The Jury in the CME-Floor Trader Suit Finds That a Deal Is a Deal–Just Like on the Floor
A jury in a class action lawsuit filed by a group of Class B shareholders (ex-floor traders) found in favor of defendant CME Group. The suit claimed that the CME had breached its contract with the would-be Killer Bs, most notably by opening the Aurora Data Center (“ADC”) and providing access to it to non-Class B shareholders. As the Court (Judge Patrick J. Sherlock) summarized it, according to the plaintiffs
the [ADC] is a trading floor within the meaning of the [Class B Shareholder’s] Core Right 2–hence, only members are eligible to exercise trading righs and privileges at the ADC. The CME Class contends that the defendants breached their Core Rights by (a) allowing non-members to trade CME products from the ADCO trading floor, (b) requiring Clas Be members to pay co-location fees from the ADC, and (c collected co-location fees from market participants without sharing them pursuant to the revenue sharing plan with members
The CME Class also contends that defendants breached their contractual rights by allowing non-members to trade at member rates, providing discounts and special rates to large customers that are better than rates available to members, and charging co-location fees to CME Class members by eliminating the benefits of the CME fee schedule.
I can’t say that the verdict was a surprise, mainly because anything can happen in a trial, especially one with a jury. Which is precisely why such large cases never make it to trial.
Although I did not hear all the testimony based on the original complaint and the record in the case (notably rulings on the motion to dismiss and summary judgment) I think that the verdict is correct. There is a yawning gap between specific facts alleged in the complaint and far broader inferences drawn therefrom.
The complaint relies heavily on quotes from Leo Melamed’s book, For Crying Out Loud: this makes for good atmospherics, but it hardly qualifies as a contractual commitment. It also contains a few facts relating to the Core Rights established under the CME demutualization plan. It then draws inferences from these that are, in my opinion, far too broad and unsupported.
It is clear that the Core Rights were focused primarily on protecting the CME floor. This was essential to get the member’s support for demutualization. Section Three of the article related to Class B Shareholders in the revised Certificate of Incorporation that formalized the demutalization is labeled, in all caps, COMMITMENT TO MAINTAIN FLOOR TRADING. It lays out various volume metrics which if met, precluded the exchange from terminating floor trading in a contract.
The CME rules at the time also provided members (who became Class Bs) “GLOBEX Screen Rights” (emphasis added for a reason that will become clear shortly), and guaranteed Class B shareholders would have access to the GLOBEX2 trading system and definted eligibility requirements for access to a GLOBEX terminal. The complaint also quotes Melamed committing that Class B shareholder clearing fees would never be higher than the clearing fees charged others.
And here’s where the yawning inferential gap appears. From these facts, the plaintiffs claim that in demutualizing the CME gave the Class B shareholders the “Right to Best Access and Closest Proximity to the Globex Platform, and to Allow the Class B Plaintiffs to Share in Globex-Related Revenues.” Not from the facts in the complaint, it didn’t.
Granting Class B shareholders the right to trade on Globex did not, when evaluated on the face of it, give them exclusive access. Nor did it give them “best access and closest proximity” under all future technological improvements to Globex. It gave them access. Via terminals (screens).
The plaintiffs attempted to argue that ackshually post-ADC, Globex was the floor, and since they had exclusive access rights to the floor (and nobody else did), they had a right to Globex on the best possible and exclusive terms under all technological conditions. They claim, based on a quote from Melamed, that since “ultimately electronic trade would become the real ‘price-discovery pit'” therefore Globex was a “virtual trading floor.” Since they had exclusive rights to the meatware trading floor, they had rights to the virtual trading floor. QED!
Nice try, but no cigar. The fulsome description of the commitments that the CME made to maintain the floor makes it abundantly clear that, within the four corners of the contract that the plaintiffs claim the CME breached, the floor was, well, the floor. Those cavernous spaces on South Wacker, not a large air conditioned building filled with humming servers.
The contract did not provide exclusive access to the locus of price discovery, whatever that turned out to be, floor, Globex, or mental telepathy. By the understanding of everybody at the time, the floor was the place where guys traded futures contracts and futures options by screaming and making funny hand gestures.
In sum, taking the allegations in the complaint as true, the CME promised to provide access to Globex to the Class B members. It did provide access to Globex, and continued to do so. It did not promise to provide “best and most proximate access” to any trading platform the CME introduced up to the end of time. That’s an assertion made by the plaintiffs that is not supported by actual evidence.
After a major technological revolution, the CME provided a new type of access–co-location, with higher fees for shorter cables. In my view, the plaintiffs made a completely unsupported, an extremely large, logical leap to assert that the access rights granted with demutualization necessarily conferred preferential access rights to the new technology.
And technological change is the crucial issue here. What this all comes down to is a failure of imagination by all concerned–including Leo Melamed. Specifically, a failure to imagine the emergence of HFT, and algorithmic trading generally. Trading that was not only not face-to-face, but was untouched by human hands via keyboard, mouse, or touch screen.
At the time of demutualiztion, computerized trading was screen trading–and it was often referred to as such. Indeed, even by the standards of screen trading as it became, 1999-era computerized trading was cumbersome. Not until TT screens were introduced (along with copycat products) in the early-2000s did screen trading become user friendly and fast. But in the minds of virtually everyone in the industry at the time, “computerized trading” meant “a human inputting orders to a computerized limit order book via a screen/terminall/keyboard/mouse.”
I was an early and pretty lonely predictor that computerized trading would eventually displace the floor, yet I did not foresee how computerized trading would evolve. My predictions were mocked when made in the mid-1990s: when I released a study (done under the auspices of Deutsche Terminborse, later Eurex) finding that liquidity on computerized and floor markets was approximately equal, the late Robin Wigglesworth, then Chairman of LIFFE, ridiculed me on the pages of the Financial Times as an ivory tower academic who had no clue about the real world. I restrained myself from gloating (publicly, anyways) when Eurex nearly killed LIFFE in 1998, and LIFFE shut its floor and went electronic.
So it’s fair to say that I was more foresighted about computers v. floor than most, and especially most actually in the industry, when I predicted the eventual triumph of electronic trading. But my foresight was far from perfect, as I did not foresee how electronic trading itself would evolve.
In other words, a lot of history was packed into the 12 years between demutualization and the opening of the ADC. The CME responded to the technological change by introducing a new form of access that was tailored to market participants who had virtually (no pun intended!) nothing in common with the Class B shareholders. The new system catered to those who specialized in virtual markets.
If the CME had not done so, it would have breached its duties to the Class A shareholders and perhaps doomed the exchange to oblivion.
The mismatch between the human capital of the Class Bs and the heaviest users of the ADC is massive. The Class Bs did not have the skills or knowledge necesssary to produce value by trading on Globex via co-lo.
Here’s what likely would have happened if the CME had provided access on the basis the Class Bs claim they were promised. Since they could not exploit the nature of the ADC themselves–most couldn’t even make the leap to screen trading–they would have arbed by selling or leasing their access rights to those who could. Or the CME would have bought out Class B shares. And what a negotiation that would have been, given the vast difference between the value of co-location to the typical Class B and its value to the Virtus and Jumps of the world (and hence to the CME). That and the personalities of the Class Bs–personalities suited to the no-holds-barred bargaining on the floor.
In the event, CME concluded that it did not owe the Class Bs anything more than providing some access to Globex (basically on the terms shaped by 1999-2000 technology) and keeping the floor open unless and until the computers killed it. As they eventually, and arguably inevitably, did.
Class B shares did plunge post-ADC. But that’s because the real value of the Class B shares was driven by the exclusive right to access the floor, not by rights to access CME trading platforms by all means imagined or unimagined. ADC at most accelerated the demise of the floor, and recognition of that dynamic cratered the value of floor trading rights, which foreshadowed the closure of the floors (for futures and many options) a few years later.
History did not work out well for the Class Bs. Like 18th-19th century weavers, they were overwhelmed by technological change that they did not anticipate, and which was not anticipated by those who displaced them. They bargained for a demutualization deal predicated on their beliefs about computerized trading technology–as did the CME. The technology changed in ways that few if any even dreamed of. This made their deal far, far less favorable than they had anticipated. In 1999-2000 they hit their drive–right into a sand trap they didn’t see.
So they wanted a mulligan, and the lawsuit was their way of demanding one. But a jury found that they got what they were promised, and a mulligan was not among the promises.
Watch the movie Floored and you can have some sympathy for this particular group of individuals who like many before were rendered obsolete for technological change. The potential for such sympathy is why I thought a jury might find for them, despite their weak legal case. But the jury instead apparently went with the principle that a deal is a deal, even if the deal doesn’t work out they way you like.
Which, ironically–was the way the floor operated.
I know all the principles on both sides. I was on the CME Board at demutualization and on the Strat Planning Committee. There was a reason I was interviewed and not called to testify by either side. I am NOT a member of the class. I told people on the floor that the value of the B share was going to be a math equation post demutualization. You get a fee break, and continue to get a fee break today. The number of round turns you trade per month determines if you should stay retail, lease, or purchase a B share.
If 90% of traders who tried trading on the floor failed, 99.9% of floor traders who tried trading on a screen failed. You lost your edge and instead of being a “professional” trader, you became “retail” and we all know what happens to retail in the trading environment. They get killed.
I only know testimony from anecdote, not actually hearing it. What you say is true, and Craig Donohue’s testimony basically said that. What I heard about the plaintiff’s testimony was not good and no one did a great job on the stand.
The structure of the exchange, A and B shares, was not strategic. It was determined by the IRS. One tenet of demutualization was that it had to be “tax-free transition”. No cap gains taxes etc. We waited a year and a half for the IRS to rule on our initial proposal-just convert memberships to stock. It didn’t fly, so we preserved the B share. Ironically, in the initial demutualization, the attorneys drew it up incorrectly. Sidley and Austin didn’t “staple” an A share to the B to make it tax free and we had to redo the entire vote. They refunded their fees entirely. Hence today, there are A shares which anyone can buy. There are B shares, and anyone can buy them, but there is one A share permanently stapled to the B. So, B’s get dividend checks for one share of stock.
The plaintiffs, for their part, either didn’t do a good job of explaining to their attorneys what the core issues were, or in the years-long run-up to the trial, the motions were denied etc and the trial focused only on a segment of their initial complaint.
Everything was going swimmingly with demutualization until 2009. In 2009, the NYMEX was purchased. CME paid $750k for their core rights. Many CME members and CBOT members wondered where their check was. Additionally, CME fenced off products in the data center the same way they did on the floor. So, if I owned a membership at CME in order to get fee breaks at exchanges CME bought, I had to buy their B share memberships. Of course, CME had no incentive to co-mingle memberships or to give preference to the original CME membership. But, seeing a $750k check written to members of another exchange while at the same time seeing your livelihood destroyed didn’t sit well. To give you some idea of the edge on the floor: By May 1 of 2003, I was up $750K trading Eurodollars. By December 31, I was down $250k because that was the year it went electronic. (might have been 2002, don’t exactly recall). In 2008, I made $2.4MM trading hogs. Market went to the screen in 2009 and I lost $1.2MM. For those that don’t know, it’s my own money (like your checking account).
At the same time, CME offered huge discounts to certain classes of traders to trade electronically, and subsidized traders on the floor in pits to move the market to the screen. They didn’t let it “naturally” transition contrary to anything they would testify to. It would have moved for sure, but maybe stayed on the floor a few more years. They also made it impossible for floor traders to creatively access the screen. As a spread trader, you couldn’t put a person on a headset and desk in front of a screen without putting a B share on them (increasing your costs since you had to buy the equipment, rent the desk space, rent the membership etc) and the handheld tech the exchange gave you was very slow compared to co-location. Everything the exchange did to the floor trader INCREASED their costs, rather than decrease them. In Strat Planning, all I ever heard about was how cheap electronic trading was….but the reality is not that.
I do think co-location fees could be argued in a way but I am not convinced a judge/jury would understand it or rule differently. The ADC is a data center but it is the defacto trading floor. CME floor members had a speed access to order flow. How does that remain when you sell out the top step? We can argue separately on this issue, but if you co-lo, you see the market and what’s in it before anyone else. On the floor, that would be like showing your entire deck to the pit and that is illegal. This is why firms will spend over $1M on an acre of land located right next to the Aurora data center. The edge is everything.
There were rumors of thousands of traders using one account to trade and not leasing B shares. The Board denied it but the rumors persisted. In court, that never came out. The other thing technology did was centralize volume. Jump Trading accounts for the lion’s share of trading in US Treasuries. In the pit, it was spread out over hundreds of traders. It isn’t “good” or “bad” it just is a function of tech. Winners take most in social media too.
The other piece no one reported on was leasing. The plaintiffs in the suit were not true market makers. They leased their seats out and collected a monthly check. Sort of like an annuity. Typically, there was a loose relationship between the value of the seat and the value of a lease. Leases in the late 80s early 90s went skyhigh as trading volume exploded. I knew people paying $7500 a month to lease. Seat values went from $580k in 1992 to 1.2MM in 1994. As electronic trading started coming in, lease and seat values plummeted to $280k in 1998 (This is for a B-1, or full membership). CME corporate has no incentive to market leasing because it cuts into their topline revenue.
Another piece of color the plaintiffs didn’t use that objective legal analysis doesn’t care about, but was used by CME to make their case since trials are as much theatre as law. Most CME/CBOT members didn’t hold on to their stock because they needed it to live. So the idea that someone got stock on an IPO and should have held it seeing it worth 30x what it was on the IPO is unrealistic. This wasn’t a bunch of rich guys suing a rich corporation. I know more than 15 people who have killed themselves due to money because the floor ended. There is a trail of divorces, drug/alcoholism etc. Remember, people on the floor had no qualifications other than that they were on the floor. It was 5000 people who couldn’t be employed by anyone. If you were a decent trader, you could make $250k a year. If you were great, you could make millions…per year. That doesn’t exist outside of those four walls. Sure people had college degrees but spend 10 years on the floor and a normal corporate company wouldn’t hire you-and lots of other firms wouldn’t either because they couldn’t identify your true skillset. It’s hard to tell people what that skillset is and how it can be applied as well.
Personally, I sold a membership as soon as I couldn’t lease it out. Kept trading on the other one until I couldn’t make money. I sold it. I lived from 2009-2016 on savings, and selling CME stock. I got an MBA at Chicago Booth—>but no one would hire me even with that credential. In 2014-15, I was at the lowest point I ever was in my life. No hope at all. Almost moved to Florida and became a realtor. Most traders went into businesses like that (insurance, realtor, wealth mgmt) I invested in startups and since that is similar to trading, I was pretty good at it fortunately and ended up on my feet and retired. But, not too many were like me.
If I were a benevolent dictator, I’d figure out a way to get rid of the B share. The B share guys wouldn’t get what they want, but at the end of the day the archaic fee system CME has could be redone based on the elasticity of demand for that product and volume. My guess is that volume would explode further, and they would make more money. CME is not filled with out of the box thinkers anymore though, so they only see legal hurdles and opportunity costs. This lawsuit won’t help.
CME and other exchanges have a regulatory monopoly, or at least monopsony. We will see if tech disrupts them like Amazon disrupted Sears but it is not likely.
Comment by Jeff Carter (@pointsnfigures1) — July 28, 2025 @ 6:59 am
Jack Wigglesworth. Robin Woodhead was his CEO.
Also there is a Robin Wigglesworth with the FT.
@ Jeff thanks, what a mess.
Comment by Ex-Global Super-Regulator on Lunch Break — July 29, 2025 @ 4:52 am
@Ex-Global Super-Regulator on Lunch Break. Thanks. Good catch. I’ll correct. It’s been 25 years + I and I was going on memory. Presumably I had seen Robin’s name more recently. I also couldn’t remember if Wigglesworth was LIFFE chair or CEO. Thanks for clarifying.
Hope all’s well.
Comment by cpirrong — July 29, 2025 @ 9:01 am
@Jeff Carter. Thanks for the long and thoughtful reply.
A couple of brief comments. I agree that ADC was “de facto the trading floor,” if by floor you mean locus of price discovery/where most of the flow went. That’s basically what plaintiffs tried to argue. But the issue in the case was whether it was a trading floor de jure, and there I think the answer is pretty clear. The jury did too, apparently.
It is somewhat odd that the CME hasn’t decided to clear the decks and get rid of the B shares, or more precisely, get rid of them by buying them out. Perhaps the trial made it impossible to proceed with that, and now that’s over with CME will move to a more conventional capital structure that doesn’t entail the governance complications that this one does. I don’t know whether tax considerations might make that complicated to do. But the Coase Theorem suggests a deal will get done. There are gains from trade between the exchange and the Bs.
Regarding the CME slashing fees to induce a move of volume to the computers, well, it was acting in its own interest in a way that was consistent with the structure set up in the demutualization. CME didn’t commit itself not to do things in the future to shift volume to Globex.
“Jump Trading accounts for the lion’s share of trading in US Treasuries. In the pit, it was spread out over hundreds of traders.” Yes, there are substantial economies of scale in electronic market making/HFT. That’s exactly one of the things that few–myself included–did foresee.
Ironically, my engagement with DTB went as follows. LIFFE and other pro-floor parties said the reason that LIFFE was more liquid was because of locals. So DTB asked me to explore the idea of creating a special member type, electronic local. My first step in that process was to see whether, in fact, DTB was less liquid. I found it wasn’t, and so told DTB there was no reason to create this new member type. They didn’t.
But it is interesting that this reveals that the mindset at this time was that liquidity providers on the screens would be like locals–individuals scalping. In the post-TT, pre-HFT days, that was arguably the case, although as you point out, and as your personal experience demonstrates, the floor local skillset didn’t translate to the screen, for many reasons.
But when the HFT/algos emerged, they blew that individual trader model out of the game. The Jumps, DRWs etc. could exploit the law of large numbers and turn a tiny edge into a way of making profit with almost no risk. (I recall Virtu reporting at one time that they hardly ever had a losing day).
Like I wrote in the post, I sympathize. There was a lot of personal suffering caused by the transition. People like you–though you seem like one of the lucky ones, given your finding something else you were good at–were what got destroyed in the process of creative destruction. There is no major technological change that doesn’t create losers.
Comment by cpirrong — July 29, 2025 @ 9:35 am
While the stories from the end of the trading floor are heartbreaking, at least those members who owned seats got stock that eventually became quite valuable. There are many other people who lost out to technology that did not have that potential lifesaver of stock, for example, those who paid over a million for a NYC taxi medallion. My understanding is that Nymex members got a check because their exchange was being bought outright unlike CME & CBOT who merged.
Comment by Andrew Stanton — July 30, 2025 @ 8:09 am
@Prof
Apologies, that message was a bit terse, I was tapping it out while on public transport.
Saw the name ‘Robin Wigglesworth’ and thought ‘He’s on the FT staff, yes?’ and went from there.
The story that you and Jeff tell is tremendously interesting: what a mix of complex and competing incentives; sorry to hear that so many people were screwed over by what should have been trustworthy counterparties.
Everything here is as well as it can be, as the world pushes inexorably towards what appears to be an impending great power conflict. If what we’re seeing in Ukraine is anything to go by, as a later-middle-aged male I might be one of the first people dragooned onto the front lines as drone-fodder! Currently learning as much Rashan, Chaynese and Japanese as I can to try to win myself a rear-echelon position in intelligence.
Comment by Ex-Global Super-Regulator on Lunch Break — August 7, 2025 @ 9:58 pm