SWP Acid Flashback, CCP Edition
Sometimes reading current news about clearing specifically and post-crisis regulation generally triggers acid flashbacks to old blog posts. Like this one (from 2010!):
[Gensler’s] latest gurgling appears on the oped page of today’s WSJ. It starts with a non-sequitur, and careens downhill from there. Gensler tells a story about his role in the LTCM situation, and then claims that to prevent a recurrence, or a repeat of AIG, it is necessary to reduce the “cancerous interconnections” (Jeremiah Recycled Bad Metaphor Alert!) in the financial system by, you guessed it, mandatory clearing.
Look. This is very basic. Do I have to repeat it? CLEARING DOES NOT ELIMINATE INTERCONNECTIONS AMONG FINANCIAL INSTITUTIONS. At most, it reconfigures the topology of the network of interconnections. Anyone who argues otherwise is not competent to weigh in on the subject, let alone to have regulatory responsibility over a vastly expanded clearing system. At most you can argue that the interconnections in a cleared system are better in some ways than the interconnections in the current OTC structure. But Gensler doesn’t do that. He just makes unsupported assertion after unsupported assertion.
Jeremiah’s latest gurgling appears on the oped page of today’s WSJ. It starts with a non-sequitur, and careens downhill from there. Gensler tells a story about his role in the LTCM situation, and then claims that to prevent a recurrence, or a repeat of AIG, it is necessary to reduce the “cancerous interconnections” (Jeremiah Recycled Bad Metaphor Alert!) in the financial system by, you guessed it, mandatory clearing. Look. This is very basic. Do I have to repeat it? CLEARING DOES NOT ELIMINATE INTERCONNECTIONS AMONG FINANCIAL INSTITUTIONS. At most, it reconfigures the topology of the network of interconnections. Anyone who argues otherwise is not competent to weigh in on the subject, let alone to have regulatory responsibility over a vastly expanded clearing system. At most you can argue that the interconnections in a cleared system are better in some ways than the interconnections in the current OTC structure. But Gensler doesn’t do that. He just makes unsupported assertion after unsupported assertion.
So what triggered this flashback? This recent FSB (no! not Putin!)/BIS/IOSCO report on . . . wait for it . . . interdependencies in clearing. As summarized by Reuters:
The Financial Stability Board, the Committee on Payments and Market Infrastructures, the International Organization of Securities Commissioners and the Basel Committee on Banking Supervision, also raised new concerns around the interdependency of CCPs, which have become crucial financial infrastructures as a result of post-crisis reforms that forced much of the US$483trn over-the-counter derivatives market into central clearing.
In a study of 26 CCPs across 15 jurisdictions, the committees found that many clearinghouses maintain relationships with the same financial entities.
Concentration is high with 88% of financial resources, including initial margin and default funds, sitting in just 10 CCPs. Of the 307 clearing members included in the analysis, the largest 20 accounted for 75% of financial resources provided to CCPs.
More than 80% of the CCPs surveyed were exposed to at least 10 global systemically important financial institutions, the study showed.
In an analysis of the contagion effect of clearing member defaults, the study found that more than half of surveyed CCPs would suffer a default of at least two clearing members as a result of two clearing member defaults at another CCP.
“This suggests a high degree of interconnectedness among the central clearing system’s largest and most significant clearing members,” the committees said in their analysis.
To reiterate: as I said in 2010 (and the blog post echoed remarks that I made at ISDA’s General Meeting in San Fransisco shortly before I wrote the post), clearing just reconfigures the topology of the network. It does not eliminate “cancerous interconnections”. It merely re-jiggers the connections.
Look at some of the network charts in the FSB/BIS/IOSCO report. They are pretty much indistinguishable from the sccaaarrry charts of interdependencies in OTC derivatives that were bruited about to scare the chillin into supporting clearing and collateral mandates.
The concentration of clearing members is particularly concerning. The report does not mention it, but this concentration creates other major headaches, such as the difficulties of porting positions if a big clearing member (or two) defaults. And the difficulties this concentration would produce in trying to auction off or hedge the positions of the big clearing firms.
Further, the report understates the degree of interconnections, and in fact ignores some of the most dangerous ones. It looks only at direct connections, but the indirect connections are probably more . . . what’s the word I’m looking for? . . . cancerous–yeah, that’s it. CCPs are deeply embedded in the liquidity supply and credit network, which connects all major (and most minor) players in the market. Market shocks that cause big price changes in turn cause big variation margin calls that reverberate throughout the entire financial system. Given the tight coupling of the liquidity system generally, and the particularly tight coupling of the margining mechanism specifically, this form of interconnection–not considered in the report–is most laden with systemic ramifications. As I’ve said ad nauseum: the connections that are intended to prevent CCPs from failing are exactly the ones that pose the greatest threat to the entire system.
To flash back to another of my past writings: this recent report, when compared to what Gensler said in 2010 (and others, notably Timmy!, were singing from the same hymnal), shows that clearing and collateral mandates were a bill of goods. These mandates were sold on the basis of lies large and small. And the biggest lie–and I said so at the time–was that clearing would reduce the interconnectivity of the financial system. So the FSB/BIS/IOSCO have called bullshit on Gary Gensler. Unfortunately, seven years too late.
Prof, I understand you want to be paid for instructing people. But can you explain to an ignoramus like me why clearing is now more risky, please. In a pre digital age it might have taken days, in a pre telegraph age it might have taken weeks. Today it probably takes seconds. So to my stupid mind surely the risk has diminished. If not, why not?
Comment by james — July 6, 2017 @ 10:02 pm
@james-Special offer, just for you: this one’s on the house 😉
The issue is not the routing, execution, or settlement of the trade: I think the communication methods that you refer to would affect those things. When I am discussing clearing, I am referring to the post-trade process related to performance risk. Clearing involves netting offsetting exposures, setting and collecting initial margin on outstanding transactions, marking positions to market, calculating, paying, collecting variation margin based on changes in mark to market, managing defaulted positions, and in the event that a defaulter’s resources are not sufficient to cover the losses on its positions, allocating the shortfall to the default fund and/or clearinghouse capital.
Digitization/computerization have obviously expedited these processes, but the crucial fact is that (a) these activities must be performed as long as a position is open, which can be years (e.g., a 30 year interest rate swap), and (b) performance risk exists through the life of a position, which again can be years.
The change in the risk that results from clearing mandates relates mainly to the forced (by regulation/legislation) increase in the scale of clearing. Positions are bigger. Positions tend to be longer-lived. Margin flows (and hence liquidity needs) are bigger.
Computerization/digitization affects where operational risks can arise in the process. Moreover, due to tight coupling, an operational failure can cause a catastrophic failure of the entire process. For example, on Black Monday, 1987, several failures of the FedWire interrupted the flow of margin payments from New York banks to the Chicago clearinghouses. This had several knock-on effects.
Hope this helps. Let me know if you have remaining questions.
Actually, this entire blog effort is free provision of a public good 😉 (Or public bad, depending on your perspective!) I am glad to answer questions, particularly to expand on or clarify what I write on the blog. So there’s no need to preface a question with an apology for asking.
Thanks, Prof. For not minding me asking.
I may be none the wiser but I am better informed, as the old joke goes.
Another Q
If clearing is subject to solvency and long term, this is like risk assessment. Is it more likely that the hovel burns down in 10 years or the manor burns down in 20. Insurance was invented for this.
As far as I can see (sub IQ) the clearing risks are so opaque and diverse that it is impossible to assess risk.
Cometh the hour cometh the opportunity. Some outfit with deep capital, GS Zurick re, BoA, AIG (ho ho!), even the Fed could take risk in exchange for the hedge. Get the insurance premium for acting as principal.
Then they go to DC and get a tweak on the rules, and bingo, they make gold out of fleece.
Do I at least get a pass mark?
Comment by james — July 7, 2017 @ 2:26 pm
I think the 3 net of all this is that in future crises the lender of last resort will have to lend collateral, not just money. the idea that centralizing relationships somehow makes them less complex or subject to cross market infection in case of panics is sooo stupid that only a Timmy or other bureaucratic self aggrandizer could delude themselves into thinking so. A quick read of Kindelberger should have fixed that!
Comment by sotosy1 — July 7, 2017 @ 2:53 pm
I think comments 3 & 4 are a tad too cynical. Clearing centralizes credit risk (which can be better managed) and adds credit enhancements (variation & initial margin & mutualized default fund) which did not exist in varying degrees in the bilateral OTC world. Unfortunately, it creates a new too big to fail in the process. I’d be interested in the counter argument from a credit risk perspective against mandatory clearing.
The FSB paper and some other recent bureaucrat-ise highlights the increase in liquidity risk. CCPs require liquidity instantly in the event of a default, which is exactly the time when liquidity will dry up. How accessible are credit lines from FCM affiliates and repo markets at that time? Who knows? I’m an arm chair QB but I’d guess they’re not. If you ask someone in Chicago, they’d probably say they wouldn’t require additional liquidity, as they’ve calculated appropriate reserves.
Also re: the lender of last resort comments…Dodd Frank already allows the fed to take deposits & provide liquidity under certain circumstances (I believe the Fed has to vote & receive Treasury approval). In both cases, the CCP must be a SIFMU and there are 8 of them. I’m not sure how that would play politically with Main Street.
Comment by Zora Duntov — July 7, 2017 @ 6:42 pm
Obviously the way to reduce CCP interdependencies is to mandate a single global CCP. Preferably issuing its own currency to guard against liquidity risk. You can send my Nobel prize to my soon to be constructed fortress designed to survive the inevitable zombie apocalypse when that defaults.
Comment by noir — July 8, 2017 @ 1:57 am