This appeared a couple of days ago.
‘They were building a Frankenstein’: how ASX’s blockchain unravelled
James Eyers Senior Reporter
Nov 17, 2022 – 3.41pm
When Elmer Funke Kupper jumped on the blockchain bandwagon in early 2016, announcing ASX would invest in New York-based start-up Digital Asset to deploy the emerging technology in an upgrade to the market’s clearing and settlement system, his intentions were good.
At the time, the ASX CEO figured a distributed ledger, similar to the technology underpinning the bitcoin cryptocurrency, would allow ASX to create a “single source of truth” around equity ownership to make the market more efficient.
It promised to streamline a multitude of complex processes connecting listed companies and their investors in the equity market by allowing everyone to see who owns a particular stock at any point in time.
The system could help reduce some of the $4 billion to $5 billion it costs to run the equities markets each year, Funke Kupper explained to The Australian Financial Review Business Summit in 2016.
The mammoth cost is due to a series of historical compromises that have created a web of inefficient relationships between many participants in the market, from brokers and banks, to custodian and registries, to wealth platforms and managers of corporate actions.
CHESS, the record of equity ownership that underpins all activity, works by the ASX and all the different computer systems used by brokers and the other players in the market sending messages to each other. Stockbroker systems have maintained copies of registers. When a company needs to pay a dividend or conduct a capital raising, it’s sometimes hard to be sure who holds the stock, especially given settlement occurs up to days after a trade is made.
In an ideal world, blockchain might compress that T+2 settlement time to real-time by allowing payments and ownership records to move simultaneously. A lot of the reconciliation and administrative processes taking place all day in broker and investor offices could be streamlined if everyone knew everything about an equity simultaneously.
Blockchain had arrived at a good time for ASX: regulators wanted it to replace an ageing CHESS system, written in an antiquated programming language called COBOL that used proprietary messaging standards that were expensive to maintain and retrofit into newer trading systems.
Funke Kupper figured ASX could help to create new data standards for market innovators to build upon. Investors and listed companies would be better off, costs would reduce, and Australia could become a genuine world leader in financial market innovation.
But this week – after a series of delays on the Digital Asset project starting in 2020 – his vision is in tatters. ASX said on Thursday it would shut down the project and take a $250 million write down, after Accenture identified a litany of problems, including excessive complexity.
Two months after ASX said it would invest in Digital Asset, then run by former JP Morgan investment banker Blythe Masters, Funke Kupper was forced to step down from his role as CEO and passed the baton to Dominic Stevens. In time, Stevens would realise his predecessor’s vision would be tough to deliver in reality.
At the heart of ASX’s spectacular failure was the unsettling nature of innovation to all those market intermediaries: what the ASX saw as potential savings, registries and other companies profiting handsomely from clipping the tickets on trading and company actions saw as an attack on their revenue bases. Australian investors pay about $1 billion each year for registry services.
‘Overly complicated’
The response to threats of disruption were then compounded by ASX management failings, especially an inability to showcase why the changes were necessary and setting unrealistic timelines on project delivery.
By 2019, it had dawned on registries like Link Market Services and Computershare they needed to mount a rearguard action to slow ASX down. They were concerned the ASX was using a monopoly power in clearing and settlement to move onto their turf.
A lobby group known as the CHESS Replacement Stakeholder Group was founded; the registries were joined by industry groups the Australian Investor Relations Association, the Governance Institute of Australia, the Stockbrokers and Financial Advisers Association and the Australian Shareholders Association. Concerns were raised that the ASX had adopted a tin ear when it came to market consultation.
Much of that criticism was justified. The ASX failed to properly explain the benefits of the new system including the sorts of services that might be built on top of it. This made it hard for companies initially supportive of the project, including several large investment banks and broking houses, to determine the return on investment as they tried to understand the new technology and how to connect to it.
A series of compromises were made. Registries role would be preserved. Delayed settlement times would stay in place. All the intermediaries in the market would keep their roles. But inevitably, this meant that the potential for the technology would never be realised.
“When you go to something like blockchain, you need to rethink everything from the ground up,” says Wayne Baskin, co-founder of Superhero, which has developed a new trading platform for stocks.
“The ASX project was overly complicated. You see this in so many projects: you think you are going to build an amazing new system, but then one stakeholder says what about this, and another says what about that. And all of a sudden, you are building a Frankenstein.
“Blockchain is a great technology. But when you are trying to cram blockchain to redo what you are currently doing, that is where you are going to struggle.”
Others agree that the ASX plan was just too ambitious, and the exchange bit off more than it can chew.
“When you are driving in the desert and your life depends on it, you need a diesel landcruiser and not a hybrid Ferrari,” says Vic Jokovic, chief exectuive of Cboe Australia, which competes with ASX on equities trading.
Stevens, while seeing the ultimate benefits to the market of more streamlined information flows, struggled to provide clarity about the project’s timetable. The ASX ran into plenty of trouble setting deadlines it could not meet.
After initially targeting a go live date of late 2020 or early 2021, when the coronavirus hit in March that year, ASX said the date would be delayed to April 2022. Then, in October 2020, it announced another year-long delay to April 2023, citing the need to ensure the system was able to withstand extreme trading volumes like those seen during the pandemic.
Market participants, who were busy training up staff on the new programming language, started to lose the faith. Compounding the issues, the ASX started to lose senior members of its executive management team with carriage of the project, most notedly deputy CEO Peter Hiom, who left in May 2021.
After giving assurances the project was almost fully complete, Stevens said in February this year he would bow out. A month later, ASX confirmed the upgrade will be delayed for the fourth time, after a three-month delay to a software upgrade from Digital Asset. Then in May, it said the April 2023 start date had been abandoned, and in August announced the fifth delay, saying it did not expect it to go live before late 2024.
New CEO Helen Lofthouse called in Accenture to assess the project. The writing was on the wall.
More here:
There are more details here:
ASX kills its blockchain project, will write off $250 million
ASX will write off $245 million to $255 million pre-tax in costs associated with the project, which has dragged on for seven years.
In a devastating blow to the credibility of the exchange and to distributed ledger technology more broadly, the independent report by Accenture highlighted significant gaps and deficiencies with the design of the system and ASX’s ability to deliver it.
Reserve Bank governor Philip Lowe described the ASX announcement as “very disappointing”.
The Accenture report was commissioned in August after the project’s fifth delay. ASX has now gone back to a “solution design” stage for CHESS replacement and will reassess all options to create a new clearing and settlement system while maintaining the legacy infrastructure.
“Current activities on the project have been paused while ASX revisits the solution design,” ASX said. “All stakeholder activity on the project will be paused, and the industry testing environment will be closed.”
“Blockchain is a great technology. But when you are trying to cram blockchain to redo what you are currently doing, that is where you are going to struggle.”
ReplyDeleteHow true is that generally when it comes to technology as a transformation tool. If I had a dollar for every project that failed because the sponsor wanted a shiny new turd and refused to change I could buy a near new French sub.
As a previous commentator pointed out a healthcare organisation has stuffed up an ERP implementation- I bet it was headed up by a CFO and run by the finance team completely ignoring the rest of the organisations functions and process - a classic pattern. Don’t envy those involved CFO’s love throwing staff under buses
Just a bump along the way, email was not perfect in the beginning, and the London stock exchange failed on its first IT day. The technology is I believe important, it is a public infrastructure, transfers trust away from middle people/organisations. The only alternative is cash but that can only happen in person.
ReplyDelete