakeholders to focus on issues surrounding cross-border payments via a hub. For a new common platform, it’s essential that all stakeholders agree on most of the platform’s activities. Because the common platform is less technically complex, it may be easier to operate and maintain than a hub-and-spoke system. A common platform usually is created on a single technical infrastructure while a hub entity and individual spoke systems may be built on completely different technical platforms. Updating a hub-and-spoke system may require tailoring a technical solution for each of the spoke systems in addition to changing how the newly updated spokes exchange information via a hub entity. A common platform can offer a more reliable service to users than a hub-and-spoke system.

RTCBP and ISO 20022

ISO 20022 adoption for cross-border payments and reporting (CBPR+) began in March 2023. This is an important milestone for the global payments ecosystem. In the pre-ISO 20022 era, many domestic payment systems across the world did not adopt ISO 15022. With ISO 20022, there is a coordinated effort by most countries to migrate to ISO 20022 with some minor variations to suit their own requirements. Along with this, ISO 20022 – CBPR+ will be adopted by all the countries and, starting in November 2025, this will be the only messaging standard to be used for all international payment transactions. Even with domestic payment systems across countries, everyone is expected to move completely to ISO 20022 by 2025.

Migration to ISO 20022 will contribute significantly to supporting the implementation of a global RTCBP. A harmonized version of ISO 20022 will help improve the quality of transmitted data and also with straight-through processing (STP), which is key for any successful real-time payment system. This will avoid the need for any messaging format conversions across participants. Additionally, new technologies like API and microservices will further help connect such platforms with existing payment systems for coverage that may be needed to make such platforms successful.

Multilateral platforms could play a key role in easing the frictions of cross-border payments and achieving the global targets for cost, speed, transparency, and easy access but some risks and challenges can limit their potential. Although several multilateral platforms like TARGET2 are in operation today, most are regional in scope. Increased public sector participation and involvement of central banks across the world can help address some of the risks and challenges.

About the Author: 

Niranjan Govindaram is a corporate banking subject matter expert with specific expertise in payments and cash management. Niranjan holds a master’s degree in business administration with specialization in strategy and management from Queens University, Kingston, Canada. He can be reached at [email protected]