Little did I expect upon going to sleep last night in Boca Raton after participating in a CME-sponsored panel on the CME-CBT merger that I would awake to a reporter’s call this morning with the news of ICE’s unsolicited offer for the CBT.
Here are my initial thoughts on the situation.
First, the economics of a CME-CBT tie up are considerably more compelling than those of an ICE-CBT merger. Far more participants trade far more volume of some combination of Treasuries, Eurodollars, S&Ps and Dow than trade Treasuries (or grains/oilseeds) and coffee, sugar, cocoa, cotton, OJ or Brent Crude (ICE Futures key products).
From an operational perspective, this means that whereas a CME-CBT combination would allow many traders to use one screen, one system, one connection, and one point of contact, with ICE-CBT most traders will continue to have to deal with two systems, two screens, two connections, and two points of contactâ€”just as they do today. Indeed, operationally things will be more complex as they will have to deal more extensively with two clearers, rather than just one.
From a clearing perspective, the potential margining and risk management benefits from joint clearing of CME and CBT products are self-evidently superior to those that could be realized from a CBT-ICE tie up. This is because the CME markets are bigger; the spreading between CME and CBT products more prevalent; and the pool of traders who participate in both markets larger. I don’t recall the industry clamoring for years for joint clearing of CBT and CSCE or NYCE products (CSCE and NYCE being NYBOT/ICE Futures predecessors); they were clamoring for a CME-CBT clearing link that was eventually established after years of back and forth. There was a good reason for that. The savings from the CME-CBT clearing link wereâ€”and areâ€”clearly much larger than a CBT-anybody else clearing link. Moreover, CME has a real chance of making inroads into OTC clearing, which would create other scope economies and increase competition in that market. I don’t think ICE has any realistic chance at doing the same.
Moreover, from a systemic risk perspective, CME’s clearinghouse is financially stronger than NYBOT’s, has experience in operating at immense scaleâ€”scale that dwarfs what NYBOT handlesâ€”and sees a bigger part of the market than a ICE-CBT clearing operation would, meaning that a CME-CBT clearing operation would have a better picture of overall risk exposures.
Second, ICE is already undertaking an integration process with NYBOT. Near simultaneous integration of another, far larger, exchange, is problematic. See my earlier post re the LCH-Clearnet integration fiasco if you need reminding how integration issues can bedevil seemingly sensible deals. Since CME already clears CBT, a good part of the integration battle is already completed there. Moreover, Globex is a proven, scalable platform already used by most CBT participants, so migrating CBT contracts to Globex should be less difficult than migrating them to ICE.
Third, with respect to competition: As I noted in my October 2006 post on the merger, how one views the competitive impact of the CME-CBT deal depends on how one defines the market. If one defines the market as “futures” the CME-CBT merger leads to a substantial increase in concentrationâ€”far greater than a ICE-CBT combination. However, I think this is the least defensible market definition. Under a narrower definitionâ€”i.e., by product line, such as Treasuries or STIRsâ€”the merger doesn’t increase concentration, as the respective exchanges already essentially have monopolies in these products. Moreover, if you are going to define the market as “futures” there is no reason whatsoever to stop thereâ€”you should expand the definition to include OTC derivatives (e.g., vanilla swaps). Indeed, in the case of the CBT, cash Treasuries should also be considered; one can create forward positions that are close substitutes for a futures contract through repo transactions. Nor should you just define the market as the US, as the futures business is increasingly global and with electronic trading location doesn’t really matter that much any more.
Many that I spoke to at Boca believe that a combined CME-CBT entity would have far more pricing power than the two exchanges do separately. Again, I am highlyâ€”emphasize highlyâ€”skeptical. Each exchange already has a dominant position in its own products, and is pricing its services accordingly. Moreover, as the Euronext and Eurex entry attempts demonstrated, the Chicago exchanges know how to defend their turf against new competitors. Entry leads to initial price cutting, but the eventual success of one exchange. Knowing this, potential entrants are likely to remain merely potential, and incumbents will continue to charge high fees unless and until some brave soul decides to trust to hope rather than experience. And the survivor will raise fees again when the loser (most likely the entrant) exits.
[Random thought: why is ICE trying to buy one dominant exchange, rather than competing with it head-to-head? What does that say about the feasibility of competitive entry?]
In my view, CME is not pricing its services now to deter a CBT threat in Eurodollars, nor is CBT pricing to deter a CME threat in Treasuries. Nor where they doing that before Terry Duffy and Charlie Carey met for dinner at Gene & Georgetti’s that fateful October night. They will cut prices when there is a real competitor to ward off, and not before. (The economics/industrial organization/game theory literatures have long recognized that potential competition is likely to have little effect on prices, except under special and arguably unrealistic assumptions about ability to adjust prices like those made in the contestability literature.)
It should also be noted that ifâ€”as has been suggestedâ€”the integration of execution and clearing in a single exchange is the impediment to competition (again something I dispute), the ICE deal is no different than the CME deal, as each involves an integrated clearing and trading operation.
Managing market structure through merger policy will not appreciably impact the competitiveness of the execution business in futures markets. Only something that radically socializes order flowâ€”as the just implemented RegNMS does in securitiesâ€”will have this effect. (Perhaps saying “mandating a centralized open access liquidity pool” is a less provocative phrase than “socializing order flow,” but they amount to the same thing.)
Absent the creation of an open access central limit order book for futures, or its near equivalent (a la RegNMS), the tippiness of order flow, and the difficulty of coordinating the simultaneous movement of order flow to a new exchange, give the incumbent execution venue decisive advantages. (An earlier post suggests that algorithmic trading may virtually link markets and socialize order flow, but this is still a possibility not a reality). Only stupidity (e.g., LIFFE c. 1998) or a refusal to adopt new trading technologies (e.g, LIFFE again, and almost NYMEX last year) gives an entrant a fighting chance.
Put differently, anti-trust policy generally, and merger policy in particular, are extremely blunt tools that, in my view, are not suited for improving competitiveness in derivatives markets, but may well impede the realization of important efficiencies. Thus, in my opinion, derailing a CME-CBT merger would not appreciably improve competition in the futures markets, but could deprive the marketplace of important operational, credit, and systemic efficiencies. For those who want to improve competition in futures in the worst way–well, scotching the CME-CBT deal would be it.
For historical reasons, derivatives markets have not experienced the same disputes over market structure that have raged for decades (three and counting) in the securities markets. Only a dramatic change in regulatory policy carried out under authority granted to the SEC by Congress, and implemented only after years of controversy, has begun to reshape the competitive landscape in securitiesâ€”and even there, it is still to early to determine whether these changes will dramatically improve competition. If futures market participants are really concerned about competition, they had better be prepared for far more drastic measures than preventing this merger or allowing that one. DOJ is incapable of doing the heavy lifting with the tools at its disposal.
I hate to say I told you so, but, well, I can’t resist. At the start of the demutualization wave in the late-1990s and 2000-2001 I opined (in writing, and at conferences like one sponsored by the Chicago Fed in 2001 or early 2002, I forget which) that by becoming more like ordinary firms, futures exchanges would become subject to the same antitrust scrutiny as ordinary firms in other industries. I don’t think this message resonated then. Regardless of what DOJ does now, however, it will soon become conventional wisdom.
I should also note that I believe that the uncertainty surrounding what DOJ will do, and the recent stories to the effect that DOJ was moving to challenge the CME-CBT deal is what precipitated ICE’s move five months after the initial merger announcement. ICE is essentially long a call on what the DOJ doesâ€”if DOJ blesses the CME-CBT merger, the call will almost certainly end up out of the money because CME will win any bidding war. But if DOJ challenges the merger, or imposes unpalatable conditions (especially involving clearing) the call will be in the money. I’d put the option at the money nowâ€”and at the money calls are acutely sensitive to volatility. Uncertainty about DOJ is creating the volatility, making this a canny move by ICE. It is buying vega (or lambda, for you Greek purists) on the cheap.
A couple of final thoughts.
Both ICE and CME are impressive exchanges that are technically progressive. ICE was a technology leader from the get go, by design. CME has also been a technological leader (especially by comparison with other traditional exchanges), but only after demutualization did it fully unleash its technological prowess. Hence, from an innovation prospective, the market would do well regardless of whether CME or ICE ends up owning the CBT at the end of the day.
And lastly something more provocativeâ€”and completely conjectural/speculative. We know that the main opponents to the CME-CBT merger are the major bank/investment bank/FCMs. We also know that a couple of same are major investors in ICE. Could it be . . . ? Just sayin’. No concrete evidence, but some interesting conjectures pop into my Machiavellian mind.