Oral argument — The Nasdaq Stock Market LLC v. SEC

Case number 21-1100 et al. The Nasdaq Stock Market LLC et al. Petitioners v. Securities and Exchange Commission. Mr. Hungar for the petitioners. Mr. Freda for the respondents. Counsel for petitioners, you may begin. Thank you, Your Honor, and may it please the Court. The Commission admits that the goal of its market data infrastructure rule was to reduce information asymmetries based on the perceived problem of a two-tier data market in which market participants who are willing to pay more receive more data faster than others. But the Commission's own findings establish that the rule will actually exacerbate the problem it set out to address. That is classic arbitrary and capricious decision-making. If the Commission's costly new regulatory regime operates as it's intended to, instead of a so-called two-tiered data market, the Commission will actually have made things worse by producing a multi-tiered market in which, as it concedes at pages 186, 221, and 241 of the Joint Appendix, different market participants will receive data at different speeds with different content and different reliability, all depending on how sophisticated they are and on how much they're willing to pay, precisely the supposed problem that the Commission said it was addressing. In fact, by codifying the privileged category of self-aggregators who have an inherent latency advantage over all other market participants, and by precluding anyone except broker-dealers and investment advisors from enjoying that specially favored status, the Commission has actually formalized two tiers of data recipients, and it expects the competing consolidators to create additional tiers through their varied product offerings with varied pricing, content, and latency. And the Commission has made things still worse again by relaxing the regulatory requirements for data reliability in the current regime and discarding the heavily regulated single-processor system. As the Commission recognized at page 241 of the Joint Appendix, some competing consolidators will offer less reliable data services than others, meaning that price-sensitive market participants, the ones most likely to hire the less expensive and less reliable consolidators, will face consolidator-specific data outages to a greater extent than their better-heeled competitors who can afford to hire more reliable consolidators, meaning that there's now going to be a whole new category of information asymmetries created by the rule that doesn't exist in the current system where everyone can rely on the heavily regulated and reliable securities information processors that provide data to the entire market. For all these reasons, the rule is arbitrary and capricious and should be vacated. Mr. Hungar, your briefing uses the language of expropriation of proprietary market data, but you're not here making a constitutional claim. Correct. Are you? That's correct. You didn't make such a claim before the agency. I believe such a claim was made before the agency, but that's not part of our submission today. We're relying here on the Administrative Procedure Act arguments that we are advancing. One, obviously, under the Administrative Procedure Act and the Exchange Act, that expropriation and its impact on competition is one of the key elements of our arguments about why what the Commission has done here is arbitrary and capricious. Unless the Court has questions about the first point, about the dramatic disconnect between what the fundamental problem the agency said it was addressing and the fact that its alleged solution actually makes the problem worse, which in and of itself is fatal to the rule, I can turn to the economic aspects, which are a separate and independent reason for invalidating the rule. Why isn't the agency, or the Commission, entitled, as it were, not to want a system where two groups control, or even one group controls, and an open competition system? Understanding that competition can be messy, and it's not as neat and clean as having a couple of monopolists run the show. But, in its judgment, there are other considerations that, in the agency's opinion, warrant a more competitive system than now exists. Even where one of the two favored groups, profitability, may be somewhat adversely affected until the new market adjusts, and we see where everyone comes out. Your Honor, the problem here is not competition. In fact, in the existing regime, the exchanges compete vigorously with each other in the proprietary data markets, as the Commission concedes at page 27 of the Joint Appendix. And the Commission's regime is eliminating that competition by imposing its centralized price controls regime on that formerly competitive market. The problem isn't that the Commission says it's bringing competition. The problem is its rationale for the regime that it has set up is directly contradicted by its own findings about how that regime is going to work. Again, the fundamental premise of this rule, the fundamental justification for it, is that the Commission says there's a problem. And what is that problem? According to the Commission, it says this even in its brief at pages 3 and 19. The problem is information asymmetries. That is, according to the Commission, and this is from page 219 of the Joint Appendix, the problem is the two-tiered market in which market participants who pay more get more data faster. But the solution that they have proffered for that perceived problem doesn't fix the problem. It doesn't ameliorate the problem. And that is arbitrary and capricious. And, again, we're relying on the Commission's own predictions and findings about how it's going to operate, which means it's going to increase latency variation, increase content variations. And, again, the self-aggregators in this new regime, the Commission is codifying for the first time a special category of market players who have a privilege that no one else is entitled to, which is this privilege of receiving this data themselves as market aggregators and acting immediately on it while everybody else is waiting for the competing consolidators to aggregate and then distribute. This is a new codified version of the two-tiered market that the Commission says it's trying to get rid of. And then the point I made before about reliability, the fact that the Commission is now admitting that it's going to create a system in which reliability varies from one provider to another, meaning that for the most price-sensitive participants, they're actually going to be at risk of suffering asymmetric data outages, a risk that does not exist in the current regime. So in all these ways, they've made the problem they claim to be trying to fix worse. That is arbitrary and capricious. Maybe they could have justified. We don't think so, but they could certainly have tried to justify the rule in a different way. And then there might well be other fatal flaws, but the one we've identified here would not be an issue. But when an agency says, we're trying to fix problem X, and here's our solution to fix problem X, and they make problem X worse, that is arbitrary and capricious. Well, worse from whose perspective? That's part of what the Commission is getting at. Well, with respect, Your Honor, they admit that there could be reduced latency, or rather increased latency, that is slower data and less content. And the competing consolidators don't have to provide all the data to everyone. Indeed, the Commission expects that they won't. And they don't have to provide it all at the fastest possible speeds. Indeed, the Commission expects that they won't. And they're going to compete based on price and reliability and all these other factors. So the Commission's own findings contemplate a multi-tiered market, which is inconsistent with their claim of reducing information asymmetries, which, again, was the alleged purpose and goal of this rule. I want to try to understand the mechanics here, and I'm not sure that I really have a handle on it. The consolidators are – is this correct? The New York Stock Exchange and NASDAQ? At present, there are two securities information processors, one of which – and they are affiliated respectively with those two entities, yes. And the idea is that maybe some other entities will enter into that – be able to enter into that market. How would that work? I mean, the information would still be coming from just the New York Stock Exchange and NASDAQ, right? Well, no. So there are approximately 16, I think, exchanges, more or less, plus a number of dark venues that are not exchanges in the traditional sense. They don't reveal their quotation data. That's why they're called dark venues. But so there are – I think there are maybe 40 or – anyway, there are many trading venues that exist. And all of that information is consolidated currently by the securities information processors. Okay. So it's not just the two exchanges, information from the two exchanges. It's information from all – however many there are, and there are a dozen more, right? Correct. Okay. Now, the self-aggregators, where do they come in? Would they be consolidators that only use the information for their own particular purposes rather than selling it to somebody else? Is that the difference? Correct. They'll be the – again, only broker-dealers or investment advisors, the entities that have been trying to obtain reductions in proprietary data fees, only they would be able to qualify as self-aggregators, and they can receive this price-fixed data directly from the exchanges in this new regime and do whatever they want with it while everyone else is waiting for the competing consolidators to aggregate and distribute the data. That raises a question that I really don't know the answer to. Why aren't they doing that now? Why aren't the broker-dealers doing that now on the current regime? Well, some broker-dealers do do that now because, again, proprietary – in addition to the securities information processors that sell the – that provide the consolidated feed, there are also – there's a very competitive market among all the exchanges that choose to participate in it, which is not just NASDAQ and NYSE but others as well, that sell proprietary data in addition to the consolidated feed. And so entities that want to do that can do that now. In fact, anyone can – I mean, any – you, if you want to, you can go buy that data yourself. It's about $15 a month. That was my next question. Day traders, for example, they must subscribe to instantaneous quotations. Otherwise, they couldn't operate. Exactly. So in the current – because, again, the market for proprietary data, separate and apart from the consolidated feed, is already competitive, as the commission admits. So you can go to NYSE or NASDAQ or SIPO as a private individual day trader and say, I want to buy the full depth of book data, all this data that the commission is talking about directly from the exchange, and you can do it. And as I say, I think for NASDAQ, my client, I think it's $15 a month for an individual private trader. So it's not like it's confiscatory or expensive for somebody who's actually engaged in that sort of practice in the current regime. $15 a month? Yes, that's my understanding. Mr. Hemgar, what about – I mean, the SEC's position is that there won't be any latency differentials between the consolidated competitors and the self-aggregators. I mean, what is your view of that? Well, I don't think that's actually – I mean, they say that in their brief, but I don't think that's actually their position because they concede in their order. I believe it's at page 374 of the joint appendix, but it's cited in our brief. In fact, there may well be a latency advantage for the self-aggregators. And simply as a matter of physics, it has to be true because the competing consolidators and the self-aggregators will receive the data at the same time. They will then process it in whatever way they choose, and then the competing consolidators will have to transfer that data to their customers while the self-aggregators are trading on it. And all you have to be is ahead of your competitors. And if you're in this world, which, again, is a very small world of the most sophisticated traders that are trading on computer algorithms where microseconds, millions of a second make a difference, right? It's indiscernible to you and me and to an individual who's trading based on what they see with their own eyes because 100 microseconds, 1,000 microseconds – 100,000 microseconds is indistinguishable to us, but to a computer it matters. And so for those people, the difference between being able to trade while the competing consolidator is sending its data to somebody else before they can trade means they're going to get there first and they're going to get the trade. So, I mean, the commission is right to concede in its order that the self-aggregators may well have a latency advantage, and it's clearly going to be the case. Mr. Hunger, a lot of – I'm just wondering, you know, kind of on the ground what's been happening since the rule has gone into place. I mean, and maybe this is something Mr. Frieda can speak to as well, but I'm just – I mean, a lot of this comes down to, you know, do the SECs properly predict what would happen with competition? And I'm wondering if you have any thoughts about what actually has been materializing in the period. I know it's being phased in, the rule, but I'm just wondering if there's any sort of on-the-ground observations. Not that I'm aware of. And remember, the rule – the phase-in – there is a phase-in, but we haven't gotten to the point yet where anybody's even decided whether they're going to enter the purported competing consolidator market. So there's not really any evidence at this point. I haven't – I see my time is running low, but I haven't had a chance to talk about the second important element of our claim, which is sections – the violation of sections 3F and 23A2, which this court has said impose a unique obligation on the commission. And it completely failed that obligation. It failed to adequately consider whether the rule will promote competition and efficiency, and failed to explain how the rule's burdens on competition are necessary or appropriate in furtherance of the Act's purposes. And it did that in several respects. First, the commission expressly found at 384 of the joint appendix that the rule could improve the competitive position of dark venues relative to the exchanges, but it made no attempt to explain how this competitive impact is necessary to further the Act's purposes, which specifically include assuring fair competition between exchanges and dark venues. That's section 11A1C2 of the Act. The commission didn't even address that, which in itself is fatal to the rule. What is a dark venue? A dark venue is a trading venue that is less heavily regulated than the exchanges and that doesn't make public quotation data. That is, exchanges – the fee that we're talking about here provides bids and asks. Here's what you can sell and buy a stock – a share of stock at on an exchange. The dark venues attract people who don't want to have to publicize their quotes and want to trade in a secretive – more secretive fashion, which injures – The second problem that the commission didn't address in terms of competitive and efficiency impacts is by favoring dark venues, the commission admits the rule may well shift additional order flow to dark venues, which of course undermines the purported goal of the rule to enhance information transparency in the market because the more order flow on dark venues, the less transparency there is in the market because the dark venues don't disclose quotation data. So again, the commission failed utterly to explain how shifting order flow to dark venues is consistent with the purpose of the act, yet the statute expressly required it to do that. So let me ask you, who uses these dark venues then? If I have some money I want to invest, why would I go to a dark venue when they won't disclose? The most large traders, sophisticated traders use the dark venues because they see an advantage in not having to disclose – not letting people see what they're doing ahead of time, basically. Which again, it undercuts the transparency. So just so I understand the operational, they just say we have X dollars invested? The large broker-dealers, the same ones that are advantaged by the other aspects of the rule, also operate many of these dark venues. I don't know exactly how they work. My client's not a dark venue, right? But you don't know how they work. A big trader will go to the dark venue with their offers or bids rather than to the exchange, knowing that the market isn't going to see what they're doing because it's a dark venue. So they perceive an advantage in being able to trade large quantities without it being known in advance what they're trying to do. Oh, I understand that aspect. What I don't understand is how the investor, the large investor or whatever, gets the information that they're going to make more money on a dark venue than on a non-dark venue. I understand secrecy, but you don't know how they work, so that's all right. All right, so let's hear from the Commission. Thank you, Your Honor. May it please the Court. Petitioner's challenge is based on three fundamental flaws. First, petitioners mischaracterize the goal of the rule. As the Commission made clear, the goal of the rule is to modernize the decades-old exclusive processor model by enhancing data content and speed and replacing it with a competitive marketplace to meet the core data needs of all market participants. Second, petitioners ignore the relevant standard. The law requires that the Commission provide a rational connection between the facts found and the choice made. And here, no one denies the need for updating that outdated model, and surely there is a rational connection between replacing that guaranteed monopoly, as NYSE referred to it, and that's a Joint Appendix 583, with a competitive marketplace. Third, petitioners ask this Court to focus on the rule's potential impact on their profits and ignore the Commission's reasonable determination, based on years of study and the comments of a cross-section of market participants supporting the rule, that the benefits of deregulating this guaranteed monopoly justify the costs. Now, specifically, the Commission found that the rule will foster competition in a number of areas, reduce information asymmetries between prop data users, so those are the proprietary data users, and the consolidated data subscribers that will benefit from this rule, reduce latency for consolidated data delivery, increase market resiliency, lower data fees and transaction costs, and lead to a more informed trading decisions, better execution quality, and a narrower NBVO. As this Court recently recognized, the equity markets experienced a sea change in these past few decades. Exchanges are demutualized for-profit companies. This is a quote from this Court's decision in the Intercontinental Matter. Trading is overwhelmingly automated and electronic, speed is key, latency is the enemy of speed, and trading speeds today are measured in microseconds. Now, it takes 350,000 microseconds for the human eye to blink. Think of that. This rule addresses a market failure that has developed under the current monopoly structure. The exclusive processors, the guaranteed monopoly that petitioners control, are far inferior in content and speed to the lucrative proprietary data products that they sell. They are on both sides of this market, Your Honors. Congress was clear in its directive to the Commission. It's an 11-cap-A, adopted in 1975, and authorized the Commission to prescribe rules and regulations to facilitate a national market system to assure the prompt, accurate, reliable, and fair collection, processing, distribution, and publication of information with respect to quotations for and transactions and security. If you could talk about innovation. So these exchanges, you know, need to develop certain kinds of proprietary data, data that obviously the Commission thinks is more valuable than the core data that was initially going out. So under this rule, I mean, what incentive is there for the exchanges to develop proprietary data? I know the Commission says, oh, well, we're not planning to include any new sources of proprietary data into core data, but I mean, you know, I don't know. I've seen how regulation works. I don't think it's unfair for the exchanges to assume that in a few years, whatever they develop is going to get, you know, sort of sucked into the core data category. Well, that's an excellent question, Your Honor. I'm glad you asked that because there's a couple of things going on here that I think are an undercurrent, and one of them is that there is no competition for proprietary data products, and that is in our release. It's in our brief. This court has held in the only court decision on the merits of whether or not proprietary data feeds compete in the net coalition decision. And this court came to the conclusion that there is no evidence that they compete, and the reason why is simple, because the data from one exchange is unique to that exchange. You can't replicate it. There's no clear substitute for every use case, and this is based on economic data in the record, and tons of commenters agree. So, you know, that's one thing that we need to make clear at the outset here, the idea that there's competition. How does that affect innovation? Okay, well, yes. I don't. I just don't. I mean, I imagine it will dampen, you know, the creation of new forms of proprietary data. Well, no. What the commission concluded, based upon its economic analysis, and that's in the appendix. There's 86 pages of Federal Register pages on economic analysis. Within that, the commission determined that the idea that the exchanges would not be able to innovate if they saw a profit-seeking or profit-providing opportunity is contrary to economic theory. You have. They can still sell proprietary data products. They will still sell proprietary data products. They will still make money from connectivity fees, and they will still innovate in order to make faster. That's not addressing my question, which is that the incentives to innovate are certainly going to be dampened if they believe there is, for instance, a very small window in which they can profit from their proprietary data. Well, another point. SEC just sort of scoops it into the, you know, core data bucket. Well, I mean, let me let me make this clear. Also, Your Honor, they have never opposed the idea of including these additional data components into core data. What they oppose is allowing competition to be the method to distribute that data. They don't want to compete on that end. And so what the commission determined was that there will be an opportunity for innovation on the consolidation side. You know, the two biggest, you know, or the two entities that are best situated to become competing consolidators are the exchange affiliates who run the monopoly system right now. So the idea that opening up the competition will actually depress innovation is contrary to reason. I understand their position is that they are enjoying their position in the market. They enjoy having the monopoly control at the same time that they're selling the products that don't compete with each other and that one would need to buy every one of them. Now, not every exchange sells a proprietary data product. Some give it away for free depending on their business model. So what we're saying is competition should dictate who are the winners and losers here, not the regulation. And that is clear, again, from the record, from the comments, we have comments spanning the securities industry in favor of this proposal. And the only opponents to this were the ones who are both controlling the monopoly and selling the data products that people are increasingly needing because of the need for information and for speed. Under the existing system, what is preventing competition and consolidation? The system. I mean, it doesn't allow for competition within consolidation. Is that because the major exchanges won't sell to somebody else? No, it's because there is an inefficiency within the market. So the way that the system was set up, and this was set up in 2005, is that there are two feeds, two exclusive processor feeds that come out. They don't compete with each other because they contain different information from different exchanges. And so those are the consolidated feeds. And they also contain regulatory and administrative data that everyone needs to trade with. But their prices are set by a committee, by the equity data plans, and those prices are set for everyone. And so that's the utility type processing system that's in place. There's no competition for that because if you were to take that feed, you cannot. It's only the base information that's provided in it. What's provided and what people do sell, people do aggregate the exchange's prop data feeds and sell them for profit. But there's no competition in the exchange's setting the prices for those feeds. So they're all the exclusive processors for their own data. And if you need all of the data in order to understand what's happening in the marketplace, well, they have market power. And that's what the commission found. And that's what economists have found. And so that's why there's no competition currently in that area. And that's the market failure, Your Honor, that that commission is is attempting to resolve in this rule. Thank you. So let me let me just address one question. Petitioners will suggest there's no one waiting in the wings to become a competing consolidator. And, you know, that's just again, that's that's contrary to economic theory, but it's also contrary to the record. There are a number of commenters who stepped up in the comments and said, yes, we're ready and willing to step into the competitive marketplace. And they run the gambit. There are there are data vendors who are willing and ready to do this. Other exchanges are waiting in the wings to enter this market. Is that footnote 209? That is 2109. 21. Yes. In the joint appendix. So, you know, it is it behooves, you know, rationality to to to decide that there's somehow no one would like to come in and and get a cut of this one hundred million dollar, you know, marketplace that, you know. Let me tell you another thing. Well, Mr. Are people waiting in the wings as we're kind of moving into this through this phase in process? I mean, I know that isn't directly before us, but I am curious to know how this if this is working the way the commission thought that it would. Sure. I mean, right now we're at the point where we're before the commission is the equity data plans, the proposals for the data products that would be coming from the exchanges. And so the way it works. And I wanted to make this quick as Judge Randolph had a question about this, too, is that is that a competing consolidator will connect directly with each exchange at their data centers. And then they will be able to consolidate and send the streams to their subscribers. We're at the point where they're setting the equity data plans are setting the fees. And that's where the fee structure will come from. And so that's before the commission right now. And after that, it's only after those fees are established that the period for competing consolidators to register opens because we understand that people want to know what the prices are that are going to be charged in the market before entering. And so once that opens, it's open for a certain period of time in which the commission said, you know, there should be a first mover advantage. And so it should be given incentive to those who are interested to enter the market. And then the competing consolidators will start running in parallel to the exclusive processors that are still running. You know, and so that that is that is where we are right now. Thank you. Another point, Judge Rogers, you had a question about who invests in dark pools or who uses a dark pool to transact to make securities transactions, equity transactions. And the answer to that is often large institutional players like mutual funds or our banks. And the reason for that is part of the problem because of the inequities of information is that if you go to a market with your transaction, if you leave your And so these these what they call them dark pools. The reason I know it sounds like it's a, you know, it's sinister, but it's not. It just means that you don't have to show your liquidity on the market. You can you can interact and you are guaranteed to get at least the best going price. If, you know, it's transacted on the on the on the dark pool that they're called alternative trading systems and they're run by large broker dealers and so on. It's just one aspect of of the market that in which we actually frankly have very robust competition for transaction services. You know, as counsel mentioned, there's like 15 exchanges dozens of alternative trading systems. And so that end of the market is very, very competitive. It's the data processing and providing. It's the idea of a consolidated feed that Congress envisioned in 1975 that everyone agrees is lagging at this point. And really the only question before the court is, is how to, how do you best address that. And, and that's what this role does the role takes the data components that no one agreed no one. No one disagrees with being helpful to the market being the information that people increase that market participants increasingly need in order to compete. The only question is how can we deliver that and the commission decided that creating a competitive marketplace, consistent with, you know, our values for hundreds of years would be the most appropriate way to go about doing this. Thank you. Any questions. Judges. Thank you. All right. Thank you, counsel. All right. This Council for petitioners have any time left. So have a minute and 30 seconds. All right, give me two minutes Council. Thank you, Your Honor, a few brief points. First of all, Council for the SEC claims for the first time today that reducing information asymmetries was not a goal of the rule that's plainly that's plainly false. We've cited numerous provisions in our, in our brief. I'll just point the court to Jay to 19 where the court says that this rule will address the concerns about and improve the content and latency differentials that currently exists. It's unquestionably a claim to benefit of the rule, and when their claim benefit is actually a detriment that renders the rule arbitrary and capricious Council claims that exchange proprietary feeds are faster than, than the consolidated feed and that this is due to alleged failure to innovate. In fact, as we noted in our brief and it's undisputed, the aggregation speed which is in the control of the steps is less than 20 millionths of a second the reason consolidated fees are slower than proprietary fees is because they have to be aggregated. And that means you have to wait for the data to arrive from from multiple different exchanges in multiple different locations and there is inherent geographic latency when you're sending data over fiber, or even microwaves from miles away and so the consolidated fees will always be slower than the direct feeds that's an inevitable consequence of geography, not a consequence of alleged failures to innovate, it will be true in the new system as well. Innovation will indeed be hindered as we've shown in our brief, and the commission utterly failed to address that concern. They say, oh well the exchanges will have no reason not to try and innovate because they can try to make money and what little is left of proprietary data, but the fact is, now that their attempts to do that have been confiscated the bait and switch is going to render them unwilling to do it again and plus they will have dramatically reduced funding and the commission utterly failed to consider the economic impact and adverse consequences of the elimination of this incentive to innovate. Mr. Freitas main position seems to be that that the petitioners here are just, you know, are just, you know, unhappy that they're losing their monopoly position. I mean, what is your response to that responses. Number one. He also said that there's that there's no evidence and the commission denies that exchanges compete in proprietary data that is completely untrue the Commission says at page 227 it talks about the ways in which it envisions the competing consolidators are going to concede can compete, including on latency and then it says competing consolidators will likely compete along all of these lines, similar to the manner in which the providers of proprietary data products have competed. That's 227 in the Joint Appendix. They do compete right now. It's not the competition is the problem. It's not that even competition with competing consolidators is the problem. Indeed, the exchanges at various points have proposed competing competition models for consolidated data. The problem is the manner in which the Commission has implemented it and the fact that the Commission's proffered justification completely underpaid undercuts what they've done because they're making the problem that they claim to be addressing worse. If unless the court has any further questions, we ask that the rule be vacated as arbitrary, capricious and contrary to law. Thank you. Council will take the case under advisement.