Thursday, December 6, 2018
Has an airline ever told you it wasn’t responsible for costs you would incur because of a flight initially delayed for “mechanical reasons” that was now delayed due to weather? In thinking about how potential losses resulting from both a clearing member default and non-default issues (such as operational, investment, or custody problems) are likely to be allocated among a clearinghouse and clearing members, I think of such past travel experiences. From my perspective, the issue of ownership is key to both.
Hence, I'd like there to be an increased amount of discussion about clearinghouse ownership. The word “ownership” was barely mentioned (maybe once) during the December 4th meeting of the Market Risk Advisory Committee, sponsored by CFTC Commissioner Rostin Behnam, which largely focused on issues related to clearinghouses. In contrast, participants frequently mentioned clearinghouse capital (“skin in the game”), and extensively discussed (Panel 2) the allocation of default versus non-default losses (which could occur nearly simultaneously). Surprisingly, there is scant legal guidance on the allocation of non-default losses in the U.S.
Clearinghouses, financial market infrastructure utilities, tend to be owned by their members or by investors (I’ve written extensively about clearinghouses for readers interested in learning more). HOWEVER, in the case of both ownership structures, members bear the tail risk of a member’s default. Hence, in investor-owned clearinghouses, there is an important misalignment of incentives. Those receiving a clearinghouse’s profits (the investors) are not those primarily bearing the risk of a clearing member’s default (the clearing members). Federal Reserve Bank of Chicago policymakers Robert Cox and Robert Steigerwald have referred to this separation of ownership from primary financial responsibility for default as “incomplete demutualization” (historically, clearinghouses were mutualized institutions, but many transitioned to investor ownership beginning in the early 1990s).
Aren’t incentive problems almost always at the heart of financial crises? If clearinghouses are member-owned, the issue of whose capital, its placement, and the appropriate amount in the clearinghouse’s default waterfall should be mitigated. Likewise, responsibility for default and non-default losses should be simplified: the members would be responsible. Member ownership should also simplify many other issues participants noted: the misalignment between investors’ commercial incentives and members’ risk-management incentives in investor-owned clearinghouses; the amount of transparency to members of margin risk models; the amount of transparency to members surrounding stress testing; members’ role in governance and risk management; members’ good faith participation in auctions of a defaulted member’s portfolio; and a clearinghouse’s need for flexibility in a crisis.
As I’ve written about, the rulebooks of many clearinghouses – for example, ICE Clear Credit – include breathtakingly expansive emergency provisions (see Rule 601) allowing the clearinghouse to take almost any action necessary to resolve an emergency. In a sense, such provisions enable the clearinghouse to become a private-market “resolution” authority empowered to resolve the crisis at hand (indeed, clearinghouses in the U.S. are self-regulatory organizations). Such flexibility could be tremendously beneficial to a clearinghouse in a crisis. Importantly, as some participants noted, future crises aren’t likely to be carbon copies of those in the past. I would think, however, that the members of an investor-owned clearinghouse are likely to be less enthused about the flexibility offered by such emergency provisions.
In sum, many issues related to clearinghouse recovery and resolution (the problem of what to do with a distressed/insolvent clearinghouse) currently being discussed could be simplified by increased consideration of a more bedrock issue: clearinghouse ownership. Indeed, it’s baffling that this foundational issue has received such scant attention (a notable exception is a paper by Professor Paolo Saguato). I’m not suggesting that member ownership of clearinghouses would solve all issues, especially that which I regard as the most important: emergency liquidity in a crisis. And, certainly, remutualization of clearinghouses would have its own non-trivial set of problems. I am suggesting, however, that at a minimum, there should be increased amount of discussion about clearinghouse ownership, and that member ownership could ameliorate many tensions between clearinghouses and clearing members.
In the remainder of this post, I highlight and comment upon additional items from the MRAC meeting.
Participants flagged a finding from an August 2018 report by several global standard-setting bodies, Incentives to centrally clear over-the counter (OTC derivatives): a mere 5 – yes, 5 – bank-affiliated clearing member firms hold 80% plus of cleared client margin for the US, UK, and Japan. That’s problematic.
However, I personally consider another August 2018 report by the same authors, Analysis of Central Clearing Interdependencies, more worrisome (just check out its figures). It details interconnections among global clearinghouses, clearing members, and service providers (often clearing members or their affiliates). Some examples: 2 clearinghouses hold 40% of the amount of global prefunded financial resources at clearinghouses, and another 8 hold 50%; the 11 largest clearing members of those surveyed (a total of 306) have connections to 16-25 clearinghouses, hence a default at one risks triggering defaults of that clearing member (or its affiliates) at 24 additional clearinghouses because of cross-default clauses; and many large clearing members (or their affiliates) provide 3+ services (in one case, 6) to the clearinghouse. Check out the figures in the report.
Participants alluded to the August 2018 default by an individual power trader clearing member at a NASDAQ clearinghouse. While research into the cause of this event appears ongoing, some commentators have suggested that the clearinghouse’s margin model could have played a role. Both clearing members and the clearinghouse suffered losses, and 2/3 of the default fund was exhausted. Hence, the clearinghouse required members to make additional capital contributions to recapitalize the default fund. During the meeting, a participant suggested that as NASDAQ is a publicly-traded company ($14 billion dollar market capitalization), it would be reasonable to suggest that it increase its contribution to the default waterfall (it reportedly contributed 7 million Euros prior to the default).
A participant briefly remarked that clearinghouse default funds for crypto assets should be kept separate from default funds for other assets. From my perspective, this makes complete sense, at least for the near future. However, the CME explains: “Bitcoin futures will fall into CME’s Base Guaranty Fund for futures and options on futures, as any newly listed futures.” The CME, Inc. (because of its division, CME Clearing) is a designated, systemically significant financial market utility under Title VIII of Dodd-Frank. This issue of crypto asset clearing is itself worthy of its own post!
At least one participant noted that some clearinghouses (those designated by FSOC as systemically significant FMUs), but not all, now have access to accounts and services at a Federal Reserve bank. As I’ve written about, this significant change to historical practice (traditionally, only depository institutions had access to Fed accounts and services) was enabled by Dodd-Frank’s Title VIII. It has financial stability, competition, and a host of other implications and concerns. Past WSJ reports (here and here) suggest that the Fed may also provide accounts for the clearing-related cash collateral of pension plans, asset managers, and hedge funds. Indeed, who wouldn’t want an account at the Fed? In fact, why not have A Public Option for Bank Accounts (Or Central Banking for All) as some legal academics have recently suggested? To be clear, I’m not taking a position on such issues in this post. I am, however, arguing that although esoteric, such issues as who has access to an account at the Fed are critical social policy choices with real world implications that merit broad-based public debate.
A final item, also related to competition concerns, is that, as a participant noted, significant clearinghouses are essentially monopolies. While clearinghouses might be natural monopolies, I’ve always also been baffled that there isn’t more discussion of anti-trust concerns in the global clearing ecosystem (one exception being a paper by Professor Felix Chang).
Today, I was heartened to read about how some airlines seem to be learning from past oversights. Let’s not wait for a catastrophe to intensify public debate about critical issues currently receiving little discussion in the clearinghouse space.