Lucas is an importer of Brazilian teas and spices in upstate New York. When paying his suppliers, his local bank charges him almost $100 per payment, which average about $1,000. When the process works well, his suppliers receive their payments almost a week later. However, they often complain about incomplete payments because destination processors have also deducted fees. Occasionally, his payments never arrive, going astray in a maze of international exchange and paperwork.
All this might soon change. Five years ago, bitcoin was capturing headlines with its promise of fast, low-cost, and secure cross-border payments for consumers and businesses. But the darker side of human nature soon emerged, generating new headlines of theft and fraud that tarnished cryptocurrency’s reputation.
Today, bitcoin lives on. But perhaps more importantly, its underlying distributed-ledger technology, known as blockchain, is being tested for its ability to deliver similar benefits for various forms of payment supervised by financial institutions and regulators. Consequently, advances in consumer peer-to-peer payments and the proliferation of pilot programs using blockchain technology are pressuring incumbent banks and traditional money-transfer operators to develop and improve their offerings. This promises to soon bring a step-change improvement in associated cost, speed, and transparency.
To provide a balanced perspective on how blockchain could disrupt cross-border payments, we examine the many challenges that today’s longstanding processes present and a promising technology that threatens to disrupt this model. Based on McKinsey’s research and experience in working with prominent payments industry participants, we evaluate the potential of three new operating models and identify key criteria for achieving their full promise.
In 2015, cross-border payment flows totaled more than $150 trillion. Most of this volume was consumer driven, but the value was largely generated by businesses. In this same period, the payments industry earned over $200 billion in revenue from services provided to payers and payees (nearly 80% resulted from B2B transactions). Most cross-border payments are still routed via bilateral correspondent banking relationships, a network of banks that use the SWIFT messaging protocol to execute transactions. Because each bank performs a function in the value chain, fees accompany each transaction. The sum of such fees occasionally exceeds 10% of a payment’s value (Figure 1).
The primary advantages of the traditional correspondent banking system are:
Generally consistent process standards
Relatively comprehensive global reach (SWIFT has more than 10,000 member institutions in more than 200 countries)
Guaranteed security, a benefit tested by recent events
Yet unsurprisingly, this network system also frequently produces pain for its users, even as it generates attractive returns for participating banks. Network users suffer on three accounts. First, pricing and foreign exchange rates are not finalized until the funds actually arrive in the recipient’s account. Second, the need to rely on multiple banks causes uncertainty about timing, accuracy, and transfer status. Lastly, both sender and receiver must have an active bank account where they live. Yet almost 40% of the world’s adult population is unbanked.
Three Aspects Cause Most of the Pain:
Speed: According to proprietary McKinsey research and analysis (2015) on cross-border payments, the average time to complete a cross-border transaction is three to five business days, which includes the final mile transfer via a domestic payment network, such as Automated Clearing House.
Cost: Fees accumulate at each step in the process, including transfers from the sender’s bank to the national correspondent bank, from one correspondent bank to another, and foreign exchange fees. Fees for cross-border payments where volumes are high usually average 2% to 3%, but can exceed 10% where payment volumes and values are low. And it’s not always clear when costs will also be charged to a recipient.
Opacity: It’s usually difficult for senders and receivers to track their payments while the funds are in transit, creating uncertainty about both delivery timing and the final payment amount. It can be especially difficult to quickly trace transactions when problems arise, such as incorrect account numbers.
Despite the often-heavy burden these disadvantages place on users, most high-value payments are still made via the correspondent banking network. This is largely due to familiarity, security, and the ability to transfer larger sums than alternative channels currently accept. Another factor is the more consistent and thorough know-your-customer compliance checks that incumbent banks provide.
Several initiatives are underway to alleviate the pain points of cross-border payments in traditional correspondent banking. For example, SWIFT is testing programs under its Global Payments Innovation (GPI) initiative that attempt to reduce cost and improve speed by requiring banks to follow strict business rules that facilitate straight-through processing. The program promises same-day use of funds, more up-front fee transparency, cloud-based end-to-end tracking and visibility, and richer payments information. In September 2016, SWIFT announced that almost 90 banks around the globe that, combined, process 75% of the cross-border payments on its network, were participating in the program. SWIFT expects to fully roll out the program in 2017.
Similarly, national efforts are underway to migrate to faster payments technologies. Some aim to enable national real-time clearing and settlement. Examples include The Clearing House’s real-time efforts and the U.S. Federal Reserve’s Faster Payments Task Force. Internationally, such programs are now live in more than 30 countries that, combined, account for almost half of all global credit transfers. As the move toward faster payment technologies gains momentum, the desire for systems interoperability and integration to enable real-time cross-border payments will grow. This will, however, require increased complexity and collaboration to handle foreign exchange functions.
Prime Time for Innovation
Inertia and strong reliance on existing network effects have long characterized the payments industry. However, for several reasons, we believe new operating models will present challenges for the cross-border segment.
Expectations: As instantaneous, consumer, peer-to-peer payments solutions such as MobilePay, Venmo, and Xoom continue to spread, businesses increasingly will seek to replicate that convenience, speed, and simplicity in the commercial world.
Innovation: The development of new payment technologies has been accelerating. This trend reflects the strong potential for revenue and profit gains, as well as the increasing adoption of mobile digital solutions for everyday tasks, including payments. The emergence of bitcoin and other cryptocurrencies introduced blockchain technology to a wider audience, increasing the potential for a fundamental change in systems architecture throughout the financial services industry.
Compliance: While users are pressing banks to keep pace with change, regulators are demanding more in equal measure. This dual pressure is causing banks to review their expansive correspondent banking relations as they seek new ways to reduce costs. Increasingly, they are seeking more efficient network participation while tightening their standards to minimize errors and manual interventions.
To successfully confront these forces for change, the payments industry must face the likelihood that traditional methods of sending and receiving cross-border payments will irrevocably change during the next few years. This will result in payments being implemented at entirely new levels of speed, cost, convenience, and transparency.
Given that today’s international correspondent banking ecosystem is effectively a decentralized global network, it’s unsurprising that innovation tends to arise from analogous decentralized network solutions. Notably, there has been significant recent discussion about the potential of using blockchain technology as the backbone of a new cross-border payments messaging system. While blockchain transfers don’t physically move digital assets, they do irrevocably record changes in asset ownership. And depending on the nature of the asset, some local exchanges are willing to provide real-time liquidity in local currency in return for ownership of the respective assets.
Ideally, blockchain technology can be used to manage digital money. In addition to ongoing discussion about the potential of bitcoin, other models involving purely digital forms of commercial and central-bank money are being advanced. In particular, we note the appearance of the following model types:
Money-transfer APIs: These are mobile peer-to-peer applications that either use a payments directory to map participants’ mobile IDs to conventional demand deposit bank account routing information, or require users to manually link their accounts using bank APIs and the domestic payments network. In terms of implementation, this solution is the most advanced. However, with a few exceptions, such as MobilePay in Scandinavia, it has limited cross-border utility.
Cryptocurrency solutions: These are closed-loop solutions that use digital global currency tokens, are registered on a distributed ledger, are protected by the latest encryption methods, and rely on a sound funds model for effective push-only payments. Such solutions depend entirely on networks of independent brokers or banks to exchange cryptotokens into local fiat currencies. Alternatively, end users must adopt and accept the digital tokens as currency. Current designs typically have no recourse provision beyond initiating a matching transaction in reverse.
Central-bank digital currency: In this model, central-bank-denominated digital notes are held and exchanged (effectively, reattributed to new owners) in real time on a central ledger (digital cash as good funds). In this model, the central bank becomes an effective administrator of deposit accounts. To be accepted as legal tender, this type of solution requires new monetary policy. However, it also offers the promise of a central database to enhance the end users’ experience (for instance, by providing accounting and purchase data) or coordinated industry services, including taxation and fraud detection.
What a Closer Look Reveals
The foregoing models differ significantly from each other in both their structure and how they could disrupt payments generally, and cross-border payments more specifically. The details are more telling.
MONEY TRANSFER APIs: The global spread of digital banking is enabling money transfer operators (MTOs) to move their services online and thereby provide more seamless, rapid, and affordable alternatives for remittance payments. Users naturally are finding these more appealing than traditional cash-based models. Venmo, Xoom, TransferWise, and other players have already increased their market shares substantially, and are expected to grow nearly 10 times faster than traditional retail MTOs from now until 2020.
Digital MTOs still use prefunded nostro accounts (accounts held in a foreign currency at another bank) in destination countries. And they can incur substantial short-term risk when relying on conventional domestic payment rails to source and deliver customer funds. Because digital MTOs can offer immediate (or next-day) funds availability, they generally must establish sufficiently large credit lines locally to honor the daily fluctuations in transaction volume demand. This presents unique difficulties when honoring commercial cross-border payments, which tend to be much larger. This, however, has not stopped traditional MTOs such as Western Union and MoneyGram from expanding their focus from consumer to more business-centric payments, with platforms such as WU.com.
CRYPTOCURRENCY SOLUTIONS: Since its launch in 2009, bitcoin has often captured headlines for the wrong reasons. Variously associated with criminal activity, theft, and a libertarian movement that focuses on disintermediating banks, it has generally been deemed inappropriate for mainstream financial services. Nonetheless, the average dollar-equivalent value of daily bitcoin transactions has risen steadily to $160 million in recent months, equivalent to 25% of global consumer-to-consumer remittance flows processed by MTOs (and larger than the daily volume processed by digital-only MTOs). Meanwhile, several technology startups have received substantial venture capital backing to launch pilot programs that offer cross-border payment services based on bitcoin transfers.
Align Commerce is a good example. It has leveraged the liquidity of bitcoin to local currency exchanges in strategic cross-border corridors to offer real-time, transparent, and low-cost cross-border payments to small and midsize enterprises. Its platform provides a unique visual dashboard that enables counterparties to track the progress of payment transfers with up-to-the-minute accuracy and implicit payment guarantees.
Abra has followed the mobile “uberization” trend by offering peer-to-peer, real-time, cross-border payments via roving mobile tellers who can accept cash payments in exchange for instant-payment bitcoin transfers to any recipient with online or mobile access. And funds can be withdrawn from a similar mobile teller in the destination country such as a teller agent in the active corridor between the U.S. and the Philippines.
Less user-friendly applications require users to actually purchase bitcoin from a public exchange before initiating a transfer. These purchases incur a cost (established providers in the U.S., such as Coinbase, Circle, and itBit, typically charge a 1% transaction fee). And moving from an originating currency to bitcoin and then back to the destination currency is subject to market bitcoin foreign exchange rates. Combined, these fees can amount to 2% to 3% of the principal, close to the costs of digital MTOs and the lowest-priced correspondent banking corridors.
It is widely believed that over 90% of bitcoin’s daily transaction volume is generated by speculators, rather than by buyers of traditional goods and services. (The average value of daily exchange-traded volume is close to $30 million). However, the substantial volumes and endurance of bitcoin are prompting some to look more closely at its underlying technology and to explore how it might be used to more rapidly and securely execute and track transfers of value while creating an immutable audit trail.
One example that employs such a modified form of distributed ledger technology is Ripple. Using its open source distributed technology protocol, in-network banks can initiate payments in Ripple cryptotokens and have them routed via the most efficient foreign exchange pathway. Payments can then be settled by in-network receiving banks. Ripple’s protocol is currency-agnostic and confirms transactions within seconds. Several large financial intuitions are already using this technology, including 15 of the top 50 global banks. Participating banks are part of a growing network of innovative financial institutions currently experimenting with blockchain technology.
However, not everyone has been quick to welcome this technology. Governments in several countries – most notably China, Russia, and Iceland – are taking more defensive positions by passing legislation that effectively makes banks’ use of cryptocurrencies illegal. Moreover, using the technology across multiple jurisdictions makes it difficult to provide robust user guarantees, and even more difficult decisions about who to prosecute in the event of failures.
CENTRAL-BANK ISSUED DIGITAL CURRENCIES: Several central banks have publicly discussed their intentions to investigate and experiment with blockchain technology, including those of Canada, England, the Netherlands, and South Africa. Each has recently clarified its policies regarding its own issuance of central bank digital currency (CBDC), and many have created teams to explore the implications of doing so.
In June 2016, the Bank of England announced that it had created its own distributed ledger as a proof of concept. Its goal is to better understand the opportunities and challenges the technology presents before adopting it. Among its specific concerns are customer privacy, scalability, data integrity, reliability, actual speed, and required volume. Despite these concerns, subsequent evaluation of CBDC macroeconomics finds that by issuing 30% of gross domestic product in the form of CBDC stock against government bonds could permanently raise GDP as much as 3%.3
Further, in July 2016, the Bank of Canada, working with the CPA and technology firm R3, successfully completed a pioneering exchange of interbank blockchain payments among Canada’s largest banks. While Bank of Canada made clear it has no plan to issue digital currency for public use, it is pursuing proof-of-concept technology for wholesale interbank payments. Similarly, within the last few months the central banks of France, the Netherlands, and South Africa have described early experiments with blockchain-based currencies.
The path to adopting digital forms of traditional fiat currencies, however, has several major hurdles. At the IMF meeting of central banks in June 2016, several core issues were identified, including legal and operational issues, monetary policy changes, privacy concerns, and transition challenges. And while theory suggests that governments should take the lead in issuing such currencies in the future, pronouncements by central bank governors, such as the Bank of England’s Mark Carney about CBDC, make clear that adopting this model will take considerable time.
Key Criteria for Realization
The three model types discussed above vary significantly in their degree of scale, adoption, and practicality of implementation (see Figure 2). That said, all three models show promise and are attracting substantial investment (close to $1 billion of venture capital in the past two years). But to achieve their promise of greater speed, affordability, transparency, security, and convenience in cross-border payments, new solutions need to satisfy such requirements as overcoming incumbent resistance, motivating user adoption, and providing regulatory compliance.
Although these new solutions could be beneficial for businesses, they could also be detrimental to the margins earned by the banks offering them. And this is the very reason that initiatives to upgrade the correspondent banking system have generally faced resistance from incumbents.
To succeed, new solutions will need to have relatively low costs, be technologically easy to integrate, offer clear advantages such as lower compliance risk, and help banks to be competitive while destroying healthy margins. More importantly, new systems will need to deliver a step change in security. A major attribute of blockchain technology is its use of leading-edge encryption combined with multifactor authentication at critical transaction stages.
The success of any new payments alternative depends on the benefits it offers to both issuers and users. For blockchain technologies, the ability to track value ownership might be a benefit to governments; however, an accompanying loss of anonymity for consumers would raise obvious privacy concerns. From the commercial viewpoint, the benefits would be much more balanced. Businesses could easily execute, track, and prove payments through low-cost channels, and be able to automate many business functions, such as tax preparation and auditing.
The payments industry is among the most heavily regulated, with efforts to circumvent compliance being promptly illuminated and punished. In 2015, for example, Ripple was fined by the Financial Crimes Enforcement Network for failing to comply with a requirement that virtual currency businesses follow established anti-money laundering regulations for money transmitters. While recent fines on non-bank MTOs have often been nominal, they reflect the industry’s desire to reinforce the need for proper compliance. Whether new entrants will ignore compliance requirements (as Uber has done) remains to be seen, but wider adoption of innovative technologies by banks must inevitably satisfy their most stringent know-your-customer and AML compliance obligations.
The correspondent banking network has served its users well for centuries, serving as the sender bank’s local presence in far-flung geographies. But the expectations of money senders and receivers are changing, requiring new cross-border payment approaches that are more rapid, economical, transparent, secure, and convenient. With many traditional correspondent banks preparing to adopt SWIFT’s GPI initiative, many hope that wider adoption will soon alleviate some of the worst pain points currently endured by businesses needing to send money overseas.
Meanwhile, recent technological innovation promises similar payments improvement, albeit without extensive networks or central administrators. In particular, blockchain technology promises the direct exchange of tokens of payment value that enable real-time messaging and clearing within a cryptographically secure and resilient environment.
Early proofs of concept suggest that central banks and large financial institutions are receptive to blockchain technology participation, but substantial challenges remain and the path to widespread adoption will likely be long and uneven. In particular, real-time settlement using blockchain tokens remains a challenge, requiring commercial and central bank money to honor such tokens.
Whether improvements in cross-border payments are primarily driven by network effects, aimed at adopting and improving the technology stack over time (such as SWIFT’s GPI) or new solutions like blockchain that can deliver compelling benefits to inspire an entirely new network of peer-to-peer bilateral agreements remains to be seen. n
About the Author:
Matt Higginson is an Associate Partner with McKinsey & Company’s Northeast Office in New York. He specializes in payments strategy and banking operations and leads the firm’s global knowledge efforts in distributed ledger technology, collections operations, and payments security. Before joining McKinsey, Higginson worked as the national business unit head for forensic testing of illicit drugs and toxicology at the U.K.’s largest privately owned forensic science provider, where he managed multiple forensic testing laboratories and gave expert witness testimony at homicide trials at Crown Court.
Higginson holds a bachelor’s degree in geography from the University of Oxford, a doctor of philosophy degree in organic chemistry from the University of Bristol, and an MBA from the University of Oxford.