Transforming the Workflow of Global Trade Transactions
12.57 Blockchain appears to be an attractive solution compared to sorting out the mess with many of the legacy computer systems upon which conventional forms of trade finance, such as letters of credit, depend because many legacy computer systems are in such a state of disrepair.[1195] Even if blockchain does not transform trade finance in quite the way some of its promoters have suggested, however, the excitement about blockchain may nevertheless make ‘boring back-office coordination sexy, which means that it might actually get done’.[1196]
12.58 Banks are by no means the only large organisations struggling to bring legacy computer systems up to par with computer systems designed today.
Industry experts estimate that 80% of the world’s corporate data reside on, or originate from, mainframe computers.[1197] A major reason that it took twenty years for US federal agencies to build a national single-window system is that so many federal computer systems are woefully out-of-date and very difficult to maintain.[1198] Most banks are in a similar position: a recent academic study of bank legacy migration problems estimated that banks are spending around 75% of their ICT budgets merely keeping their legacy systems up and running.[1199]size=1 color=black face="Times New Roman">12.59 The diversity and complexity of the ‘as is’ state of these legacy systems, the lack of a clear consensus regarding the appropriate ‘to be’ state that these systems should be moving towards, and the desire of those operating them not to make public the challenges they face in keeping them running, create significant collective action problems.
A collective action problem arises whenever the result of actors making rational decisions in response to the incentives they face as individuals produces a collective outcome that no individual wants.[1200] Some private sector efforts, such as the ‘Banking Industry Architecture Network’, are trying to help overcome these collective action problems by developing voluntary, consensus interoperability standards for financial services information systems, but progress in implementing such strategies remains slow.[1201]12.60 Cross-border trade transactions also tend to be much more complex than similar transactions carried out within a single country. A study of the export of sugar from Thailand to Bangladesh identified nineteen distinct steps required to successfully complete the entire
transaction,[1202] approximately thirty-one separate documents were required to complete the cross-border transaction and thirteen different stakeholders were involved. A sale by a Thai sugar-producer to a Thai customer would normally involve many fewer steps and stakeholders. As a result of their greater complexity, formality, and diversity of stakeholders, the challenges of business process reengineering for cross-border trade processes are usually much greater than those for commerce carried out within only one country.
12.61
The division of labour was first identified by the economist Adam Smith in 1776 as a necessary step towards increasing economic productivity.[1203] Labour segmented into many separate tasks could be combined with inflexible machine processes to produce enormous gains in productivity in the nineteenth and twentieth century. In the twenty-first century, however, the rigid segmentation of work into discrete tasks was often seen as a source of inefficiency and waste.
For this reason, business process improvement normally begins by breaking through thickets of ‘functional silos' of rigidly segmented tasks in order to see the underlying ‘end-to-end' business process.[1204]12.62
Even when the steps in business process improvement have been decided, it does not follow that it will be easy to execute them successfully. On the contrary, the conventional wisdom in management circles is that from 50% to 70% of attempts at business process improvement fail.[1205] Common causes of the failure of process improvement efforts include poor integration with an organisation's most important strategic goals; insufficient attention to the needs of stakeholders, such as customers; superficial, high-level analyses that fail to identify the challenges accurately; and poor communication.[1206] Variations in the capacity to engage in the process of ‘process improvement' or ‘quality engineering' can play a major role in the success or failure of business process reengineering efforts. ‘Process improvement' refers to ideas first described by Walter Shewhart in 1924 and later set out in detail in his 1931 book.[1207] Shewhart's protege, Edwards Deming, developed the plan-do-study-act method for implementing Shewhart's basic idea of creating a feedback loop to assist in monitoring any process that is repeated. A 2015 survey by McKinsey & Company demonstrated the continued relevance of process improvement in business: they showed that the odds of success for a ‘business transformation' project increased from 26% to 79% when managers consistently followed the engineering discipline of process improvement.[1208]
12.63 Trying to accelerate the pace of business process improvement by tinkering with bits and pieces of existing processes, rather than doing the hard work of isolating and analysing the end-to-end process that underlies many smaller functional processes, is a recipe for failure. This is because trying to optimise individual parts of a process is not the same as trying to optimise the whole process and may be worse than doing nothing at all.[1209] Technological change can intensify this problem, rather than ameliorating it.
As Bill Gates observed, ‘[t]he first rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency.'[1210] The slow progress in getting disparate federal agencies to reach agreement on common data standards for an integrated single-window system, and then actually to implement them, illustrate how difficult it can be to overcome these challenges.12.64 Blockchain technologies enclose information in a secure wrapper, but cannot themselves guarantee the quality of the information that goes into the wrapper.[1211] In order to ensure that the quality of information used is the same, or better, quality after changing the technology for trade finance as it was before that change, a thoughtful, deliberate effort at business process reengineering must be undertaken. This includes establishing process context, scope, and goals; analysing the current or ‘as-is' process by modelling its workflow and identifying factors that have an impact on process performance; designing the new or ‘to-be' process'; and developing the process for moving from ‘as-is' to ‘to-be', often at the same time as maintaining current operations.[1212]
12.65 The ‘to-be' state for many businesses engaged in cross-border trade includes a version of enterprise resource management software that combines software to manage logistics, supply chain operations, risk management, and compliance.[1213] In 2018, the Gartner consulting firm noted that ‘global trade management' (,GTM') was emerging as a distinct market for business software products that combine logistics, trade compliance, and finance functions.
Many of the anachronistic features of cross-border trade, and much of the complexity that blockchain proponents believed blockchain could alleviate, are already being alleviated by more conventional software applications that are compatible with both legacy IT systems and emerging IT systems.12.66 Global trade management software systems may help large trade intermediaries, such as global third-party logistics services and large enterprises engaged in cross-border trade, but they are too expensive and demanding for many small traders and intermediaries to be able to use, especially those in emerging economies. While no open-source, lightweight global trade management software is yet widely available for smaller intermediaries and traders in emerging economies, UNCTAD has developed an open-source software program to
help automate customs functions in emerging economies. In 2019, UNCTAD's Automated System for Customs Data (‘ASYCUDA') had been adopted in ninety countries, or more than half of emerging economies.[1214] As more emerging economies begin constructing singlewindow systems in order to come into compliance with their WTO TFA obligations, the number of emerging economies implementing ASYCUDA is likely to increase.
12.67
Efforts to modernise cross-border trade workflows have been underway for decades and are beginning to show results. The WTO TFA creates a framework within which efforts to solve longstanding collective action problems, such as migrating legacy computer systems to newer technology architectures and expanding the scope of automated processes by standardising workflows, are making progress using legacy technologies. The unique features of technologies based on blockchain—consensus decision-making and secure, tamper-resistant records—do not appear to address the most important issues in the business process reengineering of cross-border trade flows that would have emerged if a serious workflow modelling effort had been undertaken. When the chief operating officer of a blockchain trade-finance platform resigned in 2019, he publicly voiced doubts about the relevance of blockchain solutions to meeting industry challenges: ‘I have not yet seen something that shows the ultimate benefit of the technology ... We are building solutions that are perceived as valuable by the providers of the solutions, not the users.'
VI.