OTC market faces up to CCP-dependent pricing

Where multiple clearing houses cover the same products, dealers expect pricing differentials to emerge. That is causing friction with buy-side firms, and headaches for trading platforms. Joe Rennison reports

Scales with fruit

Early adopters of central clearing are encountering a problem they hoped had been left behind – when asking dealers to quote a price for the same product, they are getting different answers, or finding some are unwilling to trade at all. In the bilateral world, these differences in price and liquidity are largely driven by counterparty credit quality and the details of each client’s collateral agreement – but where dealers are facing a central counterparty (CCP) on each trade, it should remove both sources of pricing divergence.

In fact, it does – but they have been replaced by others. Currently, a bank is able to net together each client’s portfolio of trades across products and asset classes to work out its counterparty exposure. If a new trade reduces risk, then dealers can offer a better price. But in a cleared market, trades are sent to different CCPs – and, for the most part, clearing houses charge margin on each product type independently. In short, dealers are less able to achieve netting efficiencies – and while counterparty risk is greatly reduced, funding costs may increase.

Pimco is believed to be the first company to have run into these issues. Dealers refuse to confirm that on the record, but the head of clearing at one European bank describes friction between banks and a firm that started clearing large volumes of interest rate swaps at CME Group in August 2011 – up to this point, dealers had been clearing almost exclusively at SwapClear, the interest rate swap CCP run by LCH.Clearnet.

“This buy-side counterparty generated a lot of activity, and it got upset because a pricing differential emerged between SwapClear and CME,” he says.

The description fits Pimco – the firm adopted clearing early and is widely believed to be the source of the first leap in cleared volumes for both interest rate and credit default swaps (CDSs) at CME, starting in August 2011. That month saw the CCP clear notional volumes of $1.2 billion in interest rate swaps – twice the level of its previous record month – and volumes surged even higher in September. Pimco declined to comment for this article.

There is a strong buy-side preference for a single, consistent swap price to be quoted

The issue for its dealers was that they had little or no existing cleared business at CME, and Pimco’s directional positions required large amounts of initial margin. That created an incentive for dealers to adjust the price for any offsetting trades that would be cleared at CME. Because no other clients were clearing in volume, the differential appeared in the interdealer market, with brokers facilitating trades designed to flatten the exposure for Pimco’s dealers.

That apparently annoyed the California-based fixed-income giant. “After about a week, it strong-armed the dealer community to the extent that if someone was going to offer a better price, then they wanted to see it, too, and decide whether to take it,” says the European bank’s clearing head. “There is a strong buy-side preference for a single, consistent swap price to be quoted.”

Something similar happened in the market for cleared CDSs on the CDX index, says a senior trader at one European bank. “We had some big guys come in and we sold a very large chunk of index protection and our client wanted to clear it, so we did – but because we had no other clearing going on, we became a meaningful net seller to the CCP. When the client came back later and wanted to do more, we said we were only going to be buyers of protection on the CDX because we wanted to net that position out. It was pretty big,” he says.

Dealers say clients are unwilling to pay more for a trade just because of portfolio effects resulting from a bank’s activity with other customers. In addition, some clients worry they could end up losing out in other ways – fears that stem from their experience with high-frequency trading in equity and foreign exchange markets. “Clients feel like whenever there is a fractured market there will be some gamesmanship going on, in which dealers play with pricing and build algorithms that will dance between different CCPs, so I think there is a desire on the part of the buy side to see one swap price,” says the European bank’s clearing head.

Some of these issues are primarily the result of the cleared market’s current illiquidity, which ought to change this year. The Dodd-Frank Act clearing mandate is set to take effect for all US market participants during 2013 – starting with dealers and hedge funds in March, before two further deadlines extend it to include other entities – and volumes should increase exponentially as a result. In theory, that will make it easier for market-makers to run flat books at each CCP, but at any point in time a dealer may find it has an imbalance at one venue or another – and some traders expect those imbalances to be persistent, arguing CME Group will tend to attract client business, while SwapClear is favoured for interdealer trades, for example.

Other issues are still being debated. There is no consensus on whether banks will always pass along clearing-related funding costs, on the size of the resulting price differential – or on how regularly a dealer will adjust its prices to reflect portfolios that will be changing throughout the day. In addition, trading platforms are working out how to cope with a world in which pricing depends on the venue at which it is cleared.

Dealers highlight three CCP-specific factors that could affect the price of a trade: the initial margin methodology used by each clearing house, potential concentration risk charges, and – most importantly – the bank’s existing cleared portfolio.

The first issue is played down by most market participants. “The differences in margin methodology are small enough that essentially there isn’t any price differential arising purely because of that at the moment,” says Simon Wilson, deputy head of delta trading for Europe, the Middle East and Africa and global head of algorithmic rates trading at Royal Bank of Scotland (RBS) in London.

The potential for a concentration charge is described by one clearing head at a European bank as a “real stinger” as it could potentially double the amount of margin required in some cases – but the impact would only be felt once cleared risk breached a certain threshold. It should not affect trade pricing up to that point.

Instead, the key issue will be the incremental margin required to clear a trade at the available CCPs, says Bob Burke, head of global OTC clearing at Bank of America Merrill Lynch (BAML) in New York. He sketches out an example in which a bank is a net payer of fixed at one CCP, and a net payer of floating at a second venue – both portfolios have the same sensitivity to interest rates. If a client wanted to clear a big swap in which the bank would be paying fixed, then it would make sense to offer a better price at the second CCP, where it would offset the dealer’s existing portfolio, he says. “Swap pricing differentials may exist depending on which CCP a client chooses, since CCP initial margin requirements increase with risk concentrations,” says Burke.

How big could this effect be? “You could see pricing differentials around an eighth, a quarter, three-eighths, or maybe a half basis point. Not multiple basis points though, I don’t think,” says Burke.

But working out a precise add-on for each trade is not easy. In theory, it would require a dealer to recalculate its prospective margin requirement at the various CCPs available to a client for every new quote it offers. That calculation would also need to make a call on how long to price in the funding cost of the trade. Pricing to maturity would be expensive and unrealistic if the dealer expects the portfolio to return to equilibrium after a period of time. But how long is realistic – a day, a week, a month?

Dealers say this is likely to be a job for desks that calculate credit valuation adjustment (CVA). Some envisage constant communication between the CVA desk and traders, with the pricing for various CCPs changing throughout the day. Others argue this would be disruptive.

“I don’t envisage it will be managed on a trade-by-trade basis,” says Paul Hamill, head of matched principal trading at UBS. “It will be done on a day-by-day, intra-day or week-by-week basis and you will look at your positions and decide what to do at these intervals.”

This creates a new challenge for trading platforms that will need to display multiple prices for one product on central limit order books. By far the easiest way to do this is to have split screens and split liquidity pools – so a user will see one screen for prices on a 10-year euro-denominated interest rate swap at SwapClear and another screen for prices on the same product at CME. Dealers would provide prices for each cleared instrument. But critics see this as cumbersome and argue it is unrealistic to expect dealers to make prices in a number of different products on a number of different screens. The fear is that dealers would decide not to provide liquidity to all CCPs – or would quote on fewer platforms – reducing the depth of the market.

“You could have four or five different clearing houses for a given product and if you ask market-makers to stream multiple instruments then that gets resource-intensive,” says a chief operating officer of an interdealer broker. “There is a limit on how many order books someone can stream to. It’s a 1990s solution to a twenty-first century problem – why bother doing that?”

Instead, a platform may try to retain a single order book that includes prices for different CCPs. Rather than manually inputting quotes for the same instrument on multiple screens, dealers will be able to quote CCP-dependent prices for the same product in one place.

Clients, meanwhile, would only see executable prices for the clearing houses they are able to use, so a firm with a clearing relationship at Eurex and CME will not see – or, at least, will not be able to execute against – prices for the same product at SwapClear. 

Some buy-side firms accept this may be a sensible outcome – in part because the alternative is worse. If a dealer is not able to provide CCP-specific prices, then it will quote based on the worst-case margin scenario, driving up the cost for clients. But if pricing differentials exist, then it will further complicate the cleared derivatives market for buy-side firms, which will need to look at both the price they are being offered and the incremental margin demand at the clearing houses to which they have access.

“From a client perspective, when you execute, you will no longer look only at the bid-offer, you will need to look at the total cost of the trade – meaning the price we receive, fees and the margin we post – and that will be portfolio driven. We may be getting a tighter price at one clearing house on screen, but because more of the portfolio is cleared at another venue, that price may not be the best one when everything is taken into account,” says Supurna VedBrat, co-head of electronic trading and market structure at BlackRock in New York.

But CCP-dependent pricing may not be inevitable if the cleared OTC market is liquid enough – and efficient enough – for dealers to easily flatten their cleared portfolios. This is a process that Trad-X – the interest rate swap platform belonging to broking firm Tradition – is trying to facilitate by enabling dealers to trade the basis that emerges between two clearing houses.

The basic idea is simple enough – in a situation where a dealer is primarily paying floating at one CCP and fixed at a second, the bank might want to attract offsetting business at each venue. The result would be a reduction in the amount of initial margin required by both clearing houses. But rather than entering two trades into the central limit order book – in which it is 0.2 basis points cheaper for a counterparty to pay floating at the first CCP than at the second, for example – Trad-X plans to offer a screen on which users can execute both trades simultaneously, quoted as the spread differential between the two venues.

“We are looking to give the market the ability to easily move positions between CCPs so it’s easy to manage margin demand. The ability to trade one CCP versus another CCP will be vital for banks as price differentials begin to occur,” says Stuart Giles, head of business analysis at Tradition in London.

For this to work, though, dealers have to be able to execute the desired trades at each CCP – and that may not be possible. Some market participants say that while the vast majority of dealer and client interest rate swaps currently pass to SwapClear, buy-side firms are gearing up to put the majority of their flow through CME. If this bias – or a similar one – occurs, then it will split netting sets and also make it more difficult for dealers to easily flatten their risk.

“Customers that clear on CME, for example, might tend to have a bias to receive fixed – so the banks are always paying fixed into CME and there would be no way to get out of that trade. You have no choice but to price that into the cost of the trade,” says RBS’s Wilson.

If the bias does materialise, then it is likely to be permanent, meaning any price differential is likely to be more or less fixed, he says. But if this bias does not materialise, then there will not be any price differential because each venue will contain a mix of trades and dealers will be able to maintain a flat book.

“If there is only a small systemic bias, then it is likely CME and SwapClear trade with the same price, you can get out of trades easily and no one cares,” he says. “If there is a bias, then the CVA desk can work out what that is. If everyone on CME wants to receive fixed, then there is a big bias – but you still don’t need to work out your intra-day position because it will always be the same. Do I think it is likely that this bias will change intra-day? No not really. I put the highest probability on there being a basis market and there being a systemic bias.”

Some see these effects as strong enough that, over time, only one CCP will survive with any significant liquidity – the desire to maximise netting benefits and retain one pool of liquidity will be too strong. “We just don’t know what will happen, but I think ultimately one CCP will have the vast amount of trading and the other will have legacy positions, a wider bid-offer spread and less liquidity,” says one global head of risk management at a large asset management firm.

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

Switching CCP – How and why?

As uncertainty surrounding Brexit continues and the impacts of Covid-19-driven market volatility are analysed, it is essential for banks and their end-users to understand their clearing options, and how they can achieve greater capital and cross…

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here