A battle is brewing in a historically quiet corner of the US fixed-income market. Since 1986, the clearing of US Treasuries and Treasury-backed repurchase agreements has been controlled by Wall Street, via the member-owned Depository Trust & Clearing Corporation. In 2003, clearing of these transactions switched to a DTCC subsidiary, Fixed Income Clearing Corporation, which gives its members billions of dollars in rebates and fee reductions.
Some believe greater efficiencies could be unlocked, however, via futures behemoth CME Group’s planned $5.4 billion acquisition of Nex, operator of the largest Treasuries trading platform, BrokerTec. Five senior fixed-income executives Risk.net spoke to for this article say the deal creates huge incentives for the Chicago giant to bring clearing of cash Treasuries in-house, despite its public plans to continue sending BrokerTec business to FICC.
That would bring cash clearing together with CME’s vast pool of futures open interest, and its swaps clearing service.
Add in the fact CME has filed separately with the Securities and Exchange Commission (SEC) to clear repos, and it isn’t hard to see why market participants believe the industry could be about to witness a titanic battle for the soul of the US interest rates market – one that could result in a power shift from New York to Chicago.
“I really do not think the FICC was awake to the threat to their model until CME bid for Nex. They’re at risk of losing not only the US Treasuries clearing franchise, but also repo,” says a former senior US Treasuries broker.
For its part, CME has been careful to avoid stepping on FICC’s toes, stating publicly the deal “maintains BrokerTec clearing at FICC, with a goal to increase capital efficiencies for customers”.
That has done little to temper talk of a push into Treasury clearing, however. Asked whether he believes BrokerTec’s new owners will bring clearing of cash Treasuries in-house, a senior executive at a rival trading platform is unequivocal. “That’s exactly what we’re expecting,” he says. “As an execution venue, it gives you a huge competitive advantage if you can clear those trades yourself, rather than having to send them elsewhere.”
I certainly don’t believe CME would be doing this deal to allow a subsidiary to continue to clear through what they see as a potential competitorChief executive at one FICC member
The chief executive at one FICC member agrees this is the likeliest outcome. “The repercussions of the CME looking to encroach on FICC’s franchise by doing the Nex deal are going to send tidal waves through the market. Initially, BrokerTec has no choice but to go through FICC as they can’t clear through CME today. But I certainly don’t believe CME would be doing this deal to allow a subsidiary to continue to clear through what they see as a potential competitor,” he says.
Putting cash Treasury and repo clearing together with CME’s other suits could be a powerful combination. The exchange giant boasted an open interest pool of just under $1.9 trillion in Treasury futures and options as of the end of the first quarter.
In over-the-counter rates clearing, it boasts $10 trillion in open interest – putting it far behind market leader LCH's $135 trillion, for now. But with the addition of Treasuries platform BrokerTec, which averaged $160.8 billion a day in cash turnover last year, powerful savings could be wrought through cross-margining between bonds, futures, repo and swaps – perhaps ultimately allowing the Merc to strike a blow in its battle for OTC clearing supremacy, too (see box: One pot versus two pot).
FICC and CME currently maintain a so-called two-pot cross-margin agreement, in which the central counterparties (CCPs) allow offsets between cash Treasuries and Treasury futures, while still holding and managing members’ positions and collateral separately. This offers a reported 0.25% reduction in initial margin for FICC members – representing a paltry $30 million in cumulative savings given a required initial margin pool of roughly $12 billion. Both CCPs acknowledge the arrangement could be tweaked to offer more efficiency.
A more radical solution that still left the current market infrastructure intact would be to create a one-pot margining system between the two CCPs. FICC entered into such an arrangement with New York Portfolio Clearing in 2011 – the ill-fated CCP set up by NYSE Euronext and DTCC to combine cash Treasury clearing with futures. The venture failed, but had it come to fruition, estimates of average initial margin reduction were placed at 20%. If the same offsets were achievable between CME and FICC, that would account for an approximately $2.4 billion reduction margin at FICC alone. A former NYPC executive says that figure was just an average and the CCP’s analysis calculated that some clearing members’ margin reduction might reach 40% under the arrangement.
One-pot margining systems have delivered more extreme offsets in other products: CME maintains a one-pot agreement with OCC for S&P 500 index futures and CBOE’s S&P index options. CME reported a 91.24% decrease in required initial margin from the arrangement in its most recent Principles for Financial Market Infrastructures (PFMI) disclosure.
The hedge against CME’s entire bond futures complex is the underlying Treasury bond. The correlation of those products is perfectFormer senior Treasuries broker
CME and FICC are already working on revamping the existing cross-margin agreement to increase capital savings for clients. However, some market participants argue CME should move to clear cash Treasuries itself, in a bid to derive the maximum efficiencies between the two.
“I know CME has said FICC business will continue as normal with BrokerTec, but how long is that going to last? The hedge against CME’s entire bond futures complex is the underlying Treasury bond. The correlation of those products is perfect. Add the swaps component into the mix where the underlying in all spread-over business is the 10-year Treasury note, and there is no better dynamic than that,” says the former senior Treasuries broker.
Significant margin reductions are likely to be welcomed by the market. But for banks, they will have to be weighed against the loss of clearing revenues from FICC. DTCC is owned and governed by its 315 participant shareholders, which includes several large banks, who receive rebates and fee reductions during the course of the financial year. These totalled $9.7 billion between 2004 and 2009, according to DTCC data. The allocation of common shares is determined by the fees paid to the clearing house.
Waning bank influence
While there have been no immediate noises from CME about taking the next step from repo and filing a clearing proposal for US Treasuries, any future shift in the clearing landscape would be a further blow to the waning influence of dealers, given the influence they have via ownership stakes in various execution venues and clearing entities such as FICC.
On the trading side, banks have slowly become a minority on major interdealer trading platforms they once owned, such as BrokerTec and eSpeed, ceding ground to more nimble, technology-focused PTFs as the market has become more electronic, like futures.
Though banks still act as core intermediaries in the dealer-to-customer market on platforms such as Bloomberg and Tradeweb – as is the case in the repo market, where they act as go-betweens for cash providers such as money market funds (MMFs) and collateral providers such as hedge funds – some fear CME’s ownership of BrokerTec could be the final nail in the coffin of bank influence over the largest interdealer liquidity pool in the US Treasuries market.
Clearing has, until now, been a different beast, a bank-dominated service since 1986. FICC has 165 members; mostly banks and smaller broker-dealer entities.
PTFs and buy-side firms have struggled to gain access to FICC, with many firms saying its membership requirements put them off clearing their trades, while also making it difficult for new trading platforms to enter the marketplace.
More recently, FICC has allowed buy-side firms into the CCP for repo transactions. Originally, proposals had looked at giving access to MMFs as limited direct members, rather than as clients of a sponsored entity – like a bank. This has now been adjusted to allow institutions to sponsor MMFs.
However, PTFs and smaller broker-dealers that are FICC members are unhappy about new proposals, such as its capped contingent liquidity facility, that they claim are set to make the service more expensive.
The idea of the CME ultimately clearing repo transactions is therefore extremely attractive to a number of PTFs that see this as a way into the market.
“Getting access to repo would increase our abilities and make it more economic to hold positions,” says the chief executive at one PTF. “If I can have access to repo and be a liquidity provider, then maybe I can finance at better rates and enter into new markets.”
CME has been linked with setting up a repo clearing solution for its customers since 2015, pitting it directly against FICC’s solution – the sole incumbent clearer for US Treasury-backed repo trades. Last November, it filed a submission with the SEC to launch a repo clearing service, which is currently under review.
“We have been talking with the SEC this whole time about the framework for margining and governance, as well as the PFMI implementation. The final submission basically starts a process by which they will review it for public comment in the marketplace, so the timeframe as it stands is we’re waiting for publication,” says Suzanne Sprague, managing director for credit risk, banking and solutions at CME.
Sprague, however, maintains CME’s position in its announcement of the Nex deal that BrokerTec will continue to clear US Treasuries and repos through FICC.
Participants argue once a clearing solution for repo is up and running, the shift in gear to clear US Treasuries would not be far behind – and there is little FICC could do about it.
If the full benefits of bringing cross-margining to a number of different problems can be achieved, the move would be welcomed wholeheartedly by even some individuals within the largest banks to get more of the market centrally cleared; as of the end of last year, only a quarter of US Treasury trades were cleared.
“We would love to see more clearing in US Treasury and repo markets. This merger could potentially provide that, but how it plays out, I can’t say. We’d be on board, and it would be a very good piece of plumbing to have with US Treasuries, repos and futures cross-margined together,” says one liquidity expert at a US bank.
Market participants of all stripes say they have concerns about upward pressure on fees from a combined CME-Nex, with the trading and clearing of the bulk of the US Treasury market concentrated in a single entity. Should the deal receive the required regulatory approvals, CME would control the interest rate futures trade lifecycle, the execution of roughly 80% of the interdealer US Treasuries market, and a platform that year-to-date has averaged $262.7 billion in daily US repo volume.
“There may be potential efficiencies that seem like they make sense on paper, but my concern, if I was still a trader, is the places where I would go to trade duration risk and hedge duration exposure. Banks like a bit of competition because it tends to keep fees down. Now those are going to be owned by the same company, there might be less pressure on fees,” says Josh Holden, chief information officer at OpenDoor Trading, and a former US Treasuries and derivatives trader.
A spokesperson for CME says the deal, if approved, will ultimately be of benefit to clients.
“Providing new efficiencies and risk mitigation services by bringing together each company’s capabilities in clearing and post-trade services will help clients address the continuing impact of regulatory capital costs including uncleared margin rules. We are committed to maintaining and further strengthening the valued relationships with Nex’s leading exchange and clearing house partners.”
FICC has also made noises about cutting fees. At a conference held by the Federal Reserve Bank of New York in November, Jim Hraska, general manager of FICC, said the clearing house was looking to lower fees and simplify its model. Risk.net understands that it is hoping to have a proposal approved by the SEC by mid-year.
“FICC is under tremendous pressure to compete with the CME and defend its franchise. FICC gets a lot of its fees from the biggest banks, so they will be concerned about losing them to a competitor at CME,” says the chief executive at one FICC member.
Editing by Tom Osborn
One pot versus two pot
There could be a way to keep the current Treasuries clearing infrastructure intact without a battle. Both clearing houses have a historic cross-margining agreement that dates back more than a decade. This was set up to allow their respective customers to save on margin costs from offsetting bond futures and US Treasuries positions, but it has been criticised by smaller broker-dealers for not achieving desired cost savings.
Senior figures at the Fixed Income Clearing Corporation (FICC) and CME say they would like to see this agreement revamped.
“We’ve long acknowledged the agreement could do with some substantial improvement,” says Murray Pozmanter, head of clearing agency services and global operations at the DTCC. “We’ve had initial conversations with the CME about doing some work to make that a more efficient cross-margining regime. Our intent is to continue to have those conversations with them.”
The current arrangement, established between the CCPs in 2004, according to Suzanne Sprague, managing director for credit risk, banking and solutions at CME, is a two-pot approach, meaning both clearing houses still hold and manage their own positions and collateral. CME and FICC maintain separate margin algorithms and evaluate their clearing members’ positions individually for their respective products. Only after the clearing houses have margined clients will they determine the amount of margin available for cross-margining, known as the residual margin amount.
The benefit of a two-pot agreement lies in its ease of implementation, say clearing experts, because clearing members aren’t required to do anything different when trading between clearing houses with a two-pot arrangement. Instead, FICC and CME work together to facilitate additional offsets for customers.
However, the agreement’s effectiveness at freeing up additional capital remains to be seen. According to CME’s PFMI disclosures for Q4 2017, only 1.69% of the total notional value cleared at CME was cross-margined with FICC, resulting in an initial margin reduction of 1.71%. FICC’s filings reported 19% of total trade value cross-margined with CME, resulting in a 0.25% reduction in initial margin, which represents an aggregate margin savings for participants of $30 million.
CME’s Sprague acknowledges the agreement can be tweaked to better recognise offsetting positions.
“We are aware of public letters written by market participants about inefficiencies that happen today from the lack of opportunity to have additional offsets,” Sprague says. “We don’t have numbers we can run or speak to, but the feedback we’ve heard directly from market participants and seen in comment letters that are public indicate there is an opportunity for additional efficiency.”
DTCC’s Pozmanter points to the launch of new products – such as the CME’s Ultra Treasury bond futures – and changes in how the clearing houses currently operate compared to when the agreement was first signed as reasons why the arrangement could be updated.
However, despite the positive rhetoric, current CCP members are unconvinced the two clearing houses will reach any kind of agreement.
“FICC has been making changes requiring more margin, and our point is that they could do a much better job of integrating with the other clearing agencies to differentiate real risk from hedged risk. But it takes two to tango. CME claims FICC is slow-walking it, and vice versa,” says one trader at a broker-dealer.
“Paying full freight in margin for both when the risk is minimal only serves to negatively impact liquidity. In a nutshell, we would love to see the agreement updated, but I suspect it won’t happen,” the trader adds.
An easier past
Not all of CME’s cross-margin agreements have had as many difficulties. Its partnership with fellow Chicago-based clearing house the Options Clearing Corporation has offered market participants offsets between CME’s S&P 500 index futures and CBOE’s options on S&P indexes, for some 30 years. In CME’s PFMI disclosures for Q4 2017, 7.84% of the total notional cleared value at CME was cross-margined with OCC. That volume saw a 91.24% decrease in required initial margin. OCC reported 1.1% of total volume cross-margined with CME, also resulting in a 91% reduction in initial margin.
The most glaring difference between CME’s partnerships with OCC and FICC is around how the two arrangements are structured. The CME/OCC agreement is known as a single portfolio arrangement – commonly referred to as a one-pot approach. Under this cross-margining agreement, the two clearing houses agree on a common margin system and risk management methodology to net the relevant cash and futures positions together that maximises risk offsets and margin efficiency.
While the one-pot approach creates the most savings, it also necessitates significantly more work to set up. Clearing members are required to put their positions for both clearing houses into a separate account, which they have to fund collateral into.
The CCPs themselves must also agree upon a single margin methodology. Currently, CME and FICC margin interest rate futures and cash treasuries under different methodologies. CME uses its internally created Standard Portfolio Analysis of Risk (Span) while FICC employs historical value-at-risk (HVaR).
In CME’s one-pot agreement with OCC, it defers to its partner’s margin methodology – System for Theoretical Analysis and Numerical Simulations (Stans), Sprague says. Stans calculates clearing member margin requirements at the clearing house level, while CME’s Span is used by clearing members to calculate margin offsets at the account level, she adds.
Sprague says CME has experience understanding margin methodologies across different clearing houses and would “potentially” be comfortable using HVaR in a one-pot approach. However, the choice between a one-pot and two-pot approach is not as clear-cut a decision as some make it out to be.
The differences between a two-pot structure and a one-pot structure are a little bit more detailed than some people may understand currentlySuzanne Sprague, CME
“The differences between a two-pot structure and a one-pot structure are a little bit more detailed than some people may understand currently,” Sprague says. “There are pros and cons to both. It really depends on the market participants.”
If CME and FICC did choose to restructure their agreement, it would not be the first one-pot approach aimed at creating margin efficiency between futures and US Treasuries clearing.
FICC previously operated a one-pot agreement with New York Portfolio Clearing. NYPC was a joint venture between the NYSE Euronext and DTCC created to clear futures listed on NYSE Liffe and offer cross-margin opportunities with FICC. The SEC filing of the proposed cross-margining agreement, which was submitted in 2010, stated preliminary analysis found the arrangement would reduce the initial margin requirements of participants by 20%. A former NYPC executive says that figure was just an average, with the CCP’s analysis calculating some clearing members’ margin reduction reaching 40% under the arrangement.
However, the agreement drew a backlash in comment letters from some market participants and CCPs who claimed it was anti-competitive.
Despite this, in March 2011, the SEC and CFTC granted FICC and NYPC approval on their cross-margining agreement. NYPC began clearing futures contracts for NYSE Liffe later that month. The agreement was short lived, as the Intercontinental Exchange Group (Ice) bought NYSE Euronext in 2013 and wound down NYPC shortly thereafter, choosing to move futures listed on NYSE Liffe to Ice Clear Europe.
The former NYPC executive says the agreement’s failure was a product of its timing. Firms were concentrated on complying with new regulations coming out as a result of the financial crisis. CME has a better chance to succeed with FICC due to its size in the market, as long as it can come to an agreement.
“These are powerful organisations with a lot of collateral. The impact on efficiencies could be pretty significant if they were able to get together,” the executive says. “The difficulty for both is they are dragging along legacy risk methodologies that both probably have vested interests in. For this to work, you have to agree to a unified methodology, default management… because these are two very institutionalised legacy companies, it may be more difficult for them to reach an agreement on that front.”
Whether CME decides to maintain its two-pot approach or switch to a one-pot arrangement, one thing is clear: the agreement will need to be updated. One of the biggest changes, according to Sprague, would be extending the cross-margin agreement to customers. Currently, it serves only clearing members’ own positions. The change could lead to an increase in the margin efficiency of the agreement, but it’s tough to decipher by how much, she adds.
“We can’t speculate,” Sprague says.
Some market participants see CME aiming higher than the cross-margin agreement, though. “I have to believe CME isn’t just going to be a broker like BrokerTec is,” says the chief executive at a Chicago-based proprietary trading firm. “I believe CME very carefully will get into the clearing game, either through repo or through clearing Treasuries.”
The week on Risk.net, May 12-18, 2018Receive this by email