Streetwise Professor

October 17, 2017

Financial Regulators Are Finally Grasping the Titanic’s Captain’s Mistake. That’s Something, Anyways

Filed under: Clearing,Commodities,Derivatives,Economics,Financial crisis,Regulation — The Professor @ 7:11 pm

A couple of big clearing stories this week.

First, Gary Cohn, Director of the National Economic Council (and ex-Goldmanite–if there is such a thing as “ex”, sorta like the Cheka), proclaimed that CCPs pose a systemic risk, and the move to clearing post-crisis has been overdone: “Like every great modern invention, it has its limits, and I think we have expanded the limits of clearing probably farther beyond their useful existence.” Now, Cohn’s remarks are somewhat Trump-like in their clarity (or lack thereof), but they seem to focus on one type of liquidity issue: “we get less transparency, we get less liquid assets in the clearinghouse, it does start to resonate to me to be a new systemic problem in the system,” and “It’s the things we can’t liquidate that scare me.”

So one interpretation of Cohn’s statement is that he is worried that as CCPs expand, perforce they end up expanding what they accept as collateral. During a crisis in particular, these dodgier assets become very difficult to sell to cover the obligations of a defaulter, putting the CCP at risk of failure.

Another interpretation of “less liquid assets” and “things we can’t liquidate” is that these expressions refer to the instruments being cleared. A default that leaves a CCP with an unmatched book of illiquid derivatives in a stressed market will have a difficult task in restoring that book, and is at greater risk of failure.

These are both serious issues, and I’m glad to see them being aired (finally!) at the upper echelons of policymakers. Of course, these do not exhaust the sources of systemic risk in CCPs. We are nearing the 30th anniversary of the 1987 Crash, which revealed to me in a very vivid, experiential way the havoc that frequent variation margining can wreak when prices move a lot. This is the most important liquidity risk inherent in central clearing–and in the mandatory variation margining of uncleared derivatives.

So although Cohn did not address all the systemic risk issues raised by mandatory clearing, it’s past time that somebody important raised the subject in a very public and dramatic way.

Commenter Highgamma asked me whether this was from my lips to Cohn’s ear. Well, since I’ve been sounding the alarm for over nine years (with my first post-crisis post on the subject appearing 3 days after Lehman), all I can say is that sound travels very slowly in DC–or common sense does, anyways.

The other big clearing story is that the CFTC gave all three major clearinghouses passing grades on their just-completed liquidity stress tests: “All of the clearing houses demonstrated the ability to generate sufficient liquidity to fulfill settlement obligations on time.” This relates to the first interpretation of Cohn’s remarks, namely, that in the event that a CCP had to liquidate defaulters’ (plural) collateral in order to pay out daily settlements to this with gains, it would be able to do so.

I admit to being something of a stress test skeptic, especially when it comes to liquidity. Liquidity is a non-linear thing. There are a lot of dependencies that are hard to model. In a stress test, you look at some extreme scenarios, but those scenarios represent a small number of draws from a radically uncertain set of possibilities (some of which you probably can’t even imagine). The things that actually happen are usually way different than what you game out. And given the non-linearities and dependencies, I am skeptical that you can be confident in how liquidity will play out in the scenarios you choose.

Further, as I noted above, this problem is only one of the liquidity concerns raised by clearing, and not necessarily the the biggest one. But the fact that the CFTC is taking at least some liquidity issues seriously is a good thing.

The Gensler-era CFTC, and most of the US and European post-crisis financial regulators, imagined that the good ship CCP was unsinkable, and accordingly steered a reckless course heedless to any warning. You know, sort of like the captain of the Titanic did–and that is a recipe for disaster. Fortunately, now there is a growing recognition in policy-making circles that there are indeed financial icebergs out there that could sink clearinghouses–and take much of the financial system down with them. That is definitely an advance. There is still a long way to go, and methinks that policymakers are still to sanguine about CCPs, and still too blasé about the risks that lurk beneath the surface. But it’s something.

Print Friendly, PDF & Email


  1. Liquidity shortages were the reason the Federal Reserve was created in the first place (Panic of 1907). The Fed can and should play the same role for CCPs. Even Treasuries are problematic if the CCP is forced to dump them into a falling market to meet inflexible CFTC and FSOC regulations. What should happen is that the CCP gets to place the Treasuries with the Fed at the most recent market price BEFORE the crisis, and then be required to redeem them within an established window of time at the SAME price. This process injects required liquidity exactly when it is needed, and then subtracts the liquidity before it distorts asset markets. No academic equations required. CCPs should be subject to some penalty, say 1%-2% per month interest if they do not redeem their securities within the appropriate period of time. The period clock should start once the Fed or BEA or BLS declares the crisis over. This mechanism would also allow CCPs to accept a small amount of fixed income securities other than Treasuries.

    Comment by Pacy — October 19, 2017 @ 6:33 am

  2. You convinced me a long time ago that clearinghouses were the ultimate too big to fail entities (before that I had some off-base ideas). I also describe them as “financial neutron bombs” where they’re designed so that they have the best chance to survive, even if destroy the rest of the financial system. (Did I get that from you as well?)

    They’re also where collateral “goes to die” since you cannot re-lend the collateral anymore and you’ve broken the “collateral multiplier”. Do they take that into account in their crisis stress tests?

    Comment by Highgamma — October 19, 2017 @ 6:24 pm

  3. Comment for Highgamma:
    1) Rehypothecating collateral defeats the purpose of posting collateral at the CCP. The CCP needs first lien control to be able to use the collateral immediately in the event of a default. If an investor defaults, it usually means their position is underwater, so simply selling their position won’t meet all their obligations to the counterparty. Collateral is sized and posted to cover the difference between the market value of the defaulted position and the actual amount owed the counterparty under historical scenarios.
    2) I do not understand the neutron bomb comment. CCPs are processing and risk management service providers. They operate on behalf of their members who outsource the process of clearing, matching, transferring, exercising, managing, etc. derivatives trades to this central service provider. The CCP ultimately rests on the capital base of its member FCMs. If the CCP goes under, that means all of its member brokers have gone under first. The best protection for a CCP is a broad base of well-capitalized members.
    3) Nothing is too big to fail. Regulators just like to feel important, and meddling in markets is how they do it.

    Comment by Pacy — October 20, 2017 @ 7:33 am

  4. @Highgamma-I love the neutron bomb analogy. Mind if I steal it? It is spot on. You didn’t get it from me–my analogy was the levee.

    The ProfessorComment by The Professor — October 22, 2017 @ 5:10 pm

  5. @Pacy-You are missing the main issue re liquidity, which I have written about ad nauseum both here, and in my academic work. It is not so much the ability of a CCP to have access to sufficient liquid resources to cover its obligations in the event of a CM default. Instead, it is the variation margining mechanism which requires market participants to make large cash VM payments in the event of large market moves–which are likely to be precisely when liquidity dries up and it is difficult to fund these VM payments. The Brexit vote shock was the most recent illustration of this. We just passed the 30th anniversary of a more dire example: the 1987 Crash. Read the Brady Report, or the various post-Crash reports done by the GAO, and you’ll see that variation margining intended to keep the CCPs alive (CME, BOTCC, OCC) almost brought down the rest of the financial system. Greenspan intervened to provide liquidity to FCMs. Bernanke wrote a paper in the RFS in 1990 describing this, and defending it, in some detail.

    My concern is that with supersizing the clearing of derivatives (and requiring variation margining of uncleared ones) that this VM-driven liquidity risk has been supersized as well.

    The ProfessorComment by The Professor — October 22, 2017 @ 5:15 pm

  6. Response to the Professor:
    I do understand your point. In unsettling situations where markets are volatile, the various parts of the financial system play a big game of musical chairs to make sure they are holding “enough” cash and some people are left without a chair due to their portfolios (which were perfectly fine before the volatility event). Your objection to VM is simply that CCPs are pushing the market stress out to the market participants – which is where is belongs in the first place. CCPs are NOT market participants – they offer efficiency and transparency but not capital.
    The key principle at stake is the effective transmission of cash throughout the (global) financial system in times of stress, and the equally elastic absorption of the excess liquidity as markets return to whatever passes for normal. No matter how many laws, rules and regulations are written, this is a dynamic function that cannot be managed by market participants and must be handled by human judgement. If the Fed and its sister central banks were to finally acknowledge their role as the liquidity expansion tank for the modern financial system, then we could eliminate most of the financial laws written over the last 40 years. The greatest risk is our current backward situation – central banks flooding the world with unnecessary liquidity at a time of calm, strictly regulating the financial institutions so they become noncompetitive, and then telling them they are on their own in times of crisis. If the financial system is expected to operate on its own without any backstop, then whaty is the point of the central bank in the first place, and why should financial institutions accept all that regulation in return for nothing?

    Comment by Pacy — October 24, 2017 @ 7:28 am

  7. @Street. Steal away!

    Comment by Highgamma — October 26, 2017 @ 3:21 pm

  8. Do you think the policymakers are actually listening though this time around?

    Comment by Vanesa Rose — November 29, 2017 @ 12:37 pm

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress