Continuous Linked Settlement promises to alleviate the risks associated with large-value FX transactions
One of the worst scenarios for the banking industry would be if a financial institution entered bankruptcy before settlement of a large-value transaction. The failure of this firm could topple others like a row of dominoes, as counterparties scramble to meet obligations to counterparties with obligations of their own, and so on.
Settlement risk begets liquidity risk, or the risk that a wide swath of banks, leery of distressed counterparties, will refrain from trading. The result: systemic market paralysis.
Regulators and banks responsible for keeping the global financial system healthy are hoping that Continuous Linked Settlement Services, or CLSS, will help protect FX trading from these dangers. CLSS essentially ensures that both counterparties have paid in their respective legs of an FX transaction before settlement can occur (see sidebar, page 13). That way, if one side goes belly up, the other can walk away with its capital intact. The idea is to prevent financial distress from turning into a financial contagion.
To implement CLSS, over 65 of the world's largest banks have banded together to create a jointly-owned company called Continuous Linked Settlement Services, Ltd., whose New York-chartered subsidiary CLS Bank is preparing to act as the world's FX settlement intermediary and liquidity provider. The immense FX market has an estimated daily transaction volume of approximately $1.5 trillion. CLS Bank will settle almost half of that growing market by 2005, according to Celent Communications.
CLS Bank also promises to reshape the FX trading infrastructure, affecting the systems of its shareholder banks as well as those of the real-time gross settlement (RTGS) systems operated by the largest central banks plus countless smaller banks and corporate entities involved in international trade.
Yet despite its pedigree and theoretical advantages, the CLSS venture has encountered delays, technical problems and nagging questions about its ultimate viability.
Emblematic of these difficulties has been the oft-delayed launch of CLS Bank. Its debut was originally scheduled for mid-2000, then was pushed back to October 2001, and now after repeated delays has been pushed back to an unspecified date. IBM-CLS Bank's implementation partner-requested further testing time, said Mani Venmans, a spokeswoman for CLS Bank. "It's projected that we'll now go live in 2002."
Although recent market conditions show the need for protection against counterparty default, there's no undue rush to bring CLS Bank online. "We can all understand the complexities of the system and why everything has to be absolutely foolproof before we go live," said Venmans.
Industry observers share this caution. "As soon as you link together the foreign settlements of the globe, the slightest creak in the system is unacceptable," said Paul Styles, market development manager at IntraNet, a Newton, Mass.-based payments and messaging software company. "This thing has to work perfectly-otherwise it's a disaster."
Indeed, the biggest problems for CLSS are operational, not conceptual, in nature. "It's a simple concept, but it's hellishly difficult to implement when you begin to link, in a real-time way, the number of players, the number of transactions and the different currencies," said Styles.
Settlement, for example, which has traditionally occurred on a given "value date," will now occur at a specific time on that date, which could create a cascading need for changes in banks' existing systems.
Fortunately, there are a wide range of systems to handle trade tracking, schedule management, liquidity management and settlement reconciliation, among other components. Major vendors include Fundtech, IntraNet, Logica, mpct Solutions, Sungard, TSI International Software and IBM.
But this new technology will not come cheap. Banks will need to spend approximately $183 million to upgrade their IT infrastructures for CLSS, according to TowerGroup. Banks will also need to update treasury management systems and a wide range of other applications. Then there's the investment in CLS Bank itself, an estimated $3.5 million for charter members plus the cost of an upgraded connection to SWIFTNet.
Still, the long-term benefits of CLSS inclusion should outweigh any initial start-up costs, experts say. For example, CLS Bank is an early adopter of SWIFTNet services, which are based on IP, or Internet Protocol, instead of the X.25 store-and-forward mechanism used in the prior generation of financial networking. SWIFT (Society for Worldwide Interbank Financial Telecommunication) will provide SWIFTNet connections to all of the CLS Bank members, and continue by upgrading over 7,000 financial institutions from X.25 to SWIFTNet by mid-2002. "We will be able to offer not only financial messages but also VPN Virtual Private Network, Internet access and so on," said Serge Hauspy, SWIFT's CLS business manager in Belgium.
Global Crossing, a global network services company headquartered in Bermuda, is responsible for connecting the various RTGS systems to CLS Bank via the SWIFT network. "As Global Crossing takes over management of the Swift IP network, that should take over a lot of the local ingress and egress issues that CLS Bank may have had with some of the local RTGS systems," said Fritz McCormick, an analyst at Celent Communications.
SWIFT will also forge closer relationships with the largest banks. "Top-tier banks with worldwide forex operations are, to some degree, moving through individual foreign RTGS systems," said McCormick. "In some of the major money centers-London and New York - they will be hooked up directly to SWIFT."
The prospect of owning shares in a profitable CLS Bank is being used to lure banks to participate. But an even stronger incentive is pressure from regulators to mitigate settlement risk.
If not for CLSS, regulators would likely require that banks hold risk-weighted capital against their FX exposures, cutting into banks' already thin profit margins on FX trading. "Banks struggle with the business case," said IntraNet's Styles. "A lot of the larger banks are in it principally because they have to be."
By heading off regulators with a bank-owned venture for mitigating settlement risk, CLS Bank participants hope to attract transaction flow from third-party banks and corporations. "Clearly, where they can make money out of it is to bring on board the so-called third-parties, and that is where you are handling the FX transactions for those institutions that don't want to become the settlement vendors," said Styles. "Those banks that have a large correspondent banking business are the ones that are placed to gain market share and turn this whole thing into a money-generating business."
That's likely to set off a scramble among big banks for third-party business. "Third-party banks will look for service and they will look for support-someone that has staying power," said Irv Cohen, senior vice president of financial markets solutions at J.P. Morgan Treasury Services. "And then they should be looking at price" he added. "We're all in a different world, post-September 11th, in looking at service as well as price."
Since September's terrorist attacks, diversification of banks' real estate holdings has become as critical as diversification on the balance sheet. J.P. Morgan Chase recently accelerated its two-year plan to diversify its geographic footprint-moving its security operation to Texas and establishing two redundant cash-clearing facilities in separate Florida locations.
Exacerbating the revenue problem, third-party traders may not even send all of their business to a single CLS settlement bank. Smaller FX players are expected to "game the system" in order to prosper in the new environment. "Those institutions that handle the vast majority of FX trading will become even more powerful and even more controlling in the market," said Celent's McCormick. "So the secondary and tertiary dealers are going to step away and may not embrace CLS as thoroughly as predicted, because they see this as more of a power grab among the big banks."
To counter market concentration among their suppliers of settlement services, third parties may establish multiple CLS connections through different counterparties. That way, no single entity would be able to get too much control over trading costs or gain an informational advantage by having a total view of a trader's transaction flow.
LETTING THE CHIPS FALL
The ultimate success or failure of CLS Bank has far-reaching implications throughout the global economy. It's not just about settlement risk.
Payments networks have been among the first to adjust to CLS Bank, a potential 800-pound gorilla. "We recognized, when CLS first started forming back in '95, that the industry had identified a need for a vehicle like CLS to help manage FX settlement risk," said John Mohr, chief operating officer at CHIPS, the New York Clearing House's net settlement subsidiary. "We began to reposition ourselves so that we could have a system that was good and viable going forward."
As a result, CHIPS began to focus on commercial payments, the bulk of which consists of domestic activity. This year CHIPS introduced payment finality, putting the CHIPS network on a par with the Federal Reserve's Fedwire. However, Fedwire is one of the seven real-time gross settlement systems currently connected to CLS Bank; CHIPS is not.
Yet CHIPS, originally chartered to process international payments, may still find its place at the FX table. "We've talked to a number of banks that have encouraged us to look at CLS and think about it," said Mohr.
Even those central banks with RTGS already in line for CLSS have reservations. While CLSS theoretically eliminates settlement risk, it might also concentrate risk of a different kind, according to one school of thought.
"You could argue that in today's world the domestic payment settlement systems are, to a degree, hermetically sealed. If there's a liquidity problem in, for example, the sterling market, that can be largely contained within the UK and within that sterling market," said IntraNet's Styles. "As soon as you link sterling with the euro, the dollar, the yen, the Canadian dollar and all of the other currencies, you're in danger of having that knock-on effect if there's a liquidity problem in the sterling markets."
That danger increases if lowered settlement risk attracts other central banks to CLS Bank. "There are other currencies queueing up to join. There's talk about bringing on the Hong Kong dollar and the Singapore dollar," said Styles. "If you've got the major 10 currencies of the world in there, there's not going to be a lot left outside it."
Besides FX, other asset classes are tightly linked to the fate of CLS Bank.
First, continuous linked settlement in FX is practically a prerequisite for straight-through processing in the securities industry. "Without some type of compressed settlement system in the FX market, it's going to be fairly difficult to move to a real-time or T+1 world in securities," said Celent's McCormick. "You need to have, simultaneously, a compressed settlement of FX transactions and securities transactions, especially when you start thinking about cross-border trading."
Second, although the initial scope of CLS Bank includes settlement of spot transactions, swaps with options and some other exotic FX transactions, there's no reason a fully-functioning CLS Bank couldn't eventually settle trades in other asset classes. "That's a long time away," said Styles. "But there's no philosophical obstacle to using this for other asset classes."
What's more, at least a few U.S. government entities will likely welcome the consolidation of global settlement activities through a New York-based organization.
As a U.S.-chartered bank, CLS Bank will be subject to anti-money laundering regulations that require banks to proactively prevent named organizations and individuals from accessing the financial system, as per the U.S. Treasury's Office of Foreign Assets Control. "There's this implication that the OFAC regulations go global," said Styles. "Not just for the dollar but for all trades that are passed through the CLS Bank in the U.S."
That means trouble for terrorists, drug traffickers, tax evaders and other criminals around the world, and some important questions for everyone else.
"It ain't settled 'til it's settled"-and safe settlement is exactly what the designers of Continuous Linked Settlement (CLS) are after.
Continuous settlement means that there's a window of time, rather than a specific moment in time, during which counterparties to a trade can lay their respective funds on the table. For CLS Bank, a "settlement day" happens during the overnight shift (1 a.m.-6 a.m. EST) in the eastern U.S., the morning hours (7 a.m.-12 noon CET) in Europe and the late afternoon (3 p.m.-7 p.m.) in Japan.
Before the start of a settlement day, each member bank will inform CLS Bank of the agreed-upon trades of its customers and from its own operations. In return, CLS Bank will calculate and issue pay-in schedules for each of the member banks.
Throughout a settlement day, CLS Bank will stand ready to receive incoming payments from the banks via a national RTGS system. Once two "legs" of a transaction have arrived, that transaction can be safely settled on the books of CLS Bank. Then, based on the balances in each member bank's account and its own liquidity position, CLS Bank can pay out available funds, again through the central banks' RTGS systems. This cycle will continue throughout a settlement day.
As each of the central banks reach the close of business in their time zones, CLS Bank will make sure that settlements involving those currencies are complete. CLS member banks have to be ready to provide liquidity at this point, since a member bank might have a short position in, say, Japanese yen, while having a long position in the U.S. dollar.
At the end of the day, CLS Bank will have zeroed out its holdings through the various RTGS systems and all members will have their rightful funds. - Ivan Schneider