- Skip to:
Barclays and Wave claim to have closed the first letter of credit via blockchain. The new technology is now a reality and the pace of progress appears to be picking up. But will the hyped benefits – cost, speed and increased volume of trade finance transactions – be the driving force or will a race for technological market share hinder development?
Whether you are in the ‘bullish IT’ camp or believe a lot of the claims made about trade finance blockchain are a partial anagram thereof, no-one disagrees that the potential cost savings over traditional paper-based products could be significant.
Around 50% of bank costs for a letter of credit arise from manual document handling and checking. And given the extensive cross-selling potential of trade finance – according to a report by Accenture $1 in trade finance fees can bring an additional $1.7 in FX and cross border payment fees, and another $2.25 in other transactional banking revenue – a 50% cut in costs on elements of that one-stop-shop service is a major boost to the margins for trade finance lenders.
But traditional paper-based trade finance bankers have good reasons to be sceptical. The Bank Payment Obligation (BPO), a collaboration between SWIFT and the International Chamber of Commerce (ICC) to provide digital means of settlement in international trade has still not been widely adopted. According to Boston Consulting Group only 20 banks were offering BPO as of October 2015, including just six of the top 15 trade banks: The first live BPO close in the UK only happened in August (a deal via Commerzbank and UniCredit). And in some emerging markets, Myanmar for example, even paper-based letters of credit have yet to catch on.
However, scepticism – in part born of real concern, and in part because blockchain has the potential to cut jobs and move the balance of power in trade finance from relationship bankers to IT – is diminishing, particularly given the concept is moving from the theoretical to the actual quicker than many initially predicted. For a growing number of trade bankers it is no longer a case of will blockchain happen - but when it will happen.
Proof of concept
Barclays and Israeli start-up Wave are claiming the world’s first real trade financing deal via blockchain. Closed on September 8, the $100,000 letter of credit backs the export of dairy products by Irish agricultural food cooperative Ornua to the Seychelles Trading Company.
According to, Baihas Baghdadi, global head of trade and working capital at Barclays: “We've proved the reality of this technology and the client, Ornua, has asked us when they can do the next transaction, which proves how user-friendly the process was".
While the first to close a live trade deal via blockchain, Barclays is not alone in road-testing the new technology. A number of proof-of-concept initiatives have been launched in recent months, both in terms of platforms and niche add-on products.
In January, 11 member of the R3 consortium – HSBC, Barclays, UBS, BMO Financial, Credit Suisse, Commonwealth Bank of Australia, Natixis, Royal Bank of Scotland, TD Bank, UniCredit and Wells Fargo – tested a Microsoft payments platform running on a blockchain built by Ethereum.
And on August 10, both R3 and a consortium comprising HSBC, Bank of America Merrill Lynch (BAML) and Infocomm Development Authority of Singapore (IDA) announced separate proof of concepts for essentially the same thing – automated letters of credit (L/C).
The R3 testing involved 15 of its member banks - Barclays, BBVA, BNP Paribas, Commonwealth Bank of Australia, Danske Bank, ING Bank, Intesa Sanpaolo, Natixis, Nordea, Scotiabank, UBS, UniCredit, US Bank and Wells Fargo - using self-executing transaction agreements (smart contracts) on R3’s Corda distributed ledger platform to process accounts receivable purchase transactions (factoring) and L/Cs.
The HSBC/BAML/IDA initiative used Linux’s open source Hyperledger blockchain fabric, supported by IBM Research and IBM Global Business Services. The application mirrors a paper-intensive L/C by sharing information between exporters, importers and their respective banks on a private distributed ledger. The trade deal is then executed automatically via smart contracts.
The commonality between the two experiments goes beyond automating L/Cs. Both claim technological success and both claim significant cost savings over traditional paper-based trade finance. However, all these initiatives have also highlighted a number of hurdles - scalability, security, and the need for greater collaboration.
One of the major ironies of blockchain development is that the distribution and sharing element in shared distribution ledger technology could end up falling down the priority chain as a race for market share develops.
With participants to trade financings relying on different banks, for the information shared on the blockchain to be trustworthy and enabling, the different banks and parties - importers, exporters, government agencies, shipping companies, logistics operators and insurers - must agree on a common platform and set of standards for applications like smart contracts.
Competition or collaboration?
It is an issue that both R3 and HSBC/BAML/IDA are clearly keenly aware of. Despite having both worked on solutions to the same problem, and HSBC being a member of R3 (although it did not participate in R3s latest testing), both consortia were quick to issue statements highlighting that the way forward was industry-wide collaboration and partnering with complementary technology providers.
While those sentiments are echoed by many trade finance blockchain protagonists, actions by some banks would suggest this is as much a race for market share as global collaboration. For example, BAML applied for 20 blockchain patents earlier this year. Both UBS and JP Morgan have done the same and more banks are expected to follow.
The rush for patents may be defensive – no bank wants to end up being sued in the future by a start-up with a tweaked open source (otherwise known as forking) version of the same thing. And many of the patents will be for particular specific trade applications – diamond trading for example. But there is a danger that the architecture of blockchain trade finance, as opposed to specific applications, becomes a cash machine in itself rather than regaining development costs from generating improved profits via the actual business of trade finance.
If that danger becomes a reality, and recent history suggests it might given the retraction of many banks from low margin trade finance lending into higher margin transaction and cash managements services, the heavily hyped blockchain cost-savings expected to boost trade finance lending may not meet either borrower or lender expectations.
Scalability and security are also major hurdles that have yet to be overcome. Recently, the DAO - a crowdsourced venture capital platform, based on the Ethereum blockchain, which promised the ability to dispense with lawyers and financial institutions - fell victim to a $60 million hack soon after opening.
And after the HSBC/BAML/IDA trial, Vivek Ramachandran, global head of product for HSBC's trade finance business announced that the exercise had demonstrated “the technical limitations of distributed ledgers, in terms of the number of nodes you can have or the quantity of data you can have on it.”
With five major blockchain trade finance initiatives underway – the three already outlined, Ripple working with Standard Chartered and DBS; and JP Morgan with Digital Asset Holdings – and spend on solutions widely predicted to hit around $400 million per year, solutions to initial teething problems are not far off: R3 is estimating a smart contract product will be trading globally in three years.
Trade blockchain start-ups proliferate
In addition, the number of independent start-ups – for example Skuchain, Fluent, Wave and Zerado - ranging from smart contracts to sector-specific asset tracking throughout the supply chain, is snowballing.
Skuchain and Zerado are focused on unlocking cash locked up in letters of credit. While Fluent is a blockchain-based financial network and payment platform aimed at large enterprises and their global supply chains.
Fluent is currently in a pilot programme with a major bank and is also being demonstrated to potential clients in the US and abroad. The Fluent network functions via payments linked to tokenized invoices. Once a buyer approves an invoice, the goods are deemed satisfactory and the invoice is then paid directly to the financier. With the supplier never holding the funds, the risk of non-payment is eliminated.
In principle there is nothing radically new in the idea. But the system uses a custom-built, federated blockchain, where the nodes are hosted both with buyers and the financial institutions on the network, making it quicker and more secure than paper-based transactions.
In addition to the platform, Fluent has developed a suite of applications and solutions to increase efficiency, transparency, and flexibility. Its Global Payment Platform enables users to send and receive payments in real-time, both intercompany and with their suppliers worldwide. The Supply Chain Financing Platform allows suppliers to participate in one-touch receivables financing. And the Receivables Marketplace enables receivables to be sent to a multi-lender marketplace where there is competition on pricing between lenders, unlike traditional supply chain financing programs which are often hosted by one bank.
Corda or Hyperledger or both?
Of all the major initiatives underway, the Linux and IBM-backed Hyperledger and R3’s Corda-based offering are gaining the most traction. R3 has attracted 60 members, the most recent being MetLife (the first insurance multinational to join), while members on the Hyperledger project now number 82.
The difference between the memberships of both is stark. R3 claims to have over 50 of the world’s major financial institutions signed up. Hyperledger membership is considerably more industry-based with only around 10 pure financial entities in place.
Consequently, while banks grapple with the question of whether they should work together on a consensus model, or produce independent blockchain-based solutions and let market forces decide the winner, the bigger question is whether the two major approaches to blockchain – which, given their membership make-ups, arguably represent leanings toward a lender approach and a borrower approach – can or will collaborate more effectively to avoid duplication of effort.
Similarly, for all the hyped aims of blockchain trade – distribution, immutability, security and trust – development of shared distributed ledger technology requires that keyword ‘shared’ at its core if it is going to truly revolutionise trade finance.
And that revolution needs to be major – in effect bridging the $1.2 trillion (according to ADB estimates) gap between demand and availability of trade finance debt, and cutting costs to enable lenders to up margins while also passing some benefit on to borrowers. In short, future trade finance economic viability and blockchain go hand-in-hand.