Standardized Stress Tests ‘Desirable' for CCPs, Fight Continues Over Granularity

Apparently, when it comes to stress testing, it's quite easy to both completely agree and disagree, all at the same time.

Last week, the Committee on Payments and Market Infrastructures (CMPI) and the International Organization of Securities Commissions (IOSCO) announced that they are reviewing stress-testing guidelines for central counterparties (CCP).

As a result, the saber-rattling began as CCPs and market participants try to mold just how prescriptive these new standards should be. On Thursday, during a panel discussion at the FIA Boca 2015 conference, speakers agreed that baseline standards for CCP stress testing will be good for the industry.

"One size doesn't fit all, but I think it's important for market participants to have some baseline to do some average backwards comparison," noted Sharon Bowen, a commissioner at the Commodity Futures Trading Commission (CFTC).

Kevin McPartland, principal at consultancy Greenwich Associates and who recently conducted a survey on risk and central clearing, added that creating an apples-to-apples comparison for the various CCPs is important because right now that doesn't quite exist.

"An apples-to-apples comparison is important," he said. "As we were writing the research I started to dig into it and tried to show numbers side-by-side, demonstrating ‘skin in the game', for example, but I decided not to go there because it's hard to do that with apples-to-apples."

The Fight: Granularity

The two CCP participants on the panel agreed that standardized stress testing would be valuable. Where they seemingly diverged was when it came to the level of granularity.

Sunil Cutinho, president of CME Clearing, the clearing arm of CME Group, was in the corner advocating against stringent granularity because each CCP has a significantly different risk profile.

"I don't think that there's a disagreement on standardization. We love the effort from CPMI," Cutinho said. "The issue is about getting too granular. You can't get too granular because the risk profile of the CCPs are fairly different. The products that the CCPs clear are quite different. So trying to attempt one single stress test that fits all CCPs will not be a successful effort."

On the other side of the debate, David Weisbrod, chief executive officer of LCH.Clearnet, LLC, said that it was important not to dumb the test down because if it is, and the exercise isn't rigorous, the industry will be creating more risk because there will be an assumption that every has passed a stringent stress test, when they were in actuality passing a "stress-test lite" version.

"We do agree that standardized stress testing is desirable. The issue is in the detail as to how this is done," he said. "How you do a standardized stress test, across jurisdictions and multiple products with different CCPs and different risk management standards is a task. But, it's one that we are engaged in and supportive of and that we think will be beneficial.

"We agree with the goal," he continued. "We want to make sure that it incorporates all the necessary detail so that we don't dumb it down ─ in other words, reduce it to the lowest common denominator ─ and therefore inadvertently create more risk than we're trying fix."

Full Disclosure

Beyond the level of granularity, the other concern that Cutinho presented was around disclosure. Once again, he was for disclosure of the fact that a stress test was submitted, but he was against disclosing the results to the industry, outside of the regulators.

"Disclosure is very important and we agree with disclosure," Cutinho said. "But there must be a discussion around the disclosure of stress tests versus the disclosure of the results of stress tests. We have to be very careful. You cannot disclose a market participant's risk to another market participant. The disclosure is to the regulators and it's a continuous disclosure."

Finally, Cutinho said that it's important to go a step further when it comes to stress testing, and require clearing firms to undergo stress tests as well.

"Stress testing is not just at the CCP; the CCP is just one slice of purse," he says. "When a clearing member firm fails, there is an implication to the client. ...So if you're going to have stress testing, let's go all the way and not just to protect one aspect of the chain."

The Bottom Line

● CPMI and IOSCO said that they would review CCP stress testing practices and requirements, though no additional context was provided.

● Regulators, trading firms and CCPs at FIA Boca agreed that standardized stress testing was a good idea, at a basic level.

● Where they disagreed was around the granularity of information needed to run these stress tests, and to whom the results should be disclosed to.