Risk Awards 2016
In late 2014, bond traders noticed a discrepancy in the pricing of the CME's flagship US Treasury bond futures. At one point, the price of 30-year Treasury contracts expiring in June 2015 gapped more than 11% higher than those expiring in March. The discrepancy didn't reflect a sudden and dramatic shift in US rate expectations, but rather a distortion in the underlying Treasury market – a hangover from a five-year gap in 30-year bond issuance by the US debt management office between 2001 and 2006.
Because the contracts are settled physically, that presented the market with a significant challenge. The party on the short side of the trade has a choice of which T-bonds to deliver, with a bias towards delivering the cheapest – those with the shortest remaining maturity. In the March-to-June 2015 roll for 30-year US Treasury futures, the difference in term to maturity of the cheapest-to-deliver bonds between the previous and upcoming contract was five whole years – a huge jump from the usual three months.
With the roll date for the futures approaching, CME acted fast to help the contract's users.
"The market presented a number of challenges and we faced them with our clients and created new solutions and innovations to address those challenges," says Sean Tully, senior managing director of financial and over-the-counter products at CME Group in Chicago.
Many investors seek to maintain an on-the-run short or long exposure to US Treasury futures through calendar spread contracts, selling them to stay long or buying them to stay short. In response to the heightened roll risk between the March and June expiries, CME unveiled an additional calendar spread, which allowed investors to take a 3:2 exposure across the expiries, rather than the usual 1:1 ratio.
An investor using the 1:1 ratio would have to do a significant amount of work to get the final balance in its risk exposure right. With the exposure much higher in the March to June contract, the 1:1 ratio would have required a far greater number of outright transactions to achieve the same net risk exposure; a position in 1,000 of the March contracts would require a long position in 665 of the new June contracts.
The numbers tell the story of the market's receptiveness: almost 800,000 contracts with the 3:2 ratio traded during the March roll, making up 63% of Globex volumes and 52% of overall trade volumes.
This is a very significant contract in the single most liquid government bond market in the world
Sean Tully, CME Group
Some would argue such innovations have come at a price, though. Members of many major exchanges grumble that fees – whether for trading, clearing or data – crept up across the board last year. CME Group chief executive Phupinder Gill acknowledges CME was no exception in this regard, but argues the rises were necessary – and kept to a minimum.
"Fees are one of the concerns folks have with exchanges as large as ourselves, and we have been ultra-sensitive to that. When we raise fees, we do it with sensitivity of the types of products we deliver, and the quality of the risk exposure coverage we give our client base. Any increase in fees is so we can invest in technology and other solutions to bring the bid/offer spread down, which creates the liquid market that CME group has," says Gill.
Navigating the five-year gap in Treasury issuance meant reaching out to users of the contract to educate the market on the implications, and get feedback on the exchange's proposed solution. CME did so via webinars, marketing material and more direct interaction.
"It was a completely new function, and we had to go out and educate the market. This is a very significant contract in the single most liquid government bond market in the world. It was very important that participants knew how this roll was going to occur and that they were ready for the change relative to the duration. This has never happened before, and we went out and provided the technology customers needed," says Tully.
Following the closure of the bourse's open-outcry trading pits, market participants also indicated they would prefer to define their own Treasury spread ratio orders on CME's Globex platform. The result was the August 2015 release of a new user-defined spread contract, a format that attracted a volume of 80,000 in the first September-to-December roll.
"It allowed our customers to trade our product exactly the way they want to trade it. So while the amount of roll done on the floor was relatively small, we wanted to make sure each participant got the experience they wanted relative to the way they trade, so we made sure we could replicate that floor experience electronically," says Tully.
The Treasury issuance gap resulted in a situation where the existing TY 10-year Note future had a cheapest-to-deliver bond with remaining maturity of seven years, and then following the jump of the gap, the cheapest to deliver for the classic bond futures was 20 years, resulting in requests to launch a new contract somewhere in the middle.
In response, the bourse announced in October it would launch a new flavour of 10-year US Treasury note futures and options in early 2016, allowing delivery of 10-year US Treasury notes with remaining term to maturity of at least nine years and five months and not more than 10 years. For existing futures, delivery is allowed for notes with remaining terms of at least six years and six months.
The week on Risk.net, July 14–20, 2017Receive this by email