In November 2023, CCP Global – the trade body for central counterparties – co-ordinated an international default simulation to see how clearing houses and members could cope with the failure of a major member firm.

The fire drill – dubbed CIDS, or CCP international default simulation – was designed to see if the industry could put out a fake blaze on the scale of 2008’s Lehman Brothers conflagration. Its findings were issued in May.

During simulated emergency auctions, members bid on the portfolios of a supposed defaulting firm, assumed to be one of the five largest, common to no fewer than 32 CCPs worldwide. The hypothetical fire starter was dubbed ACME – A Clearing Member Everywhere.

Putting data into a risk system and pricing it systematically – rather than humans manually doing things – has to be core to making this process more robust, scalable and efficient

Head of derivatives clearing at a large bank

The European Securities and Markets Authority (Esma), among other regulators, assisted CCP Global in orchestrating the drill, which was structured so that CCPs ran their own default simulations without co-ordinating.

“This lack of artificial choreography is realistic,” explains Teo Floor, chief executive of CCP Global. “Contrary to expectations, the lack of auction co-ordination didn’t illustrate any problems.”

What’s more, Froukelien Wendt, an independent member of the CCP supervisory committee, and director for CCPs at Esma, believes that the main outcome of the drill was that no bottlenecks were identified.

And while this upbeat assessment was shared by nine participating firms that subsequently described their experience to Risk.net, there were also valuable lessons indicating room for improvement.

Clearing executives at two European banks, on the one hand, found the exercise carefully planned, and saw no serious concerns for default management practices among the results, but others pointed to the need for increased data standardisation in CCP processes. Some felt the ability of clients to move – or port – their positions to a new clearing member in the event of a default was another area that needed attention.

The exercise also brought forth suggestions for how to turn up the heat to make future fire drills even more realistic and to encompass other kinds of risk event – not solely to flag possible systemic risks, but to ensure that clearers are better prepared in the event of a real emergency.

Data deficiencies

Many participants cited standardisation and automation as significant means of reducing operational risk in CCPs’ default management processes, reports Esma, in its role as one of the drill’s co-ordinators.

Indeed, the issues that respondents raised about the lack of standardisation are similar pain points to those already identified in a 2020 report from the Committee on Payments and Market Infrastructures (CPMI) and the International Organization of Securities Commissions (Iosco).

I personally think having a surprise element is an excellent idea. But it’s up to the industry to see whether this is workable

Froukelien Wendt, Esma

Default management processes are “systemically important, in terms of trying to reduce the fallout in the event of a major member default”, observes the head of derivatives clearing at one large bank. At the top of the pile, standardising data “really stands out” as an action that could have a big influence on the outcome of an auction and on client porting.

“Putting data into a risk system and pricing it systematically – rather than humans manually doing things – has to be core to making this process more robust, scalable and efficient,” says the head of derivatives clearing.

This view is backed up by non-bank liquidity provider Optiver, which says the distinct pricing methodologies, file formats and timelines of CCPs would create inefficiencies in a true crisis situation, leading to bottlenecks.

Further recommendations included standardisation of bidding formats and conventions, the terminology, format and communication methods of CCP announcements and the timelines of CCP default management processes. Triallists also suggested that default management portals, rather than email communication, could aid in reducing operational risk.

Others said the use of market conventions and references to external data providers in auction files – as well as internal CCP references – would benefit their processes.

Gustavo Pinto, European post-trade manager at Optiver, says an easy win would be for CCPs globally to agree a standard minimum set of requirements to describe a portfolio, perhaps the 10 most basic fields: “The first 10 columns will be the same for everyone. And then if a CCP wants to add other columns for any given reason, they can add these at the end. But we know that the first few columns will be the same for everyone.”

A further point would be standardising pricing methodologies, adds Pinto. One example would be to give the previous day’s reference closing price to price a futures contract. The practice of providing reference settlement prices for futures is currently inconsistent across CCPs.

Optiver hopes that a trade association can incentivise a more standardised approach among CCPs, or that a regulatory body such as Iosco can issue guidance that will lead to less cumbersome practices.

Wendt says Esma takes the view that CCPs, clearing members and clients should come up with a plan to enhance standardisation, likely through trade associations. If there is insufficient improvement after the next fire drill in 2025, however, Esma may consider consulting with international partners from a global regulatory perspective.

The head of derivatives clearing thinks regulatory intervention would be needed to drive CCPs towards standardising data.

“I don’t think CCPs are going to do something on their own, even though members have long said: ‘This doesn’t work for us.’ The only way that would change is if there was some kind of standard issued by regulators which has consequences for non-conformance.”

Tell us the code

Bank clearing sources are sceptical that data could be easily standardised without firms incurring more cost. In many cases, auction data sent by CCPs to clearers uses exchange codes devised by post-trade service providers, such as Ion or FIS GMI. These codes then need to be mapped to the codes from data vendors, such as Bloomberg and Reuters, that traders use to bid on auction portfolios. The vendors in turn charge licensing fees for use of their codes.

“You have to bear in mind, CCPs are for profit, so assuming that auctions don’t generate any revenue, it’s just a cost,” says the head of clearing at a futures commission merchant (FCM). “Nobody will be jumping up and down to spend a ton of money to pay Bloomberg or Reuters for getting their codes.”

This lack of artificial choreography is realistic. Contrary to expectations, the lack of auction co-ordination didn’t illustrate any problems

Teo Floor, CCP Global

Gary Saunders, global head of prime derivatives services at Barclays, says there should be a multi-CCP effort to invest in the technology to standardise auction platforms and rule books, where it makes sense, “to make the system safer, given the size of risks that we’re talking about”.

“Efficiency and standardisation of process is important to help best manage those risks,” he says. Saunders wants the file format that CCPs use to send data to banks to be represented in the same way – no matter if a swap comes from CME Clearing, Eurex, JSCC or LCH – to allow traders to look at the risk they’re being asked to price in a quick and consistent way.

This efficiency is key, Saunders says: “What costs you money in crisis situations is unnecessary delays in the process. Once you’ve decided that a default is happening, then speed of action is important.”

Who’s in charge?

The other area where participants are keen to see more standardisation is in the conduct of the default auction process itself. If CCPs coalesced around a consistent approach to interaction with clearing members and those bidding on portfolios, dealers could train staff on just one CCP process, rather than several, sources suggest.

However, there’s an inherent tension between CCPs wanting to maintain some discretion in the auction process and the members and bidders looking for standardisation across the board.

“Specifying to the nth degree may pose risks to CCPs from litigation,” says Joshua Hurley, director at consultant Davies Group, alluding to the possibility that CCPs could be challenged if they deviate from expected standards. Also, potentially worse market outcomes may result if processes are too rigidly maintained. “Documentation and guidebooks should be provided, and the general process needs to be understood by all parties." 

CCP Global’s report of the CIDS exercise says participants think submission formats could be standardised across CCPs. For example, clearing houses should make it clear whether positive/negative signage on submission forms means bidder pays or bidder receives, for example.

You can’t simulate porting because you don’t know what the situation is going to be like in a real crisis, both in the market and with individual clients

Gary Saunders, Barclays

Some CCPs argue they are already well down the path of standardising data for default management purposes.

“I think some progress has actually been made in that area,” says Hannes Hauenschild, head of default management at Eurex Clearing. He points to a 2021 paper on the issue, CCP default auctions best practices, by trade associations including CCP Global.

“Eurex co-developed and adheres to that, and we are actually a strong supporter of having this consistency and common standards to the largest extent possible,” says Hauenschild. “If members tell us that a certain file format or a certain data point would be helpful for them, if possible, we’re happy to implement this.”

Lee Betsill, chief risk officer at CME Clearing, agrees that the CCP’s clients want auction data in a standardised format. But there’s a catch: they want the standardised format to conform with the coding already in use at its clearing house.

“Our participants say standardisation would be a great idea, but just don’t change what you’ve done at CME for us because we’ve already adapted to it. We’ve already built our processes for CME Clearing,” says Betsill.

Port in a storm

Client porting was tested by 13 of the 32 CCPs that took part in the 2023 CIDS exercise. A consensus among participants is that this element deserves greater focus in future.

Rafael Plata, secretary general of the European Association of CCP Clearing Houses, says the 2025 drill should expand its simulation of client porting – especially given that the tool “may be hampered in some jurisdictions”, such as the UK and the European Union, because of legal barriers.

Rafik Mrabet, managing director at derivatives risk analytics platform Cumulus9, and ex-head of risk at clearing house Ice, agrees that porting client positions during a default scenario is a critical aspect of default management that ensures continuity and market stability.

“In my experience at Ice, we found that incorporating realistic client porting scenarios significantly enhanced our preparedness and provided valuable insights for improving our processes,” says Mrabet.

Participants in the 2023 drill agree that testing porting may not produce appealing results, but is essential to understanding if default management processes are flammable.

“Porting is really hard, and quickly turns into a train wreck, so it’s actually [a] good reason to drill this more often,” says the head of clearing at an FCM.

Delays can occur in getting receiving members to sign off on accepting client portfolios, as they rightfully need to understand margin requirements, the impact on risk-weighted assets and to take into account know-your-customer (KYC) and anti-money laundering (AML) matters in onboarding.

“In a period of crisis, nobody wants to move really quickly on this, even though time is of the essence,” he adds.

But porting may be harder to simulate than a default management process where people are asked to bid. The pipes and plumbing for moving portfolios around between different clearers can be tested, and indeed often are tested daily or weekly in a business-as-usual situation.

Where it gets more complicated, says Barclays’ Saunders, is if clients have to port suddenly to a clearing member that is unfamiliar with the client, or does not have credit lines or available capital in place to accommodate the port.

“You can’t simulate that because you don’t know what the situation is going to be like in a real crisis, both in the market and with individual clients,” he says.

Porting is really hard, and quickly turns into a train wreck, so it’s actually [a] good reason to drill this more often

Head of clearing at an FCM

The bank head of derivatives clearing says porting adds “another dimension of stress”, illustrating the need for clear data from CCPs – and that more time is needed for porting than drills typically allocate.

“The fundamental questions about client porting would be: what’s the size of the risk in the portfolio? What’s the risk we already clear for this name? And what’s the liquidity drain that we’re going to see from taking on this activity?” says the derivatives clearing head. To increase the success of client porting, the windows available for it require extension, as well as ensuring that data formats are consistent in order to streamline processing, they add.

Esma’s Wendt agrees that porting is in theory a good tool in a crisis, but in practice is complex, given the operational and legal issues. It’s an exercise that CCPs could test themselves by looking at which clients with one clearing member already have a contract with another clearing member, says Wendt, and porting could form a larger feature of drills. Legal issues acting as impediments to porting deserve attention from regulators, she adds.

Taken unawares

One clearer thinks the next CIDS fire drill would benefit from being more radical – drawing on the element of surprise.

“I still don’t think it fully tests the market as to what would happen in the event of a member default,” says the head of derivatives clearing. “People could have been well-prepared and made sure that they had the relevant system access and the desk was staffed.”

They suggest notifying participants that a drill will take place at any time during a certain quarter, rather than on a specific day: “I think that would pressure-test the system. It would be an extra level to this drill. I don’t know that it would be that pretty, but that is another way to make it more realistic.”

CCPs are cautious about the idea.

Eurex’s Hauenschild says surprise tests have the merit of truly exposing how well stakeholders react, but the concept would need to garner sufficient support in the clearing community.

I think for drills, you do need to give a forewarning, rather than just ‘it’s going to happen at some point in three months’. I think the benefit of the simulation would be diminished if it was a surprise scenario

Conor Sherrard, Citi

Betsill says CME Clearing already runs unscheduled tests for some of its internal drills to prepare for auctions, liquidations or porting. But he doesn’t think the market as a whole would want the same for the external drill.

Optiver, meanwhile, doesn’t completely discount the suggestion, but would favour reducing the window to something less than a quarter. Anish Puaar, head of European equity market structure at the firm, says leaving a window as long as three months could reduce willingness to participate in the drill, “if it comes as too much of a surprise”.

Barclays’ Saunders says fire drill exercises take up a lot of resources, and a balance needs to be achieved between making them as real as possible and not overly burdensome on participants. 

Conor Sherrard, chief operating officer of prime services at Citi, agrees: “I think for drills, you do need to give a forewarning, rather than just ‘it’s going to happen at some point in three months’. I think the benefit of the simulation would be diminished if it was a surprise scenario.”

In reality, crisis events rarely happen overnight. The risk of a collapse at Lehman Brothers became apparent days or even weeks before its eventual fall, giving clearing houses and other members time to prepare.

Esma’s Wendt says: “I personally think having a surprise element is an excellent idea. But it’s up to the industry to see whether this is workable.”

CCPs are for profit, so assuming that auctions don’t generate any revenue, it’s just a cost. Nobody will be jumping up and down to spend a ton of money to pay Bloomberg or Reuters for getting their codes

Head of clearing at an FCM

To make the drill more realistic, Sherrard also thinks it could be conducted in a narrower timeframe, rather than stretched out over days. And reducing the number of CCPs would make it more manageable for participating members, rotating the CCPs for each subsequent drill to ensure any differences between CCPs are captured.

Wendt promotes the idea of evolving and improving drills in each exercise to enhance their capacity to detect risk.

She says that the 2023 drill was predicated on a relatively straightforward scenario of one large clearing member defaulting. But based on the key takeaways – including the need for further standardisation – the next exercise should see those lessons turned into actions.

“So, it’s not simply again running each CCP’s default fund management procedure, but also adding a few elements that widen the scope of this exercise compared to what we saw in November 2023. There are other scenarios than the default of a clearing member,” adds Wendt.

“There’s also a scenario where you assume a big operational risk event that calls for very different information streams,” she concludes. “I personally hope that we will not only improve in the scenario of default of a clearing member, but we also look broader to other types of scenarios where I think the operational risk scenario is a really good alternative.”

Editing by Louise Marshall