Supplements » FundTech Spring 2020

GSTPA: The death of the GSTPA

Car_under_coverNicholas Pratt looks at the failure of an industry effort to establish operational standards 20 years ago and asks if any lessons have been learned.

It’s been almost 20 years since time was called on the Global Straight Through Processing Association (GSTPA), the industry’s ambitious attempt to standardise cross-border securities transactions. “We made a beautiful car but nobody wanted to buy it,” said Jürgen Marziniak, CEO at the time of its demise. 

The story of the GSTPA highlights the difficulty of establishing an industry-wide utility but also captures the craziness of a time when banks still saw themselves as the best bodies to create and control market infrastructure, even when they ended up on competing consortiums. In essence, the technology was never the problem, it was everything else – the politics, the money, the commercial conflicts and lack of governance.

Sound familiar? Twenty years on, the subject of utilities, collaboration and global standards are alive again as the industry considers how to harness the frightening potential of blockchain and distributed ledger technology (DLT), both to completely disrupt the current post-trade landscape and its central players but also make it more efficient, transparent and fairer for its ultimate end-users – the investors. 

But what lessons have been learned from the GSTPA story and can they be used to make a success of blockchain and other new technologies that require industry collaboration?

Funds Europe talked to some of the people who were involved at the time either as a custodian, a fund manager or rival tech providers. We also contacted some of the people who served as executives at the GSTPA right up to its demise but, given the years of legal wrangling over intellectual property rights that followed, they were understandably reticent to retread this ground. “Sorry to disappoint you, but I have closed my GSTPA after nearly ten years in the courts,” said one. 

Others, though, were more forthcoming. 

In 1999, Trevor Hunt joined UBS Brinson, one of a number of asset management businesses following the merger of Swiss Bank and UBS, as head of technology in the UK. One of his primary roles was to migrate the firm’s technology from voice and paper to end-to-end automation. “After joining UBS Brinson, I was amazed to discover the lack of automation in some processes,” says Hunt. 

There had been a push by some market participants to implement electronic trade confirmations between brokers and asset managers but beyond that, the settlement process was largely manual with a host of different protocols between regions and between custodians.

“We needed industry-wide standards between all participants and a central platform that could connect brokers to asset managers and custodians so that trade confirmation, matching, settlement and reconciliation could all be shared between the right participants,” says Hunt. 

This was the vision for the GSTPA and the larger market participants that contributed to its founding. In late 1999, a consortium was appointed that included a host of buy and sell-side firms and a product was devised, a virtual matching utility to be named Transaction Flow Manager (TFM). 

By December 1999 a number of vendors were selected to build and operate the TFM including IBM, SegaInterSettle (SIS), TKS-Teknosoft/Tata Consultancy Services plus two of the industry’s bank-owned cooperatives, Swift and the Depository Trust and Clearing Corporation (DTCC).

“I remember visiting the Swift office in 2001 and seeing a demonstration of the first GSTPA system as it was being beta tested by a few banks,” says Hunt, who was by then based in the Japan office of UBS Asset Management, responsible for technology and transformation as UBS globalised its operating model. 

“It was then that I realised that such a system could deliver considerable automation,” says Hunt. “But it would also undermine the business models of many existing providers such as Thomson’s Oasys Global system and even Swift itself.”

Problems, problems
Sure enough, the GSTPA’s ambition quickly hit some major snags as it struggled to put together a convincing business case for asset managers and suffered from a series of conflicts among its own backers. A major setback was the withdrawal of the DTCC from GSTPA discussions in 2000 before announcing its plans to form a joint venture with Thomson, which was to become Omgeo. The issue was further complicated by the fact that some of the DTCC’s board members, such as Merrill Lynch, State Street, Bank of New York and so on, were also backers of the GSTPA. 

And then there were problems with the vendors. SIS was replaced by SunGard Data Systems, which was also developing its own STP product to compete with both the TFM and Omgeo’s Central Trade Manager. Then the much-hyped launch of the TFM was delayed. When it eventually did go live in September 2002, Omgeo had already been up and running for some months and had clients on the system.

The GSTPA was dissolved on the last day of 2002 when its shareholders voted to suspend operations and dissolve the business, effectively refusing to put any more money into the project after already spending $100 million. 

So, what went wrong? In essence, while there was no problem with the software, there was not a strong enough business case for asset managers. Some of this was down to external factors. The industry’s decision to scrap the move to T+1 settlement removed some of the momentum around automation and straight-through processing, as did the economic recession that followed the 2001 terrorist attack in New York.

And, as Hunt points out, many asset managers were enjoying soft dollar arrangements with their brokers who would pay for parts of their infrastructure so did not want additional costs. “GSTPA crossed between front and back-office processes, which will have made it likely that it would have increased service costs, albeit in return for better standards, automatic settlement and reduced back-office costs.”

But then some of the problems were of the GSTPA’s own making. Some felt that it was set up to deliver a piece of software rather than run a business, in contrast to Omgeo, which had more than 50 clients signed up at the time of the GSTPA’s shuttering and it is still operating today, 20 years later. 

In 2001, Kevin Milne was executive managing director at Thomson and had helped to develop Oasys Global before going on to work on Omgeo. “What I have learned from the experience is that you have to find the end-user, the client,” he says. 

“Everybody has to benefit from the product and everybody has to be treated fairly. You have to keep away from the politics and the protectionism and focus on the product. You cannot be too one-sided. It is not a case of build it and they will come. You have to sell it and market it.”

Business case
“In my view, there was not a business case for the GSTPA that made enough people want to invest and adhere to the standard,” says Pete Cherecwich, president of corporate and institutional services at Northern Trust. Back in 2001, he was working for State Street as a business analyst, where his role was suggesting how the GSTPA might actually work. 

“In our industry, there are several different players – the sell-side, the buy-side and the custodians. The problem was that there were not enough benefits for all the players,” he says. “Why would you invest if there is nothing in it for you? Even if it was free to license and there was no transaction fee, there would still be the cost of system changes. You would only invest if it minimised your risk and there was no other way to get your trades done,” says Cherecwich. 

He points to the fact that it has taken Omgeo close to 20 years to get the majority of asset managers onboard as an indication of the difficulty that the GSTPA faced back in 2001 asking asset managers to invest in its offering. “I don’t know if anything could have been done differently back then. The fundamental thing was standards and to all use the same data. But what is the way to get the industry to embrace standards?”

The issue of standards has never really gone away, but it has been given fresh impetus by the emergence of DLT and the blockchain. There are a number of initiatives under way in terms of developing DLT standards. The likes of DTCC are advocating a central distributed ledger, while the R3 consortium is promoting the idea of multiple industry-driven utilities to appeal to the different industry participants and asset classes. Meanwhile, exchanges such as the ASX in Australia are looking to implement a blockchain for its central depositary, which Cherecwich refers to as a standard by the backdoor. 

And then there are a host of new entrants such as tZero and Overstock, looking to subvert current operating models by providing a point-to-point service which includes same-day distribution, clearing and settlement. Plus there are different blockchain protocols such as Hyperledger and Ethereum. 

Given the multiplicity of initiatives and the fact that once again, big banks are involved in almost all of them, Cherecwich believes it is unlikely that the “big utility in the sky” concept of the GSTPA will work. He hopes that the lessons from the GSTPA’s demise will have been learned because, as fee pressure continues and the disruption from the pandemic add to the pain, something has to change. 

 “As an industry, our fees continue to be pushed down, we need to be more efficient across the board. The only way we can go forward and address falling margins is to agree on standards. So, I think people are more aligned to this. And for asset managers, it is more of an issue than it ever was,” he adds. 

Lessons learned?
Hunt is optimistic that lessons can be learned, not least because the market is very different now. He refers to the healthy level of co-opertition (the portmanteau of cooperation and competition) in the blockchain and crypto space, which is typically open source and can be scaled via cloud technologies. Meanwhile, the soft dollar arrangements of the past have been largely removed by MiFID II, so asset managers are making fully independent decisions about their technology and service providers. 

“This means there is a great opportunity for asset managers to embrace these new technologies to tackle some of the ongoing challenges in the value chain of trading, matching, settlement and regulatory transaction reporting. What is fascinating to contemplate is that crypto assets provide complete transparency of investments end to end. Perhaps this is the endgame – GSTPA 2.0?” Hunt muses. 

Milne too has a growing optimism that blockchain and DLT will not suffer the same fate as the GSTPA, but legacy issues remain a problem. “If you were starting from scratch today, you would put most post-trade processes on the blockchain. The reason it hasn’t taken off yet is because of the legacy systems that are still around and vested interests cannot afford and do not want to be turned off. This was a problem back in the 1980s when central settlement engines like Taurus would have put people out of a job,” he says. 

“There is a similar problem with blockchain. You may not need registrars or even CSDs. Emerging markets can go straight to that approach and then people in the developed markets will say, ‘If they can do it, why can’t we?’”

The forced changes to operating models because of the pandemic may act as a catalyst, says Milne. “When things return to a new normal, banks will have some hard decisions to make. The old way of doing many things will have to end. There is still a lot of legacy but in the interest of market efficiency, it has to happen.”

Milne also thinks the buy-side is much more powerful, larger and influential than it was. “Back then, the brokers would tell the asset managers not to worry their pretty heads about any trading or post-trade issues. But now they are much more involved in the process and have much more control and self-reliance. They are also mindful of the aftermath of the 2008 financial crisis, everything from shorting to a lack of independent research. They are now more self-reliant.”

Yet just as the shocking events of 9/11 left a long recession in its wake, followed by the global financial crisis in 2008, the uncertainty of the pandemic and the likelihood of another long recession could yet lead to a retreat from bold ideas and radical changes.

Even if the will is there, some still feel it is an impossible task. As an ex-GSTPA executive told me: “Looking at such a true global financial transaction concept today, for which you would ideally need the blockchain, I have my doubts that any player worldwide could be successful in finding capable partners, either in the US, UK, Europe or Asia.”

© 2020 funds europe