The message came from the top in September. Craig Donohue, chief executive of CME Group, signalled the exchange was going to make another attempt to break into the promised land – read promised revenues – of the $2tr a day over-the-counter foreign exchange market.
The keys to the kingdom, he said, were to be found in ClearPort, the clearing platform CME had acquired with last year’s takeover of the New York Mercantile Exchange. Central counterparty (CCP) clearing would be the way in to the FX business.
The initial market reaction to Donohue’s statement was quizzical.
Hadn’t CME trumpeted its earlier move into foreign exchange in a joint venture with Thomson Reuters called FXMarketSpace? And why had that venture – and talk of a golden age of central clearing for FX trades – only lasted two years?
Launched in May 2006, FXMarketSpace was, as far as outsiders could gauge, a costly flop. It never reached its target level of a 2% market share. Moreover, it closed in the heat of the financial chaos last October – perversely, at the very moment when counterparty risk became a headline theme and when the central counterparty clearing model was proving its worth.
So why is CME back so soon with a new approach to winning foreign exchange business? And how does it intend to make this happen?
There are three reasons for its return.
The first, the long term strategy, is the natural fit between FX and what the group does anyway. “We’ve got a diverse set of global benchmark programmes that cut across not just foreign exchange and rates but equally energy, commodities, equities and other product categories as well,” said Donohue in September.
“But foreign exchange helps us further globalise our business because it is such a global market and there’s demand for foreign exchange products and services worldwide.”
The second reason is that CME believes it has learnt a key lesson from its experience with FXMarketSpace.
Expecting any large organisation to admit making an error may be a vain hope. But Donohue may not just have been sweeping things under the carpet when he said FXMarketSpace was “too revolutionary” for its participants.
Before the collapses and rescues of Bear Stearns, Lehman Brothers, AIG, Merrill Lynch and many more, creditworthiness was a largely neglected issue in the foreign exchange markets. Only a few lone voices talked about it and they were mostly ignored. In the post-crisis markets of 2009 credit risk is top of the agenda.
Derek Sammann, CME’s managing director of FX products, says: “FXMarketSpace offered an integrated OTC FX execution and clearing service on a single platform. But what we discovered was that our customers didn’t want both clearing and execution on the same platform, but rather just clearing.
“While they were concerned about the post-execution credit risk of their counterparties, they also wanted the choice and flexibility of the product and execution venue of their own choice.”
Fear stalks markets
However – and this is CME’s third reason for dusting itself down and stepping back in the FX ring – the exchange believes now is the right time. Most importantly, credit fears are on the minds of regulators as well as the markets.
“There is now a clear government mandate for clearing CDS, and although we’ve not yet reached that point in the foreign exchange market, the case for the credit mitigation facility of a clearing house for the global FX market is becoming compelling,” says Sammann.
Indeed, CCP clearing continues to dominate the talk of regulators in the US and the European Union. Charlie McCreevy, EU internal markets and services commissioner, summed up the mood recently: “There is a consensus that we should extend the use of central counterparty clearing beyond the CDS market – the only question is how to.”
But it’s more than just regulators that are talking about managing credit risk – so too are players in the FX markets.
“In the last 12 months there has been an increased reluctance to take on counterparty risk,” says Sammann. “And we don’t see this going away.”
The head of FX at CME Group would say that, wouldn’t he? But there is some evidence to back what he says.
Transaction volumes across the global FX market are sliding – at least partly because of these credit fears. According to the latest survey released by the New York Federal Reserve in July, spot FX trading is down 25% this year, while volumes in the more risk-sensitive side of the business such as non-deliverable forwards, forwards and swaps have fallen further. FX options volumes are down 47%.
Activity on EBS, the largest OTC spot FX trading platform, has shrunk by 53% since September 2008.
Yet exchange-traded products have bucked this trend. CME Group’s September FX volumes, for example, are just 6% below those of September 2008.
“Firms still have the appetite to take on FX risk, but reluctance to take on some of the counterparty aspects means they are actively seeking the safety of the FX futures market,” Sammann argues.
Pleasing the customer
The question for CME as it prepares its new sally into forex is whether it can deliver something that the FX market wants – even if sometimes the more aggressive dealers refuse to acknowledge the need.
Perhaps the biggest problem CME has had historically with its foreign exchange business is the standardisation of its contracts. This September’s report by the Bank of England’s joint steering committee on FX says: “Traditional-style exchange trading, on recognised investment exchanges such as the Chicago Mercantile Exchange in the US, has existed for many years for FX futures and options.
“However, it has never gained more than a small percentage of overall FX market share, in large part because it necessitates focusing liquidity around a relatively small number of highly standardised contracts.”
The Bank believes that while this may be suitable for participants that simply need general financial exposure, these standard futures and options may not always meet the individual requirements of FX market users, which may need a specific amount of currency in a specific place on a specific date.
The CME has a one word reply to this challenge: ClearPort.
This clearing platform was developed to handle OTC derivatives in the wake of the Enron bankruptcy in late 2001, which gave the energy markets the kind of wake-up call about credit risk that the wider financial markets got from Lehman Brothers.
ClearPort now clears half a million highly specialised OTC oil, gas and power contracts a day, and that range will be extended across all the major asset classes CME deals with.
“ClearPort can now clear any product, any date, any notional,” says Sammann. “ClearPort means the CME Group is not just about clearing standard products on specific settlement dates, but we are fully flexible.”
The CME’s goal is to be internally ready to launch FX clearing on ClearPort by the end of the year. External readiness in terms of clients’ technology and practices is scheduled for early 2010.
One contentious issue will be pricing. Sammann says: “We haven’t agreed on what the final pricing will be, but we can be indicative. Our regular exchange fees take into account execution and settlement, whereas here we are only talking settlement, so they will be lower than our regular ones.”
Can CME choose the right prices? This is “the gazillion dollar question”, according to Jonathan Butterfield, director of communication at CLS Group, the bank that settles what is estimated to be more than half of the FX market.
At issue are two different models of clearing and settlement.
At present, payment for most classes of forex transaction on CLS is made on the PVP basis – payment versus payment. This eliminates credit risk: both buyer and seller receive their funds simultaneously when the transaction is completed.
The process is called continuously linked settlement and the present arrangement at CLS Bank went live in 2002. The initiative to set up the bank came in 1997 from a group of the biggest FX trading banks, which wanted to rid themselves of the risk of a default – commonly called Herstatt risk (see box).
Another of the prime advantages of CLS is that settlement is achieved through straight-through processing: the entire transaction flows automatically from the moment it hits the dealer’s blotter until payment and reconciliation.
En route, CLS nets the transfers between counterparties with an efficiency of around 95% – ie, for every $1tr of transactions passing through the system, only about $50bn of cash actually has to move.
To put this in perspective, in March 2008 CLS hit a record, settling $10tr of payments in a single day. It has 6,043 firms participating and handles 17 currencies.
With PVP settlement, Herstatt risk is removed – that is, you cannot be put in the situation of paying out money for a spot transaction and not receiving the acquired currency in return.
However, an agreed trade could still fail to reach completion, and that carries other risk. For example, a market participant with a hedging strategy in place may lose one leg of its trade.
The risk is higher with futures transactions, where there is time for the counterparty to go bust and hence fail to fulfil its obligation.
In the central counterparty clearing model, from the moment a trade is posted, all forms of counterparty risk are guaranteed by the CCP, the clearing house itself.
To make this possible, the counterparties have to post margin in proportion to a transaction’s risk. The risk is marked to market twice a day and margins are adjusted accordingly.
If a transaction is not completed – such as during the Lehman Brothers collapse – the clearing house will try to transfer positions to other clearing members or auction off the uncompleted trades, using the margin and its member deposits – in the CME’s case, $7bn – to fulfil obligations and contain systemic risk.
Sammann believes market participants misunderstand what CLS and CME do – how they differ and how they interact.
“As dialogue has ratcheted up regarding OTC clearing, people have had to undergo a basic education in the differences between settlement risk and counterparty risk, and the respective roles of CLS and a clearing house like CME,” he says. “CLS does not mitigate counterparty risk, but it does mitigate settlement risk. Equally, CME doesn’t mitigate settlement risk in FX transactions, but we do mitigate counterparty risk.”
Even CME’s clearing house will use CLS to finally settle transactions. “We are very much dependent on one another to create a front-to-back risk mitigation system,” Sammann says.
Warming to this theme, Sammann describes three primary actors on the FX stage – those doing the execution, such as banks and execution platforms; those doing the clearing, such as CME; and those handling settlement risk, such as CLS.
“All three of these elements have to work together to build a systemically sound global FX market,” he says.
All about price
Sammann may be right that some people misunderstand the role of central clearing. But it is nevertheless clear that many see getting the right price as the key component for success.
“For clients it’s a question of being able to work out how much the cost of a trade will become. In the last few years transparency has driven the bid-offer spread down to four decimal places and the transaction costs have had to move down in line with that,” says Butterfield. “The evidence of the last 10 years is that the market has proven to be very price-sensitive. The last CME initiative with Reuters in FXMarketSpace, an FX exchange with clearing, didn’t gain enough traction – we’re told the market voted with its feet, in part because of the price.”
Butterfield is sceptical that CME can do better this time. “Anyone attempting to come into the clearing side de novo in this market is going to have to commit to significant up-front investment cost, as well as the routine operational costs that come with clearing,” he says. “I find it hard to see how their prices could be competitive if they are not already in this business in some fashion – especially since ours are steadily going down.”
There are also going to be difficulties seeing how far CME can reach into the FX market’s more complex depths – another of the Chicago exchange’s stated aims.
Butterfield says that although one can distinguish between what might be called a standardised derivative and those that aren’t, a lot of FX options will not be of a kind that can be fitted into a simple clearing model.
“The range of FX options – the nips, tucks, collars and the like – don’t have standardised confirmation points, which means that settlement would be non-standard, and therefore not eligible for clearing,” he says. “It will be interesting to see how broad a range of FX options could be submitted for clearing.”